How can metformin help you? What the research says

Metformin, a prescription drug used to manage diabetes, has shown some benefits in lower risks of cancer and better survival, mostly among people with diabetes or high blood sugar.

How can metformin help you? What the research says

We summarize the clinical evidence for each medical benefit here. We begin with our assessment of the strength of evidence within each category, followed by a brief summary of individual studies or reviews of several studies. In assessing the strength of evidence, we consider the study design, number of participants, and the size of the treatment effect (how much outcomes changed with treatment).

To see more details, click the plus sign to the right of any section.

Our assessments of evidence for each medical benefit fall into one of these categories:

Strong evidence: consistent, significant effects in several large (or at least one very large) well designed clinical studies or at least two meta-analysesa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of clinical studies of moderate or better quality (or one large meta-analysis) finding similar results
Good evidence: significant effects in one large or several mid-sized and well-designed clinical studies ( randomized controlled trialsa study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects with an appropriate placebo or other strong comparison control or observational studies that control for confounds)
Modest evidence: significant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies), or several small studies aggregated into a meta-analysis
Preliminary evidence: significant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect
Weak evidence: one or more case studies, supported by animal evidence OR small treatment effects of limited clinical significance OR studies with no controls OR weak trends of effects
Insufficient evidence: preclinical evidence only OR clinical studies with such poor or unclear methodology that no conclusion can be drawn OR conflicting findings across clinical studies with no preponderance of evidence in one direction; conflicting evidence occurs when studies find conflicting effects (positive effect vs no effect or negative effect) with the same treatment and the same general study population (same cancer type, for example)

Learn more about how we research and rate therapies and practices in How We Rate Therapies ›

Commentary on research methodology

CancerChoices Lead Researcher Nancy Hepp and Research Associate Andy Jackson, ND: Trying to make sense of the research evidence regarding metformin’s effects on cancer incidence and outcomes is fraught with obstacles. Far too many studies do not report whether the study population includes people without diabetes. This makes a huge difference in interpreting the study findings, as people with diabetes have a much higher risk of cancer and of worse outcomes after a cancer diagnosis. See Blood sugar and insulin resistance ›

Some examples of how different comparisons can change interpretations of data:

Example 1: People with diabetes treated with metformin are compared to people without diabetes in terms of cancer incidence or survival in an observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods study. No difference is found and the researchers conclude that metformin provides no benefit.

This is a little bit like studying people who have a special booster device for running speed. The people with the device start 100 yards back from the starting line in a race, but they cross the finish line at the same time as people starting at the starting line. The researchers conclude that the special device doesn’t improve running speed because everyone finishes at the same time.

People with diabetes would be expected to have higher incidence and worse survival for some types of cancer compared to people without diabetes. Finding that people with diabetes have similar results to people without diabetes may be evidence that metformin is associated with improved outcomes among people with diabetes, but this is reported as “no evidence of an effect.” Some studies don’t adjust their data or even mention the different starting points for people with diabetes compared to people without diabetes.

Since a huge majority of people treated with metformin in observational studies have diabetes, a comparison simply of metformin vs no metformin without accounting for diabetes is not an appropriate comparison. A better comparison is between people with diabetes treated with metformin compared to people with diabetes not treated with metformin. Some of the very large observational studies do not make this comparison. In our summaries, we give as much detail about the diabetes status of study participants as the researchers provide, but too often we’re left wondering.

Designing a study observing the effects of metformin compared to a placebo on cancer outcomes among people with diabetes would be difficult. Ethical issues would arise, as withholding diabetes treatment from people who need it would lead to ethical and likely medical issues that could negatively affect the patients as well as the study outcomes. Studies comparing cancer outcomes among people treated with metformin compared to other diabetes treatments are available, but the interpretation of the comparison is not as clean as with a placebo. Some studies compare metformin to one or two specific diabetes drugs, while other studies lump all the other diabetes treatments together. Comparing across studies is difficult, as which specific drugs are used as the comparison makes a big difference in interpretation.

In this review we grouped studies into 1) those limited to people with diabetes, 2) those comparing people with diabetes treated with metformin to people without diabetes, and 3) those limited to people without diabetes. We found highly reliable differences in outcomes across cancer types. Metformin consistently shows benefits for cancer risk and survival among people with diabetes, but not much benefit among people without diabetes. Data on specific cancer types can vary, and more research is needed.

However, even just among people with diabetes, the study design may still be building in a complication (or confound, as researchers term it). People who are treated with metformin tend to have less advanced or less complicated diabetes. When metformin is no longer sufficient to provide blood sugar control, people are often switched to other treatments. Unless studies look specifically at people’s levels of blood sugar and/or insulin resistancea condition in which cells in your muscles, fat, and liver don’t respond well to insulin and can’t efficiently take up glucose from your blood for energy, they may be comparing the people with more advanced diabetes (not using metformin) to people with less advanced diabetes (using metformin). We would expect higher risk of cancer and worse survival among people with more advanced diabetes, and therefore comparatively better survival and lower risk among people with less advanced disease, who coincidentally are the people treated with metformin. The reason for the difference may not be the use of metformin, but the status of blood sugar control and insulin resistance. We found few studies controlling for this confound.

Example 2: People with cancer without a diabetes diagnosis are assigned to metformin or a placebo treatment in a randomized controlled triala study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects. No difference is found between the groups, and so the researchers conclude that metformin has no benefit.

One problem here is that not having a diabetes diagnosis is not the same as not having diabetes. In the United States, “22.8% of adults with diabetes are undiagnosed.”¹National Diabetes Statistics Report. Centers for Disease Control and Prevention. May 15, 2024. Viewed May 28, 2024. The chances are fairly high that some of the people in the study actually have diabetes or at least prediabetes.

Hidden metabolic and diabetes status shouldn’t be a big issue in a large study where randomization is expected to evenly assign people across the different treatment groups. However, in a small study, having two or three more people with prediabetes in one group compared to another could skew the results.

Even further, anticancer activity of metformin may be seen among people without diabetes but with other metabolic abnormalities compared to people with no metabolic abnormalities. Some intriguing results show that not just diabetes but the metabolic status of people with primary breast cancer—or perhaps even the tumors—can impact outcomes. Three studies we’ve reviewed have found different trends in a markera chemical or substance, such as certain proteins or genetic material, that are associated with the presence of cancer or a change in status or prognosis; these markers can be detected in blood, urine, or tissue. Tumor markers are not direct measures of clinical outcomes such as survival or metastasis, and if a therapy or treatment shows an impact only on tumor markers, we cannot surmise that it will affect survival. of tumor proliferation depending on metabolic markers among people without diabetes:

Metformin use led to a decrease in proliferation among nondiabetic women with breast cancer and higher insulin resistance, but an increase in proliferation among women with lower resistance.²Bonanni B, Puntoni M et al. Dual effect of metformin on breast cancer proliferation in a randomized presurgical trial. Journal of Clinical Oncology. 2012 Jul 20;30(21):2593-600.
Another study found a similar but nonsignificant (weak) trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently toward opposite effects on proliferation among nondiabetic people with breast cancer and insulin resistance compared to normal insulin sensitivity.³Cazzaniga M, DeCensi A et al. The effect of metformin on apoptosis in a breast cancer presurgical trial. British Journal of Cancer. 2013 Nov 26;109(11):2792-7.
A third study found lower proliferation only among nondiabetic people with breast cancer and high insulin resistance, and a weak trend toward higher proliferation among people with low insulin resistance.⁴DeCensi A, Puntoni M et al. Differential effects of metformin on breast cancer proliferation according to markers of insulin resistance and tumor subtype in a randomized presurgical trial. Breast Cancer Research and Treatment. 2014 Nov;148(1):81-90.

Another study found no evidence of effect of metformin overall, but lower risk of progression and mortality among the subset of people with tumors with higher glucose metabolism when metformin was added to chemotherapy.⁵Lee Y, Joo J et al. Randomized phase II study of platinum-based chemotherapy plus controlled diet with or without metformin in patients with advanced non-small cell lung cancer. Lung Cancer. 2021 Jan;151:8-15.

We suspect that the conflicting and sometimes confusing findings across studies may be due in part to a lack of enough attention to the metabolic/diabetes status among people in the study. Blanket conclusions about effects may be oversimplified to the point of being misleading. To help us as we interpret study findings, we separate studies into those limited to people with diabetes and those including people without diabetes. In some cases, we’ve had to make our best guess, as the study authors do not provide this information.

Other research analysis approaches find fault with many of the studies here, and in fact with many observational studies overall. One such fault is failing to account for immortal time bias created when scientists do not account for delays and dropouts that happen during the treatment process, which can make a treatment seem more effective than it really is. Bias is introduced when this period of “immortality” is misclassified or excluded during analysis.⁶Catalogue of Bias. Immortal time bias. Centre for Evidence-Based Medicine. Viewed August 4, 2022. “The association between metformin and pancreatic cancer survival has been greatly exaggerated in previous cohort studies due to the wide existence of immortal time bias.”⁷Wei M, Liu Y, Bi Y, Zhang Z. Metformin and pancreatic cancer survival: real effect or immortal time bias? International Journal of Cancer. 2019 Mar;145(7):1822-1828.

We do our best to report and interpret the results we find, but better study designs and reporting are definitely needed to come to solid conclusions about the effectiveness of metformin regarding benefits as an adjunctivea therapy connected or added to a main treatment or therapy, not used alone treatment for nondiabetic people with cancer.

Improving treatment outcomes

Is metformin linked to improved survival? Is it linked to less cancer growth or metastasis? Does it enhance the anticancer action of other treatments or therapies? We present the evidence.

People with type 2 diabetes are at higher risk for poor cancer outcomes due to specific diabetes-related processes that promote certain cancer types. Metformin thwarts some of these cancer-promoting processes and helps correct terrainthe internal conditions of your body, including nutritional status, fitness, blood sugar balance, hormone balance, inflammation and more imbalances due to diabetes. As a result, we are not surprised to find that metformin shows the biggest benefits among people with cancer types linked to type 2 diabetes, insulin resistancea condition in which cells in your muscles, fat, and liver don’t respond well to insulin and can’t efficiently take up glucose from your blood for energy, and/or high blood sugar.

Cancer as a whole

Diabetic people with cancer treated with metformin had lower cancer-specific mortality across many observational studies.

People with diabetes: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower cancer-specific mortality among people with diabetes treated with metformin

26% lower cancer-specific mortality, with most benefit for colorectal cancer, among people with cancer and diabetes treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 21 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies⁸Lega IC, Shah PS et al. The effect of metformin on mortality following cancer among patients with diabetes. Cancer Epidemiology, Biomarkers & Prevention. 2014 Oct;23(10):1974-84.
Lower cancer-specific mortality among people with both cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 28 observational studies⁹Zhang ZJ, Li S. The prognostic value of metformin for cancer patients with concurrent diabetes: a systematic review and meta-analysis. Diabetes, Obesity & Metabolism. 2014 Aug;16(8):707-10.
38% lower cancer-specific mortality among people with both diabetes and cancer treated with metformin compared to other glucose-lowering medications in a large meta-analysis of 20 observational studies¹⁰Yin M, Zhou J, Gorak EJ, Quddus F. Metformin is associated with survival benefit in cancer patients with concurrent type 2 diabetes: a systematic review and meta-analysis. Oncologist. 2013 Dec;18(12):1248-1255.

Advanced cancer

Metformin did not show an effect on survival or tumor response among people with advanced cancer as a whole, whether with or without diabetes, across many studies. Diabetic people with pancreatic cancer with locally advanced tumors but not among metastatic tumors treated with metformin showed lower mortality.

Advanced cancer as a whole

People with diabetes: no evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on survival or tumor response among people with advanced or metastatic cancer (mostly with diabetes) treated with metformin in a combined analysis of studies

No evidence of an effect on tumor response, progression-free survivalthe time during and after treatment of a disease that a person lives without disease progression (worsening), or overall survival among people with advanced or metastatic cancer (mostly with diabetes) treated with metformin with systemic anticancer therapy compared to the anticancer therapy alone in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 9 RCTsrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects¹¹Kim HS, Kim JH, Jang HJ, Lee J. The addition of metformin to systemic anticancer therapy in advanced or metastatic cancers: a meta-analysis of randomized controlled trials. Int J Med Sci. 2020 Sep 12;17(16):2551-2560.

People without diabetes: no evidence of an effect on overall or progression-free survival among nondiabetic women with metastatic breast cancer treated with metformin in a combined analysis of studies

No evidence of an effect on overall or progression-free survival among nondiabetic women with metastatic breast cancer treated with metformin compared to no metformin in a meta-analysis of 3 RCTs¹²Lusica PMM, Eugenio KPY, Sacdalan DBL, Jimeno CA. A systematic review and meta-analysis on the efficacy and safety of metformin as adjunctive therapy among women with metastatic breast cancer. Cancer Treatment and Research Communications. 2021;29:100457.

Advanced pancreatic cancer: modest evidence of lower mortality among diabetic people with pancreatic cancer with locally advanced tumors but not with metastatic tumors treated with metformin

No evidence of an effect on mortality among diabetic people with advanced pancreatic cancer treated with metformin compared to no metformin in a large meta-analysis of 21 observational studies¹³Shi YQ, Zhou XC et al. Relationships are between metformin use and survival in pancreatic cancer patients concurrent with diabetes: a systematic review and meta-analysis. Medicine (Baltimore). 2020 Sep 11;99(37):e21687.
Lower mortality among diabetic people with pancreatic cancer with locally advanced tumors but not among people with metastatic tumors treated with metformin compared to no metformin in a meta-analysis of 9 observational studies¹⁴Li X, Li T, Liu Z, Gou S, Wang C. The effect of metformin on survival of patients with pancreatic cancer: a meta-analysis. Scientific Reports. 2017 Jul 19;7(1):5825.

Bladder cancer

Diabetic people with bladder cancer treated with metformin showed lower cancer-specific mortality in several studies. People with bladder cancer and diabetes had higher rates of progression-free survival and lower cancer-specific mortality across many studies.

People with diabetes

Modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower cancer-specific mortality among people with both bladder cancer and diabetes treated with metformin

22% lower cancer-specific mortality among people with bladder cancer and diabetes treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 3 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies¹⁵Yao X, Liu H, Xu H. The impact of metformin use with survival outcomes in urologic cancers: a systematic review and meta-analysis. Biomed Research International. 2021 Oct 8;2021:5311828.

Modest evidence of better progression-free survivalthe time during and after treatment of a disease that a patient lives without disease progression (worsening) among people with both bladder cancer and diabetes treated with metformin

Better progression-free survival and 43% lower cancer-specific mortality among people with both bladder cancer and diabetes treated with metformin compared to other people in a very large meta-analysis of 9 observational studies¹⁶Hu J, Chen JB et al. Association of metformin intake with bladder cancer risk and oncologic outcomes in type 2 diabetes mellitus patients: a systematic review and meta-analysis. Medicine (Baltimore). 2018 Jul;97(30):e11596.

Brain and nervous system cancers

Diabetic people with glioma or glioblastoma treated with metformin had better overall survival in two studies. Diabetic people with glioblastoma treated with metformin saw slower cancer progression after radiotherapy, whether with or without chemotherapy, in two studies.

People with diabetes

Modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of better overall survival among people with glioma or glioblastoma and diabetes treated with metformin

25% lower progression-free mortality and 36% lower overall mortality among people with WHO grade 3 glioma (but not grade 4 as described above in Advanced cancer) and better progression-free survival among people with glioma and diabetes treated with metformin compared to no metformin in a very large observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods study¹⁷Seliger C, Luber C et al. Use of metformin and survival of patients with high-grade glioma. International Journal of Cancer. 2019 Jan;144(2):273-280.
Better overall survival among people with glioblastoma and diabetes treated with metformin compared to other diabetes treatments in a mid-sized observational study¹⁸Welch MR, Grommes C. Retrospective analysis of the effects of steroid therapy and antidiabetic medication on survival in diabetic glioblastoma patients. CNS Oncology. 2013 May;2(3):237-46.

Modest evidence of longer progression-free intervals after radiation therapy among diabetic people with glioblastoma treated with metformin

Longer progression-free intervals after radiation therapy among people with glioblastoma and diabetes treated with metformin compared to no metformin in a mid-sized observational study¹⁹Adeberg S, Bernhardt D et al. Metformin influences progression in diabetic glioblastoma patients. Strahlentherapie und Onkologie. 2015 Dec;191(12):928-35.
Better estimated progression-free survival and a weak trend toward better estimated overall survival after radio-chemotherapy (TMZ/RT) among people with glioblastoma and diabetes treated concurrently with metformin (without other antidiabetes drugs) compared to people without diabetes in a large observational analysis within an RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects; the people treated with metformin were older and in worse health than the people without diabetes, and we would have expected worse survival among those on metformin²⁰Seliger C, Genbrugge E et al; EORTC Brain Tumor Group. Use of metformin and outcome of patients with newly diagnosed glioblastoma: pooled analysis. International Journal of Cancer. 2020 Feb 1;146(3):803-809.

Breast cancer

Diabetic people with breast cancer or insulin resistance have shown better survival and tumor response when treated with metformin across many studies, and one large, rigorous study found a lower rate of relapse. Among nondiabetic people with breast cancer, metformin has not shown an effect on progression-free or overall survival, although studies have found more tumor response to treatment and less metastasis.

People with diabetes

Survival: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of better survival among people with breast cancer and diabetes treated with metformin

45% lower overall mortality among people with breast cancer and type 2 diabetes treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 11 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies²¹Tang GH, Satkunam M et al. Association of metformin with breast cancer incidence and mortality in patients with type ii diabetes: a GRADE-assessed systematic review and meta-analysis. Cancer Epidemiology, Biomarkers & Prevention. 2018 Jun;27(6):627-635.
Better overall survival among diabetic people with hormone-receptor positive but not negative breast cancer treated with metformin compared to no metformin in a mid-sized observational analysis from a larger RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects²²Sonnenblick A, Agbor-Tarh D et al. Impact of diabetes, insulin, and metformin use on the outcome of patients with human epidermal growth factor receptor 2-positive primary breast cancer: analysis from the ALTTO phase III randomized trial. Journal of Clinical Oncology. 2017 May 1;35(13):1421-1429.
No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on overall survival or cancer-specific survival among people with breast cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 3 observational studies²³Coyle C, Cafferty FH, Vale C, Langley RE. Metformin as an adjuvant treatment for cancer: a systematic review and meta-analysis. Annals of Oncology. 2016 Dec;27(12):2184-2195.
Slightly better cancer-specific survival and moderately better overall survival among people with both diabetes and breast cancer treated with metformin compared to no metformin in a meta-analysis of 11 observational studies²⁴Xu H, Chen K et al. Metformin use is associated with better survival of breast cancer patients with diabetes: a meta-analysis. The Oncologist. 2015 Nov;20(11):1236-44.
A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower overall mortality among people with breast cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 4 observational studies²⁵Lega IC, Shah PS et al. The effect of metformin on mortality following cancer among patients with diabetes. Cancer Epidemiology, Biomarkers & Prevention. 2014 Oct;23(10):1974-84.

Relapse: preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of better relapse-free survival after exemestane treatment among women with diabetes and breast tumors with high insulin-like growth factor type 1 receptor (IGF‐1R) expression treated with metformin

Better relapse-free survival after exemestane treatment among women with diabetes and breast tumors with high insulin-like growth factor type 1 receptor (IGF‐1R) expression but not low IGF‐1R expression treated with metformin compared to exemestane alone in a very large RCT²⁶Engels CC, de Glas NA et al. The influence of insulin-like growth factor-1-receptor expression and endocrine treatment on clinical outcome of postmenopausal hormone receptor positive breast cancer patients: a Dutch TEAM substudy analysis. Molecular Oncology. 2016 Apr;10(4):509-16.

Metastases: weak evidenceone or more case studies, supported by animal evidence OR small treatment effects of limited clinical significance OR studies with no controls OR weak trends of effects (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of distant metastases among people with triple negative breast cancer and diabetes treated with metformin

Weak trends toward lower risk of distant metastases among people with triple negative breast cancer and diabetes treated with adjuvant chemotherapy and also treated with metformin compared to no metformin or compared to nondiabetic people in a large observational study²⁷Bayraktar S, Hernadez-Aya LF et al. Effect of metformin on survival outcomes in diabetic patients with triple receptor–negative breast cancer. Cancer. 2012;118(5):1202-1211.

Response: preliminary evidence of better disease response rates among people with breast cancer and diabetes treated with metformin

Substantially higher rates of pathologic complete responsethe lack of any sign of cancer in biopsy samples taken after cancer treatment is completed among people with breast cancer and diabetes treated with metformin compared to no metformin, comparable or possibly better response rates compared to people without diabetes in a large observational study²⁸Jiralerspong S, Palla SL et al. Metformin and pathologic complete responses to neoadjuvant chemotherapy in diabetic patients with breast cancer. Journal of Clinical Oncology. 2009;27(20):3297-3302.

People without diabetes

Survival: no evidence of an effect on progression-free or overall survival among nondiabetic women with breast cancer treated with metformin in combined analyses of studies

No evidence of an effect on progression-free or overall survival among nondiabetic women with breast cancer treated with metformin compared to no metformin in a meta-analysis of 5 RCTs²⁹Wang Q, Ma X et al. Metformin and survival of women with breast cancer: a meta-analysis of randomized controlled trials. Journal of Clinical Pharmacy and Therapeutics. 2022 Mar;47(3):263-269.
No evidence of an effect on overall or progression-free survival among nondiabetic people with breast cancer treated with metformin compared to placebo in a meta-analysis of 3 RCTs³⁰Wu Z, Qu B et al. The potential adjunctive benefit of adding metformin to standard treatment in inoperable cancer patients: a meta-analysis of randomized controlled trials. Annals of Translational Medicine. 2020 Nov;8(21):1404.
No evidence of an effect on rates of pathological complete response among nondiabetic people with HER2-positive breast cancer treated with metformin during chemo-immunotherapy compared to chemo-immunotherapy alone in a small RCT³¹Martin-Castillo B, Pernas S, Dorca J et al. A phase 2 trial of neoadjuvant metformin in combination with trastuzumab and chemotherapy in women with early HER2-positive breast cancer: the METTEN study. Oncotarget. 2018 Nov 2;9(86):35687-35704.

Tumor response: modest evidence of a lower marker of tumor proliferation among nondiabetic women with invasive breast cancer treated with metformin before surgery

Higher objective response rate among nondiabetic people with breast cancer treated with metformin compared to placebo in a meta-analysis of 3 RCTs³²Wu Z, Qu B et al. The potential adjunctive benefit of adding metformin to standard treatment in inoperable cancer patients: a meta-analysis of randomized controlled trials. Annals of Translational Medicine. 2020 Nov;8(21):1404.
Lower markers of tumor proliferation (up-regulation of pAMPK, down-regulation of pAkt, and a drop in Ki-67) among nondiabetic women with operable invasive breast cancer treated with metformin before surgery compared to no metformin in a small RCT³³Hadad SM, Coates P. Evidence for biological effects of metformin in operable breast cancer: biomarker analysis in a pre-operative window of opportunity randomized trial. Breast Cancer Research and Treatment. 2015 Feb;150(1):149-55; Hadad S, Iwamoto T et al. Evidence for biological effects of metformin in operable breast cancer: a pre-operative, window-of-opportunity, randomized trial. Breast Cancer Research and Treatment. 2011 Aug;128(3):783-94.
A decrease in in a marker of proliferation (Ki-67) after tumor biopsy among nondiabetic women with breast cancer with higher insulin resistance (HOMA greater than 2.8), higher body mass index, higher waist/hip girth ratio, alcohol consumption, or a higher marker of inflammation (CRP) treated with 850 mg metformin twice per day compared to placebo in a mid-sized RCT³⁴Bonanni B, Puntoni M et al. Dual effect of metformin on breast cancer proliferation in a randomized presurgical trial. Journal of Clinical Oncology. 2012 Jul 20;30(21):2593-600.
A weak trend toward higher tumor proliferation among nondiabetic people without insulin resistance (HOMA index less than 2.8) treated with metformin before breast cancer surgery in a small RCT³⁵Cazzaniga M, DeCensi A et al. The effect of metformin on apoptosis in a breast cancer presurgical trial. British Journal of Cancer. 2013 Nov 26;109(11):2792-7.

Preliminary evidence of a lower marker of tumor proliferation among nondiabetic people with HER2-positive DCIS lesions, and especially ER-positive/HER2-positive DCIS, treated with metformin

A lower marker of proliferation (Ki-67) among nondiabetic people with HER2-positive DCIS lesions, and especially ER-positive/HER2-positive DCIS, treated with 1,700 mg metformin once daily for 28 days before surgery compared to placebo in a mid-sized RCT³⁶DeCensi A, Puntoni M et al. Effect of metformin on breast ductal carcinoma in situ proliferation in a randomized presurgical trial. Cancer Prevention Research (Philadelphia). 2015 Oct;8(10):888-94.

Metastases: preliminary evidence of fewer cases of metastasis after chemotherapy and hormone therapy among nondiabetic women with breast cancer treated with metformin

Fewer cases of metastasis after chemotherapy and 6 months of hormone therapy among nondiabetic women with breast cancer treated with 850 mg metformin twice daily compared to no metformin in a mid-sized RCT³⁷El-Haggar SM, El-Shitany NA, Mostafa MF, El-Bassiouny NA. Metformin may protect nondiabetic breast cancer women from metastasis. Clinical & Experimental Metastasis. 2016 Apr;33(4):339-57.

People with non-diabetes metabolic imbalances

Survival: preliminary evidence longer progression-free survival survival among people with insulin resistance and breast cancer treated with metformin in combined analyses of studies

Longer progression-free survival among nondiabetic people with insulin resistance (HOMA greater than 2.5) treated with metformin during chemotherapy for breast cancer compared to chemotherapy alone in a mid-sized RCT³⁸Nanni O, Amadori D et al. Metformin plus chemotherapy versus chemotherapy alone in the first-line treatment of HER2-negative metastatic breast cancer. The MYME randomized, phase 2 clinical trial. Breast Cancer Research and Treatment. 2019;174:433–42.

Proliferation: modest evidence of lower markers of tumor proliferation among nondiabetic people with higher insulin resistance and breast cancer treated with metformin; also see findings that women without elevated insulin resistance had worse outcomes when using metformin in Safety and precautions ›

A lower marker of tumor proliferation (Ki-67) among nondiabetic women with breast cancer and with HOMA greater than 2.8 or one of several other markers of metabolic imbalance or inflammation treated with 850 mg metformin twice a day compared to placebo in a mid-sized RCT³⁹DeCensi A, Puntoni M et al. Differential effects of metformin on breast cancer proliferation according to markers of insulin resistance and tumor subtype in a randomized presurgical trial. Breast Cancer Research and Treatment. 2014 Nov;148(1):81-90.
A weak trend toward lower cancer proliferation among nondiabetic people with insulin resistance (HOMA index 2.8 or higher) and breast cancer treated with metformin for 4 weeks before surgery compared to placebo, with lower proliferation among people with insulin resistance (HOMA index 2.8 or higher) but higher proliferation among people without insulin resistance (HOMA index less than 2.8) in a small RCT⁴⁰Cazzaniga M, DeCensi A et al. The effect of metformin on apoptosis in a breast cancer presurgical trial. British Journal of Cancer. 2013 Nov 26;109(11):2792-7.

Colorectal cancer

Diabetic people with colorectal cancer treated with metformin had lower cancer-specific mortality across many studies. They also had lower rates of metastasis and greater reductions in a tumor markera chemical or substance, such as certain proteins or genetic material, that are associated with the presence of cancer or a change in status or prognosis; these markers can be detected in blood, urine, or tissue. Tumor markers are not direct measures of clinical outcomes such as survival or metastasis, and if a therapy or treatment shows an impact only on tumor markers, we cannot surmise that it will affect survival. (CEA) in one study

People with diabetes

Modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower mortality, including cancer-specific mortality, among people with colorectal cancer and diabetes treated with metformin

20% lower cancer-specific mortality among people with colorectal cancer and diabetes treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 5 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies⁴¹Wang Q, Shi M. Effect of metformin use on the risk and prognosis of colorectal cancer in diabetes mellitus: a meta-analysis. Anticancer Drugs. 2022 Feb 1;33(2):191-199.
Moderately lower cancer-specific mortality among people with diabetes and colorectal cancer, including metastatic cancer, treated with metformin compared to no metformin (not specific to diabetes) in a meta-analysis of 5 observational studies⁴²Ng CW, Jiang AA et al. Metformin and colorectal cancer: a systematic review, meta-analysis and meta-regression. International Journal of Colorectal Disease. 2020 Aug;35(8):1501-1512.
A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower cancer-specific mortality, with stronger effects among women, among people with type 2 diabetes and colorectal cancer treated with metformin compared to no metformin in a meta-analysis of 8 observational studies⁴³Wang Y, Xiao J, Zhao Y, Du S, Du J. Effect of metformin on the mortality of colorectal cancer patients with T2DM: meta-analysis of sex differences. International Journal of Colorectal Disease. 2020 May;35(5):827-835.
Lower cancer-specific mortality among people with colorectal cancer and type 2 diabetes treated with metformin compared to no metformin in a meta-analysis of 10 observational studies⁴⁴Cheng Y, Chen Y et al. For colorectal cancer patients with type II diabetes, could metformin improve the survival rate? A meta-analysis. Clinics and Research in Hepatology and Gastroenterology. 2020 Feb;44(1):73-81.
A weak trend toward lower cancer-specific mortality among people with colorectal cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 4 observational studies⁴⁵Meng F, Song L, Wang W. Metformin improves overall survival of colorectal cancer patients with diabetes: a meta-analysis. Journal of Diabetes Research. 2017;2017:5063239.
A weak trend toward lower cancer-specific mortality among people with colorectal cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 8 observational studies⁴⁶Tian S, Lei HB, Liu YL, Chen Y, Dong WG. The association between metformin use and colorectal cancer survival among patients with diabetes mellitus: an updated meta-analysis. Chronic Diseases and Translational Medicine. 2017 Sep;3(3):169-175.
Moderately lower cancer-specific mortality among people with colorectal cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 5 observational studies⁴⁷Du L, Wang M et al. Prognostic role of metformin intake in diabetic patients with colorectal cancer: an updated qualitative evidence of cohort studies. Oncotarget. 2017 Apr 18;8(16):26448-26459.
42% lower cancer-specific mortality among people with early stage colorectal cancer and diabetes treated with metformin compared to no metformin in meta-analyses of 2 observational studies⁴⁸Coyle C, Cafferty FH, Vale C, Langley RE. Metformin as an adjuvant treatment for cancer: a systematic review and meta-analysis. Annals of Oncology. 2016 Dec;27(12):2184-2195.
35% lower cancer-specific mortality among people with colon cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 4 observational studies⁴⁹Lega IC, Shah PS et al. The effect of metformin on mortality following cancer among patients with diabetes. Cancer Epidemiology, Biomarkers & Prevention. 2014 Oct;23(10):1974-84.
34% lower cancer-specific mortality during 41 months among people with colorectal cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 3 observational studies⁵⁰Mei ZB, Zhang ZJ et al. Survival benefits of metformin for colorectal cancer patients with diabetes: a systematic review and meta-analysis. PLoS One. 2014 Mar;9(3):e91818.

Preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of substantially lower risk of metastasis and greater reductions in a tumor markera chemical or substance, such as certain proteins or genetic material, that are associated with the presence of cancer or a change in status or prognosis; these markers can be detected in blood, urine, or tissue. Tumor markers are not direct measures of clinical outcomes such as survival or metastasis, and if a therapy or treatment shows an impact only on tumor markers, we cannot surmise that it will affect survival. among people with colorectal cancer and diabetes treated with metformin

Substantially lower risk of metastasis and greater reductions in a tumor marker (carcinoembryonic antigen, or CEA) among people with colorectal cancer and diabetes treated with metformin compared to no metformin in a mid-sized observational study⁵¹Henderson D, Frieson D, Zuber J, Solomon SS. Metformin has positive therapeutic effects in colon cancer and lung cancer. American Journal of the Medical Science. 2017 Sep;354(3):246-251.

Gastrointestinal cancer

Colorectal and pancreatic cancers are listed separately.

Nondiabetic people with different types of gastrointestinal cancer treated with metformin had higher levels of a tumor suppressor (AMPK) after chemotherapy in one small study.

Metformin did not show evidence of an effect on levels of two tumor proliferation markers (Ki-67 and cleaved caspase-3) among nondiabetic people with esophageal squamous cell carcinoma in one study.

Diabetic people with liver cancer treated with metformin showed better survival across many studies. However, see evidence of worse survival among nondiabetic people with liver cancer taking metformin below.

Diabetic people with stomach cancer treated with metformin showed a weak trend toward better survival across several studies.

Gastrointestinal cancer as a whole: preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of tumor suppression among nondiabetic people with gastrointestinal cancer (including colorectal, pancreatic, gastroesophageal, and bile duct cancers) treated with chemotherapy and metformin

A higher metabolic tumor suppressor (AMPK activation) after chemotherapy among nondiabetic people with gastrointestinal cancer (including colorectal, pancreatic, gastroesophageal, and bile duct cancers) treated 500 mg metformin twice daily compared to chemotherapy alone in a small RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects⁵²Godara A, Siddiqui NS, Hachem H, Martell RE, Saif WM. First prospective study evaluating effect of metformin (M) on disease control (DC) and activation of AMP-activated protein kinase (AMPKα) in patients (pts) with GI malignancies. Journal of Clinical Oncology. 2018 Feb;36(4): 264–264.

Esophageal cancer: no evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on tumor proliferation among nondiabetic people with esophageal squamous cell carcinoma treated with metformin in a small study

No evidence of an effect on tumor proliferation (Ki-67 and cleaved caspase-3 immunostaining) among nondiabetic people with esophageal squamous cell carcinoma treated with 250 mg metformin per day for an average of 10 days compared to placebo in a mid-sized RCT⁵³Wang S, Lin Y et al. Low-dose metformin reprograms the tumor immune microenvironment in human esophageal cancer: results of a phase II clinical trial. Clinical Cancer Research. 2020 Sep 15;26(18):4921-4932.

Liver cancer

People with diabetes: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of better survival among people with liver cancer and diabetes treated with metformin

Substantially better overall survival at 1 year (3 studies), 3 years (4 studies), and 5 years (3 studies) after curative cancer therapies among people with hepatocellular carcinoma and type 2 diabetes treated with metformin compared to other antihyperglycemic agents in meta-analysesa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies⁵⁴Zhou J, Ke Y et al. Meta-analysis: the efficacy of metformin and other anti-hyperglycemic agents in prolonging the survival of hepatocellular carcinoma patients with type 2 diabetes. Annals of Hepatology. 2020 May-Jun;19(3):320-328.
A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower overall mortality among people with liver cancer with diabetes treated with metformin compared to no metformin in a meta-analysis of 7 observational studies of high quality⁵⁵Ma SJ, Zheng YX, Zhou PC, Xiao YN, Tan HZ. Metformin use improves survival of diabetic liver cancer patients: systematic review and meta-analysis. Oncotarget. 2016 Oct 4;7(40):66202-66211.

People without diabetes: see evidence of higher mortality among nondiabetic people with liver cancer treated with metformin in Safety and precautions ›

Stomach cancer: weak evidenceone or more case studies, supported by animal evidence OR small treatment effects of limited clinical significance OR studies with no controls OR weak trends of effects (this is the CancerChoices definition; other researchers and studies may define this differently) of lower mortality among people with stomach cancer and type 2 diabetes treated with metformin

A weak trend toward lower cancer-specific mortality among people with stomach (gastric) cancer and type 2 diabetes treated with metformin compared to no metformin in a pooled analysis of 3 observational studies⁵⁶Shuai Y, Li C, Zhou X. The effect of metformin on gastric cancer in patients with type 2 diabetes: a systematic review and meta-analysis. Clinical & Translational Oncology. 2020 Sep;22(9):1580-1590.

Gynecologic cancer

Ovarian cancer is listed separately.

Metformin shows conflicting evidence of an effect among people with cervical cancer and diabetes across two studies.

Diabetic people with endometrial cancer treated with metformin showed better survival across many studies. Two small studies found lower levels of a tumor proliferation marker (Ki-67) among nondiabetic people with endometrial cancer treated with metformin, but metformin showed insufficient evidence of an effect on progression or response to treatment among nondiabetic people in three studies.

Cervical cancer: insufficient evidenceconflicting findings across clinical studies with no preponderance of evidence in one direction; conflicting evidence occurs when studies find conflicting effects (positive effect vs no effect or negative effect) with the same treatment and the same general study population (same cancer type, for example) (this is the CancerChoices definition; other researchers and studies may define this differently) of an effect on survival among diabetic people with cervical cancer treated with metformin compared to no metformin

Comparable progression-free survival and cancer-specific survival among women with stage 1–4 cervical cancer treated with metformin (almost exclusively diabetic and more likely to be older, hypertensive, and dyslipidemic) compared to people not treated with metformin, whether diabetic or not, in a mid-sized observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods study;⁵⁷Takiuchi T, Machida H et al. Association of metformin use and survival outcome in women with cervical cancer. International Journal of Gynecological Cancer. 2017 Sep;27(7):1455-1463. we would expect worse cancer outcomes among the people treated with metformin, as described in the commentary above
No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on overall survival among people with cervical cancer and type 2 diabetes treated with metformin compared to no metformin in a mid-sized observational study⁵⁸Hanprasertpong J, Jiamset I et al. The effect of metformin on oncological outcomes in patients with cervical cancer with type 2 diabetes mellitus. International Journal of Gynecological Cancer. 2017 Jan;27(1):131-137.

Endometrial cancer

People with diabetes: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower mortality among people with endometrial cancer and diabetes treated with metformin

Moderately better progression-free survival (PFS) among people with endometrial cancer and diabetes treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 8 observational studies; PFS among people treated with metformin was comparable to people with endometrial cancer without diabetes⁵⁹Gong H, Chen Y, Zhou D. Prognostic significance of metformin treatment in endometrial cancer: a meta-analysis. Die Pharmazie. 2020 Aug 1;75(8):401-406.
53% lower overall mortality among women with endometrial cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 3 observational studies⁶⁰Chu D, Wu J et al. Effect of metformin use on the risk and prognosis of endometrial cancer: a systematic review and meta-analysis. BMC Cancer. 2018 Apr 18;18(1):438.
41% lower overall mortality among people with endometrial cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 3 observational studies⁶¹Meireles CG, Pereira SA et al. Effects of metformin on endometrial cancer: systematic review and meta-analysis. Gynecologic Oncology. 2017 Oct;147(1):167-180.
37% lower overall mortality among people with endometrial cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 6 observational studies⁶²Tang YL, Zhu LY et al. Metformin use is associated with reduced incidence and improved survival of endometrial cancer: a meta-analysis. Biomed Research International. 2017;2017:5905384.
46% lower overall mortality among people with endometrial cancer and diabetes treated with metformin compared to no metformin in meta-analyses of 9 observational studies⁶³Guo J, Xu K, An M, Zhao Y. Metformin and endometrial cancer survival: a quantitative synthesis of observational studies. Oncotarget. 2017 Aug 2;8(39):66169-66177.
52% lower overall mortality (7 studies) and better progression-free survival (2 studies) among people with endometrial cancer and diabetes treated with metformin compared to no metformin in meta-analyses of observational studies⁶⁴Xie W, Li T et al. Metformin use and survival outcomes in endometrial cancer: a systematic review and meta-analysis. Oncotarget. 2017 Aug 22;8(42):73079-73086.
50% lower mortality among women with endometrial cancer and type 2 diabetes treated with metformin compared to no metformin in meta-analyses of 3 observational studies⁶⁵Perez-Lopez FR, Pasupuleti V et al. Systematic review and meta-analysis of the effect of metformin treatment on overall mortality rates in women with endometrial cancer and type 2 diabetes mellitus. Maturitas. 2017 Jul;101:6-11.

People without diabetes

insufficient evidencepreclinical evidence only OR clinical studies with such poor or unclear methodology that no conclusion can be drawn OR conflicting findings across clinical studies with no preponderance of evidence in one direction; conflicting evidence occurs when studies find conflicting effects (positive effect vs no effect or negative effect) with the same treatment and the same general study population (same cancer type, for example) (this is the CancerChoices definition; other researchers and studies may define this differently) of an effect on tumor progression or response among mostly nondiabetic people with endometrial cancer treated with metformin

A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward higher complete response rates among mostly nondiabetic people with atypical endometrial hyperplasia (AEH) or endometrioid endometrial cancer, with higher response rates among people with AEH, treated with megestrol acetate plus metformin compared to megestrol acetate alone in a mid-sized RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects⁶⁶Yang BY, Gulinazi Y et alJ. Metformin plus megestrol acetate compared with megestrol acetate alone as fertility-sparing treatment in patients with atypical endometrial hyperplasia and well-differentiated endometrial cancer: a randomised controlled trial. BJOG. 2020 Jun;127(7):848-857.
Lower levels of a marker of tumor proliferation (Ki-67) among nondiabetic people with endometrioid endometrial cancer treated with metformin before surgery compared to no metformin in a small RCT⁶⁷Petchsila K, Prueksaritanond N et al. Effect of metformin for decreasing proliferative marker in women with endometrial cancer: a randomized double-blind placebo-controlled trial. Asian Pacific Journal of Cancer Prevention. 2020 Mar 1;21(3):733-741.
No evidence of an effect on markers of tumor progression among mostly nondiabetic women with atypical hyperplasia or endometrioid endometrial cancer treated with metformin for 1 to 5 weeks before surgery compared to placebo in a small RCT⁶⁸Kitson SJ, Maskell Z et al. Pre-surgical metformin in uterine malignancy (PREMIUM): a multi-center, randomized double-blind, placebo-controlled phase III trial. Clinical Cancer Research: An Official Journal of the American Association for Cancer Research. 2019 Apr 15;25(8):2424-2432.
No evidence of an effect on complete response among mostly nondiabetic women with atypical hyperplasia/endometrial intraepithelial neoplasia or early-stage endometrioid carcinoma treated with progestin therapy combined with metformin compared to progestin alone in a small observational study⁶⁹Acosta-Torres S, Murdock T et al. The addition of metformin to progestin therapy in the fertility-sparing treatment of women with atypical hyperplasia/endometrial intraepithelial neoplasia or endometrial cancer: little impact on response and low live-birth rates. Gynecologic Oncology. 2020 May;157(2):348-356.

A lower marker of proliferation (Ki-67) among nondiabetic people with operable endometrial cancer treated with metformin compared to no metformin in 2 small controlled trialsa study design in which people are assigned to either an experimental group or a control group to compare the outcomes from different treatment; assignment is not random, and so this is not as strong a study design as a randomized controlled trial, but still stronger than an uncontrolled trial⁷⁰Laskov I, Drudi L et al. Anti-diabetic doses of metformin decrease proliferation markers in tumors of patients with endometrial cancer. Gynecologic Oncology. 2014 Sep;134(3):607-14; Sivalingam VN, Kitson S et al. Measuring the biological effect of presurgical metformin treatment in endometrial cancer. British Journal of Cancer. 2016 Feb 2;114(3):281-9.

Head, neck, and oral cancer

Diabetic people with head and neck cancer as a whole, laryngeal squamous cell carcinoma, or hypopharyngeal cancer treated with metformin showed better survival, comparable to people with diabetes, across many studies.

Head and neck cancer as a whole

People with diabetes: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of better survival among diabetic people with head and neck cancer treated with metformin.

Comparable survival among presumably diabetic people with head and neck cancer treated with metformin compared to no metformin (not specific to diabetes) in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 7 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies;⁷¹Wang Y, Fu T et al. The association between metformin and survival of head and neck cancer: a systematic review and meta-analysis of 7 retrospective cohort studies. Current Pharmaceutical Design. 2020;26(26):3161-3170. our note: we would expect worse survival among diabetic people
71% higher survival among people with head and neck cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 2 observational studies⁷²Saka Herrán C, Jané-Salas E, Estrugo Devesa A, López-López J. Protective effects of metformin, statins and anti-inflammatory drugs on head and neck cancer: a systematic review. Oral Oncology. 2018 Oct;85:68-81.

People without diabetes: weak evidencestudies with no controls (this is the CancerChoices definition; other researchers and studies may define this differently) of higher tumor cell death and reduced tumor cell metabolic activity but no evidence of an effect on tumor cell proliferation in nondiabetic people with head and neck cancer treated with metformin

Higher cancer cell death (apoptosis) and markers of improved cell metabolism and function (stromal caveolin‐1 and B‐galactosidase), but no evidence of an effect on a marker of tumor proliferation (ki-67) among people with newly diagnosed head and neck squamous cell carcinoma (not specific to diabetes) treated with 2,000 mg metformin per day for 9 or more days before surgery compared to baseline in a small uncontrolled triala study in which a therapy is used, but without a comparison group against which to judge outcomes; an uncontrolled trial is considered a weak study design⁷³Curry J, Johnson J et al. Metformin effects on head and neck squamous carcinoma microenvironment: Window of opportunity trial. Laryngoscope. 2017 Aug;127(8):1808-1815.

Laryngeal cancer: preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of better survival and less advanced tumors at diagnosis among diabetic people with laryngeal squamous cell carcinoma treated with metformin

Better overall survival, less advanced tumors at diagnosis, and more favorable prognoses among diabetic people with laryngeal squamous cell carcinoma treated with metformin compared to no metformin in a mid-sized observational study⁷⁴Sandulache VC, Hamblin JS et al. Association between metformin use and improved survival in patients with laryngeal squamous cell carcinoma. Head & Neck. 2014 Jul;36(7):1039-43.

Hypopharyngeal cancer: preliminary evidence of lower overall mortality, especially late-stage mortality, and lower rate of metastases among diabetic people with hypopharyngeal cancer treated with metformin

Lower overall mortality, especially late-stage mortality, and lower rate of metastases among diabetic people with hypopharyngeal cancer treated with metformin compared to no metformin in a mid-sized observational study⁷⁵Tsou YA, Chang WD et al. The effect of metformin use on hypopharyngeal squamous cell carcinoma in diabetes mellitus patients. BMC Cancer. 2019 Aug 30;19(1):862.

Kidney cancer

People with diabetes and kidney cancer treated with metformin showed better survival and lower rates of progression across many studies.

People with diabetes

A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward better cancer-specific survival, with stronger effects with locoregional disease but no evidence of an effect with metastatic disease or on progression-free survival, among people with kidney cancer (mostly with diabetes) treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 6 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies⁷⁶Nayan M, Punjani N et al. Metformin use and kidney cancer survival outcomes: a systematic review and meta-analysis. American Journal of Clinical Oncology. 2019 Mar;42(3):275-284.
38% lower cancer-specific mortality among people with kidney cancer and diabetes treated with metformin compared to no metformin in a very large meta-analysis of 8 observational studies⁷⁷Li Y, Hu L, Xia Q, Yuan Y, Mi Y. The impact of metformin use on survival in kidney cancer patients with diabetes: a meta-analysis. International Urology and Nephrology. 2017 Jun;49(6):975-981.
Comparable cancer-specific mortality among mostly diabetic people with kidney cancer treated with metformin compared to no metformin (not specific to diabetes) in a mid-sized observational study;⁷⁸Hakimi AA, Chen L et al. The impact of metformin use on recurrence and cancer-specific survival in clinically localized high-risk renal cell carcinoma. Canadian Urological Association Journal. 2013 Nov-Dec;7(11-12):E687-91. our note: we would expect worse outcomes for the people with diabetes

Modest evidence of lower risk of progression among people with kidney cancer and diabetes treated with metformin

20% lower risk of progression and a weak trend toward lower overall mortality, but no evidence of an effect on cancer-specific mortality among people with kidney cancer and diabetes treated with metformin compared to no metformin in meta-analyses of 4 observational studies⁷⁹Yao X, Liu H, Xu H. The impact of metformin use with survival outcomes in urologic cancers: a systematic review and meta-analysis. Biomed Research International. 2021 Oct 8;2021:5311828.

Leukemia

Mainly nondiabetic people with acute lymphoblastic leukemia (ALL), including those with high ABCB1 expression, treated with metformin showed a lower risk of relapse in one study.

Lung cancer

People with lung cancer and diabetes treated with metformin showed better survival across many studies, although one study found no evidence of an effect on the risk of metastasis.

Nondiabetic people with lung cancer treated with metformin showed a higher objective response rate to treatment but no evidence of an effect on survival in several studies.

In one study that mixed people with lung cancer with and without diabetes, people whose tumors had higher glucose metabolism showed a lower risk of progression and better survival when treated with metformin, but metformin showed no evidence of an effect across the whole group on progression or survival.

People with diabetes

Survival: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower mortality among people with type 2 diabetes and lung cancer treated with metformin

35% lower cancer-specific mortality among people with type 2 diabetes and lung cancer treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 10 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies⁸¹Xiao K, Liu F et al. The effect of metformin on lung cancer risk and survival in patients with type 2 diabetes mellitus: a meta-analysis. Journal of Clinical Pharmacy and Therapeutics. 2020 Aug;45(4):783-792.
25% lower mortality among people with lung cancer and type 2 diabetes treated with metformin compared to no metformin in a meta-analysis of 10 observational studies⁸²Zeng S, Gan HX, Xu JX, Liu JY. Metformin improves survival in lung cancer patients with type 2 diabetes mellitus: a meta-analysis. Medicina Clinica (Barcelona). 2019 Apr 18;152(8):291-297.
23% lower overall mortality among people with lung cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 15 observational studies⁸³Xin WX, Fang L et al. Effect of hypoglycemic agents on survival outcomes of lung cancer patients with diabetes mellitus: a meta-analysis. Medicine (Baltimore). 2018 Mar;97(9):e0035.
27% lower mortality and longer progression-free survival among people with lung cancer and diabetes receiving epidermal growth factor receptor-tyrosine kinase inhibitor (EGFR-TKIs) therapy and treated with metformin compared to no metformin in a large observational study⁸⁴Hung MS, Chuang MC, Tsai YH, Yang YH. Metformin improves survival in lung cancer patients receiving EGFR-TKIs therapy. European Respiratory Journal. 2018;52(suppl62):2856.
21% lower cancer-specific mortality (6 studies), better progression-free survival (4 studies), and a weak trend toward better overall survival (13 studies) among people with lung cancer and diabetes treated with metformin compared to no metformin in meta-analyses of observational studies⁸⁵Zhong S, Wu Y, Yan X, Tang J, Zhao J. Metformin use and survival of lung cancer patients: meta-analysis findings. Indian Journal of Cancer. 2017 Jan-Mar;54(1):63-67.
10% lower overall mortality among people with type 2 diabetes and lung cancer treated with metformin compared to no metformin in a meta-analysis of 6 observational studies⁸⁶Tian RH, Zhang YG et al. Effects of metformin on survival outcomes of lung cancer patients with type 2 diabetes mellitus: a meta-analysis. Clinical & Translational Oncology. 2016 Jun;18(6):641-9.

Metastasis: no evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on risk of metastasis among people with lung cancer and diabetes treated with metformin in a preliminary study

No evidence of an effect on risk of metastasis among people with lung cancer and diabetes treated with metformin compared to no metformin in a mid-sized observational study⁸⁷Henderson D, Frieson D, Zuber J, Solomon SS. Metformin has positive therapeutic effects in colon cancer and lung cancer. American Journal of the Medical Science. 2017 Sep;354(3):246-251.

People without diabetes

Also see evidence of worse outcomes among people with lung cancer treated with metformin in Safety and precautions ›

Survival: No evidence of an effect on survival among nondiabetic people with lung cancer treated with metformin

No evidence of an effect on overall or progression-free survival among nondiabetic people with lung cancer treated with metformin compared to placebo in a meta-analysis of 4 RCTsrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects⁸⁸Wu Z, Qu B et al. The potential adjunctive benefit of adding metformin to standard treatment in inoperable cancer patients: a meta-analysis of randomized controlled trials. Annals of Translational Medicine. 2020 Nov;8(21):1404.

Objective response rate: modest evidence of higher objective response rate among nondiabetic people with lung cancer treated with metformin

Higher objective response rate among nondiabetic people with lung cancer treated with metformin compared to placebo in a meta-analysis of 4 RCTs⁸⁹Wu Z, Qu B et al. The potential adjunctive benefit of adding metformin to standard treatment in inoperable cancer patients: a meta-analysis of randomized controlled trials. Annals of Translational Medicine. 2020 Nov;8(21):1404.

Tumors with high glucose metabolism: preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of progression and lower mortality among a subset of people with lung cancer treated with metformin whose tumors had higher glucose metabolism, but no evidence of an effect across the group as a whole in one study

No evidence of an effect on risk of progression or mortality overall after chemotherapy among people with lung cancer (not specific to diabetes), but lower risk of progression and mortality among the subset of people with tumors with higher glucose metabolism, treated with chemotherapy plus 1000 mg metformin twice daily compared to chemotherapy alone in a mid-sized RCT⁹⁰Lee Y, Joo J et al. Randomized phase II study of platinum-based chemotherapy plus controlled diet with or without metformin in patients with advanced non-small cell lung cancer. Lung Cancer. 2021 Jan;151:8-15.

Lymphoma

Diabetic people with diffuse large B-cell lymphoma treated with metformin showed longer progression-free survival after chemo-immunotherapy in one study. In a smaller study, metformin did not show evidence of an effect on response to treatment or survival after chemo-immunotherapy among diabetic people with follicular lymphoma.

People with diabetes

Weak evidenceweak trends of effects (this is the CancerChoices definition; other researchers and studies may define this differently) of better survival among people with diffuse large B-cell lymphoma with diabetes treated with metformin

Longer progression-free survivalthe time during and after treatment of a disease that a patient lives without disease progression (worsening) and a weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward longer survival after standard chemo-immunotherapy among people with diffuse large B-cell lymphoma with diabetes treated with metformin compared to no metformin among people either with or without diabetes in a mid-sized observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods study⁹¹Singh A, Gu J, Yanamadala V, Czuczman MS, Hernandez-Ilizaliturri FJ. Metformin lowers the mitochondrial potential of lymphoma cells and its use during front-line rituximab-based chemo-immunotherapy improves the clinical outcome of diffuse large B-cell lymphoma. Blood. 2013 Nov;122(21):1825-1825.

No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on response rate or survival among diabetic people with diffuse large B-cell lymphoma treated with metformin in a small study

No evidence of an effect on response rates, progression-free survival, or overall survival after bendamustine and rituximab therapy among people with follicular lymphoma and diabetes treated with metformin compared to no metformin among people either with or without diabetes in a small observational study;⁹²Hicks AM, Singh A et al. Therapeutic effects of metformin in follicular lymphoma (FL) treated with rituximab in combination with bendamustine. Blood. 2017 Dec;130(suppl1):5152–5152. only 20 people in this study were treated for diabetes, 12 with metformin and 8 with other glucose-lowering agents—very possibly too few people to allow a meaningful analysis of results

Melanoma

Diabetic people with melanoma treated with metformin had better cancer-specific survival in one study. However, nondiabetic people with melanoma who were treated with metformin during chemotherapy showed no evidence of an effect on response to treatment, time to progression, or survival in another study.

People with diabetes: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of higher cancer-specific survival among people with melanoma and diabetes treated with metformin

Higher melanoma-specific survival among people with melanoma and diabetes treated with metformin compared to nondiabetic people in a large observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods study⁹³Urbonas V, Rutenberge J et al. The impact of metformin on survival in patients with melanoma-national cohort study. Annals of Epidemiology. 2020 Dec;52:23-25.

People without diabetes: no evidence of an effect on objective response, time to progression, or overall survival among nondiabetic people with melanoma treated with metformin

No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on objective response, time to progression, or overall survival among nondiabetic people with melanoma treated with both dacarbazine and 850 mg oral metformin 2 times a day compared to dacarbazine alone in a small RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects⁹⁴Novik AV, Protsenko SA et al. Melatonin and metformin failed to modify the effect of dacarbazine in melanoma. Oncologist. 2021 May;26(5):364-e734.

Ovarian cancer

Diabetic people with ovarian cancer treated with metformin showed better survival across many studies.

Metformin during chemotherapy has shown insufficient evidenceconflicting findings across clinical studies with no preponderance of evidence in one direction; conflicting evidence occurs when studies find conflicting effects (positive effect vs no effect or negative effect) with the same treatment and the same general study population (same cancer type, for example) (this is the CancerChoices definition; other researchers and studies may define this differently) of an effect on progression-free survival among nondiabetic people with ovarian cancer across a few studies.

People with diabetes: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower mortality and higher progression-free survival among people with ovarian cancer and diabetes treated with metformin

29% lower overall mortality and substantially better progression-free survivalthe time during and after treatment of a disease that a person lives without disease progression (worsening) among women with ovarian cancer with diabetes treated with metformin compared to no metformin, whether among people with diabetes or not, in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 8 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies⁹⁵Lu MZ, Li DY, Wang XF. Effect of metformin use on the risk and prognosis of ovarian cancer: an updated systematic review and meta-analysis. Panminerva Medica. 2023 Sep;65(3):351-361.
28% lower mortality among people with ovarian cancer and diabetes treated with metformin compared to diabetic women who did not use metformin in a meta-analysis of 6 observational studies⁹⁶Guo M, Shang X, Guo D. Metformin use and mortality in women with ovarian cancer: an updated meta-analysis. International Journal of Clinical Practice. 2022 Feb 28;2022:9592969.
45% lower mortality among people with ovarian cancer and diabetes treated with metformin compared to other hypoglycemic drugs in a meta-analysis of 7 observational studies⁹⁷Shi J, Liu B, Wang H, Zhang T, Yang L. Association of metformin use with ovarian cancer incidence and prognosis: a systematic review and meta-analysis. International Journal of Gynecological Cancer. 2019 Jan;29(1):140-146.
Comparable overall survival among mostly diabetic people with ovarian cancer treated with metformin compared to no metformin (not specific to diabetes) in a meta-analysis of 6 observational studies;⁹⁸Wang Y, Liu X et al. No effect of metformin on ovarian cancer survival: a systematic review and meta-analysis of cohort studies. Current Pharmaceutical Design. 2019;25(23):2595-2601. our note: we expect worse survival among diabetic people
49% lower overall mortality and substantially higher progression-free survival among people with ovarian cancer with diabetes treated with metformin compared to no metformin after cancer diagnosis in a meta-analysis of 3 observational studies of high quality⁹⁹Gong TT, Wu QJ et al. Observational studies on the association between post-diagnostic metformin use and survival in ovarian cancer: a systematic review and meta-analysis. Frontiers in Oncology. 2019 May 29;9:458.
A weak trend toward longer overall survival among diabetic women with ovarian cancer treated with metformin compared to no metformin (not specific to diabetes) in a small observational study;¹⁰⁰Clayton L, Keene S et al. Assessing metformin use and ovarian cancer survival from electronic medical records. Cancer Epidemiology, Biomarkers & Prevention. 2016;25. our note: we expect worse survival among diabetic people
Substantially longer median overall survival among diabetic people with ovarian cancer treated with metformin compared to no metformin in a small observational study¹⁰¹Simonelli C, Bertolotti M et al. Effect of metformin on recurrence-free survival and overall survival in diabetic patients affected by advanced ovarian cancer. Journal of Clinical Oncology 2013;31.

People without diabetes: insufficient (conflicting) evidence of an effect on survival or progression among people with ovarian cancer (not specific to diabetes) treated with metformin

No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on progression-free survival among nondiabetic people with epithelial ovarian cancer treated with paclitaxel plus carboplatin plus metformin compared to paclitaxel plus carboplatin alone in a small RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects¹⁰²Zheng Y, Zhu J, Zhang H, Liu Y, Sun H. Metformin plus first-line chemotherapy versus chemotherapy alone in the treatment of epithelial ovarian cancer: a prospective open-label pilot trial. Cancer Chemotherapy and Pharmacology. 2019 Dec;84(6):1349-1357.
59.3% relapse-free survival at 18 months, a 2.4-fold decrease in cancer stem-like cells (ALDH+CD133+), and increased tumor cell sensitivity to cisplatin ex vivo after treatment with chemotherapy and debulking surgery among nondiabetic people with advanced-stage epithelial ovarian cancer treated with metformin—results better than historical controls—in a small uncontrolled triala study in which a therapy is used, but without a comparison group against which to judge outcomes; an uncontrolled trial is considered a weak study design¹⁰³Brown JR, Chan D et al. Phase II clinical trial of metformin as a cancer stem cell-targeting agent in ovarian cancer. JCI Insight. 2020 Jun 4;5(11):e133247.
A lower marker of tumor proliferation (Ki-67) among nondiabetic people with endometrial cancer treated with 1500-2250 mg metformin per day for 4 to 6 weeks before surgery compared to baseline in a small uncontrolled trial¹⁰⁴Mitsuhashi A, Kiyokawa T, Sato Y, Shozu M. Effects of metformin on endometrial cancer cell growth in vivo: a preoperative prospective trial. Cancer. 2014 Oct 1;120(19):2986-95.

Pancreatic cancer

Diabetic people with pancreatic cancer, but not locally advanced or metastasized cancer, treated with metformin showed better survival across many studies. Metformin did not show evidence of an effect on survival or response to treatment among nondiabetic people with pancreatic cancer across two studies.

People with diabetes: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower mortality among people with pancreatic cancer and diabetes treated with metformin

17% lower mortality among people with pancreatic cancer and diabetes, although not among people with advanced cancer or among people receiving only chemotherapy for cancer, treated with metformin compared to no metformin in a large meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 21 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies¹⁰⁵Shi YQ, Zhou XC et al. Relationships are between metformin use and survival in pancreatic cancer patients concurrent with diabetes: a systematic review and meta-analysis. Medicine (Baltimore). 2020 Sep 11;99(37):e21687.
16% lower mortality among people with pancreatic cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 10 observational studies¹⁰⁶Zhou PT, Li B et al. Metformin is associated with survival benefit in pancreatic cancer patients with diabetes: a systematic review and meta-analysis. Oncotarget. 2017 Apr 11;8(15):25242-25250.
12% lower overall mortality among people with pancreatic cancer (mostly with diabetes), although not among people with advanced cancer nor among people receiving only chemotherapy for cancer, treated with metformin compared to no metformin, in a large meta-analysis of 17 observational studies¹⁰⁷Wan G, Sun X et al. Survival benefit of metformin adjuvant treatment for pancreatic cancer patients: a systematic review and meta-analysis. Cellular Physiology and Biochemistry. 2018;49(3):837-847.
23% lower overall mortality among people with pancreatic cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 6 observational studies and 1 RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects¹⁰⁸Jian-Yu E, Graber JM et al. Effect of metformin and statin use on survival in pancreatic cancer patients: a systematic literature review and meta-analysis. Current Medicinal Chemistry. 2018;25(22):2595-2607.
14% lower mortality among people with pancreatic cancer, with stronger effects in 9 observational studies involving mostly people with diabetes compared to no evidence of an effect in 2 RCTs with mixed diabetic and nondiabetic patients, and stronger effects among people with resection or with locally advanced tumors but not among people with metastatic tumors, treated with metformin compared to no metformin in a meta-analysis¹⁰⁹Li X, Li T, Liu Z, Gou S, Wang C. The effect of metformin on survival of patients with pancreatic cancer: a meta-analysis. Scientific Reports. 2017 Jul 19;7(1):5825.
23% lower overall mortality among people with pancreatic cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 6 observational studies and 1 RCT¹¹⁰Jian-Yu E, Graber JM et al. Effect of metformin and statin use on survival in pancreatic cancer patients: a systematic literature review and meta-analysis. Current Medicinal Chemistry. 2018;25(22):2595-2607.

People without diabetes: no evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on overall or progression-free survival or on objective response rate among nondiabetic people with pancreatic cancer treated with metformin in a combined analysis of studies

No evidence of an effect on overall or progression-free survival or on objective response rate among nondiabetic people with pancreatic cancer treated with metformin compared to placebo in a meta-analysis of 2 RCTs¹¹¹Wu Z, Qu B et al. The potential adjunctive benefit of adding metformin to standard treatment in inoperable cancer patients: a meta-analysis of randomized controlled trials. Annals of Translational Medicine. 2020 Nov;8(21):1404.

Prostate cancer

Diabetic people with prostate cancer treated with metformin showed lower overall and cancer-specific mortality across many studies.

People with diabetes: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower mortality among people with prostate cancer and diabetes treated with metformin

26% lower cancer-specific mortality among diabetic people with prostate cancer treated with metformin compared to no metformin (not specific to diabetes), with greater overall survival mostly among people receiving radical radiotherapy and not among those treated with radical prostatectomy and androgen deprivation therapy in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 4 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies¹¹²Yao X, Liu H, Xu H. The impact of metformin use with survival outcomes in urologic cancers: a systematic review and meta-analysis. Biomed Research International. 2021 Oct 8;2021:5311828.
34% lower overall mortality (6 studies) and a weak trend toward lower cancer-specific mortality (6 studies specific to diabetes, 1 of which involved combined use of metformin and statins) among people with prostate cancer and diabetes treated with metformin compared to no metformin (not specific to diabetes) in a very large meta-analysis of 13 observational studies¹¹³Xiao Y, Zheng L et al. The impact of metformin use on survival in prostate cancer: a systematic review and meta-analysis. Oncotarget. 2017 Oct 31;8(59):100449-100458.
No evidence of an effect on overall mortality among people with prostate cancer (mostly with diabetes) treated with metformin compared to no metformin (not specific to diabetes) in a very large meta-analysis of 5 observational studies¹¹⁴Yu H, Yin L et al. Effect of metformin on cancer risk and treatment outcome of prostate cancer: a meta-analysis of epidemiological observational studies. PLoS One. 2014;9(12):e116327.
A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower overall mortality among people with prostate cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 4 observational studies¹¹⁵Lega IC, Shah PS et al. The effect of metformin on mortality following cancer among patients with diabetes. Cancer Epidemiology, Biomarkers & Prevention. 2014 Oct;23(10):1974-84.
19% lower cancer-specific mortality among diabetic people with prostate cancer treated with metformin compared to no metformin in a meta-analysis of 9 observational studies¹¹⁶Raval AD, Thakker D et al. Impact of metformin on clinical outcomes among men with prostate cancer: a systematic review and meta-analysis. Prostate Cancer and Prostatic Diseases. 2015 Jun;18(2):110-21.
Lower overall mortality but not cancer-specific mortality among diabetic people with prostate cancer treated with metformin compared to no metformin in a meta-analysis of 9 observational studies, with a smaller effect seen in studies with lower risk of immortal time biasa study design confound that leads to a biased association; ‘immortal time’ is when participants of a cohort study cannot experience the outcome during some period of follow-up time¹¹⁷Stopsack KH, Ziehr DR, Rider JR, Giovannucci EL. Metformin and prostate cancer mortality: a meta-analysis. Cancer Causes and Control. 2016 Jan;27(1):105-13.
42% lower cancer-specific mortality among people with early stage prostate cancer and diabetes treated with metformin compared to no metformin (not specific to diabetes), with greater effects among people receiving radical radiotherapy, in a meta-analysis of 3 observational studies¹¹⁸Coyle C, Cafferty FH, Vale C, Langley RE. Metformin as an adjuvant treatment for cancer: a systematic review and meta-analysis. Annals of Oncology. 2016 Dec;27(12):2184-2195.
No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on cancer-specific mortality among diabetic people with prostate cancer treated with metformin compared to no metformin (not specific to cancer) in a meta-analysis of 3 observational studies¹¹⁹He K, Hu H et al. The effect of metformin therapy on incidence and prognosis in prostate cancer: a systematic review and meta-analysis. Scientific Reports. 2019 Feb 18;9(1):2218.
No evidence of an effect on cancer-specific mortality (4 studies) and a weak trend toward lower risk of metastases (3 studies) among people with prostate cancer (mostly with diabetes) treated with metformin compared to no metformin (not specific to cancer) in meta-analyses of observational studies¹²⁰Raval AD, Thakker D et al. Impact of metformin on clinical outcomes among men with prostate cancer: a systematic review and meta-analysis. Prostate Cancer and Prostatic Diseases. 2015 Jun;18(2):110-21.

Metformin combined with other therapies

One person combining metformin with Zyflamend™, a supplement marketed for inflammation, showed only a weak trend toward lower prostate-specific antigen (PSA) levels in a case study.

People with colorectal cancer treated with metformin and chemotherapy showed a weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward better disease outcomes in two small studies.

Zyflamend™ with metformin: weak evidenceone or more case studies (this is the CancerChoices definition; other researchers and studies may define this differently) of lower PSA levels in a man with recurrent castration-resistant prostate cancer and increasing PSA levels treated with conventional care and metformin

Dramatically lower PSA levels in a man with recurrent castration-resistant prostate cancer and increasing PSA levels after radical prostatectomy, salvage radiation therapy, and bilateral orchiectomy treated with Zyflamend and metformin for 6 months in a case studya descriptive and exploratory analysis of a person, group, or event regarding changes observed over time; because changes due to treatment are not compared to similar changes over time without treatment, a case study is considered a weak study design¹²¹Bilen MA, Lin SH et al. Maintenance therapy containing metformin and/or Zyflamend for advanced prostate cancer: a case series. Case Reports in Oncological Medicine. 2015;2015:471861.

Metformin with chemotherapy: weak evidence of disease control among people with metastatic colorectal cancer previously treated with chemotherapy subsequently treated with both chemotherapy and metformin

Disease control at 8 weeks among 22% of people with progressing metastatic colorectal cancer previously treated with 5-fluorouracil (5-FU), irinotecan, oxaliplatin, and an anti-epidermal growth factor receptor (if the tumor was RAS wild type) subsequently treated with 850 mg oral metformin continuously 2 times a day plus 5-FU 425 mg/m2 and leucovorin 50 mg intravenously weekly in a small uncontrolled triala study in which a therapy is used, but without a comparison group against which to judge outcomes; an uncontrolled trial is considered a weak study design¹²²Miranda VC, Braghiroli MI et al. Phase 2 trial of metformin combined with 5-fluorouracil in patients with refractory metastatic colorectal cancer. Clinical Colorectal Cancer. 2016 Dec;15(4):321-328.e1.
Disease control and stable disease at week 12 in more than half of people with metastatic colorectal cancer who had previously progressed on fluropyrimidine, oxaliplatin, irinotecan, and anti-EGFR antibody (if KRAS wild type) subsequently treated with irinotecan and metformin in a small uncontrolled trial¹²³Bragagnoli A, Araujo R et al. Final results of a phase II of metformin plus irinotecan for refractory colorectal cancer. Journal of Clinical Oncology. 2018;36(15_suppl):e15527-e15527.

Optimizing your body terrain

Does metformin promote an environment within your body that is less supportive of cancer development, growth, or spread? We present the evidence.

See Optimizing Your Body Terrain ›

Find medical professionals who specialize in managing body terrain factors: Finding Integrative Oncologists and Other Practitioners ›

Body weight

Nondiabetic people with cancer treated with metformin showed more weight loss across several studies. People with prostate cancer and normal blood sugar showed lower weight gain in the short term but not the long term during androgen deprivation therapy—which often results in weight gain—in one study.

People without cancer who had non-diabetes metabolic imbalances, including polycystic ovary syndrome and impaired glucose tolerance, showed more weight loss when treated with metformin across many studies.

People with cancer but without diabetes: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower body weight among nondiabetic people with cancer treated with metformin

Lower body mass index among people with breast cancer treated with metformin, with larger effects when metformin was used for more than 4 weeks, compared to no metformin (not specific to diabetes) in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 9 RCTsrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects¹²⁴Rahmani J, Manzari N et al. The effect of metformin on biomarkers associated with breast cancer outcomes: a systematic review, meta-analysis, and dose–response of randomized clinical trials. Clinical & Translational Oncology. 2020 Jan;22(1):37-49.
No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on body weight among mostly nondiabetic women with atypical hyperplasia or endometrioid endometrial cancer treated with metformin for 1 to 5 weeks before surgery compared to placebo in a small RCT¹²⁵Kitson SJ, Maskell Z et al. Pre-surgical metformin in uterine malignancy (PREMIUM): a multi-center, randomized double-blind, placebo-controlled phase III trial. Clinical Cancer Research: An Official Journal of the American Association for Cancer Research. 2019 Apr 15;25(8):2424-2432.
Greater reductions in body weight and waist circumference during tamoxifen treatment among nondiabetic postmenopausal women with hormone receptor-positive breast cancer treated with 850 mg metformin twice a day for 52 weeks compared to placebo in a mid-sized RCT¹²⁶Davis SR, Robinson PJ et al. The benefits of adding metformin to tamoxifen to protect the endometrium—a randomized placebo-controlled trial. Clinical Endocrinology (Oxford). 2018 Nov;89(5):605-612.
More weight loss among nondiabetic people with breast or colorectal cancer treated with 850 mg metformin twice a day for 12 weeks compared to controls in a mid-sized RCT¹²⁷Meyerhardt JA, Irwin ML et al. Multicenter, randomized phase II trial of physical activity (PA), metformin (Met), or the combination on metabolic biomarkers in stage I-III colorectal (CRC) and breast cancer (BC) survivors. Journal of Clinical Oncology. 2017 May;35(15):10059-10059.
A greater decrease in body mass index after surgery and (neo)adjuvant chemotherapy (if given) among nondiabetic people with breast cancer treated with 850 mg oral metformin twice a day compared to placebo for 6 months in a mid-sized RCT¹²⁸Goodwin PJ, Parulekar WR et al. Effect of metformin vs placebo on weight and metabolic factors in NCIC CTG MA.32. Journal of the National Cancer Institute. 2015 Mar 4;107(3):djv006.
A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower body weight among nondiabetic people with breast cancer treated with metformin for four weeks after biopsy in a mid-sized RCT¹²⁹Bonanni B, Puntoni M et al. Dual effect of metformin on breast cancer proliferation in a randomized presurgical trial. Journal of Clinical Oncology. 2012 Jul 20;30(21):2593-600.

Lower gains in body weight at 6 months, but no evidence of an effect at 1 year or any effect on waist circumference, radical radiotherapy and androgen deprivation therapy among men with normal glycemia and locally advanced prostate cancer treated with 500 mg oral metformin 3 times a day compared to placebo in a small RCT¹³⁰Usmani N, Ghosh S et al. Metformin for prevention of anthropometric & metabolic complications of androgen deprivation therapy in prostate cancer patients receiving radical radiotherapy: a phase II randomized controlled trial. International Journal of Radiation Oncology, Biology, Physics. 2022 Jul 27:S0360-3016(22)00747-7.

People without cancer with metabolic imbalances besides diabetes: modest evidence of lower body weight and waist circumference among people with metabolic imbalances other than diabetes treated with metformin

Lower body mass index (BMI) among nondiabetic overweight/obese premenopausal women treated with lifestyle interventions for diet and exercise plus metformin compared to lifestyle interventions alone for a year in a mid-sized RCT¹³¹Leng W, Pu D et al. Effect of metformin on breast density in overweight/obese premenopausal women. Diabetes, Metabolic Syndrome and Obesity: Targets and Therapy. 2021 Nov 3;14:4423-4432.
A decrease in waist circumference and waist-to-hip ratio, and a weak trend toward lower non-dense breast volume, among overweight/obese premenopausal women with components of metabolic syndrome treated with 850 mg metformin twice a day compared to placebo for 12 months in a mid-sized RCT¹³²Tapia E, Villa-Guillen DE et al. A randomized controlled trial of metformin in women with components of metabolic syndrome: intervention feasibility and effects on adiposity and breast density. Breast Cancer Research and Treatment. 2021 Nov;190(1):69-78.
Decreases in body weight at 6 months among overweight or obese people with cancer treated with metformin compared to self-directed weight loss in a mid-sized RCT¹³³Yeh HC, Maruthur NM et al. Effects of behavioral weight loss and metformin on igfs in cancer survivors: a randomized trial. Journal of Clinical Endocrinology and Metabolism. 2021 Sep 27;106(10):e4179-e4191.
Lower body weight among people with polycystic ovary syndrome (not specific to diabetes) treated with 1000 mg metformin per day for at least 25.5 weeks compared to controls in a meta-analysis of 28 RCTs¹³⁴Chen X, He S, Wang D. Effects of metformin on body weight in polycystic ovary syndrome patients: model-based meta-analysis. Expert Reviews in Clinical Pharmacology. 2021 Jan;14(1):121-130.
A weak trend toward more weight loss among obese nondiabetic people with breast cancer treated with metformin for 6 months compared to placebo, with stronger effects at 1000 mg than at 500 mg, but no evidence of an effect among non-obese people, in a mid-sized RCT¹³⁵Ko KP, Ma SH et al. Metformin intervention in obese non-diabetic patients with breast cancer: phase II randomized, double-blind, placebo-controlled trial. Breast Cancer Research and Treatment. 2015 Sep;153(2):361-70.
Lower body weight (BMI) and waist circumference at 6 months among people with impaired glucose tolerance treated with 1500 mg metformin per day compared to baseline, but no decrease among people treated with 500 mg, 250 mg, or no metformin in a mid-sized RCT¹³⁶Zhao X, Li Y et al. Effects of different doses of metformin treatment for 6 months on aberrant crypt foci in Chinese patients with impaired glucose tolerance. European Journal of Cancer Prevention. 2015 Jan;24(1):27-36.
More weight loss, proportional to the dose of metformin, among nondiabetic people with breast cancer and high serum testosterone levels treated with escalating doses of metformin for several months in a mid-sized controlled trial¹³⁷Campagnoli C, Pasanisi P et al. Effect of different doses of metformin on serum testosterone and insulin in non-diabetic women with breast cancer: a randomized study. Clinical Breast Cancer. 2012 Jun;12(3):175-82.

Preliminary evidence of short-term lower gains in body weight but no evidence of an effect on waist circumference during radical radiotherapy and androgen deprivation therapy among men with normal glycemia and locally advanced prostate cancer treated with metformin; people on androgen deprivation therapy often gain weight

Lower gains in body weight at 6 month, but no evidence of an effect at 1 year or any effect on waist circumference during radical radiotherapy and androgen deprivation therapy among men with normal glycemia and locally advanced prostate cancer treated with 500 mg oral metformin 3 times a day compared to placebo in a small RCT¹³⁸Usmani N, Ghosh S et al. Metformin for prevention of anthropometric & metabolic complications of androgen deprivation therapy in prostate cancer patients receiving radical radiotherapy: a phase II randomized controlled trial. International Journal of Radiation Oncology, Biology, Physics. 2022 Jul 27:S0360-3016(22)00747-7.

High blood sugar and insulin resistance

The FDA has approved metformin for reducing high blood sugar in people with type 2 diabetes, with substantial evidence of benefit for this condition. Use in people who don’t have type 2 diabetes is considered off-label use. Here we summarize only the evidence for metformin’s effects on blood glucose and insulin among people without diabetes.

Nondiabetic people with blood cancer treated with metformin showed lower insulin after eating in one study.

Nondiabetic people with breast or endometrial cancer treated with metformin showed lower blood glucose and insulin across many studies, with fewer studies showing lower insulin resistance among nondiabetic people with breast or colorectal cancer.

People with non-diabetes metabolic imbalances, such as insulin resistance, treated with metformin showed lower blood glucose, insulin, and insulin resistance across several studies.

People without cancer or diabetes treated with metformin showed lower insulin across a couple of studies, plus lower insulin resistance or fasting blood sugar in single studies.

People with blood cancers: preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of lower blood glucose after a meal but not while fasting among people with blood cancers without diabetes or prediabetes at baseline treated with metformin

A lower rate of steroid-induced high blood sugar after a meal (2-hour postprandial hyperglycemia) during high-dose prednisone treatment, but no evidence of an effect on fasting glucose, among people with blood cancers without diabetes or prediabetes at baseline treated with 850 mg metformin twice daily compared to standard care in a small RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects¹³⁹Ochola LA, Nyamu DG, Guantai EM, Weru IW. Metformin’s effectiveness in preventing prednisone-induced hyperglycemia in hematological cancers. Journal of Oncology Pharmacy Practice. 2020 Jun;26(4):823-834.

People with breast or endometrial cancers

People without diabetes: strong evidenceconsistent, significant effects in several large (or at least one very large) well designed clinical studies or at least two meta-analyses of clinical studies of moderate or better quality (or one large meta-analysis) finding similar results (this is the CancerChoices definition; other researchers and studies may define this differently) of lower levels of blood sugar, insulin, and insulin resistance among nondiabetic people with breast or endometrial cancer treated with metformin

Lower fasting blood sugar (6 studies) and a weak trend toward lower insulin levels (6 studies), but no evidence of an effect on insulin resistance (HOMA-IR, 4 studies) among nondiabetic women with operable breast or endometrial cancer or primary breast cancer treated with metformin compared to no metformin in meta-analysesa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of RCTs¹⁴⁰Farkhondeh T, Amirabadizadeh A et al. Impact of metformin on cancer biomarkers in non-diabetic cancer patients: a systematic review and meta-analysis of clinical trials. Current Oncology. 2021 Apr 6;28(2):1412-1423.
No evidence of an effect on insulin signaling pathways among mostly nondiabetic women with atypical hyperplasia or endometrioid endometrial cancer treated with metformin for 1 to 5 weeks before surgery compared to placebo in a small RCT¹⁴¹Kitson SJ, Maskell Z et al. Pre-surgical metformin in uterine malignancy (PREMIUM): a multi-center, randomized double-blind, placebo-controlled phase III trial. Clinical Cancer Research: An Official Journal of the American Association for Cancer Research. 2019 Apr 15;25(8):2424-2432.
Better markers of blood glucose and insulin resistance (HOMA-IR) among people with breast cancer, mostly without diabetes, treated with metformin compared to controls in a meta-analysis of 10 RCTs¹⁴²Zhang ZJ, Yuan J, Bi Y, Wang C, Liu Y. The effect of metformin on biomarkers and survivals for breast cancer—a systematic review and meta-analysis of randomized clinical trials. Pharmacological Research. 2019 Mar;141:551-555.
Lower insulin resistance among nondiabetic people with breast cancer treated with metformin during chemotherapy compared to chemotherapy alone in a small RCT¹⁴³Davis SR, Robinson PJ et al. The benefits of adding metformin to tamoxifen to protect the endometrium—a randomized placebo-controlled trial. Clinical Endocrinology (Oxford). 2018 Nov;89(5):605-612.
Lower insulin levels among nondiabetic people with breast or colorectal cancer treated with 850 mg metformin twice a day for 12 weeks compared to controls in a mid-sized RCT¹⁴⁴Meyerhardt JA, Irwin ML et al. Multicenter, randomized phase II trial of physical activity (PA), metformin (Met), or the combination on metabolic biomarkers in stage I-III colorectal (CRC) and breast cancer (BC) survivors. Journal of Clinical Oncology. 2017 May;35(15):10059-10059.

People with colorectal cancer: preliminary evidence of lower insulin levels among nondiabetic people with colorectal cancer treated with metformin

Lower insulin levels among nondiabetic people with breast or colorectal cancer treated with 850 mg metformin twice a day for 12 weeks compared to controls in a mid-sized RCT¹⁴⁵Meyerhardt JA, Irwin ML et al. Multicenter, randomized phase II trial of physical activity (PA), metformin (Met), or the combination on metabolic biomarkers in stage I-III colorectal (CRC) and breast cancer (BC) survivors. Journal of Clinical Oncology. 2017 May;35(15):10059-10059.

People with prostate cancer: no evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on a marker of blood sugar (glycemic) control among men with normal glycemia and locally advanced prostate cancer treated with metformin in a small study; glycemic control is often worse during ADT, but we would not expect much of an effect on glycemic control among people with normal glycemia

No evidence of an effect on a marker of glycemic control (HbA1c) during radical radiotherapy and androgen deprivation therapy among men with normal glycemia and locally advanced prostate cancer treated with 500 mg oral metformin 3 times a day compared to placebo in a small RCT¹⁴⁶Usmani N, Ghosh S et al. Metformin for prevention of anthropometric & metabolic complications of androgen deprivation therapy in prostate cancer patients receiving radical radiotherapy: a phase II randomized controlled trial. International Journal of Radiation Oncology, Biology, Physics. 2022 Jul 27:S0360-3016(22)00747-7.

People without cancer but with metabolic imbalances: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower levels of insulin, insulin resistance, and/or blood sugar among overweight or obese people or people with impaired glucose tolerance or insulin resistance treated with metformin

Lower markers of insulin resistance (HOMA-IR levels) among people with thyroid nodules and insulin resistance treated with metformin compared to no metformin in a review of 4 clinical studies of low quality¹⁴⁷He X, Wu D et al. Role of metformin in the treatment of patients with thyroid nodules and insulin resistance: a systematic review and meta-analysis. Thyroid. 2019 Mar;29(3):359-367.
Lower fasting plasma glucose, insulin resistance (HOMA-IR index), and 2-hour plasma glucose levels at 6 months among people with impaired glucose tolerance treated with 1500 mg metformin per day compared to baseline, but no decrease among people treated with 500 mg, 250 mg, or no metformin in a mid-sized RCT¹⁴⁸Zhao X, Li Y et al. Effects of different doses of metformin treatment for 6 months on aberrant crypt foci in Chinese patients with impaired glucose tolerance. European Journal of Cancer Prevention. 2015 Jan;24(1):27-36.

People without cancer or metabolic imbalances: preliminary evidence of lower levels of insulin or insulin resistance among people without cancer or diabetes treated with metformin

Lower fasting insulin, fasting plasma glucose, and HOMA-IR among nondiabetic overweight or obese premenopausal women treated with lifestyle interventions for diet and exercise plus metformin compared to lifestyle interventions alone for a year in a mid-sized RCT¹⁴⁹Leng W, Pu D et al. Effect of metformin on breast density in overweight/obese premenopausal women. Diabetes, Metabolic Syndrome and Obesity: Targets and Therapy. 2021 Nov 3;14:4423-4432.
A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower levels of insulin and insulin resistance (HOMA-IR) among people with Barrett’s esophagus and no history of diabetes treated with increasing doses of metformin to 2000 mg/day by week 4 compared to placebo in a small RCT¹⁵⁰Chak A, Buttar NS et al; Cancer Prevention Network. Metformin does not reduce markers of cell proliferation in esophageal tissues of patients with Barrett’s esophagus. Clinical Gastroenterology and Hepatology. 2015 Apr;13(4):665-72.e1-4.

Hormone imbalance

Changes in hormone levels seen in the studies here may not be beneficial in every situation. Your oncology team needs to determine whether any changes would be favorable for your condition.

Sex hormones: Nondiabetic people with breast cancer treated with metformin showed lower levels of many sex hormones.

Metabolic hormones: People with cancer or major depressive disorder treated with metformin showed lower levels of insulin-like growth factor 1 (IGF-1) across several studies. Nondiabetic people with cancer treated with metformin showed lower levels of leptina hormone that helps regulate energy balance by inhibiting hunger across several studies. People with polycystic ovary syndrome treated with metformin showed higher levels of adiponectina hormone and signaling protein involved in regulating glucose levels and fatty acid breakdown across several studies. Nondiabetic people with thyroid nodules and insulin resistance showed lower levels of thyroid-stimulating hormone when treated with metformin in one study.

Sex hormones

Breast cancer: strong evidenceconsistent, significant effects in several large (or at least one very large) well designed clinical studies or at least two meta-analyses of clinical studies of moderate or better quality (or one large meta-analysis) finding similar results (this is the CancerChoices definition; other researchers and studies may define this differently) of lower levels of many sex hormones among mostly nondiabetic people with breast cancer treated with metformin

Higher circulating levels of both estrogens and testosterone have been linked to higher risk of breast cancer among postmenopausal women.¹⁵¹Hankinson SE, Eliassen AH. Endogenous estrogen, testosterone and progesterone levels in relation to breast cancer risk. Journal of Steroid Biochemistry and Molecular Biology. 2007 Aug-Sep;106(1-5):24-30.

A greater decrease in estradiol but no evidence of an effect on sex hormone-binding globulin or bioavailable testosterone among nondiabetic postmenopausal women with hormone receptor-negative breast cancer not receiving endocrine treatment treated with metformin compared to placebo in a mid-sized RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects¹⁵²Pimentel I, Chen BE et al. The effect of metformin vs placebo on sex hormones in Canadian Cancer Trials Group MA.32. Journal of the National Cancer Institute. 2021 Feb 1;113(2):192-198.
Lower levels of sex hormones (testosterone, bioavailable testosterone, free testosterone, free androgen index, estradiol, and bioavailable estradiol) and sex hormone-binding globulin among nondiabetic people with breast cancer treated with metformin compared to controls in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 10 RCTs¹⁵³Zhang ZJ, Yuan J, Bi Y, Wang C, Liu Y. The effect of metformin on biomarkers and survivals for breast cancer—a systematic review and meta-analysis of randomized clinical trials. Pharmacological Research. 2019 Mar;141:551-555.
Greater reduction of free testosterone and estradiol and a weak trend toward a reduction of estrone, but no evidence of an effect on androstenedione among nondiabetic people with breast cancer treated with 1500 mg metformin compared to 1000 mg per day for 5 months in a small RCT¹⁵⁴Campagnoli C, Berrino F et al. Metformin decreases circulating androgen and estrogen levels in nondiabetic women with breast cancer. Clinical Breast Cancer. 2013 Dec;13(6):433-8.
Lower testosterone level and free androgen index among nondiabetic people with breast cancer treated with 500 mg metformin 3 times a day compared to 2 times a day for 5 months in a small RCT¹⁵⁵Campagnoli C, Pasanisi P et al. Effect of different doses of metformin on serum testosterone and insulin in non-diabetic women with breast cancer: a randomized study. Clinical Breast Cancer. 2012 Jun;12(3):175-82.

Endometrial hyperplasia: preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of higher rates of cessation of menstrual periods (amenorrhea) during treatment with a levonorgestrel intrauterine system among nondiabetic people with endometrial hyperplasia without atypia treated with metformin

More cessation of menstruation (amenorrhea) during treatment with a levonorgestrel intrauterine system among nondiabetic people with endometrial hyperplasia without atypia and 500 mg metformin twice daily compared to the levonorgestrel intrauterine system alone in a small RCT¹⁵⁶Ravi RD, Kalra et al. A randomized clinical trial of levonorgestrel intrauterine system with or without metformin for treatment of endometrial hyperplasia without atypia in Indian women. Asian Pacific Journal of Cancer Prevention. 2021 Mar 1;22(3):983-989.

Metabolic hormones

Modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower markers of insulin-like growth factors among people with cancer treated with metformin

Lower markers of insulin-like growth factors (IGF-1 and IGF-1:IGFBP3 molar ratio) at 3 months among both overweight and obese people, at 6 months only among obese people, but no evidence of an effect at 12 months for anyone among people with cancer, not specific to diabetes, treated with metformin compared to self-directed weight loss in a mid-sized RCT¹⁵⁷Yeh HC, Maruthur NM et al. Effects of behavioral weight loss and metformin on IGFs in cancer survivors: a randomized trial. Journal of Clinical Endocrinology and Metabolism. 2021 Sep 27;106(10):e4179-e4191.
Smaller increase of IGF-1 and maintained IGF binding protein (IGFBP-1) during treatment with paclitaxel plus carboplatin among people with epithelial ovarian cancer without diabetes treated with metformin compared to chemotherapy alone in a small RCT¹⁵⁸Zheng Y, Zhu J, Zhang H, Liu Y, Sun H. Metformin plus first-line chemotherapy versus chemotherapy alone in the treatment of epithelial ovarian cancer: a prospective open-label pilot trial. Cancer Chemotherapy and Pharmacology. 2019 Dec;84(6):1349-1357.
Lower plasma concentrations of IGF-1 among women with endometrial cancer treated with metformin compared to no metformin in a small controlled trial¹⁵⁹Cai D, Sun H et al. Insulin-like growth factor 1/mammalian target of rapamycin and amp-activated protein kinase signaling involved in the effects of metformin in the human endometrial cancer. International Journal of Gynecological Cancer. 2016 Nov;26(9):1667-1672.
Lower IGF-1 levels among people with endometrial cancer (not specific to diabetes) treated with 1500-2250 mg metformin per day for 4 to 6 weeks before surgery compared to baseline in a small uncontrolled triala study in which a therapy is used, but without a comparison group against which to judge outcomes; an uncontrolled trial is considered a weak study design¹⁶⁰Mitsuhashi A, Kiyokawa T, Sato Y, Shozu M. Effects of metformin on endometrial cancer cell growth in vivo: a preoperative prospective trial. Cancer. 2014 Oct 1;120(19):2986-95.

Preliminary evidence of lower IGF-1 levels among nondiabetic people with major depressive disorder treated with metformin

Lower IGF-1 levels among nondiabetic people with major depressive disorder treated with 20 mg fluoxetine once daily plus 1000 mg metformin once daily for 12 weeks compared to fluoxetine alone in a small RCT¹⁶¹Abdallah MS, Mosalam EM et al. The antidiabetic metformin as an adjunct to antidepressants in patients with major depressive disorder: a proof-of-concept, randomized, double-blind, placebo-controlled trial. Neurotherapeutics. 2020 Oct;17(4):1897-1906.

Modest evidence of higher levels of adiponectina hormone and signaling protein involved in regulating glucose levels and fatty acid breakdown—protective against insulin resistance/diabetes and atherosclerosis—among women with polycystic ovary syndrome (not specific to cancer) treated with metformin

Higher levels of adiponectin among women with polycystic ovary syndrome (not specific to diabetes) treated with metformin compared to no metformin in a meta-analysis of 11 controlled trialsa study design in which people are assigned to either an experimental group or a control group to compare the outcomes from different treatment; assignment is not random, and so this is not as strong a study design as a randomized controlled trial, but still stronger than an uncontrolled trial¹⁶²Duan X, Zhou M, Zhou G, Zhu Q, Li W. Effect of metformin on adiponectin in PCOS: a meta-analysis and a systematic review. European Journal of Obstetrics, Gynecology, and Reproductive Biology. 2021 Dec;267:61-67.

Modest evidence of lower levels of leptina hormone that helps regulate energy balance by inhibiting hunger among nondiabetic people with cancer treated with metformin

Under normal conditions, leptin increases after eating to signal to the body that hunger has been satisfied and suppress appetite. However, in a condition called leptin resistance, even though the leptin levels rise, the brain doesn’t receive or respond to its signals to suppress appetite. The brain senses that the body is still hungry, leading to a cycle of overeating and continuing high levels of leptin, which is linked to obesity. Chronic high leptin levels may affect factors that promote cancer, such as estrogen signaling in breast cancer.¹⁶³Leptin and leptin resistance: everything you need to know. Healthline. Viewed January 17, 2025; You and Your Hormones. Leptin. Society for Endocrinology. March 2018. Viewed January 17, 2025.

A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower leptin levels among nondiabetic people with breast or colorectal cancer treated with 850 mg metformin twice a day for 12 weeks compared to controls in a mid-sized RCT¹⁶⁴Meyerhardt JA, Irwin ML et al. Multicenter, randomized phase II trial of physical activity (PA), metformin (Met), or the combination on metabolic biomarkers in stage I-III colorectal (CRC) and breast cancer (BC) survivors. Journal of Clinical Oncology. 2017 May;35(15):10059-10059.
A greater decrease in leptin after surgery and (neo)adjuvant chemotherapy (if given) among nondiabetic people with breast cancer treated with 850 mg oral metformin twice a day compared to placebo for 6 months in a mid-sized RCT¹⁶⁵Goodwin PJ, Parulekar WR et al. Effect of metformin vs placebo on weight and metabolic factors in NCIC CTG MA.32. Journal of the National Cancer Institute. 2015 Mar 4;107(3):djv006.
Lower leptin levels among nondiabetic people with endometrial cancer treated with 1500-2250 mg metformin per day for 4 to 6 weeks before surgery compared to baseline in a small uncontrolled trial¹⁶⁶Mitsuhashi A, Kiyokawa T, Sato Y, Shozu M. Effects of metformin on endometrial cancer cell growth in vivo: a preoperative prospective trial. Cancer. 2014 Oct 1;120(19):2986-95.
Lower levels of leptin among nondiabetic, obese people with breast cancer treated with 1,500 mg metformin a day for 2 weeks before surgery compared to no metformin in a small RCT¹⁶⁷Kalinsky K, Crew KD et al. Presurgical trial of metformin in overweight and obese patients with newly diagnosed breast cancer. Cancer Invest. 2014 May;32(4):150-7.

Preliminary evidence of lower levels of thyroid stimulating hormone among people with thyroid nodules and insulin resistance treated with metformin

Lower levels of thyroid stimulating hormone (TSH) among people with thyroid nodules and insulin resistance treated with metformin compared to no metformin in a review of 4 clinical studies of low quality¹⁶⁸He X, Wu D et al. Role of metformin in the treatment of patients with thyroid nodules and insulin resistance: a systematic review and meta-analysis. Thyroid. 2019 Mar;29(3):359-367.

Immune function

Increased immune system activation is not always beneficial, so your oncology team needs to determine whether immune activation would be favorable in your situation.

Nondiabetic people with head and neck cancer treated with metformin before surgery showed higher markers of immune function in one poorly designed study. In another poorly designed study, people with esophageal cancer, whether with or without diabetes, treated with metformin before surgery also showed higher markers of immune activation.

Inflammation

Nondiabetic people with breast cancer treated with metformin showed lower levels of a marker of inflammation (C-reactive protein) across several studies. People with colorectal cancer (not specific to diabetes) treated with metformin showed lower markers of inflammation in one study.

People with polycystic ovary syndrome— a condition marked by metabolic imbalance — treated with metformin showed lower markers of inflammation (TNF-α and CRP) across many studies. Nondiabetic people with major depressive disorder treated with metformin showed lower markers of inflammation in one study.

People with cancer and no diabetes: good evidencesignificant effects in one large or several mid-sized and well-designed clinical studies (randomized controlled trials (RCTs) with an appropriate placebo or other strong comparison control or observational studies that control for confounds) (this is the CancerChoices definition; other researchers and studies may define this differently) of lower or more balanced markers of inflammation among nondiabetic people with breast cancer treated with metformin

Lower levels of a marker of inflammation (C-reactive protein, or CRP), among nondiabetic people with breast cancer treated with 1,500 mg metformin daily compared to no metformin in a small RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects¹⁷¹Sadighi S, Saberian M et al. Metformin anti-proliferative effect on a cohort of non-diabetic breast cancer patients. ESMO Annals of Oncology. 2016 Oct;27(60):vi43–vi67.
A greater decrease in CRP after surgery and (neo)adjuvant chemotherapy (if given) among nondiabetic people with breast cancer treated with 850 mg oral metformin twice a day compared to placebo for 6 months in a mid-sized RCT¹⁷²Goodwin PJ, Parulekar WR et al. Effect of metformin vs placebo on weight and metabolic factors in NCIC CTG MA.32. Journal of the National Cancer Institute. 2015 Mar 4;107(3):djv006.
Lower levels of CRP among nondiabetic people with breast cancer treated with 850 mg metformin twice per day compared to placebo in a mid-sized RCT¹⁷³Bonanni B, Puntoni M et al. Dual effect of metformin on breast cancer proliferation in a randomized presurgical trial. Journal of Clinical Oncology. 2012 Jul 20;30(21):2593-600.

People with cancer and not specific to diabetes: preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of lower markers of inflammation among people with colorectal cancer (not specific to diabetes) treated with metformin

Lower levels of some markers of inflammation among people with stage 1–3 breast or colorectal cancer (not specific to diabetes) treated with metformin compared to controls in a mid-sized RCT¹⁷⁴Brown JC, Zhang S et al. Effect of exercise or metformin on biomarkers of inflammation in breast and colorectal cancer: a randomized trial. Cancer Prevention Research (Phila). 2020 Dec;13(12):1055-1062.

People without cancer with non-diabetes metabolic imbalances: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower markers of inflammation among people with polycystic ovary syndrome—a condition marked by metabolic imbalance—treated with metformin

Lower levels of markers of inflammation (TNF-α and CRP) among women with polycystic ovary syndrome treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 11 controlled trialsa study design in which people are assigned to either an experimental group or a control group to compare the outcomes from different treatment; assignment is not random, and so this is not as strong a study design as a randomized controlled trial, but still stronger than an uncontrolled trial¹⁷⁵Duan X, Zhou M, Zhou G, Zhu Q, Li W. Effect of metformin on adiponectin in PCOS: a meta-analysis and a systematic review. European Journal of Obstet Gynecol Reprod Biol. 2021 Dec;267:61-67.

People without cancer or diabetes: preliminary evidence of lower markers of inflammation among nondiabetic people with major depressive disorder treated with metformin

Lower markers of inflammation during treatment with 20 mg fluoxetine among nondiabetic people with major depressive disorder treated with 1000 mg metformin a day for 12 weeks compared to fluoxetine alone in a small RCT¹⁷⁶Abdallah MS, Mosalam EM et al. The antidiabetic metformin as an adjunct to antidepressants in patients with major depressive disorder: a proof-of-concept, randomized, double-blind, placebo-controlled trial. Neurotherapeutics. 2020 Oct;17(4):1897-1906.

Oxidative stress

Nondiabetic people with colorectal cancer treated with metformin during chemotherapy showed lower markers of oxidative stress in one study. Nondiabetic people with major depressive disorder treated with metformin also showed a lower marker of oxidative stress in one study.

Your microbiome

Nondiabetic, obese people with solid tumors treated with metformin showed changes in the gut microbiomethe collection of microbes living on and within your body, including beneficial increases in butyrate, a fatty acid that promotes gut health, in one study.

Metformin combined with other therapies or practices

Nondiabetic people with cancer combining metformin with exercise had more weight loss and lower insulin resistance and leptina hormone that helps regulate energy balance by inhibiting hunger in one study, and lower markers of inflammation in another study. People with insulin resistance combining metformin and exercise, sometimes also with diet, experienced weight loss, lower markers of blood glucose, lower insulin resistance, and lower levels of thyroid hormones associated with metabolic disorders in a couple of poorly designed studies.

Vitamin D and metformin led to a lower marker of insulin resistance (HOMA-IR), lower body weight and testosterone, and more regular menstrual cycles among people with polycystic ovarian syndrome across many studies.

Nondiabetic people with endometrial hyperplasia combining metformin and a levonorgestrel intrauterine device (IUD) had lower body weight in one study.

Metformin and exercise, sometimes with diet

People without diabetes: preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of weight loss and lower insulin and leptin levels among nondiabetic people participating in physical activity and treated with metformin

More weight loss and lower insulin and leptin levels among nondiabetic people with breast or colorectal cancer participating in physical activity and treated with 850 mg metformin twice a day for 12 weeks compared to controls in a mid-sized RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects¹⁸⁰Meyerhardt JA, Irwin ML et al. Multicenter, randomized phase II trial of physical activity (PA), metformin (Met), or the combination on metabolic biomarkers in stage I-III colorectal (CRC) and breast cancer (BC) survivors. Journal of Clinical Oncology. 2017 May;35(15):10059-10059.
Lower levels of some markers of inflammation among nondiabetic people with stage 1–3 breast or colorectal cancer participating in exercise and treated with metformin compared to controls in a mid-sized RCT¹⁸¹Brown JC, Zhang S et al. Effect of exercise or metformin on biomarkers of inflammation in breast and colorectal cancer: a randomized trial. Cancer Prevention Research (Philadelphia). 2020 Dec;13(12):1055-1062.

People with diabetes or metabolic imbalances: weak evidencestudies with no controls OR weak trends of effects (this is the CancerChoices definition; other researchers and studies may define this differently) of lower body weight, waist circumference, blood sugar, and insulin resistance, as well as changes in thyroid hormone levels among people newly diagnosed with insulin resistance or diabetes participating in a diet and exercise intervention and treated with metformin

Lower body weight (BMI), waist circumference, insulin resistance, and thyroid-stimulating hormone, and higher levels of free triiodothyronine, but no evidence of an effect on free thyroxine (tetra-iodothyronine) among people newly diagnosed with insulin resistance following a standard diet and exercise program and treated with 1,700 mg metformin per day for 6 months compared to baseline in a mid-sized uncontrolled triala study in which a therapy is used, but without a comparison group against which to judge outcomes; an uncontrolled trial is considered a weak study design¹⁸²Anil C, Kut A et al. Metformin decreases thyroid volume and nodule size in subjects with insulin resistance: a preliminary study. Med Princ Pract. 2016;25(3):233-6.
Lower markers of blood glucose (HbA1c and methylglyoxal) and weak trends toward lower fasting glucose and body weight (BMI) among people newly diagnosed with type 2 diabetes treated with metformin increasing to 1000-2000 mg per day for 24 weeks and participating in monthly group and individual lifestyle counseling sessions providing specific calorie intake and exercise goals compared to baseline in a small uncontrolled trial¹⁸³Kender Z, Fleming T et al. Effect of metformin on methylglyoxal metabolism in patients with type 2 diabetes. Experimental and Clinical Endocrinology & Diabetes. 2014 May;122(5):316-9.

Vitamin D and metformin: strong evidenceconsistent, significant effects in several large (or at least one very large) well designed clinical studies or at least two meta-analyses of clinical studies of moderate or better quality (or one large meta-analysis) finding similar results (this is the CancerChoices definition; other researchers and studies may define this differently) of a lower marker of insulin resistance (HOMA-IR), lower body weight and testosterone, and more regular menstrual cycles among people with polycystic ovary syndrome treated with metformin and vitamin D

A lower marker of insulin resistance (HOMA-IR), lower body weight and testosterone, and more regular menstrual cycles among people with polycystic ovarian syndrome treated with metformin and vitamin D compared to controls in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 9 RCTs¹⁸⁴Xiang L, Liao M, Su Y. Efficacy of metformin combined with vitamin D in the treatment of polycystic ovarian syndrome: a meta-analysis. African Journal of Reproductive Health. 2024 Feb 28;28(2):43-54.

Levonorgestrel intrauterine system and metformin: preliminary evidence of lower body weight among nondiabetic people with endometrial hyperplasia without atypia treated with metformin

Lower body weight among nondiabetic people with endometrial hyperplasia without atypia treated with levonorgestrel intrauterine system (IUD) and 500 mg metformin twice daily compared to IUD alone compared to baseline in an uncontrolled analysis within a small RCT¹⁸⁵Ravi RD, Kalra et al. A randomized clinical trial of levonorgestrel intrauterine system with or without metformin for treatment of endometrial hyperplasia without atypia in Indian women. Asian Pacific Journal of Cancer Prevention. 2021 Mar 1;22(3):983-989.

Managing side effects and promoting wellness

Is metformin linked to fewer or less severe side effects or symptoms? Is it linked to less toxicity from cancer treatment? Does it support your quality of life or promote general well-being? We present the evidence.

Blood-related symptoms

Nondiabetic people with metastatic breast cancer treated with metformin during chemotherapy had less incidence of low levels of white blood cells (neutropenia) in one study.

Cardiovascular side effects

Two studies of nondiabetic people with breast cancer found conflicting results of effects regarding metformin’s protection of heart function during chemotherapy.

Cognitive difficulties

Two studies have found conflicting results of effects regarding metformin’s effects on cognitive function.

Preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of better performance on tests of working memory and auditory-verbal memory, and a weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward better reaction time, among survivors of pediatric brain tumors without unstable or insulin-dependent diabetes treated with metformin

Better performance on tests of working memory and auditory-verbal memory, and a weak trend toward better reaction time, among survivors of pediatric brain tumors (not specific to diabetes) treated with metformin compared to no metformin after either focal or cranial-spinal radiation in a small RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects¹⁸⁹Ayoub R, Ruddy RM et al. Assessment of cognitive and neural recovery in survivors of pediatric brain tumors in a pilot clinical trial using metformin. Nature Medicine. 2020 Aug;26(8):1285-1294.

No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on cognitive domain scores among overweight or obese postmenopausal women with breast cancer treated with metformin in a preliminary study

No evidence of an effect on scores in 5 cognitive domains among overweight or obese postmenopausal women with breast cancer treated with metformin compared to placebo in a mid-sized RCT¹⁹⁰Hartman SJ, Nelson SH et al. The effects of weight loss and metformin on cognition among breast cancer survivors: evidence from the Reach for Health study. Psycho-Oncology. 2019 Aug;28(8):1640-1646.

Fatigue

Nondiabetic people undergoing chemotherapy for breast cancer treated with metformin showed less fatigue in one study.

Gastrointestinal side effects

Nondiabetic people undergoing chemotherapy for breast cancer treated with metformin had a lower rate of fatty liver in one study.

Neuropathy and other neurological symptoms

Nondiabetic people with breast or colorectal cancer treated with metformin showed less chemotherapy-induced neuropathy in two studies.

Oral symptoms

Nondiabetic people with breast cancer treated with metformin had less oral mucositis during chemotherapy in one study.

Pain

Nondiabetic people with colorectal cancer treated with metformin showed less neuropathy-related pain during FOLFOX-4 chemotherapy in one study.

Reproductive side effects

People with breast cancer treated with metformin showed lower endometrial thickness during tamoxifen treatment.

Side effects not specific to cancer

People with diabetes (but not specific to cancer) treated with metformin had a lower risk of bone fractures in two large studies.

Nondiabetic people with major depressive disorder treated with metformin showed lower depression scores In one study.

Bone-related side effects: good evidencesignificant effects in one large or several mid-sized and well-designed clinical studies (randomized controlled trials (RCTs) with an appropriate placebo or other strong comparison control or observational studies that control for confounds) (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of fractures among people with diabetes (not specific to cancer) treated with metformin

12% lower risk of fractures among men aged 65 or older with type 2 diabetes (not specific to cancer) treated with metformin compared to no metformin in a very large observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods study¹⁹⁸Lee RH, Sloane R et al. Glycemic control and insulin treatment alter fracture risk in older men with type 2 diabetes mellitus. Journal of Bone and Mineral Research. 2019 Nov;34(11):2045-2051.
24% lower risk of fractures among people with diabetes (not specific to cancer) treated with metformin compared to no metformin in a very large observational study¹⁹⁹Josse RG, Majumdar SR et al; TECOS Study Group. Sitagliptin and risk of fractures in type 2 diabetes: results from the TECOS trial. Diabetes, Obesity & Metabolism. 2017 Jan;19(1):78-86.

Depression: preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of lower depression scores among nondiabetic people with major depressive disorder treated with metformin

Lower depression scores among nondiabetic people with major depressive disorder treated with 20 mg fluoxetine once daily plus 1000 mg metformin once daily for 12 weeks compared to fluoxetine alone in a small RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects²⁰⁰Abdallah MS, Mosalam EM et al. The antidiabetic metformin as an adjunct to antidepressants in patients with major depressive disorder: a proof-of-concept, randomized, double-blind, placebo-controlled trial. Neurotherapeutics. 2020 Oct;17(4):1897-1906.

Metformin combined with other therapies or practices

People with prostate cancer combining metformin with diet and exercise lost weight and had lower blood pressure during androgen deprivation therapy (ADT) in one study. People often experience weight gain and worsening metabolic health during ADT.

Reducing cancer risk

Is metformin linked to lower risks of developing cancer or of recurrence? We present the evidence.

People with type 2 diabetes are at higher risk for cancer due to specific diabetes-related processes that promote cancer. Metformin thwarts some of these cancer-promoting processes and helps correct terrain imbalances due to diabetes. As a result, we are not surprised to find that metformin shows the biggest benefits among people with cancer types linked to type 2 diabetes, insulin resistancea condition in which cells in your muscles, fat, and liver don’t respond well to insulin and can’t efficiently take up glucose from your blood for energy, and/or high blood sugar.

In the great majority of observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies (which is most of the available studies) of metformin’s effect on cancer risk, metformin use is limited almost entirely to people using it for diabetes. Most of metformin’s benefits related to cancer risk are found among people with diabetes, insulin resistance, and/or high blood sugar.

Cancer as a whole

Diabetic people treated with metformin had a lower risk of cancer and cancer mortality across many observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies but not across many RCTsrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects. Metformin did not have an effect on cancer mortality among people at risk of diabetes in one large study.

People with diabetes

Cancer risk: good evidencesignificant effects in one large or several mid-sized and well-designed clinical studies (randomized controlled trials (RCTs) with an appropriate placebo or other strong comparison control or observational studies that control for confounds) (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of cancer as a whole among diabetic people treated with metformin, although not when compared to treatment with sulfonylureas and rosiglitazone

Lower risk of cancer among people with type 2 diabetes treated with metformin compared to no metformin in a very large meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 67 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies²⁰²Zhang K, Bai P, Dai H, Deng Z. Metformin and risk of cancer among patients with type 2 diabetes mellitus: a systematic review and meta-analysis. Primary Care Diabetes. 2021 Feb;15(1):52-58.
Slightly lower risk of cancer among people with diabetes treated with metformin compared to people without diabetes in a meta-analysis of 53 observational studies²⁰³Campbell JM, Bellman SM, Stephenson MD, Lisy K. Metformin reduces all-cause mortality and diseases of ageing independent of its effect on diabetes control: a systematic review and meta-analysis. Ageing Research Reviews. 2017 Nov;40:31-44.
Insufficient evidenceconflicting findings across clinical studies with no preponderance of evidence in one direction; conflicting evidence occurs when studies find conflicting effects (positive effect vs no effect or negative effect) with the same treatment and the same general study population (same cancer type, for example) (this is the CancerChoices definition; other researchers and studies may define this differently) of an effect on cancer risk among people with type 2 diabetes treated with metformin compared to no metformin in the 3 studies at low risk of bias among 46 studies investigated²⁰⁴Farmer RE, Ford D et al. Metformin and cancer in type 2 diabetes: a systematic review and comprehensive bias evaluation. International Journal of Epidemiology. 2017 Apr 1;46(2):728-744.
11% lower risk of cancer among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 29 observational studies and controlled trials at low risk of bias²⁰⁵Wu L, Zhu J, Prokop LJ, Murad MH. Pharmacologic therapy of diabetes and overall cancer risk and mortality: a meta-analysis of 265 studies. Scientific Reports. 2015 Jun 15;5:10147.
No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments. on second primary cancer incidence among people with diabetes and head and neck cancer treated with metformin compared to no metformin in a mid-sized observational study²⁰⁶Kwon M, Roh JL et al. Effect of metformin on progression of head and neck cancers, occurrence of second primary cancers, and cause-specific survival. Oncologist. 2015 May;20(5):546-53.
31% lower risk of cancer among people with diabetes treated with metformin in a large meta-analysis of 47 observational studies²⁰⁷Gandini S, Puntoni M et al. Metformin and cancer risk and mortality: a systematic review and meta-analysis taking into account biases and confounders. Cancer Prevention Research (Philadelphia). 2014 Sep;7(9):867-85.
18% lower risk of cancer among people without cancer at baseline with diabetes treated with metformin compared to no metformin in a meta-analysis of 12 observational studies adjusted for body mass index²⁰⁸Gandini S, Puntoni M et al. Metformin and cancer risk and mortality: a systematic review and meta-analysis taking into account biases and confounders. Cancer Prevention Research (Philadelphia). 2014 Sep;7(9):867-85.
Lower risk of cancer among people with diabetes treated with metformin compared to no metformin in a very large meta-analysis of 18 observational studies²⁰⁹Franciosi M, Lucisano G et al. Metformin therapy and risk of cancer in patients with type 2 diabetes: systematic review. PLoS One. 2013 Aug 2;8(8):e71583.
No evidence of an effect on risk of cancer among people without cancer at baseline with type 2 diabetes treated with metformin compared to sulfonylureas and rosiglitazone in a meta-analysis of 8 RCTsrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects²¹⁰Franciosi M, Lucisano G et al. Metformin therapy and risk of cancer in patients with type 2 diabetes: systematic review. PLoS One. 2013 Aug 2;8(8):e71583.
No evidence of an effect on risk of cancer among people with type 2 diabetes treated with metformin compared to sulfonylureas and rosiglitazone in a meta-analysis of 3 RCTs, although lower risk in a meta-analyses of 21 observational studies²¹¹Thakkar B, Aronis KN, Vamvini MT, Shields K, Mantzoros CS. Metformin and sulfonylureas in relation to cancer risk in type II diabetes patients: a meta-analysis using primary data of published studies. Metabolism. 2013 Jul;62(7):922-34.

Cancer mortality: good evidence of lower cancer mortality among diabetic people without cancer at baseline treated with metformin in observational studies, but no evidence of an effect in RCTs

Lower cancer mortality among diabetic people without cancer at baseline treated with metformin compared either to other diabetes treatments or to people without diabetes in a very large observational study²¹²Gong Z, Aragaki AK et al. Diabetes, metformin and incidence of and death from invasive cancer in postmenopausal women: results from the Women’s Health Initiative. International Journal of Cancer. 2016;138(8):1915-1927.
31% lower cancer mortality among people without cancer at baseline with diabetes treated with metformin compared to no metformin in meta-analyses of 5 observational studies and controlled trials at low risk of bias²¹³Wu L, Zhu J, Prokop LJ, Murad MH. Pharmacologic therapy of diabetes and overall cancer risk and mortality: a meta-analysis of 265 studies. Scientific Reports. 2015 Jun 15;5:10147.
34% lower cancer mortality among people without cancer at baseline with diabetes treated with metformin in a large meta-analysis of 47 observational studies²¹⁴Gandini S, Puntoni M et al. Metformin and cancer risk and mortality: a systematic review and meta-analysis taking into account biases and confounders. Cancer Prevention Research (Philadelphia). 2014 Sep;7(9):867-85.
10% lower cancer mortality among people without cancer at baseline with diabetes treated with metformin compared to no metformin in meta-analyses of 5 observational studies adjusted for body mass index²¹⁵Gandini S, Puntoni M et al. Metformin and cancer risk and mortality: a systematic review and meta-analysis taking into account biases and confounders. Cancer Prevention Research (Philadelphia). 2014 Sep;7(9):867-85.
Lower cancer mortality among people without cancer at baseline with diabetes treated with metformin compared to no metformin in a meta-analysis of 6 observational studies, but no evidence of an effect on among people with type 2 diabetes treated with metformin compared to sulfonylureas and rosiglitazone in a meta-analysis of 8 RCTs²¹⁶Franciosi M, Lucisano G et al. Metformin therapy and risk of cancer in patients with type 2 diabetes: systematic review. PLoS One. 2013 Aug 2;8(8):e71583.

People without diabetes: no evidence of an effect on cancer-specific mortality among people at high risk of type 2 diabetes treated with metformin in a very large study

No evidence of an effect on cancer-specific mortality during 21 years of follow up among people at high risk of type 2 diabetes treated with metformin compared to placeboa pill, medicine, or procedure—thought to be both harmless and ineffective—prescribed for the psychological benefit to the patient or as a sham treatment in a study to allow a comparison to a therapy of interest in a very large RCT²¹⁷Lee CG, Heckman-Stoddard B et al; Diabetes Prevention Program Research Group; Diabetes Prevention Program Research Group. Effect of metformin and lifestyle interventions on mortality in the diabetes prevention program and diabetes prevention program outcomes study. Diabetes Care. 2021 Dec;44(12):2775-2782.

Bladder or urothelial cancer

Diabetic people with bladder or urothelial cancer treated with metformin showed a lower risk of recurrence across many studies.

Modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of recurrence among people with both bladder cancer and diabetes treated with metformin

A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower risk of recurrence after at least 4 years among diabetic people with bladder cancer treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 4 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies²¹⁸Yao X, Liu H, Xu H. The impact of metformin use with survival outcomes in urologic cancers: a systematic review and meta-analysis. Biomed Research International. 2021 Oct 8;2021:5311828.
45% lower risk of recurrence among diabetes people with both bladder cancer and treated with metformin compared to no metformin in a very large meta-analysis of 9 observational studies²¹⁹Hu J, Chen JB et al. Association of metformin intake with bladder cancer risk and oncologic outcomes in type 2 diabetes mellitus patients: a systematic review and meta-analysis. Medicine (Baltimore). 2018 Jul;97(30):e11596.

Modest evidence of lower risk of recurrence among people with urothelial cancer and diabetes treated with metformin

Lower risk of recurrence among people with urothelial cancer and diabetes treated with metformin compared to no metformin or to people without diabetes in a meta-analysis of 3 observational studies²²⁰Coyle C, Cafferty FH, Vale C, Langley RE. Metformin as an adjuvant treatment for cancer: a systematic review and meta-analysis. Annals of Oncology. 2016 Dec;27(12):2184-2195.

Breast cancer

Recurrence: Metformin did not show an effect on cancer recurrence among diabetic people with breast cancer as a whole or triple-negative cancer across many studies, but one large study found lower risk of recurrence of hormone-receptor positive breast cancer among diabetic people treated with metformin. Metformin did not show evidence of an effect on risk of breast cancer recurrence among nondiabetic people in one study.

Cancer risk: Metformin showed insufficient evidence of an effect on the risk of breast cancer among diabetic people across many studies. Nondiabetic, overweight people treated with metformin had lower cancer risk scores during breast imaging and lower rates of breast cancer in one study.

Recurrence

People with diabetes

Insufficient evidencepreclinical evidence only OR clinical studies with such poor or unclear methodology that no conclusion can be drawn OR conflicting findings across clinical studies with no preponderance of evidence in one direction; conflicting evidence occurs when studies find conflicting effects (positive effect vs no effect or negative effect) with the same treatment and the same general study population (same cancer type, for example) (this is the CancerChoices definition; other researchers and studies may define this differently) of an effect on recurrence among people with breast cancer and diabetes treated with metformin in a combined analysis of studies

No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on recurrence among people with breast cancer and diabetes treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 2 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies²²¹Coyle C, Cafferty FH, Vale C, Langley RE. Metformin as an adjuvant treatment for cancer: a systematic review and meta-analysis. Annals of Oncology. 2016 Dec;27(12):2184-2195.
A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower risk of recurrence (longer disease-free survival) among diabetic people with breast cancer using metformin at the time of diagnosis compared to nondiabetic people not taking metformin in a mid-sized observational study²²²Aksoy S, Sendur MA, Altundag K. Demographic and clinico-pathological characteristics in patients with invasive breast cancer receiving metformin. Medical Oncology. 2013;30(2):590.

Lower risk of recurrence (better disease-free survival and distant disease-free survival) among diabetic people with hormone-receptor positive but not negative breast cancer treated with metformin compared to no metformin in a mid-sized observational analysis from a larger RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects²²³Sonnenblick A, Agbor-Tarh D et al. Impact of diabetes, insulin, and metformin use on the outcome of patients with human epidermal growth factor receptor 2-positive primary breast cancer: analysis from the ALTTO phase III randomized trial. Journal of Clinical Oncology. 2017 May 1;35(13):1421-1429.

People without diabetes

No evidence of an effect on risk of recurrence among nondiabetic people with breast cancer using metformin

No evidence of an effect on risk of recurrence (invasive disease-free survival) among nondiabetic people with high-risk operable breast cancer treated with 850 mg oral metformin twice a day compared to placebo in a very large RCT²²⁴Goodwin PJ, Chen BE et al. Effect of metformin vs placebo on invasive disease-free survival in patients with breast cancer: the MA.32 randomized clinical trial. Journal of the American Medical Association. 2022 May 24;327(20):1963-1973.

Cancer risk

People with diabetes

Insufficient evidence of an effect on risk of breast cancer among people with type 2 diabetes treated with metformin, with possibly lower risk with use for 3 years or longer

A weak trend toward lower risk of breast cancer among people with type 2 diabetes treated with metformin compared to other diabetes treatments in a meta-analysis of 12 observational studies²²⁵Tang GH, Satkunam M et al. Association of metformin with breast cancer incidence and mortality in patients with type II diabetes: a GRADE-assessed systematic review and meta-analysis. Cancer Epidemiology, Biomarkers & Prevention. 2018 Jun;27(6):627-635.
No evidence of an effect on risk of breast cancer among people with type 2 diabetes treated with metformin compared to no metformin in meta-analysis of 3 RCTs or a a very large meta-analysis of 9 observational studies²²⁶Franciosi M, Lucisano G et al. Metformin therapy and risk of cancer in patients with type 2 diabetes: systematic review. PLoS One. 2013 Aug 2;8(8):e71583.
6% lower risk of breast cancer among people without cancer at baseline (mostly with diabetes) treated with metformin compared to no metformin in a meta-analysis of 7 observational studies²²⁷Zhang P, Li H, Tan X, Chen L, Wang S. Association of metformin use with cancer incidence and mortality: a meta-analysis. Cancer Epidemiology. 2013 Jun;37(3):207-18.
A very weak trend toward lower risk of breast cancer among people with type 2 diabetes treated with metformin compared to sulfonylurea in a large meta-analysis of 6 observational studies²²⁸Soranna D, Scotti L et al. Cancer risk associated with use of metformin and sulfonylurea in type 2 diabetes: a meta-analysis. Oncologist. 2012;17(6):813-22.
17% lower risk of breast cancer among postmenopausal women with diabetes treated with metformin compared to no metformin, with stronger associations among people using metformin for 3 years or longer compared to less than 3 years in a meta-analysis of 7 observational studies²²⁹Col NF, Ochs L, Springmann V, Aragaki AK, Chlebowski RT. Metformin and breast cancer risk: a meta-analysis and critical literature review. Breast Cancer Research and Treatment. 2012 Oct;135(3):639-46.

Modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower breast cancer-specific mortality among people without cancer at baseline with diabetes treated with metformin

37% lower breast cancer-specific mortality among people without cancer at baseline with diabetes treated with metformin compared to no metformin in a meta-analysis of 2 observational studies²³⁰Zhang P, Li H, Tan X, Chen L, Wang S. Association of metformin use with cancer incidence and mortality: a meta-analysis. Cancer Epidemiology. 2013 Jun;37(3):207-18.

Preliminary evidence of fewer grade 3 tumors or triple-negative tumors, but higher incidence of ER- or PR-positive tumors, among people with breast cancer (not specific to diabetes) using metformin at the time of diagnosis

Lower incidence of grade 3 tumors or triple negative tumors, but higher incidence of ER or PR positive tumors, among diabetic people with breast cancer treated with metformin at the time of diagnosis compared to nondiabetic people not treated with metformin in a mid-sized observational study²³¹Aksoy S, Sendur MA, Altundag K. Demographic and clinico-pathological characteristics in patients with invasive breast cancer receiving metformin. Medical Oncology. 2013;30(2):590.

People without diabetes: preliminary evidence of lower Breast Imaging Reporting and Data System (BIRADS) scores and positive pathological biopsy rate among nondiabetic overweight/obese premenopausal women treated with metformin

Lower BI-RADS scores and positive pathological biopsy rate among nondiabetic overweight/obese premenopausal women treated with lifestyle interventions for diet and exercise plus metformin compared to lifestyle interventions alone for a year in a mid-sized RCT²³²Leng W, Pu D et al. Effect of metformin on breast density in overweight/obese premenopausal women. Diabetes, Metabolic Syndrome and Obesity: Targets and Therapy. 2021 Nov 3;14:4423-4432.

Colorectal cancer

Recurrence: Diabetic people with colorectal cancer treated with metformin showed a lower risk of cancer recurrence across many studies.

Cancer risk: Diabetic people treated with metformin showed a lower risk of colorectal cancer across many studies. People with impaired glucose tolerance treated with metformin showed fewer rectal aberrant crypt foci—a risk factor for cancer—in one study. Metformin did not show evidence of an effect on the number of colorectal polyps among nondiabetic people with familial adenomatous polyposis in one study.

Recurrence

People with diabetes: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of recurrence among diabetic people with colorectal cancer treated with metformin, with stronger effects among people with early stage cancer

Substantially lower risk of recurrence among people with colorectal cancer and diabetes treated with metformin compared to no metformin in a mid-sized observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods study²³³Henderson D, Frieson D, Zuber J, Solomon SS. Metformin has positive therapeutic effects in colon cancer and lung cancer. American Journal of the Medical Science. 2017 Sep;354(3):246-251.
A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower risk of recurrence (disease-free survival) among people with colorectal cancer and diabetes treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 3 observational studies²³⁴Du L, Wang M et al. Prognostic role of metformin intake in diabetic patients with colorectal cancer: an updated qualitative evidence of cohort studies. Oncotarget. 2017 Apr 18;8(16):26448-26459.
37% lower risk of recurrence among people with early stage colorectal cancer and diabetes treated with metformin compared to no metformin in a meta-analysis of 2 observational studies²³⁵Coyle C, Cafferty FH, Vale C, Langley RE. Metformin as an adjuvant treatment for cancer: a systematic review and meta-analysis. Annals of Oncology. 2016 Dec;27(12):2184-2195.

Cancer risk

People with diabetes: good evidencesignificant effects in one large or several mid-sized and well-designed clinical studies (randomized controlled trials (RCTs) with an appropriate placebo or other strong comparison control or observational studies that control for confounds) (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of colorectal cancer among people with diabetes treated with metformin

29% lower risk of colorectal cancer among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 31 observational studies²³⁶Wang Q, Shi M. Effect of metformin use on the risk and prognosis of colorectal cancer in diabetes mellitus: a meta-analysis. Anticancer Drugs. 2022 Feb 1;33(2):191-199.
11% lower risk of colorectal cancer among people with diabetes treated with metformin compared to no metformin in a very large meta-analysis of 17 observational studies²³⁷Yang WT, Yang HJ, Zhou JG, Liu JL. Relationship between metformin therapy and risk of colorectal cancer in patients with diabetes mellitus: a meta-analysis. International Journal of Colorectal Disease. 2020 Nov;35(11):2117-2131.
23% lower risk of colorectal adenoma, 39% lower risk of advanced adenoma, and 24% lower risk of colorectal cancer among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 58 observational studies²³⁸Ng CW, Jiang AA et al. Metformin and colorectal cancer: a systematic review, meta-analysis and meta-regression. International Journal of Colorectal Disease. 2020 Aug;35(8):1501-1512.
Lower risk of colorectal cancer among people with type 2 diabetes treated with metformin compared to no metformin in a large meta-analysis of 7 observational studies²³⁹Su T, Liao B et al. [Effect of metformin on colorectal carcinoma in type 2 diabetes mellitus patients: a Markov model analysis]. Zhonghua Wei Chang Wai Ke Za Zhi. 2017 Jun 25;20(6):689-693.
Moderately lower risk of adenomas among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 10 observational studies²⁴⁰Jung YS, Park CH, Eun CS, Park DI, Han DS. Metformin use and the risk of colorectal adenoma: a systematic review and meta-analysis. Journal of Gastroenterology and Hepatology. 2017 May;32(5):957-965.
Moderately lower risk of colorectal adenomas or colorectal cancer among people with type 2 diabetes treated with metformin compared to no metformin in a meta-analysis of 19 observational studies and 1 RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects²⁴¹Liu F, Yan L et al. Metformin therapy and risk of colorectal adenomas and colorectal cancer in type 2 diabetes mellitus patients: a systematic review and meta-analysis. Oncotarget. 2017 Feb 28;8(9):16017-16026.
Moderately lower risk of colorectal adenoma (7 studies) and advanced adenomas (4 studies) among people with type 2 diabetes treated with metformin compared to no metformin in meta-analyses of observational studies²⁴²Hou YC, Hu Q, Huang J, Fang JY, Xiong H. Metformin therapy and the risk of colorectal adenoma in patients with type 2 diabetes: a meta-analysis. Oncotarget. 2017 Jan 31;8(5):8843-8853.
25% lower risk of colorectal cancer among people with type 2 diabetes treated with metformin compared to no metformin in a meta-analysis of 11 observational studies²⁴³Nie Z, Zhu H, Gu M. Reduced colorectal cancer incidence in type 2 diabetic patients treated with metformin: a meta-analysis. Pharmaceutical Biology. 2016 Nov;54(11):2636-2642.
21% lower risk of colon cancer and 42% lower risk of colon polyps among people with type 2 diabetes treated with metformin compared to no metformin in a meta-analysis of 16 observational studies and 1 RCT²⁴⁴Rokkas T, Portincasa P. Colon neoplasia in patients with type 2 diabetes on metformin: a meta-analysis. European Journal of Internal Medicine. 2016 Sep;33:60-6.
Slightly lower risk of colorectal cancer among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 15 observational studies²⁴⁵He XK, Su TT, Si JM, Sun LM. Metformin is associated with slightly reduced risk of colorectal cancer and moderate survival benefits in diabetes mellitus: a meta-analysis. Medicine (Baltimore). 2016 Feb;95(7):e2749.

People with non-diabetes metabolic imbalances: weak evidenceone or more case studies, supported by animal evidence OR small treatment effects of limited clinical significance OR studies with no controls OR weak trends of effects (this is the CancerChoices definition; other researchers and studies may define this differently) of fewer rectal aberrant crypt foci, a risk factor for colorectal cancer, among people with impaired glucose tolerance treated with higher doses of metformin

Fewer rectal aberrant crypt foci at 3 and 6 months among people with impaired glucose tolerance treated with 500 mg or 1500 mg metformin a day compared to baseline, but no decrease among people treated with 250 mg or no metformin in a mid-sized RCT²⁴⁶Zhao X, Li Y et al. Effects of different doses of metformin treatment for 6 months on aberrant crypt foci in Chinese patients with impaired glucose tolerance. European Journal of Cancer Prevention. 2015 Jan;24(1):27-36.

People without diabetes: no evidence of an effect on the number or size of colorectal and duodenal polyps among nondiabetic people with familial adenomatous polyposis treated with metformin in a preliminary study

No evidence of an effect on the percentage change in the number or size of colorectal and duodenal polyps among nondiabetic people with familial adenomatous polyposis treated with either 500 mg or 1,500 mg oral metformin a day for 7 months compared to placebo in a small RCT²⁴⁷Park JJ, Kim BC et al. The effect of metformin in treatment of adenomas in patients with familial adenomatous polyposis. Cancer Prevention Research (Philadelphia). 2021 May;14(5):563-572.

Gastrointestinal cancer

Colorectal cancer and pancreatic cancer are listed separately.

Recurrence: Diabetic people with liver cancer treated with metformin showed a lower risk of recurrence compared to other diabetes treatments across several studies.

Cancer risk: Diabetic people treated with metformin showed slightly lower risk of esophageal cancer in one combined analysis of studies, although another analysis did not find this benefit. Metformin did not show an effect on cell proliferation among diabetic people with Barrett’s esophagus—a risk factor for esophageal cancer—in one study. Diabetic people treated with metformin showed a lower risk of liver cancer or stomach cancer across many studies.

Esophageal cancer

People with diabetes: preliminary (conflicting) evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of slightly lower risk of esophageal cancer among people with type 2 diabetes treated with metformin

No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on risk of esophageal cancer among people with type 2 diabetes treated with metformin compared to no metformin, although lower risk was seen among Asian people, in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 5 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies²⁴⁸Wu HD, Zhang JJ, Zhou BJ. The effect of metformin on esophageal cancer risk in patients with type 2 diabetes mellitus: a systematic review and meta‑analysis. Clinical & Translational Oncology. 2021 Feb;23(2):275-282.
Slightly lower risk of esophageal cancer among people with type 2 diabetes treated with metformin compared to no metformin in a very large meta-analysis of 2 observational studies²⁴⁹Franciosi M, Lucisano G et al. Metformin therapy and risk of cancer in patients with type 2 diabetes: systematic review. PLoS One. 2013 Aug 2;8(8):e71583.

People without diabetes: no evidence of an effect on markers of cell proliferation or cell death among nondiabetic people with Barrett’s esophagus treated with metformin in a preliminary study

No evidence of an effect on markers of cell proliferation or cell death (apoptosis) among nondiabetic people with Barrett’s esophagus treated with increasing doses of metformin to 2000 mg/day by week 4 compared to placebo in a small RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects²⁵⁰Chak A, Buttar NS et al; Cancer Prevention Network. Metformin does not reduce markers of cell proliferation in esophageal tissues of patients with Barrett’s esophagus. Clinical Gastroenterology and Hepatology. 2015 Apr;13(4):665-72.e1-4.

Liver cancer

Recurrence: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of recurrence among diabetic people with liver cancer (hepatocellular carcinoma) treated with metformin

Substantial lower risk of recurrence (recurrence-free survival) at 1, 3, and 5 years after curative cancer therapies among people with liver cancer (hepatocellular carcinoma) and diabetes treated with metformin compared to other antihyperglycemic agents in a meta-analysis of 6 observational studies²⁵¹Zhou J, Ke Y et al. Meta-analysis: the efficacy of metformin and other anti-hyperglycemic agents in prolonging the survival of hepatocellular carcinoma patients with type 2 diabetes. Annals of Hepatology. 2020 May-Jun;19(3):320-328.
No evidence of effect on recurrence after initial liver resection among people with liver cancer (hepatocellular carcinoma) and diabetes treated with metformin compared to no metformin in a mid-sized observational study²⁵²Cho WR, Wang CC et al. Impact of metformin use on the recurrence of hepatocellular carcinoma after initial liver resection in diabetic patients. PLoS One. 2021 Mar 4;16(3):e0247231.

Cancer risk: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of liver cancer among people with diabetes treated with metformin

Moderately lower risk of liver cancer among people with diabetes treated with metformin compared to no metformin in a very large meta-analysis of 17 observational studies and 2 RCTs²⁵³Ma S, Zheng Y, Xiao Y, Zhou P, Tan H. Meta-analysis of studies using metformin as a reducer for liver cancer risk in diabetic patients. Medicine (Baltimore). 2017 May;96(19):e6888.
51% lower risk of hepatocellular carcinoma (HCC) among people mostly with diabetes treated with metformin compared to no metformin, 55% lower risk compared to sulfonylurea, and 72% lower risk compared to insulin, but no evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments compared to thiazolidinediones in a very large meta-analysis of 3 RCTs limited to people with diabetes plus 12 observational studies²⁵⁴Zhou YY, Zhu GQ et al. Systematic review with network meta-analysis: antidiabetic medication and risk of hepatocellular carcinoma. Scientific Reports. 2016 Sep 19;6:33743.
Lower risk of liver cancer among people with type 2 diabetes treated with metformin compared to no metformin in a very large meta-analysis of 8 observational studies²⁵⁵Franciosi M, Lucisano G et al. Metformin therapy and risk of cancer in patients with type 2 diabetes: systematic review. PLoS One. 2013 Aug 2;8(8):e71583.
78% lower risk of liver cancer among diabetic people without cancer at baseline treated with metformin compared to no metformin in meta-analyses of 5 observational studies²⁵⁶Zhang P, Li H, Tan X, Chen L, Wang S. Association of metformin use with cancer incidence and mortality: a meta-analysis. Cancer Epidemiology. 2013 Jun;37(3):207-18.
50% lower risk of hepatocellular cancer among people with diabetes treated with metformin compared to no metformin in a very large meta-analysis of 8 observational studies²⁵⁷Singh S, Singh PP, Singh AG, Murad MH, Sanchez W. Anti-diabetic medications and the risk of hepatocellular cancer: a systematic review and meta-analysis. American Journal of Gastroenterology. 2013 Jun;108(6):881-91; quiz 892.
76% lower risk of hepatocellular carcinoma among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 7 observational studies²⁵⁸Zhang H, Gao C, Fang L, Zhao HC, Yao SK. Metformin and reduced risk of hepatocellular carcinoma in diabetic patients: a meta-analysis. Scandinavian Journal of Gastroenterology. 2013 Jan;48(1):78-87.

Cancer mortality: preliminary evidence of substantially lower liver cancer-specific mortality among diabetic people without cancer at baseline with diabetes treated with metformin

77% lower liver cancer-specific mortality among diabetic people without cancer at baseline with diabetes treated with metformin compared to no metformin in meta-analyses of 2 observational studies²⁵⁹Zhang P, Li H, Tan X, Chen L, Wang S. Association of metformin use with cancer incidence and mortality: a meta-analysis. Cancer Epidemiology. 2013 Jun;37(3):207-18.

Stomach cancer: modest evidence of lower risk of stomach cancer among people with type 2 diabetes treated with metformin, although with a risk of bias in studies that may contribute to overestimating the benefit

No evidence of an effect on the risk of stomach (gastric) cancer among people with type 2 diabetes treated with metformin compared to no metformin in a meta-analysis of 8 observational studies without immortal time biasa study design confound that leads to a biased association; ‘immortal time’ is when participants of a cohort study cannot experience the outcome during some period of follow-up time, but moderately reduced risk in a meta-analysis of 6 observational studies with immortal time bias; the researchers conclude that this bias may be overstating the effects of metformin in many studies²⁶⁰Wang YB, Tan LM et al. Immortal time bias exaggerates the effect of metformin on the risk of gastric cancer: a meta-analysis. Pharmacological Research. 2021 Mar;165:105425.
21% lower risk of stomach (gastric) cancer, with benefit among Asian populations but not Western populations, among people with type 2 diabetes treated with metformin compared to no metformin in a meta-analysis of 11 observational studies²⁶¹Shuai Y, Li C, Zhou X. The effect of metformin on gastric cancer in patients with type 2 diabetes: a systematic review and meta-analysis. Clinical & Translational Oncology. 2020 Sep;22(9):1580-1590.
23% lower risk of stomach (gastric) cancer among people with diabetes treated with metformin compared to no metformin in a very large meta-analysis of 7 observational studies²⁶²Zhou XL, Xue WH et al. Association between metformin and the risk of gastric cancer in patients with type 2 diabetes mellitus: a meta-analysis of cohort studies. Oncotarget. 2017 Apr 8;8(33):55622-55631.
Lower risk of stomach cancer among people with type 2 diabetes treated with metformin compared to no metformin in a very large meta-analysis of 2 observational studies²⁶³Franciosi M, Lucisano G et al. Metformin therapy and risk of cancer in patients with type 2 diabetes: systematic review. PLoS One. 2013 Aug 2;8(8):e71583.

Gynecologic cancer

Ovarian cancer is listed separately.

Diabetic people treated with metformin showed a lower risk of gynecological cancer as a whole, with especially lower risk among Asian people, across many studies.

Diabetic people with cervical cancer treated with metformin showed a lower risk of recurrence in one study. Diabetic people treated with metformin showed a lower risk of cervical cancer across two studies.

Diabetic people with type 1, as opposed to type 2, endometrial cancer treated with metformin had less recurrence in one study, while overweight diabetic people with endometrial cancer treated with metformin had less recurrence in another. Several combined analyses of studies found mixed evidence on whether diabetic people treated with metformin had a lower risk of endometrial cancer. People with endometrial hyperplasia treated with metformin showed a better response to megestrol acetate in two studies, but not to progestin in one study. Metformin did not show evidence of an effect on regression among nondiabetic people with endometrial hyperplasia in two studies.

Gynecologic cancer as a whole

Cancer risk: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of gynecologic cancers as a whole among diabetic people treated with metformin

51% lower risk of gynecologic cancers as a whole, with stronger effects among Asians but no evidence of an effect among Caucasians with diabetes treated with metformin compared to no metformin in a very large meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 11 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies²⁶⁴Wen Q, Zhao Z et al. The association between metformin therapy and risk of gynecological cancer in patients: two meta-analyses. European Journal of Obstetrics, Gynecology and Reproductive Biology. 2019 Jun;237:33-41.

Cervical cancer

Recurrence: preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of recurrence among people with cervical cancer and type 2 diabetes treated with metformin

Lower risk of recurrence among people with cervical cancer and type 2 diabetes treated with metformin compared to no metformin in a mid-sized observational study²⁶⁵Hanprasertpong J, Jiamset I et al. The effect of metformin on oncological outcomes in patients with cervical cancer with type 2 diabetes mellitus. International Journal of Gynecological Cancer. 2017 Jan;27(1):131-137.

Cancer risk: modest evidence of lower risk of cervical cancer among people with diabetes treated with metformin

40% lower risk of cervical cancer among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 2 observational studies²⁶⁶Wen Q, Zhao Z et al. The association between metformin therapy and risk of gynecological cancer in patients: two meta-analyses. European Journal of Obstetrics, Gynecology and Reproductive Biology. 2019 Jun;237:33-41.

Endometrial cancer

Recurrence: modest evidence of lower risk of recurrence among diabetic people with endometrial cancer treated with metformin

50% lower risk of recurrence among people with endometrial cancer with diabetes treated with metformin compared to no metformin in meta-analyses of 3 observational studies²⁶⁷Chu D, Wu J et al. Effect of metformin use on the risk and prognosis of endometrial cancer: a systematic review and meta-analysis. BMC Cancer. 2018 Apr 18;18(1):438.

Cancer risk

People with diabetes: insufficient evidencepreclinical evidence only OR clinical studies with such poor or unclear methodology that no conclusion can be drawn OR conflicting findings across clinical studies with no preponderance of evidence in one direction; conflicting evidence occurs when studies find conflicting effects (positive effect vs no effect or negative effect) with the same treatment and the same general study population (same cancer type, for example) (this is the CancerChoices definition; other researchers and studies may define this differently) of an effect on risk of endometrial cancer among people with diabetes treated with metformin

No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on risk of endometrial cancer among people with diabetes treated with metformin compared to no metformin in a very large meta-analysis of 6 observational studies²⁶⁸Tian J, Liang Y, Qu P. Antidiabetic medications and the risk of endometrial cancer in patients. Gynecologic and Obstetric Investigation. 2019;84(5):455-462.
No evidence of an effect on risk of endometrial cancer among people with diabetes treated with metformin compared to no metformin in meta-analyses of 7 observational studies²⁶⁹Chu D, Wu J et al. Effect of metformin use on the risk and prognosis of endometrial cancer: a systematic review and meta-analysis. BMC Cancer. 2018 Apr 18;18(1):438.
13% lower risk of endometrial cancer among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 5 observational studies²⁷⁰Tang YL, Zhu LY et al. Metformin use is associated with reduced incidence and improved survival of endometrial cancer: a meta-analysis. Biomed Research International. 2017;2017:5905384.

People without diabetes: preliminary evidence of higher response rates to treatment with megestrol acetate among people with endometrial atypical hyperplasia (a precancerous condition) and metabolic imbalances treated with metformin

A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward higher rates of response to treatment with megestrol acetate (MA) among people with endometrial atypical hyperplasia and metabolic imbalances treated with metformin compared to MA alone in a small RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects²⁷¹Shan W, Wang C et al. Conservative therapy with metformin plus megestrol acetate for endometrial atypical hyperplasia. Journal of Gynecologic Oncology. 2014 Jul;25(3):214-20.
Higher response rate to 40 mg megestrol acetate among overweight women with endometrial hyperplasia without atypia (not specific to diabetes) treated with 1000 mg metformin daily for 3 months compared to placebo in a small RCT²⁷²Tehranian A, Ghahghaei-Nezamabadi A et al. The impact of adjunctive metformin to progesterone for the treatment of non-atypical endometrial hyperplasia in a randomized fashion, a placebo-controlled, double blind clinical trial. Journal of Gynecology Obstetrics and Human Reproduction. 2021 Jun;50(6):101863.

No evidence of an effect on complete response rates to progestin therapy among obese women with complex atypical hyperplasia (controlled for diabetes status) treated with metformin in one study

No evidence of an effect on complete response rates after progestin therapy among obese women with complex atypical hyperplasia (controlled for diabetes status) treated with metformin compared to no metformin in a mid-sized observational study²⁷³Matsuo K, Mandelbaum RS et al. Route-specific association of progestin therapy and concurrent metformin use in obese women with complex atypical hyperplasia. International Journal of Gynecological Cancer 2020;30:1–9.

No evidence of an effect on complete response rates or regression of endometrial hyperplasia, a risk factor for endometrial cancer, among nondiabetic people treated with metformin in two small studies

No evidence of an effect on complete response rate among nondiabetic people with endometrial hyperplasia without atypia treated with levonorgestrel intrauterine system and 500 mg metformin twice daily compared to levonorgestrel intrauterine system alone in a small RCT²⁷⁴Ravi RD, Kalra et al. A randomized clinical trial of levonorgestrel intrauterine system with or without metformin for treatment of endometrial hyperplasia without atypia in Indian women. Asian Pacific Journal of Cancer Prevention. 2021 Mar 1;22(3):983-989.
No evidence of an effect on regression of endometrial hyperplasia among people (not specific to diabetes) treated with metformin compared to megestrol in a small RCT²⁷⁵Sharifzadeh F, Aminimoghaddam S et al. A comparison between the effects of metformin and megestrol on simple endometrial hyperplasia. Gynecological Endocrinology. 2017 Feb;33(2):152-155.

Head, neck, and oral cancer

Recurrence: Diabetic people with head and neck cancer treated with metformin had a lower risk of recurrence across several studies.

Cancer risk: Diabetic people treated with metformin had a lower risk of head and neck cancer in one study. Nondiabetic people with oral premalignant lesions treated with metformin showed clinical response to metformin in one poorly designed study.

Recurrence: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of recurrence among diabetic people with head and neck cancer treated with metformin

Substantially longer recurrence-free survival after tumor resection to negative margins among people with head and neck cancer and type 2 diabetes treated with metformin compared to no metformin, either among those with diabetes or all cancer patients in a mid-sized observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods study²⁷⁶Spoerl S, Gerken M et al. Prognostic relevance of type 2 diabetes and metformin treatment in head and neck melanoma: results from a population-based cohort study. Current Oncology. 2022 Dec 7;29(12):9660-9670.
Comparable rates of recurrence (disease-free survival) among presumably diabetic people with head and neck cancer treated with metformin compared to no metformin (not specific to diabetes) in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 7 observational studies;²⁷⁷Wang Y, Fu T et al. The association between metformin and survival of head and neck cancer: a systematic review and meta-analysis of 7 retrospective cohort studies. Current Pharmaceutical Design. 2020;26(26):3161-3170. as described in the commentary above, we would expect higher rates of recurrence among people with diabetes
Lower risk of recurrence (better disease-free survival) among diabetic people with hypopharyngeal cancer treated with metformin compared to no metformin or no diabetes in a mid-sized observational study²⁷⁸Tsou YA, Chang WD et al. The effect of metformin use on hypopharyngeal squamous cell carcinoma in diabetes mellitus patients. BMC Cancer. 2019 Aug 30;19(1):862.
No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on recurrence-free survival among diabetic people with head and neck squamous carcinoma treated with metformin compared to no metformin in a mid-sized observational study²⁷⁹Lee DJ, McMullen CP et al. Impact of metformin on disease control and survival in patients with head and neck cancer: a retrospective cohort study. Journal of Otolaryngology—Head & Neck Surgery. 2019 Jul 25;48(1):34.
Lower risk of recurrence (better disease-free survival) among diabetic people with laryngeal squamous cell carcinoma treated with metformin compared to no metformin, with comparable risk to people without diabetes, in a mid-sized observational study²⁸⁰Sandulache VC, Hamblin JS et al. Association between metformin use and improved survival in patients with laryngeal squamous cell carcinoma. Head & Neck. 2014 Jul;36(7):1039-43.

Cancer risk

Modest evidence of lower risk of head and neck cancer among people with diabetes treated with metformin

29% lower risk of head and neck cancer among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 4 observational studies²⁸¹Saka Herrán C, Jané-Salas E, Estrugo Devesa A, López-López J. Protective effects of metformin, statins and anti-inflammatory drugs on head and neck cancer: a systematic review. Oral Oncology. 2018 Oct;85:68-81.

Weak evidencestudies with no controls OR weak trends of effects (this is the CancerChoices definition; other researchers and studies may define this differently) of clinical response among nondiabetic people with oral premalignant lesions treated with metformin

17% complete responses and 43% partial responses among nondiabetic people with oral premalignant lesions treated with metformin increasing to 2000 mg daily for 12 weeks, with stronger responses among current and former smokers compared to never smokers, in a small uncontrolled triala study in which a therapy is used, but without a comparison group against which to judge outcomes; an uncontrolled trial is considered a weak study design²⁸²Gutkind JS, Molinolo AA et al. Inhibition of mTOR signaling and clinical activity of metformin in oral premalignant lesions. JCI Insight. 2021 Sep 8;6(17):e147096.

Kidney cancer

Metformin showed no evidence of an effect on recurrence among diabetic people with kidney cancer in one study. However, diabetic people treated with metformin showed similar recurrence rates to people not taking metformin, including nondiabetic people, in another study. We would expect diabetic people to have worse cancer outcomes compared to nondiabetic people.

Lung cancer

Diabetic people with lung cancer treated with metformin showed a lower risk of cancer recurrence across many studies.

People with diabetes treated with metformin showed a lower risk of lung cancer across many studies.

See also evidence of higher risk of advanced disease and worse survival among people with lung cancer treated with metformin in Safety and precautions ›

Recurrence: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of moderately lower risk of recurrence (higher disease-free survival) among people with lung cancer and diabetes treated with metformin

Moderately lower risk of recurrence (higher disease-free survival) among people with lung cancer and diabetes treated with metformin compared to no metformin in a meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 18 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies²⁸⁵Xin WX, Fang L et al. Effect of hypoglycemic agents on survival outcomes of lung cancer patients with diabetes mellitus: a meta-analysis. Medicine (Baltimore). 2018 Mar;97(9):e0035.
No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments of an effect on risk of recurrence among people with lung cancer and diabetes treated with metformin compared to no metformin in a mid-sized observational study²⁸⁶Henderson D, Frieson D, Zuber J, Solomon SS. Metformin has positive therapeutic effects in colon cancer and lung cancer. American Journal of the Medical Science. 2017 Sep;354(3):246-251.

Cancer risk: modest evidence of a slightly lower risk of lung cancer among people with diabetes treated with metformin

22% lower risk of lung cancer among people with type 2 diabetes treated with metformin compared to no metformin in a meta-analysis of 8 observational studies²⁸⁷Xiao K, Liu F et al. The effect of metformin on lung cancer risk and survival in patients with type 2 diabetes mellitus: a meta-analysis. Journal of Clinical Pharmacy and Therapeutics. 2020 Aug;45(4):783-792.
11% lower risk of lung cancer among diabetic people treated with metformin compared to no metformin in a meta-analysis of 13 observational studies²⁸⁸Yao L, Liu M et al. Metformin use and lung cancer risk in diabetic patients: a systematic review and meta-analysis. Disease Markers. 2019 Feb 10;2019:6230162.
Slightly lower risk of lung cancer among diabetic people treated with metformin compared to no metformin after adjusting for smoking status in a meta-analysis of 4 observational studies²⁸⁹Zhang ZJ, Bi Y et al. Reduced risk of lung cancer with metformin therapy in diabetic patients: a systematic review and meta-analysis. American Journal of Epidemiology. 2014 Jul 1;180(1):11-4.
A weak trend toward slightly lower risk of lung cancer among diabetic people s treated with metformin compared to no metformin after adjusting for smoking status in a meta-analysis of 13 observational studies and 2 RCTsrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects, but no evidence of an effect in RCTs alone comparing metformin to antidiabetic drugs rosiglitazone or glibenclamide²⁹⁰Wu Y, Liu HB, Shi XF, Song Y. Conventional hypoglycaemic agents and the risk of lung cancer in patients with diabetes: a meta-analysis. PLoS One. 2014 Jun 12;9(6):e99577.

Ovarian cancer

Diabetic people with ovarian cancer treated with metformin showed lower rates of recurrence across many studies.

Metformin showed no evidence of an effect on recurrence among nondiabetic people with ovarian cancer in one study.

Recurrence

People with diabetes: preliminary evidencesignificant effects in small or poorly designed clinical studies OR conflicting results in adequate studies but a preponderance of evidence of an effect (this is the CancerChoices definition; other researchers and studies may define this differently) of longer recurrence-free survival among diabetic people with ovarian cancer treated with metformin

58% longer recurrence-free survival among diabetic people with ovarian cancer treated with metformin compared to no metformin in a small subgroup analysis within a mid-sized observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods study²⁹¹Simonelli C, Bertolotti M et al. Effect of metformin on recurrence-free survival and overall survival in diabetic patients affected by advanced ovarian cancer. Journal of Clinical Oncology 2013;31.
A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower risk of recurrence (better disease-free survival) among mostly diabetic people with ovarian cancer treated with metformin compared to no metformin (not specific to diabetes) in a meta-analysis of 6 observational studies;²⁹²Wang Y, Liu X et al. No effect of metformin on ovarian cancer survival: a systematic review and meta-analysis of cohort studies. Current Pharmaceutical Design. 2019;25(23):2595-2601. our note: we would expect equal or higher rates of recurrence among people with diabetes

Cancer risk

People with diabetes: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of ovarian cancer among women with diabetes treated with metformin

82% lower risk of ovarian cancer among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 3 observational studies²⁹³Wen Q, Zhao Z et al. The association between metformin therapy and risk of gynecological cancer in patients: two meta-analyses. European Journal of Obstetrics, Gynecology and Reproductive Biology. 2019 Jun;237:33-41.
Moderately lower risk of ovarian cancer among women with diabetes treated with metformin compared to no metformin in a meta-analysis of 5 observational studies²⁹⁴Lu MZ, Li DY, Wang XF. Effect of metformin use on the risk and prognosis of ovarian cancer: an updated systematic review and meta-analysis. Panminerva Medica. 2019 Jul 8.
24% lower risk of ovarian cancer among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 6 observational studies²⁹⁵Shi J, Liu B, Wang H, Zhang T, Yang L. Association of metformin use with ovarian cancer incidence and prognosis: a systematic review and meta-analysis. International Journal of Gynecological Cancer. 2019 Jan;29(1):140-146.

People without diabetes: no evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on risk of recurrence after chemotherapy among people with epithelial ovarian cancer without diabetes treated with metformin

No evidence of an effect on risk of recurrence (disease-free survival) after treatment with paclitaxel plus carboplatin among people with epithelial ovarian cancer without diabetes treated with metformin compared to chemotherapy alone in a small RCTrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects²⁹⁶Zheng Y, Zhu J, Zhang H, Liu Y, Sun H. Metformin plus first-line chemotherapy versus chemotherapy alone in the treatment of epithelial ovarian cancer: a prospective open-label pilot trial. Cancer Chemotherapy and Pharmacology. 2019 Dec;84(6):1349-1357.

Pancreatic cancer

Diabetic people with pancreatic cancer treated with metformin show a trend toward lower recurrence in a meta-analysis of two studies. We would expect equal or higher rates of recurrence among people with diabetes.

Diabetic people treated with metformin showed a lower risk of pancreatic cancer across many studies.

Recurrence: weak evidenceweak trends of effects (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of recurrence among mostly diabetic people with pancreatic cancer treated with metformin

A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward lower risk of recurrence (better disease-free survival) among people with pancreatic cancer (mostly with diabetes) treated with metformin compared to no metformin (not specific to diabetes) in a large meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 2 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies;²⁹⁷Wan G, Sun X et al. Survival benefit of metformin adjuvant treatment for pancreatic cancer patients: a systematic review and meta-analysis. Cellular Physiology and Biochemistry. 2018;49(3):837-847. our note: we would expect equal or higher rates of recurrence among people with diabetes

Cancer risk: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of pancreatic cancer among people with diabetes treated with metformin

Lower risk of pancreatic cancer among people with diabetes treated with metformin compared to no metformin in a meta-analyses of 9 observational studies²⁹⁸Hu H, Fong Y et al. Relationship of metformin with the risk of pancreatic cancer in patients with type 2 diabetes: a meta-analysis. Biomedical Research. 2017;28(10):4439-4444.
62% lower risk of pancreatic cancer among people with diabetes treated with metformin compared to no metformin in a large meta-analysis of 4 observational studies²⁹⁹Soranna D, Scotti L et al. Cancer risk associated with use of metformin and sulfonylurea in type 2 diabetes: a meta-analysis. Oncologist. 2012;17(6):813-22.
46% lower risk of pancreatic cancer among people, most of whom had diabetes, treated with metformin compared to no metformin in a meta-analysis of 7 observational studies³⁰⁰Zhang P, Li H, Tan X, Chen L, Wang S. Association of metformin use with cancer incidence and mortality: a meta-analysis. Cancer Epidemiology. 2013 Jun;37(3):207-18.
Lower risk of pancreatic cancer among people with type 2 diabetes treated with metformin compared to no metformin in a very large meta-analysis of 9 observational studies³⁰¹Franciosi M, Lucisano G et al. Metformin therapy and risk of cancer in patients with type 2 diabetes: systematic review. PLoS One. 2013 Aug 2;8(8):e71583.
A weak trend toward lower risk of pancreatic cancer among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 9 observational studies³⁰²Singh S, Singh PP et al. Anti-diabetic medications and risk of pancreatic cancer in patients with diabetes mellitus: a systematic review and meta-analysis. American Journal of Gastroenterology. 2013 Apr;108(4):510-9; quiz 520.

Prostate cancer

Diabetic people with prostate cancer treated with metformin showed lower rates of recurrence across many studies.

Metformin has shown insufficient (conflicting) evidence of an effect on risk of prostate cancer among diabetic people across many studies.

Recurrence: modest evidencesignificant effects in at least three small but well-designed randomized controlled trials (RCTs), or one or more well-designed, mid-sized clinical studies of reasonably good quality (RCTs or observational studies), or several small studies aggregated into a meta-analysis (this is the CancerChoices definition; other researchers and studies may define this differently) of lower risk of recurrence among people with prostate cancer and diabetes treated with metformin

26% lower risk of recurrence among people with prostate cancer and diabetes treated with metformin compared to no metformin in a very large meta-analysisa statistical analysis that combines the results of two or more research studies; the results of smaller research studies addressing the same or similar questions can be analyzed as though they are one bigger, more powerful study of 13 observationala type of study in which individuals are observed or certain outcomes are measured, but no attempt is made to affect the outcome (for example, no treatment is given); an example is a study that records people’s diets, but doesn’t try to alter their diets, and looks for patterns of disease or other outcomes related to different foods studies³⁰³Yao X, Liu H, Xu H. The impact of metformin use with survival outcomes in urologic cancers: a systematic review and meta-analysis. Biomed Research International. 2021 Oct 8;2021:5311828.
No evidence of an effectoverall, one or more studies did not demonstrate that a treatment or intervention led to an expected outcome; this does not always mean that there is no effect in clinical practice, but that the studies may have been underpowered (too few participants) or poorly designed. Larger, well-designed studies provide more confidence in making assessments on risk of recurrence among diabetic people with prostate cancer treated with metformin compared to no metformin in a meta-analysis of 3 observational studies³⁰⁴He K, Hu H et al. The effect of metformin therapy on incidence and prognosis in prostate cancer: a systematic review and meta-analysis. Scientific Reports. 2019 Feb 18;9(1):2218.
21% lower risk of biochemical recurrence among diabetic people with prostate cancer treated with metformin compared to no metformin in a meta-analysis of 9 observational studies³⁰⁵Stopsack KH, Ziehr DR, Rider JR, Giovannucci EL. Metformin and prostate cancer mortality: a meta-analysis. Cancer Causes and Control. 2016 Jan;27(1):105-13.
About 20% lower risk of recurrence among people with prostate cancer and type 2 diabetes treated with metformin compared to no metformin in a meta-analysis of 8 observational studies³⁰⁶Hwang IC, Park SM et al. Metformin association with lower prostate cancer recurrence in type 2 diabetes: a systematic review and meta-analysis. Asian Pacific Journal of Cancer Prevention. 2015;16(2):595-600.
26% lower risk of recurrence among diabetic people with prostate cancer treated with metformin compared to no metformin in a very large meta-analysis of 5 observational studies³⁰⁷Xiao Y, Zheng L et al. The impact of metformin use on survival in prostate cancer: a systematic review and meta-analysis. Oncotarget. 2017 Oct 31;8(59):100449-100458.
A weak trend toward lower risk of biochemical recurrence among people with prostate cancer (mostly with diabetes) treated with metformin compared to no metformin in a meta-analysis of 5 observational studies³⁰⁸Raval AD, Thakker D et al. Impact of metformin on clinical outcomes among men with prostate cancer: a systematic review and meta-analysis. Prostate Cancer and Prostatic Diseases. 2015 Jun;18(2):110-21.
19% lower risk of biochemical recurrence among people with prostate cancer (mostly with diabetes) treated with metformin compared to no metformin in a meta-analysis of 6 observational studies³⁰⁹Yu H, Yin L et al. Effect of metformin on cancer risk and treatment outcome of prostate cancer: a meta-analysis of epidemiological observational studies. PLoS ONE. 2014;9(12):e116327.

Cancer risk: insufficient evidencepreclinical evidence only OR clinical studies with such poor or unclear methodology that no conclusion can be drawn OR conflicting findings across clinical studies with no preponderance of evidence in one direction; conflicting evidence occurs when studies find conflicting effects (positive effect vs no effect or negative effect) with the same treatment and the same general study population (same cancer type, for example) (this is the CancerChoices definition; other researchers and studies may define this differently) of an effect on risk of prostate cancer among people with diabetes treated with metformin

No evidence of an effect on risk of prostate cancer among people with type 2 diabetes treated with metformin compared to no metformin in a very large meta-analysis of 18 observational studies³¹⁰Wang Y, Liu X et al. Effect of metformin on the risk of prostate cancer in patients with type 2 diabetes by considering different confounding factors: a meta-analysis of observational studies. European Journal of Cancer Prevention. 2020 Jan;29(1):42-52.
No evidence of an effect on risk of prostate cancer among mostly diabetic people treated with metformin compared to no metformin in a large meta-analysis of 18 observational studies, 16 of which analyzed risk among people with diabetes³¹¹Feng Z, Zhou X et al. Metformin use and prostate cancer risk: a meta-analysis of cohort studies. Medicine (Baltimore). 2019 Mar;98(12):e14955.
No evidence of an effect on risk of prostate cancer (not specific to diabetes) among people treated with metformin compared to no metformin in a very large meta-analysis of 11 observational studies³¹²Ghiasi B, Sarokhani D, Najafi F, Motedayen M, Dehkordi AH. The relationship between prostate cancer and metformin consumption: a systematic review and meta-analysis study. Current Pharmaceutical Design. 2019;25(9):1021-1029
No evidence of an effect on risk of prostate cancer among people with diabetes treated with metformin compared to no metformin in a meta-analysis of 9 observational studies³¹³He K, Hu H et al. The effect of metformin therapy on incidence and prognosis in prostate cancer: a systematic review and meta-analysis. Scientific Reports. 2019 Feb 18;9(1):2218.
No evidence of an effect on risk of prostate cancer among people with diabetes treated with metformin compared to no metformin in a very large meta-analysis of 24 observational studies and 2 RCTsrandomized controlled trial, a study design in which people are randomly assigned to either an experimental group or a control group to compare the outcomes from different treatments; an RCT is considered a strong design for determining a therapy’s effects³¹⁴Chen CB, Eskin M, Eurich DT, Majumdar SR, Johnson JA. Metformin, Asian ethnicity and risk of prostate cancer in type 2 diabetes: a systematic review and meta-analysis. BMC Cancer. 2018 Jan 10;18(1):65.
17% lower risk of recurrence among people with early stage prostate cancer (mostly with diabetes) treated with metformin compared to no metformin in a meta-analysis of 6 observational studies³¹⁵Coyle C, Cafferty FH, Vale C, Langley RE. Metformin as an adjuvant treatment for cancer: a systematic review and meta-analysis. Annals of Oncology. 2016 Dec;27(12):2184-2195.
9% lower risk of prostate cancer among people with diabetes treated with metformin compared to no metformin in a very large meta-analysis of 14 observational studies³¹⁶Yu H, Yin L et al. Effect of metformin on cancer risk and treatment outcome of prostate cancer: a meta-analysis of epidemiological observational studies. PLoS ONE. 2014;9(12):e116327.
No evidence of an effect on risk of prostate cancer among people with type 2 diabetes treated with metformin compared to no metformin in a very large meta-analysis of 8 observational studies³¹⁷Franciosi M, Lucisano G et al. Metformin therapy and risk of cancer in patients with type 2 diabetes: systematic review. PLoS One. 2013 Aug 2;8(8):e71583.
A weak trendan apparent change due to a therapy, close to but not achieving full statistical significance (this is the CancerChoices definition; other researchers and studies may define this differently) toward slightly lower risk of prostate cancer among people (not specific to diabetes) treated with metformin compared to no metformin in a meta-analysis of 8 observational studies³¹⁸Zhang P, Li H, Tan X, Chen L, Wang S. Association of metformin use with cancer incidence and mortality: a meta-analysis. Cancer Epidemiology. 2013 Jun;37(3):207-18.
No evidence of an effect on risk of prostate cancer among people with diabetes treated with metformin compared to no metformin in a large meta-analysis of 4 observational studies³¹⁹Soranna D, Scotti L et al. Cancer risk associated with use of metformin and sulfonylurea in type 2 diabetes: a meta-analysis. Oncologist. 2012;17(6):813-22.

Videos

Keep reading about metformin

See all our reviews of complementary therapies

Authors

Nancy Hepp, MS

past Lead Researcher

View profile

Sophie Kakarala

Research Assistant

View profile

Andrew Jackson, ND

Research Associate

View profile

Reviewers

Laura Pole, MSN, RN, OCNS

Senior Clinical Consultant

View profile

Nina Fuller-Shavel, MB, BChir, MA Hons, FBANT, IFMCP, DipIM, PG Cert RYT300

View profile

Last update: January 28, 2025

Last full literature review: March 2022

CancerChoices provides information about integrativein cancer care, a patient-centered approach combining the best of conventional care, self care, and evidence-informed complementary care in an integrated plan cancer care. We review complementaryin cancer care, complementary care involves the use of therapies intended to enhance or add to standard conventional treatments; examples include supplements, mind-body approaches such as yoga or psychosocialtherapy, and acupuncture therapies and self carelifestyle actions and behaviors that may impact cancer outcomes; examples include eating health-promoting foods, limiting alcohol, increasing physical activity, and managing stress practices to help patients and professionals explore and integrate the best combination of conventionalthe cancer care offered by conventionally trained physicians and most hospitals; examples are chemotherapy, surgery, and radiotherapy and complementary therapies and practices for each person.

Our staff have no financial conflicts of interest to declare. We receive no funds from any manufacturers or retailers gaining financial profit by promoting or discouraging therapies mentioned on this site.

Metformin

How can metformin help you? What the research says

Commentary on research methodology

Improving treatment outcomes

Optimizing your body terrain

Managing side effects and promoting wellness

Reducing cancer risk

Videos

Keep reading about metformin

Authors

Nancy Hepp, MS

Sophie Kakarala

Andrew Jackson, ND

Reviewers

Laura Pole, MSN, RN, OCNS

Nina Fuller-Shavel, MB, BChir, MA Hons, FBANT, IFMCP, DipIM, PG Cert RYT300

Learn more

Optimizing Your Body Terrain

Astragalus

Self Care