Latest Research Exploring Metformin Connection To Breast Cancer
Overview
This study aimed to investigate the relationship between metformin and other antidiabetic medications and the incidence of breast cancer in two large prospective cohort studies. It included 185,181 women from the Nurses’ Health Study (NHS; 1994–2016) and NHSII (1995–2017). Information on type 2 diabetes (T2D) diagnosis, antidiabetic medications, and other covariates was self-reported at baseline and updated biennially via follow-up questionnaires. Breast cancer cases were also self-reported and confirmed through medical record review. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using Cox proportional hazards regression models, adjusted for breast cancer risk factors. Over 3,324,881 person-years of follow-up, there were 9,192 incident invasive breast cancer cases, 451 of which were among women with T2D. Compared to women without T2D, neither metformin use (HR = 0.97; 95% CI = 0.81–1.15) nor the use of other antidiabetic medications (HR = 1.11; 95% CI = 0.90–1.36) was associated with a significantly reduced incidence of breast cancer. Among women with T2D, metformin use did not show a significant inverse association with breast cancer (HR = 0.92; 95% CI = 0.74–1.15). While past metformin use was inversely associated with breast cancer in the T2D population (HR = 0.67; 95% CI = 0.48–0.94), current use (HR = 1.01; 95% CI = 0.80–1.27) and longer duration of use did not show a significant association (each 2-year interval: HR = 1.01; 95% CI = 0.95–1.07). Overall, metformin use was not linked to breast cancer risk in the general cohort or among women with T2D.
The study concludes that metformin use does not significantly impact breast cancer risk in the overall population or in women with T2D, suggesting earlier studies may have been influenced by biases and that metformin is not a strong candidate for breast cancer prevention.
Introduction
Type 2 diabetes (T2D) is linked to an elevated risk of breast cancer. Metformin, a commonly used antidiabetic medication approved in the US in 1994 and recommended as first-line therapy by the American Diabetes Association (ADA), has garnered attention for its potential chemopreventive effects against breast cancer and other cancers. Proposed mechanisms include lowering circulating insulin levels, activating AMP-activated protein kinase, and inhibiting the mammalian target of rapamycin (mTOR) signaling pathway.
Also read: Finerenone Therapy In CKD and Type 2 Diabetes
Early epidemiological studies, primarily conducted before 2012, indicated a lower incidence of breast cancer among metformin users compared to non-users. However, recent studies have provided inconsistent results, with many showing no significant association between metformin use and breast cancer incidence. Notably, most of these studies were conducted in populations with T2D, and only a few large epidemiological studies have included both T2D and non-T2D populations, making it challenging to assess the effect of metformin in the general population and its interaction with T2D status. Additionally, the duration and recency of metformin use have been underexplored, and many studies did not adequately account for reproductive history or time-varying confounders, often having limited follow-up periods (median <10 years). Moreover, the potential variation in the metformin-breast cancer association by estrogen receptor (ER) status and cancer stage remains largely uninvestigated, and many studies may have suffered from biases such as immortal time bias.
This study aims to examine the association between the use of metformin and other antidiabetic medications and breast cancer incidence within two large epidemiological cohort studies: the Nurses’ Health Study (NHS) and NHSII. Analyses were conducted both across the entire cohort, including participants with and without T2D, and specifically among participants diagnosed with T2D.
Method
Inclusion Criteria:
- Participants: Women enrolled in the Nurses’ Health Study (NHS) and NHSII.
- Time Frame: Baseline set at 1994 for NHS and 1995 for NHSII, aligned with the approval of metformin for type 2 diabetes (T2D) treatment in the United States.
- Diagnosis and Treatment Data:
– Self-reported T2D diagnosis confirmed through supplementary questionnaires.
– Detailed information on metformin and other diabetes medication use obtained from disease confirmation questionnaires, biennial follow-up questionnaires, and supplemental questionnaires.
- Population:
– Full cohort analysis includes all women from NHS and NHSII without prior cancer diagnosis (excluding non-melanoma skin cancer).
– T2D-specific analysis includes only women with confirmed T2D diagnosis who provided information on metformin use.
- Covariate Data: Collection of demographic characteristics, reproductive history, medical history, smoking history, weight, height, physical activity, and neighborhood socioeconomic status (nSES).
Exclusion Criteria:
- Prior Cancer Diagnosis: Exclusion of participants diagnosed with cancer (excluding non-melanoma skin cancer) before the baseline date.
- Incomplete Data: Exclusion of participants without complete data on T2D diagnosis, metformin use, and other relevant covariates.
- Non-Respondents: Participants who did not respond to biennial follow-up questionnaires or supplementary questionnaires on T2D and metformin use.
- Missing Duration Data: Participants with missing metformin duration data not imputed through available methods (i.e., no updates available from any source within three consecutive cycles).
This rigorous approach ensures the integrity and reliability of the study, focusing on well-documented and consistently followed participants to assess the association between metformin use and breast cancer incidence. You can also read White Matter Brain Changes In Post-Chemotherapy Breast Cancer Patients
Analysis
Person-time of follow-up was calculated from the baseline date, corresponding to the initial metformin assessment, until breast cancer diagnosis, death, or the end of the study follow-up (June 1, 2016, for NHS and June 1, 2017, for NHSII), whichever occurred first. This prospective cohort study design helped mitigate time-window bias, as all participants were followed over a comparable period, ensuring similar exposure opportunities to anti-diabetic medication.
We used multivariable Cox proportional hazards regression models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). For the full cohort analysis, age was used as the underlying time scale. For the subset of women with confirmed T2D, time since T2D diagnosis was used as the underlying time scale to account for left truncation or immortal time bias. All analyses were stratified by cohort and 2-year follow-up intervals.
Three main models were fitted, and subgroup analyses were conducted based on breast cancer estrogen receptor (ER) status, tumor stage, weight change since age 18, physical activity, menopausal status, and age at diabetes diagnosis (for T2D cases only). The potential effect modification of metformin status and duration was tested using a likelihood ratio test. To address time-lag bias, we applied 2-, 4-, and 6-year lags to metformin status and conducted latency estimation analysis using a linear damped exponential weighting model. To reduce screening and surveillance bias, the main analyses were repeated excluding participants who did not report breast cancer screening in the last cycle (n = 15,120), adjusting for the number of questionnaires returned, and applying an inverse probability weighted model for screening mammography.
All statistical analyses were performed using SAS software version 9.4 (SAS Institute Inc., Cary, North Carolina, USA). Statistical significance was determined by p-values < 0.05, and all tests were two-sided.
Results
In two large cohorts, 15,918 women with type 2 diabetes (T2D) were identified throughout the follow-up period, accounting for 8.7% of the participants. Compared to non-diabetic women, those with T2D were older, had a lower neighborhood socioeconomic status (nSES), more weight change since age 18, an earlier age at menarche, lower physical activity levels, and were more likely to never have breastfed and be past users of oral contraceptives. Conversely, they were less likely to be current users of menopausal hormone therapy (MHT) or current smokers, had lower alcohol intake, and were more likely to use aspirin.
During 3,324,881 person-years of follow-up, 9,192 cases of invasive breast cancer were ascertained, including 451 cases among women with T2D. Compared to non-diabetic women, no statistically significant association was found between metformin use and breast cancer incidence (HR = 0.97; 95% CI = 0.81–1.15). Non-metformin diabetes medications also showed no significant association with breast cancer (HR = 1.11; 95% CI = 0.90–1.36). However, past use of metformin was inversely associated with breast cancer incidence (HR = 0.73; 95% CI = 0.54–1.00), but no association was found among current users (HR = 1.02; 95% CI = 0.85–1.22). Longer duration of metformin use did not correlate with a lower risk of breast cancer (HR per 2-year increment = 0.98; 95% CI = 0.93–1.03, p-trend = 0.41).
No significant associations were observed for other diabetes medication categories, although there was a suggestive positive association with insulin use (HR = 1.20; 95% CI = 0.82–1.75). When insulin use was used as the reference group, an inverse but non-significant association was found between metformin use only and breast cancer incidence (HR = 0.91; 95% CI = 0.63–1.30). Similar results were obtained after applying 2-, 4-, and 6-year lag periods for exposure and no heterogeneity was found between prevalent versus incident T2D cases.
Subgroup analyses showed no significant heterogeneity in breast cancer associations based on tumor estrogen receptor (ER) status, stage, weight change since age 18, and physical activity level. Among women with T2D, no significant association with breast cancer was found for ever use of metformin compared to never users (HR = 0.92; 95% CI = 0.74–1.15). Past metformin users, however, showed a lower risk of breast cancer (HR = 0.67; 95% CI = 0.48–0.94). No significant associations were found between the duration of metformin use and breast cancer risk (HR per 2-year increment = 1.01; 95% CI = 0.95–1.07, p-trend = 0.92).
Overall, the linear damped exponential model for latency estimation indicated no significant association between metformin status (ever vs. never) and breast cancer incidence (HR = 1.04; 95% CI = 0.96–1.13), and no significant latency effect (β = 0.05 ± 0.04, p = 0.24). Additional adjustments for glucose levels, other anti-diabetic medications, and number of questionnaires returned, excluding those without recent mammogram screening, and weighting by inverse probability of screening mammography, did not substantially alter these results.
Conclusion
The comprehensive analysis from two large cohorts, including 15,918 women with type 2 diabetes (T2D), reveals that metformin use is not significantly associated with breast cancer incidence. While past use of metformin showed an inverse association with breast cancer risk, current use and longer duration of metformin use did not demonstrate significant protective effects. The study found no significant associations between other anti-diabetic medications and breast cancer risk, though there was a suggestive positive association with insulin use. Subgroup analyses based on various factors, such as tumor estrogen receptor (ER) status, tumor stage, weight change since age 18, and physical activity level, also did not reveal significant heterogeneity in the association between metformin use and breast cancer risk.
Overall, the findings indicate that metformin may not play a significant role in breast cancer prevention among women with or without T2D. Further, the potential for time-lag bias and other confounding factors was considered, but adjustments did not substantially change the outcomes. Thus, while metformin remains a critical drug for managing T2D, its role in reducing breast cancer risk appears limited based on the current evidence from these large cohort studies.
Oncology Related Tools
Other
Latest Research
