Type 2 Diabetes and Risk of Prostate Cancer

A Meta-analysis of Observational Studies

D Bansal; A Bhansali; G Kapil; K Undela; P Tiwari


Prostate Cancer Prostatic Dis. 2013;16(2):151-158. 

In This Article

Abstract and Introduction


Background Emerging evidence suggests that diabetes may increase the risk of cancers. However, available evidence on prostate cancer is conflicting. We therefore examined the association between Type 2 diabetes and risk of prostate cancer by conducting a detailed meta-analysis of all studies published regarding this subject.

Methods PubMed database and bibliographies of retrieved articles were searched for epidemiological studies (published between 1970 and 2011), investigating the relationship between Type 2 diabetes and prostate cancer. Pooled risk ratio (RR) was calculated using random-effects model. Subgroup, sensitivity analysis and cumulative meta-analysis were also done.

Results Forty-five studies (29 cohort and 16 case–control studies) involving 8.1 million participants and 132 331 prostate cancer cases detected a significant inverse association between Type 2 diabetes and risk of prostate cancer (RR 0.86, 95% confidence interval (CI) 0.80–0.92). For cohort studies alone, the RR was 0.87 (95% CI 0.80–0.94), and for case–control studies alone, the RR was 0.85 (95% CI 0.74–0.96). Sensitivity analysis done by excluding one outlier further strengthened our negative association (RR 0.83, 95% CI 0.78–0.87). No evidence of publication bias was observed.

Conclusions This meta-analysis provides strongest evidence supporting that Type 2 diabetes is significantly inversely associated with risk of developing prostate cancer.


Type 2 diabetes has been associated with an increased risk of several cancers including pancreas, liver, breast, colorectal, urinary tract and female reproductive organs.[1] There is inconsistency in the reporting of prostate cancer risk in patients with Type 2 diabetes. There are two meta-analyses done earlier on the same subject by Bonovas et al.[2] and Kasper et al.,[3] reporting a statistically significant decrease in the risk of prostate cancer by 9% and 16%, respectively.

The most recent analysis done by Kasper et al. included 19 studies having 1 million population, including 20 373 prostate cancer cases. Twenty-six studies with conflicting results have been published after that meta-analysis, which raised the interest of adding new evidence to the previous analysis. Although this meta-analysis is an update of two previous meta-analysis, we have included studies published after 2006 (since the last analysis), which have shown conflicting results including decreased risk,[4–17] increased risk[18–24] and some also reported no probable association.[25–27] We also performed an additional subgroup analysis of impact of ethnic variation and duration of diabetes with risk of prostate cancer. The strength of the present study lies in inclusion of data of large scale cohorts (sample size >30 000) having less chance of observational and recall bias thus population contributing to the present analysis is more than 8.1 million, including 132 331 prostate cancer cases.

Research conducted in recent years has also elucidated genetic linkage between Type 2 diabetes mellitus and prostate cancer.[28–30] The decreased risk of prostate cancer in patients with Type 2 diabetes cannot clearly be explained, but evidence suggests that metabolic factors associated with Type 2 diabetes has an impact on hypothalamic–pituitary–testicular axis, thereby resulting in hypogonadism.[1] However, the chance of detection bias before and after the advent of PSA testing cannot be ignored. The implementation of PSA testing may have changed the scope of disease that was diagnosed as prostate cancer. With the additional studies done in the PSA testing era, we are now able to do a subgroup analysis of men diagnosed before and after this time. The present study analyzed the risk of prostate cancer in patients with Type 2 diabetes along with subgroup analysis to establish possible variation of risk in different subgroups.