Pioglitazone use and risk of bladder cancer: a systematic literature review and meta-analysis of observational studies

• Published in: Diabetology international 2019-01, Vol.10 (1), p.24-36
• Main Authors: Mehtälä, Juha, Khanfir, Houssem, Bennett, Dimitri, Ye, Yizhou, Korhonen, Pasi, Hoti, Fabian
• Format: Article
• Language: English
• Subjects: Bayesian analysis; Bladder cancer; Cancer; Diabetes; Diabetes mellitus; Diabetes mellitus (non-insulin dependent); Endocrinology; Health risk assessment; Literature reviews; Medicine; Medicine & Public Health; Meta-analysis; Metabolic Diseases; Observational studies; Original; Original Article; Pioglitazone; Sensitivity analysis; Statistical analysis; Systematic review
• ISSN: 2190-1678; 2190-1686
• DOI: 10.1007/s13340-018-0360-4

Background Studies investigating bladder cancer risk in pioglitazone-treated type 2 diabetes mellitus patients report conflicting results. Previous meta-analyses on this topic utilized publications prior to 2013. More long-term observational studies have been published since then. We reviewed the accumulated evidence and updated findings from previous meta-analyses. Methods This meta-analysis was based on a systematic review of peer-reviewed observational studies published prior to September 30, 2016. Eligible studies were identified using a specified MEDLINE search. References from included studies and from previous meta-analyses were screened for additional records. Meta-analysis hazards ratios were derived using a random-effects model. Several sensitivity analyses including hierarchical Bayesian meta-analysis with country-specific effects were conducted. Results Of 363 identified records, 23 studies were included in this review and 18 in the actual meta-analyses. For bladder cancer outcome, the estimated effect size for ever vs. never use of pioglitazone was 1.16 [95% confidence interval (CI), 1.04–1.28]. In the cumulative dose and duration analyses, highest effect was observed in the highest/longest exposure group, but substantial heterogeneity was present. In the sensitivity analysis, only studies adjusted for lifestyle-related factors were included and the frequentist effect size was 1.18 (95% CI, 1.00–1.40, p  = 0.054). However, the risk was not verified in the Bayesian framework with an effect size of 1.17 [95% credible interval (CrI), 0.94–1.54]. Conclusions In line with previous meta-analyses, we observed a small but statistically significant association between ever (vs. never) use of pioglitazone and bladder cancer risk; however, causality is not established and alternative explanations cannot be ruled out.