Beta blockers and cancer prognosis - The role of immortal time bias: A systematic review and meta-analysis

Background: Findings from experimental and observational studies have suggested beneficial effects of beta blocker (BB) use on cancer survival. Nevertheless, results have been inconclusive and there have been concerns that the observed associations might have resulted from immortal time bias (ITB). We conducted a systematic review and meta-analysis to summarize existing evidence, paying particular attention to this potential source of bias. Methods: A systematic literature search was performed in PubMed and Web of Science. Studies investigating the association between BB use and overall or cancer-specific survival were included. Summary estimates were derived from meta-analyses using random effects models. The potential influence of ITB was investigated. Results: We identified 30 eligible studies including 88,026 cancer patients in total. We deemed 11 studies to be at high or unclear risk of ITB. Including all studies in the meta-analysis, BB users had a significantly better overall (hazard ratio (HR) 0.88, 95% CI 0.79-0.97) and cancer-specific (HR 0.75, 95% CI 0.64-0.88) survival. Excluding the studies deemed to be prone to ITB resulted in HRs (95% CIs) of 1.00 (0.93-1.07) and 0.90 (0.83-0.98), respectively. Analyses on cancer site and BB type did not show beneficial associations besides overall survival among melanoma patients. However, melanoma-specific survival was not improved. Conclusion: We found no clinically meaningful evidence for an association between BB use and survival after excluding studies with a possible ITB. Our results support suggestions that the proposed beneficial effect of BBs on cancer survival might be based on ITB. (C) 2016 Elsevier Ltd. All rights reserved.