Comparative effectiveness and safety of first-line treatments of metastatic colorectal cancer in real-world settings: a systematic review and meta-analysis

Background: Comparative effectiveness analysis is critical for informed decision-making in the context of metastatic colorectal cancer (mCRC) which involves multiple treatment strategies. Randomized clinical trials are not always aligned with clinical practice, and greater use of real-world (RW) studies can inform health care decision by providing results that reflect RW practice. Aims and objectives: The aim of this systematic review and meta-analysis was to provide a synthesis of the available RW evidence on the effectiveness and safety of first-line systemic anti-cancer therapies (SACTs) in patients with mCRC. Methods: Relevant databases were searched from inception until July 2021. Inclusion criteria were: observational studies; published in English; patients ≥ 18 years; unresectable mCRC; receiving a first-line SACT for treatment of mCRC. The primary effectiveness outcome was overall survival. Safety was assessed by the occurrence of severe adverse events. The results were synthesized using the random-effect meta-analysis model based on hazard ratio and 95% confidence interval (95% CI) for survival outcomes, while risk ratio and 95% CI was used for safety outcomes. Subgroup analysis was performed to explore differences between different treatment strategies. Heterogeneity was assessed using I2 statistic.
Results: The search strategy identified 5662 studies of which 31 met the inclusion criteria and were included in the effectiveness meta-analysis. The pooled hazard ratio for overall survival including all SACTs was 1.19 (95% CI 1.1-1.29). The overall heterogeneity of included studies was 76.6%. Subgroup analysis identified a significant difference between different treatment comparisons (p =0.01). The pooled overall survival was significant for 2 subgroups: (1) chemotherapy only versus Bevacizumab+ chemotherapy and (2) Bevacizumab+ FOLFIRI versus Bevacizumab+ XELIRI (pooled estimate: 1.15 (1.05-1.26), 1.36 (1.19-1.55), respectively). For the safety outcomes, 14 studies were included in the meta-analysis. The pooled relative risk of haematological and non-haematological toxicities was 1.25 (0.89-1.76), 1.07 (0.73-1.54), respectively with no statistically significant difference between different treatment strategies (p > 0.05). The overall heterogeneity of included studies for both haematological and non-haematological toxicities was 85.5%. Conclusion: The results indicated a survival benefit for bevacizumab with no evidence of additional toxicity when compared to the other regimens used in first-line settings. Although this benefit may appear clinically modest, bevacizumab offers hope for increased survival for patients with mCRC.