Safety profile of poly (ADP-ribose) polymerase (PARP) inhibitors in cancer: a network meta-analysis of randomized controlled trials

Background: Poly (ADP-ribose) polymerase (PARP) inhibitors, which are among the most important breakthroughs in precision medicine, have played a crucial role in cancer treatment. Understanding the toxicity profiles of the different PARP inhibitors will improve strategic treatment in clinical practice. Methods: PubMed, Cochrane Library, and Web of Science were systematically searched to include related studies published in English between January 2009 and February 2020. Only prospective, phase II and III randomized controlled trials were included. The following treatment groups were analyzed: niraparib, talazoparib, olaparib, rucaparib, conventional therapy (chemotherapy), one PARP inhibitor with one angiogenesis inhibitor, and placebo. Baseline data and adverse event data were extracted from the Bayesian random-effects network meta-analysis. Results: Fourteen phase II and III randomized controlled trials (4,336 patients) were included. When considering grade 3-5 adverse events, olaparib may be a better choice (probability =57%), followed by conventional therapy (50%), talazoparib (45%), rucaparib (75%), niraparib (77%), and a PARP inhibitor with one angiogenesis inhibitor (94%). Niraparib and rucaparib had higher risks for hematological and gastrointestinal toxicities, respectively. Talazoparib was safer for gastrointestinal function. Constipation and neutropenia were less observed in olaparib, but the risks for anorexia increased. The combination of PARP inhibitor and angiogenesis inhibitor increased the risk of general, metabolic, and gastrointestinal disorders. Conclusions: This network meta-analysis suggested that the toxicity spectrum of each PARP inhibitor is different. Olaparib had the best safety profile among all PARP inhibitors because of its mild toxicity and narrow spectrum. This study may guide clinicians and support further research.