Safety of PARP inhibitors as maintenance therapy in ovarian cancer

ABSTRACTIntroduction Antiangiogenic agents and poly(ADP-ribose) polymerase inhibitors (PARP-Is) have improved the outcome of patients suffering from ovarian cancer. However, as they are associated with many adverse events (AEs), it is important to be aware of their safety and toxicity profiles.Areas covered We reviewed PARP-I therapeutical indications, mechanism of action, metabolism, and interactions. We reported on all major and minor AEs that have emerged from clinical trials (SOLO1, PRIMA, PAOLA1, ATHENA, SOLO2, NOVA, ARIEL3, NORA), their follow-ups, meta-analyses, and real-world studies, particularly hematologic toxicities and their management, and secondary malignancies (myelodysplastic syndrome and acute myeloid leukemia). We also addressed gastrointestinal, neurological, respiratory, hepatic, and renal toxicity and the use of PARP-Is in older, pregnant, and lactating patients. No specific research strategy in terms of keywords, inclusive dates and databases was used.Expert opinion PARP-Is benefits largely outweigh the risks associated with potential AEs. Randomized controlled trials produced strong good, quality data, but they enrolled a selected population and failed to capture rare events. More pharmacovigilance data and real-life studies on a larger and more heterogeneous sample are needed to understand PARP-Is differences and to clarify the incidence of late AEs to balance the risk/benefit ratio.KEYWORDS: PARP inhibitorovarian cancerolaparibniraparibrucaparibadverse eventsmyelodysplastic syndromeacute myeloid leukemia Article highlights The introduction of PARP-Is in the treatment of ovarian cancer (olaparib, niraparib, rucaparib) is modifying the natural history of the disease.Most patients with high-grade ovarian cancer receive PARP-Is in the front-line setting, while their use in the recurrence is more limited.The most common PARP-I adverse events include fatigue, nausea, and hematological toxicity (anemia, thrombocytopenia, neutropenia).Other adverse events include gastrointestinal (constipation, vomiting, decreased appetite, abdominal pain, diarrhea, dyspepsia, dysgeusia), neurological (dizziness, headache), and respiratory (dyspnea, cough) toxicities.PARP-Is also increase the risk of myelodysplastic syndrome and acute myeloid leukemia versus placebo treatment.The risk of these secondary malignancies is lower in the first line than in recurrence, although the follow-ups available for upfront trials are shorter and need to be updated.Adverse events are quite similar, and most of them are manageable with dose reductions and dose interruptions.PARP-I differences in toxicity profiles and cytochromes interaction could allow for a tailored approach, especially in the older population.There is a need to build trials with PARP-Is duration as the main focus and to update current phase III trial results with long-term adverse events.Declaration of interestsSC Cecere has received honoraria from MSD, AZ, GSK, and Clovis. SP has received honoraria from MSD, AZ, GSK, Clovis, and research funding from MSD, AZ. C Pisano has received honoraria from MSD, AZ, GSK, and Clovis. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.Reviewer disclosuresPeer reviewers on this manuscript have no relevant financial or other relationships to disclose.Author contribution statementReview conception and design: SCC, SP, CP; collection and interpretation of data: all authors; manuscript drafting: Fabio Perversi (medical writer); manuscript editing: all authors. All authors gave their approval to submit.AcknowledgmentsThe authors wish to acknowledge Fabio Perversi (Polistudium Srl, Milan, Italy) for medical writing and Valentina Attanasio and Aashni Shah (Polistudium Srl, Milan, Italy) for linguistic and editorial assistance. Medical writing and editorial assistance were supported by SP’s internal research funds.Additional informationFundingThis paper was not funded.