Comparative tolerability and dose modifications of poly(ADP-ribose) polymerase inhibitors in ovarian cancer: a retrospective cohort study of US healthcare claims data

Abstract

<h3>Objectives:</h3> Several poly(ADP-ribose) polymerase inhibitors (PARPis) are approved as maintenance therapy for ovarian cancer (OC); however, real-world evidence focusing on comparative tolerability and dose modifications is lacking. In clinical trials, the most common adverse events (AEs) leading to dose modifications were hematologic toxicities. We characterized clinical events of interest (CEIs) and dose modifications in a large population of patients with OC receiving PARPi therapy in the USA. <h3>Methods:</h3> A retrospective observational study utilizing claims data from US MarketScan Commercial and Medicare Supplemental databases was performed. Women with OC who initiated a PARPi (olaparib, niraparib or rucaparib) between 01/01/2017 and 05/31/2019 were included. Baseline characteristics were measured during the 6 months prior to starting PARPi (pre-index period). Patients were observed from first outpatient PARPi prescription (index date) until ≥30 days' follow-up. Sixteen pre-defined CEIs (Figure) were identified from claims data based on ICD-9/10 code criteria. Descriptive incidence of CEIs and dose modifications are reported during index PARPi treatment; odds ratios (ORs) were calculated to compare the likelihood of experiencing CEIs. Persistence was defined as having no PARPi treatment gaps of >90 days in patients with ≥6 months' continuous enrollment. <h3>Results:</h3> In total, 813 PARPi-treated patients with OC were included; 303, 348 and 162 received olaparib, niraparib and rucaparib, respectively. At baseline, patient characteristics and rates of CEI were similar across all groups (mean age: 58.3-59.4 years; commercially insured: 77.9-83.3%; any CEI: 88.9-94.1%; mean NCI-adapted Charlson comorbidity index: 0.7-0.9; systemic chemotherapy ever in patient record: 88.8-90.5%; median [IQR] follow-up: 7.4-9.7 [9.0-11.1] months). Overall, 89.4%, 69.3% and 93.2% of patients receiving olaparib, niraparib and rucaparib, respectively, started at the highest indicated dose. During follow-up, risk of experiencing any CEI was significantly higher with niraparib vs olaparib (OR 3.23 [95% CI 1.89-5.50], <i>P</i><0.001) and rucaparib (2.07 [1.08-3.97], <i>P</i><0.05), with no significant difference between rucaparib and olaparib (1.56 [0.89-2.74], <i>P</i>=0.1). A similar pattern was observed with hematologic CEIs (Figure). PARPi dose decreases were observed in 21.1%, 35.1% and 30.2% of olaparib-, niraparib- and rucaparib-treated patients, respectively. The proportion of patients persisting on index PARPi was significantly higher (<i>P</i><0.05) with olaparib (62.2%) vs niraparib (35.9%) and rucaparib (48.7%). <h3>Conclusions:</h3> To our knowledge, this is the largest real-world comparison of PARPi therapy in women with OC. These results suggest differences in the risk of experiencing a CEI, likelihood of dose modifications and ability to receive continuous PARPi therapy between the licensed PARPis. Further research is required to assess the influence of medication adherence on long-term effectiveness and how best to support patients on PARP inhibitors.