Safety Analysis Favors Olaparib in Meta-Analysis Comparing PARP Inhibitors in Ovarian Cancer

The results of a network meta-analysis, which compared efficacy and safety findings from studies of 3 FDA-approved PARP inhibitors as maintenance therapy for patients with relapsed platinum-sensitive ovarian cancer, demonstrated that similar efficacy results were seen across the 3 PARP inhibitors, but safety results differed.

Nonsignificant hazard ratios (HRs) for progression-free survival (PFS) emerged from efficacy comparisons of 3 FDA-approved PARP inhibitors: olaparib (Lynparza), niraparib (Zejula), and rucaparib (Rubraca). However, analyses of odds ratios (ORs) for grade 3/4 adverse events (AEs) demonstrated significant 60% to 70% reductions in favor of olaparib versus the other 2 PARP inhibitors, as well as similarly lower ORs for treatment interruption, as reported at the 2018 Society of Gynecologic Oncology meeting in New Orleans.

“This network meta-analysis allowed comparisons of efficacy and adverse events by careful selection of clinical trials of similar design and that included similar types of patients,” said Alfred Sackeyfio, a health economist and epidemiologist for AstraZeneca in Cambridge, England. “A network meta-analysis allows comparative analyses of interventions that have not been directly compared in randomized clinical trials.”

“The data available for this analysis showed no difference in efficacy among the 3 PARP inhibitors but demonstrated a more favorable safety profile for olaparib, associated with a reduced odds [ratio] of grade 3 or greater adverse events and treatment interruption,” he added. “No significant safety differences were observed between rucaparib and niraparib.”

The data will also be analyzed to determine whether the differences in AE profiles are driven by specific types of events and toxicities, said Sackeyfio.

The 3 approved PARP inhibitors received indications as maintenance therapy for relapsed, platinum-sensitive ovarian cancer on the basis of individual placebo-controlled trials: SOLO2 with olaparib, NOVA with niraparib, and ARIEL3 with rucaparib. All 3 trials had median PFS as the primary endpoint, and the results demonstrated PFS HRs of 0.20 to 0.30 in favor of PARP maintenance.

The incidence of grade 3/4 AEs in the 3 placebo-controlled trials ranged from 37% to 74%, treatment interruption from 45% to 66.5%, and dose reduction from 25% to 69%. Rates of all 3 safety outcomes were lowest with olaparib among the 3 PARP inhibitors.

For the meta-analysis, Sackeyfio and colleagues performed a systematic literature review that identified 4 studies for potential inclusion: the 3 phase III, placebo-controlled trials, plus Study 19, a phase II evaluation of maintenance therapy with olaparib in patients with relapsed platinum-sensitive ovarian cancer. The Bayesian network meta-analysis included only the phase III trials.

Baseline characteristics of patients enrolled in the 3 trials were similar, including performance status, degree of platinum sensitivity, previous exposure to bevacizumab (Avastin), best response to most recent therapy, and number of previous lines of chemotherapy.

Analyses of investigator-assessed PFS in patients withBRCA-mutant ovarian canceryielded no significant differences between agents: olaparib versus niraparib (HR, 1.11; 95% CI, 0.67-1.84), olaparib versus rucaparib (HR, 1.30; 95% CI, 0.78-2.12), or rucaparib versus niraparib (HR, 0.85; 95% CI, 0.50-1.48). Comparisons of PFS by blinded independent review committee also showed no significant differences: olaparib versus niraparib (HR, 0.93; 95% CI, 0.58-1.60), olaparib versus rucaparib (HR, 1.25; 95% CI, 0.71-2.18), or rucaparib versus niraparib (HR, 0.74; 95% CI, 0.40-1.40).

The safety analyses yielded statistically significant or numerical advantages in favor of olaparib. For the comparison of olaparib and niraparib, grade 3/4 AEs occurred significantly less often with olaparib (OR, 0.29; 95% CI, 0.14-0.57) as did treatment interruption (OR, 0.33; 95% CI, 0.14-0.63). The analysis of dose reduction produced an OR of 0.38 in favor of olaparib but with overlapping confidence intervals (95% CI, 0.07-1.45).

The safety comparison of olaparib and rucaparib produced statistically significant ORs of 0.37 (95% CI, 0.17-0.74) for grade 3/4 AEs and 0.24 (95% CI, 0.11-0.51) for treatment interruption and a numerically superior OR of 0.61 (95% CI, 0.11-2.17) for dose reduction, all favoring olaparib.

The comparison of rucaparib and niraparib yielded mixed results and no significant differences between the 2 agents with respect to grade 3/4 AEs (OR, 0.84; 95% CI, 0.43-1.43), treatment interruption (OR, 1.43; 95% CI, 0.67-2.66), or dose reduction (OR, 0.73; 95% CI, 0.22-1.96).

The safety analyses included all patients enrolled in each of the trials, and they will be updated when theBRCA­-mutant populations in NOVA and ARIEL3 have been identified, said Sackeyfio.

Reference:

Sackeyfio A, Nussey F, Friedlander M, Pujade-Lauraine E. Comparative efficacy and tolerability of the PARP inhibitors olaparib 300 mg tablets BID, niraparib 300 mg capsules QD and rucaparib 600 mg tablets BID as maintenance treatment in BRCA-mutated (BRCAm) platinum-sensitive relapsed ovarian cancer (PSROC). Presented at: 2018 SGO Annual Meeting; March 24-27, 2018; New Orleans, LA. Abstract 10743.