Results of the phase 3 ARIEL4 trial comparing rucaparib vs chemotherapy for patients with relapsed ovarian cancer and deleterious BRCA1/2 mutations were presented at 2022 ESMO and raised questions about optimal sequencing of PARP inhibitors in this setting.
Updated results of the phase 3 ARIEL4 trial (NCT02855944) that were presented at the 2022 European Society for Medical Oncology Congress indicated that patients with relapsed ovarian cancer who harbored a deleterious BRCA1/2 mutation and were sensitive to platinum-based therapy had better progression-free survival (PFS) and similar overall survival (OS) outcomes with rucaparib (Rubraca) vs chemotherapy.1
In the intention-to-treat (ITT) population, median OS favored the control arm with medians of 19.4 months (15.2-23.6) in the rucaparib (Rubraca) group (n = 233) vs 25.4 months (95% CI, 21.4-27.6) in the chemotherapy group (n = 116; HR, 1.313; 95% CI, 0.999-1.725). The difference in OS was driven by the subgroup of patients with platinum-resistant disease, with medians of 14.2 months (95% CI, 11.8-17.4) and 22.2 months (95% CI, 15.4-26.2) with rucaparib and chemotherapy, respectively (HR, 1.511; 95% CI, 1.053-2.170). In the patients with platinum sensitivity, the median OS was 29.4 months (95% CI, 23.1-37.4) with rucaparib vs 27.6 months (95% CI, 21.9-47.2) with chemotherapy (HR, 1.071; 95% CI, 0.709-1.618).
“The overall survival [results were] confounded by the high rate of crossover from chemotherapy to rucaparib which was built in [to the trial and resulted in] 90% of patients receiving rucaparib,” Amit Oza, MD, BSc, MBBS, FRCPC, head of the Division of Medical Oncology & Hematology and medical director of the Cancer Clinical Research Unit at Princess Margaret (PM) Cancer Centre in Toronto, Canada, as well as co-leader of the Ovarian Cancer Translational Research Initiative, said during a presentation of the results. “Additionally, 42% of patients in the rucaparib arm did not receive subsequent anticancer treatment, and we need to understand this pattern.”
The initial read-out of the trial that was published in Lancet Oncology earlier this year showed that median PFS was significantly improved with rucaparib vs chemotherapy in the efficacy population, or those harboring deleterious BRCA1/2 mutations without reversion mutations, at 7.4 months (95% CI, 7.3-9.1) vs 5.7 months (95% CI, 5.5-7.3), respectively (HR, 0.64; 95% CI, 0.49-0.84; P = .001). Similar PFS effect was noted in the ITT population (7.4 months [95% CI, 6.7-7.9] vs 5.7 months [95% CI, 5.5-6.7]; HR, 0.67; 95% CI, 0.52-0.86; P = .0017).2
Patients were eligible for the trial if they had high-grade relapsed epithelial ovarian, fallopian tube, or primary peritoneal cancer; 2 or more prior chemotherapy regimens, 1 of which had to be platinum based; a deleterious germline or somatic BRCA1/2 mutation; and no prior experience with wither a PARP inhibitor or single-agent paclitaxel. Randomization occurred in a 2:1 fashion to either rucaparib at 600 mg twice daily or weekly single-agent paclitaxel at 60 mg/m2 to 80 mg/m2 in those who were resistant or partially sensitive to platinum-based chemotherapy or a platinum regimen of physician’s choice in those with sensitivity. Treatment was continued until radiologically confirmed disease progression, unacceptable toxicity, death, or study termination. Optional crossover from the chemotherapy to the rucaparib arm was allowed in the event of progressive disease.
Notably, 313 of 349 patients treated on the trial received rucaparib at some point, either at first randomization or at crossover, which may have confounded the OS results in the ITT population. In total, 80 patients (69%) in the chemotherapy group did crossover to receive rucaparib. At the data cutoff of April 10, 2022, 6% of patients in the rucaparib group vs 0 in the chemotherapy group had ongoing treatment with the assigned study regimen. Patients baseline characteristics were well balances between arms.
Median duration of randomized treatment was consistently longer in the rucaparib vs chemotherapy group across those with platinum resistance (5.6 months vs 4.4 months), partial platinum sensitivity (7.6 months vs 4.5 months), and full platinum sensitivity (13.7 months vs 3.4 months). Subsequent anticancer treatment of any kind was more common in the chemotherapy vs rucaparib group in those with platinum resistance (76.3% vs 57.5%, respectively), partial platinum sensitivity (83.9% vs 61.5%), and full platinum sensitivity (84.6% vs 54.2%).
“A significant proportion of patients who were in the rucaparib group did not receive subsequent anticancer treatment [and those who] received crossover rucaparib largely stayed on treatment for more than 6 months.” Oza said.
An adjusted analysis that excluded patients who crossed over from chemotherapy to rucaparib showed favorable median OS in the experimental arm at 19.4 months (95% CI, 15.2-23.6) vs 9.1 (95% CI, 7.0-18.1) with the control arm (HR, 0.423; 95% CI, 0.276-0.650). When data were censored at crossover, OS outcomes were similar between groups (HR, 1.059; 95% CI, 0.688-1.630). OS with time-varying covariate adjustment showed no difference between treatment groups (HR, 0.91; P = .673).
Looking at results for time to second progression (PFS2), rucaparib and chemotherapy resulted in similar outcomes in those with platinum resistance (HR, 0.968; 95% CI, 0.697-1.344) and full platinum sensitivity (HR, 0.886; 95% CI, 0.492-1.596); outcomes showed slight favorability in those with partial platinum sensitivity who were treated with rucaparib (HR, 0.650; 95% CI, 0.410-1.030).
“We also tried to look at the impact of post randomization treatment. In the patients randomized to rucaparib or chemotherapy with platinum-resistant disease, there was no difference [in PFS]; whereas with subsequent treatment, there was a significant advantage to patients treated with rucaparib compared with subsequent chemotherapy,” Oza said. He continued by pointing out that those with both partially and fully platinum-sensitive disease had better PFS outcomes with initial rucaparib, and those who initially received chemotherapy had better PFS2 when they crossed over to receive the PARP inhibitor.
There were no new safety signals noted. Adverse events (AEs) of special interest included the occurrence of myelodysplastic syndrome or acute myeloid leukemia (MDS/AML) that was reported in 7 patients (3%) initially randomized to rucaparib vs 0 treated with chemotherapy.
The investigators concluded their research by raising unanswered questions about the optimal sequencing of PARP inhibitors in advanced stage disease. “Further work is also ongoing to understand the biologic basis of resistance and the optimal sequence of treatment in patients,” said Oza.