The PARP inhibitor class, featuring drugs from GSK, AstraZeneca/Merck and pharma&, has faced serious regulatory setbacks before this pair of clinical disappointments.
In a rough day for PARP inhibitors in ovarian cancer, GSK’s Zejula didn’t deliver a life extension benefit in the final analysis of a phase 3 trial. Another study showed that pharma&’s attempt to enhance its Rubraca with Bristol Myers Squibb’s Opdivo seriously backfired.GSK’s Zejula boasts a broad U.S. approval as a first-line maintenance treatment for ovarian cancer patients who have responded to initial platinum-based chemotherapy. But the phase 3 study that earned the drug that approval, PRIMA, has now shown at its final analysis that Zejula didn’t extend patients’ lives compared with placebo.After a median follow-up of about 74 months, patients in the PRIMA trial who took Zejula as a first-line maintenance treatment were essentially as likely to die as those who received placebo. In fact, Zejula was linked to a 1% higher risk of death, as Zejula takers lived a median 46.6 months, versus 48.8 months for control, according to results presented at the European Society for Medical Oncology (ESMO) 2024 annual meeting.The negative overall survival showing, however marginal, raises the possibility of an FDA crackdown like the one the agency waged on the PARP inhibitor class about two years ago. At that time, several later-line ovarian cancer indications for Zejula, AstraZeneca and Merck’s Lynparza, and the then-Clovis Oncology-owned Rubraca were withdrawn or restricted after the FDA noted a potential long-term detriment to patient survival. Zejula’s OS readoutThe risk of death in PRIMA only turned against Zejula after a small group of patients with undetermined status on the homologous-recombination biomarker were counted, GSK’s global head of oncology R&D, Hesham Abdullah, M.D., noted in an interview with Fierce Pharma. Zejula was associated with a positive 5% reduction in the risk of death in patients with homologous recombination-deficient (HRd) tumors and a 7% improvement in the homologous recombination-proficient (HRp) population. None of the above numbers carry statistical significance.Zejula secured its FDA approval for the first-line maintenance treatment of ovarian cancer—regardless of biomarker status—in 2020 after the PRIMA trial demonstrated that the drug could reduce the risk of disease progression or death by 38% versus placebo. At the time, the biggest benefit was seen in those with BRCA mutations, followed by other HRd types, then HRp tumors, with those with undetermined biomarker status experiencing the least benefit.More therapies have emerged for ovarian cancer since the PRIMA trial, and the overall survival analysis was likely confounded by the use of subsequent therapies, Abdullah said. About 38% of patients in the placebo arm received subsequent PARP inhibitor treatment in PRIMA, versus 12% for the Zejula arm. In the HRd subgroup, the difference was even bigger, with the rates at 48% and 16%, respectively. Still, at first glance, Zejula’s overall survival performance in PRIMA appears weaker than the result from its rival Lynparza in the PAOLA-1 trial, with cross-trial caveats considered. In a descriptive analysis of overall survival in HRd-positive patients, Lynparza plus bevacizumab reduced the risk of death by 38% versus bevacizumab alone in the PAOLA-1 trial.Lynparza’s FDA approval in first-line maintenance ovarian cancer is limited to HRd patients, and the AZ/Merck drug remains the PARP class leader within its approved HRd population. As doctors are more reluctant to use PARP inhibitors in HRp patients, Zejula’s sales have been lagging far behind Lynparza’s quarterly figures.PRIMA enrolled a higher-risk patient population compared with other studies, Abdullah said, pointing to such factors as stage of disease, visible residual disease and prior use of neoadjuvant chemotherapy as evidence.GSK will communicate the latest PRIMA data with the FDA, but Abdullah declined to comment further about potential regulatory actions.Zejula continued to show a benefit in slowing disease progression or death. At five years, about 22% of patients in the Zejula arm were estimated to be alive, versus 12% for placebo. The rates in the HRd population were 35% and 16%, respectively.“We’re comforted by the long-term safety profile and data that we see through survival despite the impact of subsequent therapies,” Abdullah said. Rubraca’s combo setbackAmong the three PARP inhibitors that have an ovarian cancer indication, pharma&’s Rubraca is the smallest player by market share. The Austrian company took over the struggling drug from the bankrupt Clovis Oncology last year. Before that, Clovis had tried to plump up Rubraca’s profile by combining the drug with BMS’ PD-1 inhibitor Opdivo. That now appears to have been a bad idea.Pharma& unveiled the failure of the Rubraca-Opdivo combo as a first-line maintenance treatment for ovarian cancer in May. Now, data from the phase 3 ATHENA-COMBO trial presented at ESMO 2024 show that patients who got the cocktail performed even worse than those who received Rubraca alone.The ATHENA-COMBO trial was a complete failure no matter which patient group was examined. In the overall trial group, the combo showed a 30% increased risk of disease progression or death versus solo Rubraca, as the median PFS was 15 months for the combo and 20.2 months for Rubraca monotherapy.In patients with or without BRCA mutations, with various levels of heterozygosity, or those with PD-L1-positive tumors, the results were the same: The Rubrca-Opdivo regimen was worse.In terms of treatment-related adverse events at grade 3 or higher, the combo arm saw a higher rate of elevated liver enzymes, which suggest liver dysfunction, at 21.2%, compared with 10% for the monotherapy arm. Patients in the combo arm stayed notably shorter on their treatment, as the median exposure to Rubraca was 8.4 months among them, compared with 14.7 months in the control group.How thoroughly the ATHENA-COMBO trial failed was somewhat of a surprise. The phase 3 DUO-O trial previously showed positive tumor progression results for adding AZ’s PD-L1 inhibitor Imfinzi to first-line chemo and bevacizumab and then adding Imfinzi and Lynparza to bevacizumab in the maintenance phase. However, that study didn’t establish a clear contribution from Imfinzi, and AZ doesn't seem to have made any regulatory filing for that use.“While the negative primary results were disappointing, the findings provide further support of Rubraca monotherapy” in the first-line maintenance setting, a pharma& spokesperson said about ATHENA-COMBO in a statement to Fierce Pharma.The problem is, Rubraca as a monotherapy has yet to win an FDA approval as a first-line maintenance treatment in ovarian cancer. Amid the FDA’s previous crackdown on the PARP inhibitor class, the agency last year rejected Rubraca’s application in that first-line setting, requesting long-term overall survival analysis from the phase 3 ATHENA-MONO trial. Likely as a punishment for Clovis’ rogue decision to file the application despite the agency’s prior warning, the FDA actually elevated its request in the rejection, asking for the final overall survival analysis in any potential future refiling. The final analysis from ATHENA-MONO will happen when 70% of patients in the trial have passed away, whereas the agency had previously told Clovis that a 50% maturity threshold would have been enough.Pharma& spoke with the FDA again in May this year, and the agency maintained its stance stated in the complete response letter, the pharma& spokesperson told Fierce. As such, the company will continue the ATHENA-MONO trial and explore a future FDA regulatory path, the spokesperson added.Even as the FDA wards off Rubraca’s first-line bid, the European Commission in November approved that indication. This month, the drug just won reimbursement in France for that first-line maintenance use, the company spokesperson said.