BRCA-negative ovarian cancer gets PFS gains with pembrolizumab plus olaparib

First-line treatment with chemotherapy plus pembrolizumab followed by maintenance with pembrolizumab plus olaparib, with or without bevacizumab, appears to prolong progression-free survival (PFS) in advanced epithelial ovarian cancer without BRCA 1/2 mutations, according to data from the phase III ENGOT-ov43/GOG-3036/KEYLYNK-001 trial.

At the initial analysis with a median follow-up of 30.1 months, the median PFS in the subset of patients with PD-L1 expression (combined positive score [CPS] ≥10) was 23.7 months with chemotherapy plus pembrolizumab during induction followed by pembrolizumab plus olaparib during maintenance (pembrolizumab–olaparib group) as opposed to 15.2 months with standard chemotherapy alone (control group) (hazard ratio [HR], 0.63, 95 percent confidence interval [CI], 0.49–0.80; p<0.0001). The median PFS in the total intention-to-treat (ITT) population was 22.1 vs 14.6 months (HR, 0.68, 95 percent CI, 0.58–0.81; p<0.0001), respectively. [SGO 2025, abstract LBA 07]

The clinically meaningful improvement in PFS seen in the pembrolizumab–olaparib group continued at the final analysis, reported lead investigator Dr Matthew Powell from the Washington University School of Medicine in St Louis, Missouri, US.

Over a median follow-up of 49.6 months, the median PFS was 23.9 months in the pembrolizumab–olaparib group vs 15.2 months in the control group in the CPS ≥10 population (HR, 0.66, 95 percent CI, 0.53–0.83) and 22.3 vs 14.6 months, respectively, in the total ITT population (HR, 0.71, 95 percent CI, 0.61–0.84).

The respective 48-month PFS rates in the in the pembrolizumab–olaparib and control groups were 34.7 percent and 20.6 percent in the CPS ≥10 population and 28.5 percent and 16.7 percent in the total ITT population.

“The PFS benefit was generally consistent across prespecified subgroups,” including those defined by age, race, Eastern Cooperative Oncology Group (ECOG) PS score, PD-L1 status, debulking surgery, and bevacizumab use.

In terms of the secondary outcome of overall survival (OS) at the final analysis, no notable difference was seen between the pembrolizumab–olaparib and control groups in either the CPS ≥10 population (median OS, 50.2 vs 51.6 months; HR, 0.98, 95 percent CI, 0.75–1.27) or the total ITT population (median OS, 47.7 vs 47.1 months; HR, 1.04, 95 percent CI, 0.87–1.25).

KEYLYNK-001 details

The trial included 1,367 patients (median age 61 years, 79.3 percent White) with advanced epithelial ovarian cancer without BRCA mutations. These patients were randomly allocated to one of the three treatment groups: pembrolizumab plus olaparib (n=455), pembrolizumab (chemotherapy plus pembrolizumab during induction, followed by pembrolizumab during maintenance; n=458), and control (n=454).

Carboplatin–paclitaxel-based chemotherapy was given for five cycles, pembrolizumab was administered at 200 mg every 3 weeks for 35 cycles, and maintenance olaparib was administered at 300 mg twice daily for up to 2 years. Bevacizumab use during chemotherapy and maintenance was permitted. The primary endpoint of PFS in the CPS ≥10 and the total ITT populations was assessed per RECIST v1.1 by investigator review.

At baseline, 40.53 percent of patients had an ECOG performance status of 1, 39.65 percent had stage IVA-IVB disease at screening, and 50.26 percent had CPS ≥10. Most patients (85.59 percent) had high-grade serous ovarian carcinoma and underwent debulking surgery (91.44 percent). Bevacizumab was initiated in 44.62 percent of patients.

Comparison between the pembrolizumab and control groups yielded no significant differences in PFS at the final analysis both in the CPS ≥10 population (median, 17.3 vs 15.2 months; HR, 0.95, 95 percent CI, 0.77–1.19; p=0.3339) and the total ITT population (median, 15.2 vs 14.6 months; HR, 1.01, 95 percent CI, 0.87–1.18). The same was true for OS in the CPS ≥10 population (median, 56.4 vs 51.6 months; HR, 0.94, 95 percent CI, 0.72–1.22) and the total ITT population (median, 44.2 vs 47.1 months; HR, 1.06, 95 percent CI, 0.89–1.27).

As for postprogression therapy, most patients received bevacizumab (32.0 percent in the pembrolizumab–olaparib group, 30.2 percent in the pembrolizumab group, and 34.4 percent in the control group). Others received PARP inhibitors (3.4 percent, 14.8 percent, and 17.6 percent, respectively) or immunotherapy (1.4 percent, 1.8 percent, and 5.7 percent, respectively).

“The safety profile of pembrolizumab plus olaparib was manageable and consistent with the known profiles of the individual therapies, with no new safety signals identified,” Powell said.

The most common any-grade treatment-related adverse events (AEs) were anaemia (between 41.0 percent and 54.9 percent), nausea (between 34.4 percent and 44.9 percent), neutropenia (between 28.5 percent and 35.6 percent), alopecia (between 23.7 percent and 27.3 percent), neutrophil count decreased (between 20.4 percent and 22.6 percent), fatigue (between 17.2 percent and 21.5 percent), asthenia (between 11.9 percent and 20.6 percent), and arthralgia (between 19.2 percent and 22.1 percent).

Infusion reactions and immune-mediated AEs such as hypothyroidism and hyperthyroidism, among others, were also reported.

Post hoc findings

In a post hoc, exploratory analysis, Powell noted an improvement in PFS with pembrolizumab in the subgroup of patients with low loss of heterozygosity (LOH) who did not receive bevacizumab. Median PFS was 16.4 in the pembrolizumab–olaparib group and 15.9 months in the pembrolizumab group vs 10.2 months in the control group in the CPS ≥10 population (HR, 0.56 and 0.53, respectively); and 13.7 and 11.9 vs 10.1 months, respectively, in the total ITT population (HR, 0.66 and 0.69, respectively).

With respect to OS, there was not much difference between the pembrolizumab–olaparib and control groups in the subgroup of patients with low LOH who did not receive bevacizumab, according to Powell.

However, these very specific patients who had low LOH, did not receive bevacizumab, and had CPS ≥10 “for some reason are living significantly longer” when treated with chemotherapy plus pembrolizumab alone vs chemotherapy alone, with a median OS improvement of 26 months, he added.

“We scratch our heads trying to figure out what is going on here. Who are these patients exactly? How are the choices made about bevacizumab? I think it’s exciting to try to figure out why or how this happened,” Powell said.

“I’m not telling everybody to go out and start using pembrolizumab. We need to figure out more about this, but it was quite exciting to see a 47-percent improvement in PFS and 39-percent less deaths. This represents about 10 percent of the population that we take care of with ovarian cancer and may be intriguing enough for further study,” he added.