Practice-Changing Study in Advanced Cervical Cancer

Adding pembrolizumab (Keytruda) to chemotherapy with or without bevacizumab (Avastin) significantly slowed progression of metastatic cervical cancer and improved overall survival (OS), investigators in a randomized trial concluded.

Median progression-free survival (PFS) improved from 8.2 months with chemotherapy plus or minus bevacizumab (Avastin) to 10.4 months with the addition of the PD-1 inhibitor. Comparison of 12-month PFS showed a 10%-11% absolute difference in favor of pembrolizumab in both the PD-L1-positive patients and the overall population. Median OS in all patients was 24.4 months with pembrolizumab and 16.5 months with placebo.

The results likely establish pembrolizumab plus chemotherapy, with or without bevacizumab, as the new standard of care for persistent, recurrent, or metastatic cervical cancer, said Nicoletta Colombo, MD, of the University of Milan-Bicocca, at the European Society for Medical Oncology (ESMO) virtual meeting.

“Adding pembrolizumab to chemotherapy with or without bevacizumab provides statistically significant, clinically meaningful overall survival and progression-free survival improvements in women with persistent, recurrent, or metastatic cervical cancer,” said Colombo. “Significant benefit was observed in all protocol-specified primary-analysis populations. The benefit was generally consistent across all protocol-specified subgroups, including the with- and without-bevacizumab subgroups.”

“The safety profile for pembrolizumab plus chemotherapy with or without bevacizumab was manageable. The observed adverse events were as expected based on the profiles of the individual drugs. Pembrolizumab plus chemotherapy with or without bevacizumab may be a new standard of care for women with persistent, recurrent, or metastatic cervical cancer,” she noted.

‘Should,’ Not ‘May Be’

The results pose a number of questions that require further investigation, said ESMO invited discussant Mansoor Raza Mirza, MD, of the University of Copenhagen. Even so, the results moved him to reword Colombo’s conclusion.

“I would say this should be — not may be, as Dr. Colombo said — the new standard of care for our patients,” he said.

For years, platinum-based chemotherapy has represented standard of care for advanced cervical cancer. The GOG-240 trial, conducted almost a decade ago, suggested the “preferred regimen” should consist of platinum, paclitaxel, and bevacizumab, which was associated with a median OS of 17.5 month, Colombo noted.

PD-1 inhibitors demonstrated single-agent activity in previously treated cervical cancer. Pembrolizumab resulted in an overall response rate (ORR) of 14.3% in a phase II trial of patients with PD-L1-positive disease treated with one or more prior lines of chemotherapy. Subsequently, cemiplimab (Libtayo) significantly improved OS versus chemotherapy in previously treated cervical cancer (median OS 12.0 vs 8.5 months).

No outcome data had been available for the addition of a PD-1/L1 inhibitor to chemotherapy with or without bevacizumab, said Colombo. The randomized, phase III KEYNOTE-826 trial addressed the lack of information. Investigators in the multinational trial randomized 617 patients to pembrolizumab or placebo in addition to chemotherapy with or without bevacizumab. The dual primary endpoints were PFS and OS, as assessed in patients with a combined positive score (CPS) ≥1 for PD-L1 expression, all patients, and the subgroup with CPS ≥10.

About 70% of patients had squamous-cell carcinoma, and 89% had CPS ≥1. About three-fourths of the patients had received chemotherapy or radiation with or without surgery, and about 19% had no prior therapy. Almost half the patients had stage I-II disease at diagnosis, but more than 30% had stage IVb. About 20% of the patients had metastatic disease at study entry.

The primary analysis of the CPS ≥1 population showed a 38% reduction in the PFS hazard ratio with the addition of pembrolizumab (95% CI 0.50-0.77, P<0.001). The 12-month PFS was 45.5% with pembrolizumab and 34.1% with placebo. The results were similar in the all-comer analysis (median PFS 10.4 vs 8.2 months; 12-month PFS 44.7% vs 33.5%; HR 0.65, 95% CI 0.534-0.79, P<0.001) and in patients with CPS ≥10 (10.4 vs 8.1 months; 12-month PFS 44.6% vs 33.5%; HR 0.58, 95% CI 0.44-0.77, P<0.001).

Wide-Ranging Benefits, Unresolved Issues

The subgroup analysis showed a consistent benefit for the addition of pembrolizumab with two possible exceptions. Patients with CPS <1 and those with primary metastatic disease had HRs approaching 1.0. Patients who received bevacizumab and those who did not derived similar benefit from pembrolizumab. The OS analyses yielded HRs of 0.61 to 0.67 in favor of pembrolizumab (P<0.001) for all comparisons, as well as 10%-14% absolute differences in 12- and 24-month OS.

The most common adverse events (AEs) in both arms were anemia, alopecia, nausea, diarrhea, and fatigue. As expected, immune-mediated AEs occurred more often with pembrolizumab, the most common being hypo- and hyperthyroidism, said Colombo. Treatment with pembrolizumab also significantly prolonged the time to deterioration of quality of life.

In his critique of the trial, Mirza noted that the findings represent an interim analysis. More patients randomized to the placebo group had nonsquamous histology (31% vs 24%), and immune checkpoint inhibitors are known to have only modest activity in adenocarcinoma, he noted. Pembrolizumab had no activity in the PD-L1-negative population and limited activity in patients who had metastatic disease at diagnosis, who accounted for 31% of the total population.

“One possible interpretation is that chemo- and radiotherapy-naive patients have a higher response to platinum-based chemotherapy,” said Mirza.

More than a third of the patients did not receive bevacizumab, and the PFS and OS improvement with pembrolizumab did not achieve statistical significance. The observation could mean that the addition of pembrolizumab to chemotherapy is no better than chemotherapy alone or that the study lacked statistical power to analyze the efficacy associated with bevacizumab, Mirza continued.

The pembrolizumab regimen led to significant PFS and OS benefits, was well tolerated, and should be the new standard of care, he concluded. However, future studies should aim to determine whether the combination is effective for primary metastatic disease, biomarker-negative patients, and adenocarcinoma and whether bevacizumab should be included.