Single-agent immunotherapy as first-line treatment led to a huge improvement in progression-free survival (PFS) in inoperable recurrent/advanced mismatch repair-deficient (dMMR) endometrial cancer, a large randomized trial showed.
Patients allocated to pembrolizumab (Keytruda) had a median PFS of 38.2 months, more than four times greater than the 8.4-month median with carboplatin-paclitaxel chemotherapy. A preliminary analysis of overall survival (OS) showed a trend toward improvement with pembrolizumab at 36 months. Substantially more patients in the chemotherapy arm crossed over to pembrolizumab or other immunotherapy-containing treatment at progression, but patients who started treatment with pembrolizumab still fared better.
The study did not compare single-agent pembrolizumab against the current standard of care, which is chemotherapy plus immunotherapy, Brian Slomovitz, MD, of Mount Sinai Medical Center in Miami Beach, Florida, acknowledged during a presentation at the European Society for Medical Oncology (ESMO) congress in Madrid.
“[This trial] does not show that pembrolizumab alone is equivalent to chemotherapy and immunotherapy,” said Slomovitz. “What it does show is that for selected patients with dMMR disease, especially those for whom avoiding chemotherapy is a priority, pembrolizumab monotherapy is a clinically relevant, chemotherapy-free first-line option.”
The study was published simultaneously in The Lancet.
What It Means
ESMO invited discussant Christian Marth, MD, PhD, of Innsbruck Medical University in Austria, said the KEYNOTE-C93 trial addressed the issue of whether “we really need chemotherapy for mismatch repair-deficient tumors.” He homed in on data showing that the Kaplan-Meier PFS curves did not begin to separate until after 6 months, suggesting that chemotherapy offered early disease control that was at least equivalent to that of the PD-1 inhibitor and perhaps superior.
“At 3 months there is a difference of 10% and altogether 20% inferiority for the pembrolizumab arm [by 6 months],” said Marth. “That means that 20% of the patients progress at the first CT scan. The question is, who are those 20% of primary pembrolizumab-resistant or refractory patients?”
Early survival inferiority with immunotherapy is not unique to endometrial cancer, Marth continued. In the KEYNOTE-177 trial of patients with microsatellite instability-high (MSI-H) or dMMR metastatic colorectal cancer, PFS with pembrolizumab was inferior to chemotherapy until after 6 months.
“Mismatch repair deficiency is not a target, it’s a predictive factor for better immune response, and mismatch repair deficiency is not a homogeneous group,” he said.
Marth also noted that 85% of the patients in the control arm crossed over to standard-of-care chemoimmunotherapy at progression.
“This may not be a home run [for single-agent pembrolizumab],” he said. “For overall survival, we are not comparing chemotherapy versus pembrolizumab. We are comparing chemotherapy followed by immuno-oncology versus pembrolizumab, so we will have to wait for further analysis.”
Despite the questions and limitations, Marth agreed the study showed that single-agent pembrolizumab is a reasonable option for patients who prioritize avoidance of chemotherapy.
Single-agent immunotherapies are currently approved in previously treated patients with endometrial cancer and dMMR or MSI-H disease, but none are approved as a single agent for first-line treatment. Several — including pembrolizumab, durvalumab (Imfinzi), and dostarlimab (Jemperli) — carry indications in the frontline setting in combination with carboplatin-paclitaxel.
C93 Details
Cancers with dMMR account for 25-30% of newly diagnosed endometrial cancers and exhibit a highly immunogenic microenvironment, Slomovitz noted in his introduction. The phase II KEYNOTE-158 trial showed that almost half of patients with MSI-H/dMMR endometrial cancers responded to single-agent pembrolizumab and median duration of response was not reached. The phase III KEYNOTE-C93 trial continued the evaluation of single-agent pembrolizumab in dMMR endometrial cancer in a comparison with a standard chemotherapy regimen.
Investigators at 122 sites in 24 countries enrolled and randomized 299 patients with inoperable, advanced, or recurrent endometrial carcinoma or carcinosarcoma with confirmed dMMR status. Eligible patients had received no prior systemic therapy except in the neoadjuvant setting and only if recurrence or progression occurred after 6 months or longer.
Patients received either pembrolizumab for up to 18 cycles or carboplatin-paclitaxel chemotherapy for six cycles. The primary endpoints were PFS by independent review and OS.
The patients had a median age of 61-62, 60-70% were white, and a majority had recurrent disease. About 40% of the patients had stage I disease at diagnosis, 20-25% had stage III, and 30% had stage IV.
The PFS analysis showed a 43% reduction in the hazard ratio in favor of pembrolizumab (95% CI 0.42-0.79, P=0.0002). Landmark analyses showed large advantages for pembrolizumab at 24 months (55.4% vs 31.0%) and 36 months (52.8% vs 30.0%). The restricted mean survival time for the pembrolizumab arm increased over time, from -0.33 at 6 months to 0.83 at 12 months and 6.82 at 36 months.
The prespecified interim analysis of OS showed a 23% reduction in the hazard, which did not achieve statistical significance (95% CI 0.52-1.15). The 24- and 36-month landmark analyses both favored pembrolizumab (79.0% vs 73.2% and 70.7% vs 62.0%, respectively).
Substantially more patients randomized to chemotherapy received subsequent anticancer treatment (87.5% vs 50.8%), which overwhelmingly contained immunotherapy (85.2% vs 11.1%). Patients who progressed on pembrolizumab most often received platinum-based chemotherapy.
Sensitivity analyses that accounted for crossover and subsequent immunotherapy use in the control arm produced OS hazard ratios of 0.66 to 0.55 in favor of pembrolizumab.
The advantage for the pembrolizumab arm persisted in an investigator assessment of PFS2, which showed a 37% reduction in the hazard ratio (95% CI 0.43-0.91).
Global health status and quality of life improved with pembrolizumab and declined with chemotherapy, reaching clinically meaningful differences, said Slomovitz. Symptom scores also favored pembrolizumab.
Grade ≥3 treatment-related adverse events (TRAEs) occurred in 50.7% of the control group versus 18.1% of the pembrolizumab arm. TRAEs leading to discontinuation also occurred more often with chemotherapy (14.6% vs 9.4%). Not unexpectedly, immune-mediated AEs occurred more often with pembrolizumab, but associated discontinuation rates were 2-3% in both groups.
