Pembrolizumab Improves PFS Over Chemo in dMMR Endometrial Cancer

First-line pembrolizumab (Keytruda) monotherapy significantly improved progression-free survival (PFS) compared with carboplatin plus paclitaxel in patients with mismatch repair–deficient (dMMR) advanced or recurrent endometrial cancer, according to findings from the phase 3 KEYNOTE-C93/GOG-3064/ENGOT-en15 trial (NCT05173987).¹ Overall survival (OS) numerically favored pembrolizumab but did not meet the prespecified threshold for statistical significance.

Brian M. Slomovitz, MD, lead study author, director of Gynecologic Oncology and cochair of the Cancer Research Committee at Mount Sinai Medical Center, as well as professor of Obstetrics and Gynecology at Florida International University, presented the data during a Presidential Symposium at the 2026 European Society for Medical Oncology (ESMO) Congress.

How did pembrolizumab affect progression-free survival?

Median PFS by blinded independent central review (BICR) was 38.2 months (95% CI, 17.9-not reached [NR]) with pembrolizumab vs 8.4 months (95% CI, 6.8-10.4) with chemotherapy (HR, 0.57; 95% CI, 0.42-0.79; P = .0002). At 36 months, 52.8% of patients in the pembrolizumab arm and 30.0% in the chemotherapy arm were progression free.

The PFS curves crossed at approximately 6 months. Slomovitz said the hazards were not proportional, so the HR averaged an early period that favored chemotherapy with a later period that strongly favored pembrolizumab. A post hoc restricted mean survival time analysis showed the difference growing over time, from 0.83 months at 12 months (95% CI, −0.09 to 1.75) to 4.03 months at 24 months (95% CI, 1.92-6.14) and 6.82 months at 36 months (95% CI, 3.44-10.20).

The PFS benefit with pembrolizumab was observed across the subgroups assessed. The HR was 0.45 (95% CI, 0.29-0.70) in patients with recurrent disease and 0.77 (95% CI, 0.49-1.21) in those with newly diagnosed advanced disease.

What were the overall survival results?

Median OS was NR in either arm, with an HR of 0.77 (95% CI, 0.52-1.15) in favor of pembrolizumab (P = .1000). The result did not cross the prespecified 1-sided boundary of 0.0189 (information fraction, 95%). The 36-month OS rate was 70.7% with pembrolizumab vs 62.0% with chemotherapy.

Among patients with BICR-confirmed progression, 85.2% of those in the chemotherapy arm (n = 88) received subsequent immunotherapy, including on-study crossover in 67%. In the pembrolizumab arm (n = 63), 46.0% received subsequent platinum-based chemotherapy. “In effect, the [OS] comparison is largely pembrolizumab first vs pembrolizumab second,” Slomovitz said.

Supplementary analyses adjusting for subsequent immunotherapy produced OS HRs of 0.55 to 0.66. PFS2 per investigator also favored pembrolizumab (HR, 0.63; 95% CI, 0.43-0.91).

How did response and quality of life compare between arms?

Objective response rate (ORR) was similar at 64.1% (95% CI, 55.8%-71.9%) with pembrolizumab and 62.1% (95% CI, 53.6%-70.0%) with chemotherapy. The complete response rate was 33.8% (n = 49) vs 26.2% (n = 38). Median duration of response was NR (range, 2.0+ to 40.7+) with pembrolizumab vs 10.2 months (range, 1.9+ to 43.8+) with chemotherapy, and 70.7% vs 40.0% of responders, respectively, remained in response at 36 months.

At week 12, the prespecified time point for patient-reported outcomes, global health status and quality of life improved with pembrolizumab and declined with chemotherapy, Slomovitz said. Physical, role, cognitive, and social functioning also favored pembrolizumab, while patients receiving chemotherapy reported worse fatigue, tingling or numbness, muscular pain, and body image.

What was the KEYNOTE-C93 trial design?

The open-label trial randomly assigned 299 patients 1:1 at 122 sites in 24 countries to pembrolizumab 400 mg intravenously every 6 weeks for up to 18 cycles (n = 149) or carboplatin (area under the curve of 5 or 6) plus paclitaxel (175 mg/m²) every 3 weeks for 6 cycles (n = 150). Patients in the chemotherapy arm did not receive maintenance therapy and could cross over to pembrolizumab at confirmed progression.

Eligible patients had inoperable advanced or recurrent endometrial carcinoma or carcinosarcoma, centrally confirmed dMMR status, no prior systemic therapy for advanced disease, and an ECOG performance status of 0 or 1. PFS per RECIST version 1.1 by BICR and OS were primary end points, and ORR was a key secondary end point. At the April 24, 2026, data cutoff, median follow-up was 36.7 months (range, 28.4-50.0). This was the final analysis of PFS and ORR and an interim analysis of OS.

Median age was 62.0 years in the pembrolizumab arm and 61.0 years in the chemotherapy arm. Endometrioid histology accounted for 91.3% of each arm, and 55.7% and 53.3% of patients, respectively, had recurrent disease.

“There is a strong desire among health care providers and patients and their families for a first-line, chemotherapy-free option for patients with advanced or recurrent endometrial cancer,” Slomovitz said during the presentation.

What safety findings were reported?

Treatment-related adverse effects (AEs) occurred in 71.1% of patients (n = 106) receiving pembrolizumab vs 97.9% (n = 141) receiving chemotherapy. Grade 3 or higher events occurred in 18.1% (n = 27) vs 50.7% (n = 73). Any drug was discontinued because of AEs in 10.7% vs 18.8%. Median treatment duration was 23.4 months with pembrolizumab and 3.5 months with chemotherapy.

Immune-mediated AEs and infusion reactions, considered regardless of attribution to study treatment, occurred in 41.6% (n = 62) vs 17.4% (n = 25), with grade 3 or higher events in 7.4% vs 3.5%. Two treatment-related deaths occurred with pembrolizumab (hepatic failure and bacterial meningitis; n = 1 each) and 1 with chemotherapy (pulmonary embolism). No new safety signals were observed.

What did the investigators and discussant conclude?

Slomovitz noted that the comparator was chemotherapy alone rather than the current standard of chemotherapy plus immunotherapy. “[KEYNOTE-C93] does not show that pembrolizumab alone is equivalent to chemotherapy and immunotherapy,” he said. “What it does show is that for selected patients with dMMR disease, especially for those [for whom] avoiding chemotherapy is a priority, pembrolizumab monotherapy is a clinically relevant, chemotherapy-free first-line option.”

Christian Marth, MD, PhD, head and professor of the Department of Obstetrics and Gynecology at Innsbruck Medical University, served as invited discussant.² He said the OS finding “might not be a home run” because more than 85% of patients received a checkpoint inhibitor after progression. “For overall survival, we are not comparing [chemotherapy] vs [pembrolizumab]. We are comparing [chemotherapy] followed by [immunotherapy] vs pembrolizumab,” Marth said.

Marth said the crossing PFS curves suggest approximately 20% of patients progress at the first scan, and that dMMR status is “a predictive factor for better immune response” rather than a homogeneous target. “I clearly support the conclusion of [Slomovitz] saying that pembrolizumab is an option for patients who prioritize tolerability and quality of life,” he said. “But having said that, it is important to mention that [immunotherapy] plus [chemotherapy] results in a better oncological outcome than [pembrolizumab] alone, but with the price of a higher toxicity.”

Slomovitz reported consulting fees from MSD, AstraZeneca, Daiichi Sankyo, Incyte, Seagen, Novocure, Karyopharm, Genmab, Aadi, Regeneron, Immunocure, Gilead, Eisai, BeiGene, and GSK. The trial was funded by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc.

References

1. Slomovitz BM, Backes F, Maćkowiak-Matejczyk B, et al. KEYNOTE-C93/GOG-3064/ENGOT-en15: phase 3, randomized, open-label study of first-line pembrolizumab vs carboplatin + paclitaxel in mismatch repair-deficient (dMMR) advanced or recurrent endometrial cancer. Presented at: 2026 European Society for Medical Oncology Congress; October 23-27, 2026; Madrid, Spain. Abstract LBA12.
2. Marth C. Discussion of LBA12. Presented at: 2026 European Society for Medical Oncology Congress; October 23-27, 2026; Madrid, Spain.