Chemo-Free PD-1 Win in dMMR Endometrial Faces OS and Combination Standard Test

Merck's KEYTRUDA Monotherapy Significantly Extends PFS in dMMR Endometrial Cancer

Merck announced positive Phase 3 KEYNOTE-C93 trial results for KEYTRUDA (pembrolizumab) as monotherapy in patients with mismatch repair deficient (dMMR) advanced or recurrent endometrial cancer. The anti-PD-1 therapy significantly improved progression-free survival (PFS), reducing the risk of disease progression or death by 43% (HR=0.57; p=0.0002) compared to platinum doublet chemotherapy. Median PFS was 38.2 months for KEYTRUDA versus 8.4 months for chemotherapy. This marks the first Phase 3 trial where a PD-1 inhibitor monotherapy demonstrated such a significant PFS improvement over chemotherapy in this patient population. The results were presented at ESMO Congress 2026 and published in The Lancet.

• KEYTRUDA monotherapy achieved a median progression-free survival of 38.2 months, a substantial increase compared to 8.4 months for platinum doublet chemotherapy. This translates to a 43% reduction in the risk of disease progression or death (HR=0.57; 95% CI, 0.42-0.79; p=0.0002), establishing a new benchmark for dMMR advanced or recurrent endometrial cancer.
• While the trial is ongoing for overall survival (OS), a favorable trend was observed for KEYTRUDA monotherapy (HR=0.77; p=0.1000), with 36-month OS rates of 70.7% versus 62.0% for chemotherapy. The safety profile was consistent with previous studies, showing a lower incidence of Grade ≥3 treatment-related adverse events (18.1% for KEYTRUDA vs. 50.7% for chemotherapy).
• KEYTRUDA monotherapy demonstrated a confirmed objective response rate (ORR) of 64.1% (95% CI, 55.8-71.9), comparable to chemotherapy's 62.1%. Crucially, the median duration of response (DOR) was not reached for KEYTRUDA, significantly outperforming chemotherapy's 10.2 months, with 36-month DOR rates of 70.7% versus 40.0%.

Why New Options are Crucial for Advanced dMMR Endometrial Cancer

Patients with advanced or recurrent endometrial cancer who progress after platinum-based chemotherapy face a poor prognosis, and treatment options have historically been limited. The introduction of immunotherapy, alone or in combination with targeted agents, has substantially expanded therapeutic options in subsequent treatment lines, yet survival outcomes remain modest and practical barriers persist.

• Poor prognosis after platinum failure: A real-life cohort reported median disease-specific survival for the entire cohort was 13.4 months (95% confidence interval 8.5-27.2 months), with 41 patients (95.3%) initiating second-line treatment, including 14 (34.1%) who received pembrolizumab alone or in combination with lenvatinib.

• Substantial toxicity burden: Among all treated patients in KEYNOTE-158, 76% had ≥ 1 treatment-related adverse event (grades 3-4, 12%), and immune-mediated adverse events or infusion reactions occurred in 28% of patients (grades 3-4, 7%; no fatal events), while lenvatinib plus pembrolizumab showed grade 3 or 4 treatment-related adverse events in 83/124 (66.9%) patients.

• Incomplete and variable antitumor activity: In KEYNOTE-158 the objective response rate was 48% (95% CI, 37 to 60) with median progression-free survival of 13.1 (95% CI, 4.3 to 34.4) months, and in PRIMMO the irORR was 12.0% (90% CI 3.4-28.2) in endometrial cancer, with PRIMMO not meeting its primary objective in both cohorts.

• Molecular testing implementation barriers: Only 24.2% of respondents reported that POLE testing was moderately or very easy to obtain, respondents from academic sites reported better access to molecular classification compared with respondents from community sites, and cost concerns were reported as large barriers to testing.

• Diagnostic discordance requiring complex workup: Compared to MMR-IHC, discordance was observed in 11 cases between the two methodologies, and the consistency rate was 92.2% (130/141), with discordance primarily occurring in the cases with MLH1 promoter hypermethylation and isolated PMS2 or MSH6 deficiency.

• Unresolved resistance and combination questions: For dostarlimab, its efficacy in other cancer subtypes, the development of resistance to monotherapy, and efficacy and safety in combination with other immunotherapeutic drugs have not yet been studied.

Unpacking the Pivotal KEYNOTE-C93 Trial Data

Pivotal data in mismatch repair-deficient advanced or recurrent endometrial cancer are defined by the phase III RUBY trial evaluating dostarlimab plus carboplatin-paclitaxel versus placebo plus chemotherapy, and the phase I GARNET trial evaluating single-agent dostarlimab in dMMR/MSI-H and MMRp/MSS cohorts. Both trials used RECIST version 1.1-based efficacy endpoints with extended follow-up for duration of response and overall survival.

Reshaping First-Line Treatment for dMMR Endometrial Cancer

In Part 1 of the phase III RUBY trial (NCT03981796) in patients with primary advanced or recurrent endometrial cancer, dostarlimab plus carboplatin-paclitaxel (CP) significantly improved progression-free survival and overall survival compared with CP alone. A Bayesian network meta-analysis of four trials involving 2577 patients found that dostarlimab plus chemotherapy or durvalumab plus chemotherapy with olaparib was associated with more survival benefits than other immunotherapy regimens and chemotherapy in the mismatch repair-deficient microsatellite instability-high (dMMR/MSI-H) and mismatch repair-proficient microsatellite-stable (pMMR/MSS) population, respectively, while pembrolizumab plus chemotherapy versus chemotherapy increased efficacy (cost) by 3.76 QALYs and US$540 817, which yielded incremental cost-effectiveness ratios of US$143 894/QALY in the dMMR/MSI-H population. A separate review reported that the combination of dostarlimab and chemotherapy reportedly coincides with a 2-year progression-free survival (PFS) of 61% compared to a 12-month PFS of 74% for pembrolizumab, with follow-up data for dostarlimab extended beyond 44 months and the median overall survival (OS) has not been reached. In routine practice, a French nationwide cohort of 644 patients with dMMR/MSI-H endometrial cancer (median age 71 years), most with metastatic disease (73.3%), receiving dostarlimab plus chemotherapy showed median OS was not reached, and the 1-year OS probability was 72.8% (95% CI 69.5-76.3), with median time to treatment discontinuation and real-world PFS of 8.6 and 9.7 months and 1-year probabilities of 35.7% (95% CI 32.2-39.6) and 44.4% (95% CI 40.7-48.4), respectively; the authors concluded this involved an older, more comorbid, and less selected population with more advanced disease than in RUBY, where survival outcomes were less favorable, but the safety profile was comparable.

For previously treated advanced MSI-H/dMMR endometrial cancer, pembrolizumab 200 mg Q3W monotherapy in the nonrandomized, open-label, multicohort, phase II KEYNOTE-158 study (NCT02628067) showed an objective response rate (ORR) of 48% (95% CI, 37 to 60) with median duration of response not reached (2.9-49.7+ months), median PFS of 13.1 (95% CI, 4.3 to 34.4) months, and median OS not reached (95% CI, 27.2 months to not reached). With longer follow-up to January 12, 2022, median (range) follow-up of 54.5 (14.7-71.4) months in the MSI-H/dMMR (n = 94) group, the ORR (95% CI) was 50% (40%-61%); 15 patients (16%) experienced a complete response, 32 (34%) experienced a partial response, with estimated 4-year duration of response rates of 66% and 56% in MSI-H/dMMR and non-MSI-H/non-dMMR groups respectively, median PFS of 13.1 (95% CI, 4.3-25.7) months in the MSI-H/dMMR group and 2.1 (95% CI, 2.1-2.2) months in the non-MSI-H/non-dMMR group, and median OS of 65.4 (95% CI, 29.5-not reached) and 11.1 (95% CI, 8.1-15.3) months, respectively. Dostarlimab monotherapy after progression on platinum-based treatment showed a reported overall response rate (ORR) of 43.5% with a manageable safety profile in the GARNET trial, and in the ambispective real-world DORA study of 129 patients with median follow-up of 11.6 months (range; 0.8-30.1), the ORR was 53.6% (95% CI 44.4 to 62.5), with complete response in 27 patients (21.6%) and partial response in 40 (32%), median time to response of 2.9 months (95% CI 2.6 to 3.6), median duration of response not reached, and probability of maintaining response at 12 and 24 months of 84.98% (95% CI 70.8 to 92.5) and 73.39% (95% CI 50.5 to 86.9), respectively, with median duration of dostarlimab administration of 8.8 months (interquartile range; 13.2) and 73 patients (57%) remaining on therapy at data cutoff.

Direct comparison of second-line retreatment strategies after first-line platinum-based chemotherapy showed no significant difference by regimen once platinum-free interval (PFI) was accounted for. In a retrospective Endometrial Cancer Molecularly Targeted Therapy Consortium cohort of 217 patients, ORR was 58% vs 49% for platinum vs lenvatinib/pembrolizumab (p = 0.27), while for all patients ORR was 68% vs 47% with PFI >12 months and ≤12 months, respectively (p = 0.002); at each PFI, ORR was similar regardless of treatment with platinum or lenvatinib/pembrolizumab (PFI ≤12 months ORR 49% vs 44% respectively, p = 0.75; PFI >12 months ORR 67% vs 75% respectively, p = 0.74), although patients treated with platinum had longer PFI than those treated with lenvatinib/pembrolizumab (median PFI 12.9 vs 4.6 months; p < 0.001). Lenvatinib plus pembrolizumab at the recommended phase II dose of lenvatinib 20 mg/d plus pembrolizumab (200 mg intravenously every 3 weeks) showed ORR in endometrial cancer of 52% (12/23; 95% CI, 30.6% to 73.2%). For safety, among all treated MSI-H/dMMR endometrial patients in KEYNOTE-158, 76% had ≥1 treatment-related adverse event (grades 3-4, 12%) with no fatal treatment-related events, and immune-mediated adverse events or infusion reactions occurred in 28% (grades 3-4, 7%; no fatal events); in RUBY, treatment-emergent adverse events were experienced by 100% of patients in both arms, treatment-related adverse events occurred in 97.9% of the dostarlimab arm and 98.8% of the placebo arm, immune-related adverse events occurred in 58.5% of patients in the dostarlimab arm and 37.0% of patients in the placebo arm, and dostarlimab- or placebo-related immune-related events were reported in 40.7% and 16.3%, respectively; in DORA, treatment was discontinued due to toxicity in 4.7% of patients.

Pembrolizumab Monotherapy: A New Paradigm for dMMR Endometrial Cancer

The recent Phase 3 KEYNOTE-C93 trial results for KEYTRUDA (pembrolizumab) monotherapy in mismatch repair deficient (dMMR) advanced or recurrent endometrial cancer mark a pivotal moment for patients and clinicians. Demonstrating a remarkable 43% reduction in the risk of disease progression or death and extending median progression-free survival to 38.2 months compared to 8.4 months with platinum doublet chemotherapy, these findings are poised to reshape the treatment landscape. This is the first time a PD-1 inhibitor monotherapy has achieved such a significant benefit over chemotherapy in this specific patient population in a Phase 3 setting.

For the pharmaceutical industry, this outcome underscores the power of biomarker-driven strategies. dMMR status is a well-established predictor of response to immunotherapy in endometrial cancer, and studies have indicated that dMMR patients may fare worse with chemotherapy alone. The strong monotherapy data for pembrolizumab could establish a new, less toxic, and highly effective first-line or early-line standard of care for dMMR advanced or recurrent endometrial cancer. This move could differentiate pembrolizumab from existing combination therapies, such as lenvatinib plus pembrolizumab, which are currently standard for advanced endometrial cancer after platinum failure, regardless of MMR status, but are associated with higher rates of grade 3 or higher adverse events.

However, several considerations remain. While highly effective for dMMR, pembrolizumab monotherapy has shown limited benefit in the mismatch repair-proficient (pMMR) population, meaning combination regimens will continue to be essential for a significant portion of endometrial cancer patients. Furthermore, even within the dMMR group, some patients may not respond to monotherapy and could require subsequent combination approaches. The management of immune-related adverse events, although often more tolerable than chemotherapy, will also remain a critical aspect of patient care. Ultimately, these results reinforce the evolving paradigm in oncology, where precise patient selection based on molecular markers is key to optimizing therapeutic outcomes and improving patient survival.