Endometrial cancer has overtaken ovarian cancer to become the gynecologic cancer with the most deaths in the United States. The disease has seen an alarming increase in incidence and mortality, as well as a widening racial disparity gap, resulting in a critical need for improved risk stratification and therapeutic strategies. Furthermore, endometrial cancer is the only cancer to have decreased survival over the past four decades in the United States.
Recognizing the urgent need to improve outcomes for patients diagnosed with endometrial cancer, the National Cancer Institute Gynecologic Cancer Steering Committee convened a clinical trials planning meeting in January 2024. Multidisciplinary experts were charged with addressing critical challenges to optimize treatment of endometrial cancer in the new immunotherapy landscape. Working group 1 focused on translational science and reviewed the literature for data on validated discriminants of response to immunotherapy.
Molecular classification of endometrial cancer has substantially impacted clinical practice and treatment recommendations and has now been incorporated into clinical trial designs. The Cancer Genome Atlas identified four molecular classes: POLE ultramutated, MSI hypermutated, copy-number low (endometrioid like), and copy-number high (serous like). The Proactive Molecular Risk Classifier for Endometrial Cancer (ProMisE) model currently provides the most clinically accessible testing application.
The ProMisE model overcomes the technical and bioinformatic challenges of the original TCGA classification by capitalizing on key observations. POLE status can be achieved either by a next-generation sequencing panel or directed POLE gene sequencing followed by mismatch repair immunohistochemistry or microsatellite instability testing. TP53 status is the last step in hierarchical molecular classification for both prognostic and predictive counseling.
Further subdivision of endometrial cancers by additional therapeutic targets such as HER2 expression or within molecular subtypes based on prognostic and predictive markers like estrogen receptor expression or MLH1 methylation status should continue to be considered in clinical trial design.
The mismatch repair deficiency or microsatellite instability group has had the most clinical implications from a treatment perspective. Recent positive data from the RUBY and NRG-GY018 pivotal trials, which incorporated anti-PD-1 agents with chemotherapy in metastatic and recurrent endometrial cancer, led to FDA approval and increased immune checkpoint blockade use.
However, there remains a need for novel strategies for patients whose tumors demonstrate primary or acquired resistance while on immunotherapy, including 40% of mismatch repair deficiency and 70% of mismatch repair proficient tumors observed in these trials. Potential causes for immunotherapy resistance include low neo-antigen presentation, multiple immune checkpoints at play, and immune cell-driven immunosuppression.
Notably, early data suggest that molecular mechanisms associated with response to anti-PD-1 therapy may differ in epigenetic versus mutational mismatch repair deficiency. Epigenetic silencing of MLH1 is the most common mechanism for mismatch repair deficiency function in endometrial cancer, making up 75-80% of cases. Multiple cohorts have shown a statistically significant prognostic difference with reduced progression-free survival associated with MLH1 methylated mismatch repair deficiency.
TP53 mutational status defines the clinically accessible TCGA copy-number high group, which has traditionally been considered less immunogenic than other molecular groups. Tumors in this class typically are mismatch repair proficient and may include HER2 amplification, PIK3CA mutations, and focal amplifications of MYC, ERRB2, and CCNE1. CCNE1 amplification has been associated with chemotherapy resistance and worse overall survival in uterine serous carcinomas.
HER2 amplification or overexpression is most common in the TP53-mutated class, occurring in up to 35% of serous and 12-15% of carcinosarcoma endometrial cancer patients. Preliminary clinical data from a small randomized phase 2 trial showed progression-free survival and overall survival benefit of trastuzumab, and recent data show benefit of trastuzumab deruxtecan in a single-arm phase 2 study.
Targeting TP53 and MDM2 represents an active area of investigation for this high-risk group. Novel mechanisms such as reactivation of wild-type p53 function or MDM2 inhibition are being explored, along with considerations around loss of function versus gain of function mutations in endometrial cancer.
To date, mismatch repair status and tumor mutational burden have been correlated as predictive biomarkers of response to immunotherapy in endometrial cancer. Tumor mutational burden of no less than 10 mut/Mb has been defined as tumor mutational burden high across numerous cancers. However, in KEYNOTE 158, 10 of 15 endometrial cancer tumors with high tumor mutational burden were also microsatellite instability.
A major key observation from the working group was the lack of validated discriminants for immunotherapy response beyond mismatch repair status and tumor mutational burden that have been validated for prospective treatment allocation. The quality and quantity of data specific to endometrial cancer are limited and should be evaluated with caution.
One opportunity being evaluated as a predictor of durable response is T-cell expansion, clonality, and dynamics, which can be tested through blood sampling. CD8 T-cell infiltration is associated with mismatch repair deficiency status but has not been validated as a selection biomarker. Gamma-delta T cells were recently associated with response to combined immune checkpoint blockade and anti-angiogenic therapy.
Within the no specific molecular profile group, several commonly mutated genes were discussed including PTEN, CTNNB1, and ARID1A. PTEN loss is one of the most frequent abnormalities in endometrial cancer and is especially enriched in this group, making it an appealing potential target for trials. However, PTEN loss is not associated with PD-L1 expression or lymphocyte and macrophage infiltration.
CTNNB1 mutation is associated with immune exclusion across tumor types and the impact of WNT/beta-catenin signaling on immune response warrants evaluation in endometrial cancer. ARID1A is the most frequently mutated component of the SWItch/sucrose nonfermentable chromatin remodeling complex and has emerged as a potential therapeutic target. Mismatch repair status and ARID1A have been correlated with mixed results regarding response to immune checkpoint inhibitors.
Estrogen receptor status has been identified as an opportunity for prognostic refinement within the no specific molecular profile group. Low-grade tumors and estrogen receptor-positive tumors showed a very low disease-specific death rate, while most deaths in this group were among high-grade or estrogen receptor-negative tumors.
Radiation therapy has been proposed as a strategy to enhance tumor immunity through the induction of immunogenic cell death and modulation of conditions in the tumor microenvironment. Considerations in the design of trials combining radiation with immune checkpoint inhibitors include sequencing, treatment volumes, inclusion of tumor-draining lymph nodes, and management of oligometastatic disease.
The use of neoadjuvant immunotherapy has been successful in other disease sites but data in endometrial cancer are limited. Work from Eerkens and colleagues studied this approach using pembrolizumab in mismatch repair deficiency tumors where they noted radiographic response in 3 of 8 patients and no progression events in this pilot study. Neoadjuvant treatment importantly did not delay surgery and was not associated with any clinically significant dose-limiting toxicity.
Opportunities to enhance immunotherapy response like radiation or neoadjuvant approaches are of increasing need, given the recent negative KEYNOTE B21 trial where pembrolizumab in combination with chemotherapy did not show benefit in high-risk endometrial cancer after surgery with curative intent. However, the mismatch repair deficiency group did have substantial benefit in B21.
The incidence and death rate of endometrial cancer has increased at an alarming pace. Although there have been substantial therapeutic advancements, there is an urgent need to identify novel clinical trial designs. Emphasis on molecular classification and tailored approaches will be essential, and there is a need for increasing correlative and translational science in clinical trial populations.
After extensive review of published studies, the clinical trials planning meeting working group identified no discriminants of immunotherapy response in endometrial cancer beyond mismatch repair status or tumor mutational burden that have been validated for prospective treatment allocation. Consequently, the working group concluded that there is an immediate and urgent need for endometrial cancer-specific translational research to be undertaken and prioritized.
Several factors that impact conditions in the endometrial cancer tumor immune microenvironment should also be considered in clinical trial design, including obesity, estrogen status, microbiome, and medical comorbidities. The increasing disparities and impact of race and ethnicity could not be emphasized enough. Molecular classification is recommended when able in endometrial cancer clinical trial design with further evaluation of biomarkers for prognostic and predictive refinement.