Endometrial cancer (EC) is the sixth most common cancer in women worldwide and the fourth most common in the United States, with incidence and mortality rates continuing to rise since the late 1990s. EC comprises most uterine corpus carcinomas and represents a heterogeneous group of cancers varying in pathology, histology, molecular biology, immunogenicity, and prognosis. The increasing burden of EC highlights the urgent need for improved therapeutic strategies.
Histologically, EC is subdivided into endometrioid (type I, 80% of cases) and non-endometrioid (type II) categories. Molecular classification through TCGA, ProMisE, and TransPORTEC studies has defined four subtypes: POLE mutated (best prognosis), MSI-H/dMMR (intermediate), copy number low/p53wt (intermediate), and copy number high/p53abn (worst prognosis). These molecular subtypes have become critical for guiding treatment decisions and predicting immunotherapy response.
Management of advanced EC relies on a multimodal approach involving surgery, radiation, and chemotherapy. A platinum-based chemotherapy doublet of carboplatin plus paclitaxel provides the backbone for first-line systemic therapy with a median overall survival of 37 months and median progression-free survival of 13 months. The introduction of molecular classification has led to rapid development of targeted immunotherapies.
Current NCCN guidelines recommend immune checkpoint inhibitors (ICIs) combined with chemotherapy for first-line treatment of inoperable or recurrent advanced-stage disease. Durvalumab is recommended for dMMR/MSI-H tumors based on the DUO-E trial, pembrolizumab irrespective of dMMR/MSI status based on the NRG-GY018 trial, and dostarlimab based on the RUBY trial. Additionally, trastuzumab deruxtecan is indicated for HER2-positive recurrent disease, and pembrolizumab plus lenvatinib serves as a second-line option.
Tumor antigenicity in EC is highly variable across histologic and molecular subtypes. A high tumor mutational burden (TMB) is correlated with improved prognosis and response to PD-1/PD-L1 blockade. Among molecular phenotypes, dMMR/MSI-H tumors often have high TMB while POLE mutated tumors tend to have very high TMB. In contrast, CNL/p53wt and CNH/p53abn tumors are associated with low TMB.
Antigen production is driven by specific mutational processes including defective DNA replication and repair mechanisms. Cancer-specific neoantigens such as mutant EGFR, TP53, and KRAS arise from non-synonymous substitutions or frameshift peptides. Additionally, cancer/testis antigens like MAGE-3, MAGE-4, and PRAME are overexpressed in EC, while surface antigens including HER2, TROP2, and folate receptor alpha represent targetable biomarkers for antibody-drug conjugates and immunotherapies.
The prognostic significance of tumor-infiltrating lymphocytes (TILs) in EC is well established. POLE mutated and MSI-H/dMMR tumors have a more robust antigenic footprint and greater intratumoral CD8+ TIL densities, suggesting susceptibility to TIL therapy. The extent of cytotoxic adaptive immune-cell infiltration varies among EC subtypes, with sporadic or hereditary tumor origin also influencing immunotherapeutic efficacy.
Increased intratumoral CD8+ TILs are independently associated with improved overall survival and disease-free survival across molecular and histologic subtypes. Additional subpopulations expressing CD39 (exhaustion phenotype), CD103 (resident memory phenotype), and CXCL13 (B cell-chemokine phenotype) are emerging as favorable prognostic biomarkers. Natural killer cells represent the second most dominant immune population in EC tumors after T cells, comprising three distinct subsets with varied antitumor cytotoxic potential.
Beyond the PD-1/PD-L1 axis, several emerging immune checkpoints are under investigation in EC. B7-H4 is expressed on tumor cells and antigen-presenting cells, suppressing CD4+ and CD8+ T-cell function. Its prevalence ranges from 71.5 to 100% in ECs, and positive B7-H4 status independently predicts disease-specific survival in non-specific molecular profile tumors, suggesting it may be effective for otherwise high-risk EC resistant to immunotherapy.
LAG-3 is expressed on activated T and NK cells, with immune cells positive for LAG-3 in 24% of ECs and tumor cells positive in 31.6%. TIGIT and PVRIG are distinct inhibitory receptors whose expression is linked to exhausted lymphocyte phenotypes. ECs rank within the top five solid tumors with highest PVRIG and TIGIT expression, and dual blockade of these receptors leads to increased T-cell activation and cytokine release in preclinical models.
While tumor antigenicity and immune-cell infiltration are integral to antitumor immunity, they alone are inadequate for effective malignant cell elimination. ECs generally display an immunosuppressive phenotype and employ multiple mechanisms to evade the immune system. Antitumor effector function results from the summation of inhibitory, activating, and costimulatory signals orchestrated by innate and adaptive immune cells as well as stromal cells.
The variable tumor immune microenvironment among EC molecular and histologic subtypes impacts prognosis and immunotherapy response. POLE mutated and MSI-H/dMMR tumors can be described as immune-included given their high antigenicity and immune cell infiltration. In contrast, CN low and CN high tumors align more with an immune-excluded phenotype based on limited TMB/neoantigen burden and relatively low immune-cell infiltration. Both phenotypes are susceptible to molecular and cellular immunoregulation.
Immunogenic modulation describes how anticancer treatments modify the tumor immune microenvironment and impact the susceptibility of surviving tumor cells to immune-mediated destruction. Standard-of-care therapies including taxanes, platinum-based chemotherapy, radiation, PARP inhibitors, tyrosine kinase inhibitors, HDAC inhibitors, and endocrine therapy have all shown immunogenic modulation effects. Combining these with immunotherapy provides an opportunity to sensitize tumors using drugs with established safety profiles.
Immune-subset conditioning involves three distinct mechanisms: establishing effector cells through vaccines, chemotherapy, and modified NK cells; enhancing effector function with therapeutic cytokines and costimulatory molecules like N-803 and NHS-IL12; and releasing suppression of effector function through inhibitory signaling antagonists targeting TGF-beta, MDSC migration via CXCR1/CXCR2 blockade, and CSF1/CSFR1 axis inhibition to reduce tumor-associated macrophages.
The optimization of systemic antitumor treatment in EC may ultimately require multi-drug regimens due to the complexity of the immune-excluded TIME. Preclinical studies have demonstrated the efficacy of pentatherapy combinations that increase tumor-associated T-cell activity, reduce Treg infiltration, and provide protection during tumor rechallenge. Early phase clinical trials of multi-agent combination immunotherapies are ongoing and have demonstrated safety and the ability to expand immune cell subsets.
Understanding distinctions among EC subtypes is important for optimal precision therapy, especially in mitigating worse clinical outcomes among racial and ethnic minority groups. Non-Hispanic African American patients are more likely to be diagnosed with aggressive non-endometrioid histologies, and researchers have reported conflicting results on ethnic and racial differences in genomic alterations and molecular phenotypes affecting the TIME.
The tumor immune microenvironment represents dynamic cellular and molecular interactions involving tumor antigenicity, immune-cell infiltration, and immunoregulation. Effective therapies must balance tumor cell killing with immunogenic modulation, as radiation and systemic therapy initially increase the dying fraction of a tumor but eventually remove the antigen source required for immune activation. Pre-clinical animal modeling and clinical studies are necessary to optimize therapeutic dosing and sequence.
The development of immune-intact preclinical models including patient-derived organoids and co-culture experiments is needed to address mechanistic questions. Studies using IHC, multiplexed immunofluorescence, flow cytometry, and single-cell RNA sequencing of primary and metastatic EC tumors are required to characterize the immunologic and stromal landscape. Predictive computational models may prove particularly useful for studying rare EC subtypes and designing rational combination immunotherapeutic regimens.