Targeting the immune microenvironment in ovarian cancer therapy—mission impossible?

The use of immune checkpoint inhibitors (ICIs) has been highly successful in various solid tumors but has not yielded significant advancements in the management of ovarian carcinoma (OC). The objective behind this strategy is to target the inhibitory signal that cancer cells use to suppress T-cell activity, thereby allowing the immune system to maintain an effective antitumor response. In OC, evidence has been accumulating regarding the spontaneous T- and B-cell response against tumor-associated antigens such as p53, NY-ESO-1, or LAGE-1. This leads to the generation of multiple reactional immunomodulatory signaling exposed below, to impair antitumor immunity and contribute to an immune-excluded microenvironment. Based on these observations, ICI held considerable promise in OC for restoring antitumor immunity and improving outcomes. Unfortunately, programmed cell death protein 1/programmed death-ligand 1 (PD-1/PD-L1) inhibitors, which were among the first ICI agents tested in OC, have shown very limited activity as single agents, with only ∼10% of patients experiencing an objective response in epithelial ovarian cancer and usually in clear cell carcinomas (CCOC).¹ To enhance efficacy, these agents were combined with chemotherapy and bevacizumab in the hope of priming the immune microenvironment and improving the antitumor response. In the past few years, large phase III trials recruited ∼2000 patients with OC but failed to demonstrate any significant improvement in patients’ outcomes when ICIs were combined with standard chemotherapy or bevacizumab.^2,3 Furthermore, molecular analyses have not been able to identify any relevant subgroup that might benefit significantly from ICI combination therapy. This lack of clear benefits has been a great disappointment for patients and raises concerns about the potential role of ICIs in OC management.

While the current results are discouraging, the potential of immunotherapy in OC management should not be entirely dismissed. It is essential to recognize that the reasons for such failure are complex and hampered by multiple unanswered questions. We aim to address these concerns here while also highlighting the promising novel approaches in the years to come (Figure 1), and the considerations needed to increase the chance of success of immunotherapy in OC.

Figure 1.

Novel immunotherapeutic approaches in the treatment of ovarian cancer (OC).

CAR-T, chimeric antigen receptor-T cell; CTLA-4, cytotoxic T-lymphocyte associated protein 4; LAG3, lymphocyte activation gene 3; NK, natural killer; TIGIT, T cell immunoreceptor with Ig and ITIM domains; TIM3, T-cell immunoglobulin domain and mucin domain 3; TILS, tumor-infiltrating lymphocytes.

First, unlike ‘hot tumors’ such as melanoma or non-small-cell lung cancer, OC generates a highly immunosuppressive tumor microenvironment (TME) composed of a diverse array of immune cells, secreted ligands, and nonimmune features, making it often considered a ‘cold tumor’. Altogether, these components play a crucial role in cancer progression, peritoneal dissemination, and chemotherapy resistance, thereby representing potential therapeutic targets. Among the critical factors contributing to this unique immune context, the OC TME is heavily infiltrated by regulatory T cells (Tregs), M2-polarized macrophages, and other myeloid-derived progenitors, all of which foster immune evasion and promote carcinogenic progression.

Notably, within the OC TME, Tregs exhibit high activation levels; express multiple immune-coregulators including PD-L1, T-cell immunoglobulin domain and mucin domain 3 (TIM3), and lymphocyte activation gene 3 (LAG3); and secrete numerous immunosuppressive agents such as interleukin-10 and transforming growth factor-beta. Moreover, they are found in high levels and exquisitely immunosuppressive in malignant ascites, which often drive OC spreading.⁴ In addition, tumor-associated macrophages (TAMs) are often considered a key actor of antitumor innate immunity; however, they also exhibit strong plasticity depending on the immune context. Within the OC TME, TAMs tend to transform into immunosuppressive mediators (M2-polarized macrophages) impairing effector T-cell activity while favoring Treg recruitment and proliferation. Furthermore, they contribute to protumoral effects by secreting vascular endothelial growth and epidermal growth factor, being involved in extracellular matrix degradation and spheroid formation in ascites, therefore intensively promoting cancer progression. In particular, M2-polarized TAMs are the most abundant cell type within the OC TME and are strongly associated with advanced stages and poorer outcomes.⁵ Likewise, OC cells have been shown to actively recruit myeloid-derived progenitors within the TME. Myeloid-derived progenitors exert potent immunosuppressive signals against effector T cells and NK cells through the secretion of multiple cytokines, especially interleukin-10, along with a wide range of inflammatory mediators. Collectively, the presence of multiple immunosuppressive factors within the OC TME suggests the existence of compensatory pathways to evade inflammatory responses, creating an unfavorable immune context where a single immunotherapeutic approach is unlikely to be successful.

OC is a highly heterogenous disease and considering it as a single entity will inevitably lead to failure. Approximately 70% of OC cases are high-grade serous carcinomas, with ∼40% of them displaying deficient homologous-repair (HR) pathway, making them potential candidates for poly (ADP-ribose) polymerase (PARP) inhibitor (PARPi) treatment. The remaining 30% consists of various histological subtypes, including clear-cell ovarian cancers, mucinous ovarian cancers, and low-grade serous carcinomas. Each of these subtypes is associated with distinct phenotypic and molecular characteristics, leading to substantial differences in their therapeutic strategies. Moreover, their immune contexts differ drastically from one another, while still being highly immunosuppressive.6, 7, 8 CCOC appears as the best promising subtype for ICI treatment, as suggested by subgroup analyses from the aforementioned phase III trials. However, it is worth noting that in the MOCCA trial, single-agent durvalumab (PD-1 inhibitor) did not improve progression-free survival compared with single-agent chemotherapy in patients with CCOC, and the optimal use of these drugs in this histology remains elusive.⁹

Among high-grade serous tumors, nearly half of patients present a proficient HR pathway, making them poor candidates for PARPi. Unfortunately, this subgroup experiences high recurrence rates and dismal overall prognosis and therefore represents a population with a critical unmet medical need. Interestingly, they may be the most relevant subgroup for ICI therapy. The interim results of the DUO-O trial (ASCO 2023) reported provocative findings in this population.¹⁰ This trial investigated the addition of durvalumab to olaparib and bevacizumab as maintenance therapy in patients with advanced OC. Although the trial was not specifically designed to address this question, it demonstrated that combining PARPi with ICI significantly improved outcomes in the HR pathway population, who typically do not benefit from PARPi alone. Alongside preclinical data suggesting that PARPi induces immunogenic cell death through cyclic GMP–AMP synthase/stimulator of interferon genes (cGAS/STING) activation, the DUO-O results also suggest potential synergistic activity.¹¹ As a result, a new generation of dedicated clinical trials is emerging, asking the accurate questions to the appropriate population while possessing enough power to potentially change treatment paradigms (ClinicalTrials.gov ID: NCT03602859, NCT03522246, and NCT03740165).

While ICIs have been disappointing in OC thus far, other immunotherapeutic approaches show promise. Among solid tumors, OC is also benefiting from the development of adoptive cellular therapies, including chimeric antigen receptor-T cell therapies.¹² This strategy involves infusing autologous immune cells that have been stimulated and expanded ex vivo to target a highly specific antigen, minimizing off-target effects. Promising targets are MUC16 and FOLR1, which have demonstrated strong spontaneous T-cell activation against cancer cells and promising responses in vivo. Another promising immunotherapeutic strategy relies on stimulating the patient’s own antitumor immune response through tumor-associated antigens or engineered antigen-loaded cells, such as dendritic cells.¹³ These therapies have the advantage of stimulating both the innate and adaptative antitumor immunities while using highly specific antigens to avoid severe systemic inflammation. Vaccine therapies are rapidly progressing in the OC landscape and early clinical trials have reported strong specific immune responses and prolonged remission in heavily pretreated OC cases.^14,15 Lastly, bispecific T-cell engagers represent an innovative way to redirect polyclonal T cells to the tumor by the formation of immune synapses. In OC, several compounds have been engineered targeting MUC16, WT1, or claudin 6 and have shown promising antitumor activity.¹⁶ Each of these therapies holds significant potential for incorporation into OC therapeutic strategy in the years to come. However, further efforts are required to limit off-target toxicity and enhance the accessibility of these extremely expensive therapeutics.

Taken together, despite its spectacular complexity and heterogeneity, targeting the OC TME remains highly relevant and promising. However, we face multiple challenges. First, gaining a better understanding of the spatial and temporal interactions will be critically important to anticipate immune evasion mechanisms and offer the most effective combination strategies. In this regard, novel cutting-edge technologies such as single-cell sequencing and spatial molecular profiling are expected to greatly improve current knowledge in the field. Second, inaccurate trial designs have crucially hampered the optimization of therapeutic strategies among subgroups. Certainly, the future of ovarian cancer immunotherapy lies in tailored strategies that consider individual immune characteristics coupled with novel drugs better suited to the unique environment of OC. Integration of translational endpoints into the design of clinical trials is crucial for achieving a truly personalized and effective immunotherapeutic strategy for OC patients. In addition, collaborative efforts are needed to study and advance the understanding of understudied rare subtypes, providing the best chance for meaningful progress in the treatment of OC. Lastly, while the new generation of immunotherapeutic agents is progressing rapidly, it also brings some novel technical challenges and adverse events. Considering OC’s likely status as an uncurable disease, treatment accessibility and quality of life should remain essential objectives.