Current status of cancer immunotherapy for gynecologic malignancies

Abstract

Recent cancer immunotherapy development with immune checkpoint inhibitors has shown durable clinical responses in a wide variety of tumor types. These drugs targeting programmed cell death 1, its ligand programmed death ligand 1 and cytotoxic T cell lymphocyte-associated antigen 4 have revolutionized the field of cancer treatment. It is of significant interest in optimizing the immunotherapy for cancer patients beyond the conventional treatments such as surgery, chemotherapy and radiation. Many clinical trials evaluating the safety and efficacy of various combined regimens with immune checkpoint inhibitors have been reported and are in progress. Among gynecologic malignancy, endometrial cancers have distinct subtypes with microsatellite instability-high status and polymerase ɛ mutation. These types have been shown to immunogenic tumors and appropriated candidate for immune checkpoint inhibitors. Also, recurrent cervical cancer showed a promising objective response with single anti-PD1 Ab treatment. Despite their definite outcome and considerable potential of immunotherapy, not all patients received a survival benefit and further understanding of human tumor immunology is essential to improve this type of therapy. In this review, we have summarized the updated results of clinical trials of cancer immunotherapy for gynecologic malignancies and discussed the future perspectives.

cancer immunotherapy, immune checkpoint inhibitors, PD-1, PD-L1

Introduction

Cancer immunotherapies have drastically changed the landscape of oncology. In recent decades, inducing or amplifying the antitumor immune response has been the fundamental strategy of cancer immunotherapy. However, the clinical efficacy of this strategy has been limited (1). Since the discovery of immune checkpoint molecules, drugs targeting programmed cell death 1 (PD-1) (2) (3), its ligand programmed death ligand 1 (PD-L1) (4) and cytotoxic T cell lymphocyte-associated antigen 4 (CTLA-4) (5) have provided durable clinical responses against a wide spectrum of tumors. These drugs, termed immune checkpoint inhibitors (ICIs), facilitate endogenous antitumor immunity, and because of their broad spectrum of activity, ICIs are regarded as key treatments for cancer therapy, even in patients with advanced inoperable cancers (6). Recently developed cancer immunotherapy strategies have revolutionized the treatment of cancer. However, further understanding of tumor immunology is essential for improving clinical outcomes. In this study, we have reviewed the current status of cancer immunotherapy for gynecologic malignancies and discussed the future perspectives.

Interplay between tumor cells and immune cells

Unlike traditional cancer treatments such as chemotherapy and radiation, which directly target cancer cells, immunotherapy is an innovative treatment strategy that modulates or manipulates the immune system to attack cancer (6). This concept was derived from observations and murine model studies of the correlation between the immune system and cancer (7). For example, in various cancers (e.g. melanoma, colon cancer, lung cancer and ovarian cancer), the number of tumor-infiltrating lymphocytes (TILs) prior to initial treatment was correlated with the prognosis after conventional treatments including surgery (8–11). In patients with colon cancer, as an example, the counts of TILs [e.g. CD3⁺ (pan T-cell marker) and CD8⁺ (cytotoxic T cell marker) T cells] were confirmed to be significantly correlated with prognosis after surgery in an International Collaborative Study (International ‘Immunoscore’ validation) (8). Moreover, the inclusion of the immunological status in the current tumor–node–metastasis staging classification may improve the clinical management of patients with colon cancer. A global consensus Immunoscore study validated the prediction of the risk of recurrence and survival using Immunoscore. In gynecologic malignancies, an early study reported that the presence of TILs (CD3⁺) correlated with improved overall survival (OS) in stage III and stage IV ovarian cancers (12). We then applied the TIL status (CD4⁺, CD8⁺, FoxP3⁺ Tregs) in patients with advanced cervical cancer prior to concurrent chemoradiotherapy (CCRT) and demonstrated that the TIL count prior to CCRT was a prognostic biomarker for this malignancy (9). These observations represent examples of the complex interplay between tumors and immune cells and illustrate the utility of the immune status as a biomarker for treatment.

Immunosuppressive mechanisms of the tumor microenvironment

On the basis of the aforementioned observations and other basic studies, various types of immunotherapies have been developed in recent decades to spur the immune system to activate immunomodulatory mechanisms (13). However, cancer cells detected in the clinic have already evaded the host immune system through immunosuppressive mechanisms (14). These mechanisms include the production of immunosuppressive cytokines by cancer cells, induction of immunosuppressive cells such as Tregs and induction of inhibitory molecules such as PD-1 on activated T cells or PD-L1 on cancer cells or antigen-presenting cells (APCs) (15). Under these circumstances, immunosuppressive mechanisms trigger changes in the immunogenicity of the tumor and the acquisition of immune resistance. This is a major reason why immunotherapies that only stimulate the autologous immune system exhibited little clinical effect in prior studies, and immunosuppressive strategies have been introduced to overcome these hurdles (6). One such approach is the blockade of inhibitory molecules such as PD-1, PD-L1 and CTLA-4.

Immune checkpoint inhibitors

Anti-PD-1/anti-PD-L1 therapy

The immune checkpoint molecules are located on the surface of T cells (PD-1) or tumor cells and APCs (PD-L1 and PD-L2), and they are the targets for inhibiting the overactivation of T cells (16). PD-L1 and PD-L2 bind to PD-1, which is expressed on CD4 and CD8 T cells, thereby inactivating these cells. Under normal circumstances, these checkpoint molecules prevent autoimmune disease and protect healthy tissue against damage after activation of the immune system (17). Within the tumor microenvironment, checkpoint molecules prevent T cells from attacking the tumor, weakening the ability of the immune system to recognize and destroy tumor cells (18). By blocking these molecules using monoclonal antibodies (mAbs), the immune response of T cells can be largely activated, thereby reestablishing antitumor immune effects (2).

Because of the dramatic clinical effects of ICIs, several clinical trials have been launched to assess their effects against various cancers. Inhibition of the PD-1/PD-L1 pathway can medicate the rejection of established cancers that are refractory to other treatment modalities (19–21). To date, three mAbs targeting PD-1 (pembrolizumab, nivolumab and cemiplimab) and three mAbs targeting PD-L1 (atezolizumab, durvalumab and avelumab) have been approved by the US Food and Drug Administration (FDA) across 17 different types of advanced inoperable cancers in the first- and/or later-line settings. Among anti-PD-1 mAbs, nivolumab was the first PD-1 inhibitor to enter clinical testing in a first-in-human trial in 2006 (22). Other than Hodgkin’s lymphoma, the overall response rates of malignancies to single-agent ICIs do not exceed 30% in melanoma and 10–20% in other cancer types (23). At present, hundreds of clinical trials conducted globally have described better outcomes using anti-PD-1/anti-PD-L1 mAbs.

Anti-CTLA-4 therapy

Exhausted or dysfunctional TILs express PD-1 and CTLA-4; however, these molecules have distinct inhibitory mechanisms. CTLA-4 was first cloned in 1987 (24), and Leach et al. proved that this molecule inhibits the early activation of T cells and that blockade of this molecule induced antitumor responses in murine models (25). As a negative regulator, CTLA-4 restricts the initial priming of T cells (naïve or memory) in secondary lymphoid organs by antagonizing the master co-stimulatory molecule CD28, which shares a ligand with CTLA-4 (26). An anti-CTLA-4 antibody (ipilimumab) was the first-in-class ICI to be successfully tested in melanoma, and it was approved by the FDA in 2011 for use in patients with metastatic melanoma (5). There is growing evidence that anti-CTLA-4 antibodies are associated with greater immune related toxicity and lower response rates than anti-PD-1/anti–PD-L1 therapy (5). The different toxicities can be explained by some subclinical results. For example, the phenotypes of knockout mice ablating either CTLA-4 or PD-1 are different. CTLA-4 knockout mice die of the extremely severe autoimmune disease within 3 weeks following birth (25). In contrast, PD-1 knockout mice are apparently in good health after birth. Mild and sporadic autoimmune diseases gradually begin to arise in the aged mice ~1 year following birth (27). These phenotypes demonstrated that CTLA-4 is negatively regulating a broad spectrum of immune reactions, while PD-1 is suppressing a limited phase of immune reaction.

Biomarkers of ICIs

The efficacy of ICIs is associated with several biomarkers, such as high levels of T-cell infiltration, PD-L1 expression, microsatellite instability-high status (MSI-H) and the tumor mutation burden (28). Because TILs usually recognize tumor mutation-derived peptide as foreign antigens (neoantigens), high levels of TILs indicate strong tumor immune reaction within tumor microenvironment. Several clinical trials evaluated whether PD-L1 expression could be a predictive biomarker (22). However, a significant association between clinical outcomes and PD-L1 expression was validated in limited tumor types including ovarian cancer and cervical cancer (29). Recently published report demonstrated that PD-L1 gene amplification but not polysomy was associated with response to anti PD-1 mAb (nivolumab) therapy among non-small cell lung cancer patients. Therefore, validation study including other tumor types is required for verifying preexisting PD-L1 expression as a reliable biomarker (30).

A well-known phase II study conducted by Le et al. (28,31) demonstrated that the efficacy was high in MSI-H or mismatch repair-deficient (MMR-D) solid tumors. Whole-exon sequencing revealed that the average number of somatic mutations per tumor with MMR-D was significantly correlated with the clinical outcome. According to this result, the FDA approved pembrolizumab for unresectable/metastatic solid tumors with MSI-H or MMR-D in adult and children. These indicated that biomarkers rather than tumor types themselves define treatment outcomes. Clinicians and researchers are now working on the detection of more efficient biomarkers to select appropriate candidates for immunotherapy.

Other type of immunotherapies

Adoptive cell therapy

Adoptive cell therapy refers to the treatment of tumor cells by infusing immunologic effector T cells with or without modified and amplified genes, and it has emerged as a hot research direction and important means of tumor treatment because of its strong specificity (32). In non-specific therapy, immune cells are activated by lymphocytes or cytokines in peripheral blood, and they are considered to have the ability to ablate a variety of tumors. The immune cells in specific adoptive cell therapy are induced by specific tumor antigens and stimulating factors, including TIL therapy, T-cell receptor-T cell therapy and chimeric antigen receptor T cell (CAR-T) therapy. Rosenberg and Restifo (32) have applied this concept since the 1990s and demonstrated that 49–72% of patients with melanoma achieved complete remission (CR). This type of therapy has strong specificity, strong targeting efficiency, high lethality, few side effects and no drug resistance, and thus, it can be used in patients with certain advanced cancers or those who did not respond to other curative treatments. TIL therapy is only used in patients with melanoma and cervical cancer because of difficulties in separating and collecting these cells, and CAR-T therapy is mostly used to treat blood cancers (33).

CAR-T therapy

Engineered T cells that express antigen receptors (CAR) targeting tumor cells were first established in 1993, but their true potential as anticancer agents was not revealed until 2010 (34). CAR-T-based therapies depend on the isolation, ex vivo manipulation and expansion of antigen-specific T cells, which are subsequently used in the same patients. This method has displayed several potential advantages over conventional therapies, including specificity, rapidity, high success rates and long-lasting effects. The two presently approved therapies, as well as the most current clinical trials based on the technology, are directed against CD19, a classical B-cell malignancy-associated antigen (34,35).

Immunotherapy for gynecologic malignancies

Endometrial cancer

Endometrial cancer is the most common gynecologic malignancies in Western countries and in Asian countries including Japan and a major immunogenic cancer against which ICIs have clinical efficacy. Approximately 25–30% of primary endometrial cancers are MSI-H, and 13–30% of recurrent endometrial cancers are MSI-H or MMR-D (28,31). As reported in The Cancer Genome Atlas Network, primary endometrial cancers have four distinct molecular subtypes, namely polymerase ɛ (POLE) mutant/ultra-mutant, MSI-H, copy number low and copy number high (36). Among these subtypes, endometrial cancers with first two subtypes (POLE mutant/MSI-H) are highly immunogenic, and they carry more neoantigens, resulting in increased TIL counts and a compensatory upregulation of immune checkpoint molecules (37).

In the phase II KEYNOTE-158 trial, pembrolizumab was associated with an objective response rate (ORR) of 57% in the MSI-H/MMR-D endometrial cancer cohort, consistent with its previously reported efficacy (29). In a non-randomized phase I study (NCT02715284), dostarlimab (anti-PD-1 antibody, not yet approved by the FDA) was linked to an ORR of 42% in patients with MSI-H advanced endometrial cancer with nine durable CR cases (38). In addition, in the efficacy analysis of KEYNOTE-146 (a phase Ib/II study of lenvatinib and pembrolizumab in patients with advanced solid tumors), the ORR among 108 patients with previously treated advanced endometrial cancer was 41%. Lenvatinib is an oral multikinase inhibitor that blocks VEGFR1–3, FGFR1–4, KIT, RET and PDFGRa. The ORRs for 94 patients with MMR-proficient endometrial cancer and 11 patients with MMR-D endometrial cancer were 38 and 64%, respectively, and responses were observed regardless of the MSI status, PD-L1 status or histology (39). On the basis of this result, in 2019, the FDA granted accelerated approval to lenvatinib in combination with pembrolizumab for the treatment of advanced endometrial cancer that is not MSI-H or MMR-D and that progressed following prior therapy. This combination is currently being assessed in two ongoing phase III studies: lenvatinib plus pembrolizumab versus doxorubicin or weekly paclitaxel in advanced endometrial cancer that was previously treated with platinum-based therapy (KEYNOTE-775, NCT03517449) and first-line lenvatinib plus pembrolizumab versus carboplatin and paclitaxel chemotherapy in advanced endometrial cancer (LEAP-001, NCT03884101).

To obtain better clinical outcomes for ICIs in endometrial cancer, a phase II study of combination immunotherapy with durvalumab (anti-PD-L1 ab) and poly (ADP-ribose)-polymerase (PARP) inhibitor (olaparib) is actively enrolling (NCT03951415). In endometrial cancer, ICIs has relatively modest effect. Numerous clinical trials testing combination regimen with targeted therapies or chemotherapy are currently ongoing.

Ovarian cancer

Epithelial ovarian cancer (EOC) is a most lethal disease among gynecologic malignancies and an immune-modulating malignancy. In serous ovarian cancer, the most common histological type of EOC, TIL counts were reportedly correlated with the prognosis after conventional treatment (40). We previously found that tumor infiltration by T cells was relatively weak in ovarian clear cell carcinoma (OCCC), which comprises ~25% of Japanese ovarian cancers. OCCCs produce extremely high amounts of the immunosuppressive cytokines IL-6 and IL-8 in an NF-κB-dependent manner and induce the accumulation of immunosuppressive cell populations, which might cause immunosuppression and reduce subsequent T-cell infiltration in tumors (41). An immunoreactive molecular subtype was identified by The Cancer Genome Atlas Network that displayed an enrichment of genes and signaling pathways associated with immune cells and that was associated with significantly longer OS (42).

In a phase II trial of nivolumab in patients with recurrent ovarian cancer conducted in Japan, 2 of 20 patients exhibited complete response, and the disease control rate was 45%, including a durable CR rate of 10% (21). In the KEYNOTE-100 (NCT0264061) phase II study, which enrolled 378 patients with advanced EOC, pembrolizumab was linked to a response rate of 9%, and the response differed depending on PD-L1 expression (43,44). The response rate in these studies was inferior to that in patients with EOC who display intraepithelial TILs as reported across various studies. Concerning other combination immunotherapies, ipilimumab plus nivolumab is being tested in EOC (NCT03342417, NCT02498600 and NCT03355976), and this regimen was reported to produce statistically higher response rates (31.4%) than nivolumab alone (12.2%) in a phase II study of 100 patients with persistent or recurrent ovarian cancer (45). The following section discusses combination immunotherapy using bevacizumab and PARP inhibitors (PARPis).

Bevacizumab is the first-line treatment for advanced or recurrent EOC in combination with paclitaxel and carboplatin based on the result of a phase III trial (46,47). A previous phase I trial of durvalumab illustrated that anti-VEGF therapy may enhance responses to ICIs (48). A phase II trial using the combination of nivolumab and bevacizumab reported clinical efficacy in women with recurrent EOC, with an ORR of 29% and a median progression-free survival (PFS) of 8.1 months (49). The ORR was 40% in platinum-sensitive patients and 17% in platinum-resistant participants. On the contrary, the phase III JAVELIN Ovarian 200 study, which compared pegylated liposomal doxorubicin (PLD) alone with the combination of PLD and avelumab, did not observe a statistically significant difference in PFS or OS between the arms (50,51).

PARPs are proteins involved in DNA damage repair. PARPs are critically important in cells in which the homologous recombination repair pathway has been inactivated, such as BRCA-deficient cells (52). PARPis block DNA repair in tumor cells, and their use could increase the mutational burden in patients with EOC, which is a predictive biomarker for response to ICIs. PARPis are currently being examined in combination with CTLA-4 blockade (NCT02571725 and NCT02485990) or PD-1/PD-L1 blockade (NCT03522246, NCT03642132, NCT02484404, NCT02657889, NCT03602859, NCT03737643 and NCT03740165). The results of the phase II MEDIOLA trial, which is assessing the combination of olaparib and durvalumab, and the phase I/II TOPACIO trial examining the combination of niraparib and pembrolizumab are awaited (51).

Cervical cancer

Cervical cancer has a specific human papillomavirus (HPV) etiology and relatively high tumor mutation burden, and immunotherapy is considered a treatment option (29).

From the Cancer Genome Atlas Network, detailed genomic analyses revealed amplifications of PD-L1 and PD-L2 in cervical cancer tissues, supporting the ICI approach (53). Although monotherapy with ICIs has yielded modest clinical effects, the KEYNOTE-158 phase II clinical trial demonstrated promising outcomes. In that study, 98 patients with recurrent/metastatic cervical cancer who experienced progression during treatment on intolerance to at least one line of standard chemotherapy were recruited. The response rate was 12.2% [three CRs and nine partial responses (PR)] in patients with PD-L1-positive (29). From this study, the FDA approved pembrolizumab as a second-line agent, making this drug the first ICI approved for treating recurrent cervical cancer. In another phase I/II clinical trial of nivolumab monotherapy for recurrent or metastatic cervical cancer (including vaginal and vulvar cancer, NCT02488759), the ORR was 26.3%, and the median OS was 21.9 months (54). Several phase III clinical trials of ICIs in combination with cytotoxic chemotherapy or targeted therapies are ongoing (NCT03635567, NCT03556839 and NCT03257267).

It is well known that radiation therapy impacts the tumor immune microenvironment and modulates immune system. Several clinical trials are ongoing, combination regimen with concurrent chemoradiation therapy plus ICIs (pembrolizumab or durvalumab) for advanced cervical cancer patients (NCT03830866 and NCT04221945) (55).

Other immunological approaches include adoptive cell therapy, which involves the systemic infusion of therapeutic T cells. Previous studies reported that three of nine patients experienced objective tumor responses (two CRs and one PR) (56). Cultured TILs recognize tumor-specific HPV-E6 or HPV-E7 and tumor-specific neoantigens (57), and the administration of cultured TILs has led to clinical responses in patients with chemotherapeutic resistance (56,58). Because metastatic or recurrent cervical cancers are difficult to treat, this immunotherapy approach is still limited, but further investigation is required.

Future directions

The durable clinical responses of ICIs have provided evidence of their utility in a variety of advanced/metastatic cancers, and they produced improved clinical outcomes compared with conventional chemotherapy. However, the efficacy of immunotherapy including ICIs is not satisfactory, and improving the efficacy of ICIs to achieve ‘precision’ antitumor therapy is a major direction of immunotherapy research.

In addition to the aforementioned combinations of immunotherapy and conventional treatments, a newly introduced approach is the use of ICIs in the neoadjuvant setting. The known immunologic effects of the PD-1 pathway on T-cell priming, effector functions and exhaustion suggest that neoadjuvant checkpoint blockade has clinical utility based on a different mechanism than neoadjuvant chemotherapy (59). Neoadjuvant systemic therapy is predicted to result in the enhanced detection and killing of micrometastatic tumors, which could be ultimately the source of relapse. In a study in which 21 patients with non-small cell lung cancer, neoadjuvant nivolumab was revealed to have fewer side effects, and the major pathological response rate was 45% (60). These promising results indicate that the use of neoadjuvant immunotherapy with ICIs can be expanded to other malignancies.

Concluding remarks

The treatment of cancer is drastically shifting from traditional treatments to more durable and effective treatments that enhance immune activity. The remarkable clinical efficacy of ICIs against various cancers has shifted the paradigm of cancer immunotherapy. Understanding the mechanisms underlining the antitumor effects and resistance (e.g. cancer cell-induced immunosuppression) of immunotherapy in patients and combining currently available therapies are critical for improving treatment outcomes. Further studies utilizing multiomics analyses and microbiota combined with detailed bioinformatics analyses may facilitate the development of precision medicine-based cancer immunotherapy in the near future.

Acknowledgment

We thank Ms Keiko Abe and Ms Hitomi Nishino for editing manuscript. We thank Joe Barber Jr, PhD, from Edanz Group for editing a draft of this manuscript.

Funding

This work was supported by JSPS KAKENHI grant (19K09832) to H.N.

Conflict of interest statement

The authors declare no conflict of interest.