New therapies for clear cell ovarian carcinoma

Introduction

Ovarian clear cell carcinoma is a rare subtype of epithelial ovarian cancer, accounting for 5–11% of total epithelial ovarian cancer cases. Ovarian clear cell carcinoma possesses a number of unique clinical, histopathological, and molecular features which distinguishes it from other epithelial ovarian cancer subtypes. Despite these unique features, the current treatment paradigm for ovarian clear cell carcinoma is based on clinical trials which have historically included more commonly diagnosed epithelial ovarian cancer subtypes, mainly high grade serous ovarian cancer. Given the distinct features of ovarian clear cell carcinoma there is clearly an unmet clinical need for new biomarker driven studies to improve outcomes for women with ovarian clear cell carcinoma. This review will explore current standard of care and new emerging therapies for the treatment of ovarian clear cell carcinoma.

Epidemiology of ovarian clear cell carcinoma

There is geographical variation in the incidence of ovarian clear cell carcinoma; in western populations it accounts for 5–11% of epithelial ovarian cancer cases,1 while in Asian populations it accounts for 10–30%.2 The median age at diagnosis of ovarian clear cell carcinoma is 55 years, which is younger in comparison with the overall epithelial ovarian cancer population presenting at a median age of 64 years.3 Unlike high grade serous ovarian cancer, ovarian clear cell carcinoma is not associated with a family history, but there are other risk factors to consider. The presence of endometriosis is an important risk factor for ovarian clear cell carcinoma and has been associated with a favorable prognosis.4 A pooled meta-analysis of 13 case–controlled studies reported an association between self-reported endometriosis and ovarian clear cell carcinoma with an odds ratio (OR) of 3.05 (95% CI 2.43 to 3.84, p<0.001).4 This was supported by a more recent population-wide study in Denmark in which a diagnosis of endometriosis was associated with increased risk of ovarian clear cell carcinoma with an OR of 3.64 (95% CI 2.36 to 5.38).5 Venous thromboembolism is an independent poor prognostic factor for ovarian clear cell carcinoma. It has been shown that patients with a diagnosis of ovarian clear cell carcinoma have a 2.5-times greater risk of venous thromboembolism despite thromboprophylaxis in comparison to patients with other epithelial ovarian cancer subtypes.6

Prognosis

In comparison to patients diagnosed with high grade serous ovarian cancer, newly diagnosed ovarian clear cell carcinoma are more commonly diagnosed at an earlier stage (stage I/II) (57–81% vs 19–22%).7 In a meta-analysis of 12 randomized control trials in which patients with ovarian clear cell carcinoma and non-clear cell carcinoma epithelial ovarian cancer were treated with platinum-based chemotherapy, no statistically significant difference in overall survival was observed between patients diagnosed with early stage ovarian clear cell carcinoma and high grade serous ovarian cancer (HR 0.87, 95% CI 0.75 to 1.02).8 In a subsequent case series involving patients enrolled in 12 prospective Gynecologic Oncology Group (GOG) trials, some of which were included in the meta-analysis described above, there was significantly improved progression-free survival in patients diagnosed with early stage ovarian clear cell carcinoma compared with high grade serous ovarian cancer (HR 0.69, 95% CI 0.50 to 0.96).9 This improvement in progression-free survival observed in this series did not translate into a statistically significant overall survival benefit for patients diagnosed with stage I/II ovarian clear cell carcinoma compared with high grade serous ovarian cancer (HR 0.76, 95% CI 0.53 to 1.09).9 However, in the advanced (International Federation of Gynecology and Obstetrics (FIGO) stage III/IV) and recurrent settings, ovarian clear cell carcinoma outcomes are much worse compared with other epithelial ovarian cancer subtype. Analysis of patients enrolled in the same case series of GOG studies described above along with the meta-analysis revealed significantly worse overall survival for patients diagnosed with stage III/IV ovarian clear cell carcinoma compared with high grade serous ovarian cancer (HR 1.66, 95% CI 1.43 to 1.91,9 and HR 1.71, 95% CI 1.57 to 1.86,8 respectively). In recurrent disease, a retrospective study compared overall survival of 113 patients with ovarian clear cell carcinoma with 365 patients with serous carcinoma and showed overall survival was significantly worse in those with ovarian clear cell carcinoma (HR 2.3, 95% CI 1.172 to 3.07, p<0.0001).10

Standard of care treatment for ovarian clear cell carcinoma

The current standard of care in newly diagnosed ovarian clear cell carcinoma is debulking surgery aiming to achieve full cytoreduction. Residual disease following surgery has been shown as a poor prognostic indicator in a number of retrospective studies.11 Furthermore, the relative insensitivity to chemotherapy places greater emphasis on the importance of high-quality surgery. Fertility-sparing surgery is an option for stage IA patients on the proviso that adequate staging is conducted.12 13

Adjuvant chemotherapy with carboplatin and paclitaxel is currently recommended for all patients with stage IC2 and above. The role of chemotherapy in stage IA to IC is more uncertain, however the European Society for Medical Oncology-European Society of Gynaecological Oncology (ESMO-ESGO) consensus states that no adjuvant chemotherapy is recommended for patients with stage IA, IB, IC1, and complete surgical staging.14

Ovarian clear cell carcinoma is considered relatively chemo-insensitive in comparison to other subtypes of epithelial ovarian cancer. In 27 patients with stage III/IV and residual disease post-operatively, the response rate to platinum-based chemotherapy was 11.1%.10 The response rate to chemotherapy recurrent and relapsed setting is as low as 6–8%.15 Retrospective data demonstrating the median overall survival of patients with ovarian clear cell carcinoma with ‘platinum-sensitive’ disease was 16 months and for ‘platinum resistance’ it was 7 months,15 highlighting the need for novel targeted treatments for the management of ovarian clear cell carcinoma.

Molecular features of ovarian clear cell carcinoma

In contrast to high grade serous ovarian cancer, wild-type TP53 is usually observed in cases of ovarian clear cell carcinoma, which also harbor BRCA1 and BRCA2 mutations at a much lower frequency (6%, 1% germline and 5% somatic)16 than high grade serous ovarian cancer.17 Truncating loss of function ARID1A somatic mutations are the most common genetic aberration in ovarian clear cell carcinoma, reported in 40–57% of cases.18–20 Somatic mutations affecting PIK3CA (33%), PPP2R1A (7–10%), and KRAS (5%) are also frequently observed in ovarian clear cell carcinoma, along with ERBB2 (14%) and AKT2 (14%) amplifications.18 19 21 22

Future treatment strategies for ovarian clear cell carcinoma

The poor outcomes for women with advanced ovarian clear cell carcinoma together with its unique molecular characteristics and associated cancer phenotype has demanded a stepwise change in how treatment of this epithelial ovarian cancer is approached. Attempts to improve ovarian clear cell carcinoma management has broadly focused on three main strategies:

Immune checkpoint blockade

Targeting angiogenesis

Exploiting ARID1A synthetic lethal interactions.

These novel strategies, summarized in Figure 1, will be explored in turn, including how their combination offers new therapeutic opportunities for ovarian clear cell carcinoma.

Figure 1Figure 1Figure 1

Summary of novel treatment strategies for ovarian clear cell carcinoma together with the pertinent molecular targets for each strategy. CTLA4, cytotoxic T lymphocyte-associated protein 4; PD-1, programmed cell death protein 1; PD-L1, programmed death-ligand 1; VEGF, vascular endothelial growth factor.

Immune checkpoint blockade

Cytotoxic T lymphocyte-associated protein 4 (CTLA4), expressed on regulatory T cells and activated cytotoxic T cells, and programmed cell death protein 1 (PD-1) together with its ligand, programmed death-ligand 1 (PD-L1), expressed on activated T cells and tumor cells, respectively, are common targets for immune checkpoint blockade. The presence or absence of immunoregulatory proteins is not the sole determinant of cytotoxic anti-tumor responses. The cyclic guanosine monophosphate–adenosine monophosphate (GMP-AMP) synthase (cGAS)-stimulator of interferon genes (STING) pathway responds to cytosolic nucleic acids (cNA), a by-product of cancer-associated replication stress and ruptured micronuclei, and is an important mediator of such immune responses.23 24

In addition to the anecdotal evidence from clinical trials supporting the use of immune checkpoint blockade in the treatment, there is a rationale to employ this class of drugs in ARID1A-defective cancers, including ovarian clear cell carcinoma. Loss of ARID1A function has been associated with increased microsatellite instability, tumor mutational burden, and response to immune checkpoint blockade. ARID1A is proposed to recruit the MutS homolog 2 (MSH2) to chromatin, thereby promoting mismatch repair. The loss of this interaction correlated with a microsatellite instability genomic signature with a predominant C>T mutation pattern and increased mutational load across multiple tumor types.25 Consistent with these data, a bioinformatic analysis of cases of gastric cancer found an association between ARID1A mutations and increased immune activity, which was linked to increase tumor mutational burden.26 Tumor cell-line allografts formed by an ARID1A-deficient ovarian cancer cell line in syngeneic mice displayed increased mutation load, elevated numbers of tumour-infiltrating lymphocytes, and PD-L1 expression. Notably, treatment with anti-PD-L1 antibody reduced tumor burden and prolonged survival of mice bearing ARID1A-deficient but not ARID1A-proficient ovarian tumors.25

Unfortunately, immune checkpoint blockade has not yet yielded the same improvements in epithelial ovarian cancer outcomes as it has for other tumor types such as melanoma, lung, and urothelial cancers. However, anecdotal data from the early clinical trials suggest that patients with ovarian clear cell carcinoma may benefit from this approach. In a phase II study investigating nivolumab (anti-PD-1) in platinum-resistant ovarian cancer, one of the two patients achieving complete response had ovarian clear cell carcinoma.27 In the JAVELIN phase IB study, two ovarian clear cell carcinoma patients treated with avelumab (anti-PD-L1) had a partial response.28 The KEYNOTE-100 trial enrolled 19 patients with ovarian clear cell carcinoma who had a response rate of 15.8% when treated with single agent pembrolizumab.29 In addition, the randomized phase II NRG GY003 trial, comparing ipilimumab and nivolumab with nivolumab in epithelial ovarian cancer, demonstrated that patients with clear cell histology were five times more likely to respond than other subtypes.30

Despite early trials signaling the potential efficacy of immune checkpoint inhibitors for the treatment of ovarian clear cell carcinoma, results of studies in this group of patients have been mixed. MOCCA/APGOT-OV2/GCGS-OV3 was a randomized, multicenter (sites from Singapore, South Korea, and Australia) phase II trial comparing durvalumab with physician’s choice of chemotherapy in 47 patients with relapsed ovarian clear cell carcinoma.31 Disappointingly, there was no significant difference in overall response rate (durvalumab 10.7% vs physicians choice chemotherapy 18.8%), progression-free survival (duralumab 7.4 weeks vs physicians choice chemotherapy 14.0 weeks) or disease control rate in either treatment groups; however, ongoing translational research is underway to identify any biomarkers predictive of response.31 NRG-GY016 (NCT03602586) was a single arm, two stage, phase II clinical trial, which assessed the activity of pembrolizumab in combination with the indoleamine-pyrrole 2,3-dioxygenase (IDO) inhibitor epacadostat in patients with recurrent ovarian clear cell carcinoma who had received 1–3 prior treatment lines. Pembrolizumab (200 mg) was administered intravenously every 3 weeks, while epacadostat (100 mg) was administered orally twice a day. Fourteen patients were recruited in to stage 1. In this cohort, the overall response rate was 21% (95% CI 5% to 51%), including partial responses in three patients, progression-free survival was 4.8 months (95% CI 1.9 to 9.6 months), and overall survival was 18.9 months (95% CI 1.9–not reached). This study met the predefined efficacy threshold to progress to stage 2; however, the lack of drug availability led to its termination before enrolment was complete.32 The results of the PEACOCC study (NCT03425565) investigating pembrolizumab in relapsed ovarian clear cell carcinoma appear more promising. This was a phase II trial of pembrolizumab (200 mg three weekly up to 2 years) that recruited 49 patients from the UK with advanced ovarian and endometrial clear cell carcinoma (85% ovarian clear cell carcinoma), who had progressed following at least one line of chemotherapy and were naïve to PD-1/L1 inhibitor therapy. The primary endpoint, progression-free survival rate at 12 weeks, was 43.8%. Best overall response rate was 25% (90% CI 15.1% to 37.3%), including one patient with a complete response and 11 with a partial response. The 1-year duration of response rate was 47.7% (95% CI 14.1% to 75.6%) and after a median follow-up of 2.1 years the median progression-free survival was 12.2 weeks (95% CI 5.90 to 32.9 weeks) and median overall survival was 71.9 weeks (95% 29.1 to 137.6 weeks).33 These results support further investigation of pembrolizumab and other immune checkpoint inhibitors in ovarian clear cell carcinoma. Molecular analysis including PD-L1 status has not been reported to date and will be important in order to help identify which patients obtain meaningful responses and disease control. The outcomes of NRG-GY016, MOCCA/APGOT-OV2/GCGS-OV3, and PEACOCC trials highlight the need for further trials in this area and the importance of randomized trials in rare cancers.

Targeting angiogenesis

Preclinical data have suggested that the progression of ovarian clear cell carcinoma may be dependent in part on angiogenesis mediated through vascular endothelial growth factor (VEGF) and IL-6/STAT/HF1 pathways.34 The binding of VEGF to members of the vascular endothelial growth factor receptor (VEGFR) family results in proliferation, survival, and migration of endothelial cells resulting in angiogenesis, a hallmark of cancer.35 Increased expression of VEGF in both early and late ovarian clear cell carcinoma has been reported and was associated with improved survival.36 Furthermore, in vivo studies showed that bevacizumab (Avastin, Genentech), a humanized monoclonal antibody that targets vascular endothelial growth factor A (VEGFA), inhibited the growth of ovarian clear cell carcinoma xenografts.36 Upregulation of interleukin 6 (IL-6) activating the STAT3/HIF1 pathway, a mediator of angiogenesis, has been demonstrated in ovarian clear cell carcinoma with increased circulating IL-6 in patient and mouse models.37 38

Bevacizumab was the first targeted agent to be licensed as a maintenance therapy following first-line chemotherapy in epithelial ovarian cancer, based on the results of the ICON7 clinical trial.39 ICON7 enrolled patients with both early and late stage ovarian clear cell carcinoma. In the subgroup analysis of patients with ovarian clear cell carcinoma, no significant benefit was observed with the addition of bevacizumab to carboplatin and paclitaxel chemotherapy. It should be noted that this subgroup analysis was underpowered and therefore the activity of bevacizumab for this indication cannot be excluded.39

Nintedanib is a multi-target tyrosine kinase inhibitor blocking key components of the angiogenesis pathway including platelet-derived growth factor receptor (PDGFR) α and β; fibroblast growth factor receptr (FGFR)1, 2 and 3, and FMS-like tyrosine kinase 3 (FLT 3). In a randomized phase II study involving 91 patients with recurrent ovarian clear cell carcinoma who received either nintedanib or chemotherapy (weekly paclitaxel, pegylated liposomal doxorubicin or topotecan), the primary endpoint of progression-free survival was not met (median progression-free survival 2.3 vs 1.9 months). However, there was a trend towards improved survival in the nintedanib cohort with an overall survival of 9 months compared with 4.9 months in the chemotherapy cohort. Disease control rate at 16 weeks was 23.5% and 9.1%, respectively, in the nintedanib and chemotherapy cohort (OR 5.81, 80% CI 1.79 to 18.89).40 The response rates in both arms were disappointingly low (2.1% vs 0%), highlighting the limitations of standard of care chemotherapy and the need for new treatments. While the activity of single agent nintedanib could not be demonstrated, VEGFR inhibitors, potentially in combination with other agents, should not be dismissed.

Sunitinib, a selective inhibitor of tyrosine kinase including VEGF and PDGFR, demonstrated modest benefit for the treatment of ovarian clear cell carcinoma in a phase II study. The overall response rate was 6.7% with a median progression-free survival of 2.7 months in patients who had at least one prior line of platinum-containing chemotherapy.41

Overall, the paucity of data supporting the use of anti-VEGF monotherapy in ovarian clear cell carcinoma cannot be taken to mean lack of efficacy, and clearly further research is required to identify predictive biomarkers and combination therapy should be considered.

Targeting angiogenesis has been shown to augment the anti-tumor effect of immune checkpoint blockade both through changes to the tumor microenvironment and direct effects on immune effector cells.42 Bevacizumab has been combined with atezolizumab in the treatment of both renal cell carcinoma and non-small cell lung cancer. In newly diagnosed stage III or IV ovarian cancer, the addition of atezolizumab to platinum-based chemotherapy and bevacizumab was evaluated in the phase III IMagyn050/GOG 3015/ENGOT-OV39 trial. Although there was no improvement in progression-free survival or overall survival overall (co-primary endpoints), post hoc subgroup analyses reported a numerical increase in progression-free survival with the addition of atezolizumab in non-high grade serous histology which included clear cell. There are several ongoing clinical trials of anti-angiogenic therapy in combination with immune checkpoint blockade recruiting patients with ovarian clear cell carcinoma. The INOVA study (NCT04735861) aims to recruit 38 patients with recurrent or persistent ovarian clear cell carcinoma to be treated with bevacizumab and the anti-PD-1 antibody sintilimab. Preliminary results showed an overall response rate of 40.0% (one complete response, seven partial response; 95% CI 19.1% to 63.9%) and a disease control rate of 75.0% (eight partial response, seven stable disease; 95% CI 50.9% to 91.3%). Recruitment into INOVA is ongoing and the final results are eagerly awaited.43 Lenvatinib is an oral multikinase inhibitor of VEGF receptor, FGFR1-3, PDGF receptor, RET and KIT. The combination of lenvatinib and pembrolizumab has demonstrated improved progression-free survival and overall survival in advanced endometrial carcinoma, and post-hoc analyses suggest clinical benefit in the clear cell histological subtype.44 45 A single arm, phase II, clinical trial of the combination of lenvatinib and pembrolizumab is currently recruiting patients with recurrent or persistent ovarian clear cell carcinoma who have received at least one prior line of platinum-based chemotherapy.46

Exploiting ARID1A synthetic lethal interactions

As previously discussed, the most frequently observed genetic alteration in ovarian clear cell carcinoma is loss of function ARID1A mutations. ARID1A, along with its paralog ARID1B, are components of the canonical brahma-related gene 1/brahma (BRG1/BRM)-associated factor (cBAF) complex,47 one of three eukaryotic SWItch/Sucrose Non-Fermentable (SWI/SNF) complexes described to date. Mutations in the genes encoding the SWI/SNF components are among the most frequently observed in cancer, being found in 20–25% of all cases.48 49 The vast majority of these mutations result in loss of function phenotypes, suggesting they function as tumor suppressor genes in normal cells. Mutations in ARID1A are not unique to ovarian clear cell carcinoma, being observed in approximately 6% of all cancers.50

Ataxia telangiectasia and Rad3-related (ATR) is a critical component of the DNA damage response. Using a short interfering RNA (siRNA) screen, Williamson et al identified a synthetic lethal interaction between ARID1A and ATR. This synthetic lethal interaction was observed in vitro and in vivo.51 Mechanistically, this effect was attributed to the reduced recruitment of topoisomerase 2A (TOP2A) to chromatin in ARID1A-deficient cells, which resulted in an increased reliance on ATR-induced cell cycle arrest to prevent complex chromosomal structures arising during DNA replication being carried through into mitosis and the associated reduction in cell viability51 (Figure 2).

Figure 2Figure 2Figure 2

Schematic showing a proposed mechanistic basis for the ARID1A-ATR synthetic lethal interaction. TOP2A is recruited to chromatin in an ARID1A-dependent manner. In the absence of ARID1A function, reduced chromatin bound TOP2A results in decatenation defects during DNA replication and an increased reliance on the cell cycle checkpoints enforced by ATR. In the absence of ATR inhibition, ATR enables the resolution of these complex chromosomal structures thereby maintaining cell viability. ATR inhibition leads to the accumulation of these complex chromosomal structures, as well as premature mitotic entry, which culminates in chromosomal breakage and ultimately cell death. ARID1A, AT-rich interactive domain-containing protein 1A; ATR, Ataxia telangiectasia and Rad3-related; SMARCA4, SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily a, member 4; SMARCB1, SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily b, member 1; TOP2A, topoisomerase 2A.

The identification of a synthetic lethal interaction between ARID1A and ATR led the development of the ATARI (ENGOT/GYN1/NCRI, NCT04065269) clinical trial. This international, proof-of-concept, phase II, parallel cohort trial is currently assessing the ATR inhibitor ceralasertib activity as a single agent and in combination with olaparib (Lynparza, AstraZeneca) in ARID1A stratified gynecological cancers, including patients with ovarian clear cell carcinoma.52 In the absence of ARID1A mutations, ATR inhibitor sensitivity may be enhanced by combining with PARP (poly (ADP-ribose) polymerase) inhibition, theoretically compensating for the lack of ARID1A mutations. In vitro silencing of ATR has been shown to synergise with PARP inhibition,53 due, in part, to the abrogation of cell cycle checkpoints in response to the DNA damaging effects of PARP inhibition.54 Based on the work by Khalique et al, ARID1A immunohistochemistry was used as a surrogate biomarker55 for ARID1A mutation and used to stratify patients enrolled into the ATARI trial. Patients with histologically confirmed recurrent ovarian clear cell carcinoma, who have received at least one prior line of platinum-based chemotherapy, were enrolled, with those demonstrating loss of ARID1A protein expression by immunohistochemistry receiving ceralasertib monotherapy, and patients with ARID1A proficient tumors receiving ceralasertib in combination with olaparib.

Recruitment to the ceralasertib monotherapy (ARID1A loss) and combination with olaparib (ARID1A no loss) of the ATARI trial is now complete and the results are eagerly awaited. In parallel, integrated forward and reverse translational work is already underway to refine the biomarker for ATR inhibitor sensitivity in order to better understand which patients with ovarian clear cell carcinoma are most likely to benefit from this new class of drugs as well as understand how resistance emerges. There is increasing interest in targeting DNA damage response with ATR inhibitors to enhance response to immune checkpoint blockade. The combination of ceralasertib with the PD-L1 inhibitor durvalumab is planned to be investigated in prior immune checkpoint blockade-treated endometrial cancer (including clear cell) within an additional cohort of the ATARI trial.

Aurora kinase A (AURKA) was identified as an ARID1A synthetic lethal partner through a high throughput drug screen performed in an ARID1A isogenic model of colorectal cancer. ARID1A represses AURKA transcription56 and therefore ARID1A dysfunction results in Aurora A activation together with its downstream effector M-phase inducer phosphatase 3 (CDC25C), leading to pathway addiction.56 57 Treatment with Aurora A inhibitors, both in vivo and in vitro, was associated with reduced viability of ARID1A mutant cancer.56 ENMD-2076 is an orally active AURKA inhibitor, which also targets VEGFR, which was trialed in a phase II study enrolling women with relapsed ovarian clear cell carcinoma.58 Although the trial did not meet its pre-defined primary endpoint (6-month progression-free survival rate of 22%), loss of ARID1A expression, as determined by immunohistochemistry, was associated with improved 6-month progression-free survival compared with those cases which demonstrated ARID1A expression (33% vs 12%, p=0.023).58 Further translational work is needed to refine the biomarker for response to AURKA inhibitors as well as to identify novel AURKA inhibitor combinations to enhance their anti-tumor effect in the context of ovarian clear cell carcinoma.

Another high throughout drug screen, this time performed in 12 ovarian clear cell carcinoma cells, suggested the multi-kinase inhibitor dasatinib selectively kills ARID1A mutant ovarian clear cell carcinoma.59 In vitro, dasatinib sensitivity was observed in ovarian clear cell carcinoma cell lines harboring naturally occurring ARID1A mutations and in ARID1A-deficient tumor cell line xenografts. YES1, a known dasatinib target, gene silencing rescued dasatinib sensitivity in ARID1A mutant ovarian clear cell carcinoma cell lines.59 The efficacy of dasatinib in recurrent ovarian and endometrial clear cell carcinoma associated with an ARID1A mutation was assessed in a single arm phase II study. However, the response rate in this study was 3.8% (1/28) and mean progression-free survival was 2.14 months (95% CI 1.58 to 7.29 months),60 suggesting that other clinical approaches to target ARID1A deficient ovarian clear cell carcinoma are still required.

Polycomb repressive complex 2 (PRC2) functions to repress gene expression via histone H3 trimethylation. EZH2 is a key component of PRC2 and was identified by two separate groups, using orthogonal approaches, as synthetic lethal partner of ARID1A.61 62 GSK126 is a small molecule inhibitor of EZH2, which induces apoptosis in ARID1A mutant cancer cell lines and causes tumor regression of ARID1A-deficient tumor cell line xenografts. This synthetic lethal interaction was attributed to increased expression of the PRC2 target gene PI3K-interacting protein 1 gene (PIK3IP1), an endogenous inhibitor of the PI3K-AKT pathway.61 EZH2 gene silencing was also associated with cell death in a range of ARID1A mutant tumor cell lines, including several ovarian clear cell carcinoma cell lines.62 Patients with ARID1A-deficient cancers, including ARID1A ovarian clear cell carcinoma, are currently being enrolled in a phase II basket study of the EZH2 inhibitor tazemetostat.63 A further study exploring the use of tazemetostat specifically in ARID1A mutant relapsed ovarian clear cell carcinoma is currently active, although recruitment has been halted.64

In the context of ovarian clear cell carcinoma, ARID1A mutations frequently co-occur with activating hotspot mutations in PIK3CA.19 ARID1A dysfunction leads to increased transcription of PIK3CA and results in sustained PI3K-AKT signaling, culminating in an addiction to the pathway in a process coined oncogene addiction.57 65 Therefore, in cases of ARID1A-deficient ovarian clear cell carcinoma, the ensuing dependence on PI3K-AKT signaling could be hypothesized to increase sensitivity to small molecule inhibitors of the pathway. The activity of CYH33, a phosphatidylinositol 3-kinase (PI3K) α selective inhibitor, in cases of relapsed or recurrent PIK3CA mutant ovarian clear cell carcinoma is currently being assessed.66 The interaction between ARID1A, EZH2, and the PI3K-AKT signaling pathway is summarized in Figure 3.

Figure 3Figure 3Figure 3

Schematic illustrating the interaction between ARID1A, EZH2, and the PI3K-AKT pathway. In ARID1A proficient cells, ARID1A function opposes EZH2-mediated repression of PIK3IPI transcription, which inhibits PI3K-AKT pathway signaling. In the absence of ARID1A function, EZH2-mediated repression of PIK3IPI transcription goes unopposed leading to decreased PIK3IPI levels. This in turn leads to increase signaling through the PI3K-AKT pathway culminating in pro-survival and pro-proliferation signaling as well as promoting cell cycle progression. ARID1A, AT-rich interactive domain-containing protein 1A; AKT, AKT serine/threonine kinase; BAD, Bcl-2-associated death promoter; EZH2, enhancer of zeste homolog 2; FOXO1, forkhead box protein O1, MTOR, mammalian target of rapamycin, PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha; PIK3IPI, phosphoinositide-3-kinase-interacting protein 1; PRC2, polycomb repressive complex 2.

As previously discussed, the high incidence of ARID1A mutations across tumor types has spurred a raft of preclinical research to identify ARID1A synthetic lethal partners. The benefit of some of these synthetic lethal interactions to the treatment paradigm of ovarian clear cell carcinoma is yet to be tested in the context of a clinical trial, but provide potential therapeutic strategies. Buffering between overlapping or partially overlapping functions of paralog pairs that control essential functions, such as ARID1A and ARID1B, is one way in which cancer cells can tolerate loss of function mutations.67 ARID1B along with several other SWI/SNF subunits including SMARCA4, SMARCB1 and SMARCE1, have been identified as ARID1A-specific dependencies, offering potential therapeutic targets for ARID1A deficient ovarian clear cell carcinoma.68 Another vulnerability of ARID1A deficient cancers with the potential for therapeutic exploitation for the treatment of ovarian clear cell carcinoma is glutathione metabolism. Cells rely on reactive oxygen species metabolism for survival and glutathione is an important antioxidant which aids this process. Glutathione production relies on cysteine, the levels of which are reduced in ARID1A mutant cells as a result of decreased SLC7A11 expression. This in turn creates a reliance on glutamine-cysteine ligase synthetase (GCL) to generate cysteine, a critical component of glutathione generation. Inhibitors of GCL, including APR246 and PRIMA-1, have been shown to selectively kill ARID1A-deficient cells.69 Thus, targeting glutathione metabolism is a novel potential treatment strategy for ARID1A mutant ovarian clear cell carcinoma. A further ARID1A directed concept is based on the role of ARID1A in ferroptosis regulation and the heat shock factor 1 (HSF-1) pathway. Based on preclinical work, an early phase trial of the HSF-1 inhibitor NXP800 (Nuvectis), which includes a cohort for patients with ARID1A-mutated ovarian clear cell carcinoma, is planned (ENGOTGYN5).70

Conclusion

In the advanced setting, ovarian clear cell carcinoma is characterized by resistance to standard cytotoxic chemotherapy and a poor prognosis. The unique molecular background of ovarian clear cell carcinoma and clinical behaviors distinguishes it from other epithelial ovarian cancer subtypes. Based on strong pre-clinical research, several novel treatment strategies have reached the clinical trial setting.

The overall response rate and progression-free survival benefit observed in the PEACOCC trial suggest that it is justifiable to consider pembrolizumab as a valid treatment option for ovarian clear cell carcinoma, a tumor type for which there are limited effective treatments. There is optimism that further trials of PD-1/PD-L1 inhibitors alone and in combination will help set a new standard of care option for ovarian clear cell carcinoma. The identification of predictive biomarkers, including ARID1A expression, will be of great interest and may provide strategies for augmenting the anti-tumor immune response in the context of ovarian clear cell carcinoma. Indeed, the encouraging preliminary results from INOVA support the strategy of increasing the anti-tumor effect of immune checkpoint blockade through rationally designed combinations.

It is encouraging to see that the extensive work undertaken to identify ARID1A-specific synthetic lethal interactions has already led to several clinical trials for patients with this rare cancer. ATARI represents the first biomarker-directed phase II study in ovarian clear cell carcinoma and provides a template for future clinical trial design. To date, the lack of positive outcome data for trials designed to target ARID1A synthetic lethal interactions may in part reflect our relative naivety in modeling and understanding the cancer phenotype. It is perhaps an oversimplification to predict that a single biomarker, in this case loss of ARID1A function, will be sufficient to give a clinically meaningful response to a given drug. Instead, a combination of biomarkers may be needed, in addition to ARID1A mutation or loss of expression, to impart drug sensitivity. Identifying these biomarker combinations will require close collaboration between pre-clinical and clinical researchers.

The plethora of ongoing clinical trials enrolling patients with ovarian clear cell carcinoma challenges the orthodoxy that it is difficult to initiate trials for rare tumor types. There remains a great deal of work to do in the field, not least the possibility for randomized clinical trials based on the results of those phase II studies which are ongoing or have already reported, despite the challenges in recruitment for rare cancers. Beyond this, the sequencing of different therapies in the treatment paradigm of ovarian clear cell carcinoma needs to be established, while novel treatment combinations have yet to be identified and tested in the context of clinical trials. Overcoming these challenges will require close international collaboration to ensure that the outcomes for patients diagnosed with ovarian clear cell carcinoma are improved.

Ethics statementsPatient consent for publicationEthics approval

Not applicable.

Acknowledgments

The NIHR Biomedical Research Centres at The Royal Marsden NHS Foundation Trust and The Institute of Cancer Research.

留言 (0)

沒有登入
gif