Akhil Santhosh   Nihanthy Sreenath

For years, the therapeutic armamentarium in platinum-resistant ovarian cancer (PROC) has been marred by limited efficacy. The addition of bevacizumab remains the last incremental advance, which was more than a decade ago, with the AURELIA trial showing a statistically significant progression-free survival (PFS) of 6.7 versus 3.4 months (HR, 0.48[95%CI, 0.38 to 0.6]).¹

No significant improvement in overall survival (OS) was noted. Ovarian cancer was historically regarded as an immune desert, with a very low tumor mutational burden, limited immune infiltration and a hostile microenvironment invigorated with immune-suppressive cells. Earlier trials of immune checkpoint inhibitors in PROC were met with limited success. The JAVELIN Ovarian 200 phase 3 trial, comparing avelumab alone vs addition of pegylated liposomal doxorubicin failed to meet PFS and OS endpoints.²

Even though the phase 3 NRG GY003 study showed encouraging responses for nivolumab and nivolumab-ipilimumab combinations, the phase 3 NINJA trial which tested nivolumab versus chemotherapy (gemcitabine/liposomal doxorubicin) was negative.³ 

ICIs have shown mechanistic synergy with PARP inhibitors in preclinical models, but this has not been replicated to complete fruition in larger trials, as noted with the negative outcomes from TOPACIO/KEYNOTE 162 and MOONSTONE trials testing pembrolizumab + niraparib and dostarlimab + niraparib respectively. Negative results and limited efficacy may be explained by the inability to identify predictive biomarkers. Issues with patient selection and inadequate stratification of factors like BRCA, HRD, TMB and “immune score” expression profiles have further contorted interpretation of meaningful data. Also, most of the studies in this setting suffer from suboptimal designs, weak comparator arms, insufficient power and inadequate endpoints. Till date, there remains no predictive biomarker for response to immunotherapy in PROC. PDL1 testing suffers from methodological inconsistencies, varying cut-offs and temporo-spatial heterogeneity.

On February 10th 2026, FDA approved pembrolizumab + chemotherapy with or without bevacizumab for treatment of patients with PROC based on the KEYNOTE B96 trial. At primary analysis, the pembrolizumab arm improved PFS versus comparator (8.3 vs 6.4 months, HR 0.7 p=0.0001). Landmark overall survival was in favor of the pembrolizumab arm (18.2 vs 14 months, HR 0.76, p=0.0055).

The approval is only for patients with a PDL1 score ≥1%. Exact OS numbers in CPS<1 population have not yet been revealed. Higher discontinuation rates were observed in the pembrolizumab arm with hypothyroidism being the most common immune related adverse event (17.8%), but there was no increase in fatal adverse events. The study was presented at ESMO 2025, but the full peer reviewed manuscript is not yet out and subgroup data needs more granularity. So, the obvious question arises, how did KEYNOTE-B96 succeed, while the others miserably failed to make a mark.

For starters, KEYNOTE B96 is a large international trial, led by the prestigious ENGOT group. Backbone choice was spot on, in-fact weekly paclitaxel + bevacizumab had the best response rates in the AURELIA study. Low dose paclitaxel stimulates MDSC differentiation to dendritic cell activation. Addition of bevacizumab normalizes tumor vasculature, reduces T-reg population and enhances T-cell infiltration. Also, unlike other trials, there was biomarker enrichment (PDL1 CPS ≥1). There was adequate power, with validated endpoints like OS, particularly relevant for immunotherapy. Already, there is the backing of a good phase 2 data for weekly paclitaxel with pembrolizumab in PROC with an ORR of 51.4% and median PFS of 7.23 months, with a manageable toxicity profile.⁴

Also, the study population is aptly selected for an immunotherapy trial, with patients receiving only up to 2 prior lines of chemotherapy, 73% receiving concomitant bevacizumab and >70% being enriched for PDL1. Also, the immunomodulatory property of weekly paclitaxel + bevacizumab was well documented in the subgroup analysis from the ENGOT-ov 34 trial in combination with atezolizumab.⁵

The hazards with or without bevacizumab appear similar, but progression events were lower with bevacizumab. It would be worthwhile to look at the KM curves with/without bevacizumab use when the full manuscript comes out. In case of contraindications to paclitaxel or if the patient denies intravenous chemotherapy, a reasonable alternative is oral metronomic cyclophosphamide in combination with pembrolizumab and bevacizumab as it has demonstrated encouraging activity in phase 2 trials. Interestingly, a sub-analysis of OS data indicates that patients with a platinum-free interval of < 3 months were most likely to benefit from the addition of immunotherapy. Given the fact that the real-world patient population is bound to be significantly different from trial patients, the risk-benefit ratio of this novel combination should be evaluated with caution, particularly given the risk for immune related adverse events in this heavily pre-treated population.  Recently, 2 studies reported an overall survival benefit in PROC, namely MIRASOL and ROSELLA. MIRASOL evaluated mirvetuximab soravtansine in FR-alpha high tumors (≥75% viable tumor cells with 2+ or 3+ expression), demonstrating an OS of 16.46 vs 12.5 months. Also, combinations of mirvetuximab with other therapies have also been explored, showing good synergy with pembrolizumab, paving way for future chemo-free regimens in PROC. 

ROSELLA has beautifully exploited glucocorticoid signaling in ovarian cancer, with relacorilant +nab-paclitaxel attaining an OS of 15.97 months versus 11.5 months in chemo alone arm. This may be the preferred option for those PDL1 negative patients who are not eligible for the KEYNOTE B-96 regimen. Other potential options in PROC include targeting HER2 with trastuzumab deruxtecan and new drugs in pipeline like raludotatug deruxtecan targeting CDH6 and puxitatug samrotecan targeting B7-H4.  It is high time oncologists realize that PROC has been abruptly redefined as a heterogenous and biomarker stratified disease. Incorporating comprehensive biomarker testing early on in the treatment course and diligently planning therapeutic sequencing reserving conventional cytotoxics as later options will be the way forward. Operationalizing these advances, defining rational algorithms, ensuring global access and translating these data-sets into meaningful survival gains are the next real challenges that await gynecological oncologists worldwide. For now, it is hard to say that immunotherapy has conquered the final frontier, but we sure are making meaningful in-roads and beginning to fathom the intricate complexities surrounding this disease. With three options showing overall survival benefit, we hope the oncology fraternity will be spoilt for choices when it comes to treating PROC.

References

1. Pujade-Lauraine E, Hilpert F, Weber B, et al. Bevacizumab combined with chemotherapy for platinum-resistant recurrent ovarian cancer: The AURELIA open-label randomized phase III trial. J Clin Oncol. 2014;32(13):1302-1308. doi:10.1200/JCO.2013.51.4489

2. Pujade-Lauraine E, Fujiwara K, Ledermann JA, et al. Avelumab alone or in combination with chemotherapy versus chemotherapy alone in platinum-resistant or platinum-refractory ovarian cancer (JAVELIN Ovarian 200): an open-label, three-arm, randomised, phase 3 study. Lancet Oncol. 2021;22(7):1034-1046. doi:10.1016/S1470-2045(21)00216-3

3. Hamanishi J, Takeshima N, Katsumata N, et al. Nivolumab Versus Gemcitabine or Pegylated Liposomal Doxorubicin for Patients With Platinum-Resistant Ovarian Cancer: Open-Label, Randomized Trial in Japan (NINJA). J Clin Oncol. 2021;39(33):3671-3681. doi:10.1200/JCO.21.00334

4. Wenham RM, Buras AL, Gordon SW, et al. A phase 2 study of pembrolizumab and weekly paclitaxel for platinum-resistant epithelial ovarian cancer. International Journal of Gynecological Cancer. December 2025:102856. doi:10.1016/j.ijgc.2025.102856

5. Marmé F, Harter P, Redondo A, et al. Atezolizumab versus placebo in combination with bevacizumab and non-platinum-based chemotherapy in recurrent ovarian cancer: Final overall and progression-free survival results from the AGO-OVAR 2.29/ENGOT-ov34 study. Journal of Clinical Oncology. 2024;42(17_suppl):LBA5501. doi:10.1200/jco.2024.42.17_suppl.lba5501