INTRODUCTION:In high-grade serous ovarian carcinoma (HGSOC), time to first recurrence is a major prognostic factor guiding therapeutic management. However, this parameter remains poorly studied in low-grade serous ovarian carcinoma (LGSOC), a rare subtype characterized by indolent progression and marked chemoresistance. Moreover, AGO criteria for selecting candidates for secondary cytoreductive surgery appear poorly applicable to LGSOC, highlighting the need for specific prognostic markers. METHODS:We conducted a multicenter retrospective study using the national LGSC database from the French Rare Malignant Gynaecologic Tumor Network. Patients with recurrent LGSOC were included. Associations between time to first recurrence, post-relapse survival (PRS), and post-relapse progression-free survival (PR-PFS) were assessed using Cox proportional hazards models adjusted for established prognostic factors. The association between time to recurrence and completeness of secondary cytoreduction was evaluated using multivariable logistic regression. RESULTS:Among 211 patients, the median time to first recurrence was 32 months (IQR 19-63). Longer recurrence intervals were associated with improved outcomes. In multivariable analysis, patients relapsing at 24-48 months (PRS: HR 0.50; 95% CI 0.25-0.97) and ≥48 months (PRS: HR 0.25; 95% CI 0.11-0.58) had significantly better survival compared with the 12-24 month group. Similar results were observed for PR-PFS, with a significant benefit for ≥48 months (HR 0.36; 95% CI 0.19-0.70). Incorporating time to recurrence improved model discrimination. Time to recurrence was not associated with completeness of secondary cytoreduction. CONCLUSION:Time to first recurrence is an independent prognostic factor in LGSOC. Its integration into prognostic models may improve patient stratification and refine selection for secondary cytoreductive surgery.

Patients treated for rare ovarian germ cell tumors (GCT) and sex cord stromal tumors (SCST) often experience long-term survival. Treatments include surgery with or without chemotherapy (CT), which can cause late side effects that negatively impact quality of life (QoL). This study assessed long-term QoL among GCT and SCST survivors compared to age-matched healthy women (HW), focusing on patients treated with CT. Cancer-free GCT and SCST survivors (≥ 2 years post-treatment) were recruited from the INCa French Network TMRG. HW were recruited via the “Seintinelles” platform. Participants completed validated questionnaires assessing QoL (FACT-G/FACT-O), fatigue (MFI), anxiety/depression (HADS), insomnia (ISI), neurotoxicity (FACT/GOG-NTX), cognition (FACT-COG), and sexuality (FACT-O OCS). Clinically relevant differences were defined as ≥ 5

Gynecologic carcinosarcomas (CS) are highly aggressive/rare tumors. Most patients relapse after first-line therapies and responses to systemic therapy remain low. ROCSAN (NCT03651206) is a multicentric randomized phase II evaluating dostarlimab (anti-PD1) in combination with niraparib (PARPi) compared to standard of care in the treatment of metastatic/recurrent (R/M) CS. Objective: Mandatory tumor tissue archival from naïve initial disease (ND) & baseline biopsies at relapse (R/M) were used to decipher the biological changes of CS in R/M disease vs. ND. bulk transcriptomic profiling was performed on paired samples from 60/63 included patients. Differential expression, gene ontology (GO), and immune cell deconvolution were performed. Our previously reported HOT signature was computed (GSVA) (PMID36038492). Additionally, a 7-color multiplex immunofluorescence (mIF) analysis of 52 paired samples (CD20, Ki67, CD3, CD8, FoxP3, panCK, DAPI) was done to evaluate total T cells, CD8 and CD4 Teff, Treg, and B cells cell densities and tertiary lymphoid structures (TLS). Differential expression analysis adjusted for FIGO stage, primary site and number of previous systemic therapies identified 1, 502 and 584 genes up and down regulated respectively in R/M disease vs. ND. Among 13 immune checkpoints (ICP) and 22 of their ligands (ICPL), 8 ICP (TIGIT, HAVCR2, CTLA4, VSIR, CD40LG, TNFRSF9, CD27, ICOS) and 4 ICPL (PD-L1, PDCD1LG2, CD80, CD86), were upregulated in R/M disease vs. ND. Top GO enriched pathways in R/M vs. ND included immune activating pathways, regulation of lymphocyte activation and positive regulation of cytokine production. Among 50 biological hallmarks (GSEA), R/M showed a significant enrichment in inflammatory response, TNFA signaling via NFKB, and IFNG response vs. ND. Deconvolution analysis found an increase in NK cells (P=0.024), Macrophage/Monocyte (P=0.001) and Neutrophils (P=0.023) in R/M disease vs. ND. The HOT signature score was found to be increased in R/M disease vs. ND (P=0.01). The proportion of HOT CS (score>0) was 33/60 (55%) and 23/60 (38%) in R/M and ND respectively. Among the 37/60 (62%) COLD CS (score<0) in ND, 18/37 (49%) became HOT in R/M. Among the 23/60 (38%) HOT CS at ND, 8/23 (35%) became COLD at R/M. mIF analysis showed that total T cells (P=0.037), CD8 Teff (P=0.039), CD4 Teff (P=0.034), and CD4 Treg (P=0.003) increased in R/M disease vs. ND. Intriguingly, TLS density decreased in R/M disease vs. ND (P=0.035). When compared with ND, R/M CS host a greatest immunologically active microenvironment, as suggested by the changes observed in the HOT score and HOT/COLD phenotype, implying potentially more activity for immune therapy in relapse than for localized initial therapy. Whether this could be used to select patients for immunotherapy should be investigated. Sonia Canjura-Rodriguez, Cassandra Assaf, Victor Heurtier, Annalisa Pesavento, Justine Berthet, Audrey Bellesoeur, Coriolan Lebreton, Sheik Emambux, Dominique Berton, Magali Provansal, Guillaume Bataillon, Lucas Michon, Jessie Auclair, Alexandra Lainé, Valéry Attignon, Christophe Caux, Anthony Ferrari, Ivan Bièche, Bertrand Dubois, Isabelle Ray-Coquard, Pierre Saintigny. Changes in the tumor immune microenvironment during disease progression in gynecological carcinosarcoma: Exploratory analysis from the randomized ROCSAN GINECO trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5273.

Primary antibodies used in the multiplex panels for the detection of the different immune populations

Change in GrzB expression min paired samples post-NACT

Immune cells distribution in post-placebo and post-BIBF tumors

Patients and diseases baseline characteristics in the CHIVA neoadjuvant trial

Feature plots of relevant immune populations in pre- and post-NACT clusters

Introduction/Background Low grade serous ovarian cancer (LGSOC) is a rare entity which risk factors, treatment response and risk of relapse are partially understood. Currently available data suffer from small sample size and heterogeneous management, leading to limited knowledge. Mutlivariation Information-based Inductive Causation (MIIC) is a network learning method, able to analyze and exploit simultaneously and exhaustively a large number of patients data, to identify non visible correlation, without an a priori classification on the type of reconstructed network (causal or non-causal). The aim of this study was to identify new and unknown association on patients with a LGSOC using the MIIC algorithm, in order to develop new hypothesis. Methodology We conducted a multicenter retrospective study in 31 French healthcare centers between 2010 and 2017. We included all patients with a LGSOC, and collected clinical, histological, molecular, surgical and treatment data. The MIIC algorithm was applied to this database. Results 317 patients were included. Variables were selected and pre-processed as follow: Center, age, menopausal status, tabacco consumption, abdominal surgery history, BMI, previous lesion (borderline or de novo), ascitis, initialCA125 level, initial Peritoneal Carcinosis Index, sus mesocolic involvment, digestive resection, wide peritonectomy, number of nodes removed, positive nodes, Complete resection,FIGO stage, Estrogen Receptor, Progesteron Receptor, somatic BRCA mutation, BRAF mutation, KRAS mutation, MicroSatellite Instability, chemotherapy, Hyperthermic IntraPeritoneal Chemotherapy, Bevacizumab, endocrine therapy, recurrence and death status. Known or obvious associations such as age and menopausal status, ascitis, CA125, allowed us to perform quality control of the algorithm. In addition, new associations that were previously little known or unknown, and still unexplored, have been highlighted, such as relationship between menopausal status and wide peritonectomy or nodal involvement. Conclusion The use of MIIC algorithm on a large LGSOC database has enabled the identification of interesting hypothesis, and future research topics. Disclosures The authors have no conflict of interest to declare.

Purpose: This study investigates changes in CD8+ cells, CD8+/Foxp3 ratio, HLA I expression, and immune coregulator density at diagnosis and upon neoadjuvant chemotherapy (NACT), correlating changes with clinical outcomes.Experimental Design: Multiplexed immune profiling and cell clustering analysis were performed on paired matched ovarian cancer samples to characterize the immune tumor microenvironment (iTME) at diagnosis and under NACT in patients enrolled in the CHIVA trial (NCT01583322).Results: Several immune cell (IC) subsets and immune coregulators were quantified pre/post-NACT. At diagnosis, patients with higher CD8+ T cells and HLA I+-enriched tumors were associated with a better outcome. The CD8+/Foxp3+ ratio increased significantly post-NACT in favor of increased immune surveillance, and the influx of CD8+ T cells predicted better outcomes. Clustering analysis stratified pre-NACT tumors into four subsets: high Binf, enriched in B clusters; high Tinf and low Tinf, according to their CD8+ density; and desert clusters. At baseline, these clusters were not correlated with patient outcomes. Under NACT, tumors were segregated into three clusters: high BinfTinf, low Tinf, and desert. The high BinfTinf, more diverse in IC composition encompassing T, B, and NK cells, correlated with improved survival. PDL1 was rarely expressed, whereas TIM3, LAG3, and IDO1 were more prevalent.Conclusions: Several iTMEs exist during tumor evolution, and the NACT impact on iTME is heterogeneous. Clustering analysis of patients unravels several IC subsets within ovarian cancer and can guide future personalized approaches. Targeting different checkpoints such as TIM3, LAG3, and IDO1, more prevalent than PDL1, could more effectively harness antitumor immunity in this anti-PDL1-resistant malignancy.

Change in ieCD8+ and sCD8+ populations in paired samples with NACT

Positivity of immune coregulators in pre- and post-NACT clusters

BACKGROUND:Ovarian cancer (OC) remains one of the most challenging and deadly malignancies facing women today. While PARP inhibitors (PARPis) have transformed the treatment landscape for women with advanced OC, many patients will relapse and the PARPi-resistant setting is an area of unmet medical need. Traditional immunotherapies targeting PD-1/PD-L1 have failed to show any benefit in OC. The CD47/TSP-1 axis may be relevant in OC. We aimed to describe changes in CD47 expression with platinum therapy and their relationship with immune features and prognosis. METHODS:Tumor and blood samples collected from OC patients in the CHIVA trial were assessed for CD47 and TSP-1 before and after neoadjuvant chemotherapy (NACT) and multiplex analysis was used to investigate immune markers. Considering the therapeutic relevance of targeting the CD47/TSP-1 axis, we used the CD47-derived TAX2 peptide to selectively antagonize it in a preclinical model of aggressive ovarian carcinoma. RESULTS:Significant reductions in CD47 expression were observed post NACT. Tumor patients having the highest CD47 expression profile at baseline showed the greatest CD4+ and CD8+ T-cell influx post NACT and displayed a better prognosis. In addition, TSP-1 plasma levels decreased significantly under NACT, and high TSP-1 was associated with a worse prognosis. We demonstrated that TAX2 exhibited a selective and favorable biodistribution profile in mice, localizing at the tumor sites. Using a relevant peritoneal carcinomatosis model displaying PARPi resistance, we demonstrated that post-olaparib (post-PARPi) administration of TAX2 significantly reduced tumor burden and prolonged survival. Remarkably, TAX2 used sequentially was also able to increase animal survival even under treatment conditions allowing olaparib efficacy. CONCLUSIONS:Our study thus (1) proposes a CD47-based stratification of patients who may be most likely to benefit from postoperative immunotherapy, and (2) suggests that TAX2 is a potential alternative therapy for patients relapsing on PARP inhibitors.

Primary antibodies used for IHC monomarker detection