Frederik Marmé

Atezolizumab With Bevacizumab and Nonplatinum Chemotherapy for Recurrent Ovarian Cancer: Final Results From the Placebo-Controlled AGO-OVAR 2.29/ENGOT-ov34 Phase III Trial

PURPOSE To evaluate atezolizumab combined with bevacizumab and non–platinum-based chemotherapy for recurrent ovarian cancer. METHODS The double-blind randomized phase III AGO-OVAR 2.29/ENGOT-ov34 trial (ClinicalTrials.gov identifier: NCT03353831 ) enrolled patients with first or second relapse of ovarian cancer ≤6 months after completing platinum-based chemotherapy (or third relapse regardless of treatment-free interval). PD-L1 status was tested centrally (VENTANA SP142 assay) in recent (<3 months) biopsies before random assignment. All patients received bevacizumab and investigator-selected chemotherapy (once weekly paclitaxel or pegylated liposomal doxorubicin) until disease progression or toxicity, plus either atezolizumab 840 mg or placebo once every 2 weeks until progression (maximum 2 years), randomly assigned 1:1, and stratified by number of previous lines, planned chemotherapy, previous bevacizumab, and PD-L1 status. Primary end points were overall survival (OS) and progression-free survival (PFS) in the intention-to-treat population. RESULTS Among 574 patients randomly assigned between September 2018 and July 2022, 72% were bevacizumab-pretreated, 36% had received three previous treatment lines, 26% had PD-L1–positive tumors, and 54% received paclitaxel with study therapy. After 418 patients had died, the hazard ratio for OS was 0.83 (95% CI, 0.68 to 1.01; P = .06; median 14.2 months with atezolizumab and 13.0 months with placebo) and the hazard ratio for PFS was 0.87 (95% CI, 0.73 to 1.04; P = .12; median 6.4 v 6.7 months, respectively). OS hazard ratios were similar regardless of PD-L1 status. Grade ≥3 adverse events occurred in 72% of atezolizumab-treated and 69% of placebo patients. CONCLUSION Combining atezolizumab with bevacizumab and chemotherapy did not significantly improve OS or PFS in patients with recurrent ovarian cancer ineligible for platinum. The safety profile was as expected from previous experience with these drugs.

Long non-coding RNAs as a biomarker for homologous recombination deficiency and parp inhibitor sensitivity in high-grade serous ovarian cancers

Abstract Homologous recombination deficiency (HRD) and sensitivity to PARP inhibitors are key determinants of therapeutic response in high-grade serous ovarian cancer (HGSC), yet predictive biomarkers beyond BRCA1/2 mutations or genomic HRD scores remain inadequate. Here, we investigate the potential of long non-coding RNAs (lncRNAs) as predictive markers of HRD and PARP inhibitor response. We identify a panel of lncRNAs that stratifies HGSC tumors by HRD status and drug sensitivity. Among these, ENSG00000272172.1 is significantly upregulated in HRD-positive tumors and is detectable in both formalin-fixed tissue and plasma, supporting its use as a minimally invasive biomarker. Functional analyses reveal that this lncRNA contributes to genome stability by modulating replication dynamics. These findings highlight a previously unrecognized role for lncRNAs in the HRD phenotype and suggest translational potential for ENSG00000272172.1 in guiding clinical decision-making.

Surrogate End Points for Overall Survival in Neoadjuvant Randomized Clinical Trials for Early Breast Cancer

PURPOSE To assess trial-level surrogacy value for overall survival (OS) of the pathologic complete response (pCR) and invasive disease-free survival (iDFS) in randomized clinical trials (RCTs) for early breast cancer (BC). METHODS Individual patient data of neoadjuvant RCTs with available data on pCR, iDFS, and OS were included in the analysis. We used the coefficient of determination R 2 from weighted linear regression models to quantify the association between treatment effects on OS and on the surrogate end points. RESULTS Eleven RCTs, for a total of 15 treatment comparisons and 12,247 patients, were included in the analysis. There was a weak association between hazard ratios (HRs) for OS and odds ratio of pCR overall ( R 2 , 0.07; 95% CI, 0.00 to 0.48), as well as in all the subgroups explored. Overall, the R 2 for the association between HR OS and HR iDFS was 0.46 (95% CI, 0.08 to 0.71), which is just below the cutoff of 0.5 for moderate surrogacy. In the majority of subgroups explored, the R 2 ranged from 0.5 to <0.7, while in hormone receptor–/human epidermal growth factor receptor 2– subtype, histologic grade 1-2 tumors, and lobular tumors, surrogacy was strong (ie, R 2 ≥0.7). The surrogacy value of iDFS for OS was affected by follow-up (FUP) length: R 2 substantially increased up to 36 months of FUP, with little further improvement after 48 months of FUP. CONCLUSION iDFS with sufficient FUP is an acceptable surrogate end point to confidently anticipate final OS results of neoadjuvant RCTs for early BC. This recommendation holds true across many subgroups, with the notable exception of HR+ disease. There is definite need to reassess whether OS is the optimal end point for treatment efficacy measurement in HR+ early BC.

A Biomarker-enriched, Randomized Phase II Trial of Adavosertib (AZD1775) Plus Paclitaxel and Carboplatin for Women with Platinum-sensitive TP53 -mutant Ovarian Cancer

Abstract Purpose: Preclinical studies show that adavosertib, a WEE1 kinase inhibitor, sensitizes TP53-mutant cells to chemotherapy. We hypothesized that adavosertib, plus chemotherapy, would enhance efficacy versus placebo in TP53-mutated ovarian cancer. Patients and Methods: Following safety run-in, this double-blind phase II trial (NCT01357161) randomized women with TP53-mutated, platinum-sensitive ovarian cancer to oral adavosertib (225 mg twice daily for 2.5 days/21-day cycle) or placebo, plus carboplatin (AUC5) and paclitaxel (175 mg/m2), until disease progression or for six cycles. The primary endpoints were progression-free survival (PFS) by enhanced RECIST v1.1 [ePFS (volumetric)] and safety. Secondary/exploratory objectives included PFS by RECIST v1.1 (single dimension), objective response rate, overall survival, and analysis of tumor gene profile versus sensitivity to adavosertib. Results: A total of 121 patients were randomized to adavosertib (A+C; n = 59) and placebo (P+C; n = 62) plus chemotherapy. Adding adavosertib to chemotherapy improved ePFS [median, 7.9 (95% confidence interval (CI), 6.9–9.9) vs. 7.3 months (5.6–8.2); HR 0.63 (95% CI, 0.38–1.06); two-sided P = 0.080], meeting the predefined significance threshold (P < 0.2). Clinical benefit was observed following A+C for patients with different TP53 mutation subtypes, identifying possible response biomarkers. An increase in adverse events was seen with A+C versus P+C: greatest for diarrhea (adavosertib 75%; placebo 37%), vomiting (63%; 27%), anemia (53%; 32%), and all grade ≥3 adverse events (78%; 65%). Conclusions: Establishing an optimal strategy for managing tolerability and identifying specific patient populations most likely to benefit from treatment may increase clinical benefit. Future studies should consider additional adavosertib doses within the chemotherapy treatment cycle and the potential for maintenance therapy.

Ovarian Cancer–Specific BRCA -like Copy-Number Aberration Classifiers Detect Mutations Associated with Homologous Recombination Deficiency in the AGO-TR1 Trial

Abstract Purpose: Previously, we developed breast cancer BRCA1-like and BRCA2-like copy-number profile shrunken centroid classifiers predictive for mutation status and response to therapy, targeting homologous recombination deficiency (HRD). Therefore, we investigated BRCA1- and BRCA2-like classification in ovarian cancer, aiming to acquire classifiers with similar properties as those in breast cancer. Experimental Design: We analyzed DNA copy-number profiles of germline BRCA1- and BRCA2-mutant ovarian cancers and control tumors and observed that existing breast cancer classifiers did not sufficiently predict mutation status. Hence, we trained new shrunken centroid classifiers on this set and validated them in the independent The Cancer Genome Atlas dataset. Subsequently, we assessed BRCA1/2-like classification and obtained germline and tumor mutation and methylation status of cancer predisposition genes, among them several involved in HR repair, of 300 ovarian cancer samples derived from the consecutive cohort trial AGO-TR1 (NCT02222883). Results: The detection rate of the BRCA1-like classifier for BRCA1 mutations and promoter hypermethylation was 95.6%. The BRCA2-like classifier performed less accurately, likely due to a smaller training set. Furthermore, three quarters of the BRCA1/2-like tumors could be explained by (epi)genetic alterations in BRCA1/2, germline RAD51C mutations and alterations in other genes involved in HR. Around half of the non–BRCA-mutated ovarian cancer cases displayed a BRCA-like phenotype. Conclusions: The newly trained classifiers detected most BRCA-mutated and methylated cancers and all tumors harboring a RAD51C germline mutations. Beyond that, we found an additional substantial proportion of ovarian cancers to be BRCA-like.

Primary Analysis of EPIK-O/ENGOT-ov61: Alpelisib Plus Olaparib Versus Chemotherapy in Platinum-Resistant or Platinum-Refractory High-Grade Serous Ovarian Cancer Without BRCA Mutation

PURPOSE Patients with platinum-resistant/platinum-refractory high-grade serous ovarian cancer (HGSOC) without a BRCA mutation have poor prognosis and limited treatment options. We report efficacy and biomarker data from EPIK-O, which investigated alpelisib + olaparib versus single-agent chemotherapy in these patients. PATIENTS AND METHODS EPIK-O was an open-label, phase III trial that randomly assigned patients with platinum-resistant/platinum-refractory HGSOC with no germline or known somatic BRCA mutation 1:1 to alpelisib 200 mg once daily + olaparib 200 mg twice daily or treatment of physician's choice (TPC; paclitaxel 80 mg/m 2 once weekly or pegylated liposomal doxorubicin 40-50 mg/m 2 once every 28 days). Patients had 1-3 previous systemic therapies. Previous bevacizumab was required (unless contraindicated); previous poly(adenosine diphosphate-ribose) polymerase inhibitors were allowed. Primary end point was progression-free survival (PFS) per RECIST 1.1 (blinded independent review committee [BIRC]). Secondary efficacy end points included overall response rate (ORR; per BIRC), duration of response (per BIRC), and overall survival (OS; key secondary end point). RESULTS A total of 358 patients (alpelisib + olaparib [n = 180], TPC [n = 178]) were included. The median follow-up time was 9.3 months. At data cutoff (April 21, 2023), 33 (18.3%) and 30 (16.9%) patients remained on treatment with alpelisib + olaparib and TPC, respectively. The median PFS (BIRC) was 3.6 versus 3.9 months (hazard ratio [HR], 1.14 [95% CI, 0.88 to 1.48]; one-sided P = .84) for alpelisib + olaparib versus TPC. The ORR was 15.6% (95% CI, 10.6% to 21.7%) versus 13.5% (95% CI, 8.8% to 19.4%). The median OS was 10.0 versus 10.6 months (HR, 1.22; 95% CI, 0.87 to 1.71). The safety profile of alpelisib + olaparib was consistent with that observed for the individual agents. CONCLUSION The primary objective, PFS improvement, was not met in EPIK-O. No new or unexpected adverse events were observed. Biomarker analyses provided new insights for responders to alpelisib + olaparib.

Predicting benefit from PARP inhibitors using deep learning on H&E-stained ovarian cancer slides

Ovarian cancer patients with a Homologous Recombination Deficiency (HRD) often benefit from polyadenosine diphosphate-ribose polymerase (PARP) inhibitor maintenance therapy after response to platinum-based chemotherapy. HR status is currently analyzed via complex molecular tests. Predicting benefit from PARP inhibitors directly on histological whole slide images (WSIs) could be a fast and cheap alternative. We trained a Deep Learning (DL) model on H&E stained WSIs with "shrunken centroid" (SC) based HRD ground truth using the AGO-TR1 cohort (n = 208: 108 training, 100 test) and tested its ability to predict HRD as evaluated by the Myriad classifier and the benefit from olaparib in the PAOLA-1 cohort (n = 447) in a blinded manner. In contrast to the HRD prediction AUROC of 72 % on hold-out, our model only yielded an AUROC of 57 % external. Kaplan-Meier analysis showed that progression free survival (PFS) in the PARP inhibitor treated PAOLA-1 patients was significantly improved in the HRD positive group as defined by our model, but not in the HRD negative group. PFS improvement in PARP inhibitor-treated patients was substantially longer in our HRD positive group, hinting at a biologically meaningful prediction of benefit from PARP inhibitors. Together, our results indicate that it might be possible to generate a predictor of benefit from PARP inhibitors based on the DL-mediated analysis of WSIs. However, further studies with larger cohorts and further methodological improvements will be necessary to generate a predictor with clinically useful accuracy across independent patient cohorts.

Harnessing lipid-driven immunometabolic pathways in omental metastases to enhance immunotherapy in patients with ovarian cancer

Abstract Immunotherapy with immune checkpoint blockade (ICB) in epithelial ovarian carcinoma (EOC) shows limited clinical benefit only for a small subset of patients. Overall response rates are low, so that overcoming immunotherapy resistance and improved stratification are key. In this study, we investigated the immunometabolic landscape of EOC with a focus on omental metastases, identifying lipid-laden macrophages as central elements for actionable therapeutic vulnerabilities and giving rise to biomarkers for improved patient stratification. Using patient-derived explants, we demonstrated a functional dichotomy inside the typically lipid-rich microenvironment of omental metastases: augmented maintenance of effector T cell function, while lipid uptake and processing by tumor-associated macrophages (TAMs) induces oxidative stress–dependent signaling programs, which drive macrophage dysfunction and immune suppression. Pharmacological modulation of lipid-driven signaling pathways through CCR5 inhibition (inflammation modulation through maraviroc) or blockade of the lipid scavenger receptor CD36 reprograms TAMs, restores T cell activity, and enhances antitumor immune responses within lipid-rich tumor niches. Mechanistically, studies in humanized mouse models reveal that maraviroc-mediated CCR5 inhibition induces transcriptional programs associated with immune activation in stressed, lipid-laden human TAMs. Consistent with these mechanistic insights, we demonstrated that the specific immunometabolic niche in omental metastases is clinically associated with responsiveness to ICB. We propose a non-invasive radiomics and machine-learning–based analysis of imaging data to assess omental involvement for patient stratification.

Atezolizumab With Bevacizumab and Chemotherapy vs Bevacizumab and Chemotherapy in Early Relapse Ovarian Cancer

This is a phase III, randomized, partially blinded, multicenter trial to evaluate the efficacy and safety of atezolizumab plus bevacizumab and chemotherapy compared to placebo plus bevacizumab and chemotherapy in patients with recurrent ovarian-, fallopian tube, or primary peritoneal cancer with 1st or 2nd relapse within 6 months after platinum based chemotherapy or 3rd relapse.

Prevalence of BRCA in Patients With Ovarian Cancer

The aim of this prospective registration and translational research study is to evaluate the praevalence of BRCA regarding germline and somatic mutations.