Treatment and dose modifications in patients according to their DCB status.

Additional significant differences in circulating chemokines.

Cytotoxic response in tumor biopsies of patients with No DCB upon combination treatment.

Further descriptions of tumor immune infiltrates.

Adverse events emergent during treatment and related to experimental drugs by the investigators according to CTCAE Version 4.03.

Gating strategy for flow cytometry analyses of T cells infiltration on tumor biopsies.

Immuno-Histo-Chemistry scoring of baseline tumor biopsies according to their DCB status.

Differential Gene Expression (DGE) of baseline tumor biopsies between patients with DCB and No DCB.

Baseline characteristics of the patients treated.

Abstract Background: Poly (ADP-ribose) polymerase inhibitors (PARPi) have immunomodulatory properties that may potentiate anti-PD-L1 therapy. To explore this, we developed the academic ARIANES phase 2 basket trial, which evaluated the PARPi rucaparib (R) and the anti-PD-L1 atezolizumab (A) in selected patients (pts). We performed single-cell RNA + T cell receptor sequencing (scRNA+TCR-seq) on 3 sequential tumor biopsies in a pts subset. Methods: ARIANES comprised 4 molecularly-selected DNA damage response deficient (DDRd), 3 platinum-sensitive, and 2 unselected cohorts of pts with non-small cell lung (NSCLC), prostate (CRPC), and other non-breast non-ovarian cancers. R was administered for 3 weeks, followed by R+A. Primary endpoint was overall response rate (ORR) at 12 weeks per cohort. scRNA+TCR-seq was performed on fresh biopsies collected at baseline, on R, and on R + A in the same lesion, with concomitant plasma collection. We tracked the evolution over time of tumor and microenvironment cells, differentially expressed genes and signatures within each cell population, and TCR clonal dynamics. Results: ARIANES enrolled 130 pts, including 27 DDRd CRPC and 3 DDRd NSCLC pts, on which we focus here. Best ORR by RECIST v1.1 was 22% (6/27) in CRPC pts, and mPFS: 7.9 months (95% CI 4 - 11.1) at data cut-off. ORR was 0% in NSCLC pts. We successfully profiled 3 sequential fresh biopsies in 2 BRCA2-mutant CRPC, 1 ATM-mutant, and 1 RAD54L-mutant NSCLC pts. Tumor, T and B lymphoid, myeloid, plasma, dendritic, NK, endothelial, and stromal cells were identified in all samples. Differential expression testing unveiled PARP1 upregulation in tumor cells on R. One CRPC pt with prolonged partial response displayed activation of innate immune sensing pathways (including cGAS-STING and IFN signaling) in tumor and immune cells, as well as JAK/STAT and CCL5, all increasing gradually on R and R + A; scTCR-seq revealed the expansion of specific T cell clonotypes on R + A. One ATM-mutant NSCLC pt with 6-month stable disease displayed increased cGAS/STING, JAK/STAT and CCL5 activation on R in NK cells only, and no TCR expansion. None of the above was observed in the 2 pts who progressed. Complete clinical and translational study results will be presented at the congress. Conclusion: Sequential scRNA+TCR-seq enables the dynamic assessment of tumor and immune cells on therapy. Our data confirms the potential of R to activate innate immune pathways in pt tumors, and R + A to induce T cell activation and clonal expansion. To our knowledge, this is the first study providing a longitudinal single-cell resolution analysis of the clonal evolution of tumor and immune cells in pts on PARPi + anti-PD-L1. We identify potential biological correlates of therapy response, including the activation of innate immune pathways and TCR expansion. Citation Format: Julien Vibert, Clémence Hénon, Lambros Tselikas, Baptiste Bonnet, Arnaud Pagès, Alice Bernard-Tessier, Léo Colmet-Daage, Roman Chabanon, Nicolas Dorvault, Clémence Astier, Anas Gazzah, Aurore Jeanson, Aurélien Parpaleix, Ronan Flippot, Natacha Naoun, Anna Patrikidou, Pernelle Lavaud, Giulia Baciarello, Emeline Colomba, Alina Fuerea, Laurence Albiges, Christophe Massard, Santiago Ponce, Yohann Loriot, Stéphane Champiat, Capucine Baldini, Fanny Bouquet, Nathalie Droin, Karim Fizazi, Damien Vasseur, Patricia Martin-Romano, Etienne Rouleau, Kaïssa Ouali, Sophie Postel-Vinay. Clinical and translational results of the academic ARIANES Phase 2 basket study: Longitudinal single-cell analysis of patient tumors identifies biological correlates of response to PARP inhibitors and anti-PD-L1 therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6406.

AbstractCancer immunotherapy combinations have recently been shown to improve the overall survival of advanced mesotheliomas, especially for patients responding to those treatments. We aimed to characterize the biological correlates of malignant pleural mesotheliomas’ primary resistance to immunotherapy and antiangiogenics by testing the combination of pembrolizumab, an anti–PD-1 antibody, and nintedanib, a pan-antiangiogenic tyrosine kinase inhibitor, in the multicenter PEMBIB trial (NCT02856425). Thirty patients with advanced malignant pleural mesothelioma were treated and explored. Unexpectedly, we found that refractory patients were actively recruiting CD3+CD8+ cytotoxic T cells in their tumors through CXCL9 tumor release upon treatment. However, these patients displayed high levels of somatic copy-number alterations in their tumors that correlated with high blood and tumor levels of IL6 and CXCL8. Those proinflammatory cytokines resulted in higher tumor secretion of VEGF and tumor enrichment in regulatory T cells. Advanced mesothelioma should further benefit from stratified combination therapies adapted to their tumor biology.Significance:Sequential explorations of fresh tumor biopsies demonstrated that mesothelioma resistance to anti–PD-1 + antiangiogenics is not due to a lack of tumor T-cell infiltration but rather due to adaptive immunosuppressive pathways by tumors, involving molecules (e.g., IL6, CXCL8, VEGF, and CTLA4) that are amenable to targeted therapies.This article is highlighted in the In This Issue feature, p. 799

Background We aimed to determine the safety and efficacy of nintedanib, an oral anti-angiogenic tyrosine kinase inhibitor, in combination with pembrolizumab, an anti-PD1 immunotherapy, in patients with advanced solid tumors (PEMBIB trial; NCT02856425). Methods In this monocentric phase Ib dose escalation cohort, we evaluated escalating doses of nintedanib (Dose level 1 (DL1) = 150 mg bid [ bis in die , as twice a day]; DL2 = 200 mg bid, oral delivery) in combination with pembrolizumab (200 mg Q3W, IV). Patients received a 1-week lead-in dose of nintedanib monotherapy prior starting pembrolizumab. The primary objective was to establish the maximum tolerated dose (MTD) of the combination based on dose limiting toxicity (DLT) occurrence during the first 4 weeks. Secondary objectives were to assess the anti-tumor efficacy and to identify the associated immune and angiogenic parameters in order to establish the recommended nintedanib dose for expansion cohorts. Flow cytometry (FC), Immuno-Histo-Chemistry (IHC) and electrochemiluminescence multi-arrays were prospectively performed on baseline & on-treatment tumor and blood samples to identify immune correlates of efficacy. Results A total of 12/13 patients enrolled were evaluable for DLT (1 patient withdrew consent prior receiving pembrolizumab). Three patients at 200 mg bid experienced a DLT (grade 3 liver enzymes increase). Four patients developed grade 1–2 immune related adverse events (irAE). Eight patients died because of cancer progression. Median follow-up was 23.7 months (95%CI: 5.55–40.5). Three patients developed a partial response (PR) (ORR = 25%) and five patients (42%) had durable clinical benefit (DCB), defined as PR or stable disease (SD) ≥ 6 months. At baseline, patients with DCB had higher plasma levels of Tie2, CXCL10, CCL22 and circulating CD4 + PD1 + OX40 + T cells than patients without DCB. Patients with DCB presented also with more DC-LAMP + dendritic cells, CD3 + T cells and FOXP3 + Tregs in baseline tumor biopsies. For DCB patients, the nintedanib lead-in monotherapy resulted in higher blood CCL3, Tregs and CCR4 + CXCR3 + CXCR5 − memory CD4 T cells. After the first pembrolizumab infusion, patients with DCB showed lower IL-6, IL-8, IL-27 plasma levels. Conclusion Nintedanib 150 mg bid is the recommended dose for combination with pembrolizumab and is currently investigated in multiple expansion cohorts. Early tumoral and circulating immune factors were associated with cancer outcome under nintedanib & pembrolizumab therapy. Trial registration ClinicalTrials.gov, NCT02856425 . Registered August 4, 2016 — Prospectively registered.

BackgroundWe report the results from the advanced malignant mesothelioma (aMM) expansion cohort of the PEMBIB Phase Ib trial (NCT02856425) evaluating the safety, efficacy & biomarkers of an antiangiogenic tyrosine kinase inhibitor (nintedanib) with an anti-PD1 immunotherapy (pembrolizumab).MethodsPatients with aMM relapsing after at least one line of platinum doublet chemotherapy and not previously pre-exposed to IO were treated with a combination of oral nintedanib (150mg BID) & IV pembrolizumab (200mg Q3W) with a 7 days nintedanib lead-in preceding pembrolizumab initiation. Baseline and on-treatment (cycle D2, day 1 [C2D1]) fresh tumor & blood samples were prospectively phenotyped by flow cytometry (FC). RNAseq was run on tumor samples. Immune factors were titrated on tumor secretome and plasma.Results30 aMM patients were treated and 29 evaluable for response. Median age was 68 years old (38–85) and 86% of aMM were epithelioid. The most frequent adverse events (AE) (grades 1–3) related to the combination were liver enzymes increase, fatigue, nausea, and diarrhea. 4 (13.3%) patients developed grade 3–5 immune- related AE. Patients died of cancer progression (n=14, 46.7%), myocarditis with thrombo-embolic event (n=1, 3.3%) and COVID-19 (n=1, 3.3%). Median follow-up was 14.8 months (95%CI [9.70–18.2]). Best Overall Response Rates (BORR) per RECISTv1.1 were Partial Response (PR, n=7/29; 24.1%), Stable Disease (SD, n=17/29; 58.6%) and Progressive Disease (n=5/29; 17.2%). Disease Control Rate (DCR) (defined as PR + SD) was 46.6% at 6 months. Patients with DCR at 6 months had significantly higher percentage of PDL1 expression on tumor cells (by Immuno-Histo-Chemistry, antibody clone SP263) and higher CD8+ T cells infiltrate in tumor biopsies (by FC) at screening. Upon treatment, soluble plasma rate of CXCL9 and CXCL13 increased in all patients, as well as tumor immune infiltrates estimated by deconvolution of tumor biopsies RNA-seq. But deconvoluted estimates of NK cells, T cells and myeloid dendritic cells infiltrates on baseline tumors and C2D1 biopsies were higher in patients with DCR at 6 months. Pre & on-treatment IL6 and IL8 rates in tumor secretome & plasma were higher in patients without DCR. Gene Set Enrichment Analyses on RNA-seq from screening biopsies highlighted an enrichment in E2F, MYC and KRAS gene pathways and lower expression of type 1 interferon signature in patients without DCR than those with DCR at 6 months.ConclusionsWith a BORR of 24% and a DCR of 47% at 6 months, pembrolizumab and nintedanib combination provided valuable therapeutic benefits for patients with aMM.Trial RegistrationClinicalTrialsgov, NCT02856425. Registered August 4, 2016 — Prospectively registered,https://clinicaltrials.gov/ct2/show/NCT02856425?term=PEMBIB&draw=2&rank=1.Ethics ApprovalThe protocol was first approved by the Agence Nationale de Sécurité du Médicament (ANSM) on June 24th 2016 (Ref #160371A-12). The protocol was also approved by the Ethical Committee (Comité de Protection des Personnes Ile de France 1) on Jul 12th 2016 (Ref #2016-mai-14236ND).