Work

36 Baseline systemic immune profiles predict prognosis and differential benefit from ipilimumab intensification in relapsed mesothelioma (Bio-IFCT-1501 MAPS2 trial)

Year: 2026

Type: conference-abstract

Abstract: Background Combined PD-1 and CTLA-4 blockade improves outcomes in pleural mesothelioma (PM) but is associated with substantially higher toxicity compared with anti-PD-1 monotherapy. Biomarkers to guide escalation or de-escalation of immunotherapy are therefore needed in a precision medicine approach.Methods Baseline plasma from 114 pre-treated PM patients enrolled in the randomized phase II MAPS2 trial (nivolumab n = 60 vs. nivolumab plus ipilimumab n = 54) was analyzed using the Olink immuno-oncology panel. Patients were classified as low or high for each cytokine according to the median value. Then candidate biomarkers were quantified using Ella assays, and associations with progression-free survival (PFS) and overall survival (OS) were evaluated according to the treatment arm. In a subset with available baseline biopsies (n = 72), bulk tumor RNAseq obtained at diagnosis was performed, followed by PM single-cell-based deconvolution, to explore baseline plasma immune profiles and tumor microenvironment features.Results Unsupervised clustering identified four systemic immune profiles: C1, an inflammatory-angiogenic cluster characterized by elevated CASP8, IL18, ARG1, IL8 and VEGFA; C2, a myeloid-inflamed cluster enriched for IL6, CSF1, ANGPT2, CCL23 and NOS3; C3, a cytotoxic lymphoid cluster characterized by GZMA, ICOSLG, KIR3DL1, Galectin-1 and MCP4; and C4, a cytokine-low cluster without specific cytokine enrichment. Clinically, C2 carried the poorest prognosis while C3 conferred significantly improved survival (PFS p = 0.014; OS p = 0.00047). Tumor RNAseq deconvolution revealed that clusters were independent of the histomolecular sarcomatoid S.score. The poor-prognosis C2 was enriched in CD4 + T cells and tumor-associated macrophages, particularly TREM2-high and SPP1-high populations.Treatment–biomarker interaction analyses identified CXCL9 (interaction p = 0.041 for PFS and 0.048 for OS) and TNF-α (interaction p = 0.027 for PFS and 0.12 for OS) as predictors of differential benefit from CTLA-4 blockade. These findings were confirmed using quantitative Ella assays. Patients with low baseline TNF-α (<14.1 pg/mL) derived substantial benefit from nivolumab–ipilimumab (OS HR = 0.42; PFS HR = 0.45), whereas no benefit was observed among patients with high TNF-α levels (OS HR = 1.23; PFS HR = 1.18). Similarly, low baseline CXCL9 (<1115 pg/mL) identified patients with enhanced benefit from combination therapy (OS HR = 0.53; PFS HR = 0.48), while no differential treatment effect at high CXCL9 levels (OS HR = 1.06; PFS HR = 0.94, figure 1).Conclusions Baseline circulating cytokines is associated with tumor microenvironment features, particularly macrophage composition, and are associated with clinical outcomes in PM. Notably, CXCL9 and TNF-α represent exploratory but promising biomarkers to guide treatment allocation between anti-PD-1 monotherapy and combined checkpoint blockade. Prospective validation is warranted to support a precision immunotherapy approach in this disease.Acknowledgements We thank the Intergroupe Francophone de Cancérologie Thoracique (IFCT) for sponsoring the MAPS2 trial and for enabling this translational collaboration, as well as all investigators, study coordinators, and patients who participated in the trial. M.R.D. is supported by the Fondation Nuovo-Soldati for Cancer Research, the Fondation ARC pour la Recherche sur le Cancer, an ESMO Research Fellowship, and INSERM.Trial Registration NCT02716272.Ethics Approval The research protocol of MAPS2 trial was approved by the Nord-Ouest III ethics committee (Comité de Protection des Personnes) of the University Hospital (CHU) of Caen, France, and the trial was carried out in line with the Helsinki Declaration and Good Clinical Practice guidelines. Prior to inclusion, all patients provided written informed consent to participate in the trial.Abstract 36 Figure 1

Source: Journal for ImmunoTherapy of Cancer

DOI: 10.1136/jitc-2026-sitc2026.0036