Background. Following the result of Checkmate 743, dual checkpoint inhibition is a standard of care for unresectable malignant pleural mesothelioma (MPM). There is an ongoing need for potential predictive biomarkers for both efficacy and toxicity. To determine how the tumour microenvironment impacts ICI treatment outcomes and irAEs, we have designed a pilot study to determine the role of tumour-infiltrating immune cells on Ipilimumab (IPI) + Nivolumab (NIVO) treatment.
Methods. This pilot trial was conducted as part of a prospective multicentre cohort study (NCT04631731). We performed a VISIUM spatial transcriptomic profiling (10X Genomics) of FFPE tissues. Loupe browser v.6.0.0 and GraphPad v.9 were used for data analysis with ConsensusPath-DB for functional annotation.
Results. We recruited 4 MPM male patients treated with second line IPI (1mg/kg Q6W) + NIVO (3mg/kg Q2W) and median age of 74.5 years. Patients were encoded as A1, B1, C1 and D1. Recent CT scans showed complete response and stable disease in patients A1 and C1 respectively with disease progression in B1 and D1 (both deceased by the time of this report). While being on treatment A1 and D1 developed G3 pneumonitis, C1 was admitted to hospital for G4 hepatitis with B1 had no registered toxicity. Our results established that tumours of non-responders expressed higher expression of inhibitory checkpoint TIM-3 on effector immune cells (NK and T). Moreover, the functional annotation of differentially expressed genes established that complement cascade was upregulated in tumour of responders as compared to non-responders. Regarding signatures of irAEs we determined that B1 had higher number of M2 macrophages and higher expression of HLA genes which could impact its resistance to immune-related toxicity.
Conclusions. Herein we present pilot results of NCT04631731 using cutting-edge spatial analysis in a real-world cohort of patients with cancer. Further validation of established genetic signatures of ICI is currently ongoing.