Radiomics-based prediction of treatment response to TRuC-T cell therapy in patients with mesothelioma: a pilot study

Abstract

Abstract T cell receptor fusion constructs (TRuCs), a next generation engineered T cell therapy, hold great promise. To accelerate the clinical development of these therapies, improving patient selection is a crucial pathway forward. We retrospectively analyzed 23 mesothelioma patients (85 target tumors) treated in a phase 1/2 single arm clinical trial (NCT03907852). Five imaging sites were involved, settings of evaluations were Blinded Independent Central Review (BICR) with double reads. Reproducibility of 3416 radiomics and delta-radiomics (Δradiomics) was assessed. Univariate analysis evaluated correlation at target tumor level with 1) tumor diameter response; 2) tumor volume response, according to the Quantitative Imaging Biomarker Alliance and 3) the mean standard uptake value (SUV) response, as defined by positron emission tomography response criteria in solid tumors (PERCIST). A random forest model predicted the response of target pleural tumors. Tumor anatomical distribution was 55.3%, 17.6%, 14.1% and 10.6% in the pleura, lymph nodes, peritoneum and soft tissues, respectively. Radiomics/Dradiomics reproducibility differed across tumors localization. Radiomics were more reproducible than Dradiomics. In the univariate analysis, none of the radiomics/Dradiomics correlated with any response criteria. With an accuracy ranging 0.75–0.9, 3 radiomics/Dradiomics were able to predict response of target pleural tumors. Pivotal studies will require a sample size of 250 to 400 tumors. The prediction of responding target pleural tumors can be achieved using machine learning-based radiomics/Dradiomics analysis. Tumor-specific reproducibility and average values indicated that bridging tumor model to effective patient model would require combining several target tumors models.