Embryonal and alveolar rhabdomyosarcoma immunotherapy treatment benefit prediction by CURE AI

Clinical trials are not able to study individual treatment effects as methods to do so have not yet matured into practical use. For decades, we have studied groups of patients enrolled on a trial, comparing groups of patients as large cohorts and losing signal for important, complex features that define individuals. Foundation models take a different approach. Using the CURE AI foundation model generated from clinical and multi-omics data from hundreds of thousands of patients, complex and non-linear biological patterns can be found in new datasets that can provide insights into disease biology. We previously identified a complex signature predictive of immunotherapy treatment benefit compared to chemotherapy benefit by analysis of a set of non-small cell lung cancer clinical trials (Weiss et al., AI in Precision Oncology, 2025). In the current study, we theorized that we could apply predictors of immunotherapy response/nonresponse from adult clinical trials to pediatric cancer patients. If feasible, this approach could dramatically accelerate progress in pediatric cancer treatment development by leading to the development of better treatment strategies that could immediately be assessed on pediatric clinical trials. We analyzed RNA sequencing from over 50 types of pediatric cancers to predict treatment benefit of immunotherapy relative to chemotherapy. Alveolar and embryonal rhabdomyosarcomas (RMS) stood out as notable case studies as the majority of patients in both rhabdomyosarcoma groups were, based on treatment response prediction from adult immunotherapy clinical trials, predicted to respond favorably to immunotherapy. However, about one-third of both RMS subtypes were predicted to be resistant to immunotherapy. Alveolar RMS had significantly more immunotherapy resistance-related genes upregulated in patients with predicted immunotherapy resistance compared to embryonal RMS, with largely different compositions of immune-related genes between the RMS subtypes. Interestingly, CURE AI identified 36 genes with > 4-fold upregulation that were in common between embryonal and alveolar RMS patients with predicted immunotherapy resistance including many genes not previously associated with RMS or immunotherapy resistance including TLL2 and F2RL2, suggesting that these pathways may be targeted to overcome immunotherapy resistance in RMS. Furthermore, alveolar RMS-specific and embryonal RMS-specific cytokine signatures were identified that could be implemented to guide patient selection for immunotherapy treatment. These results demonstrate that biological foundation models can be used to analyze adult clinical trials to gain novel insights into different cancer types to generate patient-level pediatric data that could justify accelerated clinical trial concept development in pediatric cancers.