BioCOMPASS: Integrating Biomarkers into Transformer-Based Immunotherapy Response Prediction
This work addresses the challenge of generalizing prediction models for immunotherapy response across varied cancer types and treatments, representing an incremental improvement over existing transformer-based methods.
The authors tackled the problem of poor generalization in immunotherapy response prediction models across diverse patient cohorts by extending a transformer-based model to integrate biomarker and treatment information through novel loss components. They achieved improved generalizability in leave-one-out evaluations, though specific numerical gains were not provided.
Datasets used in immunotherapy response prediction are typically small in size, as well as diverse in cancer type, drug administered, and sequencer used. Models often drop in performance when tested on patient cohorts that are not included in the training process. Recent work has shown that transformer-based models along with self-supervised learning show better generalisation performance than threshold-based biomarkers, but is still suboptimal. We present BioCOMPASS, an extension of a transformer-based model called COMPASS, that integrates biomarkers and treatment information to further improve its generalisability. Instead of feeding biomarker data as input, we built loss components to align them with the model's intermediate representations. We found that components such as treatment gating and pathway consistency loss improved generalisability when evaluated with Leave-one-cohort-out, Leave-one-cancer-type-out and Leave-one-treatment-out strategies. Results show that building components that exploit biomarker and treatment information can help in generalisability of immunotherapy response prediction. Careful curation of additional components that leverage complementary clinical information and domain knowledge represents a promising direction for future research.