TIRA: Tumor Immune Representation Adaptation for Zero-Shot Cross-Cancer MSI and TMB Prediction

Key points

• Microsatellite instability-high (MSI-H) and high tumor mutational burden (TMB-H) are clinically relevant biomarkers, yet their histopathological prediction remains challenging when models are transferred across morphologically distinct cancer types.1
• Immune-associated spatial patterns can persist across cancers despite these morphological differences, but foundation-model-based predictors trained on a single cancer do not explicitly use this information, limiting cross-cancer generalization.1
• We train TIRA on TCGA-COAD+READ and evaluate it zero-shot on CPTAC-COAD, TCGA-STAD, TCGA-UCEC, and CPTAC-UCEC, covering cross-site, cross-cancer, and combined cross-cancer-site distribution shifts under UNI2, CONCH, and Virchow2.1
• Source-derived spatial immune topology improved the cross-cancer robustness of frozen pathology foundation-model representations.1

Sources (1)

• [1]TIRA: Tumor Immune Representation Adaptation for Zero-Shot Cross-Cancer MSI and TMB Prediction
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 04:50 AM

  To address this limitation, we propose TIRA (Tumor Immune Representation Adaptation), a target-free framework that refines frozen foundation-model representations using spatial immune topology, without requiring target-domain data during model development or test-time adaptation.

  Microsatellite instability-high (MSI-H) and high tumor mutational burden (TMB-H) are clinically relevant biomarkers, yet their histopathological prediction remains challenging when models are transferred across morphologically distinct cancer types.

Extractive summary: sentences quoted from the sources.