CLEAR-HPV: Interpretable concept discovery for human-papillomavirus-associated morphology in whole-slide histology.

Abstract

Human papillomavirus (HPV) status is a critical determinant of prognosis and treatment response in head and neck and cervical cancers. Although attention-based multiple instance learning (MIL) achieves strong slide-level prediction for HPV-related whole-slide histopathology, it provides limited morphologic interpretability. To address this limitation, we introduce concept-level explainable attention-guided representation for HPV (CLEAR-HPV), a framework that restructures the MIL latent space to enable concept discovery without requiring concept labels during training. Within an attention-weighted latent space, CLEAR-HPV automatically discovers keratinizing, basaloid, and stromal morphologic concepts; generates spatial concept maps; and represents each slide with a compact concept-fraction vector. Its concept-fraction vectors preserve the predictive information of the original MIL embeddings while reducing the high-dimensional feature space (e.g., 1,536 dimensions) to only 10 interpretable concepts. CLEAR-HPV demonstrates consistent concept structure across The Cancer Genome Atlas (TCGA)-HNSCC, TCGA-CESC, and Clinical Proteomic Tumor Analysis Consortium (CPTAC) -HNSCC, providing compact, concept-level interpretability through a general, backbone-agnostic framework for attention-based MIL models of whole-slide histopathology.