Paper Proffered Program Therapy Physics

Habitat Imaging–Guided Drug Reversal Discovery In Head and Neck Cancer

Abstract
Purpose

Conventional radiomics models often treat tumors as spatially homogeneous entities, limiting their ability to capture intratumoral heterogeneity and its impact on prognosis and treatment resistance. Habitat imaging addresses this limitation by explicitly modeling spatially distinct intratumoral subregions. This study proposes HiG-DR (Habitat Imaging–Guided Drug Reversal Discovery), a physics-guided framework that extends habitat imaging beyond prognostic modeling toward biologically interpretable drug reversal discovery in head and neck cancer.

Methods

Radiomic features were extracted from both the entire gross tumor volume and multiple intratumoral habitats generated through spatially resolved clustering. A habitat–temporal radiomics (HTR) model incorporating time-dependent Cox regression was developed for prognostic stratification. Model development and internal validation were performed using the RADCURE dataset (n = 2,681), with external validation conducted on the Head-Neck-Radiomics-HN1/CPTAC cohort (n = 137). Genes associated with high-risk imaging habitats were analyzed using gene set enrichment analysis (GSEA), followed by gene set reversal analysis (GSRA) to identify candidate compounds predicted to reverse high-risk biological states.

Results

The HTR model significantly outperformed conventional whole-tumor radiomics, achieving higher concordance indices across validation (0.86 vs. 0.58) and external validation (0.81 vs. 0.59; all log-rank p < 0.001). High-risk imaging habitats were characterized by pronounced textural heterogeneity and phase-specific temporal patterns. GSEA revealed enrichment of stress-adaptive survival programs, including DNA damage repair, selective autophagy, nuclear transport, and cilium-associated organization. GSRA identified candidate compounds predicted to reverse these high-risk biological programs. GSRA highlighted candidate compounds targeting stress-adaptive survival and cellular transport pathways, including agents associated with signaling modulation and metabolic regulation.

Conclusion

HiG-DR demonstrates that habitat imaging can be extended beyond prognostic prediction to enable imaging-guided drug reversal discovery. By leveraging spatial and temporal intratumoral heterogeneity in a large, multi-cohort analysis, this framework provides a robust pathway for translational imaging-driven therapeutic hypothesis generation in head and neck cancer.

People

Related

Similar sessions

Poster Poster Program
Jul 19 · 07:00
Python-Based Automation Framework for Annual Machine QA Data Archiving In Qatrack+

Annual water-tank measurements help ensure beam characteristics remain consistent with commissioning baselines. However, the lack of a standardized processing workflow and decentralized data storage makes it difficult to analyze...

Syed Bilal Ahmad, PhD
Therapy Physics 0 people interested
Poster Poster Program
Jul 19 · 07:00
User Expectations and Current Availability of HDR Brachytherapy Audits In Europe

The aim of this work was to evaluate the need to implement more dosimetric audits in high‐dose‐rate brachytherapy (HDR-BT) in Europe and to identify which characteristics such audits should meet according to users.

Javier Vijande, PhD Laura Oliver Cañamás
Therapy Physics 0 people interested