AI and Cancer Research ── Cancer biology ── 2026-11-03
Cancer molecular biology and AI
- Abstract
PURPOSE Whole-exome and whole-genome sequencing (WES/WGS) datasets are skewed toward patients from white and well-resourced backgrounds, and machine learning models for cancer have advanced faster than efforts to diversify these repositories. Meanwhile, targeted sequencing panels—such as AACR Project GENIE—are comparatively abundant and clinically routine, but capture far less biological information per sample. This study asks how to maximize the utility of existing WES/WGS data to improve and equalize predictions made from the panel sequencing data we already have in relative abundance—a question of direct relevance to low- and middle-income countries where WES/WGS access lags most severely. METHODS Somatic mutations from WES/WGS and targeted panel data were annotated using two evolutionary scoring frameworks: standard impact scores (derived from cross-species conservation) and cancer effect size, a disease-specific, empirical measure of the evolutionary benefit a mutation confers to the tumor. A transfer learning approach was used to leverage feature importance learned from WES/WGS models and apply it to panel-sequenced data. Model performance was evaluated across a range of clinical cancer features and stratified by patient demographic group to assess disparities in predictive accuracy. A modest hold-out dataset from a population not represented in the data was used to compare improvement of evolutionary embedding versus representative sequencing. RESULTS Cancer effect size matched or outperformed standard impact scores across clinical prediction tasks. The evolutionary embedding transferred to panel data improved overall model performance and disproportionately benefited underrepresented demographic groups, narrowing disparities in predictive accuracy. However, direct inclusion of samples from more diverse individuals reduced disparity more effectively than any embedding strategy alone, while matching or improving performance in all groups. CONCLUSION Evolutionary embedding of somatic mutations is a powerful, immediately deployable strategy to extract more equitable signal from the limited and skewed WES/WGS data that exists—a critical lever in settings where sequencing access remains constrained. Yet this approach has a clear ceiling: computational innovation is no substitute for representative data collection. Both strategies must be pursued in parallel, but the field must recognize that inclusion at the source outperforms optimization after the fact.
Journal IF-equivalent: 2.1 (OpenAlex 2-year mean citedness, value as of 2026-11-02, retrieved 2026-11-03; not the official Clarivate IF)Reference: Fisk N, Townsend J. You Can't Embed Your Way Out of Disparities: Somatic Evolution Scores and the Case for Representative Sequencing. JCO Glob Oncol. 2026 Nov 1;12(Supplement_1):15-6. doi:10.1200/go-26-21000.Checked: Abstract only - Abstract
Liver metastasis is a lethal manifestation of diverse malignancies, but its molecular heterogeneity is not readily captured by routine histopathology. Proteomic profiling has identified functional states of metastatic lesions, although their histomorphological and tumor microenvironmental correlates remain incompletely understood. Here, we developed AI-HistoPS, a computational pathology framework for inferring proteomics-informed states from H&E-stained whole slide images. AI-HistoPS integrates pathology foundation model representations with agent-mediated multiple instance learning to capture subtype-associated morphological patterns. Using multicentre cohorts of liver metastases from multiple primary tumor types, we evaluated associations between histomorphology and three proteomic liver metastasis subtypes (LMS1-3). These states were associated with differences in tumor architecture, stromal composition, and immune cell distribution. AI-HistoPS showed subtype discrimination in the discovery cohort and was evaluated across external cohorts, demonstrating the feasibility of histology-derived representations for proteomics-informed stratification across heterogeneous clinical settings. Attention-guided analyses identified tissue regions associated with subtype-related morphological and microenvironmental features. Multiplex immunohistochemistry supported spatial associations between histological patterns and subtype-related protein markers. These findings suggest that routine histopathology may provide a lower-cost approach for approximating proteomics-informed stratification of liver metastases. Further validation in larger, more diverse, prospectively collected cohorts is required before clinical implementation or use in prognostic or treatment-related decision-making.
Journal IF-equivalent: 8.9 (OpenAlex 2-year mean citedness, value as of 2026-10-04, retrieved 2026-10-05; not the official Clarivate IF)Reference: Ren Y, He Q, Ni H, Tu C, Wu J, Sun Y, et al. AI-HistoPS: decoding proteomic heterogeneity with subtype resolution from histopathology in liver metastases. npj Precis Onc. 2026 Oct 31 [Epub ahead of print]. doi:10.1038/s41698-026-01752-z.Checked: Abstract only