Machine learning of genomic features in organotropic metastases stratifies progression risk of primary tumors

Abstract

Metastatic cancer is associated with poor patient prognosis but its spatiotemporal behavior remains unpredictable at early stage. Here we develop MetaNet, a computational framework that integrates clinical and sequencing data from 32,176 primary and metastatic cancer cases, to assess metastatic risks of primary tumors. MetaNet achieves high accuracy in distinguishing the metastasis from the primary in breast and prostate cancers. From the prediction, we identify Metastasis-Featuring Primary (MFP) tumors, a subset of primary tumors with genomic features enriched in metastasis and demonstrate their higher metastatic risk and shorter disease-free survival. In addition, we identify genomic alterations associated with organ-specific metastases and employ them to stratify patients into various risk groups with propensities toward different metastatic organs. This organotropic stratification method achieves better prognostic value than the standard histological grading system in prostate cancer, especially in the identification of Bone-MFP and Liver-MFP subtypes, with potential in informing organ-specific examinations in follow-ups.

The location and timing of metastasis are still fundamentally unpredictable. Here the authors present the Metastatic Network model, a machine learning framework that integrates clinical data and DNA alterations to predict the risk of metastasis to specific organs as well as clinical outcomes