Particle.news
Download on the App Store
3 ARTICLES
·
last mo.

MD Anderson Maps Triple-Negative Breast Cancer at Single-Cell Resolution and Unveils 13-Gene Model to Predict Chemo Response

The Nature study points to a way to match patients to chemotherapy before treatment using a concise gene signal from single-cell with spatial profiling.

Content cover image
Image
Image
Image

Overview

  • The study maps triple-negative breast cancer at single-cell resolution to flag features that predict who benefits from pre-surgery chemotherapy.
  • Researchers analyzed 427,857 cells from 101 patients and performed spatial profiling on tumors from 44 patients.
  • They defined four tumor archetypes, 13 intratumor programs, and eight cellular ecotypes that linked the cancer and its microenvironment to treatment outcome.
  • Macrophage subtypes, not T or NK cell states, tracked with response, with Mac‑IFN and Mac‑lip‑C1Q tied to complete response and Mac‑angio and Mac‑ECM tied to residual disease.
  • A 13‑gene signature and machine‑learning model correlated with response and survival in public cohorts, with authors urging prospective testing as single‑cell assays remain costly and complex.