A deep learning model of tumor cell architecture elucidates response and resistance to CDK4/6 inhibitors

Sungjoon Park; Erica Silva; Akshat Singhal; Marcus R Kelly; Kate Licon; Isabella Panagiotou; Catalina Fogg; Samson Fong; John J Y Lee; Xiaoyu Zhao; Robin Bachelder; Barbara A Parker; Kay T Yeung; Trey Ideker

doi:10.1038/s43018-024-00740-1

A deep learning model of tumor cell architecture elucidates response and resistance to CDK4/6 inhibitors

Nat Cancer. 2024 Mar 5. doi: 10.1038/s43018-024-00740-1. Online ahead of print.

Authors

Sungjoon Park^#¹, Erica Silva^#², Akshat Singhal^#³, Marcus R Kelly^{1

4}, Kate Licon¹, Isabella Panagiotou¹, Catalina Fogg¹, Samson Fong⁵, John J Y Lee¹, Xiaoyu Zhao¹, Robin Bachelder¹, Barbara A Parker^{1

4}, Kay T Yeung^{1

4}, Trey Ideker^{6

7

8

9}

Affiliations

¹ Department of Medicine, University of California, San Diego, La Jolla, CA, USA.
² Program in Biomedical Sciences, University of California, San Diego, La Jolla, CA, USA.
³ Department of Computer Science and Engineering, University of California, San Diego, La Jolla, CA, USA.
⁴ Moores Cancer Center, University of California, San Diego, San Diego, CA, USA.
⁵ Department of Bioengineering, University of California, San Diego, La Jolla, CA, USA.
⁶ Department of Medicine, University of California, San Diego, La Jolla, CA, USA. tideker@health.ucsd.edu.
⁷ Department of Computer Science and Engineering, University of California, San Diego, La Jolla, CA, USA. tideker@health.ucsd.edu.
⁸ Moores Cancer Center, University of California, San Diego, San Diego, CA, USA. tideker@health.ucsd.edu.
⁹ Department of Bioengineering, University of California, San Diego, La Jolla, CA, USA. tideker@health.ucsd.edu.

^# Contributed equally.

PMID: 38443662
DOI: 10.1038/s43018-024-00740-1

Abstract

Cyclin-dependent kinase 4 and 6 inhibitors (CDK4/6is) have revolutionized breast cancer therapy. However, <50% of patients have an objective response, and nearly all patients develop resistance during therapy. To elucidate the underlying mechanisms, we constructed an interpretable deep learning model of the response to palbociclib, a CDK4/6i, based on a reference map of multiprotein assemblies in cancer. The model identifies eight core assemblies that integrate rare and common alterations across 90 genes to stratify palbociclib-sensitive versus palbociclib-resistant cell lines. Predictions translate to patients and patient-derived xenografts, whereas single-gene biomarkers do not. Most predictive assemblies can be shown by CRISPR-Cas9 genetic disruption to regulate the CDK4/6i response. Validated assemblies relate to cell-cycle control, growth factor signaling and a histone regulatory complex that we show promotes S-phase entry through the activation of the histone modifiers KAT6A and TBL1XR1 and the transcription factor RUNX1. This study enables an integrated assessment of how a tumor's genetic profile modulates CDK4/6i resistance.