CANDiT: A machine learning framework for differentiation therapy in colorectal cancer

Saptarshi Sinha; Joshua Alcantara; Kevin Perry; Vanessa Castillo; Annelies K Ondersma; Satarupa Banerjee; Ella McLaren; Celia R Espinoza; Sahar Taheri; Eleadah Vidales; Courtney Tindle; Adel Adel; Siamak Amirfakhri; Joseph R Sawires; Jerry Yang; Michael Bouvet; Pradipta Ghosh

doi:10.1016/j.xcrm.2025.102421

CANDiT: A machine learning framework for differentiation therapy in colorectal cancer

Cell Rep Med. 2025 Nov 18;6(11):102421. doi: 10.1016/j.xcrm.2025.102421. Epub 2025 Oct 20.

Authors

Saptarshi Sinha¹, Joshua Alcantara², Kevin Perry², Vanessa Castillo¹, Annelies K Ondersma², Satarupa Banerjee¹, Ella McLaren², Celia R Espinoza¹, Sahar Taheri³, Eleadah Vidales², Courtney Tindle², Adel Adel¹, Siamak Amirfakhri⁴, Joseph R Sawires⁵, Jerry Yang⁵, Michael Bouvet⁶, Pradipta Ghosh⁷

Affiliations

¹ Department of Cellular and Molecular Medicine, School of Medicine, University of California, San Diego, La Jolla, CA 92093, USA.
² Department of Cellular and Molecular Medicine, School of Medicine, University of California, San Diego, La Jolla, CA 92093, USA; HUMANOID™ Center of Research Excellence, University of California, San Diego, La Jolla, CA 92093, USA.
³ Department of Computer Science and Engineering, Jacob's School of Engineering, University of California, San Diego, San Diego, CA 92093, USA.
⁴ Department of Surgery, University of California, San Diego, San Diego, CA 92093, USA; Department of Surgery, VA San Diego Healthcare System, San Diego, CA, USA.
⁵ Department of Chemistry and Biochemistry, University of California, San Diego, San Diego, CA 92093, USA.
⁶ Department of Surgery, University of California, San Diego, San Diego, CA 92093, USA; Department of Surgery, VA San Diego Healthcare System, San Diego, CA, USA; Moores Cancer Center, University of California, San Diego, San Diego, CA 92093, USA.
⁷ Department of Cellular and Molecular Medicine, School of Medicine, University of California, San Diego, La Jolla, CA 92093, USA; HUMANOID™ Center of Research Excellence, University of California, San Diego, La Jolla, CA 92093, USA; Moores Cancer Center, University of California, San Diego, San Diego, CA 92093, USA; Department of Medicine, University of California, San Diego, San Diego, CA 92093, USA. Electronic address: prghosh@ucsd.edu.

Abstract

Reactivating lineage commitment to differentiate, and hence eliminate, cancer stem cells (CSCs) remains a therapeutic challenge. Here, we present CANDiT (cancer-associated nodes for differentiation targeting), a machine learning framework that identifies transcriptomic vulnerabilities for differentiation therapy in colorectal cancer (CRC). Centering on CDX2-a master intestinal lineage factor lost in high-risk, poorly differentiated CRCs-we identify PRKAB1, a stress polarity sensor, as a top therapeutic target. A clinical-grade PRKAB1 agonist reactivates lineage programs, dismantles Wnt/YAP-driven stemness, and selectively eliminates CDX2-low CSCs across CRC cell lines, xenografts, and patient-derived organoids (PDOs). Multivariate analysis reveals a strong therapeutic index tied to the CDX2-low state. A 50-gene response signature, derived from integrated modeling across all platforms, predicts ∼50% reduction in recurrence and mortality risk. Like immunotherapy, CANDiT resurrects a physiologic program-differentiation-to selectively eliminate CSCs, offering a scalable, precision framework for lineage restoration in solid tumors.

Keywords: CCDC88A; CDX2 restoration; SPS; cancer stem cell; differentiation therapy; stress-polarity pathway.

MeSH terms

Animals
CDX2 Transcription Factor / genetics
CDX2 Transcription Factor / metabolism
Cell Differentiation* / drug effects
Cell Line, Tumor
Colorectal Neoplasms* / drug therapy
Colorectal Neoplasms* / genetics
Colorectal Neoplasms* / metabolism
Colorectal Neoplasms* / pathology
Gene Expression Regulation, Neoplastic
Humans
Machine Learning*
Mice
Neoplastic Stem Cells / drug effects
Neoplastic Stem Cells / metabolism
Neoplastic Stem Cells / pathology
Organoids

Substances

CDX2 Transcription Factor
CDX2 protein, human