Cohort study demonstrates multi-omics signature's ability to predict benefits of anti-EGFR therapy in colorectal cancer liver metastases, indicating potential for treatment personalization.
Key Points
This study aims to develop a multi-omics framework to predict the response of RAS wild-type colorectal cancer liver metastases to anti-EGFR therapy.
Patients receiving cetuximab were used for training and testing of the predictive model.
The study utilized a deep learning framework to integrate genetic and radiomic data.
Validation was performed on an independent cohort of patients enrolled between January and December 2018.
The combined model showed an AUC of 0.86 for predicting sensitivities to cetuximab.
The fusion signature had a hazard ratio of 17.9 for identifying treatment-sensitive cases (P = 0.003).
Patients with the fusion signature had a median progression-free survival of 9.0 months compared to 5.0 months in others (HR, 0.44; P = 0.047).