Multimodal Model for Efficacy Prediction Cetuximab in Colorectal Cancer Liver Metastasis Patient

Multimodal Deep Learning Radiomic Nomogram for Evaluation of Response to Cetuximab in Patient With Colorectal Cancer Liver Metastasis

Establishment and validation of the deep learning model of Cetuximab efficacy in simultaneous RAS wild unresectable CRLM patients

Ras wild unresectable CRLM patients with primary tumor resection followed by Cetuximab in combination with chemotherapy were included in this study. The tumor response was assessed by local MDT group. Based on tumor response, almost 100 CRLM patients were classified into two groups (Clinician drived regimen vs Multi-omics model drived regimen). They will be the prospective cohort to validate our deep learning model for predicting Cetuximab efficacy.

Inclusion Criteria:

1. Age 18-75 years;
2. Histologically proven colorectal adenocarcinoma;
3. Simultaneous liver-limited metastases;
4. Initially unresectable liver metastases determined by a local MDT;
5. Life expectancy of > 6 months;
6. RAS and BRAF V600E wild-type;
7. ECOG 0-1;
8. Available CT imaging before treatment.

Exclusion Criteria:

1. Previous systemic treatment for metastatic disease;
2. Previous surgery for metastatic disease;
3. Extrahepatic metastases;
4. Unresectable primary tumor;
5. Major cardiovascular events (myocardial infarction, severe/unstable angina, congestive heart failure, CVA) within 12 months before randomisation;
6. Acute or subacute intestinal obstruction;
7. Second primary malignancy within the past 5 years;
8. Drug or alcohol abuse;
9. No legal capacity or limited legal capacity;
10. Uncontrolled hypertension, or unsatisfactory blood pressure control with ≥3 antihypertensive drugs;