Development of Scoring model of Conversion Surgery for technically unresectable liver metastases. Young Award

Background: We assessed the long-term outcome of technically unresectable (T-UR) colorectal liver metastases (CLM) to develop a predictive model for conversion surgery (CS).
Method: This single-center retrospective study analyzed consecutive patients who were discussed to determine the treatment strategy in the multidisciplinary conference for CLMs between 2014-2021.
Results: During the study period, 892 CLMs were shared in the conference and 122 patients had T-UR CLMs. 61 patients underwent CS (CS-group) after chemotherapy and the other 61 patients did not (Non-CS-group). The median overall survival (OS) of the CS-group (5.6-year) was significantly better than that of the Non-CS-group (1.8-year) (p<0.001).
The scoring model to predict CS was constructed using four predictive factors by uni- and multivariate analyses (RAS status, number of tumors, contact with hepatic veins, and presence of preservable section), and its area under the curve for CS was 0.870.
According to the grading (A, B, C) using this scoring model, the conversion rates were 91.5% of Grade A, 32.6% of Grade B, and 10.3% of Grade C, respectively (p<0.001). And the median OS was significantly different among the three grades (5.8 vs. 2.2 vs. 1.6 years) (p<0.001).
Conclusion: Conversion surgery for T-UR CLMs had a better prognosis, and our new Scoring model was useful to predict conversion surgery and prognosis.