Background: Adjuvant therapy after liver transplantation (LT) has been reported to improve survival outcomes. However, selecting appropriate candidates who may benefit from postoperative adjuvant therapy remains challenging. This study aimed to stratify recipients who underwent adjuvant FOLFOX plus lenvatinib (F-L) therapy after LT for hepatocellular carcinoma and provide clinical guidance on adjuvant therapy selection.
Methods: This study included 106 patients in the F-L group and 229 patients in the non-adjuvant (N-A) group at Shulan (Hangzhou) Hospital from January 2017 to December 2021 for analysis. A 1:1 propensity score matching (PSM) analysis was conducted to minimize group imbalances. Recurrence-free survival (RFS) and overall survival (OS) were compared among recipients. Radiomic features with predictive value for tumor recurrence were analyzed independently and in combination with prognostic clinical features to develop predictive models. The combined model was used to classify recipients in the F-L group into low and high recurrence subgroups for further analysis.
Results: The 1- and 3-year survival outcomes for LT recipients in the F-L group (RFS: 74.75 % and 60.95 %; OS: 92.93 % and 74.58 %) were significantly better than those of recipients in the N-A group after PSM (RFS: P = 0.035; OS: P = 0.002). Six predictive radiomic features were identified to establish a radiomic model in the F-L group, and with application of the prognostic clinical feature (microvascular invasion), a combined model was developed (area under the curve: 0.81). Recipients in the low recurrence and fulfilling Hangzhou criteria subgroup had significantly better RFS to those in the high recurrence and fulfilling Hangzhou criteria subgroup (P = 0.043). Meanwhile, F-L therapy was not suitable for recipients in the high recurrence and exceeding Hangzhou criteria subgroup.
Conclusions: LT recipients benefit from F-L adjuvant therapy. Integrating the combined model with existing LT criteria may aid in identifying suitable candidates for F-L therapy, thereby optimizing postoperative adjuvant therapy selection.

