Comparative efficacy of liver resection vs. radiofrequency ablation for small hepatocellular carcinoma stratified by alpha-fetoprotein status: a retrospective cohort study using restricted mean survival time and number needed to treat analysis.

BACKGROUND

The choice between liver resection (LR) and radiofrequency ablation (RFA) for small hepatocellular carcinoma (HCC) ≤3 cm remains controversial. This study evaluated whether alpha-fetoprotein (AFP) status modulates the comparative efficacy of these modalities using absolute benefit metrics.

METHODS

Patients with T1N0M0 HCC (≤3 cm) treated with LR or RFA were identified from the Surveillance, Epidemiology, and End Results (SEER) program database. After 1:1 propensity score matching (PSM), cancer-specific survival (CSS) and overall survival (OS) were analyzed. Restricted mean survival time (RMST) and number needed to treat (NNT) were utilized to quantify the absolute clinical gain of LR over RFA.

RESULTS

Among 1,119 identified patients, 235 matched pairs were analyzed. LR demonstrated superior survival in the overall cohort (P<0.001). A significant interaction was observed between treatment modality and AFP status (Pinteraction<0.01). In AFP-negative patients, LR provided a profound survival advantage [CSS: hazard ratio (HR) =0.362; 95% confidence interval (CI): 0.220-0.594; P<0.001], translating into a substantial RMST extension of 8.1 months and a low NNT of 4. Conversely, in the AFP-positive subgroup, this surgical advantage was notably attenuated (CSS: HR =0.800; 95% CI: 0.555-1.152; P=0.23), with the RMST extension dropping to 3.5 months and the NNT increasing to 12.

CONCLUSIONS

The superiority of LR over RFA in small HCC is significantly modulated by AFP-defined tumor biology. AFP-negative patients derive profound absolute survival gains from surgery, whereas benefits in AFP-positive cases are marginal, supporting RFA as a viable alternative for biologically aggressive tumors.