Competing risk analysis of surgical resection vs. radiofrequency ablation in early-stage small hepatocellular carcinoma: a SEER-based study

Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality globally. While surgical resection (SR) is the gold standard for early-stage HCC, radiofrequency ablation (RFA) serves as an alternative for patients unfit for surgery. The relative efficacy of SR versus RFA for solitary tumors ≤5 cm remains a topic of debate. Traditional survival analysis methods, often used in comparative studies, fail to account for competing risk events, potentially biasing survival outcome estimates. This study utilizes data from the Surveillance, Epidemiology, and End Results (SEER) database, integrated with competing risk analysis, with the aim of more accurately assessing the efficacy differences between SR and RFA in patients with early-stage HCC, thereby offering clinicians more scientific and reliable evidence.

Patients with early-stage HCC [American Joint Committee on Cancer (AJCC) stage I/II] and solitary tumors ≤5 cm who underwent SR or RFA were identified from the SEER database [2004-2021]. Kaplan-Meier survival curves and Cox proportional hazards models were used to evaluate overall survival (OS) and cancer-specific survival (CSS). Competing risk models were applied to assess liver cancer-specific death (LCSD) and other cause-specific death (OCSD). Propensity score matching (PSM) was performed to minimize baseline differences between SR and RFA groups.

A total of 4,691 patients were included (1,628 in the SR group, 3,063 in the RFA group). After PSM, 1,200 patients (600 per group) were analyzed. Kaplan-Meier analysis demonstrated superior OS and CSS in the SR group before and after PSM. Competing risk analysis showed a lower cumulative incidence of LCSD in the SR group compared to the RFA group, consistent across tumor size subgroups (<3 and 3-5 cm). Multivariate analysis revealed that RFA was associated with a higher risk of LCSD [hazard ratio (HR): 1.955 in Cox model; HR: 1.791 in competing risk model].

SR is associated with better OS, CSS, and lower LCSD compared to RFA in early-stage HCC patients with solitary tumors ≤5 cm. Competing risk analysis provides more accurate and clinically relevant insights into treatment efficacy. These findings support SR as the preferred treatment for eligible patients and highlight the importance of considering competing risks in survival studies.