SURF: trial | Bending the Curve: Trial by Trial

SURF

Needs review

Surgery Versus Ablation for Hepatocellular Carcinoma: A Randomized Controlled Trial (SURF-RCT Trial) and a Nonrandomized Prospective Observational Trial (SURF-Cohort Trial)

N = 302

Population

HCC with largest diameter of 3 cm or less and up to 3 nodules (about 90% solitary; about 65% with a diameter of 2.0 cm or less)

Design

Randomised controlled trial at 49 institutions in Japan comparing surgery with radiofrequency ablation (RFA); eligible patients had a largest HCC diameter of 3 cm or less and no more than 3 nodules (Child-Pugh score of 7 or less in the 2022 report); registration 2009-2015; co-primary end points were recurrence-free survival and overall survival. Patients who declined randomisation were followed in a separate nonrandomised cohort.

Primary endpoint

RFS, OS

Arms

Surgery (experimental, n = 150): Surgery (hepatic resection)
Radiofrequency ablation (control, n = 152): Radiofrequency ablation (RFA)

People

Takayama T

RFSHR 0.92 (95% CI 0.67–1.25)Δ median RFS+0.5 yearsOSHR 0.96Δ 60-month OS+4.2 pointsRFSHR 0.9Δ 60-month RFS+0.2 points

Show the numbers

Results reported for SURF
Endpoint	Analysis	Surgery	Radiofrequency ablation	Effect
RFSSURF-RCT patients after exclusion of ineligible patients (the abstract does not use the term ITT)Recurrence-free survival, median in years (final result of RFS, OS follow-up ongoing at the time)	finalTakayama 2022	median 3.5 years (2.6–5.1)	median 3 years (2.4–5.6)	HR 0.92 (95% CI 0.67–1.25)P=0.58Hazard ratio is for surgery versus RFA. This report counts 150 patients in the surgery group and 151 in the RFA group; the 2025 report counts 152 in the RFA group.
OSSURF-RCT patients after exclusion of ineligible patients (302; the abstract does not use the term ITT)Overall survival at 5 years (final analysis of OS)	finalKawaguchi 2025	74.6% at 5 y	70.4% at 5 y	HR 0.96P=0.84The P value is the adjusted P value given in the abstract; no confidence interval for the hazard ratio is given in the abstract. Hazard ratio is for surgery versus RFA.
RFSSURF-RCT patients after exclusion of ineligible patients (302; the abstract does not use the term ITT)Recurrence-free survival at 5 years (updated RFS)	long term-updateKawaguchi 2025	42.9% at 5 y	42.7% at 5 y	HR 0.9P=0.84The P value is the adjusted P value given in the abstract; no confidence interval for the hazard ratio is given in the abstract. Hazard ratio is for surgery versus RFA.

Results reported for SURF
Endpoint	Analysis	Surgery	Radiofrequency ablation	Effect
RFSSURF-RCT patients after exclusion of ineligible patients (the abstract does not use the term ITT)Recurrence-free survival, median in years (final result of RFS, OS follow-up ongoing at the time)	finalTakayama 2022	median 3.5 years (2.6–5.1)	median 3 years (2.4–5.6)	HR 0.92 (95% CI 0.67–1.25)P=0.58Hazard ratio is for surgery versus RFA. This report counts 150 patients in the surgery group and 151 in the RFA group; the 2025 report counts 152 in the RFA group.
OSSURF-RCT patients after exclusion of ineligible patients (302; the abstract does not use the term ITT)Overall survival at 5 years (final analysis of OS)	finalKawaguchi 2025	74.6% at 5 y	70.4% at 5 y	HR 0.96P=0.84The P value is the adjusted P value given in the abstract; no confidence interval for the hazard ratio is given in the abstract. Hazard ratio is for surgery versus RFA.
RFSSURF-RCT patients after exclusion of ineligible patients (302; the abstract does not use the term ITT)Recurrence-free survival at 5 years (updated RFS)	long term-updateKawaguchi 2025	42.9% at 5 y	42.7% at 5 y	HR 0.9P=0.84The P value is the adjusted P value given in the abstract; no confidence interval for the hazard ratio is given in the abstract. Hazard ratio is for surgery versus RFA.

Hazard ratio (95% CI) by endpoint. Values below 1 favour the experimental arm.

Data table

Endpoint	HR	95% CI
RFS	0.92	0.67–1.25

Illustrative — modelled from reported values, not trial data. RFS, final.

Show curves

Surgery (experimental)
Radiofrequency ablation (control)

Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.

Data table (modelled values)

Months	Surgery	Radiofrequency ablation
0	100.0%	100.0%
26	65.1%	60.6%
53	41.7%	36.0%
79	27.2%	21.8%
105	17.7%	13.2%

Appears in

2006 Resection versus ablation for small HCC: conflicting trials

Sources

Takayama T, Hasegawa K, Izumi N, Kudo M, Shimada M, Yamanaka N, et al. Surgery versus Radiofrequency Ablation for Small Hepatocellular Carcinoma: A Randomized Controlled Trial (SURF Trial). Liver Cancer. 2022;11(3):209-218.
PubMed DOI
Kawaguchi Y, Hasegawa K, Kashiwabara K, Okamura Y, Kurosaki M, Kudo M, et al. Surgery Versus Ablation for Hepatocellular Carcinoma: A Randomized Controlled Trial (SURF-RCT Trial) and a Nonrandomized Prospective Observational Trial (SURF-Cohort Trial). J Clin Oncol. 2025;43(23):2628-2638.
PubMed DOI