OBJECTIVETo identify the effect of postoperative adjuvant transarterial chemoembolization (TACE) on late recurrence of hepatocellular carcinoma (HCC) patients after radical resection.

METHODSFrom year 2001 to 2007, 2436 HCC patients underwent radical resection were retrospectively selected. Patients underwent resection only were classified into control group, while those received adjuvant TACE within 2 months after operation were classified into intervention group. Patients were further stratified into those with tumor5 cm and presenting low or high risk factors for recurrence. Patients with single tumor and without microscopic tumor thrombus were defined as low risk for recurrence; otherwise they were defined as high risk. The effect of adjuvant TACE on late (>2 years) recurrence was evaluated.

RESULTSRecurrence rates of tumor5 cm and presenting low, high risk factors for recurrence at 2-year after resection were 20.38%, 33.06%, 30.54% and 50.82%, respectively in the control group, compared with 25.41%, 39.61%, 40.55% and 51.57%, respectively in the intervention group; there were no significant differences between intervention group and control group in each stratum. For patients recurred or died within the first 2 years after resection, the median survival of tumor>5 cm and presenting high risk factors for recurrence was 24 months in the intervention group and 12 months in the Control group. Only in this subgroup, the survival curve of the intervention group was significantly higher than that compared to the control group. For patients who remained recurrence free and survived within the first 2 years after resection, there were no significant differences in the recurrence curves between the intervention group and control group in each stratum; while cumulative survival rates in the subgroup of patients with tumor size is less than or equal to 5 cm and presenting low risk factors for recurrence were 93.95%, 91.50% and 88.42% respectively in the control group, compared with 91.70%, 81.32% and 78.19% respectively in the intervention group at 3-, 4- and 5-year after resection (P=0.0062); for other subgroups, there were no significant differences in the survival curves between the intervention group and the control group in each stratum. Cox regression model suggested adjuvant TACE was not an independent risk factor for late recurrence; however, it might have negative effect on survival [hazard ratio (HR)=1.50, P=0.062] for those patients (especially patients with tumor is less than or equal to 5 cm and presenting low risk factors for recurrence).

CONCLUSIONSThe value of adjuvant TACE was mainly due to its therapeutic actions on residual tumor or early recurrence. It had no effect on postponing or eliminating late recurrence; moreover, it could be a risk rather than a benefit in patients at low risk for recurrence (especially those with tumor is less than or equal to 5 cm and presenting low risk factors for recurrence).