Revisiting the role of postoperative adjuvant transarterial chemoembolization in hepatocellular carcinoma: a propensity score matching analysis Authors: Jia-Jun Li, Ye Xu, Hua-Hua Liu, Mao-Pei Chen, Yi Chen, Ning-Ling Ge, Lan Zhang, Zheng-Gang Ren, Xin-Rong Yang, Rong-Xin Chen Published: 2025 ABSTRACT AbstractBackgroundThe optimal adjuvant therapy after liver resection in patients with hepatocellular carcinoma (HCC) is controversial. This study aimed to revisit the efficacy of postoperative adjuvant transarterial chemoembolization (PA-TACE) in HCC patients after curative-intent hepatectomy.MethodsA total of 387 patients were divided into PA-TACE group and no adjuvant treatment control group, and follow-up data were collected. The primary endpoints of recurrence-free survival (RFS) and overall survival (OS) were assessed before and after propensity score matching (PSM) analysis. Survival data were computed by means of the Kaplan-Meier method. Multivariable Cox proportional hazards model was used to determine the independent risk factors for patients’ outcomes.ResultsThe RFS rates were higher in patients treated with PA-TACE compared to those in the control group, with borderline statistical significance [P=0.050; hazard ratio (HR) =0.75, 95% confidence interval (CI): 0.56–1.0]. Patients in the PA-TACE group had significantly higher OS rates than those in the control group (P=0.04, HR =0.70, 95% CI: 0.50–0.99). After PSM, the impact of PA-TACE on RFS and OS remained significant (HR =0.70, P=0.04 for RFS; HR =0.65, P=0.04 for OS). On multivariate Cox regression analyses in the entire cohort, age >65 years, γ-glutamyl transferase (GGT) >40 U/L, microvascular invasion (MVI)-positive, and no PA-TACE were identified as independent predictors for HCC recurrence. In further subgroup analysis, PA-TACE significantly improved RFS and OS of HCC patients in MVI-positive group (HR =0.31, P<0.001 for RFS; HR =0.47, P=0.002 for OS). The benefit of PA-TACE on RFS and OS remained significant in MVI-positive group after PSM (HR =0.32, P<0.001 for RFS; HR =0.38, P=0.003 for OS). Furthermore, patients who received PA-TACE developed recurrent HCC with less aggressive tumor characteristics.ConclusionsPA-TACE improves RFS and OS of HCC patients after liver resection, especially in MVI-positive patients. Furthermore, PA-TACE reduces the malignancy of recurrent tumors.Keywords: Hepatocellular carcinoma (HCC), resection, adjuvant, transarterial chemoembolization (TACE) Highlight box Key findings • Postoperative adjuvant transarterial chemoembolization (PA-TACE) improves recurrence-free survival (RFS) and overall survival (OS) of hepatocellular carcinoma (HCC) patients after curative-intent liver resection, especially in microvascular invasion (MVI)-positive patients. • PA-TACE reduces the malignancy of recurrent tumors. What is known and what is new? • Transarterial chemoembolization (TACE) is an effective therapy to control the progression of unresectable HCC for a long time. In Chinese HCC treatment guideline, PA-TACE is recommended for HCC patients at high risk of recurrence following resection. • In this study, we revisited the role of PA-TACE in HCC and found that PA-TACE improves RFS and OS in HCC patients after curative liver resection, especially in MVI-positive patients. What is the implication, and what should change now? • PA-TACE improves RFS and OS in HCC patients after curative liver resection, with a more pronounced effect observed in MVI-positive HCC patients. • PA-TACE may be considered as an adjuvant treatment option following liver resection in HCC patients, especially in MVI-positive patients. Introduction Hepatocellular carcinoma (HCC) is one of the most common cancers in the world, its incidence ranks the sixth among all malignant tumors, and its mortality ranks the third among cancer-related causes of death (1). Although different curative therapies have been proposed, the prognosis of HCC is far from expected. Evidence suggests that patients who undergo liver resection or radiofrequency ablation have a 5-year recurrence rate as high as 70% (2). Effective prevention of recurrence after curative resection in HCC patients is still an unmet clinical need. During the past decades, many adjuvant therapies have been applied to HCC patients after curative resection, such as chemotherapy (3), interferon, chemoembolization (4), iodine-131 labeled lipiodol, sorafenib, immunotherapy (5), and atezolizumab plus bevacizumab (6). However, none of these adjuvant treatments has been established for patients at high risk of HCC recurrence after liver resection, due to negative or inconsistent study results, or lack of high-level evidence. Transarterial chemoembolization (TACE) is an effective therapy to control the progression of unresectable HCC for a long time. TACE as adjuvant therapy after HCC resection is often applied in clinical practice in China. In Chinese HCC treatment guideline, postoperative adjuvant transarterial chemoembolization (PA-TACE) is recommended for HCC patients at high risk of recurrence following resection (7). Randomized trials, multiple retrospective studies and a meta-analysis have reported that PA-TACE may improve time to recurrence and survival outcomes in individual patients who have an intermediate or high risk of recurrence following curative resection (8-11). However, not all HCC patients at risk of recurrence following curative resection will benefit from PA-TACE treatment. Furthermore, a study has shown that the effect of PA-TACE on HCC recurrence is controversial (12). Here, we aimed to revisit the role of PA-TACE on tumor recurrence and prognosis of HCC patients after curative-intent liver resection. We present this article in accordance with the STROBE reporting checklist (13) (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-655/rc). Methods Patients The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This retrospective study was approved by the Institutional Review Boards of Zhongshan Hospital, Fudan University, Shanghai, China (No. B2025-587). Due to its retrospective nature, informed consent for this retrospective study was waived. The inclusion criteria were as follows: pathologically confirmed HCC; curative anatomical or non-anatomical hepatectomy with negative margin (confirmed by microscopic examination); Barcelona Clinic Liver Cancer (BCLC) stage A or B; Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1; Child-Pugh class A; no macrovascular tumor thrombosis, no prior neoadjuvant therapy; no lymph node or distant metastasis; and adequate organ functions. The exclusion criteria were ≤18 years old; patients with recurrent HCC; prior any anti-HCC therapy, including TACE and systemic therapy (chemotherapy, molecular targeted therapy, and immunotherapy); BCLC C stage; lack of follow-up data. Written consents for liver surgery were obtained from each patient. PA-TACE was generally recommended in patients with tumors larger than 5 cm, microvascular invasion (MVI), or multiple (two, three or more) tumors. The decision for PA-TACE was made whenever necessary according to the surgeon’s discretion. TACE All patients received curative-intent liver resection with surgical techniques and perioperative management as previously described (14,15). Approximately 1 month after liver resection, patients underwent adjuvant TACE treatment. In brief, digital subtraction angiography of the celiac artery and superior mesenteric artery was performed after a 4F or 5F RH catheter was used to catheterize arteries. Then, a catheter or microcatheter was advanced to the proper, right, and/or left hepatic artery under guide wire guidance. At last, chemotherapeutic and embolizing agents, such as doses of 1,000 mg 5-fluorouracil (5-FU) diluted with normal saline, 100–150 mg Oxaliplatin with 5% dextrose water, and 10–30 mg Epirubicin mixed with 2–5 mL lipiodol, was slowly injected. Propensity score matching (PSM) analysis A PSM analysis (16) was conducted to minimize potential bias and confounding in the comparison of outcomes between the PA-TACE group (n=183) and the no adjuvant treatment control group (n=204). Propensity scores were estimated using a logistic regression model that incorporated the following covariates: age, gender, total bilirubin (TB), alanine aminotransferase (ALT), prothrombin time (PTT), albumin (ALB), γ-glutamyl transferase (GGT), α-fetoprotein (AFP), maximum tumor diameter, tumor number, MVI, BCLC stage, and tumor capsule. Patients in both the PA-TACE treatment and no adjuvant treatment control group were matched in a 1:1 ratio using the nearest neighbor matching method, with a caliper width set at 0.05 times the standard deviation (SD) of the logit-transformed propensity scores, ensuring the optimal comparability. The balance of matching variables between the two groups was evaluated using standardized mean differences (SMDs), where SMD values less than 0.2 indicate negligible differences in the means or prevalence of covariates between the matched groups. The propensity model matched 134 patients in each group. Endpoints The primary endpoints of the study were recurrence-free survival (RFS) and overall survival (OS). RFS was defined as the elapsed time between the date of surgical treatment and either the date of tumor recurrence or death from any cause, whichever came first. OS was defined as the interval between the date of surgical treatment and either the date of patient death from any cause or the last follow-up. Data acquisition Demographic and clinical variables at diagnosis (e.g., age, gender, HBsAg, TB, ALT, PTT, ALB, GGT, AFP), details of pathology or clinical imaging (e.g., tumor diameter, number of tumors, MVI, BCLC stage, tumor capsule, and tumor differentiation), and clinical outcome data of RFS and OS were extracted from the medical records through the picture archiving and communication system (PACS) and hospital information system (HIS). Statistical analysis Descriptive statistics were employed to summarize the baseline characteristics of the patients. Continuous variables are presented as means with SD or as medians with interquartile range (IQR) as appropriate. Categorical variables are reported as frequencies and percentages. Group comparisons for continuous variables were conducted using independent-sample Student’s t-tests. The Chi-square test or Fisher’s exact test was used for comparing categorical variables. The Kaplan-Meier survival analysis was performed to estimate RFS and OS, and the log-rank test was applied to assess survival differences between two groups. Univariate and multivariate Cox proportional hazards regression models were used to identify independent prognostic factors associated with RFS and OS. Variables with a P<0.10 in the univariate analysis were entered into the multivariate analysis. Results are expressed as hazard ratio (HR) with 95% confidence interval (CI). Subgroup analyses were carried out to evaluate the potential prognostic benefit. All statistical analyses were performed using R software version 4.0.3 (R Foundation for Statistical Computing, Vienna, Austria) and SPSS version 26.0 (IBM Corp., Armonk, NY, USA). A two-sided P<0.05 were considered statistically significant. Results Characteristics of the patients For this retrospective study, we enrolled a total of 520 HCC patients who underwent curative liver resection between February 2006 and July 2017 at the Zhongshan Hospital, Fudan University, Shanghai, China. After meeting all inclusion criteria and no exclusion criteria, 387 patients were included in the final analysis (Figure 1). Table 1 summarizes the demographic and clinicopathological characteristics of the study population. Next, we evaluated the effects of PA-TACE on the tumor recurrence and prognosis of HCC patients after liver resection. The baseline clinicopathological characteristics of 183 patients treated with PA-TACE and 204 patients with no adjuvant treatment is shown in Table 1. Patients in PA-TACE treatment group had a lower level of serum ALB, larger tumor size, and a higher proportion of liver cirrhosis and tumor stage BCLC B, compared with patients in no adjuvant treatment control group (Table 1). To reduce the differences of baseline data between two groups, a 1:1 PSM was carried out based on the logistic analysis. After PSM, 134 patients in each of these two groups were identified. The difference of baseline clinicopathological features between these two groups was not significant (Table 1). The CONSORT flow chart of this study. BCLC, Barcelona Clinic Liver Cancer; HCC, hepatocellular carcinoma; LR, liver resection; PA-TACE, postoperative adjuvant transarterial chemoembolization; PSM, propensity score matching. Table 1The baseline clinicopathological characteristics of the study populationCharacteristicsPrimary cohortPSM cohortNo PA-TACE (n=204)PA-TACE (n=183)P valueNo PA-TACE (n=134)PA-TACE (n=134)P valueAge, years54.4±11.052.8±10.90.1753.5±10.452.9±10.80.64Gender0.270.63 Male164 (80.4)155 (84.7)109 (81.3)112 (83.6) Female40 (19.6)28 (15.3)25 (18.7)22 (16.4)AFP, ng/mL32.3 (4.8–381.0)51 (5–460.0)0.6642.0 (6.0–396.8)72.0 (7.0–461)0.64HBV0.170.37 Negative24 (11.8)14 (7.7)9 (6.7)13 (9.7) Positive180 (88.2)169 (92.3)125 (93.3)121 (90.3)Cirrhosis0.030.64 Negative56 (27.5)33 (18.0)23 (17.2)26 (19.4) Positive148 (72.5)150 (82.0)111 (82.8)108 (80.6)ALT, U/L 35.0 (23.1–47.8)37.0 (28.0–56)0.05336.9 (26.0–51.8)38.5 (27.7–58.3)0.50ALB, g/dL 5.1 (4.2–42.8)4.5 (4.1–5.3)<0.0014.5 (4.1–38.9)4.6 (4.2–38.8)0.26GGT, U/L46.4 (30.0–82.7)51 (32.6–86.5)0.3346.9 (28.3–83.6)48.5 (32.0–86.8)0.53TB, μmol/L13.3 (9.8–17.8) 14.4 (10.9–17.0) 0.2414.4 (10.6–18.6) 14.4 (10.8–17.0) 0.75PTT, s11.8 (11.2–12.3)11.7(11.2–12.6)0.5311.9 (11.3–12.5)11.6 (11.2–12.4)0.33Tumor size, cm3.5 (2.4–5.1)4.0 (2.5–6.5)0.0453.0 (2.0–5.4)3.5 (2.0–5.5)0.30Tumor number0.540.64 Single173 (84.8)151 (82.5)111 (82.8)108 (80.6) Multiple31 (15.2)32 (17.5)23 (17.2)26 (19.4)Tumor capsule0.430.90 Complete124 (60.8)104 (56.8)76 (56.7)75 (56.0) Incomplete80 (39.2)79 (43.2)58 (43.3)59 (44.0)MVI0.930.50 Negative143 (70.1)129 (70.5)99 (73.9)94 (70.1) Positive61 (29.9)54 (29.5)35 (26.1)40 (29.9)BCLC<0.0010.56 0–A167 (81.9)119 (65.0)105 (78.4)101 (75.4) B37 (18.1)64 (35.0)29 (21.6)33 (24.6) Data are presented as median (interquartile range), mean ± standard deviation or n (%). AFP, α-fetoprotein; ALB, albumin; ALT, alanine aminotransferase; BCLC, Barcelona Clinic Liver Cancer; GGT, γ-glutamyl transferase; HBV, hepatitis B virus; MVI, microvascular invasion; PA-TACE, postoperative adjuvant transarterial chemoembolization; PSM, propensity score matching; PTT, prothrombin time; TB, total bilirubin. The baseline clinicopathological characteristics of the study population Data are presented as median (interquartile range), mean ± standard deviation or n (%). AFP, α-fetoprotein; ALB, albumin; ALT, alanine aminotransferase; BCLC, Barcelona Clinic Liver Cancer; GGT, γ-glutamyl transferase; HBV, hepatitis B virus; MVI, microvascular invasion; PA-TACE, postoperative adjuvant transarterial chemoembolization; PSM, propensity score matching; PTT, prothrombin time; TB, total bilirubin. PA-TACE affects tumor recurrence and prognosis of patients In the primary cohort, the median follow-up duration was 62 months. By the end of the follow-up, 141 (36.4%) patients died and 246 (63.6%) remained alive. Additionally, 184 (47.5%) patients had tumor recurrence during the follow-up period while 203 (52.5%) did not. The median follow-up time was 61.2 months in the no adjuvant treatment control group and 64.3 months in the PA-TACE group. The median time to tumor recurrence (TTR) was 63.0 months in patients who underwent PA-TACE compared with 51.8 months in those who did not (Figure 2A). The 1-, 3-, and 5-year RFS rates in patients who underwent PA-TACE were 90.7%, 71.0%, and 55.7%, respectively, whereas the 1-, 3-, and 5-year RFS rates in patients who did not were 84.4%, 62.7%, and 44.1%, respectively (Figure 2A). There were borderline statistically significant differences of RFS between patients in PA-TACE group and those in no adjuvant treatment control group (HR =0.75, 95% CI: 0.56–1.0, P=0.050). The 1-, 3-, and 5-year OS rates for patients who underwent PA-TACE were 96.2%, 80.9%, and 67.2%, respectively (Figure 2B), whereas the 1-, 3-, and 5-year OS for patients who did not were 95.6%, 78.4%, and 60.8%, respectively (Figure 2B). The OS rates in patients who received PA-TACE were significantly better than those who did not (HR =0.70, 95% CI: 0.50–0.99, P=0.04). PA-TACE affects tumor recurrence and prognosis of patients in the primary cohort and the PSM cohort. (A,B) RFS and OS of patients in the primary cohort. (C,D) RFS and OS of patients in the PSM cohort. Data in parentheses are 95% confidence interval. HR, hazard ratio; OS, overall survival; PA-TACE, postoperative adjuvant transarterial chemoembolization; PSM, propensity score matching; RFS, recurrence-free survival. After 1:1 PSM matching, the median TTR was 62.6 months in patients who underwent PA-TACE compared with 60.9 months in those who did not. The 1-, 3-, and 5-year RFS rates in patients who underwent PA-TACE were 91.8%, 70.9%, and 54.5%, respectively, which were significantly higher than those in patients who did not (87.3%, 66.4%, and 52.2% for 1-, 3-, and 5-year RFS, respectively; HR, 0.70, 95% CI: 0.50–0.99; P=0.04; Figure 2C). After 1:1 PSM, the 1-, 3-, and 5-year OS for patients in PA-TACE treatment group were 97.0%, 78.3%, and 68.7%, respectively, which were significantly better than those in no adjuvant treatment control group (95.5%, 82.8%, and 64.2% for 1-, 3-, and 5-year OS, respectively; HR, 0.65, 95% CI: 0.43–0.98, P=0.04; Figure 2D). Impacts of PA-TACE on tumor recurrence and prognosis in HCC patients at high risk of recurrence To identify high-risk factors for tumor recurrence, we found that age >65 years (HR =1.517, 95% CI: 1.025–2.247, P=0.04), GGT >40 U/L (HR =1.418, 95% CI: 1.040–1.933, P=0.03), MVI-positive (HR =2.856, 95% CI: 2.055–3.969, P<0.001), no PA-TACE (HR =1.587, 95% CI: 1.178–2.137, P=0.002) were independent risk factors for tumor recurrence through the univariate and multivariate Cox analyses (Table 2). The OS rates in patients with age >65 years, GGT >40 U/L, or MVI-positive were significantly poorer compared to those without these clinicopathological characteristics (HR =1.71, 95% CI: 1.13–2.61, P=0.01 for age >65 years; HR =2.56, 95% CI: 1.78–3.67, P<0.001 for GGT >40 U/L; HR =4.73, 95% CI: 3.33–6.71, P<0.001 for MVI-positive) (Figure S1). Table 2Univariate and multivariate analysis of risk factors for recurrenceCharacteristicsTotal (N)Univariate analysisMultivariate analysisHazard ratio (95% CI)P Hazard ratio (95% CI)P Age, years387 ≤65335Reference Reference >65521.623 (1.114–2.433)0.011.517 (1.025–2.247)0.04Gender387 Male319Reference Female680.843 (0.566–1.254)0.40 AFP, ng/mL387 ≤400288Reference >400990.945 (0.669–1.333)0.75 HBV387 Negative38Reference Positive3491.007 (0.619–1.639)0.98 Cirrhosis387 No89Reference Yes2980.902 (0.640–1.270)0.55 ALT, U/L387 ≤40158Reference Reference >402290.749 (0.560–1.002)0.0520.789 (0.578–1.076)0.13ALB, g/L387 ≤45341Reference >45461.203 (0.758–1.908)0.43GGT, U/L387 ≤40181Reference Reference >402061.596 (1.188–2.145)0.0021.418 (1.040–1.933)0.03PTT, s385 ≤13.5359Reference >13.5261.116 (0.647–1.925)0.69 TB, μmol/L387 ≤17.1342Reference >17.1451.106 (0.714–1.712)0.65 Tumor size, cm387 ≤5118Reference >52690.875 (0.637–1.202)0.41 Tumor number387 Single324Reference Multiple631.197 (0.800–1.792)0.38 MVI387 Negative272Reference Reference Positive1152.871 (2.089–3.944)<0.0012.856 (2.055–3.969)<0.001Tumor capsule387 Complete228Reference Incomplete1590.932 (0.693–1.254)0.64 BCLC stage387 0-A286Reference B1011.030 (0.736–1.443)0.86PA-TACE387 Yes183Reference No2041.340 (1.000–1.799)0.0501.587 (1.178–2.137)0.002AFP, α-fetoprotein; ALB, albumin; ALT, alanine aminotransferase; BCLC, Barcelona Clinic Liver Cancer; CI, confidence interval; GGT, γ-glutamyl transferase; HBV, hepatitis B virus; MVI, microvascular invasion; PA-TACE, postoperative adjuvant transarterial chemoembolization; PTT, prothrombin time; TB, total bilirubin. Univariate and multivariate analysis of risk factors for recurrence AFP, α-fetoprotein; ALB, albumin; ALT, alanine aminotransferase; BCLC, Barcelona Clinic Liver Cancer; CI, confidence interval; GGT, γ-glutamyl transferase; HBV, hepatitis B virus; MVI, microvascular invasion; PA-TACE, postoperative adjuvant transarterial chemoembolization; PTT, prothrombin time; TB, total bilirubin. Subgroup analysis showed that PA-TACE significantly prolonged RFS and OS for MVI-positive patients in the primary cohort (HR =0.31, 95% CI: 0.19–0.53, P<0.001 for RFS; HR =0.47, 95% CI: 0.29–0.76, P=0.002 for OS; Figure 3A,3B). After PSM, PA-TACE still significantly improved RFS and OS for MVI-positive patients in the PSM cohort (HR =0.32, 95% CI: 0.17–0.62, P<0.001 for RFS; HR =0.38, 95% CI: 0.20–0.72, P=0.003 for OS; Figure 3C,3D). However, PA-TACE did not improve RFS and OS in patients aged >65 years old both before and after PSM (Figure 3E-3H). Before PSM, PA-TACE did not improve RFS, but improved OS (HR =1.51, 95% CI: 1.01–2.25, P=0.04) in patients with GGT >40 U/L (Figure 3I,3J); after PSM, PA-TACE did not improve RFS and OS in patients with GGT >40 U/L (Figure 3K,3L). Impacts of PA-TACE on the tumor recurrence and prognosis in HCC patients at high risk of recurrence. (A,B) RFS and OS of MVI-positive patients in the primary cohort. (C,D) RFS and OS of MVI-positive patients in the PSM cohort. (E,F) RFS and OS of patients with age >65 years old in the primary cohort. (G,H) RFS and OS of patients with age >65 years old in the PSM cohort. (I,J) RFS and OS of patients with GGT >40 U/L in the primary cohort. (K,L) RFS and OS of patients with GGT >40 U/L in the PSM cohort. Data in parentheses are 95% confidence interval. GGT, γ-glutamyl transferase; HCC, hepatocellular carcinoma; HR, hazard ratio; MVI, microvascular invasion; OS, overall survival; PA-TACE, postoperative adjuvant transarterial chemoembolization; PSM, propensity score matching; RFS, recurrence-free survival. Impacts of PA-TACE on tumor characteristics of recurrent HCC The tumor characteristics of recurrent HCC after liver resection were compared between PA-TACE group and no adjuvant treatment control group. The results showed that the tumor features of recurrent HCC in the PA-TACE group were less advanced compared to those in no adjuvant treatment group. Specifically, recurrent tumors in the PA-TACE group were significantly less likely to be multiple (15.9% vs. 33.3%, P=0.02), >5 cm in tumor size (17.5% vs. 33.3%, P=0.04), and beyond up-to-7 criteria (12.7% vs. 28.6%, P=0.03) (Table 3). Additionally, the proportion of MVI in patients with recurrent HCC who underwent repeat hepatectomy was significantly lower in PA-TACE group than that in no adjuvant treatment control group (23.5% vs. 45.9%, P=0.048). These results show that patients who received PA-TACE developed recurrent HCC with less aggressive tumor characteristics. Table 3Impact of PA-TACE on tumor characteristics of recurrent HCCCharacteristicsPA-TACENo-PA-TACEPTumor number, n (%)0.02 Single53 (84.1)42 (66.7) Multiple10 (15.9)21 (33.3)Tumor size, n (%)0.04 ≤5 cm52 (82.5)42 (66.7) >5 cm11 (17.5)21 (33.3)Beyond up-to-7, n (%)8 (12.7)18 (28.6)0.03Microvascular invasion, n (%)0.048 Negative26 (76.5)20 (54.1) Positive 8 (23.5)17 (45.9)HCC, hepatocellular carcinoma; PA-TACE, postoperative adjuvant transarterial chemoembolization. Impact of PA-TACE on tumor characteristics of recurrent HCC HCC, hepatocellular carcinoma; PA-TACE, postoperative adjuvant transarterial chemoembolization. Safety According to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0 (17), adverse events (AEs) in patients who received PA-TACE are summarized in Table 4. Overall, the PA-TACE treatment was well tolerated. The most common grade 1–2 AEs were nausea/vomiting (46.4%), leukopenia (10.9%), and transient hepatic toxicity (TB elevation, 37.2%; ALB decrease, 13.7%; ALT increase, 23.5%; AST increase, 22.4%; GGT increase, 43.2%). There were no grade 3–5 AEs occurred (Table 4). Table 4Adverse events in the PA-TACE group at the censor of follow-upAdverse eventsGrade†IIIIIIIVVNausea/vomiting850000 Leukopenia128000 TB increase4622000 ALB decrease178000 ALT increase367000 AST increase347000 GGT increase6514000†, grade was graded according to the US National Cancer Institute’s Common Terminology Criteria for Adverse Events version 5.0. ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, γ-glutamyl transferase; PA-TACE, postoperative adjuvant transarterial chemoembolization; TB, total bilirubin. Adverse events in the PA-TACE group at the censor of follow-up †, grade was graded according to the US National Cancer Institute’s Common Terminology Criteria for Adverse Events version 5.0. ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, γ-glutamyl transferase; PA-TACE, postoperative adjuvant transarterial chemoembolization; TB, total bilirubin. Discussion The major challenge of liver resection for HCC is the high rate of tumor recurrence after operation (18). Although advances in surgical techniques, perioperative care, and precise patient selection have reduced surgical complications and patient mortality in recent years, early recurrence rate after liver resection remains high. Therefore, it is urgent to find effective adjuvant therapies to prolong the time to recurrence and the prognosis of patients after liver resection (19). The optimal adjuvant therapy after liver resection in HCC patients is controversial. Tremendous divergences exist in the adjuvant treatment after liver resection between the East and the West, which mainly derive from the difference of the criteria for resectability. In the West, strict resectable criteria are followed and patients with BCLC stage 0/A are suggested to receive curative liver resection. In China or other Asian countries, patients with large, multiple, or BCLC stage B tumors also have the opportunity to undergo curative hepatectomy (mainly defined by complete tumor removal and microscopically negative margins). Therefore, adjuvant therapy may be more necessary for HCC patients in Asian countries. PA-TACE is considered as one of adjuvant treatment options. However, whether PA-TACE could reduce tumor recurrence and prolong survival of HCC patients after liver resection remains controversial. Studies have reported that PA-TACE may provide higher OS and RFS rates than active surveillance alone, and that TACE can be used as an effective adjuvant therapy after liver resection (10,11,20). Many previous studies have shown that PA-TACE can effectively improve RFS and OS in patients with high risk of HCC recurrence (7,21-24). However, in some recent studies, PA-TACE failed to demonstrate its efficacy in HCC patients (25,26). In our study, we revisited the role of PA-TACE and found that PA-TACE significantly improved the time to tumor progression and prognosis in the entire cohort and the PSM cohort of patients. This suggest that PA-TACE can be used as an adjuvant therapy for HCC patients after liver resection, preventing and delaying tumor recurrence, and improving the survival of patients with potential therapeutic implication. To our knowledge, it is a novel finding that PA-TACE reduces the malignancy of recurrent tumors. Interestingly, a significantly less proportion of multiple, tumors >5 cm in size, tumors beyond up-to-7 criteria, and MVI in repeat hepatectomy was observed in recurrent HCC in the PA-TACE group, indicating that patients who received PA-TACE developed recurrent HCC with less aggressive tumor characteristics. This result suggests that PA-TACE may attenuate the malignant features of recurrent tumors, including tumor multiplicity, tumor size, and MVI. This may contribute to the improved prognosis observed in HCC patients who received PA-TACE. In addition, existing studies have shown that postoperative recurrence tumor characteristics, such as tumor number, tumor size, and MVI, are associated with clinical outcomes of patients with recurrent HCC (27-30). The criterion of within up-to-7 is recognized as a prognostic tool for patients with BCLC stage B HCC (31). Saito et al. (32) previously reported that up-to-7 criterion is predictive of both tumor response and OS in patients undergoing conventional TACE. However, the specific molecular mechanisms of PA-TACE on tumor biology of recurrent HCC remain unexplored. We further conducted the univariate and multivariate Cox analysis to identify the risk factors associated with tumor recurrence in the primary cohort. Age >65 years, GGT >40 U/L, MVI-positive, and no PA-TACE were identified as independent risk factors for tumor recurrence. Furthermore, these tumor recurrence-related risk factors were found to be significantly associated with worse OS of patients. Subgroup analysis showed that patients with MVI-positive were able to benefit most from PA-TACE in the primary cohort and the PSM cohort. More previous studies have shown that MVI-positive patients can benefit from PA-TACE treatment, although some other reports suggest that PA-TACE does not offer benefit for MVI-positive patients (11,33). Our findings support the beneficial role of PA-TACE in MVI-positive patients. MVI is known to predict early recurrence of HCC after liver surgery (34-36). PA-TACE may promote necrosis of tumor cells infiltrating the blood vessels by limiting arterial blood flow to the tumor, thus preventing or delaying tumor recurrence (34,37). This may partially explain the beneficial action of PA-TACE. Based on our findings, we recommend PA-TACE for MVI-positive HCC patients after liver resection. Although a previous study has demonstrated that patients with HBV-related HCC and age >65 years old may benefit from PA-TACE (38), our findings showed that age >65 years old was associated with an increased risk of tumor recurrence and PA-TACE showed a limited efficacy in this elderly population. These results indicate that older HCC patients may not be optimal candidates for PA-TACE. In our cohort, GGT was identified as an independent risk factor. Elevated GGT level has been consistently linked to poor clinical outcomes in patients with HCC (39). While a prior study by Huang et al. (40) reported a beneficial effect of PA-TACE in patients with GGT level exceeding 150 U/L, our analysis did not show benefit of PA-TACE in reducing tumor recurrence among patients with GGT level above 40 U/L. A trend toward improved OS was observed in this certain subgroup, suggesting a potential benefit of PA-TACE in patients with elevated GGT level. PA-TACE efficacy across different GGT level strata warrants further investigation. There are several limitations in our study. First, this study was retrospective, which has selection bias and confounding factors. We minimized this limitation with PSM analysis. Second, this study is a single-center retrospective study and lacks multi-center retrospective study to validate and extend the results of this study. Third, imbalances in sample sizes were present in the subgroup analyses, which may introduce potential bias and affect the robustness of the observed associations. Last but not least, it may be interesting to characterize the site of recurrence relative to the vascular distribution of PA-TACE delivered, e.g., is recurrence in the PA-TACE group more likely to occur in liver distribution that is not selectively treated by PA-TACE? does PA-TACE have a local effect or across the entire liver? These questions clearly merit further research. Conclusions PA-TACE improves RFS and OS of HCC patients after curative-intent liver resection, especially in MVI-positive HCC patients. Furthermore, PA-TACE attenuates the malignancy of recurrent tumors, likely contributing to improved patient prognosis. Our study provides the evidence that PA-TACE may be considered as an adjuvant treatment option following liver resection in HCC patients, especially in MVI-positive patients. Supplementary The article’s supplementary files as Acknowledgments None. Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This retrospective study was approved by the Institutional Review Boards of Zhongshan Hospital, Fudan University, Shanghai, China (No. B2025-587). Due to its retrospective nature, informed consent for this retrospective study was waived. Footnotes Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-655/rc Funding: None. Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-655/coif). The authors have no conflicts of interest to declare. Data Sharing Statement Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-655/dss References Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-655/prf Supplementary Materials The article’s supplementary files as Data Availability Statement Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-655/dss