Treatment Duration of Adjuvant Immune Checkpoint Inhibitors in Hepatocellular Carcinoma Patients at High Risk of Recurrence after Resection: A Prospective, Multicentric Cohort Study

Hepatocellular carcinoma (HCC), known for its frequent occurrence and dismal prognosis, is the sixth most prevalent malignancy and the third leading cause of cancer-related mortality worldwide [1]. Global trends indicate that annual HCC diagnoses will increase by a staggering 50% between 2020 and 2040 [2], particularly in the Western hemisphere, due to the transition from disease related to virus toward disease of nonviral etiologies [3].

Hepatic resection is the first-line treatment for early-stage HCC; in Asia, some patients with intermediate-to-advanced HCC may also be recommended for resection if they have preserved liver function [4, 5]. However, up to 70% of HCC patients experience recurrence within 5 years after hepatic resection, and recurrence is particularly common among patients with high-risk factors such as tumors exceeding 5 cm in diameter, macro/microvascular invasion, microsatellite lesions, or multinodularity [6‒8]. This substantial recurrence rate contributes to significant mortality and highlights the critical need for effective postoperative prophylactic measures to enhance overall survival (OS). Accordingly, many investigators have assessed the potential of adjuvant therapies, including cytokine-induced killer cell immunotherapy and transarterial chemoembolization [9]. However, none of these modalities is recommended as a standard adjuvant therapy in most HCC guidelines [4, 7, 8]. There remains an urgent need for more effective, universally accepted adjuvant treatments to improve long-term survival.

Immune checkpoint inhibitors (ICIs), used either individually or in combination with molecular targeted agents, are currently the first-line therapeutic approach for advanced HCC [6, 10‒12]. However, there is no consensus regarding ICI efficacy in preventing HCC recurrence after curative resection or regarding the optimal ICI treatment duration. Although several studies have confirmed the efficacy of ICIs for adjuvant therapy [13‒15], the optimal treatment duration for adjuvant ICIs remains unknown. Current guidelines [6, 7] stipulate that the duration of adjuvant ICI therapy should not exceed 1 year but do not specify a minimal or recommended duration. Therefore, there is a critical need to establish the optimal ICI treatment duration for preventing tumor recurrence following curative hepatic resection in HCC patients at high risk of recurrence.

This study evaluated the efficacy and safety of adjuvant ICI treatment, on its own or combined with molecular targeted therapy, in HCC patients at high risk of recurrence. In addition, the optimal duration of ICI treatment was examined.

The study methodology adhered strictly to the principles outlined in the Declaration of Helsinki (1975) and its revisions. The Ethical Review Boards at Guangxi Medical University Cancer Hospital (LW2021103), Wuzhou People’s Hospital (2024045), and Guigang City People’s Hospital (ELW-2024-016-01) approved this study. Prior to commencing the treatment protocol, written informed consent was obtained from all participants. The study was registered with ClinicalTrials.gov (NCT05221398), and preliminary findings have already been published [16].

This prospective, multicentric cohort study enrolled HCC patients who underwent hepatic resection from January 1, 2019, to December 31, 2023, at Guangxi Medical University Cancer Hospital, Wuzhou People’s Hospital, and Guigang City People’s Hospital and who met the inclusion criteria. The clinical baseline characteristics, diagnostic modalities, therapeutic interventions, and postoperative follow-up data of all eligible patients at each participating center were rigorously and meticulously collected prospectively in a REDCap database.

The inclusion criteria were as follows: (1) eligibility for curative resection, in accordance with the definition outlined in the Guidelines for the Diagnosis and Treatment of Primary Liver Cancer (2019 edition) [17] (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000542954), (2) confirmation of HCC by postoperative pathology; (3) manifestation of at least one factor associated with a high risk of postoperative recurrence based on the criteria of the IMbrave050 study [13], which could be tumor multiplicity, tumor diameter exceeding 5 cm, Edmondson grade III–IV, or evidence of microvascular or macrovascular invasion based on preoperative imaging or postoperative pathological evaluation; (4) Eastern Cooperative Oncology Group performance status score of 0 or 1; (5) liver function assessment within Child-Pugh grades 5–7; (6) age 18–75 years; and (7) anticipated survival time exceeding 6 months.

The exclusion criteria were as follows: (1) prior treatment with antineoplastic therapies such as neoadjuvant/conversion therapy and radiotherapy before hepatic resection; (2) postoperative treatment with other adjuvant agents such as tyrosine kinase inhibitor (TKI) monotherapy or transarterial chemoembolization; (3) history of other malignancies; (4) occurrence of grade 4–5 postoperative complications before adjuvant ICI therapy; or (5) tumor recurrence within 2 months after resection.

Patients were encouraged to revisit the hospital within 4–8 weeks after their operation for a comprehensive assessment including physical examination, hematology analysis, and radiological imaging. For patients at high risk of recurrence, if their general condition was good and tests suggested no contraindications, their physicians recommended starting adjuvant ICI therapy with or without molecular targeted agents for a maximum of 12 months, with the exception of extraordinary scenarios (e.g., voluntary withdrawal of patient consent, poor compliance, disease progression, intolerable toxicity, mortality, or any other conditions specified in the protocol that necessitated termination of treatment). Adjuvant therapy for all patients who received it was initiated 4–8 weeks after hepatic resection. Since the efficacy of adjuvant ICIs with or without molecular targeted agents was not clear, the decision of whether to use adjuvant ICI treatment was made jointly by the patient or his family after consultation with the attending physician, who did not attempt to influence the choice in either direction. Therefore, fewer than 20% of patients at high risk of recurrence received adjuvant ICIs with or without molecular targeted agents.

Some patients were treated with ICIs for shorter than 6 months due to economic reasons, poor compliance, adverse events after antitumor therapy, or abnormal liver function. In this study, all patients received at least one course of adjuvant ICI therapy. Patients at different centers may have received different ICIs in accordance with routine procedures or physician experience preferences.

The ICIs most frequently prescribed were tislelizumab (BGB-A317; BeiGene, Beijing, China), toripalimab (Shanghai Junshi Biosciences, Shanghai, China), camrelizumab (SHR-1210; Jiangsu Hengrui Medicine, Jiangsu, China), sintilimab (IBI308; Innovent Biologics, Suzhou, China), atezolizumab (Genentech, San Francisco, CA, USA), and cadonilimab (Kangfang Pharmaceutical, Zhongshan, China). The agents were administered intravenously every 3 weeks following the dosage guidelines of the supplier. Patient vital signs and mental status were closely monitored during infusion and for 1 h after infusion to ensure safety.

The molecular targeted agents most frequently prescribed were lenvatinib (Eisai, Woodcliff Lake, NJ, USA), bevacizumab (Genentech), apatinib (Jiangsu Hengrui Medicine), and donafenib (Suzhou Zelgen Biologics, Suzhou, China). If combination therapy with TKIs was administered, the TKIs were administered once daily at an individualized dosage based on body weight or the supplier’s specifications. Treatment began 4–8 weeks after operation with rigorous monitoring for the occurrence of adverse events. HCC patients who were seropositive for HBsAg or hepatitis B virus DNA received long-term antiviral therapy with tenofovir or entecavir [18].

During the administration of adjuvant ICIs, indiscriminate ICI dose reduction was avoided except in exceptional cases. Immune-related adverse events were evaluated in accordance with the stringent criteria set forth in the National Cancer Institute’s Common Terminology Criteria for Adverse Events (CTCAE), version 5.0 [19]. If adverse events related to ICIs or molecular targeted therapies exceeded grade 2 during follow-up, dose reduction or temporary treatment cessation was recommended after a comprehensive and rigorous assessment. Strict monitoring continued until the adverse event grade decreased to 1 or 2 following reassessment after resumption of medication. In the event of recurrence or intolerable adverse events, adjuvant therapy was discontinued. Immunotherapy for immune-related adverse events adhered to ICI-related guidelines, ensuring that optimal protocols were followed [20].

All patients underwent rigorous follow-up starting with monthly visits for the first 2 months after resection. Subsequently, follow-up was conducted at intervals of 2–3 months and then semiannually after 2 years of surgery. Each follow-up session included physical exams, blood tests (e.g., alpha-fetoprotein), and imaging (e.g., ultrasonography, CT, or MRI). If the suspected recurrences or metastases could not be clearly diagnosed, percutaneous biopsy was performed for histopathological confirmation. The multimodal diagnosis of HCC recurrence or metastasis took into account clinical history, physical findings, and the blood test and imaging results, optionally complemented by liver tissue biopsy. Confirmed cases of recurrence or metastasis were treated with regimens tailored to the patient’s overall health, liver function, Eastern Cooperative Oncology Group performance status, tumor characteristics, and tumor quantity [21].

Cirrhosis and fatty liver were diagnosed based on postoperative pathology. The categorization of postoperative complications was based on the revised Clavien-Dindo grading system [22]. Liver function after adjuvant ICI treatment was assessed using the albumin-bilirubin scoring model [23], and the timing of the assessment was based on the first blood test results upon admission for recurrence or metastasis. Alcoholic liver disease was diagnosed based on consensus guidelines [24]. Alcoholism did not include occasional social drinking.

The primary endpoint was recurrence-free survival (RFS), defined as the time from the first day after hepatic resection to the date of first diagnosis of tumor recurrence/metastasis or death from any cause (whichever occurred first). The secondary endpoint was OS, defined as the time from the first day after hepatic resection to the date of death from any cause or the end of follow-up (May 30, 2024). Other secondary endpoints included treatment-related adverse events, which were categorized according to CTCAE, version 5.0 [19].

Normally distributed continuous variables were expressed as mean ± standard deviation and analyzed with Student’s t test or Mann-Whitney U test. Variables with skewed distributions were expressed as the median and interquartile range and analyzed by Wilcoxon or Kruskal-Wallis test. Categorical variables were described as counts and percent and analyzed with Pearson’s chi-square or Fisher’s exact test.

RFS and OS were calculated using the Kaplan-Meier method, and differences between groups were assessed by log-rank test. To identify risk factors of recurrence, a Cox regression model was used for multivariate analysis, and variables with p < 0.05 in the univariate analysis were used in multivariate analysis. The results were described using the hazard ratio (HR) and 95% confidence interval (CI). Forest plots were constructed to illustrate subgroup analyses based on factors such as age, sex, and tumor size. One subgroup analysis stratified patients who received adjuvant ICI therapy by duration of treatment. Patients with HCC recurrence within 2-6 months post-resection were excluded. This subgroup analysis did not include patients who underwent active surveillance.

Survival comparisons were conducted after propensity score matching (PSM) in order to adjust for potential confounding. Logistic regression was used to create cohorts from each of the 2 patient groups that were matched to each other in sex, age, diabetes mellitus, family history of HCC, total bilirubin, albumin, alanine aminotransferase, HBV DNA, HBsAg, anti-HCV antibodies, alpha-fetoprotein, Eastern Cooperative Oncology Group Performance Status, Child-Pugh grade, liver cirrhosis, tumor rupture, fatty liver disease, BCLC stage, maximum tumor size, number of intrahepatic nodules, mac/microvascular invasion, Edmondson-Steiner grade, and tumor satellites. The two groups were matched using the nearest neighbor method at a ratio of 1:2 or 1:1, with a caliper of 0.2 to ensure comparability. Analyses were performed in SPSS 22.0 (IBM, Chicago, IL, USA) and R 4.2.11 (R Project, Vienna, Austria), with significance set at p < 0.05.

Between January 1, 2019, and December 31, 2023, 3,304 patients with HCC underwent hepatic resection at one of the three hospitals. Among them, 589 underwent non-curative HCC resection, 373 lacked high risk factors of recurrence, and 36 experienced severe postoperative complications. Another 993 received alternative adjuvant therapies and were excluded. The remaining 1,313 patients were enrolled for further study (Fig. 1). Among them, 239 were included in the adjuvant treatment cohort: 126 (52.7%) received adjuvant ICIs alone, and 113 (47.3%) received ICIs plus molecular targeted therapy. The remaining 1,074 patients did not receive any postoperative adjuvant treatment and were included in the active surveillance cohort. Forty-two patients in the active surveillance cohort were lost to follow-up, so only the remaining 1,032 were included in the final analysis.

Most patients in our study (80.9%) were seropositive for HBsAg (Table 1). Some other patients (10.7%) were seronegative for HBsAg but seropositive for HBeAb, HBeAg, or HBcAb. These cohort characteristics are consistent with the fact that more than 90% of HCC patients in China have a background of HBV infection [25]. Only 1.8% of patients were seropositive for HCV antigens or antibodies. Nearly a quarter of patients (21.95%) had fatty liver disease, most of whom were combined with chronically infected with HBV. Few patients in our study had alcoholic liver disease, and all of them were combined with chronically infected with HBV. Subgroup analyses were performed based on HBsAg, HBV DNA, or fatty liver disease showed that patients with HBV DNA ≥20 IU/mL, positive HBsAg, and no fatty liver disease may significantly benefit from adjuvant ICI therapy (online suppl. Fig. 1–2).

In the adjuvant treatment cohort, 99 patients received tislelizumab (anti-PD-1), 59 received sintilimab (anti-PD-1), 42 received camrelizumab (anti-PD-1), 20 received toripalimab (anti-PD-1), and 19 received other ICIs (online suppl. Table 2). In the analysis of treatment duration, 39 patients who received adjuvant ICI therapy were excluded because they suffered tumor recurrence anytime within 2-6 months after resection. Among the remaining patients who received adjuvant treatment, 95 received it for up to 6 months, whereas 105 received it for longer than 6 months (Fig. 1).

To balance baseline variations, we used 1:2 PSM to select 239 patients from the adjuvant treatment cohort and 478 from the active surveillance cohort. The comprehensive baseline and preoperative clinical profiles of the two cohorts, both before and after PSM, are outlined in Table 1. Before PSM, the individuals in the adjuvant cohort were younger (50.8 years vs. 54.1 years; p < 0.001) and had more satellite nodules (16.7% vs. 9.8%; p = 0.003) compared with the patients in the active surveillance cohort. However, after PSM, no significant differences in baseline characteristics were found between the two cohorts. The rates of postoperative antiviral therapy were similar in both cohorts before or after PSM.

During a median follow-up of 30.2 months (IQR 18.5–44.1 months), 113 (47.3%) patients in the adjuvant therapy cohort and 696 (67.4%) patients in the active surveillance cohort experienced recurrence or metastasis. The recurrence locations, which are described in online supplementary Table 3, were predominantly intrahepatic. Notably, patients who experienced recurrence in adjuvant ICI or active surveillance group showed similar albumin-bilirubin scores, suggesting that adjuvant therapy did not compromise residual liver function (online suppl. Fig. 3).

The median RFS was significantly higher in the adjuvant therapy cohort (22.6 months; 95% CI 18.3–26.9) than in the active surveillance cohort (19.1 months; 95% CI 16.4–21.4) (HR 0.79, 95% CI 0.66–0.95, p = 0.019; Fig. 2a). The corresponding RFS rates at 1, 2, and 3 years were 65.0%, 49.4%, and 40.6% in the adjuvant therapy cohort and 61.7%, 45.0%, and 28.4% in the active surveillance cohort. The beneficial effect of ICI adjuvant therapy was confirmed among the patients after PSM (HR 0.75, 95% CI 0.61–0.92, p = 0.014; Fig. 2b).

The mortality in the adjuvant therapy cohort was 19.2% (46 patients), whereas it reached 29.6% (303 patients) in the active surveillance cohort. The median OS was not reached in either group, but OS tended to be higher in the adjuvant therapy cohort than in the active surveillance cohort (HR 0.72, 95% CI 0.54–0.94; p = 0.010; Fig. 2c). Specifically, OS rates at 1, 2, and 3 years were 91.9%, 82.7%, and 72.4% in the adjuvant therapy cohort compared with 86.6%, 74.7%, and 65.7% in the active surveillance cohort. This enduring survival benefit of adjuvant therapy was substantiated by the analysis conducted after PSM (HR 0.71, 95% CI 0.52–0.96, p = 0.040; Fig. 2d).

Online supplementary Table 4 outlines the post-recurrence treatments for both cohorts. Notably, the proportion of patients who received interventions after recurrence was similar between groups, indicating that these treatments did not contribute to the improved prognosis observed in the adjuvant therapy cohort.

Adjuvant ICI treatment significantly reduced recurrence in select subgroups, such as individuals ≥65 years, men, those with hepatitis B virus DNA ≥20 IU/mL, those with alpha-fetoprotein ≥400 ng/mL, those with cirrhosis, those without tumor rupture or fatty liver disease, those with advanced stage disease, those with single tumors or tumors larger than 5 cm, those with macrovascular or microvascular invasion, those with tumor satellites, and those with Edmondson-Steiner grade III/IV (online suppl. Fig. 1). In terms of OS, adjuvant ICI treatment had the best effect in individuals ≥65 years, those with low hepatitis B virus DNA, those with alpha-fetoprotein ≥400 ng/mL, those with cirrhosis, those without tumor rupture or fatty liver disease, those with advanced disease, those with tumors larger than 5 cm, those with macrovascular or microvascular invasion, and those with Edmondson-Steiner grade III/IV (online suppl. Fig. 2). We observed comparable HR trends in the post-PSM population.

Uni- and multivariate Cox regression analyses identified several independent predictors of poor prognosis: HBV DNA >20 IU/mL, Child-Pugh B, tumor rupture, tumor diameter >5 cm, multiple tumors, macrovascular invasion, and microvascular invasion for RFS; and Child-Pugh B, tumor rupture, tumor diameter >5 cm, microvascular invasion, and tumor satellites for OS (online suppl. Table 5).

Three quarters (73.6%) of patients who received adjuvant therapy experienced adverse events of any grade, including 87 patients (69.0%) who received ICI monotherapy and 89 (78.8%) who received ICIs plus molecular targeted therapy (p = 0.106). A total of 124 (51.9%) patients experienced grade 1–2 events, while 52 (21.8%) patients experienced grade 3–4 events; no grade 5 adverse event was observed (online suppl. Table 6). The most frequent adverse events involved the liver and blood system and included hyperbilirubinemia (30.1%), elevated aspartate aminotransferase (26.8%), elevated alanine aminotransferase (25.9%), neutropenia (25.5%), and decreased platelet count (22.3%). The main grade 3–4 events were hand-foot skin reaction (7.1%), fatigue (4.2%), anorexia (3.3%), nausea (2.5%), and thrombocytopenia (2.5%).

After exclusion of patients who experienced tumor recurrence or metastasis within 6 months after hepatic resection, analysis showed that patients who received adjuvant ICIs for longer than 6 months tended to have slightly higher RFS than those who received ICIs for up to 6 months (HR 0.66, 95% CI 0.42–1.04, p = 0.071; Fig. 3a). They also tended to have higher OS (HR 0.59, 95% CI 0.30–1.17, p = 0.128; Fig. 3b). In both cases, however, the survival difference was not statistically significant. Importantly, the duration of adjuvant ICI treatment did not affect the sites of recurrence (online suppl. Table 7), and no significant differences in post-recurrence treatment modalities were observed between the two duration groups (online suppl. Table 8).

Because of the imbalance in clinical baseline data between the two groups (online suppl. Table 9), 60 pairs of patients were generated through PSM 1:1. Patients who received adjuvant ICIs for longer than 6 months tended to survive longer than those who received them for up to 6 months, whether in terms of RFS (HR 0.68, 95% CI 0.38–1.23, p = 0.198; Fig. 3c) or OS (HR 0.47, 95% CI 0.20–1.08, p = 0.078; Fig. 3d).

Patients who received adjuvant ICI therapy for longer than 6 months had significantly higher frequency of adverse events of any grade than those who received ICIs for up to 6 months (83.8% vs. 64.2%, p = 0.002; online suppl. Table 10). The adverse events whose frequency differed the most between the two groups were dermatological and hepatic adverse events, including hand-foot skin reaction, reactive cutaneous capillary endothelial proliferation, hyperbilirubinemia, elevated alanine aminotransferase, and elevated aspartate aminotransferase. However, the rates of grade 3–4 adverse events were similar between the two groups.

Although several HCC guidelines recommend adjuvant ICIs for HCC patients with high risk factors of recurrence, they do not specify the optimal duration of ICI treatment. Based on a large patient cohort and an extended follow-up period, we reaffirmed the substantial positive effect of adjuvant ICI therapy, either alone or in combination with molecular targeted drugs, on the prognosis of HCC patients with one or more high-risk factors of recurrence after curative hepatic resection. Moreover, the adjuvant ICIs did not appreciably increase the risk of severe adverse events; most observed events were grade 1 or 2, indicating a favorable safety profile and excellent tolerability. Individuals who received adjuvant ICI therapy for longer than 6 months exhibited slightly higher RFS and OS than those who received such therapy for up to 6 months, although the differences were not statistically significant. However, the rate of adverse events, which included skin-related complications and liver dysfunction, was significantly higher among patients who received adjuvant therapy for longer than 6 months. Thus, extensive trials are needed to confirm the safety and efficacy of adjuvant ICIs and clarify which patients need more than 6 months of treatment.

Recent small cohort trials have focused narrowly on assessing the efficacy of ICIs after HCC resection [26‒28], resulting in a lack of consensus regarding their post-surgical benefits. The IMbrave050 trial showed a positive effect on RFS at the pre-specified interim analysis, validating the significance of adjuvant atezolizumab plus bevacizumab therapy for HCC [13]. Thus, atezolizumab plus bevacizumab is recommended by some clinical guidelines [6‒8]. However, the latest updated analysis [29] did not confirm the initially observed RFS benefit of atezolizumab plus bevacizumab. In contrast, a phase II randomized controlled trial revealed that adjuvant sintilimab significantly improved RFS among HCC patients with microvascular invasion [14]. Multiple phase II/III trials are in progress to examine adjuvant ICIs (e.g., CheckMate-9DX, KEYNOTE-937, EMERALD-2). Although retrospective studies have also revealed that adjuvant ICIs improve prognoses for patients at high risk of recurrence [30‒32], more robust evidence remains critical.

The above studies were limited to assessing the therapeutic benefits of adjuvant ICIs after hepatic resection; they did not assess the effect of ICI treatment duration on efficacy. As adjuvant ICI treatment is more frequently recommended after hepatic resection, the choice of duration can be challenging for clinicians and patients because of differences in reaction, onset, and cost. Treatment duration may play a key role in the efficacy of ICIs. Our study showed that adjuvant ICIs significantly improved patient prognosis, with a more pronounced improvement trend after longer 6 months than after up to 6 months. A clearer delineation of treatment duration for patients at high risk of recurrence after curative resection would help patients and clinicians make more informed decisions about treatment duration in consideration of multiple factors including cost-effectiveness and adverse events. Ultimately, this would lead to optimal therapeutic regimens customized for individual patients [33, 34].

The optimal duration of adjuvant therapy after curative resection remains debated. For advanced non-small cell lung cancer, Zalcman et al. [35] observed comparable therapeutic efficacy between short-term (6-month) and long-term ICI therapy, suggesting equivalence in their benefits. In contrast, Waterhouse et al. [36] suggested that sustained ICI therapy resulted in greater efficacy for advanced non-small-cell lung cancer. In current phase III trials (e.g., CheckMate-9DX, KEYNOTE-937, JUPITER 04, SHR-1210-III-325, EMERALD-2, DaDaLi) adjuvant ICIs were administered for up to 12 months. Our results also suggest that a longer treatment duration (>6 months) may be more beneficial to patients. However, the average pooled median progression-free survival in reported studies of first-line ICIs for advanced HCC is only 5.3 months (95% CI: 3.89–6.80, online suppl. Table 11), with tumor response rates ranging from 10% to 36%. Therefore, whether 12 months of adjuvant therapy after resection is necessary needs further validation.

The use of effective adjuvant monotherapies with short treatment durations would undoubtedly reduce the financial burden and national healthcare costs, especially in developing countries. Moreover, the trailing effect of ICIs coupled with their prolonged use inevitably increases the incidence of immune-related adverse events. The heightened prevalence and severity of adverse events observed in this study when adjuvant ICI therapy lasted longer than 6 months underscores this concern. The Kaplan-Meier curves show obvious differences between patients receiving treatment for different durations. It seems that patients receiving longer than 6 months had better RFS and OS, although the p values did not reach significance. This implies that 6 months of adjuvant therapy may not be enough. In contrast to our findings, prior studies support the safety and efficacy of 6-month postoperative adjuvant treatment with sintilimab [14] or shorter durations of tislelizumab plus TKI [15]. We eagerly await rigorous randomized control trials that clarify the appropriate adjuvant treatment duration.

In the present study, we found that patients with fatty liver disease did not benefit from adjuvant ICI treatment, in agreement with our preliminary findings on immunotherapy for nonalcoholic fatty liver disease (NAFLD)-related HCC [37, 38]. Similarly, Pfister et al. [39] reported that patients with NAFLD were not optimal for ICI therapy. However, in a post hoc etiological analysis of the IMbrave150 study, Espinoza et al. [40] found that the prognoses of patients with NAFLD-related HCC were comparable to those of patients with other HCC etiologies. These conflicting results demonstrate the need for further clinical trials on the efficacy of ICI therapy in NAFLD-related HCC.

Our study has several limitations. The absence of randomization in patient enrollment poses a risk of selection bias. In addition, despite the respectable sample size, the predominance of Asian patients coupled with a high proportion of virus-related HCC may undermine the generalizability of the findings. Third, due to the predominant focus on programmed death 1 inhibitors, further investigations of inhibitors of programmed death 1 ligand, cytotoxic T lymphocyte-associated protein 4, and dual-antibody inhibitors are needed for a comprehensive understanding. Ultimately, prolonged follow-up is essential to assess the long-term effects of adjuvant ICI therapy on the outcomes and quality of life among HCC patients, regardless of the drug type and treatment duration.

This prospective cohort study suggests that adjuvant ICI treatment after curative resection has the potential to improve prognosis among HCC patients at high risk of recurrence. Importantly, the findings indicate that 6 months of adjuvant treatment may be insufficient. These pivotal findings should prompt a shift in strategies for adjuvant ICI therapy, especially for patients who have factors that put them at high risk of recurrence after resection. Additional randomized, controlled, and large-scale clinical trials are critical to conclusively determine the optimal duration of adjuvant ICI treatment for HCC.

The authors thank AiMi Academic Services (www.aimieditor.com) for English language editing services.

The study methodology adhered strictly to the principles outlined in the Declaration of Helsinki (1975) and its revisions. The Ethics Review Boards at Guangxi Medical University Cancer Hospital (LW2021103), Wuzhou People’s Hospital (2024045), and Guigang City People’s Hospital (ELW-2024-016-01) approved this study. Written informed consent was obtained from all participants before the treatment protocol began. The study is registered with ClinicalTrials.gov (NCT05221398).

The authors have no competing interests to declare.

This work was supported by the Guangxi key research and development plan (GuiKe AB24010082), the National Natural Science Foundation of China (82260569), the Key Laboratory of Early Prevention and Treatment for Regional High Frequency Tumor (Guangxi Medical University), Ministry of Education (GKE-ZZ202217, GKE-ZZ202311, and GKE-ZZ202405), the Innovation Project of Guangxi Graduate Education (YCSW2023245), First-class Discipline Innovation-Driven Talent Program of Guangxi Medical University, the Project for Enhancing the Basic Research Capacity of Young and Middle-aged Teachers in Guangxi Universities (2023KY0106), and the Science and Technology Key Project of Guigang City (2300008). Funders were not involved in the study design, data collection, analysis, interpretation, or manuscript writing. The corresponding authors had full access to the data, and they assume ultimate responsibility for the publication decision. Guarantor of the article is J.-H.Z.

J.-H.Z. conceived the study. J.-Y.S., S.-P.L., X.-L.X., J.-J.O., P.-H.Y., T.-B.Z., J.-S.C., Q.-M.L., J.-R.L., F.-M.T., J.-R.L., D.-L.Y., Z.-J.D., L.-X.P., and Y.-J.L. collected and analyzed the data. L.L., Z.-M.Q., X.-M.L., and J.-H.Z. analyzed data. J.-Y.S. and J.-H.Z. drafted the manuscript. L.M., Y.-L.M., and J.-H.Z. revised the manuscript. All authors read and approved the final version to be published.

Jia-Yong Su, Shao-Ping Liu, Xiao-Ling Xu, and Jun-Jie Ou contributed equally to this work.Partial data were presented at the 2024 ESMO Conference (https://doi.org/10.1016/j.annonc.2024.08.1008).

All available data are included in the article. More original data can be obtained from the corresponding authors in accordance with privacy/ethical restrictions.