Hepatocellular Carcinoma (HCC) is one of the most prevalent malignant tumors globally and ranks as the third leading cause of cancer-related deaths worldwide, with a relative 5-year survival rate of only about 18% (1, 2). Surgical resection stands as the primary treatment choice for HCC patients, particularly for those with a single tumor and preserved normal liver function (3). However, even among resectable HCC patients, long-term survival outcomes post-surgery are less than ideal. Up to 80% face the risk of recurrence within 5 years after surgery, with many diagnosed patients already in the late stages of the disease, presenting visible vascular invasion and multifocal intrahepatic metastases (4, 5). Neoadjuvant therapy, incorporating various treatment modalities such as drugs and radiation, aims to improve overall treatment efficacy, reduce recurrence risk, and shrink lesions, transforming initially unresectable diseases into manageable therapeutic strategies, offering multiple potential advantages for HCC treatment (6, 7). Despite the absence of established consensus or standardized protocols for neoadjuvant therapy in HCC patients, evaluations have been carried out regarding potential neoadjuvant options like transarterial chemoembolization and transarterial radioembolization. However, the results have not yielded satisfactory survival benefits and are not considered optimal choices for improving overall survival in patients (8, 9). In recent years, global guidelines for HCC encompass five monotherapy molecular treatment approaches. Among these, immune checkpoint inhibitors, primarily PD-1/PD-L1 inhibitors, have opened up new avenues for neoadjuvant immunotherapy in HCC patients (10–12).The PD-1/PD-L1 pathway serves as a crucial mediator of local immune suppression within the tumor microenvironment (TME). By blocking negative regulatory mechanisms, it reactivates T-cell immune responses against tumors, aiming for antitumor effects (13, 14). The use of anti-PD-(L)1 immunotherapy in the neoadjuvant (preoperative) setting presents a potential therapeutic option for achieving “resectable cure” in tumors (15). Furthermore, combination therapies involving immune checkpoint inhibitors and anti-angiogenic agents exhibit enhanced survival benefits for HCC patients compared to standalone immune checkpoint inhibitor therapy (16). However, clinical studies supporting neoadjuvant immunotherapy for HCC are relatively scarce, and its treatment efficacy and safety remain unknown. There is a lack of standardized protocols for neoadjuvant immunotherapy in clinical practice (7).

Hence, based on the current clinical research findings, this article aims to discuss the effectiveness and safety of neoadjuvant immunotherapy in resectable HCC patients. The objective is to provide an objective and comprehensive evaluation of existing studies, offering more reliable and stable references. This discussion aims to support the feasibility assessment of neoadjuvant immunotherapy in HCC treatment and provide additional evidence-based medicine for clinical treatment.

This study followed the PRISMA reporting guidelines for systematic reviews and meta-analysis (17). This meta-analysis has been registered in PROSPERO with the identifier CRD42023474276.

Overall, this study’s results demonstrate the advantages of neoadjuvant immunotherapy in resectable HCC. First, regarding efficacy, the meta-analysis showed an overall odds ratio (OR) of 0.37 (95% CI 0.20-0.69) for ORR, with the maximum ORR reported at 54.2% (27). The summarized ORs for pCR and MPR were 0.22 (95% CI 0.14-0.36) and (95% CI 0.31-0.70), respectively, with the maximum pCR and MPR reported at 38% and 56% (28), respectively. These results align closely with neoadjuvant immunotherapy outcomes in other tumor types (29). However, considering the ongoing nature of most studies, data regarding post-tumor resection patient survival are limited. Only four articles provided statistical data, among which Kaseb, A.O. et al. (22) reported nivolumab’s PFS as 9.4 months (95% CI 1.47–not estimable [NE]), and nivolumab plus ipilimumab’s PFS as 19.53 months (2.33–NE). Other studies (21, 26) reported EFS and RFS data with a median EFS of 13.8 months (95% CI 10.3-17.3) and a 1-year RFS of 53.85% (95% CI: 24.77%-75.99%). However, due to insufficient follow-up, none of the studies reported OS data. In other tumors, the significant correlation between pathological response and survival has been validated (30). This study also conducted statistical analysis, as Ho, W. J et al. (19) found a correlation between achieving MPR (Major Pathological Response) and long-term DFS (Disease-Free Survival). So far, all patients have had a DFS interval exceeding 230 days. Kaseb, A. O. and colleagues (22) similarly reported a significant difference in recurrence-free survival based on the presence or absence of a significant pathological response (p=0.049), despite a smaller sample size. Six patients who experienced a significant pathological response did not experience recurrence at a median follow-up of 26.8 months, whereas among the 14 patients who did not show a significant pathological response, seven experienced recurrence. Xia, Y et al. (21), on the other hand, found a higher RFS (Recurrence-Free Survival) in MPR/pCR (Pathological Complete Response) patients compared to non-MPR patients, although it did not reach statistical significance, possibly due to the small sample size.

Post-hepatectomy recurrence is a common cause of disease progression in HCC (31). Early recurrence is often the result of occult intrahepatic metastasis, closely associated with tumor burden, accounting for approximately 70% of all recurrence cases. Studies indicate that early recurrence significantly decreases overall survival and disease-free survival rates (32, 33). Neoadjuvant immunotherapy is believed to induce long-term remission by eliciting antitumor immune responses before primary tumor resection (34). This therapy correlates with improved pathological responses, enhanced disease-free survival, and expansion of T and B cell reservoirs within the tumor. Additionally, compared to adjuvant therapy, neoadjuvant immunotherapy demonstrates higher efficacy in eradicating metastatic disease (35, 36). Marron, T.U. et al. (25)found that patients with tumor necrosis rates exceeding 50% after cemiplimab treatment exhibited stronger immune infiltration compared to patients with minimal or absent necrosis in surgical samples. Similarly, Kaseb, A.O. et al. (22), analyzing tissue samples pre and post-resection, correlated treatment response with immune cell infiltration. They discovered a correlation between T cell activation positive (VISTA+) bone marrow cell clusters and significant pathology reactions. This phenomenon aligns with other studies in various tumor types (37, 38). Xia, Y. et al. (21) further reported that compared to patients with large tumor lesions, neoadjuvant immunotherapy has shown promising outcomes, particularly in HCC patients, especially those with a single lesion. This may be attributed to the increased presence of immune cell infiltration and upregulation of immune pathways in smaller tumor lesions. Variations in individual responses to immunotherapy among different patients suggest a correlation between the immune status of tumor lesions and individual anti-tumor responses and survival benefits.

Regarding the safety of neoadjuvant immunotherapy, the overall odds ratio for grade 3-4 TRAEs was 0.35 (95% CI 0.24-0.51). Most immunotherapy-related adverse effects were manageable, some resolving after treatment discontinuation, and others responding well to corticosteroid therapy. Overall, neoadjuvant immunotherapy in HCC demonstrated safety akin to previous studies in gastrointestinal tumors (39).Furthermore, the highest occurrence rate of grade 3-4 adverse events reported in this study was 43% (reported by Kaseb, A.O. et al. (22)—the only RCT in this article). The study also found that the combination of nivolumab and ipilimumab led to a higher rate of grade 3-4 treatment-related adverse events compared to nivolumab alone. However, this difference, at 20% ([95% CI –14.7% to 38.7], p=0.69), was not significant, suggesting that different combination therapies might result in varied outcomes, prompting considerations on optimizing drug usage to maximize patient benefit. Furthermore, the overall surgical resection rate after neoadjuvant immunotherapy showed an OR of 3.08 (95% CI 1.66-5.72). Although most patients underwent surgery as scheduled, a small proportion might lose their surgical eligibility due to disease progression, potentially facing toxic effects. Therefore, comprehensive preparation for unforeseen risks during the perioperative period is necessary to maximize patient benefits and ensure treatment safety.

The results of subgroup analysis indicate that there is no significant difference in efficacy among the three immune checkpoint inhibitors (ICIs) in monotherapy. However, regarding different immune combination therapies, the efficacy of dual ICI treatment is superior to monotherapy, while monotherapy is better than immune combination targeted therapy. Conversely, adverse event occurrences present a contrasting scenario where the rate of Grade 3-4 treatment-related adverse events (TRAEs) is lowest in immune combination targeted therapy. Both targeted and immune therapies have gained wide clinical application in HCC, but single therapy may lead to resistance and offer limited clinical benefits. Studies suggest that their combination yields better treatment outcomes. For instance, the IMbrave150 trial (40) demonstrated significant improvements in 1-year survival rates (67.2%) and mPFS (6.8 months) in previously untreated advanced HCC patients receiving atezolizumab combined with bevacizumab, approved by the FDA and CSCO for first-line treatment in advanced HCC. Other combinations such as lenvatinib plus pembrolizumab, sintilimab plus bevacizumab, and camrelizumab plus apatinib have also shown promising results (41–44). Additionally, the anticipated IMbrave050 phase III clinical study released mid-term analysis results at the 2023 AACR conference, indicating a 28% significant reduction in the risk of recurrence, distant metastasis, or death in HCC patients post-radical treatment (surgical resection or ablation) with the immune+anti-angiogenesis combination therapy (T+A regimen) (45), demonstrating clear survival benefits. Mechanistically, studies found that immune-modulating drugs restored the immune-supportive microenvironment, while anti-VEGF drugs like bevacizumab improved immune suppression and aided in restoring vascular normalization for efficient drug delivery, allowing for lower doses of ICI to reduce adverse reactions (46–49). In our subgroup analysis, while the combination of targeted therapy and Immunotherapy did not show a significant superiority, with the short follow-up, long-term survival benefits post-surgery need attention, providing clinical evidence for selecting optimal treatment strategies in the future.

Furthermore, with the widespread application of immunotherapy in HCC, many researchers have observed that patients showing favorable responses can gain long-term survival benefits (50, 51). However, there remain numerous patients who do not respond to immunotherapy, and identifying “high-quality” patients remains a pressing issue. Studies in various solid tumors have shown that the response rate to PD-L1/PD-1 inhibitor treatment correlates with overexpression of PD-L1 in patients’ tumor tissues, serving as a predictive biomarker for immunotherapy sensitivity (52–54). In studies specific to primary liver cancer, such as CheckMate040 and CheckMate459, patients with PD-L1 expression ≥1% and PD-L1 expression<1% exhibited significant differences in median overall survival time (OS) and objective response rates (ORR) (55, 56). However, the KEYNOTE224 study showed that tumor cell PD-L1 expression levels were not associated with treatment response rates (57). Other biomarkers such as gut microbiota, circulating tumor DNA, tumor-infiltrating lymphocytes, among others, have also undergone extensive research in HCC (56, 58–61). However, most biomarkers lack validation from large prospective studies, and their specificity and sensitivity remain unclear. Within this paper, only three articles reported results related to immune biomarkers. Xia, Y. and Ho, W. J (19, 21). discovered that tumor-infiltrating B cells or high expression of DCs serve as immune biomarkers for anti-tumor immunotherapy. Additionally, they found that higher levels of DCs after neoadjuvant treatment corresponded to a lower likelihood of patient recurrence. D’Alessio, A (28). focused on endpoints, suggesting that immune cell infiltration, peripheral cell-free DNA (cfDNA), and gut microbiota composition could serve as predictive factors for the efficacy of Nivo+Ipi anti-tumor therapy. Similar findings have been confirmed in tumors such as melanoma, colon cancer, lung cancer, and others (62–64). Presently, precise and efficient prognostic biomarkers in combination or establish multidimensional predictive models to select populations benefiting most from immune therapy, thus maximizing patients’ survival prospects.

Nevertheless, this systematic review has certain limitations. On one hand, there are discrepancies in the included literature regarding treatment regimens, patient inclusion criteria, pathological and radiological response criteria, leading to study heterogeneity. On the other hand, since more than half of the data originates from conference abstract articles, most of these do not report follow-up survival indicators such as Disease-Free Survival (DFS), Overall Survival (OS), etc. Consequently, this paper cannot analyze whether neoadjuvant treatment contributes to long-term benefits for patients. Moreover, there are only a few articles reporting on biomarkers for the efficacy of neoadjuvant immunotherapy, and no established patterns have emerged yet.

In conclusion, research indicates that neoadjuvant therapy demonstrates both effectiveness and safety in resectable HCC. Regarding its efficacy, we observed that neoadjuvant immunotherapy results in pathological or radiological responses in certain patients, and these individuals seemingly experience better survival outcomes. Concerning its safety, the overall incidence of grade 3-4 treatment-related adverse events (TRAEs) with neoadjuvant immunotherapy is relatively low and does not significantly affect subsequent treatment plans. In summary, neoadjuvant immunotherapy offers benefits in the treatment of resectable liver cancer. However, substantial, high-quality trials in the future remain imperative to offer sturdy data support.

CT: Investigation, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing. YY: Data curation, Funding acquisition, Software, Writing – original draft, Writing – review & editing. YW: Conceptualization, Project administration, Supervision, Writing – review & editing. LY: Conceptualization, Funding acquisition, Methodology, Writing – original draft. CY: Formal analysis, Investigation, Methodology, Writing – original draft. YT: Funding acquisition, Resources, Visualization, Writing – review & editing. GF: Funding acquisition, Resources, Visualization, Writing – review & editing. DZ: Funding acquisition, Resources, Visualization, Writing – review & editing. XW: Funding acquisition, Resources, Visualization, Writing – review & editing.