Immunotherapy versus lenvatinib in hepatocellular carcinoma: the need for granular data and prospective validation

Abstract

Hepatocellular carcinoma (HCC) remains a leading cause of cancer-related mortality worldwide, with systemic therapies constituting the cornerstone of treatment for advanced-stage disease. While lenvatinib has been a widely used first-line agent, the advent of immunotherapy—particularly combinations such as atezolizumab-bevacizumab and the durvalumab plus tremelimumab (STRIDE regimen)—has redefined the therapeutic landscape. Recent retrospective and real-world studies have yielded conflicting results regarding the comparative effectiveness of lenvatinib and immunotherapy. Rimini et al noted a survival benefit for lenvatinib in HCC, though real-world studies meta-analysis showed similar survival outcomes to atezolizumab-bevacizumab, while Ahn et al, via target trial emulation, reported better outcomes with immunotherapy. However, limitations such as lack of key clinical variables, heterogeneous treatment regimens, and residual confounding complicate interpretation. These findings highlight the need for prospective, etiology-stratified research with comprehensive clinical data to optimize first-line therapy selection and inform personalized treatment strategies in advanced HCC.

Hepatocellular carcinoma (HCC)—a primary form of liver cancer—remains the second leading cause of cancer-related mortality worldwide, with incidence continuing to rise. Unfortunately, over 70% of patients are diagnosed at an advanced stage, at which point systemic therapies are the cornerstone of treatment. Among these, lenvatinib has traditionally served as a widely used first-line option.¹ In recent years, the emergence of immunotherapy has notably transformed the treatment landscape, with combinations such as atezolizumab plus bevacizumab and, more recently, the durvalumab plus tremelimumab (STRIDE regimen) demonstrating promising outcomes.² Despite these advances, direct comparative studies between lenvatinib and immunotherapy remain a subject of active clinical and academic interest, particularly as existing evidence presents divergent findings.

A comprehensive search of the PubMed database was conducted from its inception through June 30, 2025. The search strategy included the terms (“immunotherapy” OR “immune checkpoint inhibitor”) AND (“lenvatinib”). In addition, the reference lists of related articles were manually reviewed to identify any other relevant publications. Two independent reviewers screened the titles and abstracts of all retrieved studies. The identified studies were summarized in figure 1 and online supplemental table 1. Overall, the majority of studies suggested that lenvatinib and the atezolizumab–bevacizumab combination demonstrate comparable efficacy. Some studies indicated that the atezolizumab–bevacizumab combination may be preferred in viral HCC, whereas lenvatinib might show superiority in non-viral HCC.

Figure 1. The therapeutic effects of lenvatinib (LEN) and the atezolizumab+bevacizumab (A+B) combination in advanced hepatocellular carcinoma (HCC). The top panel illustrates the underlying mechanisms of action for both treatments. Lenvatinib inhibits angiogenesis and induces cancer cell apoptosis through the RAS/RAF/MEK/ERK and PI3K/AKT/mTOR signaling pathways. In contrast, the A+B combination enhances immune responses by blocking T cell inhibitory signals, enabling activated T cells to attack tumor cells. The left-bottom panel displays tumor response rates (ORR) across multiple studies, with the left Y-axis showing the number of patients, and the right Y-axis showing the ORR percentages for each treatment group (LEN vs A+B). The right-bottom panel provides survival data, including overall survival (OS) and progression-free survival (PFS), with a scatter plot depicting patient outcomes for each treatment modality. The size of the circles indicates the number of patients, and the color denotes the type of treatment (LEN or A+B). AKT, Protein kinase B; ERK, Extracellular signal-regulated kinase; MEK, Mitogen-activated protein kinase kinase; mTOR, Mammalian target of rapamycin; ORR, Objective response rate; PD-1, Programmed cell death protein 1; PD-L1, Programmed death-ligand 1; PI3K, Phosphatidylinositol-3-kinase; RAF, Rapidly accelerated fibrosarcoma; RAS, Rat sarcoma; TCR, T-cell receptor; MHC, Major histocompatibility complex.

A study by Rimini et al contributed valuable early insights by demonstrating a potential survival benefit for lenvatinib over the atezolizumab-bevacizumab combination, particularly in patients with HCC associated with non-alcoholic fatty liver disease (NAFLD) or non-alcoholic steatohepatitis (NASH).³ This retrospective analysis, based on prospectively collected data from patients with non-viral advanced HCC treated across 36 centers in Italy, Japan, South Korea, and the UK, evaluated 190 patients receiving atezolizumab-bevacizumab and 569 patients treated with lenvatinib. In the overall cohort, lenvatinib was associated with significantly improved overall survival (OS) (HR: 0.65; 95% CI: 0.44 to 0.95; p=0.0268) and progression-free survival (PFS) (HR: 0.67; 95% CI: 0.51 to 0.86; p=0.002). Among patients with NAFLD/NASH, multivariate analysis further supported lenvatinib’s association with longer OS (HR: 0.46; 95% CI: 0.26 to 0.84; p=0.0110) and PFS (HR: 0.55; 95% CI: 0.38 to 0.82; p=0.031).

A recent meta-analysis of real-world studies has revealed that for patients with unresectable HCC, lenvatinib monotherapy yields survival outcomes similar to those of the combination therapy of atezolizumab-bevacizumab as first-line treatment.⁴ Specifically, the median OS was 18.4 months in the lenvatinib group compared to 18.5 months in the atezolizumab-bevacizumab group, while the median PFS was 6.9 months in the lenvatinib group versus 7.3 months in the combination therapy group.

In contrast, a more recent study by Ahn et al, titled “Comparative Effectiveness of Immunotherapy versus Lenvatinib in Advanced Hepatocellular Carcinoma: A Real-World Analysis Using Target Trial Emulation,” presents conflicting findings.⁵ Using a real-world dataset and a target trial emulation framework, the authors explored the comparative effectiveness of immunotherapy versus lenvatinib in the first-line treatment of advanced HCC. After propensity score matching, each treatment group comprised 1,203 patients. The study found that immunotherapy was associated with improved OS (median 545 vs 425 days; HR: 0.86; 95% CI: 0.76 to 0.97). Subgroup analyses suggested that atezolizumab-bevacizumab outperformed lenvatinib (HR: 0.87; 95% CI: 0.77 to 0.99), with survival benefits observed in patients with viral hepatitis (HR: 0.74; 95% CI: 0.61 to 0.89) and alcoholic liver disease (HR: 0.65; 95% CI: 0.49 to 0.87). No statistically significant difference was observed among patients with metabolic dysfunction-associated steatotic liver disease (HR: 0.96; 95% CI: 0.70 to 1.31). These findings contribute meaningfully to ongoing discussions regarding optimal systemic therapies for advanced HCC and support the growing role of immunotherapy in this setting.

While I commend the Ahn et al authors for their intent and analytical rigor, several considerations may help contextualize the findings. First, clinical detail and potential confounding: the absence of key variables—such as Barcelona Clinic Liver Cancer (BCLC) stage, Child-Pugh class, and Eastern Cooperative Oncology Group (ECOG) performance status—limits clinical interpretation.^{6 7} These are essential to both prognosis and treatment decision-making in HCC. In practice, lenvatinib is often selected for patients with compromised liver function due to its manageable toxicity profile. Conversely, patients receiving immunotherapy have fewer contraindications, including portal hypertension or bleeding risk. Additionally, comorbidities such as diabetes, cardiovascular conditions, and prior variceal bleeding are not well characterized. These elements can influence both treatment decisions and outcomes.⁸ Although the authors apply propensity score methods, residual confounding is a concern, especially when surrogate markers (eg, bilirubin, cirrhosis diagnosis) may not fully capture disease severity. Second, subgroup analyses: findings such as the reduced HR in Asian patients are noteworthy. However, without formal interaction testing or discussion of potential biological and sociodemographic influences, such subgroup results are best interpreted as exploratory. Third, follow-up duration: the reported difference in median survival (545 vs 425 days) is clinically relevant, but the relatively short median follow-up (~250 days) and the convergence of survival curves around 600 days raise concerns about long-term durability. Data on post-progression therapy or survival could offer additional context. Fourth, generalizability and practical considerations: while the TriNetX network provides valuable real-world insights, further clarity on factors such as treatment access (eg, academic vs community settings, insurance status) and adverse event profiles would strengthen the study’s clinical applicability. Fifth, treatment heterogeneity: grouping distinct regimens—such as atezolizumab–bevacizumab and STRIDE—into a single “immunotherapy” category may introduce significant heterogeneity. These regimens differ in mechanisms, toxicity, and indications.⁹ While a sensitivity analysis stratified by regimen is included, more detail on its findings and statistical power would enhance interpretability. Sixth, endpoints beyond OS: OS is a key endpoint, but inclusion of metrics like PFS, response rates, or time to treatment failure would provide a more nuanced view of therapeutic impact.¹⁰

In conclusion, the recent study conducted by Ahn et al offers significant exploratory insights into the comparative efficacy of systemic therapies for advanced HCC. However, its methodological limitations—especially concerning unmeasured confounding—limit the strength of its conclusions. Further prospective, etiology-stratified studies with granular clinical data are needed to inform personalized treatment strategies in advanced HCC.