https://orcid.org/0000-0002-9116-7054
Figures
Abstract
Background and Aim
The optimal second-line therapy following first-line atezolizumab plus bevacizumab remains unascertained. In this study, we compared second-line durvalumab plus tremelimumab with lenvatinib after first-line atezolizumab plus bevacizumab in patients with unresectable hepatocellular carcinoma (uHCC).
Methods
In this retrospective, open-label, non-randomized comparative study, we analyzed real-world data derived from a prospectively registered observational cohort of patients with uHCC who received durvalumab plus tremelimumab (the Dur/Tre group, n = 14) or lenvatinib (the Len group, n = 67) as second-line therapy after first-line atezolizumab plus bevacizumab. Tumor response was assessed using the RECIST criteria version 1.1. Progression-free survival (PFS), overall survival (OS), adverse events (AEs), and changes in the albumin–bilirubin (ALBI) score were compared between the groups.
Results
The objective response and disease control rates were 7.7% and 15.4% in the Dur/Tre group and 23.3% and 76.7% in the Len group, respectively. The median PFS was 1.7 vs. 4.2 months (p < 0.001) and median OS was 5.3 vs. 14.0 months (p = 0.047) in the Dur/Tre and Len groups, both significantly favoring lenvatinib. Multivariable analysis showed that lenvatinib treatment was an independent predictor of longer PFS, and a neutrophil-to-lymphocyte ratio ≥ 3 was an independent predictor of worse OS. Grade ≥ 3 AEs were more frequent with lenvatinib than with durvalumab plus tremelimumab (70.1% vs. 21.4%; p = 0.002). At week 4, the ALBI score was maintained with durvalumab plus tremelimumab (−2.30 to −2.21; p = 0.318) but worsened with lenvatinib (−2.36 to −1.97; p < 0.001).
Conclusions
After first-line atezolizumab plus bevacizumab, second-line lenvatinib achieved superior disease control and longer survival than durvalumab plus tremelimumab. However, grade ≥ 3 AEs were more frequent with lenvatinib than with durvalumab plus tremelimumab. Careful AE management is therefore important when selecting therapy for individual patients.
Citation: Nishioka C, Tahata Y, Maesaka K, Kai M, Shirai K, Murai K, et al. (2026) Comparison of outcomes of second-line durvalumab plus tremelimumab versus lenvatinib following first-line atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma. PLoS One 21(5): e0341395. https://doi.org/10.1371/journal.pone.0341395
Editor: Keun-Yeong Jeong, PearlsInMires, KOREA, REPUBLIC OF
Received: January 29, 2026; Accepted: April 21, 2026; Published: May 7, 2026
Copyright: © 2026 Nishioka et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: Hayato Hikita received research grants from Chugai Pharmaceutical Co., Ltd., Eisai, MSD K.K., and AstraZeneca, and received lecture fees from Chugai Pharmaceutical Co., Ltd. Yasutoshi Nozaki received lecture fees from Chugai Pharmaceutical Co., Ltd., Eisai, and AstraZeneca. Tasuku Nakabori received lecture fees from Chugai Pharmaceutical Co., Ltd. and Eisai. Naruyasu Kakita received lecture fees from Chugai Pharmaceutical Co., Ltd. Atsushi Hosui received lecture fees from Chugai Pharmaceutical Co., Ltd. and Eisai. Kengo Matsumoto received lecture fees from Chugai Pharmaceutical Co., Ltd. and AstraZeneca. Tetsuo Takehara received lecture fees from Chugai Pharmaceutical Co., Ltd., Eisai, MSD K.K., and AstraZeneca. Takahiro Kodama received research grants from AstraZeneca and received lecture fees from Chugai Pharmaceutical Co., Ltd., Eisai, MSD K.K., and AstraZeneca. All other authors declare no conflicts of interest.
Abbreviations: AE, adverse event; ALBI, albumin–bilirubin; Atezo/Beva, atezolizumab plus bevacizumab; CI, confidence interval; CR, complete response; CTCAE, Common Terminology Criteria for Adverse Events; DCR, disease control rate; Dur/Tre, durvalumab plus tremelimumab; HAIC, hepatic arterial infusion chemotherapy; HCC, hepatocellular carcinoma; ICI, immune checkpoint inhibitor; IQR, interquartile range; irAE, immune-related adverse event; Len, lenvatinib; ORR, objective response rate; OS, overall survival; PD, progressive disease; PD-1, programmed death 1; PD-L1, programmed death-ligand 1; PFS, progression-free survival; PR, partial response; RECIST, Response Evaluation Criteria in Solid Tumors; RT, radiation therapy; SD, stable disease; TACE, transcatheter arterial chemoembolization; VEGF, vascular endothelial growth factor
Introduction
Systemic therapy for unresectable hepatocellular carcinoma (HCC) commenced with sorafenib, which was approved as a first-line standard of care in 2009 [1,2]. Subsequently, based on the REFLECT trial, lenvatinib was approved for HCC treatment in 2018 [3]. In 2020, the IMbrave150 trial established atezolizumab plus bevacizumab as the initial first-line regimen to incorporate an immune checkpoint inhibitor (ICI) [4]. Thereafter, in 2022, following the HIMALAYA trial, durvalumab plus tremelimumab—the first dual-immunotherapy regimen—was approved for HCC, which further expanded therapeutic options [5]. Contemporary guidelines recommend atezolizumab plus bevacizumab or durvalumab plus tremelimumab as first-line therapy for patients who are eligible for ICI therapy [6,7]. Recently, in 2025, the combination of nivolumab plus ipilimumab demonstrated a significant overall survival (OS) benefit, compared with lenvatinib or sorafenib [8]. Thus, further improvements in outcomes for patients with unresectable HCC are anticipated.
Atezolizumab plus bevacizumab is an established first-line regimen for unresectable HCC [9], with an objective response rate (ORR) of 22.0–28.2% and a disease control rate (DCR) of 69.6–70.6% [10,11]. However, most patients with HCC experience disease progression during treatment with atezolizumab plus bevacizumab and transition to subsequent therapy, and the optimal sequence of systemic treatments thereafter remains unclear [12]. In this regard, real-world studies, including a recent comparison of first-line lenvatinib and second-line lenvatinib after atezolizumab plus bevacizumab, have reported stable tumor responses and favorable survival outcomes with second-line lenvatinib after atezolizumab plus bevacizumab [13–16]. Therefore, lenvatinib may be a useful clinical option after treatment with atezolizumab plus bevacizumab. Additionally, recent real-world studies have suggested the potential utility of durvalumab plus tremelimumab as second-line therapy after first-line atezolizumab plus bevacizumab. In these reports, the DCR ranged from 54.5% to 62.5% [17,18], and anti-CTLA-4-containing therapy was suggested to effectively induce T-cell priming, regardless of prior exposure to atezolizumab plus bevacizumab [17]. Furthermore, after the initiation of durvalumab plus tremelimumab, no significant worsening of the ALBI score was observed at 6 months, at the end of durvalumab plus tremelimumab treatment, or at the end of observation [18]. These findings suggest that durvalumab plus tremelimumab may be a useful treatment option after atezolizumab plus bevacizumab.
In Japan, multiple ICI-based regimens have been approved for HCC, and both durvalumab plus tremelimumab and lenvatinib are recommended as second-line treatment options after first-line atezolizumab plus bevacizumab [19]. Therefore, a comparison of durvalumab plus tremelimumab and lenvatinib as second-line therapy after first-line atezolizumab plus bevacizumab is clinically important; however, direct head-to-head real-world evidence in this setting remains scarce. In this study, we evaluated the efficacy and safety of second-line durvalumab plus tremelimumab after first-line atezolizumab plus bevacizumab and compared these results with those of second-line lenvatinib after first-line atezolizumab plus bevacizumab.
Methods
Study design
This multicenter, registry-based, retrospective, open-label, non-randomized comparative study was performed at The University of Osaka Hospital and 18 affiliated institutions of the Osaka Liver Forum. The study used real-world data derived from a prospectively registered observational cohort of patients who received systemic therapy for HCC in routine clinical practice at the participating centers. No formal a priori sample size calculation was performed because this was an exploratory study based on real-world data. All eligible patients identified at the participating centers during the predefined study periods were included in the analysis. We compared the outcomes in patients who received durvalumab plus tremelimumab as second-line therapy after first-line atezolizumab plus bevacizumab (the Dur/Tre group) with those of patients who received lenvatinib as second-line therapy after first-line atezolizumab plus bevacizumab (the Len group).
The study protocol was approved by The University of Osaka Hospital Ethical Review Board and the ethics committees of all participating institutions (UMIN000034611). Written informed consent was obtained from all participants at each participating institution. The data were accessed for research purposes between November 16, 2024, and September 3, 2025. After data collection, the authors did not have access to information that could identify individual participants.
Study population and enrollment period
In the Dur/Tre group, first-line atezolizumab plus bevacizumab treatment was initiated between October 2020 and January 2024, and second-line durvalumab plus tremelimumab was started between April 2023 and June 2024. In routine clinical practice, the decision to initiate second-line durvalumab plus tremelimumab was made after the attending physicians comprehensively assessed the general condition, comorbidities, and prior tolerance to atezolizumab plus bevacizumab for each patient, and determined that the patient would be able to tolerate ICI rechallenge. The Len group included 67 patients who received lenvatinib following first-line atezolizumab plus bevacizumab between October 2020 and January 2023, as reported previously [13]. This cohort was derived from the period before durvalumab plus tremelimumab became available and served as a historical control. Lenvatinib was administered when the attending physicians deemed that the patient would be likely to tolerate lenvatinib after atezolizumab plus bevacizumab. Treatment-selection bias between lenvatinib and durvalumab plus tremelimumab after atezolizumab plus bevacizumab was considered to be minimized by using this historical cohort, because all patients in the Len group had been treated before durvalumab plus tremelimumab treatment became available. Accordingly, patients with unresectable HCC who received durvalumab plus tremelimumab or lenvatinib as second-line treatment following first-line atezolizumab plus bevacizumab were selected.
The inclusion criteria for this study were as follows: (1) patients with unresectable HCC who received durvalumab plus tremelimumab or lenvatinib as second-line therapy after first-line atezolizumab plus bevacizumab at hospitals participating in the Osaka Liver Forum; (2) patients with an Eastern Cooperative Oncology Group performance status of 0 or 1; and (3) patients who provided written informed consent for participation in the prospective observational cohort.
The exclusion criteria were as follows: (1) inability to receive contrast-enhanced imaging for tumor response evaluation; (2) initiation of second-line durvalumab plus tremelimumab or lenvatinib >3 months after discontinuation of atezolizumab plus bevacizumab; (3) participation in a clinical trial; and (4) observation period of < 4 weeks.
The final analytic cohort comprised 14 patients in the Dur/Tre group and 67 patients in the Len group after application of the predefined eligibility and exclusion criteria.
Lenvatinib administration
Lenvatinib was administered orally once per day. According to the approved dosage in Japan [3,20,21], the initial dose was determined by body weight: 12 and 8 mg/day for patients ≥ 60 and < 60 kg, respectively. Dose reductions or interruptions were allowed based on the severity of adverse events (AEs), in accordance with the prescribing recommendations.
Durvalumab plus tremelimumab administration
Durvalumab plus tremelimumab was administered according to the previously described clinical trial protocol [5] as follows: 300 mg tremelimumab and 1500 mg durvalumab administered on day 1, followed by 1500 mg durvalumab every 4 weeks. Dose interruptions were permitted based on the severity of AEs, in accordance with the prescribing recommendations.
Outcome measures
The primary outcomes of the study were progression-free survival (PFS) and OS. The secondary outcomes were ORR, DCR, the incidence of treatment-related AEs, changes in the albumin–bilirubin (ALBI) score, and subsequent therapy. ORR was defined as the proportion of evaluable patients who achieved complete response (CR) or partial response (PR) as their best overall response. DCR was defined as the proportion of evaluable patients who achieved CR, PR, or stable disease (SD) as their best overall response.
Treatment evaluation
Based on the contrast-enhanced computed tomography or magnetic resonance imaging findings, the tumor response was assessed using the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. CR was defined as the disappearance of all target lesions, and PR as at least a 30% decrease in the sum of the diameters of target lesions from baseline. Progressive disease (PD) was defined as at least a 20% increase in the sum of the diameters of target lesions relative to the smallest sum recorded during treatment, with an absolute increase of at least 5 mm, or the appearance of one or more new lesions or unequivocal progression of non-target lesions. SD was defined as neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify for PD [22]. Imaging evaluations were performed at baseline, weeks 4 and 8 after treatment initiation, and every 8 weeks thereafter. AEs were evaluated and graded according to the Common Terminology Criteria for Adverse Events (CTCAE), version 5.0. After disease progression, switching to another treatment regimen was allowed at the discretion of the attending physicians.
Statistical analysis
Clinical parameters are presented as counts and percentages for categorical variables, and as medians and ranges for continuous variables. Categorical variables were compared using the chi-square test or Fisher’s exact test, and continuous variables were compared with the Mann–Whitney U test. The PFS and OS were estimated using the Kaplan–Meier method, and differences between groups were assessed using the log-rank test. Independent prognostic factors associated with PFS and OS were identified using the Cox proportional hazards model. Based on previous studies [23–27], the following covariates were prespecified for inclusion in the multivariable analysis: treatment regimen, age, sex, etiology, Barcelona Clinic Liver Cancer stage, modified ALBI grades, serum α-fetoprotein level, des-γ-carboxy prothrombin (DCP) level, neutrophil-to-lymphocyte ratio (NLR), duration of prior atezolizumab plus bevacizumab and best response to prior atezolizumab plus bevacizumab treatment. Changes in the ALBI score within each group were compared using the Wilcoxon signed-rank test. p < 0.05 was considered statistically significant. All statistical analyses were conducted using SPSS statistical software (version 29.0) for Windows (IBM Corp., Armonk, NY, USA).
Results
Baseline characteristics of patients treated with second-line durvalumab plus tremelimumab versus lenvatinib after first-line atezolizumab plus bevacizumab
Initially, 35 patients who received durvalumab plus tremelimumab immediately after atezolizumab plus bevacizumab were screened for inclusion in the Dur/Tre group. After applying the exclusion criteria, 14 patients who had received first-line atezolizumab plus bevacizumab followed by second-line durvalumab plus tremelimumab were included in the analysis. The Len group comprised 67 patients who received second-line lenvatinib after first-line atezolizumab plus bevacizumab (S1 Fig). The median age was 73 and 75 years (p = 0.797), and male accounted for 64.3% and 83.6% of the patients in the Dur/Tre and Len groups, respectively (p = 0.137). Child–Pugh class A was observed in 85.7% and 86.6% of the patients in the Dur/Tre and Len groups, respectively (p = 1.000). In the Dur/Tre and Len groups, 64.3% and 68.7% of the patients exhibited five or more intrahepatic tumors (p = 0.760), 35.7% and 31.3% had macrovascular invasion (p = 0.760), and 42.9% and 47.8% had extrahepatic metastasis (p = 0.777), respectively. First-line atezolizumab plus bevacizumab had been administered for ≥ 3 months in 57.1% and 61.2% of the patients (p = 0.773). Additionally, a total of 64.3% and 53.7% of patients in the Dur/Tre and Len groups, respectively, had a best response of CR, PR, or SD to prior atezolizumab plus bevacizumab treatment (p = 0.562). All patients in the Dur/Tre group discontinued prior atezolizumab plus bevacizumab because of disease progression, whereas 89.6% patients in the Len group discontinued prior atezolizumab plus bevacizumab because of disease progression and 10.4% because of adverse events (p = 0.345). Overall, no significant between-group differences were observed in baseline characteristics (Table 1).
The median observation periods in the Dur/Tre and Len groups were 5.6 months (1.9–15.0 months) and 8.6 months (1.5–23.9 months), respectively.
Comparative efficacy and safety of second-line durvalumab plus tremelimumab versus lenvatinib after first-line atezolizumab plus bevacizumab
According to the best therapeutic response assessed using RECIST version 1.1, 0 (0%), 1 (7.7%), 1 (7.7%), and 11 (84.6%) patients in the Dur/Tre group exhibited a CR, PR, SD, and PD, respectively, while the response was not evaluable (NE) in 1 patient, yielding an ORR of 7.7% and a DCR of 15.4%. In the Len group, 1 (1.7%), 13 (21.7%), 32 (53.3%), and 14 (23.3%) patients showed a CR, PR, SD, and PD, respectively, while 7 had NE responses, yielding an ORR of 23.3% and a DCR of 76.7%. The ORR did not differ significantly between the two groups (p = 0.279), whereas the DCR was significantly higher in the Len group than in the Dur/Tre group (p < 0.001) (Table 2).
The median PFS was 1.7 (95% CI, 1.6–1.8) and 4.2 (95% CI, 2.9–5.6) months in the Dur/Tre and Len groups, respectively, and favored lenvatinib (p < 0.001). The median OS was 5.3 (95% CI, 0.3–10.4) and 14.0 (95% CI, 9.8–18.2) months in the Dur/Tre and Len groups, respectively, favoring lenvatinib (p = 0.047; Fig 1A and 1B).
Abbreviations: PFS, progression-free survival; OS, overall survival; Dur/Tre, durvalumab plus tremelimumab; Len, lenvatinib.
The incidence of any-grade and grade ≥ 3 AEs was significantly higher in the Len group than in the Dur/Tre group (any grade: 98.5% vs. 78.6%, p = 0.015; grade ≥ 3: 70.1% vs. 21.4%, p = 0.002) (Table 3). Among any-grade AEs, fatigue (56.7% vs. 7.1%, p < 0.001), proteinuria (55.2% vs. 7.1%, p < 0.001), hypertension (47.8% vs. 0%, p < 0.001), decreased appetite (52.2% vs. 7.1%, p = 0.002), thrombocytopenia (25.4% vs. 0%, p = 0.034) and hand-foot skin reaction (29.9% vs. 0%, p = 0.017) were significantly more common in the Len group than in the Dur/Tre group. Similarly, among AEs with grade ≥ 3, proteinuria was significantly more frequent in the Len group than in the Dur/Tre group (29.9% vs. 0%, p = 0.017). Treatment discontinuation due to AEs occurred in 2 patients (14.3%) in the Dur/Tre group and 18 patients (26.9%) in the Len group, with no significant difference between the two groups (p = 0.499). In the Dur/Tre group, systemic corticosteroids were administered for immune-related adverse events (irAEs) in 2 patients (14.3%)—rash in 1 (7.1%) and colitis in 1 (7.1%).
Clinical factors associated with PFS and OS in the overall cohort
The pretreatment clinical variables were analyzed to identify risk factors associated with shorter PFS and OS in the overall cohort. In the multivariable analysis including treatment regimen, only lenvatinib treatment (HR 0.271, 95% CI 0.129–0.567; p < 0.001) was identified as an independent factor associated with prolonged PFS (Table 4). In the multivariable analysis using the same covariates as that for PFS, only elevated NLR (≥ 3; HR 2.394, 95% CI 1.067–5.375; p = 0.034) was independently associated with OS (Table 5). Median OS was shorter in patients with NLR ≥ 3 than in those with NLR < 3 (8.6 vs. 20.6 months; S2 Fig).
Changes in the ALBI score
The median ALBI score in the Dur/Tre group was −2.30 (IQR, −2.51 to −2.00) at treatment initiation and −2.21 (IQR, −2.50 to −1.92) at week 4, without significant deterioration (p = 0.318). In the Len group, the median ALBI score was −2.36 (IQR, −2.62 to −2.07) at initiation and worsened to −1.97 (IQR, −2.27 to −1.61) at week 4 (p < 0.001; Fig 2). Two patients in the Len group were excluded from this analysis because their serum albumin and total bilirubin levels at week 4 after treatment initiation were not measured.
Two patients in the Len group were excluded because week 4 albumin and total bilirubin data were unavailable. Abbreviations: ALBI, albumin–bilirubin; Dur/Tre, durvalumab plus tremelimumab; Len, lenvatinib..
Subsequent therapy
In the Dur/Tre group, after the discontinuation of durvalumab plus tremelimumab, 10 of the 14 patients (71.4%) received subsequent therapy, including systemic treatment in 9 patients (64.3%) and transcatheter arterial chemoembolization (TACE) in 1 patient (7.1%). The remaining 4 (28.6%) patients received best supportive care. In the Len group, after discontinuation of lenvatinib, 48 of the 66 patients (72.7%) received subsequent therapy, including systemic treatment in 29 patients (43.9%), systemic treatment plus TACE in 3 patients (4.5%), systemic treatment plus hepatic arterial infusion chemotherapy (HAIC) in 1 patient (1.5%), TACE in 5 patients (7.6%), TACE plus continuation beyond progressive disease (PD) in 1 patient (1.5%), transcatheter arterial infusion/HAIC in 2 patients (3.0%), radiation therapy (RT) in 5 patients (7.6%), RT plus continuation beyond PD in 1 patient (1.5%), and continuation beyond PD alone in 1 patient (1.5%). The remaining 18 (27.3%) patients received best supportive care. In addition, 1 patient remained on lenvatinib at the data-cut-off timepoint (S1 Table). No significant between-group difference was observed in the proportion of patients that received subsequent therapy (p = 1.000).
Discussion
This multicenter collaborative study of patients with unresectable HCC compared the efficacy and safety of second-line durvalumab plus tremelimumab with lenvatinib following first-line atezolizumab plus bevacizumab. The results showed that the Len group had significantly better PFS, OS (Fig 1A and 1B), and DCR (Table 2) than the Dur/Tre group. However, severe AEs were significantly more frequent in the Len group than in the Dur/Tre group (Table 3). Unlike the Dur/Tre group, the Len group also exhibited a significant worsening of the ALBI score after treatment initiation (Fig 2). To the best of our knowledge, this is the first report of a direct comparison of second-line durvalumab plus tremelimumab and lenvatinib after first-line atezolizumab plus bevacizumab.
On second-line therapy after first-line atezolizumab plus bevacizumab, the Len group showed superior PFS, OS, and DCR compared with the Dur/Tre group. Maesaka et al. recently compared first-line lenvatinib with second-line lenvatinib after atezolizumab plus bevacizumab and reported comparable efficacy between the two groups [13]. Other real-world studies have also reported stable responses and favorable survival with second-line lenvatinib after atezolizumab plus bevacizumab [14–16]; therefore, lenvatinib can be expected to be effective as second-line therapy after atezolizumab plus bevacizumab. Pharmacologically, prolonged programmed death-ligand 1 (PD-L1) blockade on CD8 + T cells induced by atezolizumab plus bevacizumab may potentiate the antitumor activity of subsequent lenvatinib treatment [28,29]. In contrast, several reports suggest reduced effectiveness of durvalumab plus tremelimumab after atezolizumab plus bevacizumab [24,30]. Fujiwara et al. identified an independent association between prior atezolizumab plus bevacizumab and shorter OS (HR 27.38, 95% CI 1.92–391.50; p = 0.01) and PFS (HR 6.25, 95% CI 1.16–33.71; p = 0.03) in multivariable Cox analyses of patients treated with durvalumab plus tremelimumab [24]. Additionally, Mori et al. reported that, compared with first-line durvalumab plus tremelimumab and later-line durvalumab plus tremelimumab without prior atezolizumab plus bevacizumab, later-line durvalumab plus tremelimumab with prior atezolizumab plus bevacizumab showed significantly worse ORR/DCR and PFS rates [30]. The potential mechanisms underlying the attenuated durvalumab plus tremelimumab efficacy after atezolizumab plus bevacizumab include resistance to programmed death 1 (PD-1)/PD-L1 blockade with T-cell exhaustion [31], diminished T-cell priming owing to anti-vascular endothelial growth factor (VEGF) resistance [32], and rebound VEGF-mediated immunosuppression after bevacizumab [33]. However, recent real-world studies have reported DCRs of 54.5%–62.5% for durvalumab plus tremelimumab administered as second-line therapy after first-line atezolizumab plus bevacizumab, suggesting that this regimen may offer a certain degree of clinical benefit [17,18]. Thus, our findings should be interpreted with caution because the number of patients in the Dur/Tre group was limited in the present study. Moreover, research using large prospective datasets under uniform conditions regarding durvalumab plus tremelimumab after atezolizumab plus bevacizumab remains scarce, warranting further investigation.
In terms of hepatic reserve, the Dur/Tre group maintained the ALBI score from treatment initiation to week 4, whereas the Len group exhibited a worsening ALBI score (Fig 2). Miura et al. demonstrated the preservation of liver function before and after durvalumab plus tremelimumab in patients with HCC who were treated with durvalumab plus tremelimumab following atezolizumab plus bevacizumab [18]. Additionally, Tamaki et al. reported that durvalumab plus tremelimumab can maintain liver function throughout treatment even in later-line settings and in patients with impaired baseline function [34]. Accordingly, durvalumab plus tremelimumab is considered likely to preserve liver function. In our cohort, such preservation likely contributed to a transition rate to subsequent therapy that was comparable to that in the Len group (S1 Table). Furthermore, our observations are consistent with those of previous studies that reported an early decline in liver function in a subset of patients treated with lenvatinib [35–37]. Thus, from the standpoint of preserving liver function, durvalumab plus tremelimumab may be considered a reasonable option even after atezolizumab plus bevacizumab.
Compared to the Dur/Tre group, the incidence of AEs was significantly higher in the Len group, wherein 98.5% patients had at least one AE and 70.1% patients had grade ≥ 3 AEs; fatigue, proteinuria, decreased appetite, and hypertension were common (Table 3). Lenvatinib injures the vascular endothelium, glomerular endothelium/podocytes, and the thyroid capillary network via VEGF/VEGF receptor inhibition, thereby leading to hypertension, proteinuria, and hypothyroidism; furthermore, fatigue and anorexia have a multifactorial pathogenesis [38–40]. The AE profile observed in the Len group in this study was broadly consistent with previous reports of patients treated with lenvatinib after atezolizumab plus bevacizumab [27–29]. Although rash and hepatic dysfunction were relatively frequent in the Dur/Tre group, the overall incidence of AEs—any-grade and grade ≥ 3—was lower than that in the Len group. However, the activation of tumor- and self-reactive T cells by durvalumab plus tremelimumab can systemically injure normal tissues and induce T-cell–mediated irAEs, such as dermatitis and autoimmune hepatitis–like inflammation [41,42]. Accordingly, the AE profile in the Dur/Tre group showed a pattern similar to that reported by Miura et al. for durvalumab plus tremelimumab administered after atezolizumab plus bevacizumab. Nevertheless, the frequency of severe AEs with durvalumab plus tremelimumab after atezolizumab plus bevacizumab was low, which supports an acceptable safety profile.
This study has some limitations. First, the sample size was limited, especially in the Dur/Tre group. Although relevant covariates were adjusted for, the imbalance in sample size between the Dur/Tre and Len groups, together with the wide confidence intervals, limits the clinical applicability of the findings. Second, clinical differences in treatment availability and practice patterns between durvalumab plus tremelimumab and lenvatinib cannot be fully excluded. Therefore, the Len group was constituted from a historical cohort that was treated before durvalumab plus tremelimumab became available to mitigate this real-world bias. Nevertheless, as the patients were not prospectively allocated to either regimen, residual selection bias cannot be ruled out.
In conclusion, lenvatinib showed better treatment outcomes than durvalumab plus tremelimumab as second-line therapy after atezolizumab plus bevacizumab; however, severe AEs were more frequent with lenvatinib. Therefore, careful AE management is essential when selecting second-line therapy in patients with unresectable HCC after atezolizumab plus bevacizumab.
Supporting information
S1 Fig. Flowchart of the participant enrollment.
Abbreviations: Dur/Tre, durvalumab plus tremelimumab; Len, lenvatinib; Atezo/Beva, atezolizumab plus bevacizumab; PS, performance status.
https://doi.org/10.1371/journal.pone.0341395.s001
(PPTX)
S2 Fig. Comparisons of OS between patients with NLR ≥ 3 and <3.
Abbreviations: OS, overall survival; NLR, neutrophil-to-lymphocyte ratio.
https://doi.org/10.1371/journal.pone.0341395.s002
(PPTX)
S1 Table. Subsequent Therapy.
Abbreviations: Dur/Tre, durvalumab plus tremelimumab; Len, lenvatinib; TACE, transcatheter arterial chemoembolization; HAIC, hepatic arterial infusion chemotherapy; PD, progressive disease; TAI, transcatheter arterial infusion; RT, radiation therapy.
https://doi.org/10.1371/journal.pone.0341395.s003
(DOCX)
References
- 1. Llovet JM, Ricci S, Mazzaferro V, Hilgard P, Gane E, Blanc J-F, et al. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med. 2008;359(4):378–90. pmid:18650514
- 2. Cheng A-L, Kang Y-K, Chen Z, Tsao C-J, Qin S, Kim JS, et al. Efficacy and safety of sorafenib in patients in the Asia-Pacific region with advanced hepatocellular carcinoma: a phase III randomised, double-blind, placebo-controlled trial. Lancet Oncol. 2009;10(1):25–34. pmid:19095497
- 3. Kudo M, Finn RS, Qin S, Han K-H, Ikeda K, Piscaglia F, et al. Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet. 2018;391(10126):1163–73. pmid:29433850
- 4. Finn RS, Qin S, Ikeda M, Galle PR, Ducreux M, Kim T-Y, et al. Atezolizumab plus Bevacizumab in Unresectable Hepatocellular Carcinoma. N Engl J Med. 2020;382(20):1894–905. pmid:32402160
- 5. Abou-Alfa GK, Lau G, Kudo M, Chan SL, Kelley RK, Furuse J, et al. Tremelimumab plus Durvalumab in Unresectable Hepatocellular Carcinoma. NEJM Evid. 2022;1(8):EVIDoa2100070. pmid:38319892
- 6. Gordan JD, Kennedy EB, Abou-Alfa GK, Beal E, Finn RS, Gade TP, et al. Systemic Therapy for Advanced Hepatocellular Carcinoma: ASCO Guideline Update. J Clin Oncol. 2024;42(15):1830–50. pmid:38502889
- 7. Singal AG, Llovet JM, Yarchoan M, Mehta N, Heimbach JK, Dawson LA, et al. AASLD Practice Guidance on prevention, diagnosis, and treatment of hepatocellular carcinoma. Hepatology. 2023;78(6):1922–65. pmid:37199193
- 8. Yau T, Galle PR, Decaens T, Sangro B, Qin S, da Fonseca LG, et al. Nivolumab plus ipilimumab versus lenvatinib or sorafenib as first-line treatment for unresectable hepatocellular carcinoma (CheckMate 9DW): an open-label, randomised, phase 3 trial. Lancet. 2025;405(10492):1851–64. pmid:40349714
- 9. Kudo M. Sequential Therapy for Hepatocellular Carcinoma after Failure of Atezolizumab plus Bevacizumab Combination Therapy. Liver Cancer. 2021;10(2):85–93. pmid:33977086
- 10. Maesaka K, Hikita H, Kai M, Tahata Y, Shirai K, Doi A, et al. Hypertension as an Adverse Event Potentially Impacting the Therapeutic Efficacy of Atezolizumab Plus Bevacizumab in Patients With Unresectable Hepatocellular Carcinoma. J Gastroenterol Hepatol. 2025;40(10):2580–90. pmid:40922653
- 11. Kai M, Hikita H, Kazuki M, Tahata Y, Shinkai K, Doi A, et al. Clinical factors associated with the therapeutic efficacy of atezolizumab plus bevacizumab in patients with unresectable hepatocellular carcinoma: A multicenter prospective observational study. PLoS One. 2024;19(1):e0294590. pmid:38165900
- 12. Yoo C, Kim JH, Ryu M-H, Park SR, Lee D, Kim KM, et al. Clinical Outcomes with Multikinase Inhibitors after Progression on First-Line Atezolizumab plus Bevacizumab in Patients with Advanced Hepatocellular Carcinoma: A Multinational Multicenter Retrospective Study. Liver Cancer. 2021;10(2):107–14. pmid:33977087
- 13. Maesaka K, Hikita H, Tahata Y, Nishioka C, Kai M, Shirai K, et al. Comparative analysis of lenvatinib use after atezolizumab plus bevacizumab versus lenvatinib as first-line therapy in unresectable hepatocellular carcinoma. J Gastroenterol. 2026;61(1):68–77. pmid:41139139
- 14. Lee C-K, Yoo C, Hong JY, Park SJ, Kim JW, Tai DWM, et al. Real-World Study of Systemic Treatment after First-Line Atezolizumab plus Bevacizumab for Hepatocellular Carcinoma in Asia-Pacific Countries. Liver Cancer. 2024;14(2):127–41. pmid:40255875
- 15. Hiraoka A, Kumada T, Tada T, Hirooka M, Kariyama K, Tani J, et al. Lenvatinib as Second-Line Treatment after Atezolizumab plus Bevacizumab for Unresectable Hepatocellular Carcinoma: Clinical Results Show Importance of Hepatic Reserve Function. Oncology. 2023;101(10):624–33. pmid:37307798
- 16. Yano S, Kawaoka T, Yamasaki S, Johira Y, Kosaka M, Shirane Y, et al. Therapeutic Efficacy and Safety of Lenvatinib after Atezolizumab Plus Bevacizumab for Unresectable Hepatocellular Carcinoma. Cancers (Basel). 2023;15(22):5406. pmid:38001666
- 17. Yonemoto T, Ogasawara S, Kanogawa N, Miwa C, Fujiya M, Tsuchiya T, et al. Evaluating two rechallenge strategies of immune checkpoint inhibitors: Durvalumab plus tremelimumab in advanced hepatocellular carcinoma. Hepatol Res. 2025;55(5):718–29. pmid:40317555
- 18. Miura R, Ono A, Yano S, Amioka K, Naruto K, Yamaoka K, et al. Real-world efficacy and safety of durvalumab-tremelimumab as second-line systemic therapy after atezolizumab-bevacizumab in unresectable hepatocellular carcinoma. Medicine (Baltimore). 2024;103(34):e39289. pmid:39288227
- 19. Ikeda M, Morizane C, Ueno M, Okusaka T, Ishii H, Furuse J. Systemic therapy for hepatocellular carcinoma, from the early to the advanced stage: a Japanese perspective. Jpn J Clin Oncol. 2025;55(5):465–76. pmid:39895083
- 20. Yamashita T, Kudo M, Ikeda K, Izumi N, Tateishi R, Ikeda M, et al. REFLECT-a phase 3 trial comparing efficacy and safety of lenvatinib to sorafenib for the treatment of unresectable hepatocellular carcinoma: an analysis of Japanese subset. J Gastroenterol. 2020;55(1):113–22. pmid:31720835
- 21. Maesaka K, Sakamori R, Yamada R, Urabe A, Tahata Y, Oshita M, et al. Therapeutic efficacy of lenvatinib in hepatocellular carcinoma patients with portal hypertension. Hepatol Res. 2020;50(9):1091–100. pmid:32559342
- 22. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer. 2009;45(2):228–47. pmid:19097774
- 23. Sangro B, Chan SL, Kelley RK, Lau G, Kudo M, Sukeepaisarnjaroen W, et al. Four-year overall survival update from the phase III HIMALAYA study of tremelimumab plus durvalumab in unresectable hepatocellular carcinoma. Ann Oncol. 2024;35(5):448–57. pmid:38382875
- 24. Fujiwara Y, Kuroda H, Abe T, Kakisaka K, Nakaya I, Ito A, et al. Early Clinical Outcomes of Durvalumab Plus Tremelimumab in Unresectable Hepatocellular Carcinoma: A Real-World Comparison with First-Line or Later-Line Treatment. Drugs Real World Outcomes. 2024;11(4):701–10. pmid:39384684
- 25. Kuzuya T, Kawabe N, Muto H, Wada Y, Komura G, Nakano T, et al. Early Changes in Alpha-Fetoprotein and Des-γ-Carboxy Prothrombin Are Useful Predictors of Antitumor Response to Durvalumab Plus Tremelimumab Therapy for Advanced Hepatocellular Carcinoma. Curr Oncol. 2024;31(8):4225–40. pmid:39195298
- 26. Wu M, Fulgenzi CAM, D’Alessio A, Cortellini A, Celsa C, Manfredi GF, et al. Second-line treatment patterns and outcomes in advanced HCC after progression on atezolizumab/bevacizumab. JHEP Rep. 2024;7(2):101232. pmid:39877031
- 27. Chon YE, Kim DY, Kim MN, Kim BK, Kim SU, Park JY, et al. Sorafenib vs. Lenvatinib in advanced hepatocellular carcinoma after atezolizumab/bevacizumab failure: A real-world study. Clin Mol Hepatol. 2024;30(3):345–59. pmid:38468561
- 28. Osa A, Uenami T, Koyama S, Fujimoto K, Okuzaki D, Takimoto T, et al. Clinical implications of monitoring nivolumab immunokinetics in non-small cell lung cancer patients. JCI Insight. 2018;3(19):e59125. pmid:30282824
- 29. Aoki T, Kudo M, Ueshima K, Morita M, Chishina H, Takita M, et al. Exploratory Analysis of Lenvatinib Therapy in Patients with Unresectable Hepatocellular Carcinoma Who Have Failed Prior PD-1/PD-L1 Checkpoint Blockade. Cancers (Basel). 2020;12(10):3048. pmid:33092011
- 30. Mori N, Tamaki N, Takaki S, Tsuji K, Tada T, Nakamura S, et al. Treatment response to durvalumab plus tremelimumab after progression with previous immune checkpoint inhibitor in unresectable hepatocellular carcinoma. Invest New Drugs. 2024;42(5):559–65. pmid:39212893
- 31. Wang Z, Wang Y, Gao P, Ding J. Immune checkpoint inhibitor resistance in hepatocellular carcinoma. Cancer Lett. 2023;555:216038. pmid:36529238
- 32. Hiraoka A, Tada T, Hirooka M, Kariyama K, Tani J, Atsukawa M, et al. Efficacy of durvalumab plus tremelimumab treatment for unresectable hepatocellular carcinoma in immunotherapy era clinical practice. Hepatol Res. 2025;55(3):444–53. pmid:39526824
- 33. Ribatti D. Immunosuppressive effects of vascular endothelial growth factor. Oncol Lett. 2022;24(4):369. pmid:36238855
- 34. Tamaki N, Mori N, Takaki S, Tsuji K, Tada T, Nakamura S, et al. Change in Liver Function in Durvalumab Plus Tremelimumab Treatment for Unresectable Hepatocellular Carcinoma. Anticancer Res. 2024;44(9):3913–8. pmid:39197893
- 35. Hatanaka T, Kakizaki S, Nagashima T, Namikawa M, Ueno T, Tojima H, et al. Liver Function Changes in Patients with Hepatocellular Carcinoma Treated with Lenvatinib: Predictive Factors of Progression to Child-Pugh Class B, the Formation of Ascites and the Candidates for the Post-Progression Treatment. Cancers (Basel). 2020;12(10):2906. pmid:33050527
- 36. Hiraoka A, Kumada T, Atsukawa M, Hirooka M, Tsuji K, Ishikawa T, et al. Early Relative Change in Hepatic Function with Lenvatinib for Unresectable Hepatocellular Carcinoma. Oncology. 2019;97(6):334–40. pmid:31466068
- 37. Maesaka K, Sakamori R, Yamada R, Doi A, Tahata Y, Miyazaki M, et al. Comparison of atezolizumab plus bevacizumab and lenvatinib in terms of efficacy and safety as primary systemic chemotherapy for hepatocellular carcinoma. Hepatol Res. 2022;52(7):630–40. pmid:35417606
- 38. Enokida T, Tahara M. Management of VEGFR-Targeted TKI for Thyroid Cancer. Cancers (Basel). 2021;13(21):5536. pmid:34771698
- 39. Rini BI, Tamaskar I, Shaheen P, Salas R, Garcia J, Wood L, et al. Hypothyroidism in patients with metastatic renal cell carcinoma treated with sunitinib. J Natl Cancer Inst. 2007;99(1):81–3. pmid:17202116
- 40. Takahashi S, Matsumoto K, Ohba K, Nakano Y, Miyazawa Y, Kawaguchi T. The Incidence and Management of Cancer-Related Anorexia During Treatment with Vascular Endothelial Growth Factor Receptor-Tyrosine Kinase Inhibitors. Cancer Manag Res. 2023;15:1033–46. pmid:37771675
- 41. Postow MA, Sidlow R, Hellmann MD. Immune-Related Adverse Events Associated with Immune Checkpoint Blockade. N Engl J Med. 2018;378(2):158–68. pmid:29320654
- 42. Martins F, Sofiya L, Sykiotis GP, Lamine F, Maillard M, Fraga M, et al. Adverse effects of immune-checkpoint inhibitors: epidemiology, management and surveillance. Nat Rev Clin Oncol. 2019;16(9):563–80. pmid:31092901