Lenvatinib Might Induce Activation of Host Immunity in Patients with Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) creates a complex immunosuppressive network that evades antitumor immunity by secreting immunosuppressive cytokines that affect immune and stromal cells [[1]. In a phase 3, open-label, non-inferiority trial (REFLECT) in patients with unresectable HCC, lenvatinib (an anti-vascular endothelial growth factor [anti-VEGF] multitargeted tyrosine kinase inhibitor) was shown to be non-inferior to the standard-of-care treatment, sorafenib (an anti-VEGF multikinase inhibitor), for overall survival. In addition, the study showed that the median overall survival was longer with lenvatinib (13.6 months) than with sorafenib (12.3 months), although this difference was not statistically significant [[2]. Nevertheless, lenvatinib was later recommended as second-line systemic chemotherapy for unresectable HCC. A 2019 study on the combination systemic chemotherapy of atezolizumab, an immune checkpoint inhibitor, plus bevacizumab, a monoclonal antibody that binds vascular endothelial growth factor (VEGF), found better overall and progression-free survival outcomes than with sorafenib in patients with unresectable HCC [[3]. Later, at the 2020 consensus conference, the American Association for the Study of Liver Disease recommended this chemotherapy combination as a new first-line treatment for advanced HCC [[4].

Previously, we reported that in patients with liver cirrhosis and advanced HCC, sorafenib might prevent tumor cells from evading the host immune system by inducing T helper (Th) 1 dominance [[5]. Although lenvatinib was recommended only as a second-line treatment in patients receiving systemic chemotherapy for unresectable HCC, with the advent of atezolizumab, it became necessary to investigate the effect of lenvatinib on host immunity further. Furthermore, a phase 1b study on lenvatinib plus pembrolizumab – a monoclonal antibody inhibitor of the programmed cell death protein I receptor – in patients with unresectable HCC found promising antitumor effects of the combination therapy [[6]. Because this therapy may become a first-line approach in the future and because lenvatinib is approved for use in HCC and may therefore be used in patients who also received the immune checkpoint inhibitor atezolizumab, the present study evaluated the effect on host immunity of lenvatinib therapy in patients with chronic liver disease (CLD) and unresectable HCC.

Because the number of unresectable HCC cases is small and valuable, we collected a limited number of cases over 2 years and conducted a prospective study. We studied adult patients with CLD and unresectable HCC who received lenvatinib therapy between 2019 and 2021 at Toho University Hospital, Tokyo, Japan. Lenvatinib was administered for 4 weeks. The dose of lenvatinib was based on weight: 8 mg if bodyweight was less than 60 kg and 12 mg/day if bodyweight was 60 kg or more. Early morning blood samples were collected at baseline before the start of treatment and after 4 weeks. Potential complications of lenvatinib treatment were explained to each patient, and written informed consent was obtained.

The modified method of Jung et al. [[7] was used to analyze subsets of peripheral blood CD4-positive T cells following nonspecific stimulation with phorbol 12-myristate 13-acetate, ionomycin, or brefeldin A (Sigma Chemical Co., St. Louis, MO, USA). Cytoplasmic expression of IFN-gamma and IL-4 by peripheral blood CD4-positive T cells after culture and staining was detected using flow cytometry, as reported previously [[5]. Next, we divided the CD4-positive T cell population into IFN-gamma-positive/IL-4-negative (Th1) cells and IFN-gamma-negative/IL-4-positive (Th2) cells. Finally, regulatory T cells (Treg cells) were identified as CD25^high/CD127^low cells [[5].

We used the commercially available enzyme immunoassay (Quantikine; R&D Systems, Inc., Minneapolis, USA) to measure soluble TNF-alpha (sTNF-alpha) in duplicate. This enzyme immunoassay has a sensitivity of 2.6 pg/mL, an inter-assay coefficient of variation of +16.7%, and an intra-assay coefficient of variation of +8.8%.

Serum sTNF-alpha receptor I (sTNFr-I) levels were determined using a commercially available enzyme-linked immunosorbent assay system (Quantikine; R&D Systems, Inc.). Blood samples were analyzed within 1 week of collection, and serum was stored at −80°C until use.

The modified Response Evaluation Criteria in Solid Tumors (RECIST) [[8, 9] was used to evaluate tumor response. mRECIST evaluates only the contrast-enhanced portion of the liver lesion during the arterial phase of a dynamic imaging technique rather than the entire lesion, when evaluating target lesions, and defines progression and response in line with RECIST.

Statistical analyses were performed with the statistical package SPSS, version 11.0 (SPSS, Chicago, IL, USA). Results are expressed as the mean ± standard deviation (SD). Within-group patient characteristics were compared using the Wilcoxon signed-rank test for each group. In all analyses, p < 0.05 was considered to indicate statistical significance.

This work was supported by the School of Medicine, Toho University, Tokyo, Japan. This study was approved by the Medical Ethics Committee of Toho University Omori Medical Center (number M18127) and was conducted according to the Declaration of Helsinki and current legal regulations in Japan. Written informed consent was obtained from all patients.

Forty-three patients were enrolled in the present study. The size and clinical characteristics are shown in Table 1. In CLD etiology, 11 patients had hepatitis B virus (HBV group), 10 patients had hepatitis C virus (HCV group), 7 patients had alcohol abusers (alcohol group), and 15 patients had non-alcohol/non-HBV-/non-HCV (non-ABC group). Liver cirrhosis was more common than chronic hepatitis. Child-Pugh class A was the most common prognosis, followed by class B, and no patient had a prognosis in Child-Pugh class C. Stage IVA tumors were the most common stage, and portal vein tumor thrombus (PVTT) grade Vp3 (i.e., PVTT in the first-order branches of the portal vein) was the most common. The higher lenvatinib dose (12 mg/day) was given to more patients than the lower dose (8 mg/day). None of the patients showed CR. Twenty-one of 43 patients (48.8%) achieved a partial response (PR), but 4 patients (9.3%) showed progressive disease (PD), and 18 patients (41.9%) had stable disease (SD).

The following changes were seen after lenvatinib treatment compared to baseline: the serum ammonia and total bilirubin increased significantly. The serum level of albumin, the neutrophil count, the lymphocyte count, and the ratio of neutrophil/lymphocyte, platelet count, alpha-fetoprotein, and des-gamma carboxyprothrombin decreased significantly (Table 2).

We found significant increase of Th1 cells following lenvatinib treatment (24.83 ± 8.7% vs. 27.63 ± 8.5%; p = 0.0027 by Wilcoxon signed-rank test), although there were no significant changes in Th2 (4.25 ± 2.1% vs. 4.20 ± 1.8%) or Treg cells (8.20 ± 3.0% vs. 7.97 ± 2.6%) (Fig. 1). Moreover, serum levels of TNF-alpha (1.82 ± 1.1 ng/mL vs. 1.99 ± 0.9 pg/mL; p = 0.0033 by Wilcoxon signed-rank test), sTNF-R (2,978.64 ± 1,267.2 ng/mL vs. 3,234.55 ± 1,144.3 pg/mL; p = 0.0169), and VEGF (641.33 ± 597.3 pg/mL vs. 764.79 ± 956.5 pg/mL; p = 0.0115) increased significantly after lenvatinib treatment (Fig. 2).

Patients were divided into two groups according to tumor response. Twenty-one patients showed partial response (PR group), and 22 patients had stable disease or progressive disease (SD + PD group). There were significantly more PVTT patients in the PR group than in the SD + PD group. Moreover, there was a significant difference between both groups regarding the lenvatinib dose (Table 3).

The following changes were seen after lenvatinib treatment compared to baseline: serum ammonia levels increased significantly in the SD + PD group, although there was no significant difference in the PR group. In the SD + PD group, there was a significant increase in serum levels of AST and ALT and a significant decrease in albumin. However, serum levels of total bilirubin were increased in both groups. Neutrophils and neutrophil/lymphocyte ratio increased significantly in the PR group, while platelets decreased significantly in both groups. Regarding serum levels of tumor markers, AFP decreased significantly in the PR group, while AFP, AFP-L3, and DCP increased significantly in the SD + PD group (Table 4).

When we examined changes in host immunity according to tumor response, we found a significant increase in Th1 cells in patients with a PR following 4 weeks of lenvatinib treatment but no significant differences in Th2 cells, Treg cells, or serum levels of TNF-alpha and sTNF-R (Table 5).

The present study evaluated the effects of lenvatinib on host immunity in patients with CLD and unresectable HCC. They were seen increase in Th1 cells and increase in serum levels of TNF-alpha, sTNF-R, and VEGF in HCC patients treated with lenvatinib. We previously reported that treatment with sorafenib caused a decrease in peripheral blood Th2 cells and Treg cells in patients with CLD and advanced HCC, a change that might induce Th1 dominance [[5]. We also reported that sorafenib might prevent evasion of the host immune system by tumor cells in patients with advanced HCC and HCV-related CLD; however, this did not appear to occur in patients with CLD of other etiologies [[10]. The VEGF family includes VEGFA, VEGFB, VEGFC, VEGFD, VEGFE, and placental growth factor with three receptors: VEGFR1, VEGFR2, and VEGFR3 with associated co-receptors neutrophillin 1 and 2 [[11]. Lenvatinib targets fibroblast growth factor, stem cell factor, VEGFA, and VEGFC, reducing angiogenesis and lymph angiogenesis [[12]. VEGFR3, KIT, RET, and platelet-derived growth factor receptor (PDGF) alpha are also lenvatinib targets [[13]. In the present study, serum levels of VEGF increased significantly after lenvatinib treatment. It was reported that VEGFR2 expression and its downstream RAS/MEK/ERK signaling were up-regulated in lenvatinib-resistant HCC cells, whereas the expression of VEGFR1, VEGFR3, FGFR1-4, and PDGFR alpha/beta showed no difference [[14]. Therefore, serum levels of VEGFA combined with VEGFR1 or VEGFR2 might increase after lenvatinib treatment, while lenvatinib inhibited expression of VEGFR3. VEGF is assumed to affect antitumor immunity by three mechanisms: (1) anti-VEGF receptor inhibitors facilitate dendritic cell maturation and T cell priming [[15‒17]; (2) anti-VEGF receptor inhibitors normalize tumor vascular structure and increase tumor infiltration of cytotoxic T cells [[16, 18‒21]; and (3) anti-VEGF receptor inhibitors reduce myeloid-derived suppressor cells and Tregs and counteract immunosuppression in the tumor microenvironment [[16, 21‒25]. The TNF receptor superfamily includes the TNF receptor I (TNFr-I) and the TNF receptor II (TNFr-II) [[26, 27], which are expressed on the surface of various cells. However, following cleavage of the extra-cytoplasmic domains or alternative splicing, their soluble forms can be detected in the serum [[28], possibly reflecting TNF system activation [[29]. In the present study, serum levels of TNF-alpha increased after lenvatinib treatment in patients with PR. Hence, our results indicate that the anti-VEGF effects of lenvatinib might be more easily seen in patients who show some treatment response. We have reported that sorafenib treatment increases circulating TNF-alpha or decreases those of soluble Fas, promoting TNF-related or Fas-related apoptosis [[30]. Lenvatinib has been shown to reduce tumor-associated macrophages and increase activated CD8-positive T cells secreting IFN-gamma-positive and granzyme B [[31]. Considering all the above, we hypothesized that lenvatinib might induce activations of Kupffer cells in patients with CLD.

Th1 cells and Th2 cells cross-regulate each other during development. Cytokines released by Th2 cells were found to inhibit antitumor immunity [[32], whereas Th1 cytokines promoted it [[33‒36]. Cancer immunosurveillance mechanisms are assumed to involve 3 impairments of T cells as follows: (1) immune tolerance and impairment of T cell priming; (2) impairment of cytotoxic T cell transport and intratumoral infiltration; and (3) impaired killing function of cancer cells by cytotoxic T cells [[37‒39]. It was reported that lenvatinib increased sensitivity to T cell-mediated killing and offset IFN-gamma-induced T-cell silencing [[40]. Our results showed also that lenvatinib induce Th1 dominance in patients with CLD.

We examined and compared patients according to tumor response. In the present study, serum levels of ammonia increased significantly and liver injury worsened significantly in the SD + PD group, whereas there were no significant differences in the PR group. Furthermore, neutrophils and the neutrophil/lymphocyte ratio increased significantly in the PR group, while platelets decreased significantly in both groups. Additionally, serum levels of tumor markers increased significantly after lenvatinib treatment in the SD + PD group. It has been reported that patients with Child-Pugh class A had deteriorated to Child-Pugh class B or C liver impairment in 23.4% of patients after 4 weeks and 23.7% of patients after 12 weeks of starting lenvatinib treatment [[41]. However, in the present study, liver injury after lenvatinib treatment was observed in the SD + PD group, but not in the PR group. These findings may indicate that liver injury after lenvatinib treatment was induced by tumor progression. The neutrophil/lymphocyte ratio at the commencement of chemotherapy or ICI treatment has been reported to be prognostic [[42, 43]. Regarding lenvatinib treatment for HCC, Tada et al. [[44] also reported that the neutrophil/lymphocyte ratio could be associated with outcomes in patients with HCC treated with lenvatinib. In the present study, we found a significant decrease in neutrophil/lymphocyte ratio and a significant increase in Th1 cells in patients with a PR after lenvatinib treatment. These findings might indicate that lenvatinib promotes cell-mediated immunity, and further investigation is needed.

In conclusion, lenvatinib might induce Th1-dominant host immunity in patients with CLD and unresectable HCC treatment who had showed a PR. These changes in host immunity may be a biomarker in HCC patients treated with lenvatinib.

This work was supported by the School of Medicine, Toho University, Tokyo, Japan. This study was approved by the Medical Ethics Committee of Toho University Omori Medical Center (number M18127) and was conducted according to the Declaration of Helsinki and current legal regulations in Japan. Written informed consent was obtained from all patients.

Hidenari Nagai has received research funding from Eisai, Otsuka Pharma, and AbbVie.

The authors declare no funding regarding the publication of this paper.

Hidenari Nagai and Takanori Mukozu performed all the experiments and the data analysis and finalized the article. Hidenari Nagai drafted the manuscript. Takanori Mukozu, Kojiro Kobayashi, Akira Nogami, Hideki Nagumo, Kunihide Mohri, Go Watanabe, Makoto Amanuma, Naoyuki Yoshimine, Yu Ogino, Daigo Matsui, Yasuko Daido, Yasushi Matsukiyo, Teppei Matsui, Noritaka Wakui, and Koichi Momiyama collected blood samples and performed the data analysis. Hidenari Nagai, Koji Higai, and Takahisa Matsuda reviewed and provided feedback on the manuscript. All authors approved the final version of the manuscript.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.