Durable Stable Disease by Atezolizumab/Bevacizumab Can Provide Long-term Survival of Patients With Hepatocellular Carcinoma Lung Metastases

Abstract

Background

Multiple bilateral lung metastases secondary to hepatocellular carcinoma (HCC) are mainly treated with molecular therapy. Atezolizumab plus bevacizumab can provide excellent long-term survival for patients with a good response.

Case Report

A 67-year-old woman underwent right hepatectomy for a primary solitary HCC, 11 cm in diameter, after portal embolization. After 2 years, she developed bilateral lung metastases with >100 nodules, <1 cm in size. She had no viral hepatitis or liver cirrhosis, and the Child–Pugh Grade was A (5 points). Lenvatinib (12 mg daily) was administered as a first-line treatment and continued for 18 months. The best response was stable disease (SD). Subsequently, intravenous atezolizumab (1,200 mg) plus bevacizumab (15 mg/kg) was administered once every three weeks. The best response was SD, which continued for 26 months. After that, cabozantinib treatment was initiated and discontinued after one cycle. Subsequently, dual immune checkpoint inhibitor treatment (durvalumab + tremelimumab) was administered. She has had multiple, but lung-only, metastases over four years. She has been well as an outpatient with the Child–Pugh Grade of A and a performance status of 0.

Conclusion

Even if atezolizumab plus bevacizumab does not induce a good response, a durable SD could prolong survival in patients with metastatic HCC while maintaining liver function and a good quality-of-life.

Recurrences after liver resection for hepatocellular carcinoma (HCC) are frequently observed; approximately 80% of recurrent tumors occur within five years, with most appearing within 2 years (1,2). Among the extrahepatic recurrences, lung metastases (LM) are commonly observed; however, LMs alone are relatively rare (3). According to the US National Cancer Institute’s Surveillance, Epidemiology, and End Results Registry (SEER), the incidence of LM from HCC did not increase with the increase in HCC incidence from 2010 to 2016 (4). Before induction of effective systemic therapy, the median survival time (MST) of patients with HCC and LM is reportedly 487 days after HCC diagnosis and 179 days after LM diagnosis (5). Even after administration of molecular targeted drugs, the 1- and 3-year survival rates of HCC with LM are 13.6% and 4.9%, respectively (6).

The treatment of primary HCCs varies and includes liver resection or liver transplantation, local ablation, chemo-embolization, and systemic therapy (7,8). In addition, surgical resection, local ablation, and radiotherapy are effectively used for oligometastatic lung cancer secondary to HCC (6). However, treatment options for bilateral multiple LMs are limited (3,4). Currently, molecular therapy is the first-line treatment in such situations according to Western and Japanese guidelines (7,8). Molecular therapy consists of the lone or combination use of molecular targeted drugs and immune checkpoint inhibitors (ICIs) (9). Currently, atezolizumab with bevacizumab (ATEZO/BEV), lenvatinib, and sorafenib are commonly used as first-line treatment for metastatic HCC.

An updated IMbrave150 randomized controlled trial comparing ATEZO/BEV and sorafenib showed promising results (10). In the ATEZO/BEV-treated group, after a median follow-up of 15.6 months, the MST was 19.2 months. In patients with extrahepatic spread at the time of study enrollment, the hazard ratio (HR) was excellent at 0.60. The treatment duration was 8.4 and 7.0 months for atezolizumab and bevacizumab, respectively, and the response rate was 30% according to the Response Evaluation Criteria in Solid Tumors (RECIST) criteria. The rate of treatment-related Grade 3/4 adverse events was 43%. Recently, it was reported that an excellent response to molecular therapy is a valuable predictor of prolonged survival (11). Moreover, among the non-responders to ATEZO/BEV therapy, those with durable stable disease (SD), which was defined as SD sustained for >6 months, had excellent prognosis (12).

Herein, we present a patient in whom HCC recurred with numerous bilateral LMs after liver resection, who was successfully treated with ATEZO/BEV therapy and showed durable SD and a prolonged survival.

Case Report

A 67-year-old woman was diagnosed with bilateral LMs secondary to HCC 2 years after portal embolization and right hepatectomy. The primary tumor was a large (11-cm diameter), simple, nodular HCC (Figure 1). Histo-pathologically, the tumor showed moderately differentiated HCC with a trabecular growth-pattern and no invasion of the portal and hepatic veins. There was no evidence of fibrosis, steatosis, or steatohepatitis in the surrounding liver. She had undergone pneumonectomy for primary lung cancer (pT1, N0, M0 adenocarcinoma) 12 years prior to the hepatectomy. Over 100 nodules <1 cm in size were observed in both lungs; however, no other metastases, including liver tumors, were observed (Figure 2A). She had no history of hepatitis B, hepatitis C, or liver cirrhosis, and had a Child–Pugh Grade A (5 points) liver disease. She had a history of hypertension and diabetes mellitus and was classified as obese because of a high body mass index of 32.9. Her serum levels of alpha-fetoprotein (AFP), protein induced by the absence of vitamin K or antagonist-II (PIVKA-II), and AFP-L3 were 4.4 ng/ml (normal ≤10 ng/ml), 4418 mAU/ml (normal ≤40 mAU/ml), and 0.5% (standard ≤10%), respectively. Her C-reactive protein was 0.07 mg/dl, and her neutrophil-to-lymphocyte ratio was 5.06.

Figure 1. Contrast-enhanced abdominal computed tomography before hepatectomy was performed. Axial arterial (A) and equilibrium (B) phase images showing the typical hypervascular hepatocellular carcinoma.

Figure 2. Changes observed on axial sections of the plain chest computed tomography. Changes observed pre-lenvatinib (A1, A2), pre- atezolizumab plus bevacizumab (ATEZO/BEV) (B1, B2), and 4 (C1, C2), 7 (D1, D2), 16 (E1, E2), and 26 (F1, F2) months after ATEZO/BEV administration. The number of lung metastases did not increase during the observation period. In addition, there was a mixture of shrinking and enlarging nodules. M: Months.

She was diagnosed to have unresectable LMs secondary to HCC. Lenvatinib (12 mg daily) was initiated as first-line therapy, which was continued for 18 months. She showed Grade 2 adverse events of hand-foot syndrome and proteinuria, so she sometimes skipped the lenvatinib treatment. Chest computed tomography (CT) revealed an SD status; the LMs gradually grew (Figure 2B). Therefore, intravenous ATEZO/BEVA therapy was started (1,200 mg atezolizumab plus 15 mg/kg bevacizumab once every 3 weeks). The best response, SD, continued for 26 months (37 cycles). Thus, she was classified as having durable SD (Figure 2C-E). After the 4th cycle of ATEZO/BEVA treatment, she showed Grade 2 proteinuria, thus atezolizumab administration without bevacizumab was performed once. After that the dose of BEV was decreased to 10 mg/kg. After the 20^th ATEZO/BEVA treatment cycle, she showed Grade 1 transient hemoptysis and Grade 3 proteinuria, and BEV was skipped for three cycles. Treatment course and changes in the PIVKA-II levels were presented (Figure 3). The PIVKA-II levels well reflected the tumor response for LM. Because the LMs showed progressive disease (PD) status (Figure 2F), she was administered cabozantinib. However, cabozantinib was stopped after one cycle because of massive proteinuria (13). Subsequently, dual ICI therapy of durvalumab (anti-programmed death-ligand 1 antibody) and tremelimumab (anti- cytotoxic t-lymphocyte-associated antigen 44 antibody) was initiated (14).

Figure 3. Treatment course and changes in the tumor marker levels. PIVKA-II: Protein induced by the absence of vitamin K or antagonist-II; ATEZO/BEV: atezolizumab plus bevacizumab.

She has had multiple LMs alone over 4 years and has been doing well with a Child–Pugh Grade of A (5 points) and performance status of 0. She was admitted to the hospital for only ten days for induction of lenvatinib during the observation period after diagnosis of LM.

Discussion

For HCCs sized ≥10 cm, anterior hepatectomy using the hanging maneuver technique reportedly decreases the recurrence rate, including that of extrahepatic metastases, and improves overall survival (15). However, at a regular follow-up 2 years after curative right hepatectomy, our patient was diagnosed with numerous bilateral LMs. Histopathological examination of the primary resected tumor revealed moderately differentiated HCC without vascular invasion. PIVKA-II levels rapidly decreased after hepatectomy and rapidly increased at the time of LM diagnosis. Subsequently, the change in its concentration reflected the LM response to molecular therapy. No metastases except for LM were found over the observation periods.

The prognosis of LM secondary to HCC is very poor; a recent large population-based study using the SEER database determined that the 1-year overall survival and 1-year cancer-specific survival were 10% and 12.6%, respectively (3). Multivariate analysis demonstrated certain significant prognostic factors, including liver-directed surgery, chemotherapy, and radiotherapy (4). For patients with oligometastatic lung cancer secondary to HCC, surgical resection, local ablation, and radiotherapy are sometimes effectively used and can provide a long-term prognosis (3-6). However, few long-term survivors have been reported after cytotoxic chemotherapy (5). Moreover, a few patients with spontaneous regression were reported, mainly because of the insufficient blood flow to the tumor (16).

Recently, many studies have been published about good mid-term prognosis and prognostic factors for ATEZO/BEV therapy in a clinical setting (Table I) (10,12,17-26). Various independent factors have been reported, including preoperative liver function, cancer-relative factors, inflammation factors, and their integrated scores. She had a low CRP and AFP in CRAFITY score for immunotherapy that predicted excellent progression-free survival and overall survival (25,26); in contrast, she showed a high neutrophile-to-lymphocyte ratio associated with a worse prognosis (5,11).

Table I. Prognosis and prognostic factors for atezolizumab plus bevacizumab therapy in the recent era.

CP: Child-Pugh; ECOG-PS: Eastern Cooperative Oncology Group Performance Status; BCLC: Barcelona Clinic of Liver Cancer; PVTT: portal vein tumor thrombus; EHS: extrahepatic spread; ORR: objective response rate; SD: stable disease; OS: overall survival; PFS: progression-free survival; HR: hazard ratio; RECIST: response evaluation criteria in solid tumors; mRECIST: modified response evaluation criteria in solid tumors; ALBI: albumin-bilirubin grade; mALBI: modified albumin-bilirubin index; PIVKA-II: protein induced by vitamin K absence or antagonist-II; AFP: alpha-fetoprotein; CRP: C-reactive protein; NLR: neutrophile-to-lymphocyte ratio; CRAFITY score: CRP and AFP in immunotherapy score; irAE: immune-related adverse event; Gr: grade according to the National Cancer Institute Common Terminology Criteria for Adverse Events; aHR; adjusted hazard ratio; PSM: propensity score matching; ND: not determined.

Our patient had multiple LMs and was treated with lenvatinib for 16 months and ATEZO/BEV for 26 months. The best response of both therapies was SD. More recently, good response to ATEZO/BEV was identified as a predictor for prolonged survival; MST: 36 months or more, 16.1 months, and 8.0 months for complete or partial response, SD, and PD, respectively (12). Responders to ATEZO/BEVA survive longer than non-responders; however, it was recently reported that patients with durable SD also showed an excellent prognosis (MST, 23.7 months) (12). Our patient was treated with ATEZO/BEVA therapy for >2 years because of the long SD status with no obvious adverse events except for Grade 3 proteinuria and transient hematemesis. She exhibited PD status for 4 months; the metastases growth was slow, with no evidence of new lung or extrapulmonary metastases, including liver lesions. The synergistic anti-angiogenic effect of BEVA and ATEZO/BEVA-induced tumor immunity activation contribute to the treatment’s durable activity (27). Furthermore, the period until the deterioration of the quality-of-life is significantly longer with ATEZO/BEVA administration than with sorafenib administration (28). In addition, ATEZO/BEVA prolongs progression-free survival, maintains hepatic reserve, and lowers the treatment discontinuation rate due to severe adverse events when compared with lenvatinib (29). Lately it has been reported that Grade 1/2 irAEs are not associated with the objective response but with prolonged survival for patients with advanced HCC treated with ATEZO/BEVA (20). Further, it has been reported that molecular therapy did not improve survival of HCC patients with non-viral hepatitis, including non-alcoholic steatohepatitis (30). Our patient had no irAEs or viral hepatitis; however, she showed durable SD and long-term survival. Cabozantinib was administered as the third-line therapy in our patient; however, it was discontinued because of the severe adverse events (13). She was started on dual ICI therapy consisting of durvalumab (anti-PD-L1 antibody) and tremelimumab (anti-CTLA-4 antibody) (14). Dual ICI with a primary priming effect by tremelimumab may provide a durable response.

In conclusion, a short survival was expected in our patients because of the multiple bilateral LMs and poor response to molecular therapy. However, durable SD with ATEZO/BEVA provided prolonged survival with good PS. However, further study is required to estimate the universal advantages of durable SD in ATEZO/BEVA treatment.

Conflicts of Interest

The Authors declare no conflicts of interest in relation to this study.

Authors’ Contributions

MT and TB developed the concept and drafted the manuscript; HM, SU, HT, YN, SC, EO, RK, NO, and TI treated the patient and collected data; TM, KM, and HO revised the manuscript.

Acknowledgements

The Authors thank the patient, their colleagues, and the Yamaga City Medical Center staff for their kind cooperation.