Abstract
Novel molecular therapies using targeted drugs and immune checkpoint inhibitors for advanced hepatocellular carcinoma have been evolving. Sorafenib and lenvatinib have been commonly used as first-line therapy, followed by recent atezolizumab plus bevacizumab. The median survival time has gradually improved to over 1.5 years. The complete radiological response does not always mean a complete pathological response and a permanent cure of disease. To resolve this, conversion surgery has developed. Lenvatinib is the most suitable drug due to its high response rate. A recent large cohort study using lenvatinib had a conversion rate of 8.4% and an estimated disease-specific survival time of >80% at three years. Conversion to curative resection was an independent predictive factor for better disease-specific survival compared with lenvatinib monotherapy. In conclusion, conversion surgery following molecular therapy is a promising treatment strategy for prolonging long-term outcomes. We should discuss promising drugs and the timing for conversion surgery.
Multidisciplinary treatment is commonly administered to patients with advanced hepatocellular carcinoma (HCC); this involves hepatectomy/transplantation, local ablation therapy, transarterial chemoembolization (TACE), and systemic therapy (1-3). Systemic therapy is mainly recommended for Barcelona Clinic Liver Cancer (BCLC) stage C HCC, but recent evidence has demonstrated that the long-term prognosis of systemic therapy alone is not sufficient (1-3).
In the last 10 years, several novel molecular therapies using molecular targeted drugs and immune checkpoint inhibitors (ICIs) have been introduced (4-18). Sorafenib, a tyrosine kinase inhibitor (TKI), was approved as first-line therapy by the Food and Drug Administration in 2008 (4, 5). Lenvatinib was also approved as a first-line TKI in Japan, and its use spread worldwide in 2018 (6-10). Based on a phase-III REFLECT randomized controlled trial, the overall survival of the lenvatinib-treated group showed non-inferiority to the sorafenib-treated group (8). On the other hand, a recent covariate-adjusted analysis of the REFLECT trial clearly showed better overall survival of the lenvatinib-treated group compared with the sorafenib-treated group (11). In 2020, a pivotal phase III IMbrave 150 study (12) tested head-to-head superiority. Combination therapy of the anti-programmed death ligand 1 inhibitor atezolizumab and the anti-vascular endothelial growth factor drug bevacizumab demonstrated significantly better overall survival than sorafenib monotherapy. Recently, updated data were published in the 2021 ASCO meeting (13). Molecular therapy with ICIs (nivolumab and pembrolizumab) and combination therapy with ICIs and TKIs could possibly achieve better prognosis when given as first-line therapy (14-18). Other molecular drugs, such as regorafenib, ramucirumab, and cabozantinib, were introduced as options for second- or more-line molecular therapy and beyond (15-18).
In fact, the median survival time after molecular therapy for unresectable HCCs has gradually improved, specifically being 6.5-13.4 months for sorafenib, 13.6 months for lenvatinib, and 19.2 months for atezolizumab + bevacizumab (13, 17). Nevertheless, complete remission of HCC has rarely been reported by molecular therapy alone, despite multi-drug therapy. Several long-term HCC survivors have been reported in patients treated mainly with sorafenib (19-21) or lenvatinib (22, 23) followed by conversion surgery. Conversion is defined as the status that downstages unresectable HCC to resectable HCC. It also includes the status that unresectable extrahepatic metastases become resectable or stable.
This article summarizes the current status of conversion surgery for advanced HCC after molecular therapies focusing on long-term survival and disease cure.
Radiological Complete Response (CR) and Long-term Prognosis in Patients With HCC Treated With Molecular Therapy
After induction of molecular therapy, a small number of patients were considered to have a CR according to diagnostic imaging (24-27). Based on both the response evaluation criteria in solid tumor (RECIST) and modified RECIST (mRECIST) criteria, a CR status should be maintained for at least 4 weeks. However, not all patients achieve a complete pathological response or complete remission of HCC. When considering the utility of conversion surgery, it is important to determine the long-term prognosis of patients defined as radiological CR when surgical intervention is not performed (24-27).
Unresectable HCCs treated with first-line TKI therapy can have extremely low CR rates (0%-1%) according to the conventional RECIST criteria (15-18). Surprisingly, a study on atezolizumab + bevacizumab showed a CR rate of 7.7% (13). Conversely, CR rates assessed using the mRECIST criteria ranged from 0.4% to 2.5% for sorafenib, 2%-11% for lenvatinib, 4% for nivolumab, 12% for atezolizumab + bevacizumab, and 11% for lenvatinib + pembrolizumab (13-17). More recently, extremely high CR rates assessed by mRECIST were demonstrated in some combinations of molecular therapies: 8.2% for nivolumab + ipilimumab, 16.7% for nivolumab + cabozantinib, and 25.7% for nivolumab + cabozantinib ± ipilimumab (17, 18).
A report has summarized CR cases according to mRECIST, including 18 patients with HCC treated with sorafenib (24). Among 3,047 patients treated at multiple institutions belonging to the Liver Cancer Study Group of Japan, the CR rate was 0.6%, while the median time to CR assessment was 119 days (range=35-447 days). The significant factors predictive of CR included female sex, low body weight (≤59 kg), early tumor-node-metastasis stage (III), and a small initial dose of sorafenib (≤600 mg). Furthermore, specific adverse events, including palmar– plantar erythrodysesthesia syndrome, hypertension, diarrhea, alopecia, fatigue, nausea, and anorexia were more frequently observed in CR patients compared with non-CR patients. Another study reported seven patients (1.3%) with HCC considered to have CR by mRECIST after sorafenib therapy (25). All patients experienced normalization of the serum alpha-fetoprotein, whereas pathologic CR was never confirmed in any patient, including tumor biopsy. The median time to CR was 3 months (range=2-6 months), while the median disease-free survival was 9 months (range=3-43 months). However, recurrence at the target lesion was not noted in any of the patients. In another multicenter study (26), the CR rate according to mRECIST was 1.1% (12/1,119) after sorafenib therapy. The median time to CR was 13.3 months (range=0.9-33.3 months). The median overall survival was quite excellent (85.8 months), and six patients survived without recurrence. The treatment duration was 40.1 months, and all patients except one developed early dermatologic reactions. As for lenvatinib or other molecular therapies, summarized data have not yet been published.
Based on these studies (24-27), radiologic CR achieved by molecular therapy for advanced HCC is possible, but very rare. Furthermore, CR does not always mean complete remission of HCC. It was speculated that long-term CR may result from a sorafenib-mediated specific immune/inflammatory modulation (26). Unfortunately, none of the aforementioned studies have provided descriptive information about conversion surgery.
Impact of Conversion Surgery After Molecular Therapy on Advanced HCC
Conversion surgery following molecular therapy has become more prevalent in the clinical setting (28, 29). Conversion surgery for unresectable HCC initially developed as “downstaging resection” after interventional radiology, including TACE and hepatic arterial infusion chemotherapy (HAIC) (30). It can provide excellent long-term prognosis; radiologic PR is essential for such patients, but CR is not always required (31, 32).
Recent evidence has demonstrated that sorafenib and lenvatinib would be pivotal drugs for advanced HCC to increase the potential of conversion surgery (19-23, 33-49). Differences in the indications for both have been reported, depending on the individual patient background or HCC-related factors, including age, etiology, liver function, alpha-fetoprotein, and others (41). In this section, we summarize the current evidence regarding conversion surgery for advanced HCC.
Conversion Surgery With Sorafenib
A relatively small number of patients with HCC have reported undergoing conversion surgery following sorafenib therapy (19-21). The response rates after sorafenib monotherapy are quite low, and thus combination therapy of sorafenib with TACE or HAIC is often adopted (42-44). Curtit et al. (45) reported the first patient who achieved pathological CR (pCR) with sorafenib therapy followed by liver resection. In our previous cohort study (20), five (15.6%) patients with advanced HCC received conversion surgery following sorafenib therapy. The conversion rate was so high partially because of follow-up mainly by a hepatobiliary surgeon. Conversion surgery was one of independent prognostic factors [hazard ratio (HR)=0.07; 95% confidence interval (CI)=0.003-0.40, p=0.001] for patients taking sorafenib. We have experienced four actual 5-year survivors, and three of them are alive and disease-free, treatment-free, and have a good quality of life. We have previously reported a surprising case of initial unresectable HCC with hepatic vein tumor thrombosis proceeding to the inferior vena cava (IVC-HVTT), treated with multidisciplinary treatment including sorafenib and major hepatectomy (Figure 1) (46). Sorafenib was administered for 3 and 6 months pre- and post-hepatectomy, respectively. The patient is alive for more than 10 years without any recurrences. The STORM study failed to show the survival benefits of sorafenib as an adjuvant therapy after curative treatment for HCC (47). However, the STORM study did not include patients with apparent vascular invasion. Patients with HCC with IVC-HVTT have very poor prognosis (5-year survival <10%) despite treatment with hepatectomy (48, 49).
Contrast-enhanced computed tomography findings. (Coronal images of the venous phase). (A) On admission; (B) Before conversion hepatectomy. Reproduced from Nakamura et al. (46).
Imura et al. (21). described eight patients undergoing conversion hepatectomy after sorafenib therapy, and pCR was achieved in only three patients. All patients were alive until 6 to 54 months after the conversion surgery. One patient had a main tumor determined as the pCR and satellite nodules positive with fibroblast growth factor 4 (FGF4) for the remaining viable tumor cells. The patient was alive with no recurrence 4.5 years after surgery. A growth inhibitory assay using a FGF3/4-amplified cancer cell line clearly presented hypersensitivity to sorafenib, and it was valuable against exogenously expressed FGF4 tumors (50).
Conversion Surgery With Lenvatinib
Lenvatinib can be a suitable TKI for conversion surgery due to its high tumor regression and tumor necrosis effect (7-9). A phase-III REFLECT randomized control trial showed that lenvatinib had a higher overall response rate than sorafenib (19% vs. 7% according to RECIST 1.1 and 41% vs. 12% according to mRECIST criteria) (8). The high response rates of lenvatinib were confirmed for patients with HCC beyond the criteria of the REFLECT trial, including patients with main portal vein tumor thrombus, with >50% area of liver occupation, and classified as Child–Pugh B (51, 52). The greater response rate by conversion chemotherapy can provide more chances of liver resection for patients with initially unresectable colorectal liver metastases (53). Similarly, the greater response rate by molecular therapy can bring higher potential of conversion hepatectomy for patients with initially unresectable HCC.
The number of patients with advanced HCC undergoing radical surgery has been increasing after multidisciplinary treatment including lenvatinib. Case reports reporting a total of 12 patients have already been published (Table I) (22, 23, 33-39). Among 10 patients treated with conversion surgery, seven, two, and one patient underwent hepatectomy, microwave ablation, and removal of portal vein tumor thrombus, respectively. The remaining two patients had initially resectable HCC and underwent hepatectomy after neoadjuvant molecular therapy (33, 37). There were eight males and four females with a median age of 69.5 years (range=58-82 years). The etiologies involved hepatitis B (2 cases), hepatitis C (3 cases), alcoholic hepatitis (1 case), and non-B, non-C hepatitis (5 cases) including steatosis and steatohepatitis. The median of the largest tumor size was 10 (range=5.2-24) cm, and the median tumor number was 2.3 (range=1-many). Distant metastasis was observed in three patients. Portal venous invasion as defined by Japanese clinical practice guidelines (54) was encountered in four out of 12 (25%) patients. Combination with TACE was performed for six and two patients in the pre-lenvatinib and post-lenvatinib periods, respectively. Four and eight patients were classified as BCLC-B and BCLC-C, respectively. Immediately before surgery, the Child–Pugh score was 5 in all, except for two patients. The modified albumin-bilirubin grade (55) was 1 for five patients, 2a for four patients, and 2b for one patient. The median duration of preoperative lenvatinib treatment was 5.5 months (range=1.2-12 months). The tumor response rates were 100% according to the mRECIST (including 2 CRs) and 72.7% according to the RECIST (No CR) criteria. Pathological CR was identified in four out of 10 patients for the main tumor and in one patient for the portal vein tumor thrombus. Adjuvant therapy with lenvatinib was performed for three patients. The median survival after starting preoperative therapy and after conversion surgery was 13 months (range=9-34 months) and 9 months (range=3-25 months), respectively. All patients, except one, are alive.
Patients with advanced hepatocellular carcinoma undergoing radical surgery after multidisciplinary treatment including lenvatinib.
We also experienced a very impressive patient with two large HCCs, 24 cm in maximum diameter, with an excessive arterioportal shunt and impaired liver function (Child–Pugh score: 8 points) (22, 29) (Figures 2 and 3). Multiple TACE in combination with lenvatinib therapy was quite effective and resulted in tumor regression, complete disappearance of the major arterioportal shunt, and marked improvement in liver function. Extended right hepatectomy was successfully achieved, and the patient is alive without any recurrence 25 months after the conversion surgery.
Contrast-enhanced computed tomography findings. (Coronal images of the portal phase) (A) On admission; (B) Before conversion hepatectomy. Reproduced from Sato et al. (22).
Digital subtraction angiography. Digital subtraction angiography on the 1st (A) and 4th (B) transarterial chemoembolization. The arterioportal shunt (arrows) almost disappeared after lenvatinib therapy. Reproduced from Sato et al. (22).
Recently, Shindoh J et al. (40) summarized the prognostic impact of conversion surgery for patients with advanced HCC following lenvatinib therapy. Among 107 consecutive patients, 16 were treated with surgical intervention, including nine with curative liver resection. Therefore, the conversion rate for curative resection was 8.4%. When limited to the nine patients with curative resection, the response rates were 33% and 89% according to the RECIST and mRECIST criteria, respectively. There were no patients with PD assessed according to both criteria, and 3 CRs (33%) were recognized by the mRECIST criteria. Successful conversion to curative resection was independently associated with better disease-specific survival (HR=0.04, 95%CI=0.01-0.30; p=0.002) and longer time to treatment failure (HR=0.04, 95%CI=0.01-0.29; p=0.002) compared with no additional treatment. The estimated median disease-specific survival time of patients undergoing curative resection, non-curative resection (i.e., salvage and palliative resection), no treatment, and additional treatment including TACE and HAIC was over 80% at 3 years, 8.9 months, 11.1 months, and 19.3 months, respectively. Additional treatments other than curative resection showed less or marginal survival benefit. A decrease in plasma des-gamma-carboxyprothrombin levels compared with baseline levels was the only independent predictor (odds ratio=22.22, 95%CI=3.42-144.29; p=0.001) for curative resection after lenvatinib therapy. Based on their data, surgical intervention is strongly recommended at the timing of curative hepatectomy, if possible.
It may be difficult to produce solid evidence for conversion surgery. To conduct a randomized control trial, patients should be assigned to either conversion surgery or continuous molecular therapy when determined to be resectable. However, this is problematic from an ethical point of view. In fact, one patient was judged to be resectable after lenvatinib therapy, but he refused surgery and unfortunately experienced progressive disease after 4 months (56).
A multicenter, single-arm observation study of surgical resection following lenvatinib therapy for initially unresectable HCC is now ongoing. The study has been registered in the Japan Registry of Clinical Trials. These results can hopefully resolve questions regarding the efficacy of conversion surgery after multidisciplinary treatment, including lenvatinib.
Conversion Surgery With Atezolizumab + Bevacizumab
More recently, the first report of conversion hepatectomy for HCC was published after 15 cycles of atezolizumab plus bevacizumab (57). He presented the best tumor response of partial response by RECIST criteria, but unfortunately after that showed progressive disease. However, after the right hepatectomy, he showed disease-free survival for 19 months. Matching tumor tissues were investigated using pretreatment and post-progression samples with immunohistochemistry and gene expression analysis. As a result, high expression of PD-L1 (CD274) and T effector signature in the pretreatment sample and decreased expression in the disease progression sample were confirmed. Increased expression of PD-L1 and T effector signature may predict an improved outcome in patients with advanced HCC treated with atezolizumab plus bevacizumab.
Cessation Interval of Molecular Therapy Before Conversion Surgery
The adequate interval of TKI cessation until conversion surgery is still being debated. The half-life of the plasma concentrations is relatively short, approximately 25-48 h for sorafenib and 28-35 h for lenvatinib (58, 59). Unfortunately, the half-life has not been clarified in patients with deteriorated liver function. The Cancer Chemotherapy Manual recommends the appropriate interruption of sorafenib therapy for one week before hepatectomy (60). Therefore, approximately one week should also be a sufficient cessation time for lenvatinib therapy. Other TKIs would also need this similar interval.
A recently recommended first-line molecular therapy consists of atezolizumab plus bevacizumab. Bevacizumab can provide suppressive effects on wound-healing and liver regeneration (61, 62). Therefore, the half-life of bevacizumab is about three weeks, and thus 6-9 weeks of cessation is required (63). In contrast, the half-life of atezolizumab is 27 days from population pharmacokinetic analysis (64) and atezolizumab is not known to have such adverse events. Atezolizumab is effective without bevacizumab, thus atezolizumab monotherapy is convenient during the cessation of bevacizumab (65).
Conclusion and Future Perspectives
For patients with advanced HCC, the CR rates after molecular therapy are still low, and patients judged with CR do not always show complete remission of the disease. At present, we strongly recommend timely conversion surgery when radical resection is possible, but the best timing for the surgery is still being debated. Being a responder is desirable, but CR status is not always necessary for surgery. Recently, an innovative treatment for advanced HCC using ICIs and TKIs has been reported (12). Thus, new molecular therapies and treatment strategies with higher response rates tailored to the oncologic characteristics of each advanced HCC are required to increase the possibility of conversion surgery and to improve long-term outcomes.
Footnotes
Authors’ Contributions
KY and TB designed and drafted the manuscript. TM, HO, HN, KI, HH, and SA collected data and assisted in preparing the manuscript. All Authors read and approved the final manuscript.
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
- Received October 16, 2021.
- Revision received November 11, 2021.
- Accepted November 12, 2021.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.