Amit Mahipal, MBBS

Lewis R. Roberts, MB, ChB, PhD

As reported in The New England Journal of Medicine by Ghassan K. Abou-Alfa, MD, and colleagues, and reviewed in this issue of The ASCO Post, the phase III CELESTIAL trial has demonstrated that cabozantinib (Cabometyx) improved the median overall survival to 10.2 months in comparison to 8 months with placebo in patients with hepatocellular cancer who had disease progression after at least 1 line of systemic therapy.¹ Cabozantinib is an orally available receptor tyrosine kinase inhibitor with activity against vascular endothelial growth factor receptors (VEGFRs), MET, AXL, and other receptor tyrosine kinases involved in tumor development and progression. Cabozantinib inhibits cancer cell survival, invasiveness, metastasis, angiogenesis, and drug resistance.

Closer Look at CELESTIAL Trial

In this double-blind trial, 773 patients were randomly assigned in a 2:1 fashion to receive cabozantinib at 60 mg or placebo at 95 centers in 19 countries in Asia, Europe, and North America between September 2013 and September 2017. To be included in the study, patients must have had a pathologic diagnosis of hepatocellular cancer with Child-Pugh class A liver function (score of 5 or 6 out of 15). Patients had received previous treatment with sorafenib (Nexavar) and experienced disease progression after systemic therapy but could not have received more than two lines of systemic therapy. The primary endpoint of the study was overall survival, and the secondary efficacy endpoints included progression-free survival and response rates. The intention-to-treat population consisted of 707 patients who were randomly assigned by the data cutoff point on January 1, 2017, for the second interim analysis, with 470 patients assigned to the cabozantinib arm and 237 patients to the placebo arm. Interim analyses were planned when 50% and 75% of the expected deaths had occurred. 

At the data cutoff point for the second planned interim analysis, 484 of 707 patients (68%) had died. The median overall survival was 10.2 months with cabozantinib compared with 8.0 months with placebo (hazard ratio [HR] = 0.76, P = .005). Subsequent systemic therapy or local liver-directed therapy was administered in 26% of the patients in the cabozantinib group and 33% of the patients in the placebo group. Estimated 12-month and 24-month survival rates were 46% and 18% for cabozantinib compared with 34% and 13% for placebo. The median progression-free survival was also higher with cabozantinib than with placebo (5.2 vs 1.9 months, HR = 0.44, P < .001). Subgroup analyses of progression-free survival suggested a benefit from cabozantinib across different clinical characteristics and etiologic factors. The objective response rate was 4% for cabozantinib and less than 1% for placebo (P = .009). Disease control was achieved in 64% and 33% of the patients in the cabozantinib and placebo groups, respectively.

Dose reductions occurred in 62% of the patients receiving cabozantinib and 13% of the patients receiving placebo. The median average daily dose was 35.8 mg for cabozantinib, with a median time to first dose reduction of 38 days. Cabozantinib was discontinued in 16% of patients due to drug-related adverse events, including palmar-plantar erythrodysesthesia, fatigue, decreased appetite, diarrhea, and nausea. The most common adverse events with cabozantinib compared with placebo included diarrhea (54% vs 19%), anorexia (48% vs 18%), palmar-plantar erythrodysesthesia (46% vs 5%), fatigue (45% vs 30%), nausea (31% vs 18%), hypertension (29% vs 6%), vomiting (26% vs 12%), aspartate aminotransferase (AST) level elevation (22% vs 11%), and asthenia (22% vs 8%).

Grade 3 or 4 adverse events were reported in 68% and 38% of the patients in the cabozantinib and placebo groups, respectively. Serious adverse events occurred in 50% of the patients receiving cabozantinib and 37% of the patients receiving placebo. Drug-related death occurred in six patients in the cabozantinib group and was attributed to hepatic failure, bronchoesophageal fistula, portal vein thrombosis, upper gastrointestinal hemorrhage, pulmonary embolism, and hepatorenal syndrome. A grade 5 adverse event was reported in one patient in the placebo group due to hepatic failure.

The authors concluded that treatment with cabozantinib resulted in longer overall and progression-free survival than placebo in patients with previously treated advanced hepatocellular carcinoma. Adverse events were consistent with the known safety profile of cabozantinib, and the rate of high-grade adverse events in the cabozantinib group was about twice that observed in the placebo group.

Rapidly Changing Therapeutic Landscape

Based on the results of the phase III CELESTIAL trial, the U.S. Food and Drug Administration (FDA) has accepted a supplemental new drug application for cabozantinib, an orally bioavailable inhibitor of the VEGFRs, MET, AXL, and other receptor tyrosine kinases, for use in previously treated patients with advanced hepatocellular carcinoma. If approved, this will provide another effective option for the second-line treatment of advanced hepatocellular carcinoma. The results of planned exploratory analyses, including biomarker studies, are eagerly anticipated for their potential to help direct therapy to patients most likely to benefit from treatment with cabozantinib.

The landscape of therapy for advanced hepatocellular carcinoma and intermediate-stage disease refractory to locoregional therapy has been rapidly changing over the past 18 months.

— Amit Mahipal, MBBS, and Lewis R. Roberts, MB, ChB, PhD

Tweet this quote

After many years of stability and relative frustration in the field due to a series of negative phase III trials and no new drug approvals, the landscape of therapy for advanced hepatocellular carcinoma and intermediate-stage disease refractory to locoregional therapy has been rapidly changing over the past 18 months, with the approvals of regorafenib (Stivarga) for use in the second line in advanced disease after disease progression on sorafenib and of lenvatinib (Lenvima) in the first line after demonstrated noninferiority to sorafenib in the treatment of unresectable hepatocellular carcinoma. In a somewhat surprising move, the FDA also approved the immune checkpoint inhibitor nivolumab (Opdivo) approximately a year ago for second-line treatment of advanced disease on the basis of impressive phase II results, pending the results of a head-to-head trial of nivolumab vs sorafenib in the first line.

Within this context, the recently reported positive results of the phase III CELESTIAL trial of cabozantinib for previously treated patients with advanced hepatocellular carcinoma as well as the results of the phase III REACH-II trial of ramucirumab (Cyramza), a fully human antiangiogenic monoclonal antibody targeting VEGFR2, as a second-line treatment for patients with advanced disease and elevated serum alpha-fetoprotein levels greater than or equal to 400 ng/mL, presage a high likelihood of potential additional approvals within the next few months.

Does Tissue Remain the Issue in Hepatocellular Cancer?

Combined with recent advances in integrated molecular characterization and subclassification of hepatocellular carcinoma, the availability of these new agents for treatment of advanced disease provides the first real opportunity to develop an evidence base for biomarker-based personalized therapy for hepatocellular carcinoma. To bring this goal to fruition, there is an urgent need for a more consistent and comprehensive policy of tumor biopsy for patients with intermediate- and advanced-stage disease who are potential candidates for systemic therapy.

It is somewhat paradoxical to be making this argument in the field of oncology, where “tissue is the issue” has been considered dogma since the early development of the discipline. The increasing availability of more and more sophisticated methods of -genetic, epigenetic, and proteomic tumor characterization has led to an increasing ability to select specific therapies based on tumor molecular characteristics—this is now standard for breast, colon, and lung cancers and glioblastomas, as well as for many hematologic malignancies.

There is now a clear imperative to modify the practice of noninvasive diagnosis of intermediate- to advanced-stage disease and to move to a consistent practice of biopsy of all of these tumors.

— Amit Mahipal, MBBS, and Lewis R. Roberts, MB, ChB, PhD

Tweet this quote

However, hepatocellular carcinoma represents a special case. Due to the unique biphasic arterial supply of the liver from both the hepatic arteries and the portal vein, draining relatively highly oxygenated blood from the gut to the liver, as well as other highly specific radiologic features of disease developing within the cirrhotic liver, robust criteria have been developed for the noninvasive diagnosis of hepatocellular carcinoma using multiphasic contrast computed tomography and magnetic resonance imaging of the liver. Although the approach of noninvasive radiologic diagnosis is rational for patients with early-stage disease who are candidates for treatment with the highly effective and potentially curative modalities of liver transplantation, surgical resection, and local ablation, it is increasingly recognized that the extension of the practice of noninvasive diagnosis to patients with incurable intermediate- to advanced-stage hepatocellular carcinoma who are candidates for systemic targeted and precision therapies now stands to substantially hamper efforts to associate tumor characteristics with sensitivity or resistance to the now-increasing repertoire of systemic therapies.

From 2007 to 2017, when the only approved systemic therapy was sorafenib, there was perhaps less inclination or urgency to define the most responsive subgroup of patients, as there was evidence that a substantial proportion of patients might experience at least a partial benefit from sorafenib and there was no competing alternative. However, with the recent approval of lenvatinib in the first line as well as the increasing number of potential systemic treatment options in the second line, there is now a clear imperative to modify the practice of noninvasive diagnosis of intermediate- to advanced-stage disease and to move to a consistent practice of biopsy of all of these tumors.

A secondary question that derives from recent molecular analyses that describe substantial intratumoral and extratumoral heterogeneity of hepatocellular carcinomas is a discussion of the most effective means of obtaining sufficient characterization of tumors to allow robust characterization and the most effective personalization of therapy. For example, for patients with multinodular disease, is it necessary to biopsy multiple nodules? And if so, how many nodules are needed to obtain sufficient characterization of tumors to best select therapy?

Need for Clarity in Complex Field of Options

Thus, the new report of efficacy of cabozantinib in hepatocellular carcinoma has importance not only in the narrow sense of providing an additional alternative for second-line therapy, but perhaps more important, in bringing further to the forefront the need for a comprehensive revision of current practice standards for intermediate to advanced disease. It creates a landscape within which we will potentially have at least two options for first-line therapy—sorafenib and lenvatinib—and four options for second-line therapy of patients failing to respond to sorafenib—regorafenib, nivolumab, cabozantinib, and ramucirumab (for patients with high alpha-fetoprotein levels). Developing the clinical infrastructure for determining the optimal therapy for individual patients within this more complex field of options now becomes an important and urgent need. ■

Dr. Mahipal is a medical oncology specialist, Department of Oncology, Mayo Clinic, Rochester, Minnesota; and Dr. Roberts is Associate Professor of Medicine, Department of Gastroenterology, Mayo Clinic, Rochester.

DISCLOSURE: Dr. Mahipal reported no conflicts of interest. Dr. Roberts has received honoraria from Axis for scientific consulting, he is an advisor/consulant (honoraria to institution) with Tavec, Bayer, Redhill, Grail, and Exact Sciences; has received (institution) research funding from Redhill, Grail, Ariad, Bayer, Gilead, Wako, and Exact Sciences; and has patent or intellectual property interest with Mayo Clinic.

REFERENCE

1. Abou‑Alfa GK, Meyer T, Cheng AL, et al: Cabozantinib in patients with advanced and progressing hepatocellular carcinoma. N Engl J Med 379:54-63, 2018.