Regorafenib Survival Benefit in RESORCE Changes HCC Paradigm

Topline results from the phase III RESORCE trial¹ demonstrated that second-line treatment with regorafenib (Stivarga) represented a 38% reduction in the risk of death compared with best supportive care in patients with unresectable hepatocellular carcinoma (HCC). Additionally, findings show that the median overall survival (OS) with regorafenib was 10.6 months and 7.8 months for placebo versus best supportive care (HR, 0.62; 95% CI, 0.50-0.78; P <.001).

The patient population is of those who progressed after receiving sorafenib (Nexavar). Findings from the phase III study will be submitted to the FDA and European Medicines Agency (EMA) for potential approval, according to a statement from the developer of regorafenib, Bayer Pharmaceuticals.

Structurally related to sorafenib, which is the frontline standard of care, regorafenib is a potent inhibitor of kinases involved in angiogenic, oncogenic, and stromal/tumor environment-related signaling cascades, including VEGFR1-3 and TIE2, c-kit, Ret, BRAF, and PDGFR, among others.

The multistep oncogenic transformation underlying HCC involves alterations in several signaling cascades, resulting in molecularly heterogeneous tumors. Presumptive key carcinogenic signaling pathways include the Wnt/β-catenin pathway, chromatin remodeling, oxidative stress, and signaling involving EGF, PDGF, FGF, VEGF, and IGF, and intracellular mediators such as RAS/RAF/MAPK and PI3K/AKT.

The phase III RESORCE study randomized 573 patients with HCC in a 2:1 ratio to receive best supportive care plus either regorafenib (n = 379) or placebo (n = 194). Regorafenib was administered at 160 mg once daily for 3 weeks followed by 1 week without treatment.

The median age of patients was 63 years, with the majority being male (88%). Most patients had tumors that were BCLC stage C (87%). Prior sorafenib was administered for ≥20 days at ≥400 mg/day with documented radiologic progression. The primary endpoint of the study was OS, with secondary outcome measures focused on progression-free survival (PFS), objective response rate (ORR), and safety.

Median PFS was 3.1 months in the regorafenib arm compared with 1.5 months in the placebo group, representing a 54% reduction in the risk of progression or death (HR, 0.46; 95% CI, 0.37-0.56; P <.001). The median time to progression in the regorafenib group was 3.2 versus 1.5 months with placebo (HR, 0.44; 95% CI, 0.036-0.55; P <.001).

The ORR with regorafenib was 10.6% versus 4.1% with placebo (P = .005). When considering stable disease, the overall disease control rate was 65.2% with the multikinase inhibitor versus 36.1% with placebo.

Median duration of treatment was 3.6 months with regorafenib (range, 0.03-29.4) versus 1.9 months with placebo (range, 0.2-27.4). Grade ≥3 adverse events (AEs) were experienced by 79.7% of those treated with regorafenib versus 58.5% of patients in the placebo arm. Dose modifications to alleviate AEs were required for 68.2% of patients in the experimental arm compared with 31.1% of patients treated with placebo.

The most common grade ≥3 AEs with regorafenib versus placebo, respectively, were hypertension (15.2% vs 4.7%), hand-foot skin reaction (12.6% vs 0.5%), fatigue (9.1% vs 4.7%), and diarrhea (3.2% vs 0%). There were more deaths in the placebo arm versus regorafenib within 30 days following the last dose of treatment (13.4% with regorafenib vs 19.7% for placebo).

The availability of a second-line therapy represents the potential to greatly improve outcomes for patients. Prior to this achievement, outcomes in the setting of sorafenib resistance or intolerance were poor, with a median expected OS for the placebo arms of second-line trials in the range of 7 to 8 months.

Second-Line Setbacks Highlight Importance of RESORCE

A recent study evaluating survival among 260 patients with HCC permanently discontinuing sorafenib reported median OS durations of 4.1 months overall, 4.6 months in patients with progression, and 1.8 months in patients with decompensated disease.² Prior to the success of regorafenib in the second-line setting, there had been very little progress. Within the past few years, 3 candidate agents failed to meet primary endpoints in large phase III trials, including brivanib (BRISK-PS study), everolimus (EVOLVE-1 study), and ramucirumab (REACH study).

Brivanib, a dual inhibitor of VEGFR and FGFR, was the first agent evaluated in second-line therapy for HCC. For this agent, the phase III BRISK-PS study involving 395 patients with HCC who progressed during or after sorafenib treatment or were intolerant to the drug failed to demonstrate a survival benefit versus placebo.³

Compared with placebo, brivanib was associated with significant improvement in objective response rate (ORR; 10% vs 2% with placebo; P = .003) and other secondary endpoints, including time to progression (TTP; 4.2 vs 2.7 months; P =.001) and disease control rate (DCR). However, brivanib did not meet the prespecified primary endpoint of OS improvement (9.4 months vs 8.2 months with placebo; HR, 0.89; P =.3307). Treatment-related toxicities were substantial, with a discontinuation rate secondary to adverse events (AEs) of 23%.³

Adding this this failure, mTOR inhibition also did not demonstrate success in a large trial. In the phase III EVOLVE-1 study, the mTOR inhibitor everolimus failed to demonstrate clinical benefit.4 The survival of patients with HCC progression on or after sorafenib allocated to everolimus (n = 362) or placebo (n =184) did not differ significantly (median OS, 7.6 vs 7.3 months, respectively; HR, 1.05; P = .68). Everolimus also did not significantly affect secondary endpoints including TTP (HR, 0.93).⁴

Biomarker-Driven Approaches Under Exploration

Most recently, again on the VEGFR front, the phase III REACH trial of ramucirumab failed to detect a significant overall OS benefit versus placebo (median OS, 9.2 months vs 7.6 months; HR, 0.87; P = .14). Ramucirumab was associated with significant improvement in progression-free survival (PFS; 2.8 months vs 2.1 months; HR, 0.63; P <.001), ORR (7% vs <1%; P <.0001), and DCR (56% vs 46%; P = .011).⁵ A number of subgroup analyses were conducted on studies with disappointing results in the attempt to discover potential patients who benefited from treatment. Despite overall negative findings from the REACH trial, subgroup analysis revealed that patients with elevated baseline levels of serum alpha-fetoprotein (AFP), a marker of poor prognosis, benefited from ramucirumab.

A prespecified subgroup of patients with elevated baseline AFP >400 ng/mL had a 20% ORR and significant improvement in survival after receiving ramucirumab (median OS of 7.8 months vs 4.2 months with placebo; HR, 0.67; P = .0059). The molecular mechanisms for this correlation remain unclear, but according to the authors, high α-fetoprotein expression may be associated.⁵

Based on these findings, the phase III REACH-2 trial (ClinicalTrials.gov identifier NCT02435433) has been initiated to assess ramucirumab in a second-line setting in patients with elevated AFP. In a post-hoc analysis of REACH, ramucirumab treatment led to a greater reduction in the risk of death in patients with progressively higher baseline AFP values.

The molecular heterogeneity of HCC, which may contain as many as 30 to 40 different mutations per tumor, including 5 to 8 mutations in putative oncogenic driver genes, renders the identification of key oncogenic pathways that would guide selection of targeted treatment challenging. However, this broad heterogeneity could also contribute to the success of multikinase inhibitors, which heat several targets.

Encouraging findings for a potential enrichment strategy based on signaling pathway activation were made in a randomized phase II trial with the small-molecule MET inhibitor tivantinib: Patients with tumors expressing high levels of MET had a survival benefit, and tumor MET expression appeared both predictive and prognostic.⁶

The MET receptor tyrosine kinase and its ligand hepatocyte growth factor (HGF) are implied in carcinogenesis and metastatic progression in HCC; 25% to 87% of HCC overexpresses MET. High-MET expression is particularly frequent in advanced tumors with vascular invasion.⁶

In the phase II trial, 107 patients with advanced HCC with progression on or intolerance to sorafenib were randomly allocated at a 2:1 ratio to tivantinib (n = 71; 39 with initial starting dose of 360 mg twice daily, amended to 240 mg twice daily due the incidence of severe neutropenia) or placebo (n = 36). In the overall study population, tivantinib was associated with longer TTP (1.6 vs 1.4 months; HR, 0.64; 90% CI, 0.43-0.94; P = .04).⁶

Tumor MET expression was assessed by immunohistochemistry in 77 patients (72%), revealing high-MET tumors in 37 patients (48% of the study population). Patients with high-MET tumors had significant benefits with tivantinib treatment, resulting in prolongation of median TTP (2.7 months vs 1.4 months with placebo; HR, 0.43; P = .03) and of median OS (7.2 months vs 3.8 months with placebo; HR, 0.38; P = .01).⁶

Tumor MET status appeared predictive for response to treatment: Patients with low-MET tumors did not benefit from tivantinib treatment (median TTP, 1.5 months vs 1.4 months with placebo; HR, 0.96; P = .92; median OS, 5.0 months vs 9.0 months with placebo; HR, 1.33; P = .92).⁶

Tivantinib as second-line treatment in HCC is currently being investigated in a randomized phase III study in a selected population of patients with high-MET advanced HCC after sorafenib failure (METIV-HCC trial; ClinicalTrials.gov identifier NCT01755767). During a preplanned interim analysis, an independent data monitoring committee recommended the continuation of the phase III trial. The interim analysis was conducted following 60% of required events. The next planned analysis will be conducted after 100% of target events have been reported.

Promising outcomes in the second-line treatment of advanced HCC have also been observed with cabozantinib, a small-molecule inhibitor of MET, RET, and VEGFR. A phase II study including 41 patients with advanced HCC reported clinical activity, with a 12-week DCR of 68%, median PFS of 4.4 months, and median OS of 15.1 months.⁷

Immune Checkpoint Blockade Under Exploration

Based on these results, a randomized phase III trial (CELESTIAL; ClinicalTrials.gov identifier NCT01908426) has been initiated to evaluate cabozantinib versus placebo in patients with advanced HCC that progressed following sorafenib. With a targeted recruitment of 760 subjects, the study projects 90% power to detect 31.6% increase in the primary endpoint, OS (HR, 0.76). Cancer immunotherapy has emerged as a highly effective therapeutic approach in multiple solid malignancies, and recent evidence suggests promise in the treatment of advanced HCC. Data from an ongoing phase I/II trial of nivolumab, a fully human mAb directed against PD-1 indicate safety and efficacy in patients with advanced HCC.

Both the PD-1 and CTLA- 4 inhibitory immune checkpoints normally temper T-cell responses to avoid damage to self; however, overexpression of these receptors and their ligands in cancers mediates escape from an antitumor response.

Nivolumab prevents interaction of PD-1 with its ligands, augmenting antitumor immunity. Overexpression of PD-L1 is observed in 45% to 95% of HCC and has been associated with a poor prognosis.

In a phase I/II study, ascending doses of nivolumab were explored that ranged from 0.1 mg/kg to 10 mg/kg every 2 weeks for up to 2 years. In an expansion cohort, nivolumab was administered at 3 mg/kg across 4 cohorts, based on their hepatitis infection status and whether they had received prior sorafenib.

Of 206 enrolled patients, 91 had ≥18 weeks of follow-up and were evaluable for responses. The ORR was 9%, with a median duration of response of 17 months (range, 6-24). When including stable disease, the disease control rate was 65%. The 6-month OS rate was 69% (95% CI, 0.43-0.85).⁸

Recruitment to the phase I/II study (ClinicalTrials.gov identifier NCT01658878) continues, with a goal of enrolling 600 patients. In the frontline setting, a phase III study is also comparing nivolumab with sorafenib for patients with HCC. The study hopes to enrolled 726 patients with and estimated completion date of July 2017 (ClinicalTrials.gov identifier NCT02576509).

References

1. Bruix J, Merle P, Granito A, et al. Efficacy and safety of regorafenib versus placebo in patients with hepatocellular carcinoma (HCC) progressing on sorafenib: results of the international, randomized phase 3 RESORCE trial. Presented at: 2016 World Congress on GI Cancer; June 28 - July 2, 2016; Barcelona, Spain. Abstracts LBA03.
2. Iavarone M, Cabibbo G, Biolato M, et al. Predictors of survival in patients with advanced hepatocellular carcinoma who permanently discontinued sorafenib. Hepatology. 2015;62:784-791.
3. Llovet JM, Decaens T, Raoul JL, et al. Brivanib in patients with advanced hepatocellular carcinoma who were intolerant to sorafenib or for whom sorafenib failed: results from the randomized phase III BRISK-PS study. J Clin Oncol. 2013;31:3509-3516.
4. Zhu AX, Kudo M, Assenat E, et al. Effect of everolimus on survival in advanced hepatocellular carcinoma after failure of sorafenib: the EVOLVE-1 randomized clinical trial. JAMA. 2014;312:57-67.
5. Zhu AX, Park JO, Ryoo BY, et al. Ramucirumab versus placebo as second-line treatment in patients with advanced hepatocellular carcinoma following first-line therapy with sorafenib (REACH): a randomised, double-blind, multicentre, phase 3 trial. Lancet Oncol. 2015;16:859-870.
6. Santoro A, Rimassa L, Borbath I. et al. Tivantinib for second-line treatment of advanced hepatocellular carcinoma: a randomised, placebo-controlled phase 2 study. Lancet Oncol. 2013;14:55-63.
7. Verslype C, Cohn AL, Kelley RK, et al. Activity of cabozantinib (XL184) in hepatocellular carcinoma: results from a phase II randomized discontinuation trial (RDT). J Clin Oncol. 2012;30:(suppl; abstr 4007).
8. Sangro B, Melero I, Ya TC, et al. Safety and antitumor activity of nivolumab (nivo) in patients (pts) with advanced hepatocellular carcinoma (HCC): Interim analysis of dose-expansion cohorts from the phase 1/2 CheckMate-040 study. J Clin Oncol. 2016;34 (suppl; abstr 4078).