For Cyramza, a small portion of the hepatocellular carcinoma market is in reach for patients with high baseline alpha-fetoprotein expression

by Anna Boudoures | Jun 5, 2018

Co-authored by Arnold DuBell, Ph.D., MBA

Hepatocellular carcinoma (HCC) has been a notoriously challenging disease to develop drugs for due to its biology and the inherent coinciding liver disease. Only patients with the least amount of liver damage – who have a Child-Pugh A liver function score – typically receive systemic therapy. Patients with Child-Pugh B receive limited benefit from first-line chemotherapy with Nexavar® (sorafenib, Bayer), which is currently the only FDA-approved agent for first-line therapy[i]. Patients with Child-Pugh C status are in end-stage liver disease and are only eligible for supportive care. Worsening liver function through disease progression further excludes patients from treatment. Further complicating treatment, at relapse not all patients will still have this same Child-Pugh A liver function because disease progression may have damaged the liver too much. In 2017, only about 49% of patients with Child-Pugh A liver function who relapsed went on to receive a second-line therapy[ii].

In April 2018, Eli Lilly announced via a press release that their pivotal REACH-2 trial (NCT02435433) of Cyramza® (ramucirumab, Eli Lilly) versus placebo, in combination with best supportive care, met its primary endpoint of overall survival as well as its secondary endpoint of progression free survival in a biomarker-selected group of HCC patients (BCLC[iii] Stage B or C / Child Pugh A) who had received prior Nexavar. On Monday afternoon at ASCO 2018, data were finally presented, which showed that Cyramza (8 mg/kg IV Q2W) increased both the overall survival (OS) and progression-free survival (PFS) of HCC patients with ≥400 ng/mL baseline serum α-fetoprotein (AFP) by 1.2 months (median OS: 8.5 months versus 7.3 months, HR 0.710, p=0.0199; median PFS: 2.8 months versus 1.6 months, HR 0.452, P <0.0001). Although the disease control rate (DCR) was significantly improved with Cyramza, the overall response rate (ORR) was not. The Cyramza arm was associated with an 4.6% ORR, compared to an ORR of 1.1% for patients on the placebo arm (p=0.1156), and patients on the Cyramza arm were associated with a 59.9% DCR, compared to a DCR of 38.9% for patients on placebo (p=0.0006). Neither arm was associated with a complete response.  Almost two-thirds (58.9%) of patients who received Cyramza experienced at least one grade ≥3 treatment-emergent adverse event, and 10.7% discontinued treatment due to a treatment-related adverse event. The only grade ≥3 toxicities that occurred in more than 5% of patients were hypertension (12.2% versus 5.3%) and bleeding / hemorrhage events (GI hemorrhage events and epistaxis; 5.1% versus 3.2%). In a pooled analysis of patients from the REACH trial[iv] with AFP ≥400 ng/mL and patients from REACH-2, the median OS was 8.1 months on Cyramza versus 5.0 months on placebo (HR 0.69, p=0.0002)[v].

But will this statistical significance in survival and disease control be enough for physicians to consider “reaching” for Cyramza in their post-Nexavar HCC patients? Just a few years ago, Cyramza’s survival data may have been sufficient to capture significant market share. But after a decade of calm, the HCC market has become a small hotbed of development. Since Nexavar’s 2007 approval, numerous drugs have tried and failed to enter the HCC market. But in the past year, this changed. Stivarga® (regorafenib, Bayer) managed to break the barrier to the HCC market in April 2017, when it was FDA-approved based on an improvement in median overall survival for relapsed patients compared to placebo in the RESORCE trial (10.6 months versus 7.8 months, HR 0.63, p<0.0001)[vi], setting a new benchmark for therapies entering the market. This agent was also approved for the same setting in Japan (May 2017) and Europe (August 2017).  Stivarga was followed closely by the hottest new development in oncology, checkpoint inhibitors. In the phase I/II CheckMate-040 trial, Opdivo® (nivolumab, Bristol-Myers Squibb / Ono Pharmaceuticals) administration resulted in a 13.2-month median overall survival for HCC patients previously treated with Nexavar, which resulted in accelerated FDA approval in September 2017 for relapsed / refractory HCC[vii]. Unlike Stivarga, Opdivo is not yet approved in Europe or Japan. These two agents are likely to be joined in the United States in 2019 by two other agents.  Cabometyx® (cabozantinib, Exelixis) was filed for regulatory approval in May 2018 for Nexavar pre-treated patients based upon positive overall survival data compared to placebo in the phase III CELESTIAL trial that was previously presented at the ASCO Gastrointestinal Cancers Symposium but updated in a poster Sunday morning at ASCO (median OS: 10.2 months versus 8.0 months, HR 0.63, p = 0.0049)[viii].  Lenvima® (lenvatinib, Eisai), was filed for approval in 2017 in the United States, Europe and Japan based on positive data in the REFLECT trial that evaluated the non-inferiority of Lenvima to Nexavar as a first-line therapy[ix].  Lenvima has subsequently been approved (March 2018) by Japanese regulators.

The mild safety profile and strong efficacy data for Opdivo suggests that physicians will likely utilize Opdivo as their primary second-line therapy, relegating additional competitors, including Cyramza, to use in later lines of therapy. With multiple new entrants in the past year, Cyramza must “reach to new hand-holds” to capture market share. Cyramza’s two defining features are its demonstrated efficacy in patients who express a biomarker, α-fetoprotein (AFP), and its mild safety profile. The rationale for use of this biomarker-defined patient population lies in Lilly’s previous attempt at getting Cyramza approved for HCC patients via the REACH trial. The REACH trial enrolled all patients following Nexavar use regardless of AFP levels; however, this did not meet its primary endpoint of overall survival (9.2 months versus 7.6 months, HR 0.87, p=0.14). However, a retrospective sub-analysis of patients with high AFP (≥400 ng/mL) enrolled in the REACH trial suggested that Cyramza could be effective in this subgroup (median OS: 7.8 months versus 4.2 months, HR 0.67, p=0.006)iv. These high AFP patients comprise about 20% of relapsed HCC patients, and selects for a group of patients with worse prognosis than patients with lower levels of the circulating protein[x].

The poor prognosis of patients with elevated AFP may offer some explanation as to why median overall survival reported for Cyramza in REACH-2 (8.5 months) was two to five months shorter than that reported for Cabometyx (10.2 months), Stivarga (10.6 months), and Opdivo (13.2 months). Additionally, the patients in REACH-2 on the placebo arm achieved a higher than expected overall survival, further diminishing the benefit of Cyramza for patients with high AFP. For example, the median overall survival for patients with high baseline serum AFP on the placebo arm in the REACH trial was 4.2 months versus 7.3 months for patients in the placebo arm of REACH-2. The discussant (Dr. Bergsland) noted that differences in the two patient populations enrolled in REACH and REACH-2, as well as post-progression treatments, may partly explain the higher survival seen in patients enrolled in the placebo arm of REACH-2. Although it will be important for Eli Lilly to message the benefit of Cyramza compared to competitors in this selected subgroup with poor prognosis, the biomarker requirement could diminish physician’s interest in the use of the agent, especially considering that the benefit for Cyramza relative to placebo, even in the biomarker-selected population, appeared to be minimal.

With such stiff competition, another differentiating factor for Cyramza may be its toxicity profile compared to the other approved (or soon-to-be approved) tyrosine kinase inhibitors, which appears to be milder than Stivarga or Cabometyx. Of note, the only grade 3 or higher adverse event that occurred in more than 10% of patients was hypertension. However, patients enrolled in the CELESTIAL trial and treated with Cabometyx not only reported grade 3-4 hypertension (16%), but also reported higher occurrence of grade 3-4 palmar-plantar erythrodyesthesia (17%), diarrhea (10%) and fatigue (11%). However, this highlights another reason Opdivo, which only reported higher occurrence of grade 3-4 increases in lipase (13%) and aspartase aminotransferase (10%) levels, will likely generate strong interest in HCC patients in the second-line setting.

However, the apparent improved toxicity profile relative to the other tyrosine kinase inhibitors makes Cyramza an attractive partner for combination strategies with PD-1 and PD-L1 inhibitors such as Opdivo. For example, there is currently an ongoing Phase I trial of Cyramza in combination with Tecentriq® (atezolizumab, Genentech/Roche; NCT02572687). At least in gastric cancer patients, this combination exhibited a manageable toxicity profile[xi].  

While the future for Cyramza remains unclear – Eli Lilly had only suggested in their April press release they would “initiate regulatory submissions in mid-2018” – one certainty is that patients with HCC are the winners in the current market environment. New development affords physicians choices which they can tailor to patient characteristics, including a drug for a biomarker-selected patient population. Just a few years ago, HCC patients had no choice once they experienced disease relapse following Nexavar. Should Eli Lilly decide to file for regulatory approval based upon the REACH-2 data, then Cyramza may become another agent to benefit patients with an incredibly hard-to-treat disease.


[i] da Fonseca LG, Barroso-Sousa R, Bento AD, et al., “Safety and efficacy of sorafenib in patients with Child-Pugh B advanced hepatocellular carcinoma;” Mol Clin Oncol, 3 (4):793-796, 2015.

[ii] CancerMPact® Treatment Architecture, Kantar Health. Available from www.cancermpact.com. Accessed 4 June 2018.

[iii] BCLC: Barcelona clinic liver cancer staging system

[iv] Zhu AX, Ryoo BY, Yen CJ, 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 16 (7): 859-870, 2015.

[v] Zhu AX, Kang Y-K, Yen C-J, et al., “REACH-2: A randomized, double-blind, placebo-controlled phase 3 study of ramucirumab versus placebo as second-line treatment in patients with advanced hepatocellular carcinoma and elevated baseline alpha-fetoprotein following first-line sorafenib;” J Clin Oncol, 36 (suppl): Abstract 4003, 2018.

[vi] Bruix J, Qin S, Granito A, et al., “Regorafenib for patients with hepatocellular carcinoma who progressed on sorafenib treatment (RESORCE): a randomised, double-blind, placebo-controlled, phase 3 trial;” Lancet, 389 (10064): 56-66, 2017.

[vii] Crocenzi TS, El-Khoueiry AB, Yau TC, et al., “Nivolumab in sorafenib-naïve and -experienced patients with advanced hepatocellular carcinoma: CheckMate 040 Study;” J Clin Oncol 35 (suppl): Abstract 4013, 2017.

[viii] Abou-Alfa et al., “Cabozantinib versus placebo in patients with advanced hepatocellular carcinoma who have received prior sorafenib: Results from the randomized phase 3 CELESTIAL trial;” J Clin Oncol, 36 (suppl): Abstract 4019, 2018.

[ix] Kudo M, Finn RS, Qin S, et al., “Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomized phase 3 non-inferiority trial;” Lancet, 391 (10126): 1163-1173, 2018.

[x] Farinati F, Marino D, De Giorgio M, et al., “Diagnostic and prognostic role of alpha-fetoprotein in hepatocellular carcinoma: both or neither?” Am J Gastroenterol, 101 (3): 524-532, 2006.

[xi] Bang Y-J, Golan T, Lin C-C, et al., “Interim safety and clinical activity in patients with locally advanced and unresectable or metastatic gastric or gastroesophageal junction adenocarcinoma from a multicohort phase I study of ramucirumab plus durvalumab;” J Clin Oncol, 36 (suppl 4S): Abstract 92, 2018. 

Leave a comment

More than one Google Analytics scripts are registered. Please verify your pages and templates.