The investigational cancer drug brivanib did not significantly increase survival for people with hepatocellular carcinoma over existing standard therapy, researchers reported at the recent Liver Meeting 2012, the 63rd Annual Meeting of the American Association for the Study of Liver Diseases (AASLD) in Boston. But another drug, tivantinib, did appear beneficial for a subset of patients.
Over years or decades chronic hepatitis B or C, heavy alcohol consumption or other causes can lead to severe liver disease including cirrhosis and hepatocellular carcinoma (HCC), a form of primary liver cancer. One of the most common cancers worldwide – especially in Asia – it is often not detected until advanced stages and is therefore difficult to treat.
The multikinase inhibitor sorafenib (Nexavar) is currently the only drug shown to extend survival for liver cancer patients who are not eligible for resection or surgical removal. Like many cancer drugs, however, it can be difficult to tolerate.
Philip Johnson, from the University of Birmingham, and an international team of colleagues conducted the phase 3 BRISK-FL trial to evaluate the safety and efficacy of sorafenib versus brivanib, a selective inhibitor of receptors for vascular endothelial growth factor and fibroblast growth factor, two chemical messengers that play a role in liver cancer development. In previous trials brivanib demonstrated anti-tumour activity, though it did not extend overall survival compared with placebo.
BRISK-FL enrolled 1155 people with advanced HCC who had received no prior systemic treatment. Most (84%) were men, about two-thirds were Asian and the median age was 60 years. About 45% had hepatitis B, about 20% had hepatitis C and about 15% had alcoholic liver disease. More than 90% were Child-Pugh stage A. Half had distant metastasis, or spread beyond the liver, and just over one-quarter had regional lymph node metastasis.
Participants were randomly assigned to receive 800mg oral brivanib or 400mg oral sorafenib, both twice daily. They were followed until they experienced disease progression or unacceptable toxicity. The researchers looked at overall survival as a primary endpoint, as well as time to progression, objective response rate, disease control rate and quality of life.
The median duration of treatment was 3.2 months in the brivanib arm and 4.1 months in the sorafenib arm. Median overall survival was 9.5 months in the brivanib arm compared with 9.9 months in the sorafenib arm, not a statistically significant difference. No significant survival differences were seen in subgroups based on geographic region, cause of HCC or disease severity.
The median time to progression was 4.2 months in the brivanib arm compared with 4.1 months in the sorafenib arm, again not a significant difference. Less than 1% of people in the both treatment groups experienced complete response. Partial response was seen in 12% of patients in the brivanib arm and 8% in the sorafenib arm. About half of participants in both arms (54 vs 56%, respectively) had stable disease, whilst disease progression occurred in 16 and 24%, respectively.
Objective response rates as reported by patients were 12% in the brivanib arm and 9% in the sorafenib arm, a difference that fell just short of statistical significance; 46% of brivanib recipients and 53% of sorafenib recipients discontinued the study early due to disease progression.
Both brivanib and sorafenib were difficult to tolerate, with more that half of patients in both arms experiencing serious adverse events (59 vs 52%, respectively. Rates of adverse events leading to treatment discontinuation were 43 and 33%, respectively. Most deaths during the study were due to cancer progression; 1.6 vs 0.3%, respectively, in the two treatment arms were attributed to drug toxicity.
Loss of appetite, fatigue, nausea, vomiting, high blood pressure and low blood sodium were at least 10% more common in the brivanib arm, whilst hair loss, rash and hand or foot skin reactions were more common in the sorafenib arm. Patients’ quality of life declined considerably in both arms by week 12 of treatment, slightly but significantly more so in the brivanib arm.
Based on these findings the researchers concluded: “The primary endpoint of non-inferiority in overall survival for brivanib vs sorafenib was not met.” However, they added, “Brivanib had antitumour activity similar to sorafenib”, based on time to progression, objective response rate and disease control rate. “Brivanib had an acceptable safety profile” but was “generally less well-tolerated than sorafenib”.
At a conference overview for the media, AASLD president Guadalupe Garcia-Tsao said that, since brivanib showed no improvement in survival over sorafenib but was less well-tolerated, “sorafenib will continue to be the standard of care for these patients”.
Another liver cancer study looked at tivantinib, a selective oral tyrosine kinase inhibitor of the hepatocyte growth factor receptor known as MET. In phase 1 studies it showed promise alone and in combination with sorafenib.
Ivan Borbath from Cliniques Universitaires Saint-Luc in Brussels and colleagues conducted a phase 2 multicentre clinical trial of patients with unresectable (not surgically removable) HCC who either experienced disease progression after first-line therapy or were unable to tolerate first-line treatment, usually sorafenib.
A total of 107 participants were randomly assigned to receive 360mg or 240mg twice-daily tivantinib or placebo; the initial 360mg dose was lowered to 240mg after the first set of patients developed neutropenia or low white blood cell count.
In an intent-to-treat analysis, the median time to progression was 1.7 months for the tivantinib arm compared with 1.5 months for the placebo arm. This difference was not statistically significant, nor was the difference in overall survival.
Benefits were greater, however, for the subset of participants with high MET expression. For this subgroup, time to progression was 2.7 months with tivantinib vs 1.4 months with placebo. Overall survival durations were 7.2 vs 3.8 months, respectively. Adverse events included fatigue, weakness, loss of appetite and blood cell deficiencies.
The researchers saw no relation between MET expression and blood biomarkers or hepatitis B or C status, noting that it must be determined by liver biopsy. In the absence of treatment, liver cancer patients with high MET expression have a higher risk of disease progression and death.
Results from this study should be interpreted cautiously as they rely on a subgroup analysis, but a growing body of evidence in various types of cancer indicates that individual genetic factors play a key role in determining treatment response.
“Tivantinib demonstrated a nearly doubling of progression free and overall survival in the MET high group compared to placebo in a Phase 2 study in patients with advanced HCC as second-line treatment”, Jörg Trojan and Stefan Zeuzem from Johann-Wolfgang Goethe-Universität in Frankfurt summarised in a review in the November 21, 2012, advance edition of Expert Opinion on Investigational Drugs. “The activity of tivantinib in combination with sorafenib is also promising.”
Larger studies of tivantinib are being planned specifically for high MET hepatocellular carcinoma patients.
Johnson P et al. Brivanib (BRI) versus sorafenib (SOR) as first-line therapy in patients with unresectable, advanced hepatocellular carcinoma (HCC): Results from the phase 3 BRISK-FL study. 63rd Annual Meeting of the American Association for the Study of Liver Disease, Boston, abstract LB-6, 2012.
Borbath I et al. Randomized controlled phase 2 study (RCT) with tivantinib in pre-treated hepatocellular carcinoma (HCC): efficacy, safety, and MET-analysis. 63rd Annual Meeting of the American Association for the Study of Liver Disease, Boston, abstract 114, 2012.
Trojan J and Zeuzem S. Tivantinib in hepatocellular carcinoma. Expert Opinion on Investigational Drugs, 2012 (epub ahead of print). The study abstract is available here.
Article also posted at: http://www.aidsmap.com/Brivanib-fails-to-match-sorafenib-for-liver-cancer-but-tivantinib-looks-promising/page/2557612/