Title: No Acceptable Surrogates for NASH Cirrhosis
Author: Peter G. Traber, MD
The FDA released draft guidance on developing drugs for compensated NASH cirrhosis (Nonalcoholic Steatohepatitis with Compensated Cirrhosis: Developing Drugs for Treatment, Guidance for Industry, U.S. FDA, June 2019). This document has important implications for current drug development programs and future development in this area of unmet medical need. As with all draft guidance document, this reflects the current thinking of the FDA on the topic.
Compensated NASH cirrhosis includes patients with histologic evidence of cirrhosis on liver biopsy, who have not had complications of cirrhosis including but not limited to ascites, variceal hemorrhage, or hepatic encephalopathy. Once complications of cirrhosis occur, it is referred to as decompensated cirrhosis and this guidance does not address patients with this stage of disease.
The FDA recommends that patients enrolled in clinical trials should have cirrhosis due to NASH and not other etiologies of cirrhosis. The most critical part of the draft guidance is the recommendation for clinical trial endpoints. The drug development program should evaluate the effect of the investigational drug relative to placebo on the composite endpoint of time from randomization to the first of any one of the following outcome events:
- Complication of ascites (bacterial peritonitis, diuretic-resistant ascites, hepato-pleural effusion, etc.)
- Variceal hemorrhage
- Hepatic encephalopathy
- Worsening in the MELD score to ≤15 (this assumes the MELD at enrollment is ≤12)
- Liver transplantation
- Death from any cause
In common terms, patients in any clinical trial would be monitored during therapy (or in follow up after therapy has concluded) and would be identified as drug failures if any outcome on this list occurs. Patients only need to have one of these to be labeled a drug failure and prevention of one outcome on this list is not sufficient for an improvement of composite clinical outcome. For a drug to be effective, percentage of drug treated patients would need to have a lower percentage of reaching this composite outcome than placebo-treated patients, and the difference would need to be statistically significant and clinically meaningful. “The FDA strongly recommends clinical outcome trials to support a marketing application.”
What the FDA says next is very important for ongoing and planned clinical trials in this space. The FDA states “Histological improvements in fibrosis can be proposed and justified; however, at present the relationship between histological changes in cirrhosis and clinical outcomes has not been characterized, and further, reversal of cirrhosis (e.g., fibrosis stage F4) may not be feasible. Because currently there is insufficient evidence to support the use of histological improvements as a surrogate endpoint that is reasonably likely to predict clinical benefit to support accelerated approval, in general, the FDA expects to evaluate drugs for the treatment of compensated NASH cirrhosis under the traditional approval pathway.” In other words, the FDA does not recognize any surrogate endpoints for accelerated approval under subpart H, but rather only development programs for full approval using the composite clinical outcome endpoint described above. Note that there are no other potential surrogate endpoints even mentioned in the draft guidance.
Guidance Implications for Ongoing Registration Programs
In February, Gilead Biosciences ($GILD) reported failure of the phase III STELLAR4 trial that evaluated oral selonsertib in patients with compensated NASH cirrhosis (stage 4 fibrosis). While this program failed to meet its primary endpoint, it should be noted that the primary endpoint of this phase III registration trial was a one stage reduction of fibrosis stage in compensated cirrhotic patients, essentially reversal of cirrhosis. Similarly, $ICPT is conducting the REVERSE trial in compensated NASH cirrhosis with the primary endpoint of a one-stage reduction in fibrosis with data expected in July 2020. It is now clear from the draft guidance that the FDA views this endpoint as not sufficient for accelerated approval in compensated NASH cirrhosis.
Galectin Therapeutics ($GALT) has plans to initiate a phase 3 study in compensated NASH cirrhosis patients without varices at baseline using belapectin treatment for two years. While there are no study details to review on clinicaltrials.gov, the company indicated the protocol is being finalized with target start in fall 2019. In a March 2019 presentation, the company indicated that the clinical trial would have the “surrogate endpoint of the proportion of patients in treatment groups who develop esophageal varices vs placebo after 2 years of treatment under an accelerated approval (Subpart H) pathway as a surrogate for the development of large varices.” However, the FDA draft guidance does not indicate this is an acceptable surrogate endpoint for compensated NASH cirrhosis, so the proposed endpoint appears contradictory to the guidance.
In the ENCORE-PH study, Conatus Pharmaceuticals ($CNAT) evaluated patients with compensated NASH cirrhosis with HVPG ≥12 mmHg with 24 weeks of placebo versus 3 doses of emricasan with the primary endpoint of change in HVPG (hepatic venous pressure gradient). While there was no significant difference between treatment and placebo in HVPG in the total patient group, the group of patients with HVPG ≥16 mmHg had a significant reduction of HVPG While it is unclear about the future of this clinical development program, the FDA guidance document makes no statement about using HVPG in a surrogate endpoint in compensated NASH cirrhosis.
The latest FDA draft guidance has an important impact on compensated NASH cirrhosis drug development. At this time, only a composite clinical outcomes endpoint is acceptable for approval. There are no acceptable surrogates. Although the FDA does not feel there is adequate evidence to support reversal of cirrhosis on liver biopsy as a surrogate at this time, I am sure there will be intense discussion on this issue. While there are certain to be many industry comments on this draft, everyone should be pleased that the FDA is leading the discussion by issuing a guidance. As is one of the FDA’s goals, this draft guidance should stimulate creative thinking about endpoints and research to validate surrogate endpoints in this challenging and important area for medical therapy. In the meantime, registration trials will focus on composite clinical outcomes.
To read and receive other insights and commentary on drug development, please visit my personal page and subscribe to my mailing list: https://www.alacrita.com/consultants/peter-g-traber/