Author + information
- Ori Ben-Yehuda, MD⁎ (, )
- David Pizzuti, MD,
- Andrea Brown, MD,
- Marcus Littman, MBA,
- Hunter Gillies, MD,
- Noreen Henig, MD and
- Tobias Peschel, MD, PhD, MBA
- ↵⁎Gilead Sciences, Inc., 333 Lakeside Drive, Foster City, California 94404
To the Editor: Liver toxicity is the most common reason for drug withdrawals (1). Drug-related hepatotoxicity is rare, however, and therefore may not be detected in pre-marketing clinical trials. In the case of drugs developed for orphan diagnoses, the clinical trial experience is invariably even smaller. Post-marketing experience then becomes essential in detecting a possible safety issue. Post-marketing surveillance is typically hindered, however, by significant under-reporting, estimated at upward of 90% (1).
Endothelin receptor antagonists (ERA) are approved for the treatment of pulmonary arterial hypertension (PAH). Historically, ERA as a class have been associated with hepatocellular injury. In short-term trials of bosentan (Tracleer, Actelion Pharmaceuticals, Allschwil, Switzerland), the incidence of alanine or aspartate aminotransferase elevations above 3× the upper limit of normal (ULN) in patients taking 125 mg of bosentan twice daily was 13.7% over 16 weeks, compared with 0% in those taking placebo (2). Concerns over hepatic safety led to the withdrawal of sitaxsentan (Thelin, Pfizer, New York, New York) from global markets in 2010.
The endothelin A selective antagonist ambrisentan (Letairis, Gilead Inc., Foster City, California; Volibris, GlaxoSmithKline, Brentford, United Kingdom) is the most recent ERA agent to be approved by the U.S. Food and Drug Administration (FDA) for the treatment of PAH. No evidence of an association between ambrisentan therapy and hepatotoxicity was identified in pre-registration clinical trials. Nonetheless, at approval by the FDA in 2007, the potential safety risk was assessed as high for the drug class and the database for ambrisentan was not considered large enough to rule out possible hepatotoxicity. Therefore, a “black box” warning and Risk Minimization Action Plan requiring patient enrollment in a controlled distribution program (LEAP [Letairis Education and Access Program]) and monthly monitoring of liver function tests (LFT) was mandated by the FDA.
The availability of rigorous post-marketing monitoring has allowed for a unique opportunity to assess the hepatic safety of ambrisentan.
Spontaneous as well as prescribing healthcare provider reports of abnormal LFT in LEAP were considered “medically confirmed” and compose the focus of this analysis (LFT elevations that were determined by the healthcare provider as not related to ambrisentan treatment were excluded). Hepatic adverse events/reactions were further considered as clinically significant if they satisfied 1 or more of the following criteria: 1) serum aminotransferase levels were elevated >3× ULN; 2) serum total bilirubin was elevated >2× ULN; 3) serum aminotransferase levels were elevated and accompanied by possible signs or symptoms of hepatic injury; or 4) the event was assessed as medically serious by medical reviewers for any other reason.
In the 3.5-year period from June 15, 2007, to December 14, 2010, 10,927 patients were exposed to marketed ambrisentan in the United States with a total exposure representing 9,893 patient-years (Fig. 1). The mean time on ambrisentan was 330.5 days, with 62% of patients on ambrisentan for longer than 3 months and 50% longer than 6 months. The LEAP database included 314 (2.87%) post-marketing spontaneous reports of possible hepatic injury, of which 156 (1.43%) were medically confirmed. A total of 77 medically confirmed cases did not meet the criteria of clinically significant hepatic events (all of these cases reported small or unspecified elevations in liver function tests) with 79 (0.72%) remaining as clinically significant hepatic events. Six of the 79 clinically significant hepatic events had both elevated aspartate and alanine aminotransferase >3× ULN and serum total bilirubin >2× ULN and were considered as potential Hy's Law cases (Hy's Law is defined as >3× ULN for serum alanine or aspartate aminotransferase and serum total bilirubin >2× ULN), with no other reason to explain the combined increase and is considered a marker of possible drug-induced liver injury (3). In 5 of the 6 reports, probable alternative causes of the hepatic events were reported, with the sixth case having intermittent enzyme elevations for 2 years and a poorly documented medical history. Possible alternative causes or contributory factors for the hepatic events were also present in 55 of the 73 remaining medically confirmed clinically significant cases, and in 38 of these, ambrisentan was successfully restarted.
PAH is itself associated with congestive liver disease due to transient or progressive right ventricular dysfunction. PAH patients may also have abnormal serum aminotransferases due to concomitant medical conditions such as underlying diseases that cause secondary PAH. In placebo-treated patients in published clinical trials for ERA, background incidence of elevated aminotransferase levels of up to 6% has been reported in patients with PAH (4). In normal populations, 1% to 4% may also have an isolated or transient elevation of liver transaminases (5).
Differences in the structure of different ERAs may underlie differences in hepatic safety profiles. Bosentan and sitaxsentan are both sulfonamide-based, whereas ambrisentan is propanoic acid–based. As such, the compounds may have different effects on hepatocytes. In vitro analysis using sandwich-cultured human hepatocytes reveals that both bosentan and sitaxsentan reduced hepatic transport pumps. Ambrisentan, in contrast, showed no inhibition of influx or efflux pumps (6).
Based on the LEAP data, in March 2011, the FDA removed the requirement for mandatory monthly monitoring of LFT with ambrisentan therapy (the monitoring for pregnancy and the black box warning against the use of ambrisentan in pregnancy was maintained, in keeping with the known teratogenicity of ERA).
Please note: All authors are employees of Gilead Sciences, Inc., and own Gilead Sciences stock and/or stock options.
- American College of Cardiology Foundation
- Barst R.J.,
- Langleben D.,
- Badesch D.,
- et al.,
- STRIDE-2 Study Group
- Hartman J.,
- Brouwer K.,
- Mandagere A.,
- Melvin L.,
- Gorczynski R.