52

Real-World Issues: Fialuridine

Fialuridine (FIAU) was a drug used to treat hepatitis B virus (HBV) in the early 1990s. Its use in one clinical trial in particular produced seven cases of severe hepatic toxicity (including five deaths) in 15 patients. This trial produced major fallout in the world of clinical research in the United States. The events are briefly reviewed here, and a few comments are made in regard to the drug safety aspects of the study.

FIAU is a pyrimidine nucleoside analogue believed to be a promising treatment for HBV patients. It had previously been studied in a major cancer hospital against other viral infections, including cytomegalovirus (CMV) and various herpes viruses, with some promising results. Animal studies revealed vomiting, diarrhea, mild cardiac toxicity at high doses, and bone marrow toxicity. No liver toxicity was noted in any of the animal models (rat, mouse, and monkey).

In 1989, a small pharmaceutical company, together with the National Institutes of Allergy and Infectious Diseases, did a phase I/II dose escalation trial in human immunodeficiency virus (HIV) patients with positive CMV cultures using fiacitabine (FIAU is an active metabolite of fiacitabine). The study of 12 patients did not show an effect on the CMV. Adverse events (AEs) reported were nausea, fatigue, and an increased creatine phosphokinase in one patient. No one died during the study. Follow-up showed four deaths, including one patient who had hepatitis B about 6 months after the end of the study. His death was believed to be due to the underlying hepatitis and other hepatotoxic medications he was taking.

In 1990, a short-term treatment (2 weeks) study of FIAU was done on HIV patients. During the study, it became clear that there was a significant effect on HBV but probably little CMV efficacy. The protocol was amended to allow HBV patients to enter also. Investigators at the National Institutes of Health (NIH) joined the study team. Significant decreases in HBV DNA were noted. Some patients also developed doubling or tripling of serum transaminase levels. It was unclear whether the transaminase elevations were AEs due to the drug or due to the so-called flare phenomenon of elevated transaminases seen when HBV viremia drops with other drug treatment (e.g., interferon).

In 1991, the positive results from this trial led to a trial of 4 weeks of FIAU therapy. The trial showed excellent results in groups receiving the highest dosages, with major reductions in HBV DNA. No AEs producing dropout or dose modification occurred. “Flares” and HBV DNA rebound were noted in several patients, but three of nine patients had continued suppression of HBV DNA and normal liver tests and were HBV-antigen negative. One patient developed abdominal pain (but with normal liver and pancreas tests) that resolved 4 months after the trial. One patient was diagnosed with cholelithiasis, and another had an episode of peripheral neuropathy. One patient died. He had chronic hepatitis B and had a good result with FIAU in the trial as judged by his HBV DNA level. He had nausea and fatigue during the trial. One month after the last dose of FIAU, his transaminase was noted to be four times normal, and he complained of nausea, fatigue, and abdominal pain. A nonstudy physician recommended cholecystectomy, which was done under general anesthesia. Liver biopsy revealed chronic active hepatitis and steatosis. He deteriorated after surgery and died a few months later. An autopsy revealed steatosis. His death was attributed primarily to the anesthesia drugs administered.

On the basis of these results, the NIH and Eli Lilly, Inc. (now developing the drug) began a study in 1993 of carefully selected hepatitis B patients with a planned treatment period of 6 months of FIAU. Within the first few weeks of the study start, some patients complained of fatigue, nausea, cramps, and diarrhea. Some patients had dose interruption. No abnormal laboratory tests were noted. However, all patients had decreases in HBV DNA, and 6 of 10 became HBV DNA negative. The data were reviewed at about 6 weeks into the trial, and because of the positive results, the investigators opted to continue.

A few days later one patient who had discontinued the drug 2 weeks previously presented to an emergency room with nausea, weakness, and hypotension. The transaminases were normal, but the bilirubin and lactic acid levels were elevated. The patient went on to develop liver and renal failure and died, despite liver transplantation. The autopsy revealed pancreatitis, glomerulonephritis, esophageal varices, and pneumonia. The liver showed micro- and macrovascular steatosis, cholestasis, and chronic active hepatitis.

Within 2 or 3 days of the patient’s first showing up at the emergency room, the study was stopped, and all patients still on FIAU were told to stop the drug. All patients (15) were admitted to the NIH Clinical Center for observation. Despite stopping the drug, seven patients developed hepatic failure, pancreatitis, neuropathy, and myopathy and were to have liver transplantation. Five died (some after transplantation) and two survived. Eight of the 15 patients had no AEs.

As a result of this disaster, several inquiries were set up. The U.S. Food and Drug Administration (FDA) established an internal task force and issued a report in November 1993. It was believed that there were many episodes of “missed toxicity” and that the AEs were attributed to the disease rather than to the study drug. The task force also felt that the informed consents and the study monitoring and oversight were not adequate. The FDA issued warning letters to the investigators charging, among other things, failure to immediately report serious AEs to the sponsor and investigational review boards, failure to reduce or terminate dosing in subjects with moderate toxicity, failure to describe all foreseeable risks in the informed consent, failure to follow up on serious AEs, failure to include complete and accurate safety data in the investigator brochure, and failure to adequately monitor by not ensuring that all AEs were reported in the case report forms.

NIH did its own investigation and concluded that the rationale for the studies was strong, especially in light of the lack of other (oral) therapy for the disease in question. The NIH investigators believed the protocols were “meticulously” prepared and implemented and that fatal outcomes could not have been predicted from the AEs. They also noted that the AEs were adequately reported. Thus, NIH and the FDA investigations reached opposite conclusions.

Next, an investigation was done by the Institute of Medicine (IOM) at the request of the secretary of health and human services. The IOM, a private, nonprofit, nongovernmental organization, is part of the National Academy of Sciences. Its review largely agreed with that of the NIH. It concluded that excellent attention was paid to safety monitoring and that there were no significant violations of study conduct or informed consent. They concluded that rapid action to stop the study actually saved lives and prevented an even worse tragedy.

These events were extensively discussed and criticized. Press coverage was vivid and lurid, and a congressional investigation in 1994 occurred with strong charges thrown about. Further work revealed that the FIAU toxicity was due to mitochondrial damage by the drug.

Much has occurred since then in regard to tightening and harmonizing regulations and clinical trial oversight. It is not the intent here to review the issues in clinical trial regulation, oversight and monitoring, or politics. Instead, the drug safety implications are discussed.

• Serious and nonserious AEs due to (study or marketed) drugs may occur that mimic the disease being treated. The implications of this are important. It is not adequate to attribute serious AEs to the disease or condition being treated, background medical conditions, intercurrent problems, or other nondrug causes without a careful consideration of a drug-related etiology. A very high level of suspicion that the drug produced the serious AE must be maintained at all times when evaluating whether a drug has produced a particular AE. The drug is not “innocent until proven guilty.”
  
• Clinical trial safety oversight by the company (sponsor) and the investigator must be “meticulous.” All regulatory requirements (protocol design, investigator qualification, AE collection, reporting and review, consent forms, investigational review board oversight, safety data review committees, etc.) must be followed strictly and completely.
  
  
  
  • A data safety plan (risk management) must be drawn up and in place before the study starts.
    
  • Prestudy signals (from animal data or other clinical trials or class drugs) must be followed carefully.
    
  • Investigator (and sponsor, and monitor) training must be done before the study starts and during the study if new personnel become involved. The training should be of high quality, customized to the study, and done by a training specialist (not simply printed material).
    
  • A sponsor physician must be designated as clearly in charge of the ongoing safety review. Qualified investigational review boards with high-quality experienced personnel who have sufficient time to review safety data must be used. Sponsors and investigators must supply the investigational review boards with easily reviewable data sent at frequent periods.
    
  • Companies and institutions doing clinical trials must have a crisis management plan in place to do a preliminary investigation and to take appropriate actions on critical safety issues within a few hours.
    
  • Larger political and governmental solutions should also be considered (presumably they are), including full-time dedicated national safety monitoring committees, a separate safety organization within the federal government, involvement of academia in ongoing safety monitoring (as in France), limitation of proprietary secrets, an AE reporting system in a federal database for clinical trial AEs combining the safety efforts put into case report form safety reporting with those for regulatory serious AE reporting (this refers to duplication of reporting in clinical trials: to the drug safety group and in the case report form), continued research into trend analysis and early signaling, and so on.

See an excellent review on the FIAU safety issues:

Nickas J. Clinical trial safety surveillance in the new regulatory and harmonization environment: lessons learned from the “Fialuridine crisis.” Drug Inform J. 1997;31:63–70.

An excellent review of the situation is available:

Saag M. A review of the FIAU tragedy and its effect on clinical research. J Clin Res Practice. 1999;1:21–32.

Also see the IOM report:

Manning FJ, Swartz M, eds. Review of the Fialuridine (FIAU) clinical trials. Committee to Review the Fialuridine (FIAU/FIAC) Clinical Trials. Division of Health Sciences Policy. Institute of Medicine. Washington, DC: National Academy Press, 1995.

Stay updated, free articles. Join our Telegram channel

Join