News

Memo suggests Upjohn concealed Halcion side effects

An investigation by the United States Food and Drug Administration (FDA) into possible misconduct by Upjohn, makers of the sleeping pill Halcion (triazolam), "strongly suggested" that the company knowingly concealed side effects in an early clinical trial, according to an FDA memorandum. Subjects in protocol 321, a study of 28 healthy prisoners for 42 days in 1972, experienced more adverse reactions than were shown in the summary that the pharmaceutical giant submitted to the FDA in its application for approval. The true incidence of side effects emerged only in 1991, when the original data came to light during a court case in the US over Halcion.

The memo of December 1992, obtainable under the US Freedom of Information Act, shows that the FDA considered referring the case to the US Department of Justice for an investigation by a grand jury. But it decided not to go ahead because the study was considered to be marginal importance in Halcion's approval for marketing.

The FDA decided to investigate the circumstances surrounding the submission of the report on protocol 321 after Britain banned the drug in October 1991. The memo says that Halcion was banned in Britain "as a result of the information pertaining to study 321." In its letter to Upjohn announcing the suspension of its product licence the United Kingdom's Committee on Safety of Medicines said that the licence would never have been granted had adverse reactions been reported accurately.

The errors emerged when an American woman, Ilo Grundberg, sued Upjohn, claiming that she had shot her mother under the influence of Halcion. Ian Oswald, emeritus professor of psychiatry at Edinburgh and an expert witness for Mrs Grundberg, discovered the discrepancies when he compared the summary that the firm had submitted to the FDA with individual case reports.

Upjohn reported the errors to the FDA and drug regulators worldwide, blaming them on "transcription error." According to the FDA's memo the firm did not reanalyse the data, claiming that this would not change the frequency of side effects. In fact, the summary reported four events of nervousness or anxiety and two incidents of paranoia in the 28 prisoners, while case report forms show at least 70 events of nervousness or anxiety and seven prisoners with paranoia. Upjohn is suing Professor Oswald, a long time critic of Halcion, and the BBC for libel in the High Court in London. The trial, over a BBC Panorama programme and an article in the New York Times which quotes Professor Oswald, started at the end of January and is expected to last four months (22 January, p 221).

The memo, from George Prager, consumer safety officer in the FDA's division of scientific investigations, says: "FDA's investigation strongly suggests that the firm was aware that the report was inaccurate and incomplete but continued to withhold notification to the FDA that the report was flawed. Study 321 took place 20 years ago and because of the expired statute of limitations any case developed would be predicated on evidence of a conspiracy to cover up the data from study 321. We were unable to develop such evidence from the current inspection. However, it was felt that case may be appropriate for an investigative grand jury."

But officials decided not to take further action because the facts withheld would not have been material to Halcion's application for approval. The memo cites the unusual nature of the study subjects - prisoners - and the high dosage, which was never seriously considered for marketing in the US.

The libel trial has focused attention again on protocol 321 and other studies of Halcion over which question marks hang. Documents which the company fought for years to keep secret are emerging in the trial and through lawsuits in the US.

Four of the investigators in clinical trials of the drug have been disqualified by the FDA. One, a Mississippi doctor, Samuel Fuerst, was disqualified in 1984 for professional misconduct. The FDA found no evidence that the people he had listed as patients had ever taken the drug. Upjohn agreed to strike the results of the study from its database in 1984, but the company's president, Ley Smith, admitted last month that it had submitted Dr Fuerst's results to British drug regulators in 1991 without mentioning the FDA's action seven years before.

Upjohn refused to comment until the end of the libel trial.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?