Nevirapine Misinformation: Will It Kill?
December 31, 2004
Summary: In mid December 2004 three Associated Press stories created widespread doubts about nevirapine, a well-known, critically important drug that can prevent HIV in many of the 1,800 babies now infected every day by their mothers in childbirth. The media allegations that went around the world grew out of a bitter personal and personnel dispute between two employees at the U.S. National Institutes of Health. No new information about nevirapine was released; doctors know that it still has the same risks and benefits after the newspaper stories as before. But many experts fear that the emotions released by the worldwide misinformation will result in many HIV-positive mothers getting no treatment and unnecessarily infecting their children with HIV. Here is background that has been missing in many of the news reports.
On December 14 and 15 the Associated Press touched off a media firestorm with stories charging that side effects of single-dose nevirapine (to prevent mothers with HIV from infecting their babies during childbirth) had been covered up. The next day it reported on the August 2003 death of a woman in a U.S. clinical trial of continued treatment with nevirapine (not single dose), due to a rare liver failure probably caused by the drug, after an abnormal blood-test result was not noticed in time. Later the AP quoted responses -- one comparing nevirapine's distribution in Africa to the notorious Tuskegee Experiment, another charging that Africans were treated like guinea pigs. In fact there never was any evidence of a significant risk of side effects from only a single dose of nevirapine. There is a risk of HIV drug resistance, but this is well known to all AIDS doctors and experts and has never been covered up.
Every day about 1,800 babies are born with HIV, mostly to women who have no treatment options either for themselves or to prevent the infection of their child. There is no reason to doubt that single-dose nevirapine works, and could prevent about half of these infections. Because of the resistance problem, single-dose nevirapine is not the first choice -- but sometimes it is the only choice possible.
The brief media storm that still threatens the lives of thousands of children grew out of a bitter dispute between two officials of the U.S. National Institutes of Health -- Jonathan M. Fishbein, M.D., a physician with clinical-trials monitoring expertise, and his supervisor, Edmund Tramont, M.D., director of the Division of AIDS (DAIDS) at the U.S. National Institute of Allergy and Infectious Diseases (NIAID), part of the NIH. The falling out happened rapidly; Dr. Fishbein was hired by NIH in July of 2003, and notified in February 2004 that he would be fired. Dr. Fishbein sought whistleblower status and released documents to Congress that he said showed "scientific and professional misconduct" at NIAID. The AP published selected internal NIAID emails, memos and reports (see links to these documents below). Dr. Fishbein, still a Federal employee today (earning about $178,000 a year, according to a December 29 story in The Washington Post), set up a Web site, www.honestdoctor.org, which alleges wrongdoing by NIAID officials and provides documents that had been released elsewhere; he "did not provide non-public documents to the Associated Press," according to a statement from his attorney, Stephen M. Kohn.
The danger now is that misleading nevirapine stories published around the world will cause patients, doctors, or even governments to reject single-dose nevirapine to prevent mother-to-child HIV transmission, in cases when no other treatment is possible.
Background on Nevirapine to Prevent HIV Transmission in Childbirth
Nevirapine was approved in the U.S. in June 1996, for use in combination with other antiretrovirals for treating HIV. For this use it is taken twice a day for as long as the virus is under control.
Later, a study in Uganda from 1997 to 1999 (the HIVNET 012 clinical trial) found that a single dose of nevirapine given to the mother and a single dose to the infant reduced HIV transmission (from childbirth or breastfeeding) during the first 14 to 16 weeks of life to about half of what it was with a very short course of AZT. This study in 645 mother-infant pairs, conducted as a collaboration between researchers from Johns Hopkins University and Uganda and funded by the U.S. National Institutes of Health (NIH), was published in September 1999. It showed that HIV transmission at childbirth could be greatly reduced by a very inexpensive and easy regimen, even when the mother had little or no prenatal care. It is rightly considered one of the great successes in HIV prevention.
Nevirapine alone is not the best regimen, however. Later it was learned from the same study that even the single dose sometimes selects for resistance mutations in the mother's HIV -- a serious problem because it could make her treatment more difficult in the future. This can be prevented by treating the mother's HIV if she needs antiretroviral treatment, which of course should be done anyway -- or by using a much more difficult regimen of AZT to prevent transmission -- or by adding other drugs (usually AZT plus 3TC) to suppress the virus while the nevirapine is slowly eliminated from the body. But still today the great majority of women with HIV do not have access to any antiretroviral treatment. Single-dose nevirapine is inexpensive and easy to use -- and in some areas many women will not accept a longer course of medication, because they are afraid of the consequences if people around them learn or suspect that they have HIV.
Background on the Recent Controversy
The December 2004 controversy developed because after the Uganda study had been published, an NIH audit found that data on possible side effects had not been reported correctly by the Ugandan staff. This problem in one trial did not change the known safety of single-dose nevirapine -- which has been tested in many other clinical trials and widely used to prevent maternal transmission, without side effects. In continuous, long-term use in HIV treatment, serious or fatal side effects can occur, as with any antiretroviral. But these are rare, they can be prevented with proper medical care, and they do not happen with one dose. Aside from the HIV resistance problem, there is no evidence of any significant safety risk from a single dose of nevirapine.
NIAID hired Dr. Fishbein in July 2003, to help it correct the kinds of deficiencies that had been found in the Uganda study. A key disagreement seems to be whether the reporting problems should invalidate the conclusion from that study that single-dose nevirapine is safe and effective for preventing maternal-infant transmission of HIV.
Links to the three AP document-release Web pages, one for each day's story, are http://wid.ap.org/nevirapine1.html, http://wid.ap.org/documents/nevirapine2.html, and http://wid.ap.org/documents/nevirapine3.html. An unsigned email from Honestdoctor.org to this writer, in response to our request for comments on an early draft of this article, specifically asked AIDS Treatment News to direct our readers to the documents on these pages in order to show Dr. Fishbein's side of the issue, and noted that "all the documents about 012 available on the site [www.honestdoctor.org] are public, many having been posted by AP last week." It is unusual for a wire-service story to link to a page set up by the wire service to release documents. If AP later takes them offline, check www.honestdoctor.org.
This whole dispute concerns a nevirapine trial that was completed and published over five years ago -- and recent disagreements over how to report flaws in the research that were discovered after publication. These flaws are universally acknowledged and were being addressed well before Dr. Fishbein arrived at NIH. They almost certainly do not affect our current understanding of the risks and benefits of nevirapine.
We looked through all the documents on Honestdoctor.org as of December 22, 2004, including those on the AP pages, and found nothing there that raised any new doubt on single-dose nevirapine -- now established by much more than the one trial in Uganda. Instead, the documents on that site show the extensive work that NIH and others were doing, both before and after Dr. Fishbein was hired, to correct universally acknowledged reporting problems. The goal was and is to re-analyze the Uganda trial in the light of all available information, both to re-check its conclusions when possible, and also to improve clinical research in the future, particularly in developing countries, which often have a steep learning curve in applying standards created for pharmaceutical-company research at sites with far more resources. We do not know why Dr. Fishbein alleged "widespread scientific and professional misconduct at the NIH Division of AIDS (DAIDS)" (quote from Honestdoctor.org).
The biggest public controversy concerned the rewritten safety report that was the subject of the second AP story, on December 14. (Both versions are available on the AP document-release page http://wid.ap.org/documents/nevirapine2.html.) We do not know NIH rules and procedures, but it is our understanding that Dr. Tramont was responsible for that report, not the team as a whole. Dr. Tramont's version provided more overview, while the previous version more deeply analyzed the problems -- and was repeatedly critical of management decisions not to investigate certain problems further. Dr. Tramont's also differed in noting something the study did right:
"These health visitors [who assisted in the trial in Uganda] knew each patient individually and used culturally sensitive methods of making the contact. As a result of their efforts, maternal and infant follow-up overall for the first six weeks of the study was 97.4% for those who received ZVD [AZT] and 98% for those in the NVP [nevirapine] group. The 18 months follow-up of the study was also high, 93.8% for the ZVD group and 96.1% for the NVP group."
A separate issue, not part of the public controversy but being discussed among some activists and researchers, is whether the current U.S. FDA's GCP (Good Clinical Practice) research standards (which were required but not always followed in the nevirapine trial) are always appropriate for research in developing countries, for which they were not designed. The goal is not weaker standards, but different ways to get at least equally good data, with better patient protection than the current system affords.
For example, simply clarifying which U.S.-government standard for adverse-event reporting should have been used when, and designing reporting forms appropriately so that dates would clearly be missing if they were stamped on the back side where they would not be faxed, and having enough blank forms so they would not be re-used to fax multiple reports, would have improved adverse-event reporting from HIVNET 012 in Uganda (see the original version of the safety review, April 3, 2003, at http://wid.ap.org/documents/nevirapine2.html). Also, according the Dr. Tramont's version (available on the same Web page), the Uganda study team consistently defined "serious" adverse events as those leading to hospitalization or death, which was the customary practice in that medical community, instead of using the more complex NIAID research definitions; the change was never formally approved by NIAID, however. Dr. Tremont's report suggested that with modifications currently accepted by the team (also counting as "serious" those problems that needed treatment to avoid hospitalization, or those that needed hospitalization though the patient refused it) the simpler definition could work.
A strong case could be made that imposing the same research requirements regardless of infrastructure and environment can result in second-class standards for developing countries, since there was little or no attempt to make the standards they must use appropriate and workable for them -- while there was such flexibility in the U.S. and other rich countries where the standards developed. Instead of fighting over how strictly to enforce rules that are sometimes unworkable, why not design rules that will better protect people and data, while helping staff get their work done correctly?
Despite the problems in this trial five years ago in Uganda, there is no reason to doubt that single-dose nevirapine works and reduces HIV transmission to about half of what it would be without treatment. (It may do better than that, since the comparison group was not a placebo but a very short course of AZT, which may have had fewer HIV transmissions than a placebo would have.) The management of NIAID's Division of AIDS, like almost all other AIDS experts, wants to focus on public-health efforts to make preventive and other treatment available, and not derail these efforts by fighting over technical problems in a trial that ended five years ago, when the medical and scientific results of that trial remain firmly established regardless. This is not "scientific and professional misconduct."
AIDS organizations including the Elizabeth Glaser Pediatric AIDS Foundations did well in answering the misinformation about nevirapine. But the damage had already been done. The news story was unexpected because it was tied to no medical or scientific development; it went around the world immediately and no answer could catch up. It is possible that children have already been born with HIV as a result, and that many more will be infected unnecessarily.
Later Dr. Fishbein told Science that "he is 'not in disagreement' that nevirapine saves lives. 'My issue is not nevirapine, but the process'" (December 24, 2004; see reference below).
What Can We Learn for the Future?
This is not the last time the AIDS world will face mass-media storms that carry serious misinformation throughout the world. What can we do about it?
AIDS needs a major organization dedicated to consensus development, and able to offer reporters a single entry point to learn what credible consensus exists on almost any AIDS issue. No position will speak for everybody, but the process should be open to hearing and understanding all dissenting views. Two or more incompatible consensus clusters could emerge, and they would need to be represented by different organizations. But reporters could immediately find broadly credible statements, and talk with experts about them. They might still publish misinformation, but at least an answer could go out with it -- or be clearly missing from their story.
Years ago AIDS had more influence through policy organizations in Washington DC than it does now. Often they represented insiders with their own interests more than a national or world community; for example, treatment and international issues were mostly locked out for years, and usually the only way to have a voice was to be part of the scene in Washington, to be at the right meetings and dinners. Groups like the AIDS Action Council became trade associations, only without admitting it -- and with a deep fear of grassroots activity, and no way for non-specialists to get involved. Still they served an important purpose in providing reporters and others with a common starting point for policy discussion. We miss that today.
Now we need a new kind of organization that prides itself on listening and learning from different people (almost like social scientists exploring what is out there instead of imposing their own view) -- but then finds and suggests practical, creative ways these views and movements can work together in a larger whole. And we need funded, top-quality media outreach that reflects consensus of those working on the epidemic, is on duty at all times, and can answer misinformation immediately.
Ten days into this controversy Dr. Fishbein had a better Web site than most AIDS organizations do after many years -- immediately raising the communication standard. AIDS will face media attacks in the future, and must get its house in order.
Honestdoctor.org is well organized, allowing readers to see immediately what is available and navigate to what they want. The site has an extensive collection of recent press articles, consistently and attractively laid out. Under "Definitions" it has a list of acronyms, a list of people with their titles, and even organizational diagrams of NIAID and DAIDS; it will also have a glossary. When documents are photographed and displayed as images, they are processed correctly, so that they are entirely readable and yet download rapidly on any Internet connection. And last but not least this site has clearly legible type on its main pages, when most sites have at least some text that is too small, too light, or without enough contrast between text and background.
AIDS organizations should ask for volunteer or professional Web help that can do at least as well. Remember that our visitors have millions of other pages a few clicks away, and if a site is poorly organized or otherwise hard to use, many will leave.
For More Information
Here are sources for more information on the recent nevirapine controversy. Except for the last one, they are December 2004 statements or articles in chronological order.
Note: This writer had not worked with or communicated with either Dr. Tramont or Dr. Fishbein before the AP stories. We emailed each a draft asking for comments, and made some but not all of the changes each side suggested. We checked the Boehringer Ingelheim Web site but did not contact the company about this article.
Copyright 2004 by John S. James. Permission granted for noncommercial reproduction, provided that our address and phone number are included if more than short quotations are used.
Copyright 2004 by John S. James. Permission granted for noncommercial reproduction, provided that our address and phone number are included if more than short quotations are used.
South Africa to Stop Recommending Nevirapine for Use by Itself for Mother-to-Child HIV Transmission, Medicines Control Council Says
This article was provided by AIDS Treatment News.