AIDS and Alternative Medicine: A Journalist's Perspective

Remarks to AIDS and Alternative Medicine: Current State of the Science, Bastyr University, Seattle, April 28, 1996:

I started AIDS TREATMENT NEWS 10 years ago, and the newsletter has now published 245 twice-monthly issues. When we started, we thought that AIDS TREATMENT NEWS would focus mainly on alternative/complementary treatments -- thinking that physicians and patients were already flooded with mainstream treatment information from journals and from industry marketing departments.

It did not turn out as expected. While some of our most important articles have been about alternative treatments -- for example, the September 1, 1995 interview with acupuncturist John Sinclair of the Immune Enhancement Project in San Francisco -- we have published many more articles about mainstream experimental treatments, or mainstream approved treatments, than about alternative approaches. Partly this is because the journals and pharmaceutical companies have done poorly in getting physicians the mainstream information they need, so most of our readers' questions concern mainstream treatments.

But also, there is much less research on alternative treatments, due to well-known funding difficulties. Research money, from government as well as from industry, is funneled preferentially into development of the most expensive treatment options -- a major though unrecognized upward pressure on the overall cost of medical care. Our reporting has been affected, because most of the new research findings have come from mainstream research. And we have felt compelled to focus on what is new, because no one has had the answer. Existing treatments, both mainstream or alternative, can be helpful, but most of the answers required to save lives will have to come from new research and development.

What has been lost from failing to research alternative treatments?

The big Vancouver international AIDS conference will happen in two months. After one of the earlier meetings, a physician from the developing world, who had just wandered through the high-tech exhibits and research reports, commented, "There is nothing for us here."

That is not surprising, since every medical tradition on Earth except for one has been excluded from any real role in developing new treatments. Only the corporate, high-tech, big-money, FDA-approved research process has been taken seriously. And this year at Vancouver there will be another issue: the unaffordability of the fruits of the big-money tradition, now that it is producing results arguably worth fighting for. Being excluded from AZT monotherapy seldom became a major concern.

But the loss is not only to the developing world, or to the excluded communities within the United States. For if there are important leads to new approaches within indigenous healing traditions -- and certainly there are -- we are unlikely to hear about them. Excluding almost everybody from the search for better treatments will slow that search greatly.

Major Problems in Drug Development

Some of the biggest problems in medical research and drug development affect mainstream and alternative alike. To simplify the recent history of drug development and approval, several years ago there were two major bottlenecks. One of them was obvious and painfully pressing, and has now largely been fixed, at least for AIDS. The other problem was more subtle, and it remains one of the major challenges for today.

When AIDS TREATMENT NEWS started reporting 10 years ago, the obvious bottleneck was in the late stage of drug development -- the later steps leading up to FDA approval. It was clear that clinical-endpoint trials were going to be required -- and that if a drug did work, it would take years for all but a few people to get it. The drug had to already look good enough that hard-nosed companies would invest many tens of millions of dollars in its development. And people could read about the drug in the newspapers, but could not get hold of it. This bottleneck, of course, affected only mainstream treatments, since anything which gets that close to approval would be considered mainstream by definition.

In most cases we do not have this problem in AIDS today. The current FDA leadership has developed regulations for accelerated approval, which allows a drug to be approved based on easily measured markers like CD4 and viral load -- and encouraged various forms of expanded access, which allow companies to provide a drug to many people before approval. The FDA now goes out of its way to find official, legal ways to make available AIDS drugs that the medical community as a whole has good reason to want to get. There are still problems, however: for example, expanded access can work commercially for some large companies, but others are uninterested, and for small companies it may be out of the question.

But the other bottleneck, early in the drug development process, has not been solved -- and it affects mainstream and alternative treatments as well. This bottleneck, which I believe is the fundamental problem in drug development today, is the difficulty of going from early work to the first credible human data on possible activity or efficacy.

In the case of mainstream treatments, the early work is usually in the laboratory. Treatment leads which show promise commonly do get published in mainstream, peer-reviewed journals. They get published because publication is the lifeblood of academic careers, so the researchers will fight to reach this stage. But then, in almost all cases, development stops cold, no matter how promising the laboratory results may be. With no human data, no one will invest millions of dollars which current rules usually require to be spent before that human data is available.

For alternative treatments, the bottleneck occurs after the treatment is already in human use, sometimes widespread human use, but there are no generally accepted scientific studies or data to support its benefit. Usually those studies never happen -- despite the obvious public health importance of researching treatments which are in widespread use, whether they work or not.

Part of the tragedy is that if a new drug or other treatment approach or idea could reach the stage of early data -- the first credible showing of activity in humans, even if less than a controlled trial -- then the existing system of drug development might respond passably well. Pharmaceutical companies, physicians, academics, and others will start being interested, and the treatment will have the momentum to carry it into further stages of research. But before it has that first credible human data, there is no sufficient evidence, no momentum, to start the ball rolling. Laboratory results, biological plausibility, or popular interest and use, have usually not been enough.

Much of the advance of medicine comes from lucky accidental discovery, in which the unexpected is allowed to reveal itself. Today's system poisons this well of potential discovery, by making research so cumbersome and expensive that it can only test ideas which already have widespread support in advance.

What Can Be Done?

How might research be organized to overcome this particular gap between widespread use of an alternative treatment, and widely credible data showing possible activity or effectiveness? We see several possible approaches. They are relevant not only to researchers, but also to physicians, patients, advocates, public officials, and potential donors who might support research.

* Use of viral load. Viral load testing, now widely used although not quite yet officially approved by the FDA, has made it far easier than before to tell whether or not a drug is working as an antiviral. A major antiviral effect would ordinarily show up in a few days, or a couple weeks at most. This means that the first human trials can now be far shorter than before. The treatments that don't work could be stopped within a couple weeks -- freeing resources to continue the trials for the minority of treatments which did reduce viral load, to get long-term data for them.

* Developing less cumbersome and expensive ways than controlled clinical trials to get credible initial data. The U.S. National Cancer Institute has published guidelines for "best case series" reports from physicians who use unconventional cancer therapies. And in surgery (where randomized controlled trials of operations are now being seriously considered) -- there is also interest in improving case series reports (see THE LANCET, April 13, 1996). The purpose of using case series is not to replace clinical trials, but to help make a preliminary judgment as to what treatments should go into the trials, which will be conducted at great financial and human cost.

* Investigating existing obstacles to credibility. For ten years AIDS TREATMENT NEWS has been answering phone calls for information about particular treatments. Often the calls come in groups; we will suddenly get a number of calls about some particular treatment, mainstream or alternative. We have had a chance to look at what is behind the calls.

For mainstream treatments, and sometimes for nutritional approaches as well, a flurry of calls usually means that research just published in a high-status journal was reported by major newspapers or TV networks. What is driving this interest is scientific discovery.

But behind a flurry of calls about an alternative treatment, in most cases, is not a scientific advance, but a marketing campaign to make money for someone. A new incarnation of some old pill or machine is being pushed.

Recently we used the Internet's World Wide Web to do computer searches on alternative treatments. We used a search engine called Alta Vista (described below), which immediately checks over 20 million existing Web pages for whatever words we specify, and returns results immediately. About 4,000 pages include the phrase "alternative medicine." But a great many of them were listings of health-food type products for sale. When we asked the search engine to eliminate all those that contained the word "product" or "products," we got the total down to a more manageable 2,000.

The fact that a product is commercially promoted does not mean it doesn't work. But it can make it hard for us to get a sense of whether it works or not. We do not know the people behind the campaign; do they believe in their own product? Do they have evidence, and are they evaluating it well? Or are they only pursuing a marketing challenge, in which their goal is to sell as much as they can, without any need to know whether or not it works?

Occasionally a flurry of phone calls or letters results not from a commercial promotion, but from activities of a true believer. Here the credibility problem is largely the same. It is hard to tell what you can trust and what you cannot.

The pharmaceutical-industry mainstream deals with this problem by having third parties, usually academic or clinical research physicians, conduct its research. While the company pays the money, the researcher's career depends only slightly if at all on whether there are positive results -- but depends centrally on a continued reputation for objectivity.

Perhaps the alternative-treatment world could borrow this system, too. The difference would be that we will never have the money that pharmaceutical companies have, and therefore must find research strategies which are far less expensive. This is possible, since most alternative treatments are already in frequent human use, so trials are much safer than those testing a new chemical entity which no human has ever encountered before.

But the history so far is not always promising. Often academics are misused in a way which constitutes a sale of their name for use in a marketing campaign. Hungry researchers or institutions which have a public reputation are hired to do some laboratory test or other piece of the research on an alternative product -- a test which has little or nothing to do with whether the product works as claimed. The academics do their job competently, but meanwhile their name is plastered on ads or promotional articles, to make the product appear to have credibility it would not otherwise have had. This practice complicates the search for methods to do the research right.

* Better standardization and handling of patient data. Much could be learned without the human and material costs of trials by analysis of data already generated in medical practice. But the data needs to be computerized, in ways which avoid unnecessary inconsistencies between the records of different practices. Then the data might be pooled anonymously -- or if that was not sufficient to protect confidentiality, researchers could get statistical summaries from the different offices and analyze those.

* More openness to and more equal communication with indigenous and traditional healers and traditions. Usually we will have to start our learning process on their terms -- not start with trying to force what they are doing into our preconceived scientific mindset.

Ultimately, to make knowledge widely transportable and useful, we will need to find credible ways of testing proposed approaches. But these ways need to be developed in collaboration. It will often not work to bring in a preconceived piece of scientific methodology developed in other contexts and for other purposes.

One potential resource for starting this learning process is ExtraMED, a new database created by the World Health Organization, of over 200 medical journals in developing countries. These journals are excluded from Medline, Embase, and other major bibliographic databases, and therefore largely unknown to researchers in the developed world. (For more information about ExtraMED, see below).

* Case studies of successful medical advancement. We can learn from history and analysis of important medical advances in the Western scientific tradition, but in other traditions as well.

* Building educated donor networks. We must be ready to explain research issues to persons who could contribute major support for research, but who may want to be sure that their money does not go down the same ratholes where so much has gone already. We need to learn what it is about a research project which disposes it to success, and what disposes it to failure -- and be ready to share this knowledge with others.

Appendix: ExtraMED Database

ExtraMED is a database of over 220 biomedical journals selected by the World Health Organization as among the best in the developing world. But almost all of these journals are not indexed in Medline, Science Citation Index, or any other Western indexing service. Therefore most researchers never find out about this published research.

An eye-opening article in the August 1995 SCIENTIFIC AMERICAN, "Lost Science in the Third World" by staff writer W. Wayt Gibbs, compiles appalling statistics on the exclusion of developing-country research from the world's scientific literature. A handful of indexing services have come to serve as gatekeepers of whose work can reach the scientific community. They have allowed only two percent participation in science from 80 percent of the world. And the numbers have become seriously worse during the last ten years.

ExtraMED, now in its second year, claims particular strength "in such topics as AIDS, tropical medicine, communicable diseases, emergency conditions, and traditional medicine." The database is distributed on CD-ROM, which stores the complete text of all pages of all articles as high-quality images -- together with a cumulative index which allows all disks to be searched with one command.

ExtraMED costs $1800 a year for 10 disks (50% less in the developing world). For more information, contact DV Trimmer, Informania Ltd., P.O. Box 1359, London W3 0ZU, England; fax 011-44-1730-265398, email 100060.172@compuserve.com.