Are protease inhibitors miracle AIDS drugs, or are researchers rushing down a dead-end street?

By Mark K. Anderson

The headlines were straightforward enough. "New AIDS therapies prolonging lives," "Triple-drug therapies are changing patterns, costs of AIDS treatment," "The end of AIDS? Not yet, but new drugs offer hope."

This front-page coverage in newspapers and magazines throughout the country came in conjunction with the Fourth Conference of Retroviruses and Opportunistic Infections, which met in Washington, D.C. from Jan. 22 to 26. The therapies heralded at the conference involve combinations of old AIDS drugs with a new, experimental type of drug termed "protease inhibitors."

Despite the bold, auspicious headlines, however, protease inhibitors have proven to be some of the most controversial as well as the most expensive AIDS drugs yet concocted. First put on the marketplace in late 1995 after the fastest and least stringent approval process in the history of the Food and Drug Administration, protease inhibitors have a mixed clinical record at best -- one that most AIDS clinicians have overlooked for the time being in favor of the increased treatment options the drugs offer.

The story of protease inhibitors parallels the story of the HIV virus they were designed to trammel. Both have inspired considerable dissent within the AIDS research community, although that dissent has inspired surprisingly scant coverage in the mainstream media. And both have subverted the traditional route of disseminating information through peer-reviewed scientific journals in favor of the unmediated forum of press conferences and press releases.

Of the two, though, HIV takes top honors for breaking rules, standards and precedents. Big-budget science, even in the age of hype, has never seen its like before.

Without having published a single paper to back up his claims, cancer researcher Robert Gallo held a press conference at the National Institutes of Health in April 1984 announcing that the newly discovered retrovirus LAV/HTLV-III -- later renamed HIV -- was the "probable cause" of AIDS. Before his announcement, Gallo had spent 20 years in the failed search for a viral cause of cancer and at one point claimed to have discovered a sexually transmitted retrovirus that causes leukemia -- after an absurdly untenable 45 year latency period. It was later revealed that more than half of Gallo's AIDS patients had in fact tested HIV-negative, and a Congressional investigation uncovered fraudulent data in the papers Gallo did eventually publish.

To this day, Gallo's hypothesis -- that HIV causes AIDS -- remains unproven. More than 100,000 papers have been published on HIV, and yet even the outlines of a proof that the virus causes AIDS are lacking. In fact, virologist Dr. Stefan Lanka wrote an article for the AIDS dissident journal Continuum in May 1995 arguing that no one has even proven that the HIV virus itself exists. In conjunction, Continuum posted a £1,000 "Missing Virus" reward for anyone who can locate a paper establishing the independent existence of HIV. To date, two leading researchers have submitted claims to the reward. Neither succeeded. Continuum's contest follows the criteria for isolating any retrovirus, established at the Pasteur Institute in 1973.

As of 1997, HIV-AIDS dissidents have coalesced around two groups, the Group for the Scientific Reappraisal of the HIV-AIDS Hypothesis and the Health Education AIDS Liaison. The former consists of more than 450 scientists, including two Nobel Prize winners, while the latter is a rapidly growing activist organization with 23 chapters around the world. In the past two months alone, chapters have formed in Atlanta and Philadelphia.

Then there's Dr. Peter Duesberg, a widely vilified virologist who is nonetheless considered among the world's leading experts on retroviruses -- the class of viruses to which HIV belongs. Last year he released the book Inventing the AIDS Virus to considerable harrumphing from the AIDS research orthodoxy. In it he argues that AIDS is a syndrome of 30 previously existing illnesses and conditions brought on by a combination of factors, including recreational drug use (an idea first proffered by the U.S. Centers for Disease Control and Prevention), the use of clotting factor given to hemophiliacs (already known to suppress immunity), and the AIDS treatments themselves (such as AZT, which was shelved as a cancer chemotherapy in 1964 because of its extreme toxicity).

Other scientists, including the French researcher whom Gallo was found to have stolen his data from, now argue that HIV is only one cause among many possible factors leading to AIDS.

Press coverage of protease inhibitors, however, typically omits this background. Instead, it begins here: In a virus -- such as HIV -- protease is an enzyme that helps the virus make copies of itself inside infected cells. A protease inhibitor is a synthetically created drug that "gums up" the copying process, preventing the virus from replicating and infecting other cells.

AIDS patients who take these drugs -- or more likely their private and/or public insurance underwriters -- can expect to spend upwards of $10,000 a year on the therapy. And yet there is precious little scientific evidence to suggest that AIDS patients who use these drugs feel any better or live any longer than those who don't.

That may be for a good reason too. One expert in the field sees HIV protease inhibitors as a miracle drug, but not because he thinks they'll cure AIDS. He argues instead that protease inhibitors are the ultimate test of Gallo's hypothesis.

Yes, these drugs appear to stop HIV, and that's precisely the problem. Ultimately, he argues, protease inhibitors may prove the unpopular stance he and a growing number of dissident scientists have taken: that HIV does not cause AIDS and that in its urgent search for effective treatment -- and ultimately a cure -- mainstream science is rushing down a dead-end street.

Dr. David Rasnick is a visiting scientist at the University of California, Berkeley and a pioneer in protease research and protease inhibitor development.

Rasnick has studied proteases and protease inhibitors since the mid-1970s, researching those related to HIV since 1984. He has also followed the AIDS epidemic since its initial days in late 1980. After founding a bio-tech company in the San Francisco area, he and his colleagues began to hear about a strange "cancer" that was afflicting a subset of the city's gay community. Later called GRID (for Gay-Related Immune Deficiency), the syndrome of diseases was eventually given what is now one of the most ubiquitous acronyms in the English-speaking world: AIDS.

And then came the watershed moment in AIDS history.

"In 1984 Robert Gallo had his press conference, and I'll never forget that," Rasnick recalled. "I was just as tickled as I could be, because he claimed that it was a virus that was the probable cause of AIDS. I knew viruses had all sorts of proteases. I had been interested in them for many years. Then I said, 'Wow! It's a retrovirus. I'm a protease inhibitor guy. I'm going to jump in on this.' This was within days or weeks of Gallo's press conference."

Since then, he's become much more skeptical of Gallo's claims, harboring doubts about the HIV-causes-AIDS hypothesis since 1987. But now he sees his area of expertise as providing a key test to check the validity of the claims of Duesberg et al.

His thesis is that HIV protease inhibitors work so outstandingly in stopping the virus that they provide an effective "toggle" in turning off the virus' contribution to a patient's symptoms. If protease inhibitors show no long-term clinical benefits in AIDS patients, then farewell to the assumed direct correlation between HIV and AIDS. (Of course, the solution could be more complex than a "yes" or "no" answer. That is, HIV could be one of several "co-factors" that by themselves are benign but together lead to AIDS -- a position that some AIDS researchers have now taken in favor of the original Gallo model.)

So what do the data say?

Recent media coverage of protease inhibitor studies have been largely positive -- some even playing it like a victory at sea or a Babe Ruth shot out of the ballpark. "There's no toxicity. It's a home run!" Crowed one Newsday article.

But did anyone actually see the ball in the air? Or is the catcher holding it?

"This is research by press release," Rasnick cautions. "People are making all these conclusions from the studies. But there's nothing published on this ... . The only thing that matters to an AIDS patient is whether he feels better, he does better and he lives longer. And that isn't published anywhere in the scientific literature."

Indeed, combing through the journals and conference abstracts, one finds hundreds of references to protease inhibitors and protease inhibitor research. But tracking down a study that actually finds protease inhibitors providing patients with a longer life or better health is like trying to find a passenger pigeon in midtown Manhattan. Accept anything that looks like a pigeon and there are flocks upon flocks to choose from; stick to the task at hand and ... well ... good luck.

One recent National Institutes of Health study, for instance, was publicized as part of the protease inhibitor success stories from the recent Retroviruses Conference. AIDS Clinical Trials Group 315 studied the consequences of putting AIDS patients on a "cocktail" of zidovudine (AZT), lamivudine (3TC) and the protease inhibitor ritonavir. The first two drugs are known as "reverse transcriptase inhibitors" and are in general much more toxic than protease inhibitors.

The study, like so many of the headline-grabbing studies of late, considered a mere three-month trial period.

"You can't determine anything in three months," Rasnick noted. "It's within the background noise. Yet people are drawing tremendous, unwarranted conclusions from this stuff."

And the bottom line -- if the patients on the drug cocktails lived longer or felt better -- was nowhere to be found.

Dr. Michael M. Lederman of Case Western Reserve University was the protocol chairman and author of ACTG 315. He says the study was never designed to consider how the patients fared as a result of the therapy. Instead, his team measured the popular "surrogate markers" CD4 count and viral load (See: 'Surrogate Marked'). However, Lederman does think the extrapolation to a patient feeling better and living longer (morbidity and mortality, in clinical-speak) is a reasonable one.

"We believe that changes in viral load measurements in response to these therapies do predict a response in morbidity and mortality," he noted.

But belief is something entirely different from proof. AIDS researchers never seem to be short on the former. Without the latter, though, the clinician is left with no solid foundation on which to stand -- other than the hallowed ground of the faith healer.