Washington Waves

The article "Shock Waves" by Carolyn Lochhead (Jan.) is simply the best analysis of the bushwhacking that occurred on November 8 throughout this great country. You do not need to read any more articles to get the complete picture. Only two pages long, it should be inserted into every new history book published in the next century. Bravo, Lochhead!

C. Russell Farmer
Spartanburg, SC

Rick Henderson ("Sobriety Test," Jan.) offers several issues the Republicans must confront if they are serious about cutting government. All good thoughts. But I'll believe all this talk about giving government back to the people only when I read that Washington-area real-estate prices are in free fall. And when I see angry government employees marching on the Capitol to protest massive layoffs. Yes, Virginia, this is no time to be responsible or polite ("Washington Traps," Jan.). We need insurgents, guerrillas. Are these new Republicans up to the job? My optimism is leavened with skepticism. History shows us that few can resist the siren song of power.

Marc Beauchamp
Palm Desert, CA

AIDS Causes, Cont.

As a physician and medical researcher who is not a part of the AIDS-research establishment, but who has made a study of the HIV/AIDS controversy, I was pleased to see REASON tackle the issue ("What Causes AIDS?," June 1994 and "What Causes AIDS: The Debate Continues," Dec.). Let me add a few points:

1) The idea that AIDS in gay men is caused by illicit drug use rather than a virus can be tested by examining the statistical effect of drug use on AIDS risk after HIV infection. For AIDS risk to be associated with drug use before HIV infection but then become independent of drug use after HIV infection suggests that HIV is the critical element, and drug use is simply a marker for behaviors that transmit HIV. We have at least six studies which find exactly such results (for non-IV drugs). Also, the 1993 study discussed in REASON found that immune decline over time in groups of San Francisco men is purely a function of HIV status and has no association with drug use after HIV status is taken into account. This pattern held for the years before AZT became available in 1987, so HIV status was not merely a marker for AZT use, as critics occasionally suggest.

2) Many AIDS risk groups such as hemophiliacs and transfusion recipients and their spouses are not heavy drug users. To explain hemophilia-associated AIDS, authors Thomas, Mullis, and Johnson abandon Occam's razor and postulate that the hemophilia AIDS epidemic results from a second mechanism (this time, gross immune failure due to clotting-factor treatment), which is supposed to have begun to operate at significant levels in hemophiliacs just as the new epidemic of immune deaths exploded among gay men. Even though AIDS appeared epidemically in hemophiliacs and homosexuals a few years after we know a silent epidemic of HIV infection hit both groups almost simultaneously, skeptics nevertheless reject the simple body-fluid borne infection hypothesis in favor of a complicated one, in which the new epidemic of AIDS has as many separate causes as risk groups, and appears at the same time in each coincidentally.

Can we test the clotting-factor causation idea? Just as is the case with drugs, at least two studies find that hemophiliacs who become HIV-positive go on to develop AIDS, years later, independently of how much clotting factor they use after becoming HIV-positive. This again implies that any association between AIDS and clotting-factor consumption prior to HIV infection results simply from the increased probability of contracting HIV the more infected clotting factor is used. After the virus is in the body, clotting-factor use becomes statistically irrelevant, as a viral hypothesis would predict.

Nor does AIDS look toxin-mediated to the epidemiologists, who know that exponentially rising rates of disease are typical of beginning infectious disease epidemics. Ten to 15 years of previous drug use in gay men, even if immunotoxic in ways that cannot be seen in animals, hardly explains the exponential increase in AIDS cases noted in the first four years of AIDS. Similarly, the crude death rate in hemophiliacs during the first quarter of 1984 was nine times what it had been for each of the two preceding years—an abrupt phenomenon not explainable on the basis of a treatment which had first been introduced into this population with wide time variability, more than a decade before. The AIDS plague in hemophiliacs was both real and sudden.

Which brings us to a factual correction: Thomas et al., following Peter Duesberg and Robert Root-Bernstein, have absurdly claimed that "hemophiliacs in the age of AIDS are living longer than they ever did in the past," a myth apparently universal among HIV/AIDS skeptics. In fact, due to clotting-factor treatment, life expectancy in hemophiliacs had risen to nearly normal by 1979, just before the era of AIDS, but fell for the first time in the later 1980s and now has fallen to a figure less than it was before World War II. About half of all clotting-factor-using hemophiliacs were infected with HIV in the seven years before factor concentrate was purified of virus in 1985; since then, between one-fourth and one-third of these infected people have died of AIDS.

3) Hemophilia is not the only problem for HIV/AIDS critics, for data from transfusion AIDS victims also needs explaining. In 1984, even before a commercial HIV-virus test had appeared, a landmark study was published which compared people who developed AIDS after blood bank transfusion with people who had been transfused equally from the same blood bank but remained well. It was found that AIDS victims were far more likely than healthy transfusion control cases to have gotten blood which came from "high-risk activity" donors, such as self-injecting drug users or promiscuous gay men. Only a hypothesis involving infectious microbe(s) plausibly explains a firm statistical link between disease development in blood recipients years after transfusion and the lifestyles of blood donors they had never met.

4) There are reasons other than epidemiologic ones to think that HIV causes AIDS. Despite denial by Thomas et al., there are animal models for HIV/AIDS of similar quality to animal models for most diseases. Since viruses are generally more species specific in action than drugs, lab-animal virus infection models must usually use animal viruses. Fortunately, HIV is but one member of a family of related nine-gene slow-acting brain-infecting mammalian retroviruses ("lentiviruses") which are very similar to each other in structure, genetics, and appearance. Several of these also (just coincidentally) cause animal immunodeficiency syndromes as close to human AIDS as it is realistic for an animal to exhibit. The feline immunodeficiency virus (FIV) and simian immunodeficiency virus (SIV) cause deadly immunodeficiency diseases in domestic cats and Asian monkeys, respectively.

The destruction of immune systems by FIV and SIV models the human immune destruction seen in AIDS better than that produced by any other known immunosuppressant treatment in animals. Cats and Asian monkeys infected with FIV or SIV preferentially lose CD4 lymphocytes, and are afflicted with viral brain lesions, wasting, lymphadenopathy, lymphomas, and infections. Monkeys develop CMV viral retinitis, Pneumocystis pneumonia, M. Avium disease, disseminated candida, and cryptosporidial intestinal disease. All these are opportunistic infections not common in monkeys, and almost unknown in young humans before AIDS (they are very common in AIDS). Most significantly, all these pathologies happen in monkeys due to levels of lentiviral activity much like those in HIV-infected humans, so there is nothing about human HIV infection that contradicts the possibility of a similar virus-caused immune destruction mechanism operating in humans.

Ignoring such similarities, Thomas et al. deny the animal models of AIDS by focusing on latency period. They incredibly call latency studies "tall tales," though even the classic original slow virus findings in Icelandic sheep have by now been well confirmed, with up to eight-year latencies being demonstrated for experimental infections. Duesberg denies the existence of any latent viral disease syndrome following host serum immunity, thus denying the findings of dozens of research groups on slow retrovirus infection in sheep, goats, horses, cows, cats, and monkeys (Duesberg, disbelieving in slow viruses, insultingly calls all these scientists "slow virologists"). Thomas et al., following Duesberg's idiosyncratic and unaccepted view in the matter, go so far as to proclaim that in monkeys "SIV disease follows the primary infection closely"—a generalization contradicted by many independent reports. While experimental SIV disease in monkeys may cause simian AIDS rather immediately (just as HIV does occasionally in humans), SIV may also cause immune-failure death up to three years after initial infection and after a previous good immune response. Deaths in closed colonies of monkeys due to SIV infection (and also in the face of good immune response) have been recorded after latencies of up to seven years. In FIV in cats, another slow virus model, the latency for immune failure, lymphoma, and death after experimental infection is 18-24 months and may be as long as three years. Again it occurs long after the apparent full recovery with good immune response that follows primary FIV infection in cats.

5) Can it be coincidence that HIV, which is the closest human relative to the simian immune-system-destroying retrovirus SIV, is the virus most commonly found throughout every class of AIDS patient? It is true that the more familiar strain of HIV, called HIV-1, infects but does not cause severe illness in monkeys (just as yellow fever does not). However, preliminary reports from several labs now show that HIV-2, a related retro-virus isolated from many AIDS patients in West Africa, does indeed cause death from immunodeficiency when injected into Asian monkeys. It would be a stupendous coincidence if HIV-2 causes immune failure and death in both humans and animals, while at the same time HIV-1, the related strain found in U.S. serum-exposed patients who are severely immunodeficient, is a harmless bystander—yet this is the suggestion of the skeptics. May we now hear from them an admission of probability that at least HIV-2 is a likely cause of some AIDS cases, if not HIV-1?

I suspect no such admission will be forthcoming, when Thomas et al. assert that lab animals "tend to have weakened immune systems," a statement perhaps meant to invalidate any lab-animal experiment showing immune suppression results which threatens the authors' hypotheses. In domestic cats, FIV was first isolated from cats in a household with dying pets, and it also causes disease in feral animals, so it is hardly an artifact of the laboratory setting. Still, if we cannot trust laboratory retrovirus infection data, what are we left with? The authors dismiss epidemiological results they don't like as mere "correlation." Skeptics have thus carefully set a task for science which may be impossible, for it is always possible to find faults and differences in any laboratory animal model and always possible to find something not carefully enough controlled in any epidemiologic study.

6) HIV/AIDS skeptics say that what they want most is new definition of AIDS not involving HIV, so let us here suggest one for heuristic purposes. Our task is to decide what disease entity it is we're trying to nail down with a definition, and why. Two characteristics stand out: 1) The new disease which emerged from obscurely low levels in 1981 in the United States was acquired, meaning that it occurred epidemically in previously well people who had had intimate contact with body fluids of others who were to develop the same syndrome, and 2) this new plague was a severe immunodeficiency which was invariably fatal.

How do we gauge "severe" (life-threatening) immune problems, for purposes of definition? We can crudely quantify immune function and project what degrees of loss are dangerous and predictive of death with a simple count of the CD4 blood lymphocytes. An immunologically healthy person's CD4 count is perhaps 1,000. Less than 200 is critically low and can be termed "immune failure," because this is the approximate number below which opportunistic-infection fatalities are increasingly seen. Most AIDS patients actually travel far beyond this boundary: Today, fully 85 percent of AIDS patients survive long enough to see their CD4 counts fall below 50, with some going all the way to a count of zero before death. It is these patients with life-threatening immunological destruction and very poor prognosis who constitute the core of what most people mean by "AIDS."

If we define "AIDS," then, as otherwise unexplained immune failure in body-fluid-exposed people, we will capture in our definition at some point during life almost everyone who ultimately dies as a result of this syndrome, and we will diagnose almost no one who has more than several years life expectancy. Thus, our definition will do the main job. Moreover, using this definition, we find that the fraction of "AIDS" patients who just happen to have HIV is essentially 100 percent, so we do not have to deal with "HIV-negative AIDS." Why, then, don't critics who don't want to talk about HIV tests use this stringent HIV-free definition and be done with it?

The answer, possibly, is that it would leave them with nothing to complain about. Critics want the existence of HIV-negative AIDS, and in order for this phenomenon, so dear to skeptics, to exist, "AIDS" must be defined by much less-strict criteria than we just proposed. Critics are always happy to do just that. As example, consider the 90 or so HIV-negative people in the United States who have been identified so far with CD4 counts less than 300 (not the critical 200 for AIDS, but approaching it). It was pointed out in REASON by the letter from CDC scientists (Dec.) that these people with "ICL" are epidemiologically different from AIDS sufferers, a statement which Thomas et al. dismiss as meaning "apparently that they are not linked to HIV." Not so. The problem is that ICL sufferers have no evidence of an acquired problem—they are not only HIV-negative but they are also not gay men, drug users, hemophiliacs, or transfusion recipients either, so their existence is as detrimental to the REASON authors' alternate AIDS-causation case as it is to the case for HIV.

There are similar problems with modifying the entire CDC AIDS definition by simply removing any mention of HIV—also a favorite ploy of critics. The problem is that HIV status in the CDC definitions is not there just to make skeptics unhappy and uphold the HIV theory; it also has a prognostic value for people with moderate immune problems and disease. If HIV-negative, such people may recover or at least get no worse over the long term, but if HIV-positive they invariably become more immunocompromised, and die within a few years.

Although Thomas et al. do not pause overlong to wonder why HIV is the only virus infection of moderately immunocompromised people that has such prognostic value for future total immune failure, they do admit the association. Certainly, then, they should understand why it is unfair to remove a criterion which is doing prognostic work in a disease definition, then make much of the fact that the resulting mutilated set of criteria no longer does what it was designed to do. It is this kind of thinking which allows critics to suggest that many HIV-negative people in the 1993 San Francisco study really had "undiagnosed AIDS" because they developed some of the milder "AIDS-defining diseases." Again, however, no one ever suggested that all diseases that are AIDS-associated secondary diseases are by themselves prognostic of future immune failure and death. The point of the San Francisco study was not to find which group(s) has any opportunistic-type diseases, but rather to see which group(s) inexorably suffers progressive immune function loss and high mortality, which is closer to what we all recognize as AIDS.

It is best to leave secondary diseases out of the definition of AIDS entirely if HIV status is also not included, yet critics persist in doing this, and may even go further. Duesberg has gone so far as to count many clinically healthy people with only mild immune abnormalities as "HIV-negative AIDS," when, contrary to what the REASON authors assert, such cases would not be called AIDS today, even if HIV-positive. Duesberg has gotten a lot of mileage out of his "HIV-negative AIDS" cases (as we see), and he chooses to make his AIDS definition broad enough that as many as possible of his "AIDS" cases don't have HIV. We, however, are certainly not obligated to do the same.

In summary, if we are not allowed to use HIV status, then defining AIDS stringently in terms of an acquired worsening immune failure (a CD4 count of less than 200) still preserves the prognostic boundaries of the fatal syndrome of AIDS long familiar to the man on the street. Thomas et al. may carp about problems with official CDC AIDS definitions, but they are actually carping about modifications of official definitions which they made themselves by subtracting HIV testing. Let them make better definitions, for surely there is something perverse about choosing a definition that is so overbroad as not to be useful, then blaming the establishment that it is so overbroad as not to be useful. The important fact for the rest of us is that people dying of immune failure for no other obvious reason in body-fluid-exposed groups are invariably HIV-infected.

7) I thought the Thomas/Mullis/Johnson article an unusually bitter one for REASON to be running, since its authors suggest not only that the U.S. government has failed with AIDS but that just about everyone else has failed to think properly as well. The authors essentially suggest that the scientific method itself failed when it has come to this problem—that instead, hundreds of independent working groups in public and private labs in scores of countries are even now participating in the same shared conspiracy or mass delusion. Supposedly, every major HIV study lab has been afraid to acknowledge openly that the emperor has no clothes. Does this really seem likely to the reader? I am strongly reminded of the conspiracy theories of creationists and UFO-coverup enthusiasts when I read this kind of thing.

Although I live in the world of institutional biomedical research, I don't find it to be the same conspiratorial one I read about in the Thomas/Mullis/Johnson article. Knowing how competitive biomedical researchers are, I prefer a different explanation for the present state of AIDS understanding—HIV infection just happens to be a really hard biological problem. However, progress is being made. According to recent news reports, a second controlled study has now confirmed that sterilized plasma from well HIV-positive people improves health and extends life considerably in AIDS patients. This is just as we might expect from knowing how HIV normally escapes its host's antibody immune response in AIDS, but yet might still be susceptible to anti-HIV antibodies transferred in plasma from someone else. Globulin protein (antibody) therapy has been tried in AIDS patients in the past, but was not very helpful, apparently since it contained no anti-HIV antibodies. I predict that the success of immunologic therapies targeted specifically against HIV in AIDS is shortly to make drug-induced AIDS theories even harder to believe.

Steven B. Harris, M.D.
Department of Pathology
UCLA School of Medicine
Los Angeles, CA

(Footnotes to Dr. Harris's letter are available upon request. Please send a stamped, self-addressed envelope to REASON's Los Angeles office. —The Editors)

Charles Thomas, Kary Mullis, and Phillip Johnson reply: To evaluate Dr. Harris's specific points, one must have in mind what the HIV hypothesis is and why it is in question. The hypothesis attributes extraordinary powers to the retrovirus HIV, powers that no other organism has ever been suspected of possessing. This virus multiplies rapidly after the initial infection until the immune system does its job and neutralizes the infection. HIV does no damage to the immune system while it is active in the body; rather, it is supposed to cause immune system damage years later, when it is latent and often very difficult even to find. HIV is also supposed to cause disease conditions, such as dementia, that are not a result of immune system damage, and is even held responsible for diseases like Kaposi's sarcoma which have been linked to other causes and occur in men who are HIV-infected and non-infected alike. Researchers have no idea how HIV accomplishes all these feats. The initial assumption that HIV directly attacks and kills T-cells has been discarded by many since HIV infects only a fraction of the cells it is supposed to be killing, but the direct-killing scenario is still pursued because all alternatives are speculative.

The proof that HIV causes AIDS is said to be correlation, but the correlation is mainly an artifact of the definition. AIDS is defined as one of 29 different disease conditions, or low T-cell counts, when accompanied by real or suspected HIV infection. The correlation is to antibodies and not to active virus, and the antibody tests generate many false positives.

In short, there are many problems with the HIV theory, and the problems have never been fairly investigated. Persons who are disposed to question the official theory have been treated as enemies to be ridiculed and punished. There are no truly independent researchers, since to question the HIV theory is to lose all research funding.

Papers dealing with all of Dr. Harris's points and many others will appear in the February issue of Genetica devoted to questions about the HIV/AIDS theory and its difficulties. For present purposes we can give only brief responses to his specific points, as follows:

1) As to gay men: The original AIDS-defining disease in gay men was Kaposi's sarcoma (KS). It is now known, however, that KS is found in many persons who have never been HIV-positive, and just about everybody agrees now that KS is not caused by HIV. A herpes virus is suspected by many, but drugs (especially poppers) are very much under study as a causative factor despite a press blackout on the subject. Even in cohorts of gay men with multiple health risks, authorities now admit that at least 5 percent (some say 10 percent or more) of antibody-positive subjects are AIDS-free 15 years after infection and show no signs of ever developing AIDS. Among persons who are otherwise healthy, the percentage of so-called non-progressors is undoubtedly much higher. The claimed perfect correlation between HIV and AIDS was established by biased studies that failed adequately to investigate alternatives. What is needed are unbiased studies that compare otherwise similar HIV-positive and HIV-negative persons from the general population; this has never been done.

2) Hemophiliacs: In the upcoming issue of Genetica, Peter Duesberg reviews all the studies on the subject and shows persuasively that the data do not support HIV causation. Life expectancy for hemophiliacs went up steadily following the HIV infection that occurred with the use of Factor VIII, and began to decline in the late 1980s when hemophiliacs began being "treated" with the toxic drug AZT.

3) Transfusion recipients: There are no studies comparing otherwise similar groups of HIV-positive and HIV-negative transfusion recipients. There is no "firm statistical link," only anecdotal evidence.

4) So-called SIV disease: After chimps infected with HIV failed to develop AIDS, HIV researchers turned to claims that a different virus (SIV) causes a different pattern of disease in monkeys. SIV is harmless in these animals in the wild but sometimes causes disease under laboratory conditions. The condition is unlike AIDS in the most important respect: HIV is supposed to destroy the immune system years after neutralization by antibodies, often when viral activity in the body is undetectable. SIV causes disease when it is present in quantity and active.

5) HIV-2: This virus has not been shown to cause AIDS. Most persons infected with it are healthy.

6) New definition: We heartily endorse Dr. Harris's proposal of a new definition of AIDS not presupposing HIV causation, and predict that use of proper definition would cause the claims of correlation to disappear. Incidentally, the KS sufferers originally diagnosed with AIDS in the early 1980s were not "otherwise healthy." They were mainly highly promiscuous, drug-using, bathhouse-frequenting gay men with a multitude of diseases.

7) Cures and treatments: The HIV theory has led to no health benefits. Claims of treatments just around the corner have been made again and again. They have always proved false.


In the March editorial, "Future Shock," the first name of the author of Complexity was inorrect. The book is by M. Mitchell Waldrop.