VIRUSMYTH HOMEPAGE



DEBATING AZT

Why the ‘AIDS Test’ is Useless, and Pathologists Agree

By Anthony Brink


15 March 2000

To the pathologist:

The Professional Provident Society requires me to take an HIV test for the purpose of increasing my life insurance. An ‘Informed Consent’ document supplied by the Life Offices Association invites me to ask you to explain its contents if I have any problems understanding it. I do have problems understanding it and I have several questions.

According to the face of the document, the test to be administered is an ELISA 3, which I understand to be a third-generation enzyme immunoassay for HIV antibodies. I wish to be informed of the name of the test kit employed and its manufacturer, and I require a copy of the operating/information booklet in order to inform myself fully about the test which I am obliged by my insurer to take.

Under the heading “Is the test always accurate? Can there be mistakes?” I am told that “the tests used are very accurate.” Even more categorical is the explanation under the heading “What does it mean if the test is positive?”: “this means that you have been infected with the AIDS virus.”

Does the mere presence of HIV antibodies in the absence of any clinical symptoms of illness signify an active infection with HIV? Are significant levels of such antibodies not consistent with a successful immune response? Are any other diseases diagnosed purely on the basis of antibody detection in the absence of clinical presentation?

I have looked up the specificity of four different third-generation ELISA HIV antibody test kits, and all claim specificity of about 99.8%.

Two senior medical technologists with the Natal Blood Transfusion Service tell me that the HIV seroprevalence among white people is this province is negligible and less than one in a thousand. I was told that the seroprevalence among Indian and Coloured people was likewise very small.

With a sensitivity of 100%, as all the test kits claim, the true positive in a thousand test subjects will be detected (allowing for present purposes one in a thousand ‘true positives’). With a specificity of 99.8%, two in a thousand non-infected test subjects will also register positive.

It follows that for every thousand people like me tested, there will be three reactive results, one true positive and two false positives. In other words, for people from my low- risk category in Natal-KwaZulu, HIV-positive test results will be wrong twice as often as they will be right. Am I right? If not, in what respect is my arithmetic unsound?

When I look at the specificity data for the antibody tests of the kind under discussion, I find no indication that any have been validated for specificity by comparing reactive results with confirmed viral infection in test subjects. In a pregnancy test for instance, the incidence of reactive urine tests would have been compared with actual confirmed pregnancies to determine sensitivity, and non-pregnant cases to establish specificity, that is the false-positivity rate. But looking at the scientific literature cited by the test kit manufacturers and other research papers, I find that this elementary control has never been performed for any HIV antibody test kit. Is there any reason why the specificity of HIV antibodies can’t be determined by comparing the incidence of reactive antibody test results with actual cases of confirmed HIV infection, ascertained by viral isolation in the suspected case?

I assume that we are agreed that viral infections can be directly confirmed by harvesting and dismantling putatively infected cells, by purifying and isolating the suspected virus by zonal ultracentrifugation into isopynic density gradients, electron photomicrography to confirm expected particle morphology, analysis of the proteins and nucleic acids of the purified particles to establish their exogenous origin, and confirmation of their infectivity by inoculation of virgin cell lines and then repetition of this procedure.

Can you refer me to any literature reporting that this has ever been done for HIV? Or am I correct in understanding from Abbott Laboratories’s statement, “there is no recognized standard for establishing the presence or absence of antibodies to HIV-1 and HIV-2 in human blood”, that HIV has never actually been isolated, and that no gold standard for the specificity of HIV antibody tests exists?

How does the claim in the informed consent form that “a positive test result means infection with the AIDS virus” square with Abbott’s warning, “All enzyme immunoassays…may yield non-specific reactions due to other causes” and therefore such results are required by Abbott to be “investigated further in supplemental tests”?

One of the test kit manuals that I have read states that the proteins employed as antigens by the test kit for the detection of HIV-1 antibodies are p24 and gp160. I assume that other HIV ELISA tests employ these same antigens, and/or p41 and its polymers, p80 and p120.

Have you any idea why p24 is described as an HIV-1 protein when Professor Luc Montagnier himself points out that p24 is not unique to HIV, and that it is also a constituent of HTLV-1 and HTLV-2 viruses as well as of endogenous retroviral sequences that form up to 2% of the human genome?

Since the glycoprotein with the molecular weight of 160 daltons is a polymer of p41, and Gallo has pointed out that Professor Luc Montagnier’s favoured ‘HIV-protein’ p41 is a ubiquitous cellular protein (which he now admits), can you explain why gp160 is described as an HIV protein? If the ‘co-discoverers of HIV’ are right, HIV antibody test kit reactivity to p24, p41, p80, p120 and p160 would represent no more than the detection of antibodies to cellular and other viral proteins from any number of sources, whether endogenous or exogenous.

What prevents HIV antibody test kits from lighting up to one or more of these non-HIV proteins?

I have difficulty understanding why ELISA HIV antibody test kit results need interpretation, and why reactivity or non-reactivity is determined not by reference to absolute on/off values but to a cut-off value on a continuum. In plain terms, if I am slightly reactive I am not infected, but if I am moderately strongly reactive I am. How can this be? If the proteins employed in the test as antigens are uniquely constituent of HIV-1, and HIV-1 antibodies are specific and monoclonal - the fundamental assumptions underlying HIV antibody testing - how can the test be reactive at all if I am not infected? How was this cut-off value fixed?

Under the heading “What is HIV?” I am told, “HIV is the virus that causes AIDS…” I have copies of and have studied Luc Montagnier’s 1983, and Gallo’s 1984 Science papers on LAV and HTLV-3 (now called HIV), and referred to as authority for this proposition by the test kit manufacturers, and I think you’ll concede that none come even close to establishing (a) that any virus was isolated under the well settled protocol for the purification and isolation of viruses, discussed at a symposium on this procedure at the Pasteur Institute in 1973, and (b) HIV-AIDS aetiology, except by weak reliance perhaps on the post hoc, ergo propter hoc fallacy that has so often has fooled medical researchers. Could you please refer me to any other literature that establishes HIV isolation by the Pasteur method, and the HIV-AIDS causality claimed in the ‘Informed Consent’ document. I believe his quest for such literature has occasioned some difficulty to Nobel laureate Kary Mullis Phd, inventor of the PCR technology adapted to your ‘HIV viral load’ tests. He complains that not even Luc Montagnier could refer him to any such literature, and that medical experts just ‘know’ HIV causes AIDS, just like they ‘knew’ bad air caused malaria. Because they ‘see’ it.

Under the heading “Is there a cure for HIV and AIDS?” I am informed that “there is no known cure” but that with careful management “you can greatly enhance the quality of your life before AIDS sets in.” Am I to understand from this that a person who is HIV-positive will invariably die a premature death from an AIDS indicator disease, and that his life will deteriorate even before such disease develops? If so, what research reports establish this?

What research reports establish that any of the licensed AIDS drugs improve quality of life? Isn’t it trite that they are all so poisonous and their ill-effects so severe that a very high proportion of patients are unable to comply with their treatment regimens and suffer dangerous toxicity injuries?

The ‘Informed Consent’ document restates a basic legal principle that persons urged to undergo any medical procedure are entitled to the fullest information about it, and that medical practitioners are required to supply it. Please consider this request for clarification and deal with my queries in the light of this. I reiterate my request for a copy of the information or operating manual supplied by the test kit manufacturer as I wish to study it closely myself.

Yours faithfully

ANTHONY
BRINK

Postea:

Pietermaritzburg pathologist, Dr Michael King, agreed unreservedly with my points made in paragraphs 1 and 3, and told me that pathologists have been conducting “a running battle with the Life Offices Association for years” regarding the sufficiency of the test as a basis for an HIV-positive diagnosis. At least five people preceded me for my ELISA test as I waited my turn including young black middle class folk who presumably lead not dissimilar lives and enjoy a similar healthy standard of living as their professional and business counterparts among the other ‘low risk’ races. None were alerted to the misinformation contained in the “Informed Consent” form that all were required to sign. Orthodox ‘AIDS expert’ Professor Gerald Stine of the University of North Florida made the same criticisms contained in paragraphs 1 and 3 above in AIDS UPDATE 1999 An Annual Overview of the Acquired Immune Deficiency Syndrome in his article The Performance Rate for the Combined Elisa and Western Blot HIV Test – Is 99% accuracy good enough? The Answer Is No. As the title tells, and we’ll discuss below, a follow up Western blot test doesn’t plug the holes.

Imagine my surprise then to see King asserting in the Natal Witness newspaper on 28 June 2000, Diagnosis of HIV highly specific: “A number of conditions have been described that can give positive HIV Elisa results... Fortunately, these false positives are uncommon and are excluded by the highly specific confirmatory tests… Occasional samples give indeterminate results on Western Blotting and further patient follow-up or testing with highly sensitive and specific nucleic amplification techniques (PCR) may be required. Despite the admission by mainstream medicine that occasional difficulties with diagnoses can occur, the serological diagnosis of HIV infection using the combination of enzyme immunoassays and Western Blotting is highly sensitive and specific (99%). Ref: Mandell: Principles and Practice of Infectious Diseases, 5th ed, 2000, Churchill Livingstone.” Roma locuta, ergo finita est!

Before we look at these “highly specific confirmatory tests”, you might be interested to learn that Lynn Morris of the National Institute for Virology told us at the second meeting of President Mbeki’s AIDS Advisory Panel in July 2000 that two reactive ELISA’s suffice for an HIV-positive diagnosis. You might wonder, “How can one unvalidated test possibly confirm another? To which another expert might offer the riposte, “We follow up with a different kind of test, the Western blot; it’s more specific.” Actually, the manufacturers of HIV Western blot tests do not make claims for better specificity than contemporary HIV ELISA kits. And in England and Wales, positive HIV ELISA test results are not confirmed or disconfirmed with an HIV Western blot test precisely because such tests are regarded by the ‘AIDS experts’ there as being too non-specific. The manual for one such HIV Western blot test (Epitope/Organon - Teknika Corporation) warns, “Do not use this kit as the sole basis of diagnosis of HIV-1 infection.” That’s how much confidence the manufacturer has in the specificity of its test. But don’t King’s “highly sensitive”, “highly specific” and “occasional” just roll off the tongue so nicely? No good upsetting the customers. Can’t have them thinking for themselves. Trust us. We know. Anyway Western blot is no different in principle from ELISA; it’s just that with Western blot antibody testing, you get to see which supposed ‘HIV proteins’ on the test strip react with the antibodies in your blood, whereas with ELISA the proteins are served mixed. Both kinds of tests presuppose that the test proteins have been shown to be uniquely constituent of a virus called HIV. But that’s not true. Quite the opposite in fact. It gets worse. Western blot test results for ‘HIV antibodies’ are interpreted differently in different places, kit to kit, lab to lab, country to country. By these different diagnostic criteria, you will be ‘infected with the AIDS virus’ and doomed to die in this country but not that. According to one pathologist but not another. What an incredible mess.

Some really clever guys like Dr William Makgoba, president of the Medical Research Council, puff the sophisticated technology of modern ELISA HIV antibody tests by treating you to a little lesson on the purity of the proteins used in them as antigens to fish for the presence of ‘HIV antibodies’. “They don’t use purified proteins anymore”, he lectured us at the AIDS Panel’s second meeting. “They use recombinant proteins now.” That big drop-dead word is sure to impress, until your thoughts stray and you wonder, “What is the point of producing magnificently pure proteins, all with precisely the same molecular weight by means of bio-engineering techniques before ascertaining whether such proteins are unique to HIV?”

King’s statement that one can confirm or disconfirm HIV infection with “highly sensitive and specific nucleic amplification techniques (PCR)” will be a shocker to anybody who has read the contrary admonitions by the manufacturers of such tests. Makgoba spoke the same way in an interview in Focus in June 2000: “I have every confidence that the antibody test is so specific now that we don’t get many false positives. And if you take that with the identification of the virus by DNA techniques, there will be an abundance of correlative results.”

The only HIV PCR test licensed by the FDA for clinical (as opposed to experimental) use by pathologists is Roche Diagnostics Corporation’s AMPLICOR HIV-1 MONITOR Test, version 1.5. The manual says: “The AMPLICOR HIV-1 MONITOR Test, version 1.5 is not intended to be used as a screening test for HIV-1 or as a diagnostic test to confirm the presence of HIV-1 infection.” That’s because the manufacturer recognises that it is not specific enough. No, no, the ‘AIDS expert’ points out. That’s the wrong kind of PCR test. We don’t use quantitative monitoring tests for diagnosing HIV infection; we use a qualitative test. Like Roche Diagnostics Corporation’s other PCR test, their AMPLICOR HIV-1 Test. Well, it would help if the ‘AIDS experts’ read the manual: “For research use only. Not for use in diagnostic procedures.” As for “an abundance of correlative results” between HIV PCR and HIV antibody tests, in the only comparative study of its type yet performed - reported in AIDS in 1992 by the Multicenter Quality Control of PCR Detection of HIV DNA - the concordance of reactive results when the same blood was tested with both kinds of tests ranged unpredictably, hit and miss, between 40% and 100%. Odd isn’t it?

Dr King relies on a textbook for his statement that “the serological diagnosis of HIV infection using the combination of enzyme immunoassays and Western Blotting is highly sensitive and specific (99%).” All I can think is that by the time he wrote that, he had forgotten our little chat - specifically our discussion of the Grand Canyon of a difference between specificity and reliability in a low sero-prevalence cohort.

Let’s have a closer look at the significance or otherwise of King’s “99%” specificity figure. I learned that the specificity of the test used on me - an Abbott HIV gO EIA - was claimed to be 99.8%. Just how little such a specificity figure really means is well set out by Christine Maggiore in Los Angeles in a letter she wrote at the end of May 2000 to the webmaster of an AIDS information website: “The fact is that the specificity and the accuracy of HIV tests were determined by assuming that 100% of people with AIDS-defining illnesses who tested positive had actual current infection with HIV. The specificity was established by assuming that 100% of symptomless blood donors who tested HIV-negative did not have a current infection with HIV.”

Abbott Laboratories’s HIVABtm HIV-1 EIA test manual tells how the ‘specificity’ of the test was established: “Sensitivity and Specificity: At present there is no recognized standard for establishing the presence and absence of HIV-1 antibody in human blood. Therefore sensitivity was computed based on the clinical diagnosis of AIDS and specificity based on random donors. The ABBOT studies show that:

Sensitivity based on an assumed 100% prevalence of HIV-1 antibody in AIDS patients is estimated to be 100% (144 patients tested).

Specificity based on an assumed zero prevalence of HIV-1 antibody in random donors is estimated to be 99.9% (4777 random donors tested).”

The stunning implications of this are highlighted when we recall our pregnancy test illustration. The test gets tried out on 1000 women chosen because they are plump around the middle. They are presumed pregnant because they are tubby. Nobody thinks to establish by means of a scan whether they are actually pregnant. Then the test is tried out on 1000 slender women. They are presumed not to be pregnant because they have flat tummies. Nobody ascertains whether any are in their first terms of pregnancy. The test reacts for all the big-bellied women, and on this basis is declared 100% sensitive for pregnancy. It reacts for only two of the slim women and so gets declared 99.8% specific. Were such junk to be marketed for pregnancy testing, think how women’s groups would freak out. Can you just imagine?

Suppose that after well over a decade of use of this test it just coincidentally entered the heads of two independent teams of researchers on separate continents each to do a scan on one of these plump women who light up the test, and to publish their photographs in their leading trade rag. Imagine if the photograph showed nothing that looked like a foetus, in size and shape, bearing in mind how foetuses look through these scanning devices. The analogy is not as wild as one might think. Australian medical physicist Eleni Papadopulos-Eleopulos and her colleagues at the Royal Perth Hospital tells what happened when two separate teams of researchers went looking for HIV in the preparations of what they thought would be masses of concentrated, purified retroviral particles (Virology, March 1997)(*). And the astonishing concession made in the same year by the ‘discoverer of HIV’, Dr Luc Montagnier of the Pasteur Institute, concerning why he never published any electron photomicrograph of purified virus when making his claim to having isolated HIV (then called LAV) in 1983(#). Papadopulos-Eleopulos’s collated papers – all published in fine journals – are archived on the www.virusmyth.com/aids/perthgroup/ website.

When we leave our pregnancy test analogy and return to ‘HIV antibody tests’, the tale curdles even more. What if ‘AIDS defining illnesses’ in the absence of ‘HIV infection’ frequently cause the ‘HIV antibody test’ to react as well? Like the state of being plump setting off a pregnancy test. Such as the state of being thin lighting up an HIV antibody test. It does, actually – simple malnutrition is a reported cause of ‘false-positives’. As is tuberculosis. About seventy other conditions too, amply documented in the medical literature from ‘flu through to malaria. That’s the problem: ‘HIV antibody tests’ have never been validated against confirmed infection, and what’s more, just about anything can set them off. It’s something the ‘AIDS experts’ never get into. The manual for the test kit used on me rightly concedes, “False positive test results can be expected with a test of this nature” - contradicting the ‘Informed Consent’ form on the meaning of a reactive result: “What does it mean if the test is positive?”: “this means that you have been infected with the AIDS virus.”

Dr Desmond Martin wrote an article on the subject of ‘HIV diagnosis’ for the January 2000 issue of the South African Medical Journal. Reading it, you’d think you were in good hands going for an ‘HIV test’. That these guys know what they are doing. That their expert pronouncements on the state of your health can be confidently relied on. That they are cleverer than mediaeval doctors who wrote up an elaborate body of arcane learning on the exquisite variety of diagnostic meanings that could be pegged, for a fee, upon the qualities of your urine - its taste, colour, scent, density, viscosity and so on.

King was unable to answer or ducked the rest of my questions (in fact I had researched and knew the answers already) and referred me to the National Institute for Virology, university virology departments, and to an outfit called TOGA Laboratories, where Dr Desmond Martin currently makes his living. As far as the first two went, I’m afraid it was a case of ‘been there, done that’. Without any luck. None at all. The quality of my exchanges with ‘the experts’ at these places would bring tears to your eyes. University of Durban Virology Professor Alan Smith’s article on ‘HIV testing’ published alongside Dr King’s in the Natal Witness on 28 June 2000 is a good example of the brown-outs you encounter when ‘AIDS experts’ get asked simple questions at the root of this business. I didn’t bother Dr Martin or Dr Sim at TOGA Laboratories. Can you blame me? Nor did I go to the Life Offices Association again. I had approached their Medical Underwriters Committee before. They didn’t know what I was talking about. It all went right over their heads. Meant nothing to the Natal Blood Transfusion Service’s chief medical technologist, Dr Ravi Reddy, either, so I thought it would be pointless asking him: Why should race rather than class and environmental factors predispose one to contracting an infectious disease? (Are blacks really hornier than whites?) How can you determine the seroprevalence (infection rate) in a given community with an indirect (antibody) test before you have established the specificity of the test - by comparing how closely its performance (reactive, non-reactive) matches the incidence of infection (pathogen directly isolated in the patient)? Because until you do this, you’re just chasing your tail.

Think of an antibody test as detecting the fire fighting service out on a call. Usually to put out fires. But also to rescue kittens from trees. Or coax suicide jumpers away from high ledges. Or free drivers pinned inside their crashed cars. Apart from all this, there is an additional problem with relying on antibody tests as the sole basis for a diagnosis of infection: antibodies are often more partial to antigens other than those that stimulated their production. The assumption is that antibodies are specific, like faithful spouses. But as we all know, some husbands prefer their girlfriends to their wives. In short, antibodies are generally poly- not monoclonal. They are faithless partners. So you can’t just assume that a reactive antibody test indicates infection with a particular bug. You have to establish the specificity of the test first. Properly. Not in the asinine manner in which the HIV antibody test kit manufacturers have done. Imagine just assuming that a person lying in a hospital bed is ‘infected with HIV’ and has AIDS, just because he has one of the age-old diseases arbitrarily pulled under the CDC’s ever expanding bureaucratic umbrella as an AIDS indicator disease, and because he is an inner-city queer, black, or junkie – so in a ‘risk group’. Maybe just socially unpopular, marginalised and poor. Just the kind of person to feed AZT.

Why the blood of the impoverished black Africans (as opposed to the black middle classes and elites) makes HIV antibody tests light up like Christmas trees is a matter elucidated for the scientifically intrepid in papers that can be read on the internet: AIDS in Africa: distinguishing fact and fiction (World Journal of Microbiology & Biotechnology (1995) Vol. 11), Is a positive Western blot proof of HIV infection? (Bio/Technology June 1993, Vol. 11), HIV antibodies: further questions and a plea for clarification (Current Medical Research and Opinion Vol. 13: 1997), and HIV Antibody Tests and Viral Load - More Unanswered Questions and a Further Plea for Clarification (Current Medical Research and Opinion Vol. 14: 1998) all by Papadopulos-Eleopulos et al and archived at the website mentioned above. Frankly, after these papers, anybody who tells you that a positive result to an ‘HIV antibody’ test means that you are infected with a deadly virus is, to quote John Lauritsen, “either ignorant, lazy or stupid.”

The ‘three or four million South Africans infected’ figure, which drives the hysteria in this country and elicits funds galore for AIDS careerists, is based on the extrapolation of anonymous HIV antibody test results of mostly poor black pregnant women at antenatal clinics. Unfortunately ‘AIDS experts’ haven’t thought to figure into their thrilling sums the fact that past pregnancy itself is a documented cause of ‘false positives’, reported in five separate research papers. And warned against by Abbott. Or, messing up the sums even more, that HIV infectivity is eight times lower for men than women according to top ‘AIDS experts’ (Padian et al 1997) - a curious notion for an allegedly sexually transmitted disease, but then HIV-AIDS is a curious affair. Whose mounting anomalies need interminable excuses, like that other rotting paradigm in its death throes, Ptolemy’s geocentric model of planetary motion, adjusted ad hoc to answer every Copernican challenge, until it all became just too ridiculous, and the whole thing finally collapsed, vehemently defended by the experts to the end.

If you are beginning to suspect that ‘HIV antibody’ testing is nothing more than a vicious form of high tech mumbo jumbo, bone throwing, divination, and death spell casting, with modern witchdoctors keeping suckers like us terrified and in their power - and their pockets full - I should emphasise that this little essay only scratches the surface. In her papers to which I have referred above, Eleni Papadopulos-Eleopulos and her colleagues take ‘HIV antibody testing’ comprehensively to task. And blow it to smithereens.

This much is certain: HIV antibody test results are no more significant an indication of health or disease than a phrenologist’s skull-chart. They’re worth a bowl of cold spit. But while they shatter countless lives they sure rake in the cash. And the Life Offices Association’s ‘Informed Consent’ form for HIV tests creates litigation possibilities for psychic trauma claims enough to keep lawyers in business for years.

(*)  www.deltav.apana.org.au/~vturner/aids

(#)  http://www.virusmyth.com/aids/data/dtinterviewlm.htm


CONTENTS DEBATING AZT


VIRUSMYTH HOMEPAGE