See what one wants... 7 April 2005
Previous Rapid Response Next Rapid Response Top
Nicholas Bennett,
Infectious Disease Postdoc/Clinician
Department of Pediatrics, University Hospital, Syracuse NY

Send response to journal:
Re: See what one wants...

It appears as if my point was made rather too obviously.

I chose to give examples of high-risk cohorts specifically because of the simple fact that none of the HIV-negative members demonstrated a chronic decline in immune function, or more specifically a chronic yearly loss of CD4 T cells.

Mr Tyler asks "However, how could [Ascher] have known [drugs had no effect] when all 'HIV' positives were gay men, and according to Table 2, 100% of these gay men were also either 'heavy' or 'light' nitrite users."

Firstly, as I think Ascher made clear in followup correspondance, the "light" users did included a number of non-users. Secondly the light users actually had worse CD4 T cell counts, as a group, compared to heavy users, in direct contradiction to the drug-AIDS hypothesis. Thirdly, epidemiology allows one to consider individual factors one at a time with the factor-less group acting as a control. Only those with HIV had losses in their CD4 counts.

The simple reason why "heavy" popper usage correlated with higher rates of KS is as I've explained before, they have a far higher risk of acquiring HHV8 due to the association with popper use and anal sex. The relationship is not poppers ->KS but rather poppers -> anal sex - > HHV8 -> KS. The missing link fits all the cases, including those that aren't related to poppers, and therefore is a better explanation.

The reader should be reminded that lifestyle factors were considered as a cause for AIDS until the point where low-risk individuals were found to have AIDS - these included heterosexual partners of IV drug users and the hemophiliacs. At that time a blood-borne sexually transmitted infection was considered - the epidemiology simply didn't fit a lifestyle cause. That's not to say that non-HIV factors can't contribute (e.g. high risk sexual practises increasing the risk of infection) but they themselves are not sufficient nor necessary.

The longitudinal studies are the most striking, because unlike a chronic exposure, these demonstrate that individuals _start_ to lose their CD4 T cells, after several years of normal levels, coincidentally at the same time as they start to produce antibodies to HIV. One must presume that the two events are linked.

For sure, some HIV-negative people will get AIDS-indicator diseases, in the same way as some non-smokers get lung cancer. This is why epidemiology teaches us the merits of Relative Risk. When deciding whether a factor is or is not contributary to an outcome we judge how frequent the outcome is with and without the risk factors. When controlled for HIV status, drug use had no effect on whether or not someone got AIDS. When controlled for drug use, HIV status had a large effect (the only independant risk factor). Mr Tyler quotes some studies apparently showing decline in CD4 T cell counts, but a count of 800 is still in the normal range! Can he provide examples of HIV-negative people with CD4 counts of 200 or lower? Also if an immune system is damaged by some other factor (no-one is denying that it can happen) then an infection with a virus will of course be at a higher rate. I'm not surprised that people with lower CD4 counts prior to infection had a higher risk of acquiring HIV - it is well established that immune responses can protect against HIV infection even with multiple exposures (e.g. Kaul et al from 2000 and 2001)

I wonder though why Mr Tyler quotes from Des Jarlais but omits the following:

"We studied CD4 cell counts and percentages from 1984 to 1992 among 1,246 HIV-seronegative injecting drug users in New York City, a population at very high risk for exposure to bloodborne pathogens. Severe CD4 lymphocytopenia was rare, and there was no evidence of an increase over time. Of 229 subjects with longitudinal data, only four met the surveillance definition for "idiopathic CD4 lymphocytopenia" (ICL)."

Hardly glowing evidence of support for the drug-AIDS hypothesis!

Marion et al don't actually demonstrate any immune deficiency at all, aside from lack of responses to DNCB (a chemical that can cause skin reactions). Responses to TB protein, candida and trichopyhton were all normal.

Lang et al actually say:

"The three groups were 37 HIV seroconverters, 304 prevalent HIV seropositives remaining free of the acquired immunodeficiency syndrome (AIDS), and 69 men who developed AIDS during observation. Six months before seroconversion, CD4 levels were similar among HIV seroconverters and 356 seronegative controls. Within 18 months of seroconversion, mean CD4 levels fell to the level of the prevalent seropositives at study entry."

I do not think these are very good arguments for the drug-AIDS hypothesis. Lang et al actually is excellent evidence for HIV infection leading to immune failure.

The question dissidents have to answer is why these antibodies so effectively predict the progression to fulminant immune failure. Why don't HIV-negative people with the same risk factors have the same immune decline? They have to answer why these antibodies are associated with standard assays of viral culture and PCR which cannot detect anything in matched control subjects. They have to ask why they have to propose multiple causes which make little biological sense to explain all of these findings, when one (a retroviral infection) rather neatly explains them all!

As regards the two women in question, the average time of progression in untreated individuals is 8-10 years. They are still within normal variation - they won't even qualify for long-term non-progressor status (of which around 5% of HIV+ people will do) for another 7 years (the year 2012). This has been well known for over a decade, but the dissident views are based upon Duesberg's comments along the lines of "Infections do not cause illness after 10 years". If only that were true! Besides, Ms Maggiore has said that she has also had HIV-negative and HIV-indeterminate tests, so frankly one wonders which to believe.

Sadly I haven't the time to address the important points raised about the viral load techniques, but judging by the quality of the points I have got around to addressing so far, I do not think that there is much substance to the concerns.

Nick Bennett njb35@cantab.net

Competing interests: None declared