Infectious Disease Postdoc/Clinician
Department of Pediatrics, University Hospital, Syracuse NY
Send response to journal:
My comment on the Montagnier graph was simply that after the message was posted the html-coding on the Rapid Responses made the page unreadable for myself with both Firefox and Internet Explorer. This was fixed after a day. The pun on my opinion of the interpretation may have been lost.
The graph is entirely accurate in representing a standard individual's laboratory history during infection. However the Perth Group were trying to argue that this showed no relation between viral load and CD4 T cell rate of decline. This graph is not able to add or detract from that argument, since it doesn't address that point. The Mellors analysis of over 1600 people, comparing their viral loads with rate of CD4 T cell decline, is the kind of work required to refute or support the idea.
The KS association with gays in the early part of the epidemic is to me easily explained by the co-epidemic of HHV8 infection in that sexually- distinct group. KS is also seen at higher CD4 counts than other AIDS- OI's, hence its vastly increased risk of occurance. A lower level of immune suppression is required for the virus to cause clinically apparent disease. Of course KS can occur with "normal" CD4 counts, in say those on anti-rejection chemotherapy, but it's all about relative risk. In such a multifactorial disease, no clear-cut on and off "causative" factor can be expected. There are three other types of KS separate from the more aggressive AIDS-KS, one of which was known for some time as an indolent disease of elderly Mediterraneans. Here the only risk factor, aside from demonstrable HHV8 infection, is apparently merely old-age. This is a form of immune-suppression in that the immune system is less responsive, but I don't think anyone would accept it as such a dire state as AIDS.
I've previously noted the relatively low rate in Thailand, which I agree is quite different from the Western situation. I agree that something else, some other risk/protective factor is at work, but no solid evidence is out there to suggest what that is.
As regards CD4 counts, it's well known that the initial response to therapy is due to redistribution from lymph nodes. In this situation the link between viral load and CD4 count may well be uncoupled. This does appear especially interesting when a non-antiviral approach is used, such as antioxidants. It does not detract from the _fact_ that in over 1600 people (in that one Mellors paper) the level of viral load predicted their rate of CD4 T cell loss. Other groups have found similar results.
The data by Millazo and Levy are letters which I cannot access online, but this abstract appears to directly contradict it:
Puro and Ippolito J Acquir Immune Defic Syndr. 2000 Aug 15;24(5):440- 3. "Brief report: effect of antiretroviral agents on T-lymphocyte subset counts in healthy HIV-negative individuals.
"A slight increase in mean CD4 (range, 4.8%-6. 7%) and CD8 (range, 1.4%-9.3%) cells/mm3 was observed in each group."
"These findings seem to reject any direct effects of antiretroviral agents, independent of retroviral inhibition, on proliferation and redistribution of T lymphocytes, as well as the hypothesized braking of lymphocyte apoptosis. The observed variations could reflect biologic variability."
CD4 counts do vary considerably, which is why no decent clinician should use a single slightly low CD4 count as a basis for diagnosis of AIDS or initiation of therapy. The normal range is 500 to 2000 per ul, and the average CD4 count immediately post-seroconversion with HIV is well within this range at 700 (the average uninfected person in the same cohorts, or the same person prior to seroconversion, has counts around 1000 per ul). However I've read reports of KS appearing with CD4 counts as high as 600 - in contrast to most AIDS diseases which are only at a significantly increased risk at counts below 200.
It seems relatively clear though that a longer-term increase in CD4 counts occurs, which is combined with a return of previously lost immune responses [1-4]. The phrase "chronic immune activation" appears more than once it seems, supporting the theory that it is the HIV infection driving the abberations in T cell function and number, and that controlling the infection relieves the effect. That of course is not to say that it is the only possible way to assist the immune system. In the same way as one can prevent heart disease by blocking noradrenaline, Angiotensin II and HMG CoA reductase, so it may be that preventing AIDS will also turn out to be best approached by multiple paths.
But of course we're back with Mr Whitehead's call for further formal research into alternative/complimentary therapies aside from the current antiretrovirals. There is a "middle way" perhaps, depending on your point of view, but that middle way is not going be found by people such as the Perth Group arguing points using misinterpreted data. Such behaviour is potentially detrimental to any kind of sensible discussion regarding non- antiviral approaches to managing HIV infection.
Nick Bennett firstname.lastname@example.org
1. Pahwa et al. AIDS Res Hum Retroviruses. 2003 Jun;19(6):487-95. "CD4+ and CD8+ T cell receptor repertoire perturbations with normal levels of T cell receptor excision circles in HIV-infected, therapy-naive adolescents."
2. Kostense et al AIDS. 1998 Dec 24;12(18):F235-40. "Diversity of the T-cell receptor BV repertoire in HIV-1-infected patients reflects the biphasic CD4+ T-cell repopulation kinetics during highly active antiretroviral therapy."
3. Giovannetti et al Clin Exp Immunol. 2001 Apr;124(1):21-31. "T cell responses to highly active antiretroviral therapy defined by chemokine receptors expression, cytokine production, T cell receptor repertoire and anti-HIV T-lymphocyte activity."
4. Hainaut et al Pediatr Infect Dis J. 2003 Jan;22(1):62-9. "Age- related immune reconstitution during highly active antiretroviral therapy in human immunodeficiency virus type 1-infected children."
Competing interests: None declared