Re: "HIV", HHV-8 AND KS 21 August 2004
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: Re: "HIV", HHV-8 AND KS

The Perth Group really are good at demonstrating their lack of appreciation of the current literature, not to mention reading of my previous posts!

I described the increased risks in my post entitled "KS risks" (now there's a clue for you) of July 21st. There is clearly a many-thousand fold relative risk of KS in those who are HIV+, and this is skewed such that those who acquired HIV through homosexual sex are more likely to have it (in complete agreement with a parallel infectious agent). If I and my friends and colleagues can provide the references to support this in a matter of a few minutes, one wonders what the Perth Group have been doing with their time.

There is huge evidence that HHV8 is found in 95-100% of KS cases, of all types [1-3]. The references provided are merely a smattering of those that exist, chosen at random. As with HPV and cervical cancer, the senstivity of detection depends on how hard you look - without detracting from the specificity. If they would attend a single infectious disease discussion regarding HHV8 at their local hospital or university they would learn all this and more in about an hour.

The Perth Group are quite right that the correlation alone cannot say that HHV8 causes KS. Clearly it would help if seroconversion preceded KS disease but was not associated with other diseases [4] or HHV8 contained cancer-promoting genes [5,6,7]. Hmmm, looks pretty good to me. Incidentally the seropostive rates in reference [4] clearly show how "popper" use in homosexuals might confound the higher incidence of KS seen in that group: "Homosexual men were more likely to be HHV8 positive (61.4%) than individuals who were infected with HIV through heterosexual contact (21.4%) and intravenous drug users (15.0%)."

I greatly question whether nitrites and semen are carcinogenic for KS. HHV8 is present in 95-100% of cases whereas the other two factors are not (okay, semen is 100% associated but then so is water, and I give semen about the same carcinogenic properties!). Seroconversion to HHV8 precedes KS. Antibody titer to HHV8 actually predicts rate of progression! Rare cases of demonstrable KS without HHV8 antibodies or DNA might indeed be due to other causes (KS being only a clinical/histological definition). Paper [4] below did indeed include a single case of KS in the HHV8 "seronegative" group although he seroconverted to HHV8 about 1 year prior to KS appearing. Whether nitrites contribute to KS is another question, but they are clearly not necessary whereas HHV8 has a much stronger case.

In my post entitled "Re: Where is the infectious molecular clone" of the 14th July I clearly spelled out for the benefit of the Perth Group and other lay readers what a molecular clone was, how it was obtained for HIV and how we know it belongs to an exogenous retrovirus. I also showed that people who test positive for HIV by antibodies also grow HIV and/or have HIV genetic material present in 100% of cases (2 cases in the series I presented were positive by culture but not PCR and vice versa - all AIDS cases were positive for both, no seronegative controls were positive by either). This, along with other evidence, is more than ample to convince the vast majority of the scientific establishment that HIV+ people are infected with HIV. The same criteria applied to any other infection would suffice!

I used 19 peer-reviewed and widely accepted references in the article, and I was under the apparently misguided impression that reading it and them might be educational for the Perth Group.

Through their repeated and circular responses without recognising the facts as presented to them, the Perth Group are clearly wasting the time of ourselves as well as those of the BMJ Rapid Responses editorial team. I echo Chris Noble's response by asking whether the Perth Group are willing to admit that they were misleading when they said that Montangier didn't properly characterise RT in the original "isolation" paper.

Their lack of honesty and the fact that their work is clearly influencing a small number of HIV-infected and uninfected people is potentially putting lives at risk.

As regards two of their questions to Chris Noble:

I have said REPEATEDLY that there is endogenous RT activity in cells, stimulated or not. However the Perth Group seem unable to recognise the principle of RELATIVE amounts. HIV infected cells produce many-fold higher amounts of RT activity compared to identically stimulated cells, in addition to viral particles and other specific proteins. To suggest that the same scientists who discovered endogenous RT activity would then ignore it says a lot of the Perth Group's opinion of their own ability to distinguish scientific fact compared the original researchers. It seems exceedingly arrogant to imply that they have somehow thought of something that the actual researchers failed to consider!

Secondly KS is still an AIDS indicator disease because it is vastly (greater than 10,000 fold) more common in HIV+ patients than normal controls. It is an indicator of immune deficiency and a predictor of future opportunistic infections and death (from OIs or KS itself). KS is perhaps one of the best cases to make for an AIDS indicator disease! The fact that the Perth Group still fail to grasp this speaks volumes.


1. Buonaguro et al Int J Cancer. 1996 Jan 3;65(1):25-8. Herpesvirus- like DNA sequences detected in endemic, classic, iatrogenic and epidemic Kaposi's sarcoma (KS) biopsies.

2. Schalling et al Nat Med. 1995 Jul;1(7):707-8. A role for a new herpes virus (KSHV) in different forms of Kaposi's sarcoma.

3. Cathomas et al. J Clin Pathol. 1996 Aug;49(8):631-3. Detection of herpesvirus-like DNA by nested PCR on archival skin biopsy specimens of various forms of Kaposi sarcoma.

4. Rezza et al. J Natl Cancer Inst. 1999 Sep 1;91(17):1468-74. Human herpesvirus 8 seropositivity and risk of Kaposi's sarcoma and other acquired immunodeficiency syndrome-related diseases.

5. Sturzl et al. J Natl Cancer Inst. 1999 Oct 20;91(20):1725-33. Expression of K13/v-FLIP gene of human herpesvirus 8 and apoptosis in Kaposi's sarcoma spindle cells.

6. Yang et al. J Exp Med. 2000 Feb 7;191(3):445-54. Transgenic expression of the chemokine receptor encoded by human herpesvirus 8 induces an angioproliferative disease resembling Kaposi's sarcoma.

7. Li et al. J Virol. 1997 Mar;71(3):1984-91. Kaposi's sarcoma- associated herpesvirus encodes a functional cyclin.

Competing interests: None declared