More responses to Christopher Noble 24 February 2004
Previous Rapid Response Next Rapid Response Top
Eleni Papadopulos-Eleopulos,
Department of Medical Physics, Royal Perth Hospital, Western Australia,
Valendar F Turner, John Papadimitriou, Barry Page, David Causer, Helman Alfonso, Sam Mhlongo, Todd Miller, Andrew Maniotis, Christian Fiala

Send response to journal:
Re: More responses to Christopher Noble

More responses to Christopher Noble

In his rapid response "Re: Still no proof for a correlation between the HIV PCR and antibody tests", 11th February, Christopher Noble wrote: "The Perth Group write "Nowhere in the text of the Owens et al paper do the authors confirm or even mention whether any of the studies in their analysis included controls from a "non-diseased population", "at risk for HIV infection". Under the subtitle, "Recommendations for Study Design" they wrote: "Many of the studies we analysed had design limitations?incomplete representation of the spectrum of patients in the study population, insufficient sample size, and incomplete reporting of test results. To increase the generalisability of study results, the study sample should reflect the entire spectrum of disease encountered in the clinical population of interest. For example, the nondiseased population should include persons who are at risk for HIV infection and would be candidates for testing rather than healthy controls".

What would be Christopher Noble's interpretation of the above?"

The above passage means exactly what it says no more and no less. If The Perth Group could be bothered to read the studies that Owens et al used in their metastudy they would indeed find that many of the studies did include non-diseased (not HIV infected) groups that were in high risk groups. The correlation between HIV antibody tests and HIV PCR tests remained.


Let us repeat our request once again:


1. Of the "many" studies could Christopher Noble please give us one which included "non-diseased (not HIV infected) groups that were in high risk groups" and which satisfied all the criteria set by Owens et al.
2. One study which included non-HIV/AIDS individuals but who nonetheless were sick and which satisfied all the criteria set by Owens et al.


Christopher Noble wrote: "It is extremely obvious that it is impossible to "prove" anything to someone that will never accept any evidence that contradicts their dogmatic beliefs.

The problem is not that we won't "accept any evidence that contradicts" our "dogmatic beliefs" but that we have not been given any evidence.  No amount of convincing and references (including Owens et al) would make us accept Christopher Noble's claim that the PCR and the antibody tests are highly specific and that a correlation exists between the "HIV" PCR and antibody tests if he does not give us a reference which satisfies some basic scientific principles.  These include randomisation, blind performance and the use of proper controls and a gold standard.  One does not need to be a scientist to realise that any study which does not satisfy these conditions would be flawed.


Christopher Noble wrote: “It is fruitless and pointless to even attempt to prove anything to the Perth Group. Any paper that presents evidence that contradicts their hypotheses will be deconstructed and "flaws" will be found. "


A response to this accusation can be found in the rapid response: "That is a Scientist’s Responsibility", 19th February by Mark Bartlett.  We thank him for his support and cannot improve upon his answer.

Christopher Noble wrote: "The Perth Group write "If a positive HIV-2 antibody test was proof for HIV-2 infection and if the primers were amplifying the env gene, then: (i) each primer pair should have given positive results with all 42 samples; (ii) all samples should have tested positive with all primer pairs."

Hughh. What are you talking about? You have previously drawn our attention to the genetic diversity of HIV. Now you expect one PCR primer to amplify all strains of HIV-2. The only thing that your response demonstrates is your determination to ignore any evidence that contradicts your hypotheses.


It is true that we have drawn Christopher Noble's attention to the genetic diversity of "HIV-1". It was Christopher Noble who, all along in this debate, tried to convince us that a virus can vary by more than 50% at the nucleic acid level and by more than 81% at the amino acid level and:

1. it would still be possible to define "HIV-1" infection in molecular terms.
2. "HIV-1" infection could be diagnosed using the same antibody test. (Although he claimed that different tests are used to prove infection with "HIV-1" group 0 he could not provide such evidence).
3. the proteins would still perform the same function.
4. it would still be possible to develop a vaccine.

Our view was, and still is, the opposite.

In particular we have claimed that, given the genetic diversity of "HIV-1", it would not be possible to define "HIV" infection in molecular terms, that is by hybridisation and PCR.  The Damond et al findings support our claim. Although they used a small number of patients (42) and 5 primers they could not detect the "HIV-2" env gene in all the 42 "HIV-2" positive individuals.  To detect the env gene in all "HIV-2" positive individuals many more primers would be necessary.  If similar numbers of primers are necessary for the detection of all the other genes the detection of the whole "HIV-2" genome would require the use of a very large number of primers.

Given that the "HIV-1" genome is more diverse than that of "HIV-2" and that many more people are infected with "HIV-1" the number of primers needed to detect the whole "HIV-1" genome in all "infected" individuals, that is, to define "HIV-1” infection in molecular terms in these individuals, would be prohibitory. 

Christopher Noble wrote: "The Perth Group write: "However, after repeated questioning he [Montagnier] admitted that reverse transcriptive activity is not specific to retroviruses and thus, indirectly, that he did not isolate HIV"

Montagnier has not, directly or indirectly, admitted that he did not isolate HIV. This is purely your spin, your interpretation and your lies.

A response to this accusation can be found in the rapid response: "Re: Re: Still no proof for a correlation between the HIV PCR and antibody tests", 14th February by Murali Mohan.  We thank him for his response and are unable to offer any improvement to his reply.

Christopher Noble wrote: "The Perth Group write: "Repeat, although maybe not to his liking, we have honestly answered all Christopher Noble's questions. If he wishes he may repeat them and we will answer them again."

Half a year ago I asked you to provide a citation for your claim that "the genomes of the most variable RNA viruses do not differ by more than 1%". So far you have not done so. You have repeatedly cited a paper that says nothing of the kind.


In the study we have cited the authors even gave several reasons, based on sound biological principles, why the variability of the viral genomes cannot be much larger than 1%.  In fact, to avoid any misinterpretation we have repeatedly quoted these authors.


Christopher Noble wrote: "It is imperative in science that you accurately report the contents and findings of the papers you cite.

Once again you do exactly the opposite.

You write: "The largest, longest, best designed and executed studies conducted in the USA and Africa show that HIV is not heterosexually transmitted.24-27"

The first reference you give is 24. Gray RH, Wawer MJ, Brookmeyer R, et al. Probability of HIV-1 transmission per coital act in monogamous heterosexual, HIV-1 discordant couples in
Rakai, Uganda. Lancet 2001; 357:1149-1153."

It is a blatant lie to claim that this study shows that HIV is not heterosexually transmitted.


Firstly, 24-27 means 4 references.  Christopher Noble claims that ref. 24 does "not show that HIV is not heterosexually transmitted".  Does it mean that he agrees with us that the other 3 references show "that HIV is not heterosexually transmitted"?

Secondly, the only way to obtain epidemiological evidence for heterosexual transmission is to follow a large number of discordant couples in prospective studies where other confounding, non-sexual routes of transmission are excluded.  No such studies have been published from Africa.  The claims of heterosexual transmission are based on the prevalence of "HIV-1" seropositivity and the equal number in men and women, or at best, in cross-sectional studies.

In the Gray et al community based study, "174 monogamous couples in which one partner was HIV-1 positive, were retrospectively identified from a population cohort" of 15,127 individuals aged 15-59 years of age in a period of 4 years.  Like in the Padian et al cross-sectional part of the study, in the Gray et al study non-sexual routes of transmission cannot be totally excluded.  Neither did the authors exclude anal intercourse.

The estimated infectivity for male-to-female transmission was the same in both studies, 0.0009 per sexual contact. According to this estimate, it would take 770 or 3333 sexual contacts respectively to reach a 50% or 95% probability of becoming infected. If sexual contacts were to take place repeatedly every three days this would require a period of 6.3 and 27.4 years respectively. Padian estimated that male-to-female transmission was "approximately eight-times more efficient than female-to-male", while Gray et al estimate that the probability of transmission per sexual contact for female-to-male was 0.0013.  Again, assuming sexual contact occurs on average once every three days, according to the Padian estimates would require 51 and 222 years to reach a probability of 50% and 95% respectively and to Gray's et al estimates 4.4 and 19.5 years [1]. 

Gray et al concluded: "The probability of HIV transmission per sex act in Uganda is comparable to that in other populations, suggesting that infectivity of HIV subtypes cannot explain the explosive epidemic in Africa". If: (i) there is no more heterosexual transmission in Africa than anywhere else; (ii) in the Padian prospective study the transmission was zero;
then in Africa, like anywhere else, "HIV" is not sexually transmitted.
If only for this fact it is imperative to discover the reason(s) for the very high level of seropositivity in Africa.



  1. Papadopulos-Eleopulos E, Turner VF, Papadimitriou JM, et al. Mother to Child Transmission of HIV and its Prevention with ATZ and Nevirapine. pp 204.  Perth: The Perth Group, 2001.



The Perth Group

Competing interests: None declared