Chief Science Officer, Boveran, Inc.
San Ramon, CA 94583
Send response to journal:
Peter Flegg uncritically accepts viral load as a measure of infectious HIV in spite of the documented problems with using it for that purpose (see my letters of March 7 and April 23, 2003). For example, he says that, "[viral load] levels are often extremely high during PHI [Flegg's reference(5), reference (1) below], and some individuals have seminal viral loads that consistently exceed plasma levels and are considered to be highly infectious hyperexcretors" even though they were either HIV negative at entry or HIV positive within 30 days of entry.
Apparently, Pilcher et al. were not aware of the pioneering study of Piatak et al. where high HIV viral load did not translate into infectious virus (2). Commenting on that study, Shepard et al. said that, "...the high level of plasma virus observed by Piatak et al. was about 99.9 per cent non-culturable, suggesting that it was either neutralized or defective... . [W]e question the longitudinal conclusions some of these investigators have drawn from cross- sectional data. The results presented are equally consistent with the conclusion that higher viraemia is a consequence of, rather than the proximate cause of, defective immune responses" (3).
Piatak et al. were unable to culture HIV from 53% of patients with viral loads ranging from 10,000 to over 2,000,000 copies of HIV RNA per ml of blood plasma (2). The authors said that, "Plasma virus levels determined by QC-PCR correlated with, but exceeded by an average of 60,000-fold, virus titers measured by endpoint dilution culture. ... [And] For HIV-1 propagated in vitro, total virions have been reported to exceed culturable infectious units by factors of 10,000 to 10,000,000, ratios similar to those we observed in plasma."
With the Piatak et al. results in mind, I noticed that Pilcher et al. said that, "All subjects with primary HIV infection tested had positive PBMC HIV culture". However, they provided no data and it is unclear whether they cultured HIV from seminal plasma or only from blood plasma. If "positive PBMC HIV culture" meant the presence of p24 antigen in the supernatant, then (as I reported in the April 23, letter) the presence of transmissible HIV is in doubt because p24 antigen is not a reliable indicator of infectious HIV (or non-infectious HIV, for that matter).
Flegg goes on to say that, "Pilcher has also model[ed] transmission and estimates that during the peak viral shedding that is associated with PHI, transmission would be 20-fold higher than that during stable chronic infection. For individuals with peak seminal viral loads of log 8.85, the probability of transmission per act would be 1.0 (i.e. transmission would occur with every single exposure), dropping to 0.03 for individuals with levels of 5.43log, down to 0.0015 for levels of 3.75log [Flegg's reference (5), reference (1) below]. However, I could not find this information in the Pilcher et al. reference that Flegg cited (1). Perhaps he meant a different reference or I may have overlooked it, but I scanned the paper three times with no luck.
Flegg's last citation was to Chakraborty et al. who have "looked at HIV's infectiousness in 86 males and 24 females, and developed a model for calculating the probability of HIV transmission per heterosexual coital act [Flegg's reference (6), reference (4) below]. As levels of HIV-RNA rise in semen, probability of transmission increases (seminal level of log5 equating with probability of transmission of 1 per 100 episodes of intercourse)."
Chakraborty et al. begin by saying that, "Epidemiological and mathematical models have been developed to estimate the likelihood of HIV-1 transmission during a single episode of sexual intercourse. Such models are confounded by difficulty in collecting appropriate empirical data from discordant couples (when HIV-1 positive and negative people engage in sex) and from limitations in different kinds of estimation."
The "difficulty in collecting appropriate empirical data from discordant couples" is indeed a problem when the best controlled studies of discordant couples have failed to observe even a single HIV-negative sex partner becoming HIV-positive after years of observations (e.g. Padian et al. (5), where the observed efficiency of heterosexual transmission of HIV was zero).
Chakraborty et al. say that, "The probability of per-partner sexual transmission of HIV-1 has been examined in 11 different studies, whereas the per-sex-act probability of transmission has been reported in 13 studies. The probability of transmission of HIV-1 from male to female during an episode of intercourse has been examined in seven of these studies. Analysis of data from North America and European studies of heterosexual couples provide estimates of per-sex-act HIV-1 transmission of approximately 1 in 1000...".
Nevertheless, in 2001, the authors deemed it necessary to come up with yet another estimate because "the transmission probabilities presented [by the 7 studies in their Table 2] are so low that it becomes difficult to understand the magnitude of the HIV-1 pandemic, especially in developing countries. An alternative approach to explain the epidemic is development of mathematical models." Yet, the authors admit that, "The greatest limitation of this and other models lies in the tremendous difficulty in clinical validation."
The authors "assumed that the best predictor of infectiousness of the male partner is the cell-free virus measured in seminal plasma. It is not known whether HIV-1 is transmitted from cell free virus in the seminal plasma, or from cellular HIV-1." The unstated assumption, of course, is that HIV-1 is sexually transmitted in the first place. But of course, being that honest would never have gotten past the gatekeepers reviewing the manuscript.
Here is a summary of the stated and unstated assumptions that the authors used to construct their model: 1) HIV is sexually transmitted, 2) viral load is measuring the presence of HIV, 3) viral load measures the level of infectious HIV, 4) "the risk of HIV-1 transmission remains the same for each episode of intercourse..., although some have argued that exposure leads to some degree of immunity", 5) "that total non-synctium-inducing (NSI) HIV-1 RNA concentrations (x1) and CD4+CCR5 receptor cells (x2) were represented by a Pearson's type-1 distribution that could be transformed into a Beta distribution (subtracting the minimum value and divided by the range)", 6) "that 100% of the HIV-1 variants in the semen express the NSI phenotype, 7) "that HIV-1 variants which use CCR5 receptors (NSI isolates) are preferentially sexually transmitted", and 8) "The number of receptors for HIV-1 will also determine the efficiency of transmission".
Armed with these assumptions, an uncritical acceptance of the uncontrolled HIV-1 viral load data for seminal plasma from Chapel Hill, Seattle, and St. Gallen, plus additional subjects, and a considerable degree of mathematical hand-waving, the authors came up with an estimate of the probability of sexually transmitting HIV that is 10-fold greater (i.e. "1 per 100 episodes of intercourse") than the numerous estimates produced by others that they cite in Table 2 as being unable to explain "the magnitude of the HIV-1 pandemic, especially in developing countries". However, the authors make no comment as to whether their 10-fold higher estimate is sufficient to explain "the magnitude of the HIV-1 pandemic, especially in developing countries".
1. Pilcher, C. D., Shugars, D. C., Fiscus, S. A., Miller, W. C., Menezes, P., Giner, J., Dean, B., Robertson, K., Hart, C. E., Lennox, J. L., Eron, J. J., Jr., and Hicks, C. B. (2001) HIV in body fluids during primary HIV infection: implications for pathogenesis, treatment and public health, Aids 15, 837-845
2. Piatak, M., Saag, M. S., Yang, L. C., Clark, S. J., Kappes, J. C., Luk, K.-C., Hahn, B. H., Shaw, G. M., and Lifson, J. D. (1993) High levels of HIV-1 in plasma during all stages of infection determined by competitive PCR, Science 259, 1749-1754
3. Sheppard, H. W., Ascher, M. S., and Krowka, J. F. (1993) Viral burden and HIV disease, Nature 364, 291-292
4. Chakraborty, H., Sen, P. K., Helms, R. W., Vernazza, P. L., Fiscus, S. A., Eron, J. J., Patterson, B. K., Coombs, R. W., Krieger, J. N., and Cohen, M. S. (2001) Viral burden in genital secretions determines male-to-female sexual transmission of HIV-1: a probabilistic empiric model, Aids 15, 621-627
5. Padian, N. S., Shiboski, S. C., Glass, S. O., and Vittinghoff, E. (1997) Heterosexual transmission of human immunodeficiency virus (HIV) in northern California: results from a ten-year study, Am J Epidemiol 146, 350-357
Competing interests: None declared