Here are two interesting studies on Women & HIV reported at today's Retrovirus Conference - Mon. January 31st, 2000
Jules Levin, NATAP

Do Gender Differences in Viral Load Predict Differences in HIV Disease Progression?

J. BLAIR*, D. HANSON, J. JONES, M. DWORKIN, D. SMITH, P. DENNING, A. MOORMAN, and K. DECOCK. CDC, Atlanta, GA.

Background: Reports of different viral load (VL) levels among men and women at a given level of immunosuppression, the impact of these differences on HIV disease progression, and whether they are important in recommending gender-specific VL thresholds for initiation of antiretroviral therapy (ART) have been inconclusive.

Methods: Our study population included HIV-infected, ART-naive men and women greater than or equal to 13 years of age who had at least one VL measurement. We analyzed combined data from four CDC-supported projects: the Adult and Adolescent Spectrum of HIV Disease Project (ASD), the HIV Epidemiology Research Study (HERS), the HIV Outpatient Study (HOPS), and the Viral Load Surveillance Project. We assessed differences in VL by gender using a CD4 -stratified multiple regression model (MR) with robust variance estimates. Factors in the model included VL assay type, CD4 count, mode of transmission, age, race, and gender. For patients in the ASD, HERS, and HOPS projects, we examined the association of gender with time from first VL to diagnosis of initial AIDS-opportunistic illness (OI) (2,238 persons, 92 OI's) and to death (2,450 persons, 118 deaths) prior to ART initiation using Cox proportional hazards models (PH). Two-way interaction terms between gender and viral load were also included in the PH models.

Results: Of the 3,776 persons included in this analysis, 2,467 (65%) were men and 1,309 (35%) were women, for whom a total of 7,185 VL assays were performed. In the MR model, for CD4 strata of 0-199 cells/ml, 200-499 cells/ml, and >500 cells/ml it was estimated that women had VL levels that were 40%, 48%, and 57% lower than men, respectively (p<.05). Gender was not significantly associated with time to AIDS-OI (female/male hazard ratio [HR]=1.2; 95% CL=0.6, 2.4) nor survival time (HR=0.6; 95% CL 0.3, 1.2).

Conclusions: Although differences in VL exist by gender prior to initiation of ART, these differences were not found to be associated with survival time to AIDS-OI or death nor with the prognostic value of first VL on HIV disease progression. Our data do not suggest that a change in the treatment guidelines based upon gender and viral load is warranted at this time.

Note from Jules Levin-- In previous email I sent: excerpt from study reported in Lancet 1998, Homayoon Farzadegan et al, "Ö.another finding of the present study was that whereas time to AIDS was similar for men and women univariately, and even after controlling for CD4 cell count (table 3, models 1 and 2), the addition of HIV-1 viral load to these analyses resulted in a raised relative risk for AIDS in women (table 3, model 3). These results can be interpreted in two ways. First, results show directly that single measures of viral load are lower in women than in men, whereas time to AIDS is similar for men and women, which implies that at the same viral load women have a higher risk of AIDs. Second, the same time to AIDS would be predicted at a lower viral-load threshold (about 50%) in women than in men. The figure supports this interpretation by showing that a doubling of the viral-load thresholds for women not only eliminates the sex difference in viral load (table 2), but also attenuates the adjusted relative hazard for AIDS for women when controlling for CD4 concentration and viral load (table 3, model 3). Given that earlier studies5,6,13 have shown that both rate of CD4 cell decline and time to AIDS after HIV-1 seroconversion are similar in men and women, our results are best interpreted not as showing differences in time to AIDS according to sex, but rather that there is a different relation between viral load and AIDS in men compared with women.

No Seroconversion Occurred In Circumcised Males: Viral Load Higher in Transmitters Than Non-Transmitters

Viral Load and Risk of Heterosexual Transmission of HIV-1 among Sexual Partners. Abstract 193

T. C. QUINN*, M. J. WAWER, N. SEWANKAMBO, D. SERWADDA, C. J. LI, F. WABWIRE-MANGEN, M. MEEHAN, T. LUTALO, and R. H. GRAY« NIAID. Johns Hopkins Univ., Baltimore, MD; Columbia Univ., New York, NY; and Makerere Univ., Rakai Project Team, Uganda.

Objective: To determine the effect of viral load on the risk of heterosexual transmission of HIV, 415 HIV-discordant (HIV-positive/HIV-negative) couples were followed over a 30-month period in a study of 15,127 people in Rakai district, Uganda. Incidence of transmission and acquisition/100 person-years (py) were determined in relation to HIV viral load, sociodemographic and behavioral characteristics, and active STDs.

Results: The male partner was HIV infected in 228 (55%) and the female was HIV positive in 187 (45%). 90 (21.7%) of 415 discordant couples seroconverted (incidence 11.8/100 py). The transmission rate from males to females was 12.0/100 py and not statistically different from female-to-male transmission (11.6/100 py). Incidence was highest among 15 to 19-year-olds (15.2/100 py) and among uncircumcised males (16.7/100 py); no seroconversions occurred in circumcised males (p<0.0001). Mean viral load was significantly higher in transmitters vs. non-transmitters (90,254 vs. 38,029; p=0.01). There was a dose response of higher transmission with increasing viral load (p<0.0001), with no transmissions <1500 copies/mL, and up to 23.0/100 py for a viral load >50,000 copies/ml. In multivariate analyses, HIV viral load was the major predictor of HIV risk; each log_{10}increment in viral load was associated with a 2.45-fold increased risk of transmission (95% CI 1.85-3.26).

Conclusion: HIV-1 viral load was the major predictor of heterosexual transmission, with no differences by gender. However, circumcision among men reduced risks of acquisition. With no transmissions at viral loads <1500 copies/ml, reductions in viral load with antiretrovirals or vaccines may reduce transmission of HIV-1.