March 22, 2011
From CROI, March 2011: A large African study (Partners in Prevention) that tried to prevent HIV by using aciclovir to prevent herpes infections failed to show any effect on reducing HIV transmissions, as reported last year at CROI. This year, however, three new analyses of that study made three valuable observations: 1) the risk of transmission to the negative partner, 2) how viral load in genital fluids affects the risk of transmission, and 3) how viral load in the positive partner affects the viral load of the infected partner.
Although aciclovir did not reduce the risk of transmission, the study did show a 92% lower risk for HIV in the negative partners of 349 positive partners who had started HIV treatment during the study. The three analyses looked at different aspects of this reduced risk in 3,297 couples.
The risk of transmission to the negative partner
As part of the study design, all negative partners were given antibody tests every 3 months. The positive partners had viral load tests taken every 3 to 6 months. At each study visit, everyone was asked about their sexual activity and condom use since their last visits, including sex with partners outside the primary relationship.
A total of 151 new HIV infections occurred during the study: 108 from primary partners and 43 from non-primary partners. Of those partners who reported using condoms 100% of the time, 56 infections actually occurred while another 15 infections occurred between partners who said they hadn’t had sex since their last visits. The results showed the following rates of risk:
| Viral load of positive partner | Risk of infection per act of sex |
| 1,000 copies | 1 in 3,571 acts * |
| 10,000 | 1 in 1,220 acts * |
| 100,000 | 1 in 434 acts * |
| 1,000,000 | 1 in 149 acts * |
| *Figures from Gus Cairns, NAM/aidsmap.com | |
Factors that increased infections were herpes (more than doubles risk) and genital ulcer disease (nearly triples risk). Factors that decreased infections included undetectable blood viral load and use of condoms by the positive partners as well as male circumcision and fewer sexual infections in the negative partners.
How viral load in genital fluids affects infectiousness
Although high HIV viral load in blood has been shown to increase the risk of infection, the infectiousness of viral load in genital fluids is not well understood. In the study, viral loads were measured from cervical fluids and semen collected from 2,521 positive partners, although only 1,805 cervical samples (46 transmissions) and 716 semen samples (32 transmissions) were used.
The results showed that for every 10-fold increase (1 log) in genital viral load, there was a 2.20-fold higher risk of transmission from positive women and a 1.79-fold higher risk from positive men. After adjusting for blood viral load, genital viral load was still independently associated with a 70% higher risk for HIV transmission.
Although detectable genital viral load increased risk, 7 cases of transmissions from positive women and 4 cases from positive men occurred in those with undetectable genital viral load but with detectable blood viral load. No infections occurred in anyone with undetectable blood but detectable genital viral loads.
How the viral load of the positive partner influences the viral load in the infected partner
Studies in gay and bisexual men show a relationship between the viral load of the positive partner and viral set point of the infected partner. The viral set point, established during early infection, relates to that person’s level of disease progression. The Partners in Prevention study provides new information on this viral set point in heterosexual people.
One hundred cases of transmissions were used. For every 10-fold increase (1 log) in blood viral load of the positive partner, there was a 2.5-fold increase in the viral set point in the infected partner. If the positive partner took aciclovir, then the average set point was 60% lower. If the positive male partner was circumcised, the average set point was 80% lower. Pregnancy also greatly increased the risk of a higher viral set point.
Final thoughts
These data will probably help guide prevention services, especially in mixed-status couples. Although a natural impulse may be to try to draw similar conclusions about HIV transmission through anal sex (both heterosexual and homosexual), this study did not evaluate those levels of risk. These data must come from other randomized studies that also take fluid samples from the anus and rectum.
RESEARCH STUDIES:
Determinants of Per-act Infectivity of HIV-1 in the Partners in Prevention Study. Hughes J, et al. Abstract #135.
Genital HIV-1 RNA Levels Predict Risk of Heterosexual HIV-1 Transmission. Baeten J, et al. Abstract #154.
Infected Partner’s Plasma HIV-1 RNA Level and the HIV-1 Set Point of Their Heterosexual Seroconverting Partners. Lingappa J, et al. Abstract #134.