Viral subtype may influence the likelihood of HIV transmission between heterosexual couples, and concurrent genital ulcer disease (GUD) plays a clear role, according to a study from Uganda presented yesterday at the XVII International AIDS Conference in Mexico City.
Noah Kiwanuka presented an analysis of data from a long-term collaborative Uganda/U.S. study assessing rates of and risk factors for HIV transmission among serodiscordant heterosexual couples in Rakai, Uganda.
In the present analysis, the investigators sought to determine whether particular HIV-1 subtypes are associated with a higher rate of transmission.
As background, they noted that laboratory studies have shown different HIV subtypes to have varying properties related to viral pathogenicity (such as viral fitness and CCR5 vs CXCR4 tropism), and that these differences may influence transmission efficiency, disease progression, and response to antiretroviral therapy.
The Rakai cohort has been under study since 1994. About once a year, participants complete regular questionnaires about their health and sexual behaviour; give blood, urine, and vaginal fluid samples for HIV and sexually transmitted infection testing; and are offered condoms and risk-reduction counselling.
The present analysis included data from 271 serodiscordant couples in which the HIV-negative partner was monogamous (i.e., reported having sex only with the HIV-positive partner) and for whom subtype data were available. Three-quarters of the couples consisted of an HIV-positive man and an HIV-negative woman.
The HIV-negative partners were categorised as being exposed during the early (within one year or less of the partner’s seroconversion), latent (more than 12 months after seroconversion but before the onset of AIDS), or late (after the onset of AIDS) stage of the partner’s disease. However, only the latent group was included in the analysis due to very small numbers of HIV transmissions in the other groups.
HIV-1 subtype D was most common among the HIV-positive partners in this analysis (73%), followed by various recombinant or hybrid forms (14%) and then subtype A (11%).
During the course of follow-up, a total of 92 HIV transmissions occurred, representing just over one-third of the couples (incidence rate 11.6 per 100 person years). There was no difference in the rate of male-to-female compared with female-to-male transmission.
After adjusting for age, viral load and genital ulcer disease (GUD), HIV subtype A was transmitted nearly twice as often as subtype D (adjusted rate ratio 1.98). Recombinant strains and subtype D were transmitted with similar frequency (rate ratio 1.07). The recombinants were transmitted less often than subtype A, but this difference did not reach statistical significance.
Other significant risk factors for HIV transmission were lower reported condom use, higher viral load and younger age of the positive partner (transmission was about four times more likely if the positive partner was under 30 compared with over 40 years old; rate ratio 3.98). Very few men were circumcised, and circumcision was not a significant factor in this study.
The presence of genital ulcers, especially in the HIV-negative partner, also had a strong effect. If either partner had GUD, the risk of HIV transmission nearly doubled (8.7 vs 14.2% per 100 person-years; rate ratio 1.76). But if both had GUD, the likelihood rose by close to four-fold (29.6 per 100 person-years; rate ratio 3.70).
Discussing these findings, Kiwanuka said it remains unclear why, if subtype A is more easily transmitted, subtype D is so much more prevalent in the population.
Data on subtype differences in transmission “are crucial for HIV vaccine development programs, for understanding the dynamics of HIV-1 epidemics in different geographical regions, and for future prediction of the pandemic,” the investigators stated.
Kiwanuka also suggested that researchers should consider HIV subtype as a new parameter when constructing mathematical models of transmission.