icon-folder.gif   Conference Reports for NATAP  
  International HIV and Hepatitis Drug Resistance Workshop
June 8-12, 2010,
Dubrovnik Croatia
Back grey_arrow_rt.gif
US/Canadian HIV Transmission Clusters Largely Male, White, and Gay
  International HIV and Hepatitis Drug Resistance Workshop, June 8-12, 2010, Dubrovnik, Croatia
Mark Mascolini
Gay, white men largely populated 43 HIV transmission clusters identified at 69 centers across the US, Canada, and Puerto Rico in 2007 [1]. People in clusters--groups of infected individuals with genetically related HIV--accounted for 109 of 709 people (15%) analyzed by Lisa Ross (GlaxoSmithKline) and colleagues at US and Canadian clinics.
Transmission clusters identified by phylogenetic relationships between viruses can help virologists and public health officials understand how HIV is spreading through large or small populations. Of course everyone infected with HIV belongs to a cluster because everyone is infected by someone else. But clustering analyses can be enlightening by defining the traits of phylogenetically identifiable transmission groups.
This study involved 709 people seeking antiretroviral therapy at US, Canadian, and Puerto Rican clinics in 2007. Because of this focus, the clustering analysis may say little about how HIV transmissions cluster among people who remain undiagnosed or who know they have HIV but are not in care.
Ross and coworkers determined that 109 of these 709 people (15%) belonged to a cluster including at least 2 people. Notably, only 10 of 43 identified clusters (23%) included more than 2 people, so the study definition of a cluster is loose and not what a nonscientist would colloquially consider a cluster.
Median pretreatment viral load was the same in clustered individuals and the whole study group at 5.04 log (a little over 100,000 copies). Forty-eight clustered participants (44%) and 335 of the entire cohort (47%) had a viral load below 100,000 copies. Median pretreatment CD4 count was a little higher in clustered patients than in the whole group (217 versus 204).
Comparing clustered and nonclustered people, the investigators found that clustered individuals were more likely to be male and white and less likely to have major protease or reverse transcriptase mutations or to be coinfected with hepatitis B or C. Also, Canadians studied were more likely to be in clusters:
-- Male: 90% clustered versus 83% nonclustered
-- White: 65% clustered versus 60% nonclustered
-- Canadian: 30% clustered versus 13% nonclustered
-- Major resistance mutation: 9% clustered versus 14% nonclustered
-- Hepatitis C: 4% clustered versus 9% nonclustered
-- Hepatitis B: 0% clustered versus 2% nonclustered
Of the 43 clusters, 42 included men. Only one cluster consisted entirely of women, and only 6 clusters (14%) included at least one woman. Thirty-six clusters (84%) included at least one man who identified himself as gay. Only eight clusters (19%) included people who listed either gay or straight sex as an HIV risk factor. Thirty clusters (70%) included people seeking treatment in the same city, and 33 (77%) included people seeking care in the same state or province.
Among the 709 people studied, 519 (73%) started antiretroviral therapy with two nucleosides and a boosted protease inhibitor, including 93 of 109 clustered patients (85%) and 426 of 600 nonclustered patients (71%). During 120 weeks of therapy, 29 treated people (6%) had virologic failure, and only 2 of them (7% of 29) were in clusters.
Together these findings indicate that in the US and Canada it is easier to identify HIV transmission clusters in gay men starting antiretroviral therapy than in people infected heterosexually or by injecting drugs. The high clustering rate in Canadian patients may point to more closely knit communities of people at risk of HIV infection, or it may simply be a sampling artifact. (The investigators did not identify the 69 sites.)
The lower rate of transmitted resistance in clustered people may seem surprising since gay white men generally have better access to antiretrovirals than other groups in the US and Canada and thus are more likely to be treated, to acquire resistant virus during treatment failure, and to pass that resistant virus to gay sex partners. But because this study focused on gay men seeking care in 2007, many may have been infected by people taking recently prescribed strong regimens that suppressed viremia and resistant viral populations.
1. Ross LL, DeJesus E, Potter M, et al. Epidemiological and genotypic clustering of HIV infection within North America during 2007. International HIV and Hepatitis Drug Resistance Workshop. June 8-12, 2010. Dubrovnik, Croatia. Abstract 150.