Characteristics of patients recently infected with HIV-1 non-B subtypes in France: a nested study within the mandatory notification system for new HIV diagnoses

Abstract

The presence of HIV-1 non-B subtypes in Western Europe is commonly attributed to migration of individuals from non-European countries, but the possible role of domestic infections with non-B subtypes is not well investigated. The French mandatory anonymous reporting system for HIV is linked to a virological surveillance using assays for recent infection (<6 months) and serotyping. During the first semester of years 2007 to 2010, any sample corresponding to a non-B recent infection was analyzed by sequencing a 415-bp env region, followed by phylogenetic analysis and search for transmission clusters. Two hundred thirty-three recent HIV-1 infections with non-B variants were identified. They involved 5 subtypes and 7 circulating recombinant forms (CRFs). Ninety-two cases (39.5%) were due to heterosexual transmissions, of which 39 occurred in patients born in France. Eighty-five cases (36.5%) were identified in men having sex with men (MSM). Forty-three recent non-B infections (18.5%) segregated into 14 clusters, MSM being involved in 11 of them. Clustered transmission events included 2 to 7 cases per cluster. The largest cluster involved MSM infected by a CRF02_AG variant. In conclusion, we found that the spread of non-B subtypes in France occurs in individuals of French origin and that MSM are particularly involved in this dynamic.

FIG 1

Study population and results of the phylogenetic analysis. NT, not typeable; QC, quality control.

FIG 2

Phylogenetic tree of 106 gp41 sequences from HIV-1 variants involved in transmission clusters. Multiple-sequence alignment including reference sequences (1 to 3 sequences representative of each subtype or CRF for which a cluster was identified) was conducted using Clustal W (26). The phylogenetic tree was constructed using the neighbor-joining method. Genetic distances were computed using the Kimura two-parameter method in MEGA5 (28, 29). The optimal tree is drawn to scale, with branch lengths in the same units as those of the distances. The reliability of the branching orders was assessed by bootstrap analysis of 500 replicates. Viruses isolated in heterosexual patients, MSM, injecting drug users (IDU), and individuals with unknown risk factors are indicated by triangles, circles, diamonds, and squares, respectively. The region of residence is shown for each case. The color code corresponds to five regions of metropolitan France: Paris area (yellow), northeast (red), southeast (green), southwest (orange) and northwest (dark blue). One case was a patient living in the French West Indies (in gray), and the region of residence was not known for one case (empty symbol).

FIG 3

Time interval between HIV-1 infections within each cluster. Large clusters are shown on the left part of the scheme. Each cluster is represented by a vertical bar (B, orange; non-B, green). Each infection is represented by a large horizontal bar.