You are here
Ongoing HIV-1 subtype B transmission networks in the Netherlands
Daniela Bezemer, Oliver Ratmann, Ard van Sighem, Bas E. Dutilh, Nuno Faria, Rob van den Hengel, Luuk Gras, Peter Reiss, Frank de Wolf, Christophe Fraser, ATHENA observational cohort *Research, HIV Monitoring Foundation, Amsterdam, Netherlands, Imperial College, London, United Kingdom, Centre for Molecular and Biomolecular Informatics, Nijmegen, Netherlands, University of Oxford, Oxford, United Kingdom
Background: The HIV-1 epidemic amongst men having sex with men (MSM) in the Netherlands has been shown to be resurgent despite the widespread use of effective combination antiretroviral therapy (cART). This epidemic is dominated by HIV-1 subtype B. In this study, using phylogenetic analysis of HIV-1 subtype B polymerase sequences, we identify the transmission networks that constitute the epidemic. Methodology: As of June 2011, the ATHENA database contained polymerase sequences from 5,852 subtype B infected patients in the Netherlands. For every sequence, the 10 most similar sequences were selected from the Los Alamos database and a phylogenetic tree was created of in total 8,320 unique sequences. Transmission networks were defined as clusters of sequences with a bootstrap value ≥90%, within the 5th percentile threshold of the wholetree patristic distance distribution including at least 10 patients in the ATHENA cohort. Results: Amongst the 5,852 patients, 4,288 (73%) were MSM registered in the Netherlands. We identified 106 different transmission networks which included 3,061 (52%) ATHENA sequences. 50% (2,128) of 4,288 HIV cases amongst MSM in the Netherlands and 380 of the ‘Los Alamos’ sequences were included in 91 MSM majority networks (≥50%). Strikingly, 64 (70%) of these 91 MSM transmission networks were already circulating before 1996, the year that cART became widely available. Of the total 3,507 sequences among MSM with diagnoses after 1996, 41% (1,436) were found in these 64 networks, and 89% (57) of these networks include patients with proof of infection after 1996. Only in three of the 64 networks no new cases were diagnosed after 2006. The people pertaining these three networks had a higher median age at the study end date; 51 years (IQR: 45-57) compared to 46 (IQR: 39-52) in the persisting networks. Further, 68 MSM were in 10 of non-MSM dominated networks, and 1,324 (31%) were in 550 smaller clusters. 768 (18%) sequences amongst MSM in the Netherlands could not be identified as belonging to a network (singletons). Of MSM in a network, 4% have probably been infected abroad (according to their own report), versus 10% in the smaller clusters and 19% of singletons. If we regard every cluster, network, or singleton as a new HIV introduction, this totals to 1,419 HIV-1 subtype B introductions amongst MSM. Roughly indicating that 33% of infections amongst MSM were imported, and that 6% of introductions amongst MSM in the Netherlands gave rise to a local network. Conclusions: The majority of the identified MSM-dominated transmission networks were already present before the introduction of cART and are still ongoing. Our analyses suggest that the resurgent epidemic is sustained by well-established networks, not brought to an end by the widespread use of cART.