Open Access Highly Accessed Research article

Epidemiological study of phylogenetic transmission clusters in a local HIV-1 epidemic reveals distinct differences between subtype B and non-B infections

Kristen Chalmet1, Delfien Staelens1, Stijn Blot2, Sylvie Dinakis2, Jolanda Pelgrom2, Jean Plum1, Dirk Vogelaers2, Linos Vandekerckhove12 and Chris Verhofstede1*

Author Affiliations

1 AIDS Reference Laboratory, Ghent University, De Pintelaan 185-Blok A, B-9000 Gent, Belgium

2 AIDS Reference Center, Ghent University Hospital, De Pintelaan 185-1P2, B-9000 Gent, Belgium

For all author emails, please log on.

BMC Infectious Diseases 2010, 10:262  doi:10.1186/1471-2334-10-262

Published: 7 September 2010

Abstract

Background

The number of HIV-1 infected individuals in the Western world continues to rise. More in-depth understanding of regional HIV-1 epidemics is necessary for the optimal design and adequate use of future prevention strategies. The use of a combination of phylogenetic analysis of HIV sequences, with data on patients' demographics, infection route, clinical information and laboratory results, will allow a better characterization of individuals responsible for local transmission.

Methods

Baseline HIV-1 pol sequences, obtained through routine drug-resistance testing, from 506 patients, newly diagnosed between 2001 and 2009, were used to construct phylogenetic trees and identify transmission-clusters. Patients' demographics, laboratory and clinical data, were retrieved anonymously. Statistical analysis was performed to identify subtype-specific and transmission-cluster-specific characteristics.

Results

Multivariate analysis showed significant differences between the 59.7% of individuals with subtype B infection and the 40.3% non-B infected individuals, with regard to route of transmission, origin, infection with Chlamydia (p = 0.01) and infection with Hepatitis C virus (p = 0.017). More and larger transmission-clusters were identified among the subtype B infections (p < 0.001). Overall, in multivariate analysis, clustering was significantly associated with Caucasian origin, infection through homosexual contact and younger age (all p < 0.001). Bivariate analysis additionally showed a correlation between clustering and syphilis (p < 0.001), higher CD4 counts (p = 0.002), Chlamydia infection (p = 0.013) and primary HIV (p = 0.017).

Conclusions

Combination of phylogenetics with demographic information, laboratory and clinical data, revealed that HIV-1 subtype B infected Caucasian men-who-have-sex-with-men with high prevalence of sexually transmitted diseases, account for the majority of local HIV-transmissions. This finding elucidates observed epidemiological trends through molecular analysis, and justifies sustained focus in prevention on this high risk group.