Hotspots of Transmission Driving the Local Human Immunodeficiency Virus Epidemic in the Cologne-Bonn Region, Germany

Abstract

Background: Geographical allocation of interventions focusing on hotspots of human immunodeficiency virus (HIV) transmission has the potential to improve efficiency. We used phylogeographic analyses to identify hotspots of the HIV transmission in Cologne-Bonn, Germany.

Methods: We included 714 HIV-1 infected individuals, followed up at the University Hospitals Cologne and Bonn. Distance-based molecular network analyses were performed to infer putative relationships. Characteristics of genetically linked individuals and assortativity (shared characteristics) were analyzed. Geospatial diffusion (ie, viral gene flow) was evaluated using a Slatkin-Maddison approach. Geospatial dispersal was determined by calculating the average distance between the residences of linked individuals (centroids of 3-digit zip code).

Results: In sum, 217/714 (30.4%) sequences had a putative genetic linkage, forming 77 clusters (size range: 2-8). Linked individuals were more likely to live in areas surrounding the city center (P = .043), <30 years of age (P = .009). and infected with HIV-1 subtype B (P = .002). Clustering individuals were nonassortative by area of residency (-.0026, P = .046). Geospatial analyses revealed a median distance between genetically linked individuals of 23.4 kilometers (km), lower than expected (P < .001). Slatkin-Maddison analyses revealed increased gene flow from central Cologne toward the surrounding areas (P < .001).

Conclusion: Phylogeographic analysis suggests that central Cologne may be a significant driver of the regional epidemic. Although clustering individuals lived closer than unlinked individuals, they were less likely to be linked to others from their same zip code. These results could help public health entities better understand transmission dynamics, facilitating allocation of resources to areas of greatest need.

Trial registration: ClinicalTrials.gov NCT02149004.

Keywords: HIV transmission; geospatial dispersial; phylogeographic analyses; public health.

© The Author(s) 2018. Published by Oxford University Press for the Infectious Diseases Society of America. All rights reserved. For permissions, e-mail: journals.permissions@oup.com.

Figures

Figure 1.

Geospatial distribution of (A) number of sequences in the study population (n = 714), and (B) number of clustering individuals based on the 3-digits zip code of residency (n = 217).

Figure 2.

Transmission network of HIV in the Cologne-Bonn region. Sex of the individuals (nodes) is indicated by shape. In (A) the color indicates the reported risk group. In (B) the color indicates the residence location, ie, living in the city center (orange) or surrounding areas (blue) of Cologne-Bonn. All edges represent a genetic distance of ≤1.5%. Abbreviations: HTS, heterosexual sex; MSM, men having sex with men; NA, not applicable; PWID, people with injection drug use.

Figure 3.

Geographic inferred transmission network by the location of residence overlaid on the maps of the Cologne and Bonn area. The size of the dots defining the number of clustering individuals in each specific area. The golden lines representing the links between different areas. Most of the larger clusters (>20 individuals) are located in the central parts of Cologne and Bonn (n = 217).

Figure 4.

A, Median Euclidean distance between linked sequences vs median distance between random selection of sequences. Simulation were repeated with 1000 iterations. B, Viral gene flow within the sampled epidemic between central district of Cologne and the surrounding areas. Viral migration was inferred using the Slatkin-Maddison index on phylogenetic trees constructed from 1000 random subsets of equal number of sequences per location to identify the diffusion of the epidemic in the region between central Cologne and surrounding areas. C, Geospatial mixing between genetically linked individuals using assortativity coefficients by area of residency (n = 714).