Combining Phylogenetic and Network Approaches to Identify HIV-1 Transmission Links in San Mateo County, California.

Sudeb C Dalai,Vivian Levy,Renee Mehra,Dennis Israelski,Sergei L Kosakovsky Pond,Eduan Wilkinson,Tulio De Oliveira,David Katzenstein,Dennis Maletich Junqueira

doi:10.3389/fmicb.2018.02799

Abstract

The HIV epidemic in San Mateo County is sustained by multiple overlapping risk groups and is an important hub for HIV transmission in northern California. Limited access to care has led historically to delayed clinical presentation, higher rates of opportunistic infections, and an increased prevalence of antiretroviral drug resistance. The virologic and clinical consequences of treatment within these multiple ethnic and behavioral groups are poorly understood, highlighting the need for efficient surveillance strategies that are able to elucidate transmission networks and drug resistance patterns. We obtained sequence data from a group of 316 HIV-positive individuals in the San Mateo AIDS Program over a 14-year period and integrated epidemiologic, phylogenetic, and network approaches to characterize transmission clusters, risk factors and drug resistance. Drug resistance mutations were identified using the Stanford HIV Drug Resistance Database. A maximum likelihood tree was inferred in RAxML and subjected to clustering analysis in Cluster Picker. Network analysis using pairwise genetic distances was performed in HIV-TRACE. Participants were primarily male (60%), white Hispanics and non-Hispanics (32%) and African American (20.6%). The most frequent behavior risk factor was male-male sex (33.5%), followed by heterosexual (23.4%) and injection drug use (9.5%). Nearly all sequences were subtype B (96%) with subtypes A, C, and CRF01_AE also observed. Sequences from 65% of participants had at least one drug resistance mutation. Clustered transmissions included a higher number of women when compared to non-clustered individuals and were more likely to include heterosexual or people who inject drugs (PWID). Detailed analysis of the largest network (N = 47) suggested that PWID played a central role in overall transmission of HIV-1 as well as bridging men who have sex with men (MSM) transmission with heterosexual/PWID among primarily African American men. Combined phylogenetic and network analysis of HIV sequence data identified several overlapping risk factors in the epidemic, including MSM, heterosexual and PWID transmission with a disproportionate impact on African Americans and a high prevalence of drug resistance.

Highlights

Molecular epidemiologic analyses of the HIV-1 epidemics in Africa (Dalai et al, 2009; Gray et al, 2009; Mir et al, 2016), Asia (Neogi et al, 2012; Pang et al, 2012), and North America (Smith et al, 2010; Mehta et al, 2015) have provided evidence for distinct epidemic dynamics and patterns of transmission within defined communities
Between 1996 and 2010, 637 HIV sequences were obtained from 316 people living with HIV/AIDS (PLWHA) receiving antiretroviral treatment (ART) in the San Mateo County AIDS Program
By using a minimum branch support of 90% (TBE) and an intra-cluster genetic distance threshold of 8%, we identified nine supported clusters in Cluster Picker

Summary

Introduction

Molecular epidemiologic analyses of the HIV-1 epidemics in Africa (Dalai et al, 2009; Gray et al, 2009; Mir et al, 2016), Asia (Neogi et al, 2012; Pang et al, 2012), and North America (Smith et al, 2010; Mehta et al, 2015) have provided evidence for distinct epidemic dynamics and patterns of transmission within defined communities. In the past 10 years, the HIV epidemic in California has shifted from a primarily white MSM (men who have sex with men) population to include a diverse range of overlapping risk groups, where heterosexual women comprise the fastest growing demographic for new infections and an increasing fraction of new diagnoses occur among diverse racial and ethnic groups (California Department of Public Health, 2015). The epidemic is sustained by multiple ethnic, migratory, and behavioral networks, including MSM, migrant populations from Asia and Latin America, and PWID, each having distinct patterns of HIV acquisition and transmission (San Mateo County Department of Public Health, 2017). Within these communities, access to care, including antiretroviral treatment (ART) has historically been constrained by financial, cultural, and linguistic barriers. The implications of prolonged, undiagnosed HIV infection, viremia, 1http://www.countyhealthrankings.org/

Methods

Results

Discussion

Conclusion