CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

Methods: Using 2000-2018 HIV surveillance data from Public Health–Seattle & King County, HIV-1 pol gene sequences were linked to demographic, clinical, and epidemiological information. We identified genetic similarity clusters of 2+ individuals using TN93 pairwise genetic distance with a 0.015 threshold, and assessed correlates of clustering using multivariate logistic regression. We conducted probabilistic phylodynamic modeling to estimate transmission flows between YMSM (age <25) and older MSM (categorized for analyses as age 25-34, 35-44, and >45). Results: From 2000-2018, 4597 MSM were diagnosed with HIV in King County, with 654 (14%) diagnoses among YMSM. Among 2851 (62%) of MSM with an available sequence, 1435 (50%) clustered in 277 genetically similar clusters: 9 clusters were comprised of only YMSM, 166 of only older MSM, 102 of both older and YMSM. YMSM had higher odds of clustering compared to those >25 years old (AOR 1.6; 95% CI: 1.3, 2.0). Older MSM were more likely to cluster with other MSM>25 years old (AOR 4.3; 95% CI: 2.3, 3.1) and less likely to cluster with YMSM (AOR 0.4; 95% CI: 0.3, 0.5), compared to YMSM. Phylodynamic modeling suggest that the majority (47%) of HIV transmissions occurs among MSM age 25-34 and 35-44 years old. The overall assortativity coefficient was 0.08. YMSM had the highest probability of acquiring HIV fromMSM aged 25-34 years old (39%) and 35-44 years old (31%), with a 19% probability of acquiring HIV from other YMSM. Phylodynamic models estimated that YMSM acquire HIV from MSM with probability-weighted mean age difference of 11.2 years older (IQR 4 to 18 years). Conclusion: Both molecular epidemiology and phylodynamic methods were suggestive of age-assortative mixing among older MSM, among whom the majority of HIV transmissions occurred. However, molecular cluster analyses were suggestive of high relative rates of transmission among YMSM. Phylodynamic models also found that YMSM frequently acquire HIV from older partners, suggesting that age-discrepant partnerships play an important role in HIV dynamics among YMSM.

Indigenous: h=.29; Asian: h=.33; Other: h=.11), residential neighborhood (Downtown Vancouver: h=.39; Vancouver: h=.37; Outside Vancouver: h=.31), education (High school or greater: h=.34; Less than high school: h=.09), patronage of gay bars and clubs (About once per month or more: h=0.33; Less than once per month: h=.19), and use of online sex seeking apps (About once per month or more: h=.32; Less than once per month: h=0.16), and use of GHB (Yes: h=.30; No: h=.25), LSD (Yes: h=.41; No: h=.14), crystal methamphetamine (Yes: h=.37; No: h=.29), and crack (Yes: h=.14; No: h=.40). Low homophily (h<.30) was observed for perceived HIV transmission risk (Low Risk: h=0.27; High risk: h=0.11), STI history (Ever Diagnosed: h=.27; Never Diagnosed: h=.09), and for patterns of condom use: (No anal sex: h=.07; No condomless anal sex (CAS): h=.04; CAS with only sero-concordant partners: h=.01; CAS with serodiscordant/unknown status partners: h=.11). Conclusion: We observed moderate to high homophily across demographic characteristics, substance-use, and dating-venues, but low homophily of sexual behaviours. 918 PHYLOGENETIC INSIGHTS ON HIV-1 TRANSMISSION DYNAMICS AMONG MSM AND MIGRANTS IN QUEBEC Bluma G. Brenner 1 , Nathan Osman 1 , Ernesto Cuadra Foy 1 , Antoine Chaillon 2 , Ruxandra-Ilinca Ibanescu 1 , Isabelle Hardy 3 , Nadine Kronfli 4 , David Stephens 5 , Michel Roger 3 1 Lady Davis Institute for Medical Research, Montreal, QC, Canada, 2 University of California San Diego, La Jolla, CA, USA, 3 Centre de Recherche du CHUM, Montreal, QC, Canada, 4 Research Institute of McGill University Health Centre, Montreal, QC, Canada, 5 McGill University, Montreal, QC, Canada Background: Phylogenetic analyses of the interrelationships of viral sequences, using novel statistical tools, provide molecular epidemiological frameworks to reconstruct HIV transmission networks. We applied these methods to gain novel insights on HIV transmission patterns in Quebec, uncover cryptic at-risk populations, and elucidate epidemic drivers that cannot be identified by traditional epidemiological approaches. Methods: Genetic analyses were performed on subtype B pol sequences derived from newly-infected Men having Sex with Men (MSM, n=4800) and Heterosexuals subgroups, including People who Inject Drugs (PWID) and Migrants from Haiti and the Americas (n=1836). Phylogenetic analyses were also conducted on non-B viral subtypes originating fromMigrants from Africa, Asia and Europe (n=1578). Growth trajectories of transmission networks (6+ members/cluster) were analyzed using Maximum-Likelihood in MEGA10 and/or HIV-TRACE (TRAnsmission Cluster Engine) platforms. Results: Half of new infections (n=2328) among MSM segregated as solitary “dead-end” transmissions (n= 1478) or small transmission networks having 2–5 members/cluster (n=850). The remaining half of new infections (n=2371) were in large transmission networks (6–150 members, mean 42 members/ cluster). Phylodynamics showed a marked decline in singleton transmissions and small cluster outbreaks post-2008, concomitant to advances in Treatment- as-Prevention paradigms. This was offset by an increase in large cluster transmissions rising from 37% of infections in 2004 to 65% of new infections among MSM in 2017. HIV-TRACE maps showed differential features of forty large cluster sub-epidemics (members’ age, sex, sequence diversity). Heat maps of individual clusters distinguished “actively-growing” clusters and “newly emerging” clusters from older low-risk clusters. Phylogenetics uncovered the cryptic introduction and spread of subtype B and non-B subtype sub-epidemics in recent migrants to the province. Conclusion: The ability to predict, identify and respond to emerging “active” HIV transmission clusters in close to real-time may inform public health interventions to avert transmission cascades and control the HIV epidemic. 919 NON-B SUBTYPE HIV INFECTIONS IN GERMANY BEFORE AND AFTER THE EUROPEAN MIGRANT CRISIS Melanie Stecher 1 , Antoine Chaillon 2 , Christoph Stephan 3 , Jan-Christian Wasmuth 4 , Josef Eberle 5 , Georg M. Behrens 6 , Julia Roider 7 , Christoph D. Spinner 8 , Matthias C. Müller 9 , Elena Knops 10 , Guido Schäfer 11 , Sanjay R. Mehta 2 , Joerg Janne Vehreschild 1 , Martin Hoenigl 2 , for the Translational Platform HIV (TP-HIV) Molecular Epidemiology Team 1 Cologne University Hospital, Cologne, Germany, 2 University of California San Diego, La Jolla, CA, USA, 3 University Hospital Frankfurt, Frankfurt, Germany, 4 Bonn University Hospital, Bonn, Germany, 5 Max von Pettenkofer-Institute, Munich, Germany, 6 Medizinische Hochschule Hannover, Hannover, Germany, 7 Medical

Poster Abstracts

917 HOMOPHILY IN THE SOCIOSEXUAL NETWORKS OF GAY AND BISEXUAL MEN Kiffer G. Card 1 , Zishan Cui 1 , Jordan M. Sang 1 , Heather L.Armstrong 1 , Nathan J. Lachowsky 1 , David M. Moore 1 , Robert S. Hogg 1 , for the Momentum Health Study 1 British Columbia Centre for Excellence in HIV/AIDS, Vancouver, BC, Canada Background: Kenyon & Delva (2018, “It’s the network, stupid”) argue that the elevated prevalence of sexually transmitted infections (STIs) in sub-populations is due to the structure of their socio-sexual networks. Homophily, which measures the degree to which individuals associate with those like themselves, has been regularly identified as a key determinant of socio-sexual network structure. Thus, we aim to describe patterns of homophily within the networks of gay and bisexual men (GBM). Methods: Sexually-active GBM, aged 16+, were recruited between 2/2012 and 2/2015 using respondent-driven sampling. Participants recruited up to six participants in their social or sexual networks. Homophily estimates (h), based on recruitment patterns, were calculated in RDSAT and ranged from -1.00 (completely heterophilous) to +1.00 (completely homophilous). Results: Among 774 GBM, high homophily (h>.50) was observed by HIV serostatus (Positive: h=.62; Negative: h=.31; unknown h=.03), gender (Cis-man: h=.59; Trans-man: h=.57), and age (Age < 30: h=.57; Age > 30: h=.55). Moderate homophily (h>.30) was observed for ethnicity (White: h=.39;

CROI 2020 344