CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

show no significant differences in partnership risk behaviours in the four cluster groups. SPOT cohort data showed an inverse relationship between partnership risk behaviour and testing propensity. Conclusions: The Montreal MSM epidemic is sustained by primary/recent stage infection with an ongoing genesis of large cluster viral lineages (median 16 infections/cluster) showing protracted infectivity. Overall 7% of lineages account for 70% of onward transmissions. These findings substantiate the need for frequent testing among increasingly younger and mixed risk MSM populations. 252 Sources of HIV-1 Transmission in the Ongoing, Concentrated HIV Epidemic Among Men Having Sex With Men in the Netherlands Between July 1996 and December 2010 Oliver Ratmann 1 ; Ard van Sighem 2 ; Daniela Bezemer 2 ; Alexandra Gavryushkina 3 ; Peter Reiss 2 ; Frank deWolf 1 ; Christophe Fraser 1 1 Imperial College London, London, United Kingdom; 2 Stichting HIV Monitoring Foundation, Amsterdam, Netherlands; 3 University of Auckland, Auckland, New Zealand Background: The HIV epidemic among men having sex with men (MSM) remains out of control. To better target HIV prevention efforts, it is critical to quantify the proportion of HIV transmissions that originate throughout the HIV treatment cascade, from undiagnosed to treated individuals. Methods: We conducted a combined analysis using anonymized, molecular genetic and clinical data from HIV infected individuals in care in the Netherlands between 1996 and 2013. Patients were followed at high frequency in the open, national opt-out, clinical ATHENA cohort. Partial HIV-1 pol sequences were collected for 46% of all MSM in care. Using viral evolutionary analyses, we determined potential transmitters to 667 recipient MSM that were diagnosed with recent HIV infection up to December 2010. Using clinical data, we associated treatment cascade stages with potential transmission intervals. Results: A total of 1,660 person-years of potential transmission intervals were associated with phylogenetic evidence for direct HIV-1 transmission. Overall, the estimated proportion of transmissions from undiagnosed individuals decreased from 66% (95% confidence interval: 52-79%) fromMay 2006 to December 2007 to 58% (43%-74%) from July 2009 to December 2010, while the estimated proportion of transmissions from individuals after ART initiation increased from 8% (4-12%) to 17% (12-22%) in the same observation periods. Considering the most recent period July 2009 to December 2010, 15% (10-20%) of all transmissions are estimated to originate from undiagnosed individuals with recent HIV infection at diagnosis, 24% (19-29%) from those undiagnosed with chronic HIV infection at diagnosis, 19% (14-25%) from those undiagnosed with no data on recency of HIV infection, 4% (2-5%) from diagnosed, recently infected individuals within the first 3 months after diagnosis, 6% (4-8%) from diagnosed untreated individuals with CD4 count >500, 15% (12-19%) from diagnosed, untreated individuals with CD4 count <500, and 10% (7-12%) from treated individuals with a viral load above 50 copies/ml plasma blood. Conclusions: Due to its national opt-out policy, the open, clinical ATHENA cohort enabled the worldwide largest analysis to date into the sources of HIV infection amongst MSM. A combined approach appears to be required to bring this HIV epidemic under control, including expanded testing, universal ART coverage, and frequent monitoring of the treated population. 253 A Direct Comparison of Two Densely SampledWestern European HIV Epidemics: The UK and Switzerland Manon L. Ragonnet-Cronin 1 ; Mohaned Shilaih 2 ; Huldrych F. Günthard 2 ; Jurg Boni 2 ; SabineYerly 3 ;Valerie Delpech 4 ; David Dunn 5 ; Roger Kouyos 2 ; Andrew J. Leigh Brown 1 Swiss HIV Cohort Study/UK HIV Drug Resistance Database 1 University of Edinburgh, Edinburgh, United Kingdom; 2 University Hospital Zurich, Zurich, Switzerland; 3 University Hospital Geneva, Geneva, Switzerland; 4 Public Health England, London, United Kingdom; 5 MRC CTU at UCL, London, United Kingdom Background: The UK and Swiss (CH) HIV epidemics have both historically been driven by transmission of subtype B among men who have sex with men (MSM). The CH population is 1/8 the size of the UK and HIV prevalence in CH is nearly double that of the UK. Both epidemics are densely sampled, by the UK HIV Drug Resistance Database and the Swiss HIV Cohort Study respectively. Previous independent analyses have suggested dramatically different epidemic dynamics. Methods: A bioinformatics pipeline to compare HIV transmission patterns was developed which included: maximum likelihood phylogenetic trees for subtype A1, B and C pol sequences against a background of global sequences; cluster detection at a range of bootstrap (70%-95%) and genetic distance (1.5% and 4.5%) thresholds; analysis of degree distributions by risk group and characterisation of HIV import for each country independently. We use univariate and multivariate logistic regression to predict cluster membership based on country, sampling date, risk group, ethnicity and sex. Results: We analysed >8000 subtype B sequences from CH and >25000 from the UK. A genetic distance of 1.5% yielded mainly pairs in both. Clustering was much higher among UK sequences at all thresholds (35% vs 15% on average, p<10 -91 ) suggesting that the UK database is more likely to capture transmitting partners. The number of links for each clustered sequence (degree distribution) followed a power law in both, but shifted downwards for CH relative to the UK. Down sampling the UK dataset to match the size of the CH dataset revealed a more similar degree distribution (p<10 -5 vs p<10 -12 ), remaining higher for the UK in MSM (p<10 -8 ), but not for heterosexuals (p=0.2). Adjusting for sampling time in the logistic regression models reduced the clustering odds-ratio between countries from 3.1 to 2.2. Both countries showed extensive intermixing with other European countries (80% of direct links for CH and 60% for UK), and the UK also displayed linkage with other Anglophone countries (Australia, Canada and USA, 26% of links). Within Europe, Spain was the most frequently linked country (12% and 15% of all links for the UK and CH, respectively). Conclusions: In conclusion, epidemic size, insularity and sampling time play a part in explaining the differences in clustering patterns between the CH and UK epidemics. Reduced clustering in CH is most significant among MSM, who seemmore likely to have acquired the virus outside the country.

Poster Abstracts

THURSDAY, FEBRUARY 26, 2015 Session P-B5 Poster Session

Poster Hall

2:30 pm– 4:00 pm Next Generation of Next-Generation Sequencing 254 Present Applications of a High-Throughput, Single Measure HIV Genomic Incidence Assay SungYong Park 1 ;Tanzy Love 2 ; Nolan Goeken 1 ; Robert Bolan 3 ; Alan S Perelson 4 ; Michael Dube 1 ; Ha Youn Lee 1 1 Keck School of Medicine at University of Southern California, Los Angeles, CA, US; 2 University of Rochester School of Medicine and Dentistry, Rochester, CA, US; 3 Los Angeles Gay and Lesbian Center, Los Angeles, CA, US; 4 Los Alamos National Laboratory, Los Alamos, CA, US Background: Annual HIV incidence is the primary assessor for monitoring the epidemic’s rise and decline. In pursuit of an accurate assay, we have proposed a genomic incidence assay utilizing fingerprints harbored in the HIV sequence population, which showed over 95% accuracy among 182 incident and 43 chronic samples. Still, two major hurdles must

224

CROI 2015