CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

system based on recently acquired HIV infections. The aim is to assess the current prevalence of transmitted drug resistant (TDR) variants and HIV-1 subtypes in the main HIV- transmission groups: men who have sex with men (MSM), women/men with heterosexual contacts (HET) and persons with intravenous drug use (PWIDs) with respect to their origin and place of infection. Methods: Newly diagnosed cases are reported to the RKI as a statutory duty for anonymous notification. Diagnostic laboratories provide dried serum spot (DSS) of ~60% of all newly diagnosed HIV infections reported. DSS serologically classified as “recently acquired infections” (<140 days; BED-CEIA, Sedia) were genotyped in the HIV- pol -region to identify TDR and to determine the HIV-1 subtype. The results are linked to notification data from the report. Results: In 2013 and 2014 a total of 1,963/6,371/DSS originated from a recent infection. Of these, 881 were successfully sequenced and analysed. Total TDR was 10.7%, comprising 4.3%with mono resistance to nucleotide reverse transcriptase inhibitors (NRTIs), 2.7% to non-NRTIs, 2.7% to protease inhibitors and 0.6% and 0.3%with dual and triple class resistances, respectively. HIV-subtype B was most prevalent with 76.2%. Non-B infections were identified more often in HET compared to PWIDs or MSM (79%; 39%; 12%, all p<0.05). Non-B subtypes were also more frequently found in patients originating in countries other than Germany (49% vs.15%; p<0.05) and in patients infected outside of Germany (65% vs.14%; p<0.05). Conclusions: TDR prevalence in recent HIV infections among notified newly diagnosed HIV patients in Germany remained high (>10%) in 2013/2014 and is comparable to other European countries, including with regard to the proportions of resistance classes. Therefore, genotypic resistance testing of HIV prior to first-line treatment should be continued. Our data also demonstrate that subtype B infections remains the most frequently transmitted subtype in the country based on its high prevalence in MSM. 486 Influence of Transmitted Drug Resistance on CD4 Decline Among ART-Naïve HIV Patients Anna Schultze 1 ; CarloTorti 2 ; Alessandro Cozzi-Lepri 1 ; Anne-MiekeVandamme 3 ; Maurizio Zazzi 4 ; Helen Sambatakou 5 ; Andrea De Luca 4 ; Anna Maria Geretti 6 ; Anders Sönnerborg 7 ; Giuseppe Lapadula 8 1 Univ Coll London, London, UK; 2 Univ Magna Graecia, Catanzaro, Italy; 3 Katholieke Universiteit Leuven, Leuven, Belgium; 4 Univ of Siena, Siena, Italy; 5 Hippokration General Hosp, Univ of Athens, Athens, Greece; 6 Inst of Infection & Global Hlth, Univ of Liverpool, Liverpool, UK; 7 Karolinska Inst, Stockholm, Sweden; 8 Infectious Diseases Clinic, San Gerardo Hosp, Monza, Italy Background: The presence of transmitted drug resistance mutations (TDRM) may influence the natural history of HIV infection. We evaluated the effect of TDRM on CD4 count decline in a large European cohort collaboration. Methods: Data from several European HIV clinics (ViroLAB, EuResist and EuroSIDA contributing clinics; Royal Free and St Mary’s Hospital, London; University of Bari) were merged. Individuals were included if they were aged ≥18 and had ≥1 CD4 count and ≥1 genotypic resistance test before starting antiretroviral therapy (ART). Baseline was defined as the date of the first available CD4 count. TDRM were identified using the WHO 2009 surveillance list, and were presumed to have been present since the time of infection. Linear mixed models with a random intercept and slope were used to estimate the effect of TDRM on CD4 slopes. Results: 6326 individuals were included: 74%were male and 65% infected with a subtype B virus. The median follow-up was 1.2 (IQR=0.07-3.4) years. Overall, 623 individuals (9.9%) had at least 1 TDRM (NRTI:7.0%, NNRTI:3.0% and PI:2.5%). The most common mutations were thymidine analogue mutations for the NRTIs (5.9%), K103N for the NNRTIs (1.8%) and L90M for the PIs (0.9%). The median baseline CD4 count was 418 (IQR=284-580) cells/mm 3 , and there was no evidence that this differed according to the detection of TDRM (426 in those with TDRM v. 417 in those without, p=0.14). The viral set point (median of the mean distribution of all pre-ART RNA measurements) was 4.4 log 10 cp/ml among individuals with TDRM and 4.5 among those without (p=0.07). In unadjusted models, the overall estimate of CD4 decline was 54 cells/year in the whole population; 56 cells/year among those with TDRM and 54 cells/year among those without (difference=-2.30 cells/year, 95%CI=-9.67; +5.03, p=0.54). After adjustment for potential confounders (Table 1), there was no evidence to suggest that the rate of CD4 decline differed according to TDRM presence (p=0.29). There was also no evidence to suggest that CD4 slopes differed according to the class of resistance present (Table 1). 71 (1.1%) individuals had M184V, and we could not find any evidence that this was associated with baseline CD4 counts (p=0.15) or CD4 slopes (p=0.68). Conclusions: In one of the largest European datasets of resistance tests results from ART-naive individuals, we were not able to find any evidence supporting the hypothesis that the rate of CD4 decline in the absence of ART is different between patients with and without TDRM.

Poster Abstracts

487 No Evidence of Sexual Transmission of Minority HIV Drug Resistance Mutations in MSM Antoine Chaillon ; Sanjay R. Mehta; Joel O.Wertheim; Ben Murrell; Susan J. Little; Douglas D. Richman; Davey M. Smith; Sara Gianella Univ of California San Diego, San Diego, CA, USA Background: During primary HIV infection, the detection of minority drug resistant mutations (DRM), defined as <20% of sampled population, may be a consequence of (i) sexual transmission, (ii) de novo mutations, and/or (iii) technical errors. Methods: Baseline blood samples were collected from 32 HIV+ antiretroviral-naive, phylogenetically- and epidemiologically-linked male source and recipient partners, within a mean of 64 days (range: 11-170) after the recipient’s estimated date of infection; 23 longitudinal samples were available for 11 recipients over a mean of 267 days (range: 11-1,520). Next generation sequencing (NGS) of HIV reverse transcriptase (RT) was performed (Roche 454); filtered reads were screened for nucleoside and non-nucleoside RT inhibitor DRM (Stanford >35). NGS error was estimated using confidence limits for %DRM and a binomial model was used to determine background error rate for each site. The likelihood of sexual transmission of minority DRM was assessed using Bonferroni-adjusted paired t-tests. We estimated the site-wise mutation rate, selection (dN/dS ratio), and linkage between minority DRM and accessory mutations (AM). Longitudinal persistence of minority DRM was assessed by mixed-effects regression analysis adjusting for HIV RNA levels. Results: NGS identified minority DRM in baseline samples from all sources and recipients (mean: 3, range: 1-13), with a total of 139 DRM from 22 sites (9 NNRTI, 13 NRTI) and an average frequency of 4.07% (range: 0.02-15%with average NGS error of 0.0017 mutations/site). In samples evaluated shortly after infection, we found: 1) no association between the presence of minority DRM in the source and recipients (all p>0.1), 2) no increased mutation rates (permutation test p >0.05), 3) no enrichment for diversifying or purifying

189

CROI 2016