CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

528 INTEGRASE GENOTYPIC TESTING AND DRUG RESISTANCE AMONG NEW HIV DIAGNOSES IN NEW YORK Zhengyan Wang , Randall V. Collura, Mark Rosenthal, Jayleen K. Gunn, Joanne Gerber, Brenda Moncur, Bridget J. Anderson New York State Department of Health, Albany, NY, USA Background: HIV treatment guidelines state that genotypic resistance testing should be obtained at diagnosis. Integrase strand transfer inhibitors (INSTIs) have emerged as initial regimens for persons newly diagnosed with HIV because of their clinical effectiveness and tolerability. However, with widespread use of INSTIs, the concerns of transmitted integrase (IN) drug resistance and risk of virologic failure are rising among clinicians. The aims of this analysis were to explore 1) the frequency of IN testing and risk factors associated with IN testing, 2) the rate of transmitted IN drug resistance, and 3) common clinically significant INSTI-resistance mutations among persons with newly diagnosed HIV in New York State (NYS). Methods: Persons age 13 and older diagnosed between 2013-2017 and reported to the NYS HIV registry were included in the study. The first IN nucleotide sequence for an individual was identified and flagged as an “initial” test if ordered within 3 months of the HIV diagnosis date. Persons with 1) incomplete diagnosis or test dates or 2) invalid sequences were excluded. Multivariable analysis was used to test the association between IN initial testing and sociodemographic factors. Sequences were analyzed using the NYS in-house Resistance Analysis System and compared with major INSTI resistance mutations published on Stanford HIVdb Programwebsite. Results: Overall, 15,345 persons were included; 59.2% had any resistance testing within 3 months of diagnosis. 20.9% (3,209) had initial IN testing; 2.5% had only IN testing. Initial IN testing increased significantly from 5.6% in 2013 to 32.4% in 2017. The likelihood of having initial IN test was lower in minorities than whites (RR:0.87, 95%CI:0.79-0.96), and higher among males with a history of male-to-male sexual contact than heterosexuals (RR:1.31, 95%CI:1.09-1.58). Resistance to ≥1 IN drug was seen in 0.7% (24) of 3,209 persons with initial tests. The most common clinically significant INST-resistance mutations were: E138A/K, N155H/S, Q148H/K/R, E92G/Q, T66A/I, G140C/S, Y143C/R. Conclusion: Clinician ordering of initial resistance testing lags current guidelines. These data indicate that initial IN testing has increased among persons newly diagnosed with HIV. While IN drug resistance remains low, clinically significant major mutations observed suggests that transmitted IN resistance is emerging; it is importance for clinicians to order IN test at time of HIV diagnosis for treatment decision. 529 HIV-TRANSMITTED DRUG RESISTANCE IN CISGENDER MSM AND TRANSGENDER WOMEN IN LIMA, PERU William L. Trebelcock 1 , Javier R. Lama 2 , Ann Duerr 3 , Hugo Sanchez 4 , Robinson Cabello 5 , Sari Reisner 6 , Kenneth H. Mayer 7 , James Mullins 1 , Rachel A. Bender Ignacio 1 1 University of Washington, Seattle, WA, USA, 2 Asociacion Civil Impacta Salud y Educacion, Lima, Peru, 3 Fred Hutchinson Cancer Research Center, Seattle, WA, USA, 4 Epicentro, Lima, Peru, 5 Asociacion Via Libre, Lima, Peru, 6 Harvard University, Boston, MA, USA, 7 The Fenway Institute, Boston, MA, USA Background: Transmitted drug resistance (TDR) mutations threaten the efficacy of first-line antiretroviral therapy (ART) in individuals initiating treatment. In Peru, genotypic resistance profiling is not routinely performed at ART initiation, and administration of a partially effective regimen can select for further resistance and lead to virologic failure. In Peru, previously reported TDR prevalence ranged from 1.0 – 4.7% as last reported before 2012. Methods: We obtained HIV sequence data from 3 parent studies conducted in 2013 – 2017 of ART naïve cisgender men who have sex with men (cis-MSM; n=332) and transgender women (TW; n=144) in Lima, Peru. Consensus gene sequences of the 2,510 – 3,209 region of HIV pol (not codifying the entire protease and integrase genes) were interrogated for TDR using the Stanford HIVdb interpretation algorithm and scored for resistance to common nucleoside reverse transcriptase inhibitors (NRTIs) and non-nucleoside reverse transcriptase inhibitors (NNRTIs). We calculated binomial proportions with a 95% confidence interval. χ² and Fisher’s exact tests or generalized linear models were used to examine possible predictors of TDR. Results: Eighty (16.8%) of 476 individuals had TDR (95% CI: 13.6, 20.5). Twenty- two unique base changes totaling 94 TDR mutations were present. Mutations conferring resistance to NNRTIs represented 88% of total TDR, and prevalence of a singular mutation (15.1%) was more common than 2 (1.3%), or 3+ (0.4%)

sequence reporting and the prevalence of transmitted drug resistance- associated mutations (TDRMs). Methods: For persons with HIV infection diagnosed during 2013−2016 and no evidence of prior antiretroviral therapy use, we analyzed sequences collected within 3 months of diagnosis and reported to NHSS by 12/2017. We included states in which ≥20% of HIV diagnoses during the 4-year period had an analyzable sequence and defined TDRMs using the CDC HIV-1 surveillance mutation list. We examined reporting by sequence type, prevalence of TDRMs and temporal trends for sequence types reported and TDRMs detected from 2013−2016. Results: The 23 states reported sequences for 36,288 (32%) of 113,121 HIV diagnoses from 2013–2016. Among persons with eligible sequences, prevalence of IN sequences obtained increased from 3.7% in 2013 to 23.0% in 2016 while prevalence of PR/RT sequences decreased from 99.2% to 93.0%. TDRMs were detected for 6,880 (19.0%) sequences, including TDRMs to non- nucleoside reverse transcriptase inhibitors (NNRTIs) (11.9%), nucleoside reverse transcriptase inhibitors (nRTIs) (6.8%), protease inhibitors (PIs) (4.3%), and INSTIs (0.8%). INSTI TDRM prevalence did not differ by sex, age group, or race/ ethnicity. Prevalence was low for TDRMs to 2 drug classes (2.4%) or ≥3 drug classes (0.3%). TDRM prevalence increased from 2013 to 2016 for NNRTIs (11.3% to 12.4%, p=0.012) and INSTIs (0.8% to 1.1%, p=0.041) but not for other drug classes. Conclusion: NNRTI TDRM prevalence continues to increase, outpacing all other HIV drug classes. During this period of increasing INSTI use (and IN sequence reporting) INSTI TDRM prevalence also increased. Though drug resistance testing based on PR/RT gene sequencing is recommended for all new HIV diagnoses, an increasing proportion have only an IN sequence reported, precluding detection of TDRMs for nRTIs, which remain a critical backbone of multidrug therapy. 527 IMPACT OF PRETREATMENT DRUG RESISTANCE ON TREATMENT OUTCOME IN THE ITREMA TRIAL Lucas E. Hermans 1 , Laura Marije Hofstra 1 , Rob Schuurman 1 , Rob ter Heine 2 , Hugo Tempelman 3 , Willem D. Venter 4 , Monique Nijhuis 1 , Annemarie Wensing 1 1 University Medical Center Utrecht, Utrecht, Netherlands, 2 Radboud University Medical Center, Nijmegen, Netherlands, 3 Ndlovu Care Group, Groblersdal, South Africa, 4 Wits Reproductive Health and HIV Institute, Johannesburg, South Africa Background: Prevalence of pre-treatment drug resistance (PDR) in sub- Saharan Africa has risen during scale-up of antiretroviral treatment (ART) and may result from either exposure to previous ART or infection with resistant viral strains. We assess prevalence of PDR and its impact on treatment outcomes in the first year of ART. Methods: The ITREMA open-label randomized clinical trial (ClinicalTrials registration NCT03357588) compares treatment monitoring approaches in response to viral rebound in rural South Africa. Of 501 participants, 294 were on stable first-line ART, and 207 initiated first-line ART. For these 207, plasma collected prior to initiation was analysed batchwise. Population-based RT sequencing was performed. PDR was defined as detection of at least one 2017 IAS-USA listed major mutation. Viral load testing was performed at week 24 and week 48 of ART, and annually thereafter. Logistic regression adjusted for gender, age and baseline CD4-count was used to estimate adjusted odds ratios (aOR) for viral rebound (viral load ≥1000 copies/mL) within the first year of ART. Results: All 207 newly initiated patients received efavirenz-based ART. 60.4% (125/207) were female. Median age was 38.8 years (IQR: 31.4–46.7). Median CD4-count at ART initiation was 191 cells/mm3 (IQR: 70–355). 194 patients had a baseline sample with viral load >250 copies/mL available for sequencing. PDR was detected in 12.9% (25/194). 20.6% of patients (34/165) with available follow-up had viral rebound during the first year of ART. Patients with PDR more frequently experienced rebound (53.3% versus 17.4%, p=0.003). 13 patients reported prior use of ART, which was associated with PDR (aOR 1.37 [95%CI: 1.13–1.67], p=0.0017). When correcting for sex, age, baseline CD4 and disclosed previous ART exposure, PDR remained associated with viral rebound (aOR 1.42 [1.22–1.64], p<0.0001). Upon differentiation between NNRTI-PDR and dual- class PDR, dual-class PDR was strongly associated with viral rebound (aOR 2.56 [2.00–3.27], p<0.0001), but NNRTI-PDR was not (aOR 1.12 [0.96–1.31], p=0.16). Conclusion: PDR was detected in 13% of patients initiating first-line ART in this study. Dual-class PDR increased the risk of viral rebound, but solitary NNRTI-PDR did not. Reported prior ART use increased the risk of PDR. Efforts to uncover previous ART use should be made before initiating first-line treatment.

Poster Abstracts

CROI 2019 198