Dear Editor: We estimated the prevalence of transmitted drug resistance mutations (TDRMs) and drug-relevant polymorphisms (DRPs) in primary HIV infection in a cohort of youth (ages 18–24) and determined whether demographic or clinical characteristics were associated with prevalence of TDRMs or DRPs. We report the results of a substudy of Adolescent Medicine Trials Network for HIV/AIDS Interventions (ATN) 061, an ongoing study of highly active antiretroviral therapy (HAART) initiation in HIV-positive youth who have CD4+ T cells (CD4) of>350 cells/mm3 with subsequent deintensification. Subjects were screened from 15 ATN U.S. and Puerto Rico sites and from 7 International Maternal Pediatric Adolescent AIDS Clinical Trials (IMPAACT) U.S. sites. The study was approved by each participating institution's Institutional Review Board. Subjects were eligible for ATN 061 if they were 18–24 years of age, had a CD4 T-cell count>350/mm3, HIV-1 RNA viral load (VL)≥1000 copies/mL, and were HAART-naive with the exception of ARVs for the prevention of mother to child transmission for a single pregnancy with documentation of virologic suppression (n=5). Enrollment began in 2006 and was completed in 2010. All subjects had a screening genotype and those without specific mutations that were exclusion criteria for ATN 061 (I50L; I84V; N88D/S, D30N; V32I; L33I/F/V; M46I/L; I47V/A; G48V; I50V/L; I54V/L/A/M/T/S; L76V; V82A/F/T/S/L; L90M, or major RT mutations Q151M and 69 insertion complex) were eligible. The screening genotypes from all patients were evaluated for protease inhibitor (PI), non-nucleoside reverse transcriptase inhibitor (NNRTI), and nucleoside/nucleotide reverse transcriptase inhibitor (NRTI) major transmitted drug resistance mutations (TDRMs) and drug relevant polymorphisms (DRPs) according to World Health Organization (WHO) and International AIDS Society (IAS) algorithms.1,2 Alone, DRPs may not confer resistance, however, they may lead to compensatory changes that may confer a fitness advantage to the virus. All of the 130 eligible subjects had screening genotypes. The median age was 21.4 years, 86.9% were male, 69.2% black, and 99.2% acquired HIV sexually (Table 1). Twelve of 130 subjects (9.2%) had a total of 24 TDRMs: 5 of 24 (20.8%) NRTI; 9 of 24 (37.5%) NNRTI; and 10 of 24 (41.7%) PI. Two of 12 subjects with TDRMs (16.7%) had single, 8 of 12 (66.7%) double, and 2 of 12 (16.7%) triple class resistance; 10/130 (7.7%) had 2 or more class resistance. No factors were significantly associated with identification of TDRMs. Table 1. Demographic and Clinical Characteristics of Patients with Transmitted Drug Resistance Mutations and Drug Relevant Polymorphisms DRPs were found in 105 of 130 (80.8%) subjects including the 12 subjects with TDRMs who all had additional DRPs (Tables 1 and and2).2). Geographic region was significantly associated with having a DRP (Table 2). Due to a significant interaction of VL with the MSM HIV acquisition risk category (approximately 0.3 log10 higher VL in MSM risk group, p=0.0448), two multivariate models were created, one that includes HIV VL with geographic region and a separate model that includes HIV acquisition risk with geographic region. In both models, Midwestern and Western geographic regions were associated with a decreased likelihood of DRPs as compared with the South (adjusted odds ratio [AOR] 0.166 [95% confidence interval {CI} 0.047–0.591] and AOR 0.093 [95% CI 0.024–0.360]), respectively). For each log10 increase in VL, the odds of having a DRP increased by 2.118 (AOR 2.118 [95% CI 1.033–4.344]). Additionally, MSM HIV acquisition risk category versus heterosexual acquisition risk was independently associated with an increased likelihood of having DRPs (AOR 4.016 [95% CI 1.251–12.890]). No other demographic or clinical associations were identified. Table 2. Factors Associated with the Presence of Drug Relevant Polymorphisms at Baseline in the Univariate Analysis In this geographically diverse study of U.S. youth, 9.2% had TDRMs and 7.7% had multiclass resistance. These rates are an underestimation as youth known to have harbored particular resistance mutations were excluded by the parent study and not captured by the substudy. The multiclass resistance rate was higher than previously reported from similar sites.3 Reported rates of DRPs have ranged from 60% to 90%, in line with the 80.8% rate we found.4,5. The significance of these high DRP prevalence rates is unclear, however, it may be important to continue to monitor as there may be an impact on treatment response to both current and future PIs. Due to their potency and higher genetic resistance barrier, PI-based regimens are particularly important for youth who tend to have lower treatment adherence.6 Additionally, as youth are a group where HIV vaccines may be particularly targeted both for treatment and primary prevention, high prevalence rates of polymorphisms may also be of significance. We report that MSM HIV acquisition risk and higher screening VL were associated with an increased risk of having DRPs, while being in the Midwest and West was associated with a decreased risk of having DRPs compared to the South. Other studies of resistance in youth have not shown differences by geographic location.3,7 However, the Southern region encompasses locales (e.g., Florida) with high incidence and prevalence of HIV infection8; the finding of significant regional differences in the DRP rates may be of concern and in need of further evaluation. Increasing drug resistance among newly infected adult MSM has been reported; and 45% of known HIV-positive and previously treated adult men (age 26–64) in a community sample in a large, urban city were found to harbor resistance to at least one ARV9. The increased risk of DRPs among those with MSM HIV acquisition risk in this study is noteworthy as this mode of transmission is responsible for increasing proportions of infections among behaviorally infected youth; and age bridging (having sex with older men) is a known risk factor for HIV acquisition.10 Many of these older individuals may have long treatment histories and thus antiretroviral exposure with risk of transmitting viral variants with resistance. The association of increasing viral load with DRPs in this study is of concern as it may reflect greater fitness of viral variants with resistance mutations, as well as higher or possibly more efficient risk of transmission of resistant variants in those individuals with increased viremia.11 As this study was not designed to ascertain cause and effect behind this association, this instead may also reflect that viral quasispecies with higher replication capacities are likely to be more error prone. Our study is limited by relatively small numbers, which diminishes our power to examine significant associations. Primary drug resistance is commonly found among HIV-infected youth, with a significant proportion acquiring and harboring multi-drug resistant viruses. These findings are worrisome, underscoring the importance of genotypic testing prior to ARV initiation for treatment of HIV infection, particularly if ARV pre- and postexposure prophylaxis becomes more frequently prescribed as part of HIV prevention.12 The significance of the high prevalence of DRPs is unclear and needs further study.