416. Comparison of World Health Organization and Stanford Drug Resistance Criteria in HIV-1 Genotypic Testing
Session: Poster Abstract Session: HIV - Antiretroviral Therapy
Friday, October 21, 2011
Room: Poster Hall B1
Background: Most commercially-available genotypic testings for HIV-1 utilize the Stanford algorithm to interpret drug resistance. The WHO recommends their genotypic definition of HIV-1 drug resistance for surveillance. HIV-1 drug resistance presenting as transmitted drug resistance mutations (TDRM) increase the risk of virologic failure. We compared differences in the prevalence of TDRM and virological response to combination antiretroviral therapy (cART) with interpretation of HIV genotypes using these two different methodologies.
Methods: This was a retrospective study of 801 treatment naïve HIV-1-infected patients from 2001 to 2009 who had pre-cART genotypic testing results available to guide initial antiretroviral selection. The prevalence of TDRM was assessed by both the WHO definition and Stanford algorithm. The impact of TDRM on virologic failure (HIV RNA level > 400 copies/mL) at 48 weeks was assessed by multivariate Cox proportional hazards regression models.
Results: The Stanford algorithm identified a greater number of subjects with TDRM than the WHO definition (18% vs. 14%, p=0.031). Specifically, nucleoside reverse transcriptase inhibitor (NRTI)-resistance was more prevalent (8% vs. 5%, p=0.015) . No differences were seen with non-nucleoside reverse-transcriptase inhibitor (NNRTI)-resistance or protease inhibitor (PI)-resistance between the two methodologies (both p>0.05). Among 611 patients started on cART, virologic failure did not differ (p=0.96) for those with resistance identified by either Stanford algorithm (37%, n=50/135) or WHO definition (38%, n=42/110). When adjusted by age, sex, CD4 cell count, HIV RNA level and cART regimen (PI vs. NNRTI-based regimen), TDRM interpreted by the Stanford algorithm and the WHO definition both conferred a1.36-fold increased risk of virologic failure.
Conclusion: The prevalence of TDRM was significantly lower when interpreted by the WHO definition, but predicted virologic failure with the same likelihood as the Stanford algorithm. Overall, patients with TDRM had a 1.36-fold increased risk of virologic failure despite administration of a cART regimen guided by the genotypic testing. Use of the Stanford algorithm may overestimate the prevalence of TDRM.

Subject Category: H. HIV/AIDS and other retroviruses

Toshibumi Taniguchi, MD, Jessica Grubb, MD, Nur Onen, BSc, MBChB and E. Turner Overton, MD, Division of Infectious Diseases, Washington University School of Medicine, St. Louis, MO

Disclosures:

T. Taniguchi, None

J. Grubb, None

N. Onen, None

E. T. Overton, Gilead: Consultant, Consulting fee
Bristols-Myers Squibb: Consultant, Consulting fee
Glaxo-Smith-Kline: Consultant, Consulting fee
Tibotec: Speaker's Bureau, Research support
Merck: Consultant, Consulting fee
Monogram Science: Consultant, Consulting fee
Boehringer Ingelheim: Consultant, Consulting fee
Abbott: Research Contractor, Research support

Findings in the abstracts are embargoed until 12:01 a.m. EST Thursday, Oct. 20 with the exception of research findings presented at IDSA press conferences.