15th Conference on Retroviruses and Opportunistic Infections Boston, Massachusetts - February 3-6, 2008

Conf Retrovir Opportunistic Infect 2008 Feb 3-6;15: (abstract no. 23)

Zabrina Brumme ¹, C Brumme1, H Streeck¹, J Carlson^2,3, M Markowitz^4,5, H Jessen⁶, A Kelleher⁷, M John⁸, D Heckerman², and B Walker^1,9
1Partners AIDS Res Ctr, Boston, MA, US; ²Microsoft Res, Redmond WA, US; ³Univ of Washington, Seattle, US; ⁴Aaron Diamond AIDS Res Ctr, The Rockefeller Univ, New York, NY, US; ⁵Rockefeller Inst, New York, NY, US; ⁶Jessen-Praxis, Berlin, Germany; ⁷Univ of New South Wales, Sydney, Australia; ⁸Murdoch Univ, Perth, Australia; ⁹Howard Hughes Med Inst, Chevy Chase, MD, US

BACKGROUND: The HLA class I-restricted cytotoxic leukocyte (CTL) response plays an important role in HIV immune control during acute infection; however, selection of escape mutations may compromise long-term antiviral control by CTL. We define rates of recognition and escape among published CTL epitopes in the first year of infection to identify the most frequently targeted and most rapidly evolving epitopes in Gag/Pol/Nef.

METHODS: Kaplan-Meier methods were used to characterize selection of human leukocyte antigen (HLA) -associated substitutions within published, optimally defined epitopes in Gag/Pol/Nef (n=162) in the first year of infection in a longitudinal, untreated, subtype-B-infected seroconverter cohort (n=98). “HLA-associated substitutions” were defined in an independent, cross-sectional, phylogenetically corrected analysis of a large chronic untreated cohort (n>1200). Escape rates were correlated with CTL recognition frequencies of these epitopes as measured by interferon-gamma (IFN-γ) ELISpot in a cohort of 289 untreated persons in primary infection.

RESULTS: Evidence for HLA-associated HIV evolution was observed in 68 of 162 (42%) epitopes. Of the 10 most rapidly evolving epitopes, 5 were restricted by “protective” HLA-B alleles (B13/51/57/58), consistent with immunodominant CTL responses in primary infection. B57 TW10-Gag was the most rapidly evolving epitope, with 80% of B57+ persons selecting T242N or G248A within the first 6 months, corresponding to a crude first-year “epitope escape rate” of 38% per month. B57 epitope escape rates fell in the order TW10-Gag > IW9-RT > HW9/HQ10-Nef > IW9-Gag > KI8-Gag, with all but KI8 falling in the upper half of the distribution of evolving epitopes. B27 KK10-Gag escaped at a rate of 11% per month, with most substitutions being L268M. A significant positive correlation was observed between frequencies of recognition vs escape in Gag (Spearman R=0.5, p=0.003) and Pol (R=0.8, p=0.0008), but not in Nef epitopes, indicating that escape rates generally reflect the level of CTL pressure on these epitopes in primary infection.

CONCLUSIONS: Marked differences in the kinetics of escape within CTL epitopes correlate with frequency of epitope targeting during primary infection. The finding that certain HLA alleles remain protective despite rapid selection of mutations within targeted epitopes underscores our need to assess the consequences of CTL-driven evolution on HIV disease progression. The inclusion of rapidly selected escape variants into immunogen design deserves consideration.

Download abstract

2008-02-03
23

Copyright © 2008 - Foundation for Retrovirology and Human Health. Reproduction of this abstract (other than one copy for personal reference) must be cleared through the Foundation for Retrovirology and Human Health.