HIV adaptation to HLA-restricted immune responses - implications for vaccine design and evaluation
Mallal, S., Moore, C., Carvalho, F., Patterson, A., Liu, C., Goodridge, D., Sayer, D., James, I. and John, M. (2003) HIV adaptation to HLA-restricted immune responses - implications for vaccine design and evaluation. In: 10th Conference on Retroviruses and Opportunistic Infections, 10 - 14 February 2003, Boston, U.S.A.
Background: HIV and ancestral retroviruses have evolved under intense selective pressure from HLA restricted immune responses, consistent with recent evidence of HLA Class I associated viral polymorphisms at a population level. The efficacy of an HIV vaccine will depend on the breadth and strength of HLA restricted immune responses it elicits, and the extent to which the infecting HIV sequence has escaped those responses.
Methods: Full-length HIV genome sequencing and high resolution HLA typing is being undertaken on pre-ART samples in 249 sequential patients (pts) enrolled in the WA HIV Cohort Study since 1996. Analysis of the first 80 pts identified HLA associations in all coding regions. 1) The proportion of Class I associated polymorphic sites adapted to the relevant Class I allele present (non-consensus amino acid at sites with positive HLA associations or consensus amino acid at sites with negative associations) was calculated; and 2) an optimal vaccine was designed for this population using the population consensus at positive HLA associations and the second most common amino acid at negative associations. A simulation was done to determine the likely efficacy of different vaccine candidates assuming the target population had the same HLA diversity, and was exposed to the same viral diversity, as in our cohort.
Results: 1) Pts with HIV sequences well adapted to HLA restricted immune responses had higher viral loads (mean 26,000 cps/ml in least adapted and 110,000 cps/ml in most adapted, p = 0.02); and 2) percentages of pts in the highest quartile for matches between the candidate vaccine and potential infecting viruses at residues likely to afford protection were: optimal 45%, our population consensus 35%, Clade B 28%, Clade C 14%, Clade A 25% and SIV 2% (p < 0.05). Similar results were obtained if the strength of HLA-restricted immune responses for these matches was estimated using the relevant data obtained in our HIV infected cohort.
Conclusions:These data demonstrate a rational approach to vaccine development guided by associations between HLA Class I and viral polymorphisms at a population level. It is anticipated that the number and strength of HLA associated polymorphisms, the degree to which these explain pre-treatment viral load variability, and the ability to refine and evaluate an optimal vaccine for this population should improve as data on a larger number of pts becomes available.
|Publication Type:||Conference Item|
|Murdoch Affiliation:||Centre for Clinical Immunology and Biomedical Statistics|
|Item Control Page|