Background: Although Chinese-origin Rhesus macaques (Ch RhMs) infected with simian immunodeficiency virus (SIV) have been used for many years to evaluate the efficacy of AIDS vaccines and therapeutics, the bio-clinical variability of such a nonhuman primate AIDS model was so far not established.
Methodology/principal findings: By randomizing 150 (78 male and 72 female) Ch RhMs with diverse MHC class I alleles into 3 groups (50 animals per group) challenged with intrarectal (i.r.) SIVmac239, intravenous (i.v.) SIVmac239, or i.v. SIVmac251, we evaluated variability in bio-clinical endpoints for 118 weeks. All SIV-challenged Ch RhMs became seropositive for SIV during 1-2 weeks. Plasma viral load (VL) peaked at weeks 1-2 and then declined to set-point levels as from week 5. The set-point VL was 30 fold higher in SIVmac239 (i.r. or i.v.)-infected than in SIVmac251 (i.v.)-infected animals. This difference in plasma VL increased overtime (>100 fold as from week 68). The rates of progression to AIDS or death were more rapid in SIVmac239 (i.r. or i.v.)-infected than in SIVmac251 (i.v.)-infected animals. No significant difference in bio-clinical endpoints was observed in animals challenged with i.r. or i.v. SIVmac239. The variability (standard deviation) in peak/set-point VL was nearly one-half lower in animals infected with SIVmac239 (i.r. or i.v.) than in those infected with SIVmac251 (i.v.), allowing that the same treatment-related difference can be detected with one-half fewer animals using SIVmac239 than using SIVmac251.
Conclusion/significance: These results provide solid estimates of variability in bio-clinical endpoints needed when designing studies using the Ch RhM SIV model and contribute to the improving quality and standardization of preclinical studies.