Ochieng', W: Immune and intrinsic correlates of protection i

Efficacy assessment of AIDS vaccines relies both on pre-clinically challenging immunised monkeys with a pathogenic virus and subsequent monitoring of infection rates in large human trials. Conventional parameters of vaccine-induced immune responses do not completely predict outcome. Moreover, existing methods for testing cellular immunity are sophisticated and difficult to establish in resource-limited settings, thereby constraining large studies. There is a need for study models that bridge the gap between preclinical and clinical vaccine testing, and which are able to predict a virus-specific vaccine effect before actual challenge. Virus replication kinetics (VVR) on ConA-stimulated peripheral blood mononuclear cells (PBMC) was used as an ex vivo model to mimic the interaction between different components of the immune system and viral infection. PBMCs were obtained from the 17 experimental rhesus monkeys before immunisation and subsequently at 12, 26 and 44 weeks during immunisation (wdi). Before immunization, VVR of vaccine-naïve PBMCs varied between individual animals by between >430-fold and >60-fold after 7 and 10 days of infection cultures. VVR of sham-vaccinated control monkeys remained constant over 44 the weeks. However, VVR of immunised animals was significantly attenuated during this follow-up period. This effect was not influenced by the MHC- class 1 Mamu-A*01 allele, which is normally associated with slow disease progression. VVR was instead dependent on the number of IFN?-producing cells (p=0.001), CD8+ T-cell non-cytotoxic antiviral response (CNAR) (p=0.01) and MIP-1a (p=0.013). High VVR correlated with increased CXCL10, IL-1ß and MIP-1ß. Importantly, pre-challenge VVR, CXCL10, IL-1ß and MIP-1ß but not IFN? correlated directly with acute plasma viremia and inversely with memory CD4+ T-cell counts after SIVmac239 challenge. VVR was thus able to predict disease progression and the protective capacity of the vaccine