Abstract
Polysaccharide responsiveness is tested by measuring antibody responses to polysaccharide vaccines to diagnose for humoral immunodeficiency. A common assumption is that this responsiveness is biased by any previous exposure to the polysaccharides in the form of protein-coupled polysaccharide vaccines, such as those used in many childhood vaccination programmes. To examine this assumption, we investigated the effect of protein-coupled polysaccharide vaccination on subsequent polysaccharide responsiveness. HIV-infected adults (n = 47) were vaccinated twice with protein-coupled polysaccharides and six months later with pure polysaccharides. We measured immunoglobulin G responses against three polysaccharides present in only the polysaccharide vaccine (non-memory polysaccharides) and seven recurring polysaccharides (memory polysaccharides). Responsiveness was evaluated according to the consensus guidelines published by the American immunology societies. Impaired responsiveness to non-memory polysaccharides was more frequent than to memory polysaccharides (51% versus 28%, P = 0.015), but the individual polysaccharides did not differ in triggering sufficient responses (74% versus 77%, P = 0.53). Closer analysis revealed important shortcomings of the current evaluation guidelines. The interpreted responses number and their specificities influenced the likelihood of impaired responsiveness in a complex manor. This influence was propelled by the dichotomous approaches inherent to the American guidelines. We therefore define a novel more robust polysaccharide responsiveness measure, the Z-score, which condenses multiple, uniformly weighted responses into one continuous variable. Using the Z-score, responsiveness to non-memory polysaccharides and memory-polysaccharides were found to correlate (R(2) = 0.59, P<0.0001). We found that polysaccharide responsiveness was not biased by prior protein-coupled polysaccharide vaccination in HIV-infected adults. Studies in additional populations are warranted.