
Full text loading...
Category: Bacterial Pathogenesis; Immunology
Establishing Immune Correlates of Protection, Page 1 of 2
< Previous page | Next page > /docserver/preview/fulltext/10.1128/9781555815820/9781555814083_Chap23-1.gif /docserver/preview/fulltext/10.1128/9781555815820/9781555814083_Chap23-2.gifAbstract:
This chapter discusses the possible immune measurements that could be used to develop a correlative model, the functional forms of statistical models that have been used, and the results of correlative models for invasive pneumococcal disease (IPD), acute otitis media (AOM), pneumonia, and colonization. The most common immune measurement that is used in the development of protective correlates is the enzyme-linked immunosorbent assay (ELISA) for immunoglobulin G (IgG) class anticapsular polysaccharide antibodies. Generalization to other polysaccharide-based conjugate vaccines regardless of the carrier protein(s) is most likely acceptable, but the models have uncertain validity for nonconjugate pneumococcal vaccines that are protein or polysaccharide based or for populations that differ in important ways from those studied in the clinical efficacy trials, such as infants infected with human immunodeficiency virus. The probability of acquisition as a function of IgG antibody concentration was modeled using logistic regression. Only vaccine serotypes 9V, 14, 19F, and 23F were modeled because of the limited number of acquisition events for the other serotypes. In a separate analysis, serotype 6A was modeled using the immune responses to 6B. The antibody levels needed to protect against IPD, AOM, and colonization apparently differ. Higher levels are required for protection against AOM and colonization than for protection against IPD.
Full text loading...
Theoretical relationship between risk of disease and concentration of protective antibodies. The step function represents the simplifying assumption required to calculate a protective concentration, Cprot.
RCD curves for IgG antipneumococcal capsular polysaccharide antibody concentrations aggregated for the seven vaccine types in three controlled PCV efficacy studies and the studies with pooled data weighted for the number of study subjects ( 17 ).
Effect of antibody concentration on the yearly incidence of serotype-specific AOM. Fitted values are from the GLM for each serotype. The risk of AOM caused by serotype 6A is associated with the concentration of cross-reactive 6B antibodies. Data are from Jokinen et al. ( 12 ).
Percentages of new acquisition events (in 1-log increments) and percentages of new acquisition events predicted (curve) by the logistic regression model by Streptococcus pneumoniae serotype (9V, 14, 19F, and 23F) among subjects who received PCV9-CRM (black squares) and subjects who received the control vaccine (black circles). Data are from Dagan et al. ( 5 ).
Percentages of new acquisition events (in 1-log increments) and percentages of new acquisition events predicted (curve) by the logistic regression model for Streptococcus pneumoniae serotype 6A (using serotype 6B serologic values) among subjects who received PCV9-CRM (black squares) and subjects who received the control vaccine (black circles). Data are from Dagan et al. ( 5 ).
Three controlled double-blind efficacy trials of PCV used in meta-analysis of protective pneumococcal antibody concentration a
Number of children with and without 6A, 6B, 19F, or 23F AOM events and GMCs of corresponding IgG antibodies among children immunized with PCV a
Reduction in risk of AOM associated with a 10-fold increase in ELISA-measured antibody concentration or OPA a
Results of the logistic regression model of the probability of a new acquisition of pneumococcal serotypes 6A, 9V, 14, 19F, and 23F during follow-up of 129 subjects who received PCV9-CRM a