Evaluation of the continuing effectiveness of the United Kingdom's Haemophilus influenzae type b national immunisation programme

Abstract

Haemophilus influenzae (Hi) is a gram negative bacillus that commonly colonises the upper respiratory tract of humans. While the serotype b capsular form (Hib) of this organism accounts for less than 5% of carriage episodes, it is responsible for more than 95% of all invasive Hi infections. The capsular polysaccharide of Hib, polyribosylribitol phosphate (PRP), is an important virulence determinant. Serum antibodies specific for PRP correlate with bactericidal activity against the organism, and their acquisition with age in populations has an inverse relationship with the experience of invasive disease. The first vaccines to be developed against Hib were based on the capsular polysaccharide, but were limited in their application due to the T cell independent nature of the antigen, which was not immunogenic in children less than two years of age. Chemical conjugation of PRP with a carrier protein such as diphtheria or tetanus toxoid converted it into a T cell dependent antigen, capable of eliciting antibody production in infants only a few months old, and resulting in the formation of populations of ‘memory’ B lymphocytes believed to provide long term protection from infection. Efficacy of these conjugate vaccines against both acquisition of the organism and progression to invasive disease was subsequently demonstrated. Hib was a significant cause of morbidity and mortality among British children before inclusion of Hib conjugate vaccines in the routine UK infant immunisation schedule in October 1992. In contrast to other developed countries, it was scheduled as a three dose accelerated primary course at 2, 3 and 4 months of age without a booster dose in the second year of life. Vaccine introduction was accompanied by a catch-up immunisation campaign in which a single dose was given to children under four years of age. Despite an initial rapid and dramatic fall in Hib disease reports, a resurgence of cases occurred from 1999 onwards, predominantly in vaccinated children. This rise has since been attributed to a number of factors related to vaccine use. Wearing off of the initial catch-up campaign’s effect, rapid waning of immunity induced by the accelerated infant schedule and the use of less immunogenic acellular pertussis containing Hib combination vaccines in 2000/2001 have all resulted in a reduction in protective antibody titres in the population at risk. In keeping with the notion that this should correspond with greater likelihood of disease, infants bom prematurely, whose response to immunisation is less than term-delivered counterparts, have been shown to have a four fold increase in risk of invasive Hib infection. Hib vaccines have further been demonstrated to delay acquisition of carriage, thereby reducing transmission in populations. This indirect effect of immunisation, or ‘herd immunity’ was hailed as a major contribution to disease reduction. It may also impair the maintenance of long term immunity at a population level, however. Following an initial decline in Hib reports associated with vaccine introduction among older unimmunised cohorts, adult case numbers in the UK have now returned to prevaccination levels. This rise has occurred in association with a reduction in specific antibody titres to Hib in adult age groups which was observed within only two years of vaccine introduction. In a surprising finding, a national case-control study of invasive Hib infections found a reduced risk of invasive disease among children attending day nurseries, or exposed to cigarette smoke. Prior to vaccine use, both of these exposures were associated with higher rates of disease, presumed to be mediated through higher carriage prevalence. It is possible that repeated antigenic challenge through colonisation may result in improved persistence of antibodies elicited by immunisation. The anticipated medium to long term effects of Hib conjugate vaccine as used in the UK on transmission and maintenance of population immunity were simulated using a mathematical model. The model further provided a framework in which to investigate key assumptions regarding the mechanism of vaccine protection. Good correlation of outputs with seroepidemiologic data from the UK was achieved. Exploratory analysis of the protective efficacy of immunologic memory within this system reinforced individual and population level observations of relatively poor protection against invasive disease in individuals whose post-immunisation Hib antibody titres had waned to undetectable levels. The anticipated effects of a range of potential vaccination strategies which may be employed in the UK to raise antibody titres in the age groups at greatest risk and thereby reduce disease incidence were investigated.