Effects of quadrivalent human papillomavirus vaccination – Authors' reply

Abstract

Cost-effectiveness analysis, as referenced by Davide Mauri and Nikolaos Polyzos, constitutes one of several sources of information considered by policymakers in developing and developed worlds in making decisions about the optimum efficient use of health-care resources. The WHO Commission on Macroeconomics and Health has suggested that interventions costing less than three times a country's per capita gross domestic product per disability-adjusted life year gained can be regarded as good value, and analysts have equivalently applied this threshold to analyses that use quality-adjusted life years (QALYs).1Resch SC Salomon JA Murray M Weinstein MC Cost-effectiveness of treating multidrug-resistant tuberculosis.PLoS Med. 2006; 3: e241Crossref PubMed Scopus (89) Google Scholar Preliminary results from a cost-effectiveness analysis of vaccination with quadrivalent HPV 6/11/16/18 vaccine in Mexico suggest a cost/QALY ratio well below this threshold in that country.2Insinga RP, Dasbach EJ, Elbasha EH, Puig A, Reynales-Shigematsu LM. Cost-effectiveness of quadrivalent human papillomavirus (HPV) vaccination in Mexico: a transmission dynamic model-based evaluation. 1st ISPOR Latin America Conference, Cartagena, Colombia, 2007.Google Scholar Previous analyses in developed world settings have consistently shown that vaccination of girls and young women has a cost-effectiveness ratio within the range typically regarded as cost-effective.3Dasbach EJ Elbasha EH Insinga RP Mathematical models for predicting the epidemiologic and economic impact of vaccination against human papillomavirus infection and disease.Epidemiol Rev. 2006; 28: 88-100Crossref PubMed Scopus (135) Google Scholar In countries with the fewest resources, direct assistance and public-private partnerships can help deliver needed medicines to the population at or below development costs—eg, the ivermectin donation for river blindness. Marc Arbyn states that if the cases of vaccine-type-related disease are subtracted from disease due to all types, there are a larger number of cases in women who received vaccine than in those who received placebo. This subtraction assumes that the subset of disease cases due to vaccine HPV types and the subset of cases due to non-vaccine HPV types are mutually exclusive, which is not the case. Coinfections with vaccine and non-vaccine types are common. In the presence of coinfection, the effect of such a subtraction is to ignore the presence of non-vaccine HPV types in disease where a vaccine-type HPV has also been detected. The effect of the subtraction is to preferentially attribute co-infected disease cases only to the vaccine HPV types. Individuals in the placebo group are more likely to have their non-vaccine type-related disease discounted in this way. Owing to the high efficacy of the vaccine, individuals in the vaccine group have less vaccine-type-related disease, and so those in the vaccine group have fewer such coinfection cases. To illustrate this point, an analysis of the numbers of individuals with disease due to vaccine and non-vaccine HPV types in the intention-to-treat population of protocols 0134Garland SM Hernandez-Avila M Wheeler CM et al.Quadrivalent vaccine against human papillomavirus to prevent anogenital diseases.N Engl J Med. 2007; 356: 1928-1943Crossref PubMed Scopus (1532) Google Scholar and 0155The FUTURE II Study GroupQuadrivalent vaccine against human papillomavirus to prevent high-grade cervical lesions.N Engl J Med. 2007; 356: 1915-1927Crossref PubMed Scopus (1697) Google Scholar is presented in the figure. The parts shaded blue would be the result of subtraction, similar to Arbyn's subtraction. However, the total numbers of cases of disease related to non-vaccine HPV types are 226+56=282 cases in the vaccine group and 193+106=299 cases in the placebo group. There is not an excess of cases caused by non-vaccine HPV types in the vaccine group. We declare that we have no conflict of interest other than that noted in the original paper. Effects of quadrivalent human papillomavirus vaccinationK A Ault and the Future II Study Group (June 2, p 1861)1 report a 99% efficacy rate for the quadrivalent HPV 6/11/16/18 vaccine against lesions associated with human papillomavirus (HPV) 16/18. However, when the intention-to-treat population was considered, efficacy was only 44% and the reduction in the detection of prevalent and incident high-grade cervical disease, due to either vaccine or non-vaccine types, was only 18% compared with placebo. This is a substantial decrease, but far from target, considering that HPV 16/18 vaccines are generally developed to prevent about 70% of cervical cancers. Full-Text PDF Effects of quadrivalent human papillomavirus vaccinationThe pooled per-protocol results of the phase II and III trials, at 36 months of follow-up,1 confirm that the quadrivalent Gardasil vaccine offers quasi-100% protection against cervical intraepithelial neoplasia of grade II or worse (CIN2+) or adenocarcinoma in situ (AIS) related to the human papillomavirus (HPV) vaccine types, if women are not infected with these types at enrolment. In the intention-to-treat (ITT) population, also involving non-naive women, protection against any disease or HPV-16/18-related disease was 18% and 44%, respectively. Full-Text PDF