HPV vaccination not cost effective

Knowledge of the etiology of cervical cancer has been developed over the last century and during this time many lifestyle and environmental factors have been implicated in the etiology of this disease [1,2]. In 2002 it was stated that Human Papillomavirus (HPV) genotypes 16 and 18 are responsible for causing approximately 70% of cervical cancer worldwide [3]. This claim was made in 1995 based on the use of new hybridization technology for detecting HPV DNA that was stated to be ‘truly sensitive and specific’ [4]. Prior to 1995 the detection of HPV DNA in different tissues was known to be unreliable and the sensitivity of the results varied with the different techniques [5]. A causal theory based only on the presence of HPV genotypes is strongly dependent upon the accuracy and precision of the biotechnology used for detection [6]. Identification of HPV genotypes in the anogenital tract is also complicated by the fact that there are at least 40 HPV types present making it difficult to distinguish the causal factors for cervical cancer [6].

In 1995, Bosch et al. set out to characterise the global distribution of HPV genotypes because they knew this was ‘essential to the development of vaccination strategies to curb the burden of cervical cancer’ [7] p.797. In this study of 1000 cervical cancer tumors it was found that 93% contained HPV DNA [7]. This international study used new polymerase chain reaction-based (PCR) assays to detect more than 25 HPV types in 1000 specimens. In 1999, the 7% of tumors that were originally found to be HPV negative in the Bosch et al. study were re-analyzed using different techniques and assumptions [8]. After re-analysis Walboomers et al. claimed that 99.7% of tumors contained HPV DNA [8]. This evidence and other case–control studies led scientists to claim that persistent infection with HPV 16 and 18 is the main and determining factor in the etiology of most cervical cancer [3,8]. Consequently it was considered that a vaccine might be beneficial in reducing the global burden of cervical cancer [3]. This conclusion was based solely on the accuracy of the detection methods and assumptions that were used to attribute causality to HPV genotypes 16 and 18.

Whilst PCR methods are more sensitive and specific than liquid hybridization techniques and enable the identification of different genotypes, the specificity of this technique depends upon the type of primer used: type-specific or broad-spectrum [6]. The Bosch et al. 1995 study used the broad-spectrum MY11/09 method to genotype HPV-DNA [7]. The nascent technology used in this study was only available from the mid-nineties so the evidence for the causality of different HPV genotypes was based on a small number of studies between 1995 and 2002. Prior to 2002 a multi-factorial etiology was believed likely with HPV being a necessary factor but not a sufficient cause[2].

By 2002 scientists were proposing that human papillomavirus (HPV) Type 16 and 18 was the ‘first ever identified necessary cause of human cancer’ [3]. This suggests that cervical cancer does not and will not develop in the absence of HPV DNA [3]. The claim hasn’t been sustained because some investigators observe that HPV infection cannot be found in every patient with cervical cancer [2]. Scientists have found that persistent infection with one of 15 genotypes of HPV can lead to cervical cancer and it is stated that HPV genotypes 16 and 18 are the cause of the majority of cervical cancer worldwide [3,8]. The International Agency for Research on Cancer (IARC) working group also acknowledged in 2005 that there are cofactors that are associated with HPV infection and cancer development [9]. In other words, an HPV infection does not progress to cancer without the co-factors being present.

Several risk factors have been identified including infection with other sexually transmitted infections (STI’s), high parity, smoking and hormonal contraceptives [9]. The strength of these risks (co-factors) is variable and even though it was known in 2002 that co-factors were required for pathogenesis it was still believed that a vaccine targeting HPV 16 and 18 would prove effective[9].

Clinical trials to test the hypothesis that the quadrivalent HPV vaccine would be effective against cervical cancer, and not just the prevention of HPV 16/18 infection, were started in phase 3 trials in 2003 [10,11]. These trials investigated pre-cancerous lesions in 12,167 women (15–26 years old) and were completed in 2007 [10]. Yet the quadrivalent HPV vaccine was approved for the European and US market in 2006 [12,13]. The HPV vaccine, Gardasil® was developed and marketed to women as an effective prevention for cervical cancer after only 4 years of testing for efficacy against pre-cancerous lesions [11,14].

This paper examines the epidemiology of HPV infection and its progression to cervical cancer in different countries. It analyses the assumptions that have been made to claim that a vaccine against cervical cancer is cost-effective (CE) in countries with already established Pap screening programs. The cost-effectiveness of HPV vaccines has been determined using mathematical models that are limited by the assumptions they are based on [15]. An independent assessment of these assumptions is essential to population health and the effective distribution of health resources to the community. This paper provides an independent assessment of these assumptions and re-evaluates the cost effectiveness of broad vaccination programs that have been implemented in many countries.

The electronic version of this article is the complete one and can be found online here.

© 2013 Wilyman; licensee BioMed Central Ltd.