HPV vaccination- what impact on cervical cancer can we expect?

The problem

HPV vaccines have been approved for marketing and use in young girls throughout the world including low income countries to prevent cervical cancer. The vaccine is on patent and is one of the most expensive vaccines in the world. Concerns have been raised in the literature about efficacy and effectiveness especially in countries lacking data on HPV and cancer epidemiology and because of the reliance on surrogate outcome measures. This presentation reviews the evidence for its efficacy.

The approach

A literature search was done on Medline and Embase for RCTs relating to vaccine efficacy against cervical cancer. 29 papers were found linking to 12 RCTs. Studies were analysed using the CASP criteria. A meta-analysis was planned however due to significant heterogeneity between trials the number of comparisons that can be made is limited.

Findings

Five of the trials were based in one country; the other seven were multi-national however none were conducted in Africa. Ten trials focused on women aged 15-26 years but two trials focused on older women. The trial mean follow-up length varied from 13.6 months to 8.9 years. The number of trial participants ranged from 433 to 18,729.Seven trials restricted eligibility on number of previous sexual partners. For the According to Protocol (ATP) groups, there were differing requirements on whether subjects needed to remain HPV negative throughout the vaccination period and whether subjects needed to be seronegative to HPV16/18. Most trials had multiple subgroups and the proportion of trial participants included in the ATP analyses varied from 62.9-86.8%.There were 59 primary and secondary surrogate endpoints overall. In 2003, the WHO decided CIN2+ should be the main trial endpoint, however CIN2+ results are published for only nine of 12 trials and 15 of 29 papers. In 2014 the IARC proposed persistent infection to be the main endpoint, results are given in nine trials. The definition of persistent infection varies between studies from 4-12 months. Different trials have analysed CIN2+ and persistent infection due to vaccine-HPV types, oncogenic types or regardless of HPV type, limiting comparisons.For six-month HPV16/18 persistent infection in the HPV-naïve groups vaccine efficacy varied from 82.9% 95%CI[53.8,95.1] to 100% 95%CI[84.1,100]. For CIN2+ irrespective of HPV type in the HPV-naive groups vaccine efficacy varied from 52% 95%CI[<0,82] to 70.2% 95%CI[54.7,80.9].

Consequences

There is significant heterogeneity between the 12 trials in terms of eligibility criteria, choice of endpoints and sub-group analyses making comparisons between the trials difficult. The trials were often not powered to consider the main relevant endpoints of persistent infection and CIN2+. There remains significant uncertainty whether the observed heterogeneous efficacy findings for surrogate endpoints will translate into reduction in cervical cancer rates, which is the goal of the vaccination programme.

Credits

• Claire Rees
• Allyson Pollock
• Petra Sevcikova