by ecancer reporter Clare Sansom

The linkage between infection with certain subtypes of the human papilloma virus (HPV) and the development of cervical cancer is now well known.

About seventy percent of all cases of this cancer worldwide are thought to be caused by infection with one of two HPV subtypes, HPV 16 and HPV18.

Two HPV vaccines have been developed for prophylaxis against cervical cancer; they are generally given to pre-adolescent girls at an age when they can be presumed to have no previous exposure to the virus.

One of these vaccines, GlaxoSmithKline’s Cervarix® protects against HPV 16 and HPV 18 only, whereas Merck’s Gardasil® is quadrivalent, also offering protection against two further subtypes, HPV 6 and HPV 11.

These subtypes have both been implicated in the development of genital warts.

So far, clinical trials of HPV viruses have generally used procedures and selected study populations that were designed to optimise vaccine compliance and that may therefore not be representative of the population in general.

Salaheddin Mahmud of the University of Manitoba, Winnipeg, Canada and his colleagues have now conducted an observational study of the effectiveness of the quadrivalent vaccine, Gardasil, using data routinely collected in Manitoba.

This vaccine was introduced into the province in 2006 and very high quality data on vaccinations and cases of cervical cancer for the subsequent seven years can be obtained from registries there.

Mahmud and his colleagues identified a vaccinated cohort of 3,541 females aged 15 or over and living in Manitoba who had received Gardasil privately between August 2006 and March 2010.

Each vaccinated female was linked by age to up to three females in Manitoba who had not received the vaccine during this period; this non-vaccinated cohort comprised 9,594 individuals.

Most of the vaccinated females were aged between 15 and 19 when they received the vaccination.

The study participants were followed up for a mean of 3.1 years following enrolment for the occurrence of pre-cancerous conditions that can be detected by standard cervical screening (the Pap test): abnormal squamous cells (ASCUS), low-grade and high grade squamous intraepithelial lesions (LSIL and HSIL respectively).

In the whole cohort, the 3-year cumulative probability that a participant would develop squamous intraepithelial lesions was slightly higher if she had not been vaccinated (3.7% higher for HSIL and 2.6% higher for LSIL); the probability of developing ASCUS did not depend on vaccination status.

No cases of invasive cervical cancer were detected during the study.

Subgroup analysis indicated significant differences in vaccine effectiveness for women of different ages and with different histories of virus exposure before vaccination.

Vaccine effectiveness was highest in girls aged between 15 and 17 at enrolment, being estimated at 35%, 21% and -1% against HSIL, LSIL and ASCUS respectively.

Estimated effectiveness was higher still when girls who did not undertake Pap screening after vaccination were excluded from the analysis, being 46% against HSIL and 35% against LSIL: only these figures reached statistical significance.

The vaccine appeared to be significantly less effective in women aged at least 18 at enrolment, particularly if they had a history of abnormal cervical cytology prior to vaccination.

There was some evidence that the vaccine was effective in protecting women over 18 without abnormal cytology, with a 23% reduction in HSIL risk estimated for these women.

These results are consistent with those of clinical trials conducted before the vaccine was licensed, and indicate that the vaccine is moderately effective at least in protecting girls vaccinated at a young age against higher grade lesions.

However, its effectiveness in protecting older women cannot be confirmed by this study, particularly in women with a previous diagnosis of abnormal cervical cytology.

Mahmoud and his colleagues conclude by recommending that Gardasil should be administered before girls become sexually active, and that all sexually active women should continue to be screened for cervical dysplasia, even if they have been vaccinated.

Reference

Mahmud, S.M., Kliewer, E.V., Lambert, P., Bozat-Emre, S. and Demers, A.A. (2014). Effectiveness of the Quadrivalent Human Papillomavirus Vaccine Against Cervical Dysplasia in Manitoba, Canada. J. Clin. Oncol., published online ahead of print 6 January 2014. doi: 10.1200/JCO.2013.52.4645