Two or three doses of human papillomavirus vaccine?

BMJ 2015; 350 doi: https://doi.org/10.1136/bmj.g7778 (Published 07 January 2015) Cite this as: BMJ 2015;350:g7778
  1. Julia Brotherton, medical director
  1. 1National HPV Vaccination Program Register, PO Box 310, East Melbourne, VIC 8002, Australia
  1. jbrother{at}vcs.org.au

Switching to two doses looks feasible, but only with careful monitoring

Human papillomavirus (HPV) vaccines have the potential to prevent the considerable morbidity and mortality caused by oncogenic HPV types. In the eight years since the vaccines were first licensed, we have seen remarkable reductions in genital warts, HPV infections, and pre-cancerous cervical lesions in vaccinated populations.1 2 3 4 5 6 7 8 However, achieving high coverage with three doses of vaccine is challenging in many populations, and the cost of the vaccine has kept it out of reach for many countries. In a linked paper (doi:10.1136/bmj.g7584), Jit and colleagues explore, through modelling, the potential cost effectiveness of a two dose HPV vaccination schedule.9

Both the bivalent and quadrivalent HPV vaccines were initially registered for use as three dose courses given over six months, using the model of subunit vaccines for which multiple doses are needed to generate a sufficient immune response. However, HPV vaccines are notably immunogenic, producing very high and durable antibody responses, and the virus-like particle structure of the vaccines, with their repetitive antigen display, may be stimulating immunity that is more akin to the response generated by viral infections or live vaccines.10

Review of evidence about the vaccines, including results from trials of two dose schedules, led the World Health Organization in 2014 to endorse the use of two dose HPV vaccination schedules for girls aged under 15 years. In this age group, two doses spaced at least six months apart produce antibody titres as high as those shown to be effective against HPV infection and disease in older women.11 Clearly, fewer doses is an attractive strategy, offering the potential for cost savings and higher coverage. However, trade-offs need to be considered if the duration of protection offered by two doses is lower than that offered by three or if cross protection against related HPV types is reduced.

In their analyses, Jit and colleagues have adapted established models of HPV transmission, vaccination, and cytology based screening to project the impact of a switch from three to two dose schedules on HPV related and other health costs. Their models use a 100 year time frame and assume HPV vaccination at age 12 with 80% coverage and an initial catch-up programme to age 18. They compare the effect of a three dose schedule—assumed to give lifelong protection and cross protection against related HPV types consistent with existing data—with a two dose schedule under a range of plausible assumptions about duration of protection: 10, 20, or 30 years with cross protection or lifelong protection without cross protection.

By comparing outcomes and costs, the potential cost effectiveness of two dose vaccination compared with no vaccination, and the incremental cost of three dose compared with two dose vaccination, can be assessed objectively. The analysis assumes that each vaccine costs the UK list price, which is higher than most government negotiated prices for use in national programmes. Helpfully, the authors did a threshold analysis to indicate at what price two or three doses become cost effective, using £30 000 per quality adjusted life year as the subjective threshold for cost effectiveness.12

In the baseline UK based model, a two dose schedule using either vaccine will dramatically reduce the burden of HPV related disease compared with no vaccination, even if protection lasts only 10 years, and will be cost effective even at full list price. If protection after two doses lasts 20 years, the additional benefit of a third dose is minimal. A third dose can be cost effective only if two doses give only 10 years’ protection. Under this scenario, the threshold price for the third dose to be cost effective is around £147, well above the list price of around £80. Conversely, if two doses provide more than 10 years’ protection (which is not yet known), a third dose can be cost effective only if the price falls to below £31 (and in many scenarios much lower).

These analyses emphasise the importance of considering the incremental price of the third dose and its attendant opportunity cost. They do not resolve the tension between giving three doses, with relative certainty about impact, and moving to two doses, with the attendant uncertainty about duration of protection and cross protection. Decision makers must consider what would happen if protection wanes after a two dose schedule. Access to comprehensive vaccination records would be critical, along with a reliable way of finding and recalling women for a booster and a strategy to minimise loss of confidence in vaccination programmes.

Moving to two doses would also require clear communication about when a third dose is still needed, an ability to monitor adherence to the minimum dose interval, and a strategy to manage the risk of only ever delivering one dose given the six month or greater separation in time between doses. Given the need for long term records of vaccination to evaluate effectiveness in cohorts offered different vaccines or schedules, is difficult to comprehend developed countries proceeding with a two dose vaccination strategy without an established vaccine registry and an ability to link vaccination records to health data in the future.

With new developments in both cervical screening technology and HPV vaccines coming, we are living in an era of great achievement and promise in cancer prevention but also a movable feast of prevention programmes. Both prospective modelling, such as that of Jit and colleagues,9 and comprehensive infrastructure for monitoring and evaluation are needed.


Cite this as: BMJ 2015;350:g7778


  • Research, doi:10.1136/bmj.g7584
  • Competing interests: I have read and understood the BMJ policy on declaration of interests and declare the following interests: I have been an investigator on investigator designed, unrestricted HPV epidemiology research studies partially funded by bioCSL/Merck but have received no personal financial benefits.

  • Provenance: Commissioned; not externally peer reviewed.


View Abstract

Sign in

Log in through your institution