UK: Cervarix is cost-effective

Yesterday, The Guardian Science Blog has helped me launch the debate on HPV vaccination. There you can read a blog post on my first findinds and thoughts.

In the past weeks I’ve come across very interesting studies about the two vaccines that can prevent cervical cancer.

First of all, in this Cancer Research UK’s page you can read very clear information about the human papillomavirus.

Secondly, it is worth reminding the main differences between the two vaccines.

1 – Cervarix:

2 – Gardasil:


Just a few days ago, the Department of Health has responded to my Freedom of Information (FOI) request. My main questions were related to the scientific evidence that led the UK Government to choose Cervarix over Gardasil. They clearly admitted that the Joint Committee on Vaccination and Immunisation (JCVI) recommended the use of Gardasil as it protects against not only cervical cancer but also genital warts, so the decision was ultimately determined by matters of cost-effectiveness. We all would like to know how much the UK has paid for the vaccine, neverhteless the FOI response stated that:

The information in question has been identified at all stages as being commercially sensitive and we have considered your request very carefully, taking account of all the circumstances.  In short, as explained below, we have reached the conclusion that the information you have requested should be withheld as exempt under FOIA section 43(2) which is available to protect commercial interests.

Now with the help of Dr Peter English, who is editor of Vaccines in Practice magazine and has been a consultant on vaccine boards for various vaccine manufacturers, I found out that the scientific evidence is not yet enough to permit any manufacturer to make claims of duration of efficacy beyond that which has been proved. One should not forget that such HPV vaccination programmes have been implemented a few years ago only, so we need to wait and remain monitoring the effectiveness of the programmes in order to draw conclusions.

One other thing I found out is that scientific evidence released after the UK Government went for Cervarix is that this GSK vaccine may offer cross-protection to other cancer-causing HPV strains that Gardasil doesn’t. Check out this study:

Human Papillomavirus (HPV) Vaccines: Limited Cross-Protection against Additional HPV Types, published in The Journal of Infectious Diseases

Gardasil and Cervarix
As you can see from this table, Cervarix offers cross-protection to HPV 45, which according to this study should be given priority in cancer prevention.

In addition, there are other worth mentioning aspects raised by Dr Peter English. Please note that these are his own opinions and not those of his employer.

Cervarix and immunity

Cervarix, which only contains the two oncogenic types, also contains an “adjuvant” – a chemical which makes the vaccine stimulate a higher level of immunity. GSK, who make Cervarix, claim that this means that the vaccine generates a stronger immune response, with more antibodies produced (a higher concentration of antibodies to the vacine in the blood stream – “higher serum levels”). As serum levels fall gradually over time, they suggest that this means that people given Cervarix will have immunity that will last for longer.

Gardasil, by contrast, while it protects against genital warts, is unadjuvanted. The antibody levels it generates are not as high as following Cervarix. Sanofi Pasteur Merieux, who make (or market) Gardasil in Europe claim that the levels are sufficient to give good immunity that is likely to last as long as it’s needed.

Are three doses necessary?

If antibody levels continue to fall as the graphs – and experience with other vaccines etc. – suggest, both vaccines are likely to provide protection for 10-25 years – but it’s hard to know for how much longer than Gardasil Cervarix will actually protect women. A single booster dose years after the original course could be expected to push antibody levels back up to the levels they were at after initial course of vaccine; and with the quantity of vaccine that is being produced for world-wide use, it’s likely that the cost will be considerably lower in 15 years time.

Have you found this interesting? Have you got any information that may be relevant to the investigation? I am particularly interested in knowing which countries use which vaccine in their (perhaps non existing) vaccination programmes. We know that both vaccines have been approved in most countries, but which one do their governments pay for – and which one does private care tend to go for?

I’ve received many emails and tweets so far. They all have been very useful in my research. Please contribute as well by commenting this blog post!


3 thoughts on “UK: Cervarix is cost-effective

  1. Really enjoying the site, you’re working in an important area that demands scrutiny.

    The HPV vaccine is an interesting topic, not least because in the UK we have such an established screening programme. The complexity (and cost) of integrating a vaccination programme into the existing smear test is a formidable challenge and one that health economists no doubt go grey over.

    For me there are several questions that I am interested in.

    Inevitably once the vaccine programme is in full swing, the girls being vaccinated now are older and coverage of succeeding generations is high then the screening programme can shrink but until then both need to be run together. There will be a burden of high cost while this happens. However will this be offset by the decline in demand for colposcopy for example? It takes two years (if memory serves) to train a microscope technician to diagnose cervical cancer, people who will need to be retrained or made redundant as the vaccination programme reduces the scope of the screening programme. Smear test diagnostics are on their way to being highly automated, with expensive equipment. I guess my question is, how is vaccination going to affect the infrastructure of screening?

    I would also be keen to know what you can find out about the cost-benefits of extending vaccination to boys. The vaccine has been proven safe (according to several papers including in boys. The expense has always been an overriding factor, as far as I am aware, however do we not want to generate herd immunity?

    Finally, cervical cancer deaths are relatively low in this country however it is a disease that is extremely insidious in that when symptoms eventually present it is usually too late. The really high death rates are in the developing world. Do you plan to find out what, if any, programmes exist to try and take vaccination to the developing countries?

  2. This is a complex issue.

    Medicines including vaccines are licensed to be given according to what’s been proven to work. Which is sensible. Except that it limits you to giving the medicines according to the trials that have been done.

    In this case, both manufacturers did trials where they gave three doses of vaccine, according to a particular schedule. It worked, so a three dose schedule was licensed, and is used.

    Is this a problem? Yes, it is, because it means that three doses have to be paid for, when they may not be necessary.

    Of course, not everybody will have three doses. Over time it may be possible to study people who get cervical cancer (or intermediate markers, such as abnormal smears or the presence of HPV in smears) to see how many doses of vaccine they had.

    I believe that some studies have been done, looking at antibody levels; and that one or two doses may stimulate a sufficient immune response (as measured by antibody levels) to prevent infection. If this were the case, then we could, for example, give teenagers two doses of vaccine rather than three. We could then monitor antibody levels in some of them, over time (as well as monitoring incidence of abnormal smears, HPV detection, and cervical and other HPV-related cancers. (The cancers take longer to develop, so are less useful in the short to medium term.) If antibody levels start to fall, or diseases start to occur, we could adjust the programme as appropriate – which could mean returning to a three dose initial schedule; or possibly to sticking to a two-dose initial schedule, followed by a routinely offering a booster at an appropriate interval (say 15-20 years after the initial course – depending, of course, on what the studies show).

    When trying to answer the question “is [a proposed treatment] cost-effective”, you have to add up the costs of the treatment, and the costs that will be saved or prevented if the treatment is given.

    If you’ve decided to vaccinated anyway, and are then deciding whether to use the vaccine that will prevent genital warts, then it’s rather easier.

    For the costs, you only have to consider the difference in cost between the two vaccines: they have pretty much the same doseage schedule, and require similar call-recall schemes, public information campaigns, vaccine storage and distribution arrangements, staffing, etc.

    For the benefits you have to consider any difference between efficacy against cancer-causing types; and the benefits (converted to financial costs) of preventing genital warts. There are wide estimates for the costs to the health service of genital warts. In purely physical, medical terms, they are more of a nuisance, than a serious threat to physical health (except when they cause recurrent warts on a baby’s larynx – recurrent respiratory papillomatosis. But they are distressing; and they generally occur in young adults people, and can have a serious impact on their sex-lives, at a time when many people have enough problems in this area already. So they may cause significant psychological distress. They are also difficult to treat, usually requiring repeated, time-consuming (and thus expensive) uncomfortable treatments.

    In effect, what the government did was to say that the estimated marginal cost of using Gardasil over Cervarix was going to be £x per year. (The actual cost would depend on e.g. uptake, which could only be estimated.) The estimated additional value would be £y per year. If y is greater than x, then you’d choose the vaccine that prevents warts; if x is greater than y, then you’d choose the other vaccine.

    So what is the value of preventing genital warts (y)? Many studies have estimated the costs caused by genital warts, and thus (after taking into account the approximately 90% vaccine efficacy against them, likely vaccine uptake, effects of herd immunity, and so forth) the likely value of vaccinating against them. Run a search at if you’re interested. The figures reached vary widely: and those used in the studies quoted by Department of Health in the information they’ve given on the reasons for their choice of vaccine were very much at the lower end of the spectrum. Some doctors working in sexual health services – not least the well known media doctor, Phil Hammond[1] – think they undervalued the benefits.

    If they’d used a different value for the benefits of preventing genital warts, then y would be different; and if Phil Hammond is right, it would be greater, perhaps to the extent that it would be have been more cost-effective to choose Gardasil than Cervarix.

    Of course, cost-efficacy isn’t necessarily the only factor to bear in mind. Reliability of supplies, likelihood of interfering with other vaccines, ability to “sell” the vaccine to the public (with suggestions that a vaccine that a sexually transmitted infections vaccine might generate antagonism that a cancer vaccine might not.

    The reduction in the number of abnormal smears that will be detected, and the number of colposcopy examinations (using a microscope to look at the patient’s cervix) required was a major consideration in calculating the benefits of vaccination. It will prevent many more abnormal smears and colposcopy exams than it will cases of cervical cancer. It may also change the methods used for screening, with a switch to (or addition of) tests which can detect the presence of cancer-causing virus types, further reducing the need for expensive, invasive, and unpleasant tests.

    It’s actually very hard to run an effective screening programme for cervical cancer, based on cervical screening. And expensive. It requires good access to facilities for doing the test, good lab systems, and good programmes to identify and call for screening the women who need it. The inverse care law applies: poorer people are generally more likely to develop cervical cancer, and less likely to be screened. Many developing countries do not have universal screening programmes – although people who can afford it may choose to be screened.

    It may well be more cost-effective for a developing country, wanting to reduce cervical cancer, and without either a vaccination or a screening programme, to introduce a vaccination programme. (We can afford to do both; but not all countries can.) I believe that arrangements may have been (or are being) made to provide cheaper vaccine for developing countries.

    Boys don’t get cervical cancer, of course. But they do get penile, head-and-neck, and anal cancer (and evidence suggests that anal cancer rates in men who have sex with men are greater than cervical cancer rates). HPV vaccine would prevent all of these; as well as genital warts if the tetravalent vaccine were used.

    But we return to costs, again. Would the amount of disease prevented justify the considerable additional cost? If not, we could get better health benefit by spending the money on something else.

    What about herd immunity? (There’s a good video explaining this at the NHS choices website: ). This is a slightly more complex concept when you’re considering a largely-sexually-transmitted infection than when spread is via respiratory droplets. Modelling studies done using a so-called “dynamic model” – which takes into account herd immunity – suggest that vaccinating boys would achieve herd immunity considerably faster than just vaccinating girls; but that the benefits still aren’t enough to justify the considerably greater cost. That said, it’s not clear how accurately the studies have evaluated all the costs – several only looked at the benefits from reducing cervical cancer and abnormal smear rates. It is possible that this might change as better information on the potential benefits comes in, and (especially) if vaccine costs fall significantly (or a single- or two-dose schedule were to be adopted).

    1. Hammond P. (Not) warts and all. BMJ 2008;337(oct23_1):a2186- PMID: PMID: 18948345. (

    There have been few head-to-head studies looking at cross-protection. I believe that the virus-like-particles are very similar for the two vaccines; so the difference in the paper cited probably comes from the adjuvant used in Cervarix.

    This is, of course, just looking at intermediate measures (presence of virus in cervical swabs); not cancer cases, or even cervical abnormalities.

    For cancer to develop, there has to be sustained infection. In many instances a person will be infected by and HPV, and their immune system will clear the infection. Such individuals will not be susceptible to cancer. The studies indicate that Cervarix might be more effective at preventing cancers from non-vaccine HPV types; but it will be a long time before we can be sure if this is really true, or of the true size of the effect.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s