Shot in the armSome additional remarks. (Click here for part 1. Thanks to mhatrw for the NEJM link.)

This Gardasil is not as effective as they say. Claims that it is 98% effective against two of the 13 HPV strains that can cause cervical cancer may technically be true but they fail, monstrously, to reflect the true picture.

As mentioned in my earlier post, none of the women in the study (control group or vaccinated group) actually contracted cervical cancer. The analysis of Gardasil’s efficacy was based on rates of precancerous lesions, also called neoplasias, which were treated as “cancer surrogates” for the purpose of analysing the study data.

The lesions that can be caused by HPV are categorised into three grades. Grade 1 lesions are generally not considered precancerous and treatment is not usually indicated. Grade 2 lesions are usually but not always treated and 40% of them regress spontaneously. Some experts consider that Grade 2 lesions are sufficiently “precancerous” to be a good indicator of a vaccine’s efficacy against cervical cancer, others do not. Grade 3 lesions are the least likely to regress and the most likely to result in cancer.

The two main studies used to demonstrate the efficacy of Gardasil are called FUTURE 1 and FUTURE 2 (acronym for some puke-making proper title).

FUTURE 1 found that vaccination reduced the total number of lesions from 5.9% in the control group to 4.7% in the vaccinated group, reducing your total risk for such lesions by around 20%. However, analysis showed that most of this reduction related to the more-or-less benign Grade 1 lesions. There was a modest different in the rates of Grade 2 lesions and no detectable difference in Grade 3 lesions.

The larger study FUTURE 2 found that, looking just at Grades 2 and 3, the risk in the vaccinated group was reduced to 3.6% from 4.4% in the control group, a reduction of 17%. There was no data identifying the effect on Grade 3 lesions.

And, in case this didn’t sink in the first couple of times you read it – there is NO EVIDENCE as to the number of cancers that were or might have been prevented by the vaccinations in these trials. We do not know how many of the lesions prevented would or might have gone on to cause cancer. What if the vaccination only targets lesions of a more benign type? What if, as the limited data from FUTURE 1 suggests, the vaccination actually has NO impact on the total number of Grade 3 lesions, those most likely to end up as cancer?

The results of these studies show that at best Gardasil can modestly reduce your chances of getting HPV lesions that may or may not lead to cervical cancer. That’s it.

These figures – a maximum of 20% risk reduction even on the most generous interpretation – show something very different from the 70% risk reduction that we are led to expect by the headlines. If the vaccine is nearly 100% effective against the two virus strains causing 70% of all cancers – why did the rate of precancerous lesions reduce by a far, far lower percentage?

There are a few possible explanations for the very modest benefits of Gardasil despite its apparently very good results in creating immunity to the two strains of HPV that are currently implicated in around 70% of all cervical cancer:

Pre-existing, undetected lesions

It is possible that some women who developed lesions during the fairly short period of the study had undetected lesions before the study started which only showed up during the study. These would be equally prevalent in each group, and accordingly they would make the relative difference between the vaccinated and control groups appear to be lower than is fairly the case. Would it make the efficacy appear to be a mere 20% or, probably, less? Only a longer term study would give any indication of an answer, and no longer term studies have yet been conducted.

Pre-existing HPV infections

As previously mentioned, the vaccination only works if you don’t already have an infection with one of the two strains of HPV that Gardasil is supposed to protect you from.

If you include in your data population people who have a pre-existing infection and who we therefore know will get no benefit from the vaccination, this is bound to skew the figures by making the vaccine look less good than it really is.

BUT (1) it remains to be seen whether there were so many women with a pre-existing infection that they could skew the results this badly and (2) if pre-existing infections are so common, shouldn’t we be recommending that tests be done before giving the patient a $360 shot of something that has limited safety data? To avoid exposing them to unnecessary risk and cost?

Oh, wait. The proposed patient is a girl of 11 or 12, and the test involves vaginal swabs. Stuff it, let’s jab them all – it’s all money in the bank, right?

Effect of vaccination on other strains

Of even greater concern, to me, is that the vaccinated group, while suffering almost no lesions attributable to the two vaccinated strains, suffered nearly as many lesions in total. So it appears that the vaccinated group developed more lesions, compared with the control group, caused by the other 11 strains of HPV: the reduced number of lesions caused by the vaccination strains was counterbalanced by an increased number caused by other strains.

Did the reduction in some strains therefore simply make way for others which became (speculating here) more successful in the absence of “competition” from the vaccination strains? Did the vaccination itself in some way strengthen the other strains? Or what? I guess we don’t know why the non-vaccination strains increased in the vaccinated group, but I think it is safe to say that this is something we should find out before we let the vaccination loose on an unsuspecting public.

So is it worth the risk?

A lot has been made of Gardasil in cost-benefit terms, in addition to the possible cancer-prevention effect. Even ignoring cervical cancer, if the vaccination only reduces the total number of lesions this would save the cost and distress of those lesions. One course of treatment at $360 could prevent who knows how much expenditure in treating lesions. Or so the argument goes.

However, taking the 0.8% absolute difference in risk level shown in the FUTURE 2 study, you would on average have to vaccinate 129 women just to prevent one lesion at Grade 2 or 3.

(We do not know how many women would need to be vaccinated in order to prevent one Grade 3 lesion. Given that lesions do not necessarily lead to cancer, we do not know how many women would need to be vaccinated in order to prevent one case of cervical cancer.)

So if you have to give 129 courses of Gardasil vaccination at $360 a time, you would need to weigh that cost, all $46,440 of it (before you consider distribution and administration costs or the overhead of compensating vaccine damage victims), against the cost of treating a lesion. Does it really cost the health service more than $46,440 to treat the average lesion? I doubt it.

What about the girls? Please, won’t somebody think of the children!

It is quite instructive to review the information on Merck’s own website. There is an awful lot about “why you need Gardasil” – but very little about the evidence for its safety and efficacy. Even the patient information leaflet doesn’t tell you how good the drug is – or isn’t.

What you get is a whole lot of sales talk glossing over any possibility of any concerns over either safety or efficacy. What you get is information in an easy explanatory tone, skilfully pitched to create patient demand without creating any significant awareness of possible disadvantages or reasons for caution.

Take the “Who should receive Gardasil” page, for example:


GARDASIL is for girls and women ages 9 to 26*. GARDASIL works when given before you have any contact with HPV Types 6, 11, 16, and 18.

If you’ve already been infected with HPV**, you may still benefit from GARDASIL because it is unlikely that you have been infected with all 4 types of the virus covered by the vaccine. Your doctor or healthcare professional can help you understand more.


Like other vaccines your daughter has received, GARDASIL works to help prevent illness. GARDASIL works when given before there is any contact with HPV Types 6, 11, 16, and 18. That’s why it’s important that you talk to your daughter’s doctor or healthcare professional about getting GARDASIL now—not later. You’ll be helping to protect her future*** from cervical cancer and genital warts before she’s even old enough to worry about them.


Talk to your daughter’s doctor or healthcare professional about getting her vaccinated with GARDASIL.

* Actually, it is not “for” that age range at all. It is licensed for that age range, but it is targeted at the young girls in whom a pre-existing infection is much less likely, and it is these young girls who are likely to be the subject of any mass/mandated vax programme.
** Oh – really? Pre-existing infection is not a problem? Forget everything the studies have shown. Forget the risks. Forget the adverse reactions. Jab them all regardless!
*** Yeah, you better get your daughter vaccinated, or is that you don’t want to protect your daughter? Isn’t she worth that measly $360?
**** By whom? Being officially licensed is not the same as being officially recommended – so who is doing the recommending? Merck researchers perhaps? It is worth pointing out that the license granted to Merck was on the basis that further trials be carried out and therefore, presumably, on the basis that there was not as at the date of the grant of the FDA licence sufficient evidence as to its safety and/or efficacy. Some recommendation.

It is worth reiterating that:

  • This vaccination has never been tested for efficacy in girls of the target age.
  • This vaccination has had only limited safety trials for girls of the target age.
  • Girls of the target age would not be offered vaginal swabs to identify whether they have any pre-existing infection that would prevent the vaccination from working.
  • Nobody knows how long the limited protection provided by the vaccination would last. Girls are being vaccinated to provide long term protection on the basis of trials lasting about three years, or less.

Sources: – Merck website.

HPV Vaccination — More Answers, More Questions
New England Journal of Medicine – Sawaya and Smith-McCune, 10 May 2007

Gardasil Not the Dream Vaccination Women Expected
Associated Content – Summer Banks, 10 May 2007