Rapid Responses to:

RESEARCH:
Mark Jit, Yoon Hong Choi, and W John Edmunds
Economic evaluation of human papillomavirus vaccination in the United Kingdom
BMJ 2008; 337: a769 [Abstract] [Full text]
*Rapid Responses: Submit a response to this article

Rapid Responses published:

[Read Rapid Response] The paradox effects of human papillomavirus vaccination on cervical cancer incidence and mortality
Christian A Gericke   (18 July 2008)
[Read Rapid Response] Excellent paper; but is it the last word?
Peter M English, Keith Neal   (23 July 2008)
[Read Rapid Response] Over-optimistic benefits of HPV vaccination: HPV vaccination is not that good
Germaine Wong, Kirsten Howard, and Jonathan Craig   (7 August 2008)
[Read Rapid Response] Informed vaccine choice
Mark A Henderson, Christina E Ryan   (11 August 2008)
[Read Rapid Response] Re: Informed vaccine choice
Mark Struthers   (12 August 2008)
[Read Rapid Response] Why has the DoH chosen Cervarix?
Christopher E Clark   (18 August 2008)
[Read Rapid Response] Re: Re: Informed vaccine choice
Peter J Flegg   (1 September 2008)
[Read Rapid Response] Authors' response
Mark Jit, Yoon H. Choi, W. John Edmunds   (18 September 2008)
[Read Rapid Response] HPV 6-11 and undetermined/low grade cervical lesions:something to be ignored?
Silvano Costa, Francesco Saverio Mennini, Giampiero Favato, and Mauro Picardo   (9 October 2008)

The paradox effects of human papillomavirus vaccination on cervical cancer incidence and mortality 18 July 2008
Next Rapid Response Top
Christian A Gericke,
Professor of Public Health Policy
The University of Adelaide, Adelaide, SA 5005, Australia

Send response to journal:
Re: The paradox effects of human papillomavirus vaccination on cervical cancer incidence and mortality

The development of the human papillomavirus (HPV) vaccines and their potential for reducing cervical cancer incidence and mortality in the future is generally considered an important step to improve population health worldwide. For countries without well performing cervical cancer screening programmes and lack of treatment options the current epidemiological and economic models seem applicable.

However, the assumptions underlying published economic evaluations in countries with high coverage, well functioning screening programmes such as the study published in the BMJ on July 17 by Jit et al1 from the UK neglect a major limitation to the validity of their models: there is a non -negligible risk that a high HPV vaccine coverage in adolescents will lead to a decrease in screening uptake in later years because vaccinated women will see themselves no longer at risk and will consequently avoid the screening procedure, which is not that pleasant after all. As about 30% of cervical cancers are not caused by the current HPV vaccines against HPV types 16 and 18, there is a real risk of increased cervical cancer incidence and mortality rates in women who do not take up screening, counterbalancing the positive effects of the HPV vaccination. This is fuelled by health promotion material from manufacturers, Cancer Councils and Cancer Research UK who market the HPV vaccines as ‘cervical cancer vaccines’. This is reflected in the general media and even in some of the leading scientific journals.2-6

Research investigating the potential impact of the HPV vaccines on screening uptake is urgently needed to establish the magnitude of this paradox effect on cervical cancer incidence and mortality and improve our epidemiological and economic models to guide future policies for cervical cancer prevention.

1. Jit M, Choi YH, Edmunds WJ. Economic evaluation of human papillomavirus vaccination in the United Kingdom. BMJ 2008;337:a769.

2. Kmietowicz Z. Opportunity was missed in choice of cervical cancer vaccine, health campaigners say. BMJ 2008;336(7659):1456-7.

3. Keim B. Controversy over cervical cancer vaccine spurs safety surveillance. Nat Med 2007;13(4):392-3.

4. Cohen J. Public health. High hopes and dilemmas for a cervical cancer vaccine. Science 2005;308(5722):618-21.

5. Rai MA, Ali SH. Cervical cancer vaccine: the Indian sub- continental context. Vaccine 2006;24(49-50):7024.

6. Zimet GD, Shew ML, Kahn JA. Appropriate use of cervical cancer vaccine. Annu Rev Med 2008;59:223-36.

Competing interests: None declared
Excellent paper; but is it the last word? 23 July 2008
Previous Rapid Response Next Rapid Response Top
Peter M English,
Consultant in Public Health Medicine
Surrey KT19 9XF,
Keith Neal

Send response to journal:
Re: Excellent paper; but is it the last word?

It is good to see, in the article by Jit et al,1 more of the details underlying the Department of Health’s (DH’s) choice of vaccine for the human papillomavirus (HPV) vaccination programme. (I had responded to an earlier BMJ article, commenting on the fact that professionals have been unable to scrutinise the DH decision.2 Since then we have also had an opportunity to learn more about the criteria for the decision.3)

Nevertheless, there is much that I assume will remain “commercial in confidence”, and will therefore not be put into the public domain, which detracts from our ability to scrutinise this decision; not least, the deals offered by the two companies involved: the process remains a “black box”. There have been significant concerns about how the decision was reached; these concerns might have been mitigated if the process had been more transparent.

The paper by Jit et al is very welcome. It is particularly good to see a dynamic model, taking “herd immunity” into account.

The paper makes some assumptions that a worthy of further discussion.

Duration of immunity

The paper appears to assume that the duration of efficacy is the same for each vaccine. While there are insufficient data to be confident about the duration of immunity for either vaccine, GlaxoSmithKline claims that Cervarix™ uses a superior adjuvant, which gives higher antibody levels and which will provide longer-lasting immunity. There is currently no known immunological correlate of protection, and both vaccines generate immune responses that considerably exceed those resulting from natural infection, so it may be that both vaccines will produce similar, long-lasting immunity. Nevertheless, if GlaxoSmithKline’s claims are true, it would tend to tip the balance more in favour of choosing Cervarix™.

Number of doses

The paper assumes that “those receiving fewer than three doses received no protection”. This seems very conservative. We gather that data from intention to treat trials show that, for at least one of the vaccines, there may be relatively little additional benefit in the third dose: the antibody response to two doses is sufficient, in most women, to provide an excellent level of protection. There may well be studies ongoing, perhaps in the developing world, to assess the efficacy of a two-dose regime. These, and good quality surveillance of the programme, allowing us to differentiate outcomes between women who received one, two, or three doses of vaccine, may allow us to assess this further, in due course: indeed, we may even be able to switch to a two-dose programme if data support this. A less conservative approach would tend to tip the balance further in favour of introducing HPV vaccination.

Changes to cervical screening programme

While the paper models the vaccination programme’s effect on the screening programme caused by a decreased incidence of positive screening results, it is not clear that it models the savings that might result from other changes to the programme. Papers have suggested that a better screening strategy, after the introduction of the vaccine, might be a combination of HPV testing and cytology, with a reduction in the number of cytological samples required, and/-or an increase in the screening interval. This would be likely to reduce the cost of the screening programme, and thus to tip the balance further in favour of introducing HPV vaccination.4

Genital warts

The paper estimated the cost of treating an episode of genital warts (GWs) at £134 (2006 prices). Other papers have estimated the cost of a single successful episode of treatment of a case of GWs to be considerably higher - £216 in one paper. 8 9 Given, also, the high and increasing incidence of GWs world-wide,8 10 and in the UK,11 there is reason to wonder whether the estimated cost savings from GWs if the quadrivalent vaccine (Gardasil™) had been selected should not have been greater than stated.

Table 5 shows the effect of using different discounting rates; but it’s not clear that it distinguishes between the late benefits wrt anogenital cancer, and the early benefits from GW prevention. There are good arguments that, while future costs should be discounted, future benefits should be discounted at a lower rate, if at all.12-14 Since the lead time for preventing GWs will be much shorter than the lead time for preventing anogenital cancers, discounting benefits would tend to tip the balance more in favour of a vaccine which prevents GWs.

Other benefits, not considered

Other cancers (particularly anogenital and head and neck cancers) are also caused by the viruses prevented with these vaccines, including cancers which affect men. Indeed there is some (albeit limited) data suggesting that the rate of anal cancer in men who have sex with men might exceed the rate of cervical cancer in women; and that a higher proportion of anal cancers than cervical cancers can be prevented by these vaccines.5-7

Consideration of the possible benefits of vaccination on these cancers would have tipped the balance further in favour of introducing the vaccine. Indeed, it might also have tipped the balance closer towards being in favour of vaccinating boys as well as girls, thus mitigating the equity issues that otherwise arise; and avoiding giving the impression that sexual health is only an issue for girls. Furthermore, if vaccine uptake is low, then herd immunity will not be achieved by vaccinating girls alone; another argument which could tip the balance further towards vaccinating boys as well as girls.

Summary

This is a very interesting and useful paper, which provides us with more of the evidence on which the introduction of the planned HPV vaccination programme was based.

Careful scrutiny of the paper raises questions, however, about the weight to be placed on its conclusions.

References

1. Jit M, Choi YH, Edmunds WJ. Economic evaluation of human papillomavirus vaccination in the United Kingdom. BMJ 2008;337(jul17_2):a769- (http://www.bmj.com/cgi/content/abstract/337/jul17_2/a769).

2. English PM. How can we know if we can't scrutinise the details? 2008;Accessed: 2008(June 24 2008):Rapid response to article "Kmietowicz Z. Opportunity was missed in choice of cervical cancer vaccine, health campaigners say. Br Med J "Online First" 2008:bmj.a451." (http://bmj.com/cgi/eletters/bmj.a451v1#197698).

3. Primarolo D. Parliamentary Questions on HPV Vaccine (02/07/08). House of Commons Hansard2008(http://www.parliament.the-stationery-office.co.uk/pa/cm200708/cmhansrd/cm080703/text/80703w0021.htm).

4. Franco EL, Cuzick J, Hildesheim A, de Sanjose S. Chapter 20: Issues in planning cervical cancer screening in the era of HPV vaccination. Vaccine 2006;24 (Supplement 3):S171-S177 ( http://www.sciencedirect.com/science/article/B6TD4-4K4PVW5-5/2/716cde2f963dceda7ab1b0fc4c815e26).

5. Parkin DM, Bray F. Chapter 2: The burden of HPV-related cancers. Vaccine 2006;24 (Supplement 3): S11-S25 ( http://www.sciencedirect.com/science/article/B6TD4-4KSV4T8-6/2/01502eeb2bac9ad580cb5061926ecbf2).

6. Daling JR, Madeleine MM, Johnson LG, Schwartz SM, Shera KA, Wurscher MA, et al. Human papillomavirus, smoking, and sexual practices in the etiology of anal cancer. Cancer 2004;101(2):270-80.

7. Chin-Hong PV, Vittinghoff E, Cranston RD, Buchbinder S, Cohen D, Colfax G, et al. Age-Specific prevalence of anal human papillomavirus infection in HIV-negative sexually active men who have sex with men: the EXPLORE study. J Infect Dis 2004;190(12):2070-6.

8. Lacey CJN, Lowndes CM, Shah KV. Chapter 4: Burden and management of non-cancerous HPV-related conditions: HPV-6/11 disease. Vaccine 2006;24 (Supplement 3): S35-S41 ( http://www.sciencedirect.com/science/article/B6TD4-4KSV4T8-7/2/47baa7e29ca7bbbbfdf88ce4b07c556b).

9. Langley PC, White DJ, Drake SM. The costs of treating external genital warts in England and Wales: a treatment pattern analysis. Int J STD AIDS 2004;15(8):501-8.

10. Kjaer SK, Tran TN, Sparen P, Tryggvadottir L, Munk C, Dasbach E, et al. The burden of genital warts: a study of nearly 70,000 women from the general female population in the 4 Nordic countries. J Infect Dis 2007;196(10):1447-54.

11. Health Protection Agency Centre for Infections. Sexually transmitted Infections and Young People in the United Kingdom: 2008 Report. London: Health Protection Agency, 2008 ( http://www.hpa.org.uk/webw/HPAweb&HPAwebStandard/HPAweb_C/1216022460726?p=1158945066450 ).

12. Crott R. Economic analysis of HPV-vaccines: Not so simple? Vaccine 2007;25(45):7717-7717 ( “ http://www.sciencedirect.com/science/article/B6TD4-4PPNK23-3/1/46064d5714f8dab6b9cbcc57d755a64e).

13. Torgerson DJ, Raftery J. Economics notes: Discounting. BMJ 1999;319(7214):914-915 ( http://www.bmj.com/cgi/content/full/319/7214/914).

14. Brouwer WBF, Niessen LW, Postma MJ, Rutten FFH. Need for differential discounting of costs and health effects in cost effectiveness analyses. BMJ 2005;331(7514):446-448 ( http://bmj.bmjjournals.com/cgi/content/full/331/7514/446).

Competing interests: Between them the authors have given occasional lectures for, received expenses for professional conferences from, and participated in advisory boards for various pharmaceutical companies, including GlaxoSmithKline, Sanofi Pasteur MSD, and others. The views expressed are our personal views, and not necessarily those of our employers, or of any company from whom we have received sponsorship.
Over-optimistic benefits of HPV vaccination: HPV vaccination is not that good 7 August 2008
Previous Rapid Response Next Rapid Response Top
Germaine Wong,
PhD candidate
2145,
Kirsten Howard, and Jonathan Craig

Send response to journal:
Re: Over-optimistic benefits of HPV vaccination: HPV vaccination is not that good

The complex dynamic model, rather than the usual static approach, developed by Jit el al is comprehensive and novel.1 Few other studies have assessed the benefits and effects of “herd immunity”,2;3 and none have extensively evaluated the effect of cross-protection on other non-vaccine HPV types and its impact on non-cervical cancers. The methodology used is also robust. Instead of varying single influential variables at a time using univariate sensitivity analyses, the authors have considered lists of scenarios and uncertainties around the key parameter estimates of the model using probabilistic multivariate sensitivity analyses. However, we have a number of concerns about the validity of the clinical inputs, all of which tend to over-estimate the true effects of the vaccine. First, the effect of this vaccine on patient-relevant outcomes such as the incidence of cervical intraepithelial neoplasia and cancer related to any HPV serotypes were not assessed in this model. The assumption of 100% protection against vaccine type HPV infection and cervical dysplasia in HPV naïve women may be valid but prevention on any HPV serotype-related cervical dysplasia is a better indicator of effectiveness and a more realistic approach to inform women and policy-makers about the true benefits of the vaccine. Detailed analyses of the two large, recent well- powered randomised controlled trials by the U.S. Food and Drug Administration (FDA) showed only a modest reduction by 36.8% for all CIN2+ cervical dysplasia in HPV naïve women. These data show that waning benefit of HPV vaccination along the causal pathway to cervical cancer. From 20% for all-HPV infection to 18% for CIN1, 45.7% for CIN2 and 36.9% for CIN3. In absolute terms these differences were even smaller.

Second, the potential effect of strain replacement was not considered by the authors. It is clear from historical evidence that 70% of all cervical dysplasia are related to HPV 16 and 18 infections,4;5 but there are at least 15 other oncogenic viruses identified,6 and the effects of the vaccine on these viruses are largely undefined. It is likely that eradication of HPV-6 and 11, which have been shown to exert protective effect against other oncogenic viruses such as HPV-16,7;8 could potentially increase the risk of other oncogenic viruses and reduce the overall effectiveness of the vaccine.

Third, although the authors have extensively considered the effect of reduced protection by the vaccine over time in the sensitivity analyses, there are no trial-based information about the duration of protection (long follow up being 52 months) or the long-term adverse effects of the vaccines. The model by Jit el al1 shows vaccination for 12 years old girls is favourable if life-long immunity is maintained. However, if immunity is short-lived, vaccination may not be good value for money. It is uncertain whether there is any an-amnestic response from a challenge dose if vaccine efficacy wanes over time; if a booster dose is required the effect on the overall cost-effectiveness is unclear. Third, although the authors have extensively evaluated the effects of “herd immunity” in the model, there is currently no trial-based data informing what the magnitude of such an effect is, if any.

We believe the net effect of the vaccine observed in this analysis is an over-estimation of the true efficacy for cervical dysplasia and cancer and may mislead women, clinicians and policy makers. We must be cautious about the hidden uncertainties such as the duration of protection, efficacy for all viral types and the unwanted adverse effects that might evolve over time. Future research is needed to assess its longer-term benefits/potential harms to inform both the public and decision-makers about policies for cervical cancer prevention.

Reference List

1. Jit M, Choi YH, Edmunds WJ. Economic evaluation of human papillomavirus vaccination in the United Kingdom. BMJ.337:a769, 2008.

2. Dasbach E, Insinga R, Elbasha E. The epidemiological and economic impact of a quadrivalent human papillomavirus vaccine (6/11/16/18) in the UK. British Journal of Obstetrics and Gynaecology 8 A.D.;115:947-56.

3. Taira AV, Neukermans CP, Sanders GD. Evaluating human papillomavirus vaccination programs. Emerging Infectious Diseases.10(11):1915-23, 2004.

4. Ho GY, Bierman R, Beardsley L, Chang CJ, Burk RD. Natural history of cervicovaginal papillomavirus infection in young women. New England Journal of Medicine.338(7):423-8, 1998.

5. Koutsky LA, Holmes KK, Critchlow CW, Stevens CE, Paavonen J, Beckmann AM et al. A cohort study of the risk of cervical intraepithelial neoplasia grade 2 or 3 in relation to papillomavirus infection. New England Journal of Medicine.327(18):1272-8, 1992.

6. Munoz N, Bosch FX, De Sanjose S, Herrero R, Castellsague X, Shah KV et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer.[see comment]. New England Journal of Medicine.348(6):518-27, 2003.

7. Luostarinen T, af G, V, Bjorge T, Eklund C, Hakama M, Hakulinen T et al. No excess risk of cervical carcinoma among women seropositive for both HPV16 and HPV6/11. International Journal of Cancer.80(6):818-22, 1999.

8. Silins I, Wang Z, Avall-Lundqvist E, Frankendal B, Vikmanis U, Sapp M et al. Serological evidence for protection by human papillomavirus (HPV) type 6 infection against HPV type 16 cervical carcinogenesis. Journal of General Virology.80 ( Pt 11):2931-6, 1999.

Competing interests: None declared
Informed vaccine choice 11 August 2008
Previous Rapid Response Next Rapid Response Top
Mark A Henderson,
Foundation House Officer
Ninewells Hospital, Dundee, DD1 9SY,
Christina E Ryan

Send response to journal:
Re: Informed vaccine choice

We read the article by Jit et al (1) with interest although we are saddened that some populations will not have the benefit of an informed choice between a bivalent or quadrivalent vaccine. For example Rick Perry, the Governor of Texas, has mandated that from September 2008 all Texan 6th Grade girls will be offered Gardasil® rather than any other form of HPV vaccination. (2, 3) Although the decision to implement a vaccination strategy is commendable, ensuring that this protection is provided by one particular brand-name vaccine may be a mistake. If other manufacturers develop more other vaccines which offer more extensive protection against other oncogenic strains, the population of Texas will be disadvantaged. Interestingly, it has been suggested that government lobbying may have had a role in this decision. (3)

1. Jit M, Choi YH, Edmunds WJ. Economic evaluation of human papillomavirus vaccination in the United Kingdom. BMJ. 2008;337:a769 doi:10.1136/bmj.a769

2. Hopkins Tanne, J. Texas governor is criticised for decision to vaccinate all girls against HPV. BMJ. 2007; 334:332-333

3. Cabluck H. Texas governor orders anti-cancer vaccine for schoolgirls. USA Today [Internet] 2007 Feb 2 [Cited 2008 Aug 10]. Available from: http://www.usatoday.com/news/health/2007-02-02-texas_x.htm

Competing interests: None declared
Re: Informed vaccine choice 12 August 2008
Previous Rapid Response Next Rapid Response Top
Mark Struthers,
General Practitioner
Bedfordshire, UK mark.struthers@which.net

Send response to journal:
Re: Re: Informed vaccine choice

Dr Henderson appears more concerned that the Texas schoolgirl can only choose the single quadrivalent vaccine available in the world rather than be offered the choice of having an HPV vaccine at all.

This is curious.

Judicial Watch, a US public interest group has been looking carefully at the safety profile of Gardasil and the lobbying tactics of Merck in seeking state mandates for their vaccine.

http://www.judicialwatch.org/story/2008/may/judicial-watch-investigates-side-effects-hpv-vaccine

Tom Fitton, Judicial Watch president, said in May this year,

"The FDA adverse event reports on the HPV vaccine read like a catalog of horrors. Any state or local government now beset by Merck’s lobbying campaigns to mandate this HPV vaccine for young girls ought to take a look at these adverse health reports."

Competing interests: None declared
Why has the DoH chosen Cervarix? 18 August 2008
Previous Rapid Response Next Rapid Response Top
Christopher E Clark,
GP
Mid Devon Medical Practice, School Surgery, Fore St, Witheridge, Devon, EX168AH

Send response to journal:
Re: Why has the DoH chosen Cervarix?

Sir

This interesting article concludes that use of the quarivalent vaccine (Gardasil) is cost effective for HPV vaccination in the UK and that the bivalent vaccine (Cervarix) may be as cost effective at £13-£21 a dose less.

Since both vaccines remain priced the same at £80.50 per dose (eBNF accessed 18/08/08) it remains unclear to me why Cervarix, as reported in the same issue, has become the vaccine of choice on either clinical or cost-effectiveness grounds?

Since the article is introduced as the cost effectiveness model on which the UK DoH decision was based. One can only conclude that this evidence was either ignored or was outweighed by other contrary facts. It would give me some faith in the choice if the full reasoning behind the choice were made public.

Yours faithfully

Competing interests: None declared
Re: Re: Informed vaccine choice 1 September 2008
Previous Rapid Response Next Rapid Response Top
Peter J Flegg,
Consultant Physician
Blackpool FY3 8NR

Send response to journal:
Re: Re: Re: Informed vaccine choice

Tom Fitton from Judicial Watch may well be of the opinion that the US FDA adverse event reports on HPV vaccine read like a "catalogue of horrors". This is however only his opinion, and it is a medically-uninformed one and is not one borne out by the facts.

Adverse events are collated by the FDA through their system of Vaccine Adverse Event Reporting System (VAERS) by which any event, however unlikely it is to be causal, may be reported by any member of the public. A study of the VAERS reports on the 20 deaths reported in association with Gardasil HPV vaccine is quite enlightening. Many of these reports are entirely anecdotal, a form of: "A friend of a friend says a girl died after getting vaccinated", and they consist of no detail whatsoever on which to draw any conclusions. For the cases where there is medically-verifiable information, many of the cases are clearly unrelated to the fact that vaccine had been given at some time in the recent past (e.g. deaths due to Influenza B with Staphylococcal sepsis, Meningococcal septicaemia and meningitis, Brugada syndrome with arrhythmia and seizures, severe diabetic ketoacidosis, arrhythmia with underlying QT syndrome). There were also a number of deaths due to pulmonary embolism in those taking oral contraceptives. There were however a small number where no obvious cause of death could be determined, although information such as toxicology is unavailable in some of these cases. It is not unexpected that young women occasionally die, but it is rather speculative and it may be erroneous to assume, as Tom Fitton presumably does, that Gardasil was the causative factor in all these deaths.

What is needed is hard evidence that any of these deaths are occuring at a higher than background rate. Initiatives such as the CDC's Vaccine Safety Datalink are a better way to ascertain accurate epidemiological information regarding vaccines that is the VAERS system. The CDC and FDA have recently issued a statement about the safety of Gardasil which discusses the safety issues.

Competing interests: None declared
Authors' response 18 September 2008
Previous Rapid Response Next Rapid Response Top
Mark Jit,
Health economist and mathematical modeller
HPA Centre for Infections, 61 Colindale Avenue, London NW9 5HT,
Yoon H. Choi, W. John Edmunds

Send response to journal:
Re: Authors' response

We thank the authors of the responses to our article for their comments, and would like to make the following points in response:

1. HPV vaccination may lead to a decrease in coverage (Gericke). We agree that this is a potential issue, and indeed have highlighted the importance of future changes to screening in the Discussion. However, at present no behavioural data have been collected on this so it is difficult to know on what to base any predictions. There is certainly a need to find out more about attitudes of vaccinated females towards screening and to incorporate this in future models that can inform policy before the oldest vaccinated cohort reaches the age of screening in 2015.

2. Cervarix™ may have a longer duration of immunity than Gardasil™, individuals receiving fewer than three doses may be partially protected, post-vaccination changes to the screening programme may enable further cost savings, some non-genital cancers may be prevented (English and Neal). We agree that these are potential benefits of vaccination (although the first two remain speculative until further data are available). If these assumptions are correct, then they imply that our model was conservative, particularly in its treatment of the cost-effectiveness of Cervarix™, so they would not have affected the ultimate decision to proceed with vaccination using Cervarix™. We believe that it is highly unlikely that vaccinating boys will be cost-effective even after considering non-genital cancers, because most of the burden of disease (in both girls and boys) can be prevented by vaccinating girls alone.

3. Some studies have suggested that the cost of treating warts may be higher than the cost used in our analysis (English and Neal). In the Langley et al paper mentioned (1), the cost to the health service per episode of care was £136 for males and £146 for females which is very similar to the figures we used of £134 (even after taking into account price inflation between 2000/1 and 2006/7). The figure in the abstract of £222 for males and £211 in the Langley et al paper refers to the cost per successful outcome.

4. Discounting benefits at a lower rate would favour Gardasil™ (English and Neal). We agree in principle with the statement, and indeed have presented incremental cost-effectiveness ratios for a range of discounting scenarios. However, we have applied in our base case the discount rate used by the National Institute for Health and Clinical Excellence (NICE) for its recommendations. Using a different rate in our analysis would mean that HPV vaccination is being considered under different criteria from most health care spending decisions in the UK.

5. The assumption of 100% protection against vaccine type HPV infection and cervical dysplasia is optimistic (Wong et al). To clarify, we assumed that vaccination provides 100% protection in HPV-naïve individuals against vaccine-type infection only (Types 16 and 18 for Cervarix™, and types 16, 18, 6 and 11 for Gardasil™). In a separate scenario, we assumed an additional 27% protection against non-vaccine types, but this was not done in the base case. Hence the 100% efficacy assumption we used should not be compared against figures for efficacy against endpoints for all HPV types.

We are aware that the vaccine shows less than 100% efficacy against short-term infection and low grade neoplasias, and so acknowledge that our assumption was a simplification. However, clinical trials indicate that when HPV-naïve individuals are vaccinated, these neoplasias do not progress to more severe neoplasias or to cancer. Prevention of grade 1 cervical intraepithelial neoplasias (CIN1) is estimated to account for under 5% of the quality adjusted life years (QALYs) saved due to vaccination (assuming the base case scenario, 20 years average vaccine protection and taking median values), and most of them would still be prevented even under more pessimistic assumptions of vaccine protection against low grade neoplasias. Hence, our assumption is unlikely to have had much influence on the economic results.

6. HPV vaccination may be less effective than predicted due to type replacement (Wong et al). There is no evidence for type replacement at present, and the biological mechanism by which such an effect could occur is uncertain since multiple HPV types are able to co-exist in the cervix. Hence it is not known whether removing some types would enhance selective pressures on other types. However, we do agree that there is a need to closely monitor the epidemiology of HPV post-vaccination for such effects. The Health Protection Agency has plans to put an appropriate surveillance strategy in place for this (2).

6. There are no epidemiological data on the magnitude of the herd immunity benefit from vaccination (Wong et al). We agree that there are no data at present. However, since vaccinated individuals are protected against infecting others with HPV, and not just against disease endpoints, such benefit would be expected, as has been the case for all other vaccines with this characteristic. Indeed, one of the main purposes of conducting mathematical modelling prior to introducing large scale vaccination is in order to estimate the effect of herd immunity. Unlike a static model, a dynamical model such as the one we used does not require any observational data on the magnitude of herd immunity as it simply assumes that vaccinated individuals are unable to be either infected or to transmit infection to others for the strains being considered.

7. The duration of protection from HPV vaccination is unclear at present (Wong et al). We agree that the duration of protection is a key uncertainty in models of HPV vaccination, which is why we have taken care to present results using three possible values for this parameter (10 years, 20 years and lifelong). Our analysis indicates that vaccination is cost-effective as long as the duration of protection is at least 20 years, and may even be marginally cost-effective (using a £30,000 per QALY willingness to pay threshold) if duration of protection is only slightly greater than 10 years. We do not believe that there is significant evidence that the duration of protection will be less than 10 years, since clinical trials with almost 5 years follow-up have yet to show a decline in protection. If the average duration of protection was 10 years then some waning may be expected to be detected several years earlier due to variations in individual responses to vaccination.

8. It is unclear why Cervarix™ was chosen when it has the same catalogue price per dose as Gardasil™ (Clark). Although the catalogue prices for both vaccines are £80.50, the choice of vaccine was made after a commercially confidential competitive tendering process in which manufacturers of both vaccines participated. The criteria used for adjudication were based partly on our analysis and are publicly available (3).

References

1. Langley PC, White DJ, Drake SM. The cost of treating external genital warts in England and Wales: a treatment pattern analysis. Int J STD AIDS 2004; 15:501-508.

2. Howell-Jones R, Jit M, Beddows S. What will HPV immunisation do for health protection in the UK? Health Protection Matters 2008; 11:16-19.

3. House of Commons Hansard Written Answers for 2 July 2008, columns 943W-944W and 3 July 2008, columns 114W-118W. Available on www.publications.parliament.uk.

Competing interests: None declared
HPV 6-11 and undetermined/low grade cervical lesions:something to be ignored? 9 October 2008
Previous Rapid Response Top
Silvano Costa,
MD, Professor of Gynaecology Oncology, School of Obsetrics and Gynaecology, University of Bologna
Ob/Gyn Dpt. S.Orsola-malpighi University Hospital, via G. Massarenti,13-40138- Blogna, Italy,
Francesco Saverio Mennini, Giampiero Favato, and Mauro Picardo

Send response to journal:
Re: HPV 6-11 and undetermined/low grade cervical lesions:something to be ignored?

Economic modelling is regarded as one of the most reliable tools for the assessment of the effectiveness of healthcare interventions. Currently, however, the uncertainties related to some specific model parameters selected to predict the economic impact of vaccination strategies, represent a limit which has to be carefully taken into account in the public health decision-making process[1-3]. These parameters include the algorithms simulating the natural dynamic transmission of Human Papillomavirus (HPV) infection (including the progression/regression transition states), the actual vaccination coverage rate, and the duration of vaccine protection.

In the model proposed by M. Jit et al[4] for the economic evaluation of HPV vaccination in UK, genital warts have been exclusively incorporated as events induced by HPV 6-11. Therefore the model ignored other cervical events caused by HPV 6-11 infection, such as abnormal Pap tests, ASCUS (atypical squamous cells of undetermined significance), AGUS (atypical glandular cells of undetermined significance), and LSIL (low-grade squamous intraepithelial lesions). These events should be included in the clinical and economic evaluation: in absence of a primary prevention plan[5-6], therapeutic interventions following LSIL or CIN1 (low-grade cervical intraepithelial neoplasia) could lead to considerable, unnecessary costs. A growing body of evidence shows that a relevant percentage of LSIL (in a range between 10% and 20%) is associated with HPV 6-11: this evidence simply confirms previous observations regarding the significant frequency of HPV 6-11 in women with borderline cytology[7-11]. In the Region of Emilia-Romagna in northern Italy, an observational retrospective cohort study that we primarily designed to accurately estimate the frequency of HPV 6-11 associated with borderline cytology (abnormal Pap tests, ASCUS, and AGUS) is currently ongoing. Early results related to 1,041 women with a mean age of 37.5 years (range 23-65 yrs) indicated that HPV 6-11 overall accounted for approximately 14.5% of all evaluated outcomes.

In Italy, the implementation of a multi-cohort vaccination programme with the quadrivalent vaccine is expected to contribute to further reduce the expenses associated with the management and treatment of low-grade cervical lesions and anogenital warts by approximately 34 million Euro. A significant cost-saving could be realised as early as in the first five years, determined by the prevention of outcomes induced by HPV 6-11[11- 12]. Although the realised cost reduction corresponds to just 26% of the total projected cost-savings, nonetheless it would represent a significant economic resource that could be allocated to other priorities of public health.

In general, the effective allocation of resources in healthcare should not be merely based on the persistent search for the lowest treatment price. The provision of innovative diagnostic instruments, drugs and vaccines should require an objective «spending review» that includes all significant costs, in order to support decision-makers to make economically effective and efficient strategic choices. The World Health Organisation (WHO) recognises the limitations of health care decisions based exclusively on price, eventually leading to anti-economic and ineffective outcomes[13]. The WHO concerns definitely apply to the choice of vaccines, as primary prevention is regarded as a core objective in public health.

References

1. Philips Z, Ginnelly L, Sculpher M, Claxton K, Golder S, Riemsma R, et al. Review of guidelines for good practice in decision analytic modelling in health technology assessment. Health Technol Assess 2004; 8(36): 1-172

2. National Institute for Health and Clinical Excellence (NICE). Guide to the Methods of Technology Appraisal. April 2004. www.nice.org.uk. Accessed June 2006

3. Guidelines for the economic evaluation of health technologies: Canada [3rd Edition]. Ottawa: Canadian Agency for Drugs and Technologies in Health; 2006

4. Jit M, Choi YH, Edmunds WJ. Economic evaluation of human papillomavirus vaccination in the United Kingdom. BMJ 2008; 337: a769

5. Russo JF. Controversies in the management of abnormal Pap smear. Current opinion in Obst. Gynecol 2000; 12: 339-343

6. ASCUS-LSIL Traige Study (ALTS) Group. Results of a randomized trial on the management of cytology interpretations of atypical squamous cells of undetermined significance. Am J Obstet Gynecol 2003; 188: 1383-1392

7. Clifford GM, Rana RK, FranceschiS et al. Human papillomavirus genotype distribution in low-grade lesions: Comparison by geographic region with cervical cancer. Cancer Epidemiol Biomarkers Prev 2005; 14: 1157–1164

8. Evans M, Adamson C, Papillo JL. Distribution of human papillomavirus types in Thin prep Papanicolaou tests classified according to the Bethesda 2001 terminology and correlations with patient age and biopsy outcomes. Cancer . 2006; 106:1054-1064

9. Panotopoulou E, Tserkezoglou A, Kouvousi M, et al. Prevalence of human papillomavirus types 6, 11, 16, 18, 31, and 33 in a cohort of Greek women. J Med Virol 2007; 79: 1898-1905

10. Del Prete R, et al. Prevalence and genotypes identification of human papillomavirus infection in a population of South Italy. J Clin Virol 2008; http://dx.doi.org/10.1016/j.jcv.2008.01.011

11. Costa S, Favato G. Evaluation of the economic impact produced by the prevention of events induced by the HPV 6-11 virus types contained in the quadrivalent vaccine. January 2008. Social Science Research Network (SSRN): http://ssrn.com/abstract=1080113

12. Favato G, Pieri V, and Mills R. Cost-Effective analysis of anti-HPV vaccination programme in Italy: A multi-cohort Markov model. February 2007. Social Science Research Network (SSRN): http://ssrn.com/abstract=961847

13. World Health Organization. Immunization financing, supply and procurement. Posted by: WHO, Geneva, Switzerland, 3 February 2006. http://www.who.int/immunization_supply/en/ Accessed December 2007

Competing interests: None declared