Quadrivalent HPV vaccination (qHPV) for adolescent girls is recommended and publicly financed in a number of countries.  This intervention promises to prevent up to 70% of HPV generated cancers in those vaccinated, as well as vastly reducing the burden of genital warts (GW).  In relation to prevention of HPV generated cancer and cancerous lesions, its effective contribution will need to be evaluated in relation to cervical screening, and as part of an overall cancer prevention strategy: the question of the right mix of interventions, including vaccination and its relation to the method and interval of screening, are questions of ongoing debate.  Regarding HPV generated GW, however, it may already be possible to score up certain gains.

A recent Swedish national cohort study of 2.2 million women aged 10-44 years offers evidence of the effectiveness of qHPV in relation to genital warts (GW) that complements the findings of clinical trials regarding its therapeutic efficacity (http://jnci.oxfordjournals.org/content/105/7/469.abstract?sid=e0d178cd-8985-42e4-b934-201c858ca42e). The results of efficacity trials follow pre-specified inclusion criteria and are consequently not generalizable.  Now, however, thanks to the rigorous registration of patients and treatment in Sweden, these researchers have been able, so they claim, to gain an insight into the effectiveness of the vaccine – better than would been possible in many other countries.  Results are stratified by age: vaccine effectiveness was 93% against GW where administered at under 14 years;  80% for girls aged 14-16 years; 71% for those aged 17-19; 48% for those aged 20-22 years; and effectively zero for those above 22 years.  This study confirms at the level of national surveillance the general picture offered, at the level of the STI clinic, by Read & Fairley’s account of the “near disappearance” of heterosexually transmitted GW from  vaccinated <21 years, four years after the introduction of a qHPV program in Melbourne, Australia (vaccination restricted for three years to girls ≤26 years and thereafter to 12-13 years) (http://sti.bmj.com/content/87/7/544.abstract?sid=58e7950b-ccdf-4899-b67b-c7ad3b00081d).

Given the steep decline in effectiveness of qHPV for girls aged >13 years as indicated by these studies, it is disquieting to discover in a recent study of US parents expressing their non-intention to have their teens vaccinated with qHPV that as many as 11% of them gave as the reason that their children were “not yet sexually active”.  The study (Darden & Jacobson) was initiated on the basis of data from the National Immunization Survey, and involved asking parents of children who were “not-up-to-date” with qHPV (along with Tdap and MCV4) and who further indicated their non-intention to vaccinate, the reasons for their decision (http://pediatrics.aappublications.org/content/131/4/645.abstract?sid=082b138f-5a8d-47c5-bf41-5c45d343b129).  The percentage of US girls in the immunization cohort taking up the full three doses of qHPV has increased over the three years 2008-2010 from 17.9% to 32%, but remains substantially below that for Tdap and MCV4 (81.2% & 62.7% for 2010 as against 32%); there is a rising percentage giving as reason for non-vaccination “safety concerns/side effects” – rising over the three years from 4.5% to 16.4%.

The long term implications of incorporating qHPV into the STI toolkit are very unclear, though the reduction of GW is, of course, very welcome.  Aside from the much discussed question of how the choice of cervical screening method should complement vaccination in a post qHPV era (http://sti.bmj.com/content/87/Suppl_1/A14.3.abstract?sid=190a5031-ecc8-4f7f-a4bc-1411d34798f1) there is the question of how a woman’s knowledge that she has been vaccinated is likely to impact on her behaviour in relation to cervical screening (http://jms.rsmjournals.com/content/18/1/41.full.pdf+html?sid=9162f5ac-9762-43c5-bcd8-5e09044c2368;

http://ruizscience.webs.com/HPVvaccineCervicalCancerScreening.pdf).  If we suppose a moderate level of HPV cover combined with a partial collapse in cervical screening compliance what will be effect on the incidence of cervical cancer? Given many health systems are already embarked on the route of qHPV vaccination, it is imperative to maximize the vaccination uptake.  The study of Darden and Jacobson may indicate the challenge posed, with the integration of qHPV into the toolkit, by the urgent need to ensure we communicate the right public health messages.

Older women who have had ≥5 lifetime sexual partners could turn out to be a relatively high-risk group for Human Papilloma Virus (HPV) and associated cervical lesions – regardless of declining sexual activity.  If – that is – Gravitt, Viscidi et al. (http://jid.oxfordjournals.org/content/207/2/272.full.pdf+html?sid=4bc8141c-9b7c-4a13-8a43-451d6fa85fcf; http://jid.oxfordjournals.org/content/207/2/211.full.pdf+html?sid=4bc8141c-9b7c-4a13-8a43-451d6fa85fcf) are correct to see the results of their cohort study of Baltimore women attending obstetric-gynaecology clinic as supportive of the hypothesis that the HPV experienced by these older high-risk women is a reactivated form of the infection.  The virus, they suggest, may remain undetectable in the body for years in order to emerge in later years with immune-senescence – like the varicella zoster virus that reactivates as shingles.  This would explain the double peak in age-specific HPV prevalence (debut and menopause) in some cultures; while the absence of this pattern in the US and N. Europe, they hypothesize, may be an effect of the more restrictive sexual mores of the pre-sixties generation.

What is the evidence?  When the results are stratified according to number of lifetime sexual partners, it is discovered that the Population Attributable Risk (PAR) of high-risk (HR) variants of HPV (i.e. those causing cancerous lesions) due to ≥5 lifetime partners is 87% among the older study participants, and 28% among the younger ones; while the PAR of HR-HPV due to a new sex partner is 7.7% among the older group and 28% among the younger.  This, as the authors argue, is consistent with reactivation of the virus in later life, though they admit reactivation is difficult to prove.

If this hypothesis is supported by further research, it will certainly impact on a number of issues that have been a concern for our readers and contributors.  First, the question of women’s perceived risk of cervical cancer – important because it has been shown to predict cancer screening attendance and has been associated with HPV vaccination uptake (http://sti.bmj.com/content/88/6/400.abstract?sid=2f5a46a3-43d9-4648-a3f6-9f1a46384f61, pp.1-2).  If Gravitt et al. are right, then currently less sexually active older women, who are nevertheless in the higher-risk group on account of past sexual activity, are in serious danger of underestimating their risk; furthermore, that risk could, in reality, be greater that we imagined owing to the behavioural impact of the sexual revolution.  Second, the question of stigmatization, especially in relation to HPV testing at routine “smear” tests (http://sti.bmj.com/content/82/2/169.abstract?sid=2f5a46a3-43d9-4648-a3f6-9f1a46384f61).  If HPV is often present, but undetected, and, when detected, may turn out to have no relation to current sexual activity, then the stigmatizing link between HPV status and current sexual behaviour is, at least, weakened – which ought to have a de-stigmatizing effect.  Third, HPV vaccination, and the concern of certain parents that it might lead to sexual disinhibition (http://sti.bmj.com/content/87/4/349.abstract?sid=44f0f0a6-c872-4cb1-9ca1-afa0aa8db872).  Here, again, the idea that sexually inactive, or less active, people can be at serious risk runs counter to the public perception of a direct link between current sexual behaviour and cervical cancer risk – which ought to make HPV vaccination easier to justify to a public concerned about sexual disinhibition.

The US-based journal Obstetrics and Gynecology carries a front-page editorial: “Abuses in Human Papillomavirus DNA Testing”. A study featuring in July’s issue based on a survey of nationally representative samples of smear-test (pap test) providers reports widespread overuse of HPV testing in the context of routine smear tests. The editorial dramatically concludes that this “seriously wastes government, payer, and patient dollars, but also potentially harms patients, therefore violating the first tenet of the Hippocratic oath”.

To understand the problem, we need to bear in mind:
i. that only some of the HPV genotypes are associated with cervical carcinogenesis, and that HPV DNA tests have recently become available capable of screening for exclusively “high-risk” strains;
ii. that US national guidelines recommend HPV testing (“reflex testing”) of women with atypical squamous cells of undertermined significance (ASC-US): HPV-positive women are referred to colposcopy and HPV-negative women rescreened in 12 months;
iii. that US national guidelines also recommend HPV testing in the context of a routine three-yearly smear-test (“co-testing”), but only in the case of women aged 30 years and older. (In the younger age-groups occurrences have been shown to be pervasive but transient). Negative cytology and positive HPV – or, conversely, ASC-US cytology and negative HPV – triggers re-screening after a year; ASC-US cytology plus positive HPV warrants referral for colposcopic evaluation. More abnormal cytology (i.e. low-grade intra-epithelial lesions and anything more serious) triggers colposcopic evaluation

This study discovered three areas of widespread “abusive” testing:
1. 28.5% of routine test providers employed HPV DNA tests for “low-risk”, as well as “high-risk strains of HPV. The low-risk HPV test is not recommended in this setting – or, for that matter, in any other health setting – though for some reason the test remains widely available;
2. 60% of health-care providers, and 66% of clinics, used routine HPV co-testing in women aged under 30. This presumably leads to over-testing and unnecessary referrals;
3. health care providers performed reflex testing in cases of intra-epithelial lesions, warranting immediate referral for colposcopy. This might indicate that women were being put at unnecessary risk of cervical cancer.

So why is practice so frequently in breach of current (2006) guidelines? The authors mention amongst possible factors: the availability of low-risk HPV DNA tests, test marketing, health care provider confusion over the difference between high-risk and low-risk tests, financial gain in the form of liberal test reimbursement of private practice by private insurance, patient requests to test HPV status. They point to the need for wide-reaching interventions such as limiting test reimbursement to facilitate recommended uses of HPV DNA tests and eliminate low-risk HPV testing, and education materials for patients and health care providers to address patient demand for HPV testing.

Jennifer Wai-Yin Lee et al., “Low-Risk Human Papillomavirus Testing and Other Nonrecommended Human Papillomavirus Testing Practices Among U.S. Health Care Providers”, Obstetrics and Gynecology, vol. 118, no.1, July 2011
Editorial, Philip E. Castle, “Abuses in Human Papillomavirus DNA Testing”, Obstetrics and Gynecology, vol. 118, no.1 July 2011

http://journals.lww.com/greenjournal/pages/currenttoc.aspx