HPV vaccine results do not apply to UK

Vaccination programmes to protect against cervical cancer are “likely to cut the numbers who get the disease”, The Guardian has reported. In 2008, the UK began such a programme, offering teenage girls a vaccine against the human papilloma virus (HPV), the viral cause of cervical cancer.

The news is based on an important Australian study that monitored the incidence of high- and low-grade cervical abnormalities at cervical screening tests, both before and after the HPV vaccination programme was introduced. However, while this well-conducted research monitored the situation in Australia, the UK's vaccination and cervical screening programmes are different. The UK vaccination programme was introduced more recently, uses a slightly different vaccine, and does not target as wide an age group as Australia's. Therefore, the results do not apply to the UK.

In addition, the success reported by the newspapers was a decrease in incidence of high-grade cervical abnormalities (which may or may not progress to cancer) seen only in under-18s, but not in other age groups. The study reported the results of screening tests only, and did not include data on confirmed cervical cancers.

Overall, the situation in the UK is different to that in Australia, and the effects of the UK's vaccination programme cannot be assumed to be the same without appropriate study.

Where did the story come from?

This Australian study was carried out by researchers from the Victorian Cytology Service and the Biosciences Research Division of the Department of Primary Industries. No funding was received for this research.

The study was published in the peer-reviewed medical journal The Lancet.

This important research monitored the effect of introducing vaccination on rates of cervical abnormalities in Australia. However, newspaper reports of this research could be interpreted as implying that the results have a direct application to the UK’s use of the HPV vaccine. This is not supported by the study as Australia has a different vaccination and cervical screening programme to the UK. The newspapers also failed to make it make it immediately clear that the results were only observed in girls aged under 18. This age group is not offered routine cervical screening, either in Australia or the UK. Therefore, the girls in the study were not representative of the teenage population as a whole who have been given the vaccination.

What kind of research was this?

This before and after study compared the incidence of cervical abnormalities in the female population of Victoria in Australia in 2003–2007 (before the introduction of the HPV vaccination programme) and 2007–2009 (after its introduction).

The quadrivalent human papilloma virus (HPV) vaccine was introduced for all women aged 12-26 years old in Australia in April 2007, after it was licensed for use in mid-2006. The quadrivalent vaccine provides protection against high-risk HPV types 16 and 18 that are detected in 70% of cervical cancers, as well as the low-risk types HPV 6 and 11 that cause 90% of genital warts.

Australia is said to be the first country to introduce an extensively funded HPV vaccination programme. Other countries have variably used the quadrivalent or bivalent vaccines (the latter protecting against the high-risk HPV types 16 and 18 only), and have chosen to implement short-term catch-up programmes aimed at older age groups, ranging from 13–18 years to 26 years. The NHS HPV vaccine programme began in 2008 and, unlike Australia, uses the bivalent vaccine.

What did the research involve?

The study used data available from the Victorian Cervical Cytology Registry between 2003 and 2009, which promotes regular participation of women in the National Cervical Screening Program and follow-up of women with abnormal smear tests. It also compiles statistics for the purpose of monitoring and research. It covers a population of over 2.7 million girls and women, of whom fewer than 1% request that test results are not held on the registry.

The researchers looked at records of histology (tissue sample analysis) to compare the incidence of high-grade cervical abnormalities (the primary outcome) and low-grade cell abnormalities (the secondary outcome) across five different age groups (under 18 years old, 18–20, 21–25, 26–30, and 31 or over). These were taken from January 2003 to March 2007 (before the programme began) and April 2007 to December 2009 (after it was introduced).

High-grade cervical abnormalities were defined as cervical intraepithelial neoplasia (CIN) of grade 2 or worse, or adenocarcinoma in situ. CIN refers to abnormal cells that are not yet cancerous, but which may progress to squamous cell cancer of the cervix. Abnormalities are graded as 1 to 3, depending on the thickness of the surface layer of the cervix that is affected (one-third, two-thirds, and full thickness respectively). Adenocarcinoma in situ is a precancerous lesion that can also be described as CIN 3. It is not yet classed as cancer if all the cells are contained within the surface layer of the cervix and if they have not yet spread to the underlying tissue of the cervix. However, treatment must be performed as soon as possible and further investigations are needed to see whether there has been any spread to the deeper tissues of the cervix (i.e. whether cervical cancer is confirmed).

In Australia, national guidance and protocols are used to guide the management of these different grades of CIN or adenocarcinoma in situ. However, in general, all cases will be immediately referred for further investigation. This report only presented the incidence of these high abnormalities and did not provide data on the number of cases of confirmed cervical cancer.

The study also considered the incidence of low-grade cell abnormalities, which in Australia are usually managed by arranging another smear test 12 months later to see whether the abnormality has resolved or whether further investigations or treatment are needed.

What were the basic results?

The researchers compared the proportions of screened women in the different age groups who were found to have high-grade cervical abnormalities. In the four years before vaccination and the 2.5 years after screening, the incidence of high-grade abnormalities were:

under 18 years old – 0.80% before vaccination and 0.42% after
18–20 years – 1.20% before and 1.17% after
21–25 years – 1.53% before and 1.71% after
26–30 years – 1.26% before and 1.43% after
31 years and over – 0.35% before and 0.37% after

The researchers noted that since the introduction of the vaccination programme, there has been a significant decrease in the incidence of high-grade cervical abnormalities among girls aged younger than 18 years (0.38% decrease in incidence, 95% confidence interval 0.61% to 0.16%). No significant decrease in incidence was seen for the 18–20 year group, though a small increase in incidence since the introduction of vaccination was seen for older age groups.

The proportions with low-grade cell abnormalities for the same time periods were:

under 18 years – 12.2% before and 12.5% after
18–20 years – 11.0% before and 10.9% after
21–25 years – 7.9% before and 7.3% after
26–30 years – 5.0% before and 4.4% after
31 years and over – 2.5% before and 2.0% after

In contrast to high-grade abnormalities, there was no decline in incidence of low-grade tissue abnormalities in the under-18s or 18–20 year group. While there was a decrease in the older age groups, the researchers say that these reflect longer-term trends that began before the vaccination programme.

How did the researchers interpret the results?

The researchers say that this is the first report of a decrease in incidence of high-grade cervical abnormalities among under-18s, which was seen within three years of implementation of a population-wide HPV vaccination programme. However, they say that there is a need to cross-reference vaccination and screening registers to confirm that this observation is attributable to vaccination. The study looked at incidences of vaccination and abnormalities among a population, but did not confirm that a lower rate of abnormalities was specifically seen in only those women vaccinated.

They also say that they need to monitor participation in cervical screening programmes among women who have been vaccinated.

Conclusion

This important research from Australia monitored the effect of introducing vaccination on rates of cervical abnormalities. Although it had strengths, such as its use of extensive population-based data, caution should be applied before using this data to hail either the UK’s or Australia’s vaccination campaigns as a success:

This study only assessed the effect of introducing a vaccination programme in Victoria in Australia. No assumptions should be made from this about the effect of the programme in the UK, particularly as the two programmes use different types of HPV vaccines, the UK programme was introduced more recently and has not targeted such a wide age group.
Establishing the outcome of the UK’s HPV vaccination programme will require separate research using histological databases in the UK, as well as examining the incidence of cervical abnormalities from cervical screening before and after the vaccination was introduced.
In this Australian study, a decrease in incidence of high-grade abnormalities was only observed in under-18s, the group with the lowest incidence of cancer and cervical abnormalities. No effect was seen in older age groups, though it is assumed that, in time, the benefits would extend to them too.
Under-18s are not called for routine cervical screening in the UK, where screening is currently offered to women aged 25 and over. In Australia invitation for cervical screening commences at 18. Why these teenagers under 18 in Australia have received a smear is unclear, but it may be because they were experiencing some gynaecological symptoms. The true incidence of high- or low-grade abnormalities in girls under 18 before and after introduction of vaccination in Australia is therefore unknown. This is because girls of this age who have had a smear test are likely to represent only a small proportion of the total girls of this age group. As invitation for screening currently starts at age 25 in the UK, any benefit for young women here will take longer to detect (until girls who have received the vaccine in the UK start attending regular smear tests).
The report only examined high-grade cervical abnormalities or low-grade abnormalities, but not confirmed cancer. This is because the study only had results of screening, and no high-grade cervical abnormalities can be classed as cancer without further investigation. The Daily Mirror ’s headline that the vaccine could halve the number of girls who get cervical cancer is, therefore, incorrect.

In the UK, the national vaccination programme for girls aged 12-13 years old was introduced in September 2008, with a catch-up campaign for older girls up to the age of 18 introduced shortly after. In the UK, the bivalent vaccine, which protects against high-risk HPV types 16 and 18, is used. Also, while Australian cervical screening routinely begins at 18 years of age, in England and Northern Ireland it begins at 25 and in Scotland and Wales at 20 years. The UK's situation is therfore different from Australia's, and it will take several years to detect any effect the HPV vaccine has on cervical abnormalities detected through screening in this country.

Overall, due to key differences between the vaccination and screening programmes in the UK and Australia, the effects of the UK’s vaccination programme cannot be estimated without appropriate study in the UK.