2. The rationale for primary HPV screening

Over 100 different types of human papillomavirus (HPV) have been identified and there are more than 40 anogenital HPV types, 15 of which are classified as ‘high risk’ or oncogenic.[1][2] HPV infections can induce the development of either benign or malignant lesions. Benign lesions (including non-genital and anogenital skin warts, oral and laryngeal papillomas and anogenital mucosal condylomata) are caused by HPV types designated ‘low risk’ (notably HPV 6 and 11, which cause anogenital warts).[3] Persistent infection with oncogenic HPV types is generally subclinical, but can result in the development of a range of anogenital tumours including cancers of the cervix, anus, penis, vulva and vagina.[3] HPV infection is also associated with squamous cell carcinomas of the head and neck, particularly oropharyngeal cancers.[4] 

Anogenital HPV infections are transmitted mainly by skin-to-skin or mucosa-to-mucosa contact.[5] Penetrative sexual intercourse is not strictly necessary for transmission and HPV can be transferred to the cervix from original infection at the introitus.[6] Therefore, genital skin-to-skin contact, vaginal sex, oral sex, and anal sex represent types of sexual activity that may facilitate the person-to-person transmission of anogenital types of HPV. 

To the implementation of HPV vaccination, cervical HPV infection was common in sexually active women. A study of pre-vaccination cervical HPV prevalence found multiple infections were common in Australian women, with a wide range of HPV types detected (HPV 16 being the most common), and incidence peaking in the years following the start of sexual activity.[7] International data show prevalence is high even in young women who are with their first partner and are monogamous, with HPV infection rates of 30% within 1 year of becoming sexually active and 48% within 3 years.[8] HPV prevalence peaks soon after the average age of first sexual intercourse. In Australia the median age of sexual debut is 17 years for females born from 1965 onwards.[9] Prevalence among women aged over 30 years is much lower than among younger women.[10] Most infections are cleared by the immune system within 1–2 years.[11]

Back to top

There is overwhelming evidence that HPV infection is necessary for development of cancer of the cervix. The International Agency for Research on Cancer has classified certain HPV types as Group 1 carcinogens (agents for which there is sufficient evidence that it is carcinogenic to humans).[12] While HPV infection is necessary for the development of cervical cancer, it is certainly not sufficient.[13] Worldwide, it has been estimated that pre-vaccination, there were about 100 million adult women infected with oncogenic HPV types.[14] This compares with approximately 528,000 new cases of cervical cancer worldwide each year.[10] However, the risk of developing cancer increases significantly with persistent HPV infection.[15] 

Women with persistent infections, especially with HPV 16, are at significantly higher risk of cervical cancer and its immediate precursor lesion, cervical intraepithelial neoplasia (CIN) grade 3 (CIN3).[16][17][18] However, although a majority of women are infected with HPV within a few years of sexual debut, incidence of cervical cancer peaks at about age 45 years,[19] which suggests that progression from persistent infection to invasive cervical cancer is generally slow. More than 70% of cervical squamous cell carcinomas and about 78% of cervical adenocarcinomas are caused by oncogenic HPV types 16 and 18.[20] HPV 16 is the most carcinogenic, accounting for about 55–60% of cervical cancers, while HPV 18 accounts for a further 10–15% of cervical cancers.[2][21]

The four major steps in cervical cancer development are HPV infection/acquisition, viral persistence (versus clearance), progression to cervical pre-cancer, and invasion.[22] The natural history of HPV and cervical intraepithelial neoplasia (CIN) is summarised in Figure 2.1.. It has been estimated that persistent HPV infections and pre-cancer are established, typically within 5–10 years, from less than 10% of new infections.[22] However invasive cervical cancer arises only rarely, in a small proportion of women with pre-cancer. If invasive cancer arises, this generally occurs over many years – often decades – with the peak risk occurring after about age 35–55 years.[22] 

Low-grade squamous intraepithelial lesions (LSIL) are manifestations of acute HPV infection with any type (oncogenic types or other types such as 6, 11), rather than cancer precursors,[23] and most will resolve spontaneously within 12 months.[24] Some high-grade squamous intraepithelial lesions (HSIL [CIN2]) will regress over time, but these lesions are associated with a higher risk of progression compared with LSIL. At the molecular level, pre-cancerous lesions occur when oncogenic HPV is not cleared, infects immature cells and prevents maturation and differentiation, resulting in the replication of immature cells and the accrual of genetic changes that can lead to cervical cancer (Figure 2.1).[25] Lesions histologically classified as CIN2 represent a heterogeneous mix of low-grade and high-grade abnormalities at the molecular level. Clinically, however, lesions classified as CIN2 or above (CIN2+) are often termed ‘high grade’ or ‘precancerous’ and are treated.

In women with oncogenic HPV infection, current cigarette smoking significantly increases the risk of squamous cell carcinoma, but not of adenocarcinoma.[26] Other co-factors that increase the risk of progression to cervical cancer in women who have a persistent oncogenic HPV infection includes multiparity (more than five full-term pregnancies),[27] early age at first full-term pregnancy,[27] and the use of oral contraceptives.[28] Immune deficiency (e.g. acquired by HIV infection) contributes significantly to persisting HPV infection and cervical cancer risk.[29]

Figure 2.1. HPV to cervical cancer

Acknowledgment: Adapted from Schiffman M, 2005.[30]

Back to top

Since 2007, prophylactic vaccination against HPV in pre-adolescent females has been introduced in most developed countries, supported by modelled evaluations of the cost-effectiveness of this intervention.[31] Two first-generation vaccines are available: the quadrivalent vaccine (Gardasil, CSL/Merck) and the bivalent vaccine (Cervarix, GSK). These have been shown to be effective in preventing persistent infection and histologically confirmed HSIL (CIN2/3) in females naïve to HPV vaccine types[32][33] and at preventing persistent infection, external genital lesions and anal intraepithelial neoplasia in males.[34] First-generation HPV vaccines protect against oncogenic HPV types 16 and 18, which are together responsible for approximately 70% of invasive cervical cancers.[2] The quadrivalent vaccine also protects against oncogenic HPV types 6 and 11, which cause more than 90% of anogenital warts. As nearly 80% of adenocarcinomas are associated with the HPV types 16/18,[35] prophylactic HPV vaccination is also expected to be effective in preventing these cancers.[36] 

Large-scale studies have shown the HPV vaccines to be safe and well tolerated. Gardasil, the quadrivalent vaccine distributed via the National Immunisation Program, has been assessed as safe and effective by the Australian Therapeutic Goods Administration, the US Food and Drug Administration and the European Medicines Agency. For further information about HPV vaccine safety and efficacy, as well as dosage and administration, please refer to the Australian Immunisation Handbook.

For maximum efficacy, prophylactic vaccines need to be administered to individuals prior to HPV exposure.[14] Current vaccines do not have a therapeutic effect in those already infected with HPV. It is recommended that HPV vaccines be provided before sexual activity commences. In Australia, the National Immunisation Program targets ongoing vaccination towards adolescent/pre-adolescent girls, aged 11–13 years. Young males were included on the National HPV Vaccination Program from 2013,[37] Australia-specific modelling has suggested this will increase the level of ‘herd immunity’ protection to females.[38] 

In late 2014, the US Food and Drug Administration approved a second-generation 9-valent vaccine, which targets the quadrivalent oncogenic HPV types and five additional oncogenic HPV types (31, 33, 45, 52, and 58). Together, oncogenic HPV types included in the 9-valent vaccine are found in approximately 90% of cervical cancers globally.[39] Compared with the quadrivalent vaccine, the 9-valent vaccine has been shown to be 97% effective for prevention of high-grade cervical, vulvar, and vaginal disease caused by types 31, 33, 45, 52, and 58 in individuals naïve for these types, and to be associated with non-inferior seroconversion for the oncogenic HPV types included in the current quadrivalent vaccine: 6, 11, 16, and 18.[40] The Australian Pharmaceutical Benefits Advisory Committee will evaluate the 9-valent vaccine for inclusion in the National Immunisation Program. As of early 2016 this had not yet occurred.

Back to top

Australia was the first country to initiate a national public vaccination program, which began in 2007. Female vaccination uptake is approximately 71–72% for three-dose coverage in girls aged 12–13 years, and catch-up in women aged 18–26 years (conducted from 2007–2009) achieved coverage rates of approximately 30–50%.[18][41] From 2013, males aged 12–13 years have also been vaccinated at school with a 2-year catch-up to Year 9 (age approximately 15 years). Via herd immunity, male vaccination will also provide incremental benefits to females, and is expected to lead to further reductions in rates of infection with vaccine-included oncogenic HPV types and high-grade cervical abnormalities in females.[38] Several factors have come together to achieve a more rapid impact of vaccination on cervical screening in Australia than in many other countries. These include the early introduction of HPV vaccination, the extended catch-up to age 26 years, the early age of screening commencement at 18–20 years, with the consequent overlap of vaccinated and screened populations from the inception of the vaccination program, and the relatively high coverage rates for vaccination and cervical screening. 

After the introduction of vaccination, Australia experienced rapid falls in rates of infections with vaccine included oncogenic HPV types, in anogenital warts and in histologically confirmed HSIL. These reductions have now been documented extensively in young females and also in heterosexual males due to herd immunity effects.[42][43][44][45][46] Between 2004–2006 and 2012, rates of CIN2/3 among women aged less than 20 years decreased by 53%, while rates of confirmed CIN2/3 among women aged 20–24 years were stable until 2010, then decreased by 21% in the following year.[42] 

It is expected that rates of HSIL (CIN2/3) will continue to decline, and that the decline will extend to older age groups as the cohorts offered vaccination continue to age. As successive cohorts of girls are vaccinated, and the vaccinated cohorts mature, the risk of cervical cancer will continue to fall.

However, cervical screening will remain necessary, since the current vaccine does not cover all oncogenic HPV types that can lead to cervical cancer and may not be effective in women exposed to HPV prior to vaccination. If 9-valent vaccines are introduced, the extent of these reductions in HSIL (CIN2/3) would eventually be expected to increase further, but this is not expected to occur until 2030, since cohorts aged 12–13 years offered next-generation vaccines will not reach the new target age group for cervical screening for some years.

Back to top

Even in completely unvaccinated populations, rates of invasive cervical cancer are low in women younger than 25 years[42] (see also Cervical cancer in Australia). A substantial body of evidence has found that cervical screening in this age group has little or no impact on the risk of developing invasive cancer before age 30 years.[47] Almost all countries with organised programs recommend that cervical screening commences at age 25 or 30 years and the International Agency for Research on Cancer (IARC) recommends regular cervical screening begin at the age of 25.[48] This starting age achieves the best balance of benefits and harms for cervical screening, as detailed in the report of effectiveness modelling and economic evaluation[49] undertaken during renewal of the NCSP.[50] 

The evaluation undertaken for renewal of the NCSP considered a range of screening strategies starting at age 25 years.[50] The evaluation predicted that, compared with pre-renewal NCSP based on cytology screening in sexually active women starting at age 18–20 years, 5-yearly HPV screening starting at age 25 will be associated with reductions in cervical cancer incidence and mortality rates of at least 15%; this reduction is predicted even if the population had never been offered HPV vaccination. Subsequent modelling, taking into account post-colposcopy management as recommended in these guidelines, has predicted reductions of 31-36% in cervical cancer incidence and mortality in unvaccinated cohorts, and reductions of 24–29% in cohorts offered vaccination (see Appendix A. Modelled evaluation of the predicted benefits, harms and cost-effectiveness of the renewed National Cervical Screening Program (NCSP) in conjunction with these guideline recommendations). 

In the post-vaccination era the risk of cervical cancer in women aged 25 years or less has been reduced even further.[50] In Australia, the prevalence of infections with vaccine-included oncogenic HPV types decreased by 78% among women aged 18–24 years between 2005–2007 (pre-vaccination era) to 2010.[44]This substantial reduction occurred within a few years after vaccination was introduced. Furthermore, the impact of the vaccination program has not been confined to those who are individually vaccinated. Even prior to the implementation of male vaccination, females showed herd immunity due to the vaccination of other females in the community. The effect has been documented as a fall in the prevalence of infections with vaccine-included oncogenic HPV types in unvaccinated women aged 18–24 years that occurred by 2012; the effect of herd immunity is expected to be even further increased following the implementation of male vaccination in 2013. As detailed above, a dramatic fall in histologically confirmed HSIL (of over 50%) has also been documented in women under 25 years of age in Australia.[44]

Therefore, several factors have combined to support a starting age of 25 years in the renewed NCSP, including:

• the relatively lower rates of cervical cancer in women less than 25 years of age
• the lack of evidence for the effectiveness of cervical screening in this age group
• the impact of HPV vaccination on further substantially lowering the risks for both vaccinated and unvaccinated young women.

Back to top

Due to the relationship between persistent infection with oncogenic HPV types and the development of cervical cancer, testing for the presence of oncogenic HPV DNA in cervical cell specimens has the potential to identify women at increased risk of developing cervical cancer. Women in whom oncogenic HPV types are not detected are at very low risk of CIN3 or cancer for at least 5 years.[51][52] HPV DNA testing in cervical screening is more sensitive than cytology and detects high-grade lesions earlier, thus preventing more cervical cancers.[53][54] Screening using HPV testing has the potential to improve identification of adenocarcinoma and its precursors.[51][55] A large body of evidence, including data from randomised trials in developed countries, has shown HPV testing in primary screening is superior to cytology.[55][56][57] Analysis of four European randomised controlled trials found that, compared with cytology, HPV-based screening provided greater protection against invasive cervical cancers.[55] Using the HPV test as a primary screening tool allows for development of population-based screening recommendations based on individual risk assessment rather than vaccination status, which will change over time as vaccinated cohorts reach screening age.[58] Given oncogenic HPV types 16 and 18 account for the greatest proportion of infections causing cervical cancer, screening tests with partial genotyping for oncogenic HPV types 16/18 are expected to improve risk stratification of women with a positive oncogenic HPV test result in cervical screening programs.[23] 

In the renewed NCSP, HPV testing at 5-year intervals from age 25 years has been recommended and adopted as the preferred pathway for screening in Australia.[59] The review of strategy and policy undertaken for renewal of the NCSP identified options for HPV screening in Australia that were predicted to result in life–year savings, compared with current practice.[59] The greatest gains in effectiveness were associated with strategies based on primary HPV testing with partial genotyping for HPV 16/18, in which women with these HPV types are referred directly for diagnostic evaluation.[59] An Australian trial of HPV screening with partial genotyping, Compass, is providing information on resource use and outcomes of the renewed NCSP in both unvaccinated and vaccinated women. This information has informed the development of these guidelines. Conducted in the state of Victoria by the Victorian Cytology Service Ltd and Cancer Council NSW , Compass has two phases: Phase I (the pilot), which recruited 5000 women, and Phase 2 (the main trial) which is recruiting 121,000 women.

Back to top