Disease Outcome Indicators (2022)

Cervical cancer incidence is the measure used by the WHO to determine whether elimination has been achieved (2).

In 2018, 937 women were diagnosed with cervical cancer in Australia, an incidence rate of 6.5 per 100,000. Incidence varied between groups and rates (over the period 2012-2016) were three times as high as the elimination target in Indigenous women, and at least twice as high as the elimination target in women living in remote or very remote areas (Figure 1.1). There is a socioeconomic gradient in cervical cancer incidence, with the incidence rates becoming higher with increasing levels of socioeconomic disadvantage. In Indigenous women, a 69% reduction is required to reach the elimination target.

In most demographic groups, cervical cancer incidence rates in 2012-2016 were broadly similar to those in 2011-2015, although in some groups there was an apparent increase (Aboriginal and Torres Strait Islanders, very remote areas, ACT, NT). Rates in these groups were based on fewer than 80 cases over a 5-year period in all cases apart from Aboriginal and Torres Strait Islanders (156 cases).

Cervical cancer increased somewhat in 2018 (from 5.9 per 100,000 to 6.5 per 100,000 in 2018) (Figure 1.2). A transient increase in cancer diagnoses was expected to occur in 2018 as a result of Australia transitioning from 2-yearly cytology-based screening to 5-yearly primary HPV screening on 1st December 2017 (8).

The impact of HPV vaccination on cervical cancer incidence would be expected to occur gradually over time and affect younger age groups first. Cervical cancer incidence rates are extremely low in women aged less than 25, and the number of cases in women aged less than 25 is also very small (6 cases in 2018), making rates in these age groups unstable and difficult to interpret. Cervical cancer incidence has decreased over recent years in women aged 25-29 years (Figure 1.2).

Source: AIHW Australian Cancer Database, published in AIHW 2021 and AIHW 2022 (3,11).

In many population groups, very few cases occur in a year, and so rates are subject to random fluctuation. For this reason, in order to report on and compare cancer incidence across a range of population groups, data are reported across a 5-year period. Results by Indigenous status are based on four jurisdictions (NSW, Qld, WA and NT) where data quality was considered sufficient for analysis. Cervical cancer incidence reported here is based on age-specific rates published in AIHW reports (3). Age-standardised rates are lower than in the AIHW reports, because the age-standardising method and standard population recommended by WHO for measuring cervical cancer elimination has been used (World 2015 female population), to enable comparison with the WHO target and internationally (2,12). Age-specific rates were not published for a jurisdiction when they were non-zero and based on a cell size that was less than five; for this reason, rates for Qld, WA, Tas, ACT and NT are based on imputed cases, or if necessary an assumption there were 2.5 cases, in some age groups. In addition to this, rates for the ACT and NT are based on fewer than 100 cases over a 5-year period. Trends for these jurisdictions should therefore be interpreted with caution.

There is currently a delay of around five years in the availability of national cancer incidence data stratified by population groups in Australia, which limits our ability to conduct timely reporting of cervical cancer incidence in Australia against the primary WHO elimination target. In the context of this delay, and reporting across a 5-year period, it is not surprising that cervical cancer rates in most population groups have not reduced compared to the previous report. This is because the period reported on (2012-2016) precedes the change to the screening program and the impact of HPV vaccination was likely limited in this time period, especially in the all-ages results, due to the long time between a prevented HPV infection and the cancer which otherwise would have occurred (this period is 5-9 years after the vaccination program commenced).  National data (without further breakdown by population groups) is available for 2018, and as expected (8) the transition to HPV screening was accompanied by an increase in cervical cancer incidence. This would be due to the more sensitive screening test leading to the detection of some prevalent cancers and represents a success of the new program (given that earlier detection leads to downstaging of cancer and improved survival). More timely cancer data have been published for Victoria (13). These Victorian data indicate lower cervical cancer incidence in 2020 across a range of age groups compared to 2018 and 2019, but this is consistent with a 7% decline in diagnoses across all cancer types in Victoria in 2020, and likely due to a reduction in people seeking health services and therefore being diagnosed during the COVID-19 pandemic, rather than a true reduction in cervical cancer.

Reducing premature mortality from cervical cancer is the overarching aim of the WHO elimination strategy, so cervical cancer mortality has been identified as a key impact indicator in the global strategy to accelerate elimination (2). Cervical cancer mortality will capture two things that will not be reflected in cervical cancer incidence rates: downstaging (due to screening) and the WHO target of treating 90% of women diagnosed with cervical cancer.

In 2019, there were 231 deaths from cervical cancer. Cervical cancer mortality was 1.4 per 100,000 women in 2015-2019 with some variation in mortality across the population, most notably higher rates for Aboriginal and Torres Strait Islander women, women living in very remote areas, and women in the NT; and a socioeconomic gradient (Figure 2.1).

Source: AIHW Australian Cancer Database, published in AIHW 2021 (3).

In order to report on cancer mortality across a range of population groups, data are reported across a 5-year period (the most recent 5-year period for which data are available). Age-specific rates were sourced from the most recent national report (3). Age-standardisation is performed across all ages using the 2015 World female population, consistent with the methodology recommended for incidence as part of the WHO elimination strategy (2). As a result of this difference in methodology, mortality rates in this report are not comparable to AIHW reports.

National mortality data are more timely than national cancer incidence data. Cervical cancer mortality is relatively low in many groups but many of the groups with higher rates correspond with those with a higher incidence (higher rates in Aboriginal and Torres Strait Islander women and increasing rates with increasing level of socioeconomic disadvantage). There is considerable overlap in the groups identified as having the highest mortality rates (Aboriginal and Torres Strait Islander women, women living in very remote areas, women living in the NT). 

Reductions in cervical cancer mortality over time will reflect a range of factors, including improved prevention, earlier detection (downstaging), improvements to treatment, and providing more women with optimal treatment.

Cervical screening aims to detect specific types of cervical disease (known as high-grade -see Methods below for details), that are considered to be the precursors of cervical cancer, so that women can have treatment to prevent progression to cancer. High-grade abnormalities of the cervix include cervical intraepithelial neoplasia (CIN) that has been graded as moderate (CIN 2) or severe (CIN 3), or for which the grade has not been specified, as well as endocervical dysplasia and adenocarcinoma in situ (3).

This indicator measures the proportion of women screened who are found to have high-grade disease as defined by a positive result on biopsy/histopathology. It will be determined by the underlying proportion with high-grade disease in the screened population, the accuracy of the screening method used, and the subsequent diagnostic steps and follow up undertaken with women found positive on the screening method.

Nationally, detection of high-grade cervical disease in 2020 was 15.7 per 1,000 screened women aged 25-74 years (age adjusted rate 14.0 per 1,000) (Figure 3.1). The rate of CIN3 detection was 8.9 per 1,000 (age adjusted 8.0 per 1,000 and 56.7% of all high-grade disease detected) and of was 0.6 per 1,000 (3.8% of all high-grade disease) (Figure 3.2). Women residing in the least socioeconomically disadvantaged quintile had a high-grade detection rate lower than the national average (14.0 per 1,000). Women living in outer regional and remote areas had higher high-grade detection rates than the national average (at 16.8 and 19.0 per 1,000, respectively). Queensland had the highest detection rate at 20.1 per 1,000 screened women (Figure 3.1).

Detection of high-grade cervical disease, and of CIN3, was highest in the 30–34-year-old age group at 24.2 and 14.3 per 1,000 screened women respectively (Figure 3.2). Detection of AIS was highest at aged 35–­39 years (1.1 per 1,000). The age group with the highest proportion of high-grade disease that was CIN3 was the 65-69-year-old age group (63.6%) and the lowest proportion was those aged 25-29 at 51.8%. The age groups with the highest proportion of high-grade disease that was AIS was the 35-39- and 40-44-year-old age groups (both at 5.6% - the only age groups with proportions over 5%). No cases of AIS were detected in the 65-69-year-old age group in 2020.

High-grade detection rates by Indigenous status were unavailable for publication.

Source: Analysis of NCSR data (NCSR RDE 3.4.1 07/08/2021); data prepared by the Screening Analysis and Monitoring Unit of the Australian Institute of Health and Welfare.

High-grade cervical disease was defined as a histology result of S3.1 (High-grade intraepithelial lesion (HSIL) (CIN NOS)), S3.2 (HSIL (CIN 2)), S3.3 (HSIL (CIN 3)) or E3.1 (Endocervical dysplasia), E3.2 (Adenocarcinoma in situ) or E3.3 (Mixed carcinoma in situ/adenocarcinoma in situ) (14).

Crude rates are presented, except for the national high grade and CIN3 rate, due to the suppression of small cell sizes preventing calculation of age standardised rates. Cell suppression impacted some age groups for AIS and for stratifications by State and Territory, remoteness and socioeconomic status. Age-standardised rates were the number of women with a high-grade abnormality detected by histology per 1,000 women screened, age-standardised to the Australian 2001 Standard Population (15).

In the final six months of the previous Pap based screening program (Jan-July 2017), the crude high grade detection rate for women aged 25-74 years was 6.6 per 1,000 screened women and highest in the 25–29-year-old age group (at 14.4 per 1,000) (16). The 2020 rate was significantly higher, as was predicted by modelling (8), both due to the move to HPV based screening with its higher sensitivity and because fewer women were expected to attend in 2020 and also those attending were all at relatively higher risk of high-grade disease: women screening in 2020 were either entering the program at age 25, were attending because they were under surveillance for a previous positive HPV test, or else were overdue for screening (due to the transition from the previous 2nd yearly screening interval). The peak rate occurred in women aged 30-34 for the first time in the program’s history, likely due to significant vaccine impact preventing disease in 25–29-year-old women (who would have been offered HPV vaccine in 2007 at age 12-16, prior to the average age of sexual debut in Australia). Women offered HPV vaccination as catch up as adults (aged 18-26 in 2007) were aged 31 to 39 in 2020.

There is known under-enumeration of histopathology reports held in the NCSR, as safety net follow-up (and attempts to retrieve not yet notified histopathology) cease when an MBS item number consistent with cervical histology is located (indicating the woman has received further assessment). Additionally, due to teething issues during initial register implementation, 2018 data is known to be incomplete. It is likely that high grade rates estimated from the early years (2018-2019) may more significantly underestimate the true high-grade detection rate, with reporting to the register improving over time, so these data should be interpreted with caution. The AIHW 2021 report using these data estimated an age standardised high grade rate of 7.5 and 8.6 per 1,000 women aged 25-74 years screened in 2018 and 2019 respectively, with the peak incidence rate in both years seen in women aged 25-29 (3).

There was some evidence in 2020 of variation in high grade detection rate by remoteness of residence, with women in remote areas having higher rates, and by socioeconomic quintile of residence, with women living in the least disadvantaged quintile having lower rates. Women residing in Queensland, WA and SA had higher rates that in other jurisdictions in 2020. National data by Indigenous status remain unavailable for inclusion in this report, with no reporting of high-grade rates amongst Aboriginal and Torres Strait Islander women published since Renewal. This is an important priority for monitoring, given that previous research in Qld, the NT and SA under the cytology-based program demonstrated that Indigenous women had higher rates of high-grade disease at screening than non-Indigenous women (about twice as high, although with a closing of this gap in the most recent period in the NT, the jurisdiction with the largest proportion of the population who are Indigenous) (17-19).

The ability to achieve cervical cancer elimination is dependent upon the reduction in prevalence of oncogenic HPV types through HPV vaccination. Monitoring infection prevalence, in settings where this is feasible, provides a direct measure of the extent to which this reduction is being achieved and sustained over time.

In 2021, 2.3% of women screened aged 25-74 years were positive for HPV16 and/or 18, the most oncogenic HPV types prevented by vaccines utilised in National HPV Vaccination Program (three doses of quadrivalent until 2018, then a switch to two doses of nonavalent). Positivity was low across all socioeconomic groups, remoteness areas and jurisdictions. Positivity for oncogenic HPV types other than HPV16 or 18 was 8.8%, with higher positivity for women living in the highest socioeconomic quintile areas, remote and very remote areas, and women in the NT. HPV prevalence was similar in 2021 and 2020, but slightly higher than 2019, with the greatest difference seen for HPV types other than HPV16 or 18 (Figure 4.1).

The additional five HPV types targeted by the nonavalent HPV vaccine (HPV31, 33, 45, 52 or 58) and the remaining HPV types not targeted by vaccination (HPV35, 39, 51, 56, 59, 66 or 68) were detected among 5.1% and 7.4% of screening tests respectively, with little variation across socio-demographic characteristics (remoteness, and socioeconomic group) (Figure 4.2).

HPV16 and HPV18 positivity was low, and relatively stable across age groups. In contrast, positivity for remaining oncogenic HPV types was highest among women aged 25–29 years and decreased with increasing age (Figure 4.3). These patterns reflect the population-level impact of Australia’s quadrivalent HPV vaccination program on reducing infection with HPV types 16 and 18 (20,21). The impact of the nonavalent vaccine is yet to be seen because these cohorts are not yet at screening age. 

Breakdowns by Indigenous status were not available for publication. 

Source: Analysis of NCSR data (NCSR RDE 3.4.1. 08/01/2022 for 2021 and 2020 results; NCSR RDE 3.4.1 07/08/2021 for 2019 results); data prepared by the Screening Analysis and Monitoring Unit of the Australian Institute of Health and Welfare.

Calculation of proportions in Figure 4.1 were based on national data reported to the NCSR. Data presented are based on the clinical management pathways of the NCSP: proportion of women with HPV16 and 18 (regardless of positivity for other types) versus proportion of women with other oncogenic types (in the absence of HPV16 and 18). Figure 4.2 and Figure 4.3 were derived from a small subset (n~4,500) of women whose samples was submitted to VCS Pathology in 2020 and tested on the Seegene Anyplex II HR HPV assay, which provides discrimination between individual HPV types. This subset represents 35% of all screening tests conducted on that platform and 2.6% of tests conducted in Victoria in 2020. Compared with national data, this group had similar overall HPV16 and 18 prevalence (2.4% vs 2.3%), but slightly higher prevalence of oncogenic HPV types other than HPV16 or 18 (10.3% vs 8.6%). Results were derived using the standard method for calculating population prevalence: sum of women positive for a given HPV type divided by the total count of women with individual genotyping data.

HPV prevalence in 2021 (and 2020) was higher than in 2019. This was expected, as women screened in those years were either entering the program at age 25 or else were overdue for screening (due to the transition from the previous 2nd yearly screening interval).

Many HPV vaccinated women have now attained the age at which they are participating in cervical screening. Women offered HPV vaccines in the catch-up program are turning 42 from the second half of 2022 and those in the school-based program are turning 32-33 years of age. Linkage between the NCSR and AIR will in future permit comparisons of prevalence of HPV infection detected at screening between vaccinated and unvaccinated women.

Surveys prior to the start of the vaccination program among women attending cervical screening suggested similar prevalence of HPV16/18 in Indigenous compared with non-Indigenous women at all ages, but a higher prevalence of other oncogenic types among Indigenous women aged 31 to 40 years (22). A survey conducted after the program had begun found declines in prevalence similar to those observed in non-Indigenous women, with an HPV16/18 prevalence among Indigenous women aged 18–26 of 1.4% in 2014-15 (23). A 2021 study of women 18-35 years found that the prevalence of HPV16/18 was very similar in Indigenous and non-Indigenous women (2.3% versus 1.9% respectively) (24).