Beyond the blind spot: considering the benefits of comprehensive skin cancer surveillance

Australians are diagnosed with skin cancer at the highest rates worldwide, including melanoma,¹ Merkel cell carcinoma² and keratinocyte cancers (KCs; basal cell carcinoma [BCC] and squamous cell carcinoma [SCC]).³ The KCs are by far the most common, and they impose a significant health and financial burden on the Australian population and healthcare system. The most contemporary estimates of incidence are for the period 2011–2014, at 770/100,000 for BCC and 270/100,000 for SCC.⁴ At least two in three Australians will be diagnosed with one or more of these cancers in their lifetimes,⁵ and many people are multiply affected, with lesion-based rates approximately two-fold higher than the person-based rates.⁴ KCs also represent the most expensive cancer for our health system to diagnose and treat.⁶ The total cost to the Australian health system to treat KCs in 2019–2020 was in excess of A$1.4 billion,⁶ accounting for 12% of all cancer-related health system expenditure and about 1% of total health system costs for all diseases.⁶ The incidence of KC is expected to increase in the coming decades, partly due to population ageing.

Long-term surveillance of trends in cancer incidence and mortality forms the basis of estimates of the cancer burden used to establish priorities for cancer control programs, so decision-makers can make informed and evidence-based policy decisions. Accurate and up-to-date population-based incidence data enable the monitoring of temporal trends in incidence, the identification of disparities across different populations, the targeting of prevention and early detection efforts to high-risk groups and the evaluation of the impact of cancer control and public health education initiatives. High-quality data from cancer registries are seen as the gold standard for these comparisons.

Unlike other cancers (including melanoma), where reporting diagnoses to cancer registries is mandated by law in all states and territories, KCs are not notifiable by law and are therefore not routinely registered by cancer registries except in Tasmania. Thus, despite causing substantial morbidity and treatment costs, reliable and up-to-date incidence statistics for these cancers are not available, and there is no accurate record of the disease burden.⁵ The primary challenge to national registration is the high number of diagnoses and constraints on available resources in managing the extensive volume of notifications.

Keratinocyte cancer registration in Australia: the current situation

Although there is no legislative requirement, the Tasmanian Cancer Registry (TCR) has registered histologically confirmed BCC and SCC diagnoses by pathology laboratories, hospitals and oncology clinics since its establishment in 1977. Registration was conducted manually until 2005 when it paused because of resource constraints. Although the registry continued to receive notifications, they were not processed until 2020 when an automated coding system was implemented, and the TCR now has a record of all KCs registered in Tasmania between 1978 and 2018.⁷

The Queensland Cancer Register is the only other jurisdictional registry that is moving towards notification of BCCs and SCCs.⁸ Legislative changes implemented in 2025 will mean that BCCs and SCCs with perineural invasion and/or metastases will become notifiable. Although lesions with these features represent a very small proportion of all BCCs and SCCs diagnosed in Queensland, they account for the greatest burden in terms of morbidity and mortality, often requiring specialised surgery, hospitalisation and radiation therapy.

In the absence of national cancer registration data, policymakers rely on indirect measures of incidence, including administrative claims data from the Medicare Benefits Schedule (MBS) or local surveys. A recent systematic review of KC incidence in Australia concluded that all recent population-based reports of the incidence of BCC and SCC in the Australian population have limitations and are likely underestimates.⁵ Until 2002, periodic national surveys provided the only data on the burden of KCs on the Australian population.^9–11 The most recent survey was conducted more than 22 years ago now, however. More recent estimates of incidence have relied on administrative data from the MBS.^4,5,12,13 The item codes used by Medicare to reimburse claims for excision of histologically confirmed KC are not specific to cancer type (i.e. BCC or SCC), however. Moreover, the MBS data do not cover services provided in public hospitals. Hospital inpatient morbidity data have been used in analyses of the economic burden of KCs, but as for the MBS database, published data on hospital admissions have historically been episode-based rather than person-based, and thus are unsuitable for estimating incidence. Individual-level data are now accessible through linked health datasets curated by the Australian Institute of Health and Welfare (i.e. the National Health Data Hub) for approved researchers, and they may provide a new resource for investigating the burden of skin cancer.

Benefits of the keratinocyte cancer incidence data collected by the Tasmanian Cancer Registry

Population-based data on BCC and SCC incidence gathered by the registry have enabled quantification and description of temporal trends over four decades (1978–2018)¹⁴ and trends within important subgroups of the population, considering socioeconomic status and indicators of geographical remoteness.¹⁵ A concerning trend of increasing incidence of both BCC and SCC was identified.¹⁴ Rural residence was associated with a high risk of SCC, particularly among younger Tasmanians, a finding that may reflect higher cumulative sun exposure (and occupational exposure) and possibly lower access to health services in those regions.¹⁵ The data have also enabled estimation of the absolute and relative risk of being diagnosed with a subsequent KC following a first diagnosis,¹⁶ identifying a significant burden of multiple occurrences of these cancers. Approximately one in three Tasmanians with a history of KC will develop another KC within 5 years, and half will do so within 10 years.¹⁶ The registry data have made it possible to identify high-risk groups on the basis of age, socioeconomic advantage and location of residence. The data provide crucial information for establishing priorities for skin cancer control programs and for directing prevention activities towards the populations that need them most. There is evidence that tailoring prevention messages to high-risk groups may have the greatest impact on sun-safe behaviours.^17,18

Benefits of incidence data collected by cancer registries in other countries

There are numerous examples of how data on KC incidence have informed public health efforts internationally. Monitoring trends in skin cancer incidence (including SCC and BCC) has been used as a tool by the United Nations Environment Panel to assess the progress of the Montreal Protocol and its amendments in protecting the ozone layer.¹⁹ The World Health Organization (WHO) and the International Labour Organization (ILO) used population-based registry data on BCC and SCC incidence from multiple countries to quantify the global burden of non-melanoma skin cancer attributable to occupational exposure to UV radiation, estimating that 29% of deaths from KC are attributable to occupational exposure.²⁰ In Finland, registry data on BCC incidence revealed a concerning rise in this cancer among women, likely due to an increase in tanning bed use. This prompted the Finnish Radiation and Nuclear Safety Authority to launch national campaigns, directed at both professionals and the general public, promoting sun safety.²¹ Finally, highlighting the evolving landscape of skin cancer, a recent report comparing trends in the incidence of SCC with those of melanoma in seven populations with SCC registration (Finland, Norway, Sweden, Denmark, Scotland, the Netherlands and Tasmania, Australia) revealed faster-rising incidence of SCC when compared with melanoma, particularly among women.²²

The example of melanoma

National cancer registry data have been a powerful tool in combating melanoma. The data have informed research priorities, facilitated treatment development and helped improve clinical practices. Although much less common than KCs, melanoma has historically carried a greater threat to life with a significantly higher mortality rate. Because all melanoma diagnoses are reported to cancer registries across the country, trends have been monitored for the entire population and specific subgroups for over four decades. The data have also enabled quantification of the economic burden to the government and patients.²³ The trend of increasing incidence in Australia and worldwide informed the development of primary prevention campaigns and led to the discovery that early childhood was the critical period for sun exposure to cause melanoma.²⁴ It also led to increased investment in research, facilitating clinical trials of new innovative treatments for advanced melanoma (immunotherapy, targeted therapy and combination therapies) that have had a profound impact on survival rates.²⁵

Potential barriers to national reporting of keratinocyte cancer incidence

Structured pathology reporting is an important requirement for the successful implementation of automated registration of the KCs. Because they are so numerous, the labour resources required to manually review all pathology reports for BCC and SCC are likely to be prohibitive. There is ongoing development towards standardised reporting of cancer to the cancer registries across Australia. The Royal College of Pathologists (RCPA) recently published the first edition of a structured pathology reporting protocol (SPRP) for non-melanoma skin cancers (NMSC, which includes BCC, SCC and other rare types of skin cancer) of the head and neck.^26,27 The protocol is restricted to excision specimens of the head and neck, and it excludes low-risk cancers such as superficial BCC. This decision was based on concerns from pathologists regarding the variety of NMSC specimen types (from curettes to radical resections) and significant differences in the biological behaviour of different tumours, making it difficult to use a single protocol. A further concern was inadequate compensation from the Australian MBS for the use of an SPRP for low-risk NMSC. The RCPA has identified significant challenges associated with the digital implementation of structured reporting currently within Australian anatomical pathology laboratories.²⁸ Advances in artificial intelligence may help to overcome the barriers to digital implementation. For systematic national reporting of KCs to succeed, we advocate for the use of SPRP for all BCCs and SCCs removed via excision, with appropriate recompense for reporting.

The cancer registries will require extra personnel and other additional resources to clean and check automated registrations from pathology laboratories. A phased approach may be more feasible, perhaps following the example of the Queensland Cancer Register by first registering KCs that have defined characteristics associated with a poor prognosis. This approach would require significantly fewer resources, but it would result in the establishment of systems and procedures that might later be expanded to include the registration of lower-risk lesions. The costs associated with collecting, storing and reporting the additional data on BCC and SCC should be carefully evaluated.

Summary and recommendations

Accurate data on skin cancer incidence will allow the Australian Government to monitor trends (particularly for high-risk subgroups of the population) and develop targeted prevention strategies. It will also allow evidence-based decision-making with regard to resource allocation and policy development, and provide a baseline against which progress can be monitored. We encourage all state and territory cancer registries to explore the feasibility of implementing statutory notification and routine reporting of KCs, considering the model currently being established in Queensland. Standardised pathology reporting, coupled with artificial intelligence, presents a significant opportunity to automate KC registration, reducing labour-intensive tasks, increasing efficiency and reducing the time and resources required, thus increasing the likelihood of a favourable balance between resource investment and public health benefits. Successful implementation will require a collaborative approach with clinicians, public health officials, data scientists and policymakers working together to improve skin cancer outcomes for all Australians.