Australia is deciding whether to screen for melanoma by genotype. It is not ready yet

Australia does not have a national melanoma screening programme. What it has instead is opportunistic screening: skin checks that happen when a patient asks for one, or when a GP notices something during a visit for an unrelated reason. Two papers published early this year describe what that system currently produces, and what it would take to replace it with something targeted.

The proposal: screen by risk, not by population

The idea under evaluation is targeted screening of high-risk individuals rather than screening everyone, on the argument that it would use healthcare resources more efficiently [s1]. The stratification tool would be an integrated risk score — a combination of a polygenic risk score with non-genetic risk factors, which offers the best performance for melanoma risk stratification among available approaches [s1].

A narrative review published in the Medical Journal of Australia on 2 February set out to establish whether that is implementable at population scale [s1]. Its conclusion is a qualified no: integrated risk scores have the potential to facilitate targeted high-risk melanoma screening in Australia, but significant evidence and infrastructure gaps must be addressed before a programme could be implemented [s1].

The evidence gaps

The review identifies five research needs before implementation [s1]:

Understanding the progression rate of melanoma in situ to invasive disease. This is the overdiagnosis question in its most concrete form — a screening programme that finds more in situ lesions is only beneficial to the degree those lesions would have progressed.

Defining who should be offered integrated risk scores in the first place.

Addressing performance issues across ancestries. Polygenic risk scores are derived from the populations they were trained on, and Australia's population is not the population most melanoma genetics research was conducted in.

Developing clearly defined risk thresholds and the clinical advice that should accompany each.

Investigating the clinical utility and impact of a person receiving an integrated risk score — that is, whether being told a number changes anything for the better.

The infrastructure gaps

The second list is not about evidence at all, and it is arguably the harder one [s1]. A programme would require equitable access to post-screening care; guidelines and quality standards for generating polygenic and integrated risk scores; healthcare rebates for polygenic testing; computational and data storage infrastructure; workforce training and clinical decision support resources; clearer protections around the use of polygenic scores in risk-rated insurance; and clear plans for programme quality and performance management [s1].

The insurance item is worth pausing on. A national programme that generates genetic risk stratification for millions of people creates data that risk-rated insurers have an obvious interest in, and the review lists protection against that use as a precondition rather than an afterthought [s1]. Equitable access to post-screening care is the other structural precondition: a risk score that identifies someone as high-risk is of no value if the follow-up dermatology appointment is unavailable or unaffordable.

What the current system actually delivers

Against that, a study published in Cancer Epidemiology in March measured what opportunistic screening produces in practice [s2].

It drew on the 45 and Up Study, which recruited 267,357 New South Wales residents between 2005 and 2009, and analysed self-reported clinical skin checks in the previous 12 months among the 43,799 participants who responded to a 2020 follow-up survey — a 52.8% response rate [s2]. Participants' mean age was 70.3 years (SD 8.3, range 56–103), and 55.9% were female [s2].

In the previous 12 months, 43.2% had a whole-body skin check, 21.0% had a partial check (part-body or a specific mole or spot), and 35.8% had no skin check at all [s2]. Just over one in five — 21.8% — reported more than one check during the year [s2].

Who gets checked

The strongest predictors of a whole-body check were the ones a clinician would want: a personal history of melanoma (adjusted OR 3.74, 95% CI 3.39–4.13), a personal history of non-melanoma skin cancer (aOR 4.05, 95% CI 3.83–4.29), having many moles (aOR 3.26, 95% CI 2.79–3.80 versus no moles), and very fair, fair or olive skin (aOR ≥ 2.6 versus black or brown skin) [s2].

But the list of other significant associations is where a targeting problem appears. Whole-body checks were also associated with being male, age 70–79, birth in Australia or New Zealand, university education, private health insurance, being retired, household income above $90,000, family history of melanoma, inability to tan, longer time spent outdoors, participation in other cancer screening programmes, and frequent sunscreen use [s2].

University education, private health insurance and household income above $90,000 are not melanoma risk factors. Their presence alongside genuine clinical predictors is the signature of a system in which access, not risk, partly determines who gets screened. That is precisely the inefficiency the targeted-screening proposal is meant to correct — and also the reason the readiness review lists equitable access to post-screening care as a precondition [s1].

What these two papers do not establish

Neither study is a trial. The readiness review is a narrative review identifying gaps, not an evaluation of whether integrated risk score screening improves outcomes [s1] — no such evaluation is reported. The skin check study is cross-sectional and self-reported, in a cohort with a mean age above 70 drawn from one state, with roughly half the invited follow-up sample responding [s2]. It describes older New South Wales residents, not Australians generally, and it cannot say whether those checks found anything.

Neither paper recommends a screening interval, and nothing here is guidance about whether any individual should seek a skin check.

What to watch

The decision point is whether Australian health authorities commission the specific evidence the review names — particularly progression rates from in situ to invasive disease, and cross-ancestry performance — or whether integrated risk scores enter practice piecemeal through private testing before any of that exists. The review's framing suggests its authors consider the second outcome the likelier risk.

Sources

• [s1] Population-Based Melanoma Screening Using Integrated Risk Scores in Australia: A Narrative Review to Determine Readiness, Medical Journal of Australia, published online 2 February 2026. https://doi.org/10.5694/mja2.70133
• [s2] Prevalence and correlates of clinical skin checks for skin cancer among Australians, Cancer Epidemiology, available online 27 March 2026. https://doi.org/10.1016/j.canep.2026.103060