Diving beyond surface perceptions of melanoma diagnosis and screening

Abstract

During the last few years, there have been significant changes in the diagnosis of melanoma. This new era was ushered by the introduction of dermoscopy, digital dermoscopy and other skin imaging techniques and is continuing to thrive with the addition of innovative artificial intelligence applications. However, a new question has recently arisen: Have these improvements positively affected patient outcomes and survival? Unfortunately, the answer to this question may not be so simple.

The Schleswig-Holstein (Germany) initiative famously yielded ambiguous results. A total of 360,000 people were screened with 90% of melanomas detected being less than 1 mm thick. 5 years following the 12-month screening effort, overall melanoma mortality seemed to decline while in adjacent geographic areas there were no changes. Unfortunately, the initial mortality decline was not sustained; 2 years later, mortality rates returned to the pre-screening level.¹ The U.S. Preventive Services Task Force published an evidence update on available data regarding melanoma prevention and screening.² Screening was not considered harmful for patients, but no direct benefit was demonstrated in regard to survival. Overall, the evidence was inconsistent, but an association of screening with diagnosing thinner lesions could be made.² One of the few studies that compared melanoma trends between screened and unscreened patients, reported that screened patients were more likely to be diagnosed with in situ melanoma (MIS) or thin invasive (≤1 mm) melanoma.³

Overdiagnosis is not a new concept in the epidemiology of cancer. A recent meta-analysis has shown that up to 27% and 17% of breast and ovarian cancer may be overdiagnosed.⁴ In addition, for every life saved due to breast cancer screening, there are 136 false positives, 21 redundant biopsies and 3 overdiagnoses.⁵ Interesting data about lung, liver, breast, ovarian and prostate cancers are provided, but melanoma is nowhere to be found.⁴ This exclusion is not based in discrimination against skin cancer but simply in a lack of robust epidemiologic data. There are no randomized clinical trials regarding melanoma and therefore, at this time, there is no way to quantify the outcomes of skin cancer screening efficiently.

With regard to melanoma, there are no official guidelines on who to screen and how often. In addition, dermatology seems to be a popular medical specialty as a recent study reported dermatologists as the second most often visited specialists in a 24-month interval and the first healthcare providers for skin cancers in most European countries. As a matter of fact, naevi check-up or skin cancer screening was the most common reason for visiting a dermatologist.⁶ The Schleswig-Holstein initiative reported that 620 people had to be screened to detect 1 melanoma. These findings, also corroborated by other studies, point out that screening an entire population is simply not effective.⁷ In addition, such results may be inherently biased as people who seek to be screened tend to be more health-oriented. A recent study that examined mortality in patients with primary MIS demonstrated that patients showed better overall survival compared with the general population indirectly corroborating the previous assumption.⁸ Therefore, it is not unlikely that screening programmes do not actually reach the population needing to be screened the most. The same study showed that patients older than 80 years had higher risk of melanoma-specific mortality compared with patients aged 60–69 years (7.4% vs. 1.4%, respectively).⁸ Similarly, it has been shown that up to 20 excisions may be required to find 1 melanoma in men ≥65 years, but more than 50 excisions to find 1 melanoma in men aged 20–49 years.⁷ A recent consensus statement strongly supported a risk-stratified approach to melanoma screening in clinical settings and public screening events.⁹

Besides identifying high-risk populations, improved and standardized documentation of melanoma cases in Europe could also help identify skin cancer trends and areas for improvement. A recent paper investigated the global incidence and mortality of melanoma and reported that highest incidence was found in Australia, Western Europe and North America, but highest mortality in Eastern Europe.¹⁰ One could argue that this may be due to differences in access to treatments. However, the fact that in many countries only mortality from melanoma is reported, skewing epidemiological data about incidence, should also be taken under consideration.

Several questions still persist. However, the most important of all is whether skin cancer screening can improve patient outcomes without causing harm. A need for conducting high-quality clinical trials able to produce robust data on overdiagnosis of skin cancers, identification of malignant potential of in situ melanomas and appropriate risk stratification of thin melanomas, among others, is highlighted. Creating skin cancer registries with data for all melanomas including treatment approaches and efficacy could help in better understanding the behaviours of these tumours in real life. Additionally, expert panels could help in designing realistic screening programmes targeting high risk population groups as well as defining appropriate screening intervals specifically customized per patient needs. Melanoma screening is important; however, as in every other type of cancer, fine tuning based on scientific data is required to achieve the best possible outcome for patients, national healthcare systems and physicians alike.

Both authors contributed to the conception, gathering of relevant data and authoring the paper; drafting and critically revising the article; and in final approval of the version to be published.

None.

The authors have no conflicts of interest to declare.