Risk-based triage could help target skin cancer screening

A significant proportion of patients who seek routine skin cancer screenings without any specific skin complaints are at low near-term risk of being diagnosed with the disease, according to a cross-sectional study published in JAMA Dermatology. The findings suggest simple triage criteria could play a role in directing surveillance toward higher-risk groups.

Researchers Yin Li, PhD, and Robert A. Swerlick, MD, of Emory University, analyzed pre-visit survey data from new patients seen at the Emory Healthcare Dermatology Clinic between March 2021 and October 2022. Of 4,875 new patients who completed the survey, 1,074 (22%) reported no specific skin concerns and said the purpose of their visit was to have their skin checked for cancer.

The cohort skewed younger and lower risk than might be expected for a cancer surveillance population. Patients had a mean age of 50.3 years, and just over half (50.4%) were 50 or younger. Fewer than 10% of those 50 or younger reported a personal history of skin cancer, compared with 42.1% of patients older than 70.

Among the full cohort, 131 patients (12.2%) underwent 146 biopsies, which yielded 38 skin cancers. The distribution of those diagnoses underscored the influence of age and skin type: 92.1% of the cancers were detected in patients older than 50, and 37 of 38 occurred in patients with skin phototypes I through III. Only three patients younger than 50 were diagnosed with skin cancer. The authors quantified screening efficiency using the number needed to examine (NNE) to detect one skin cancer. For patients 50 or younger, the NNE was 181; for those older than 70, it dropped to 7.

Personal history was also an impactful factor. Patients with prior skin cancer had an NNE of 12, compared with 52 for those without. Patients with skin phototypes I through III and those with blue, green, or hazel eyes had lower NNEs than patients with darker skin or brown eyes.

The findings arrive amid continued uncertainty about the value of broad skin cancer screening. The authors noted that the U.S. Preventive Services Task Force concluded in its 2023 update that there is inadequate evidence that visual skin examination by a clinician reduces morbidity and mortality, and that no U.S. professional organization currently recommends visual screening for skin cancer. Even so, the authors cited survey data showing that nearly 40% of dermatologists recommend routine checks regardless of a patient's risk factors.

“The earliest review done by the USPSTF more than 20 years ago encouraged the development of risk-based approaches to identify patients at high risk for SC and highlighted the disease burden specifically in older adults,” the authors explained. “This has been advocated for repeatedly, and multiple risk models have been developed primarily as research tools. However, standardized and practical risk-based tools have not been broadly deployed in general dermatology practice, although use of age and phototype has been advocated for risk assessment.”

Li and Swerlick noted that age, skin phototype, and personal history of skin cancer are easily captured data elements that could form the basis of a practical, risk-stratified approach to surveillance. They pointed to age as a logical starting point, drawing a parallel to the age-based frameworks that have historically anchored other cancer screening programs.

“The first steps in adopting population-based screening almost invariably are to follow uniform age-based approaches, making standard recommendations for individuals within age ranges,” the authors wrote. “Given the clear relationship between [skin cancer] development and age, adopting an age approach to [skin cancer] screening is a logical first step in deploying a risk-based strategy for [skin cancer] surveillance, especially in asymptomatic patients without skin complaints.”

The authors highlighted several limitations of the study. It was conducted at a single academic practice, which may limit generalizability, and the 52.6% survey response rate could introduce selection bias. The short data-collection window also prevented analysis of how often individual patients returned for checks or of longer-term outcomes such as skin cancer deaths. Still, the findings suggest there would be value in a triage system to determine which asymptomatic patients would benefit from surveillance.

“The implementation of triage criteria for asymptomatic patients seeking [skin cancer] surveillance, based on risk factors such as age, skin phototype, and [skin cancer] history, may be beneficial for selecting individuals who are at higher risk for [skin cancer] diagnosis and therefore most likely to benefit from routine surveillance,” the authors concluded.