PCP–based screening increases uptake among men, improves detection of thin melanoma
NEW YORK — Primary care physician–based screening may increase screening rates for melanoma among men and older individuals, according to a speaker at HemOnc Today Melanoma and Cutaneous Malignancies.
Screening also may improve the detection of thin, in situ and invasive melanomas.
“With [our] large dataset, we hope we can predict the number of people needed to screen by age and sex, and target those more likely to have the thickest melanoma and patients least likely to find their own melanoma,” Laura K. Ferris, MD, PhD, assistant professor of dermatology at the University of Pittsburgh, said during her presentation.
In 2014, University of Pittsburgh Medical Center integrated a large primary-care based screening program designed in collaboration with primary care leaders. It is a nonrandomized voluntary intervention, because a randomized control trial is not feasible for melanoma screening, Ferris said.
Patients were invited for screening when visiting their PCP. After the first year of program implementation, more than 333,735 eligible patients aged 35 years and older were seen, and over 53,000 screenings (16% of the eligible population) were performed. Of those screened, 43% were men, and their median age was 60 years (range, 35-91).
“This was encouraging because men in general are very hard to reach for screening intervention,” Ferris said.
There were 50 melanomas detected in the screened group and 104 in the unscreened group. The median thickness for invasive melanomas were significantly thinner in the screened population (0.37 mm vs. 0.65 mm; P = .0006).
There also were fewer cases of thicker melanomas in the screen group: 10 melanomas thicker than 2 mm were observed in the unscreened group compared with none in the screened group.
Incidence rates of in situ (45.1 vs. 12.5; RR = 3.3; 95% CI, 2-5.6) and invasive (48.9 vs. 24.6; RR = 1.9; 95% CI, 1.2-3) melanomas were significantly higher in the screened group.
Further, a significantly greater proportion of melanomas detected in the screening group were considered thin, defined as having a Breslow thickness of 0.01 mm to 1 mm (43.2 vs. 16; RR = 2.6; 95% CI, 1.6-4.4), and insignificantly fewer had a Breslow thickness greater than 1 mm (5.6 vs. 7.8; RR = 0.7; 95% CI, 0.2-2.2).
Screenings also were performed by dermatologists.
“We found equal distribution of detection by all people, and there were no significant differences in thickness of melanomas found by these groups,” Ferris said.
The results do not suggest a significant increase in health care costs measured by benign biopsies, dermatology visits and skin surgeries, Ferris added.
These study results can help inform efforts to increase screening. However, according to the United States Preventive Services Task Force, there is not enough evidence to determine the balance of benefits vs. harms of visual skin screening by a clinician for cancer in adults.
“We need to try to work toward getting more data to understand risk and benefits of melanoma screening,” Ferris said.
The pros of melanoma screening are that lesions are visible to the naked eye, screening can result in early detection, and earlier melanomas are thinner and can result in a higher cure rate; however, cons include specificity and sensitivity of the test depends on the experience of screener, and there is a risk for overdiagnosis.
“It seems like a no brainer in some ways,” Ferris said.
SEER data show melanoma incidence continues to increase, whereas mortality rates have remained flat. However, research shows the greatest number of deaths occur in patients with thinner melanomas.
“Whereas the mortality rate is higher among thicker melanomas, we see more deaths from thin melanomas,” Ferris said. “This is an opportunity to improve early detection and find these melanomas earlier at a more curable stage.”
There are two strategies to screen for melanoma: a full-body exam performed by a clinician, or patient detection.
Researchers in Belgium evaluated individuals using these strategies to see which is a more suitable approach. A total of 1,668 individuals had a total body exam and 248 underwent lesion-directed screening after they had found something on their skin that looked suspicious.
Skin cancer detection rate among the two groups was similar. However, more individuals who underwent total body exam were diagnosed with melanoma (n = 8) compared with individuals who performed their own first level of screening (n = 1; 0.08; 95% CI, –1.78 to 0.65).
“Just asking patients to find their own melanoma is not going to result in the same level of detection,” Ferris said.
Another study by researchers in France showed that in 2008, one-third of PCPs (32%) were trained in melanoma detection and sent quarterly monographs to update their training. By 2011, a trend of thinner melanoma detection was found after this intervention without significant increase in incidence.
Ferris said “the best piece of evidence to date” is the nationwide screening initiative in Germany, in which residents aged 20 years and older were invited to undergo screening (n = 360,288). Single visual skin cancer screening examinations were performed by dermatologists and nondermatologists, and if patients presented with suspicious lesions and/or risk factors for skin cancer, they were referred to dermatology.
After 5 years of follow-up, melanoma mortality decreased by 50%.
“This seemed to be a good success and, based upon this, Germany started a national screening program for individuals aged 35 and older,” Ferris said. “Among those melanomas detected, they were primarily thin melanomas.” She added that almost 90% of invasive melanomas detected were smaller than 1 mm in thickness.
Ferris noted that the mortality benefit did not persist over time, which may have been attributable to the absence of screening younger patients. – by Melinda Stevens
Ferris LK. Primary care-based skin cancer screening. Presented at: HemOnc Today Melanoma and Cutaneous Malignancies; March 24-25, 2017; New York.
Disclosure: Ferris reports no relevant financial disclosures.