Balance between benefits, harms key to limiting overdiagnosis in melanoma screening
NEW YORK — Although overdiagnosis is common in cancer screening, it is not excessive if the benefits outweigh the harms, according to a presenter at the HemOnc Today Melanoma and Cutaneous Malignancies meeting.
That is typically the case with regard to melanoma, because the harms associated with overdiagnosis are relatively minor compared with other malignancies, Martin A. Weinstock, MD, PhD, professor of dermatology and community health at Brown University in Providence, Rhode Island, said during a presentation.
Martin A. Weinstock
“Overdiagnosis is very common and it is a consequence of the system of screening for cancer, but it often can be a misunderstood term,” Weinstock said. “It is not a mistake and it doesn’t mean there was malpractice. It can be defined as a diagnosis that would not have caused morbidity or death to a patient had the diagnosis not been made.”
Overdiagnosis can lead to increased costs, patient distress, morbidity and mortality due to unnecessary procedures. Consequently, the risk–to–benefit ratio of screening for several cancer types — such as prostate cancer, breast cancer and lung cancer — has been the subject of extensive debate.
SEER data show melanoma incidence has risen steadily since 1975, whereas mortality has remained relatively stable, and the trends create “a classic case that makes you suspect overdiagnosis,” Weinstock said.
No randomized trials have been conducted to assess the benefits of melanoma screening, but several times series provide valuable data that support the benefits of screening for reduction of melanoma incidence.
Schneider and colleagues assessed the value of an educational campaign that promoted self-examination and offered targeted screening to employees at the Lawrence Livermore National Laboratory in California.
The study spanned 27 years, and researchers divided their investigation into three parts: a pre-awareness period from 1969 to 1975, during which employees were not aware of melanoma risks; an early awareness period from 1976 to 1984, during which the educational campaign helped increase participants’ knowledge of melanoma risks; and the targeted screening period from 1984 to 1996.
The results, published in 2008 in Journal of the American Academy of Dermatology, showed incidence of melanoma in situ increased during the study period; however, incidence of thicker melanomas decreased as melanoma awareness increased.
Crude incidence of melanomas 0.75 mm or thicker declined steadily, from 22.1 cases per 100,000 person-years in the pre-awareness period, to 15.13 cases in the early awareness period, to 4.62 cases in the screening period (P = .001).
Researchers determined 3.4 melanoma deaths were expected during the screening period; however, none were observed (P = 0.34).
The SCREEN project, an observational study that evaluated the effect of population-wide skin cancer screening in the German state of Schleswig-Holstein, also demonstrated the benefits of screening.
More than 360,000 participants aged 20 years or older underwent screening via whole-body examination between July 2003 and June 2004.
Researchers compared melanoma incidence and mortality among screened individuals with that reported in individuals from adjacent regions in which population-based skin cancer screening was not conducted.
Results showed the screening program led to greater increases in detection of invasive melanoma among women (53% vs. 18%) and men (26% vs. 10%) in Schleswig-Holstein compared with the regions where screening was not offered.
A study by Aitken and colleagues — conducted in Queensland, Australia —showed whole-body clinical skin examinations were associated with increased diagnosis of melanomas less than 0.75 mm (OR = 1.38; 95% CI, 1.22-1.56). However, the researchers reported increased detection of melanomas from 0.75 mm to 1.5 mm (OR = 0.93; 95% CI, 0.73-1.19), 1.5 mm to 3 mm (OR = 0.83; 95% CI, 0.66-1.05), and larger than 3 mm (OR = 0.6; 95% CI, 0.43-0.83).
The examinations also were associated with a 23% reduction in deaths within 10 years.
A randomized trial Weinstock and colleagues conducted to assess whether a multicomponent intervention increased monthly thorough skin self-examination (TSSE) illustrated the potential for overdiagnosis.
The analysis included 1,356 patients who attended routine primary care visits in southeastern New England. Patients assigned to an intervention arm received instructional materials, counseling, a follow-up phone call and tailored feedback letters.
At baseline, performance of TSSE was comparable between the intervention group and a control group. However, researchers observed a significant increase in performance of TSSE in the intervention group at 2 months, 6 months and 12 months post intervention (P < .0001 for all).
Only one melanoma was diagnosed during the study, but participants in the intervention group underwent twice as many surgeries on their skin within 6 months of randomization.
“These surgeries were not done for melanoma. They were done for all sorts of other reasons,” Weinstock said. “There seems to be an effect of getting people to look at their skin could potentially be related to this overdiagnosis issue.”
The USPSTF is re-evaluating its recommendations on melanoma screening, and its updated recommendation is due in about a year.
“The key questions are, how much overdiagnosis is too much, and how do you do these interventions so you get reductions in death and minimize overdiagnosis,” Weinstock said. “We have to approach it by looking at each individual case and determining whether the benefits outweigh the harms. I think that’s the approach the USPSTF will take. I believe the harms are relatively minor for melanoma compared with other cancers, but we will see to what extent the task force changes its recommendation based on all of this evidence.” – by Mark Leiser
Weinstock MA. Overdiagnosis of melanoma from screening: How much is too much? Presented at: HemOnc Today Melanoma and Cutaneous Malignancies; April 10-11, 2015; New York.
Disclosure: Weinstock reports consultant fees from Castle Biosciences and MELA Sciences Inc.