While highly sensitive technology is important for detecting melanoma, methods with poor sensitivity can lead to unnecessary biopsies, treatments, cost, as well as morbidity. Several technologies are available to improve sensitivity and specificity, some of which are used as second-level tests.
A number of new imaging applications for identifying melanoma have become available to dermatologists in the last 15 years, which can improve the sensitivity and specificity of total body skin examinations when used in the appropriate clinical setting, according to Michael Marchetti, M.D., assistant attending physician at Memorial Sloan Kettering Cancer Center in New York.
“Screening efforts over the last 20 to 30 years have increased the number of melanomas we’re diagnosing at earlier stages, but there still remains a significant number of people diagnosed at later stages, as well as people who die from melanomas caught at relatively early stages,” says Dr. Marchetti, who spoke to colleagues about imaging technologies for pigmented lesions during the summer meeting of the American Academy of Dermatology (New York, 2015).
Dr. Marchetti notes that up to 85% of melanomas are not initially found by dermatologists.
“In general, high sensitivity of an imaging technology is important, but if used as a screening device specificity may even be more important,” he says. “The objectives of my talk included also pointing out the potential risks of melanoma overdiagnosis associated with technology use, which are well-recognized in other medical conditions such as ultrasound with thyroid cancer and CT angiography with pulmonary embolism, and may lead to unnecessary treatment, cost, and morbidity” he adds.
ALSO READ: Survival with melanoma subtypes varies
To improve the sensitivity and specificity of skin examinations, dermatologists have access to a wide variety of technologies, some old, and some emerging, including total body photography, sequential dermoscopic imaging, and confocal laser microscopy, among others.
NEXT: Guidelines in a changing world
Imaging technologies can be used either in the doctor’s office or by the patient at home. “One of the challenges is the low prevalence of disease. Some estimate that there are 35,000 nevi per melanoma diagnosed, or 200 atypical nevi per melanoma diagnosed,” Dr. Marchetti says. “This is not adding in solar lentigines, seborrheic keratosis, and other benign lesions that can mimic melanoma. We need something to have not only sensitivity but also extremely high specificity, and unfortunately, very few technologies have emerged that have such high specificity,” he adds.
RECOMMENDED: Skin clues help identify cancer metastases
One application of imaging technologies is to use them as a second-level evaluation procedure, he says. After physicians identify a suspicious lesion, they can use something like sequential dermoscopic imaging before deciding to perform a biopsy. In patients at high risk for disease, dermatologists can be more aggressive with imaging.
For example, he notes, the British Journal of Dermatology recently published a systematic review of clinical practice guidelines for melanoma screening.
“They gave a high level of evidence recommendation for dermoscopy and sequential dermoscopic imaging, and a lower level of evidence recommendation for total body photography,” Dr. Marchetti explains. “They also gave a consensus recommendation for a protocol of six-monthly surveillance with total body photography and sequential dermoscopic imaging for patients at very high risk of melanoma,” he says.
YOU MIGHT LIKE: Viruses may lead to skin disease, cancer
“My suggested approach today looks like this: The baseline exam should include clinical and dermoscopy examination of all lesions, and anything that’s frankly suspicious for melanoma should be biopsied. Anything that’s an outlier or something less than that can undergo sequential dermoscopic imaging or confocal microscopy. In terms of follow-up, total body photography in experience is very useful for detecting new or changing lesions, followed by dermoscopy to evaluate those lesions. Total body photography can be particularly useful in people who have particularly complex nevus phenotypes, in patients where every nevus is kind of scary because anything that is not changing over time is very unlikely to be harmful, despite its appearance,” according to Marchetti.
Each technology comes with its own limitations, he suggests.
“People have argued that total body photography, which in its initial application used two-dimensional photographs that catalogue the entire skin surface, doesn’t accurately represent raised nevi or lesions on curved surfaces and leads to images that are partially overlapping where things look different in each photograph,” Dr. Marchetti says. As a result, three-dimensional total body photography has emerged as a solution.
“Three-dimensional is what we’re using at my institution, in which 46 cameras fire simultaneously, and it builds a three-dimensional image that can be rotated along any axis,” he explains. “In addition, multiple groups are trying to develop automatic change detection of total body photography images.”
NEXT: Improving specificity
The second imaging technology Dr. Marchetti describes in sequential dermoscopic digital imaging (SDDI), in which dermoscopic images of lesions are taken and repeated over time for purposes of comparison.
“The key thing is that the image has to be put on the monitor. You have to do a side-by-side comparison, because the changes you’re looking for are extremely subtle. You can’t just look at the patient with a dermatoscope and then look at the screen,” he explains.
Dr. Marchetti describes a study published by a group of researchers in Belgium, which pointed out the beneficial use of SDDI.
“The dermatologists were split into three groups: dermoscopy novices, a second group with some experience at dermoscopy, and a third group that had experience and access to SDDI. In the group with access to sequential imaging, the number of lesions needed to biopsy was about 2.4 for every melanoma, which is pretty incredible. For the dermoscopy novices, the number needed to biopsy was about ten or 11 to one,” he notes.
“This technology can really be used to really monitor lesions over time. You select maybe the 20 lesions that are most atypical, and take dermoscopy images of all of those lesions,” Dr. Marchetti says. “Some groups repeat all 20 of those images every time a patient comes in. This technology can improve the sensitivity of the exam. It identifies featureless melanomas and slow growing melanomas.” Imaging may be repeated every three months, for example.
NEXT: Assessing histological detail
When doctors analyze the clinical and dermoscopic features of some melanomas they find that they are virtually indistinguishable from other nevi, he notes.
“There do remain a subset of melanomas that are particularly difficult to diagnose, even with these imaging technologies,” Dr. Marchetti says. Therefore, dermatologists should be aware of all of the techniques available.
Another technology he describes is in vivo reflective confocal microscopy.
“This comes in two forms: one is handheld, while the other is a little bit more cumbersome but allows you to image a wider horizontal view of the skin surface at one time. This technology produces images that approach histologic detail, and you really start to see things at the cellular level,” he explains. “However, melanin is used as a natural contrast agent, so it may be more challenging to use in those with a lighter phenotype,” he cautions. “Also, it’s limited to a depth of about 200 microns.”
“The application of confocal microcopy can be used to improve diagnostic accuracy, mostly by reducing unnecessary procedures. In one particular study, there was a 50% reduction in benign lesions biopsied. It can also be used after SDDI, almost like a third level test,” Dr. Marchetti suggests.
Disclosure: Dr. Marchetti reports no relevant disclosures.
 Tromme I, Sacré L, Hammouch F, et al. Availability of digital dermoscopy in daily practice dramatically reduces the number of excised melanocytic lesions: results from an observational study. Br J Dermatol. 2012;167(4):778-86.