Unaided visual inspection is not enough to detect and diagnose all cutaneous melanomas. Numerous instruments are being utilized or investigated to increase diagnostic sensitivity and specificity and improve in vivo differentiation of melanoma from its simulators.
Detecting Suspect Lesions
During screening examinations, some melanomas may remain unnoticed by the clinician. Some simply lack clinical features suggestive of melanoma. Performing total-body skin examination and comparing the findings to baseline photographs is the only technique that can aid clinicians in identifying new or changed lesions. (1) These suspect lesions can then be evaluated further with the new technologies to determine whether a biopsy is warranted. Thus, clinicians can identify subtly changing or new melanomas even if they lack classic "ABCD" features.
Evaluating Suspect Lesions
Magnifying Lens and Wood's Light
A simple magnifying glass can assist in evaluation of pigmented lesions. For example, it allows visualization of comedo-like openings in seborrheic keratoses and telangiectasias in pigmented basal cell carcinomas (BCCs), helping to exclude melanoma from the differential diagnosis. (2) The wavelength emission from a Wood's lamp is absorbed by melanin and may assist in diagnosis of lentigo maligna melanoma (LMM) by revealing irregular pigment distribution. It can also help define LMMs' subtle clinical borders. (2)
Dermoscopy
Dermoscopy uses a hand-held instrument called a dermoscope, equipped with a transilluminating light and standard magnifying optics. It can be performed using a fluid interface or polarized light. Both reduce light reflection, refraction, and diffraction, essentially making the epidermis translucent and allowing in vivo visualization of subsurface epidermal and papillary dermal structures. These structures have generated new terminology and clinical criteria for assessing pigmented lesions. Properly interpreted, these criteria can increase diagnostic accuracy for melanoma 10% to 20%. (2) [For more on dermoscopy and melanoma, see Volume 22, No. 3, 2003.]
Image Analysis and Computer-Assisted Diagnosis
Numerous computer programs can objectively document the clinical and dermoscopic features of digitized pigmented lesion images. (2) Most rely on sophisticated programs for lesion segmentation, determining the boundary separating the lesion from "normal" skin. Once the computer identifies the lesion, the program extracts features programmed in as characteristic of melanoma, eg, colors, texture, asymmetry, border irregularity, diameter, linear dimensions, perimeter, length, and area. Most systems analyze the images, providing "computer-assisted diagnosis" approaching or exceeding the sensitivity and specificity achieved by expert dermoscopists.
Multi-Spectral Imaging and Automated Diagnosis
Knowing that light of different wavelengths penetrates skin to different depths led investigators to evaluate pigmented lesions under wavelengths from infrared to near ultraviolet. Two systems--SIAscope[TM] (Spectrophotometric Intracutaneous Analysis system) and MelaFind[TM]--utilize these "multi-spectral" sequences of dermoscopic images for computer analysis. Melafind[TM] provides diagnosis in a completely automated system, while Siascope[TM] requires physician interpretation.
SIAscope[TM] performs in vivo examination of a 12 mm diameter area, capturing images at 4 different, narrow-spectrum wavelengths from 400 to 1000 nm peaks. These images provide information on concentration, distribution, and position of skin chromophores--collagen, melanin, and hemoglobin. This information is then displayed in different SIAgraphs. The combined features of dermal melanin, collagen "holes," and "erythematous blush" with blood displacement have resulted in 80% specificity and 83% sensitivity for melanoma. (2)
MelaFind[TM] also involves multi-spectral illumination of clinical and dermoscopic images, using dedicated software for automatic differentiation between melanoma and benign pigmented lesions. It employs 10 narrow-spectrum wavelengths, from near infrared through visible light, to obtain information on absorption and scattering at different lesion depths. Within 3 seconds, this provides information unavailable to the naked eye about lesion border, size, and morphology. The aim is to produce a fully automated, non-operator dependent, objective measurement instrument for melanoma diagnosis, and to differentiate in situ from invasive melanoma as well as determine Breslow thickness. Studies have achieved over 95% sensitivity and 68% to 85%. (2)
Confocal Scanning Laser Microscopy
Confocal scanning laser microscopy (CSLM) allows in vivo examination of the epidermis and papillary dermis at a resolution approaching histologic detail. A low-power visible or near-infrared laser is focused on a specific point, detecting the light reflected from that point through a pinhole size filter. This beam is scanned horizontally over a two-dimensional grid to obtain a horizontal microscopic section. Focal length can be adjusted, allowing the microscope to image a series of horizontal planes stacked vertically, with an axial thickness of 2 to 5 [micro]m. This in vivo axial section thickness correlates closely with that of excised histological sections. The imaging depth in normal skin is limited to 200 to 300 [micro]m.
Though the lateral resolving power of CSLM mainly permits imaging of cellular structures, sometimes subcellular structures such as melanosomes and nucleoli are visualized. Pigmented lesions can be viewed with good resolution because free cytoplasmic melanin pigment and cytoplasmic pigmented and nonpigmented Melanosomes provide strong contrast for the infrared laser. Due to the presence of Melanosomes, the melanocytic cytoplasm in pigmented and amelanotic melanomas appears bright on CSLM, allowing easy detection. (2)
CSLM can assess cellular components of intact skin lesions noninvasively. A lesion can be examined anytime to determine if it has melanoma features, and biopsy-proven melanomas can be examined to determine lateral margins. Many lesions can be examined in the same visit, which helps in evaluating patients with multiple suspicious pigmented lesions (atypical mole syndrome).
Ultrasound
High-frequency ultrasound is being increasingly applied to clinical dermatology. High-frequency sound impulses are transmitted into the skin, then reflected, refracted, or inflected to varying degrees based on the acoustic impedance of the tissue interface they encounter.
Dermatologic ultrasonography uses high-frequency scanners of 20 to 50 MHz for sharp focusing in superficial focal regions such as the epidermis and upper dermis, allowing evaluation of melanocytic lesions. The newer 50 to 100 MHz scanners have an axial resolution approaching 10 [micro]m, compared to 80 [micro]m with the 20 MHz scanners, and their lateral resolution is <30 [micro]m compared to 200 [micro]m for the lower frequency scanners. They are more suitable for epidermal study, allowing imaging of nests of nevus cells or tumor vessels. Melanomas generally appear as solid homogenous hypoechoic lesions.
Although ultrasound cannot diagnose melanoma reliably, it may help determine in vivo maximum melanoma thickness, volume, and/or vascularity, offering guidance in operative planning. (2)
Optical Coherence Tomography
Optical Coherence Tomography (OCT) parallels ultrasound imaging, using light rather than sound. A pulse of near-infrared, low-coherence light is split, with half the beam sent to the specimen and half to a scanning reference mirror. The light to the specimen is focused on superficial skin layers and backscattered, then recombines with the beam reflected from the mirror system. Measuring the interference pattern can determine where in the tissue the light was reflected. This reflectiveness of different tissue components, such as cell membranes and melanin, provides the contrast in the images.
A two-dimensional cross-sectional image is built up by lateral scanning across the tissue. The axial resolution depends on the light's coherence length, reportedly about 15 [micro]m, but has attained 2 to 4 [micro]m resolution with technologies using ultrashort pulse lasers. The resolution does not routinely reach the capabilities of routine microscopy or CSLM. However, the latest systems may be able to resolve cellular detail.
Many skin diseases have been studied using OCT, showing distinctive architectural changes compared to normal skin. Like other tumors, melanocytic skin tumors show more homogenous signal distribution, as well as increased light scattering. However, current resolution limits do not allow differential diagnosis between benign and malignant lesions. Depth of invasion can be better measured with OCT than CSLM; however, OCT is limited to thin tumors up to 1 mm. (2)
Magnetic Resonance Imaging
Magnetic resonance imaging (MRI) has also been used experimentally for examining pigmented skin lesions, with specialized surface coils allowing higher resolution. However, the technology has no specific clinical dermatologic applications yet. Experimentally, criteria suggesting malignancy in skin tumors, such as nonhomogeneity and surrounding soft tissue edema, have been reported but are not reliable. To date, MRI cannot dependably discriminate benign from malignant melanocytic lesions. However, it may provide sufficient resolution for information on depth and extent of underlying tissue involvement, which may assist in preoperative assessment. (2)
Conclusions
New techniques are needed to help diagnose early melanoma. Photography, magnifying lens, Wood's light, dermoscopy, CSLM, ultrasound, MRI, OCT, and multi-spectral imaging may all prove useful. However, pigmented lesions must be evaluated in the context of an entire clinical skin examination. It would be impractical to evaluate every single pigmented lesion with these new techniques. Only lesions considered different or suspicious should be subjected to in-depth evaluation.
Although these new technologies may improve differentiation of benign from malignant lesions, nothing substitutes for an experienced clinician's knowledge and intuition in deciding which of potentially hundreds of lesions must be investigated further. Patient history and physical examination remain the cornerstone, and the clinician must maintain a high index of suspicion for melanoma in deciding which lesions require in-depth evaluation. Periodic screenings of high risk patients with comparisons to baseline total-cutaneous photographs can help detect the most probable melanomas. These "suspect" lesions can be further evaluated with the new in vivo instruments. However, the final decision on whether or not to biopsy always rests with the clinician.
(From The Melanoma Letter Vol. 22, No. 2, 2004)
References
1. Feit NE, Dusza SW, Marghoob AA. Melanomas detected with the aid of total cutaneous photography. Br J Dermatol. 2004;150(4):706-14.
2. Marghoob AA, Swindle LD, Moricz CZ, Sanchez Negron FA, Slue B, Halpern AC, Kopf AW. Instruments and new technologies for the in vivo diagnosis of melanoma. J Am Acad D ermatol. 2003;49(5):777-97.
Ashfaq A. Marghoob, MD
Clinical Assistant Professor
Department of Dermatology
Memorial Sloan-Kettering Cancer Center
New York, NY
The Skin Cancer Foundation
245 5th Ave., Suite #1403, New York, NY 10016
To obtain copies of this article and other educational materials on skin cancer prevention, detection, and treatment, contact The Skin Cancer Foundation by phone at 212-725-5176 or by e-mail at info@skincancer.org.
You can also find information on all forms of skin cancer and learn about Membership benefits on the Foundation's comprehensive website at www.skincancer.org.
Since its founding in 1979, The Skin Cancer Foundation has set the standard for educating the public and professionals about skin cancer, its prevention by means of sun protection, and the need for early detection and prompt effective treatment. It is the only national and international organization devoted solely to combating the world's most common cancer, now occurring at epidemic levels. More than a million new cases of skin cancer are diagnosed in the United States every year. The Foundation's mission is to decrease the incidence, illness, disfigurement, and mortality of the disease by means of public and professional education, medical training, and research.
Outstanding articles from the Foundation's award-winning publications, The Skin Cancer Foundation Journal, The Melanoma Letter, and Sun & Skin News appear regularly.
The Skin Cancer Foundation Journal, published annually, is a centerpiece of the public and professional education program. It features articles by leading physicians on diagnostic and treatment advances, research on the cutting edge, and new trends in sun protection. Medical Editor: Pearon G. Lang, Jr., MD, Editorial Director: Joyce Weisbach-Ayoub, Editor: Lucy Kavaler
Sun & Skin News, a quarterly newsletter, offers the latest approaches to sun protection, skin cancer prevention, and skin care. Medical Editor: Arnold W. Klein, MD, Managing Editor: Joyce Weisbach-Ayoub, Publications Editor: Mark Teich
The Melanoma Letter, published quarterly, is the only clinical bulletin exclusively devoted to keeping physicians up-to-date on developments in diagnosis, treatment, and scientific research on this life-threatening form of skin cancer. Editor-in-Chief: Allan C. Halpern, MD, Associate Editor: Ashfaq A. Marghoob, MD, Consulting Editor: Alfred W. Kopf, MD, Managing Editor: Joyce Weisbach-Ayoub, Editor: Mark Teich
COPYRIGHT 2005 Journal of Drugs in Dermatology, Inc.
COPYRIGHT 2005 Gale Group