Discriminating Nevi from Melanomas




Reflectance confocal microscopy (RCM) together with dermoscopy enables improved differentiation of melanomas from most nevi. The resulting high sensitivity for detecting melanoma with RCM is complemented by a concomitant increased specificity, which results in the reduction of unnecessary biopsies of nevi. Although RCM can achieve high diagnostic accuracy for early melanoma detection, false-negative and false-positive cases of melanoma are occasionally encountered. This article reviews the essential clues and pitfalls for the diagnosis of melanoma via RCM and highlights the importance of evaluating RCM findings in light of the clinical scenario and dermoscopic features.


Key points







  • •

    The improved ability to differentiate nevi from melanoma via reflectance confocal microscopy (RCM) has the potential to greatly impact the management of patients with multiple atypical nevi, changing nevi, and hypomelanotic or amelanotic lesions.


  • •

    Clinically subtle melanomas usually reveal architectural disarray of the dermoepidermal junction (DEJ) with nonedged papillae and atypical nucleated cells along the basal layer; in addition, the presence of pagetoid cells consisting of large roundish and/or dendritic refractile cells is a prominent feature seen in many melanomas.


  • •

    Most nevi manifest edged papillae with a combination of benign patterns at DEJ (ie, clod pattern, ringed pattern, meshwork pattern). Mild focal architectural disarray may also be seen.


  • •

    False-positive cases of melanoma on RCM are often encountered when evaluating nevi that prove on histology to have a high degree of dysplasia, a spitzoid morphology, or are inflamed.


  • •

    False-negative cases of melanoma often reveal minimal architectural disarray on RCM. Although they tend to lack pagetoid cells, they also do not display any of the benign RCM nevus patterns.






Introduction


Challenges in Early Detection of Melanoma


Despite great advances made in the treatment of late-stage melanoma, the best chance of survival hinges on early detection. However, detection pressure leading to a heightened sensitivity for finding thinner and smaller melanomas is usually coupled to a lowering of specificity, which results in the biopsy of many nevi. Ideally, surveillance of high-risk patients for melanoma via total-body skin examination aided by technology (eg, dermoscopy, reflectance confocal microscopy [RCM]) should aim to maintain a high sensitivity for the detection of melanoma while at the same time prevent the excision of as many nevi as possible. Parameters that can be used to track surveillance efficiency is monitoring ones benign to malignant biopsy ratio (B:M ratio), ratio of thick to thin melanomas, and mean/median melanoma thickness. Total-body photography was one of the first technologies introduced aimed at improving the sensitivity and specificity for melanoma detection. It has been shown that total-body photographs lead to the detection of thinner melanomas and less biopsies of benign nevi with a B:M ratio of 17:1 compared with 45:1 when the examination is performed without photographs. Dermoscopy has also been shown to lead to the diagnosis of thinner melanomas. Dermoscopy and digital monitoring has further improved the B:M ratio to between 4 to 7:1 and RCM has continued to improved this ratio to about 2:1. RCM has demonstrated an improvement in the diagnostic accuracy of physicians for melanoma detection with a mean sensitivity of 93% and specificity of 76%.


Of course, although RCM can greatly impact diagnostic accuracy, it does require learning the features associated with melanoma and nevi. The first section of this article reviews the common clinical scenarios in which RCM can enhance the detection of early melanoma. The second part focuses on the main RCM features used to differentiate nevi from melanoma. Lastly, the RCM pitfalls, including the false-positive nevi and false-negative melanomas, are discussed.




Introduction


Challenges in Early Detection of Melanoma


Despite great advances made in the treatment of late-stage melanoma, the best chance of survival hinges on early detection. However, detection pressure leading to a heightened sensitivity for finding thinner and smaller melanomas is usually coupled to a lowering of specificity, which results in the biopsy of many nevi. Ideally, surveillance of high-risk patients for melanoma via total-body skin examination aided by technology (eg, dermoscopy, reflectance confocal microscopy [RCM]) should aim to maintain a high sensitivity for the detection of melanoma while at the same time prevent the excision of as many nevi as possible. Parameters that can be used to track surveillance efficiency is monitoring ones benign to malignant biopsy ratio (B:M ratio), ratio of thick to thin melanomas, and mean/median melanoma thickness. Total-body photography was one of the first technologies introduced aimed at improving the sensitivity and specificity for melanoma detection. It has been shown that total-body photographs lead to the detection of thinner melanomas and less biopsies of benign nevi with a B:M ratio of 17:1 compared with 45:1 when the examination is performed without photographs. Dermoscopy has also been shown to lead to the diagnosis of thinner melanomas. Dermoscopy and digital monitoring has further improved the B:M ratio to between 4 to 7:1 and RCM has continued to improved this ratio to about 2:1. RCM has demonstrated an improvement in the diagnostic accuracy of physicians for melanoma detection with a mean sensitivity of 93% and specificity of 76%.


Of course, although RCM can greatly impact diagnostic accuracy, it does require learning the features associated with melanoma and nevi. The first section of this article reviews the common clinical scenarios in which RCM can enhance the detection of early melanoma. The second part focuses on the main RCM features used to differentiate nevi from melanoma. Lastly, the RCM pitfalls, including the false-positive nevi and false-negative melanomas, are discussed.




Role of reflectance confocal microscopy in detecting melanoma: clinical applications


Patients with the Atypical Mole Syndrome


Numerous studies have shown that individuals with many nevi and individuals with large acquired nevi (>5 mm in diameter) are at increased risk for developing melanoma. The presence of many nevi displaying increased variability of size, shape, and color allows easy identification of this high-risk population. Although melanoma may develop in association with any nevus, most melanomas develop de novo; thus, prophylactic excision of nevi is an inefficient strategy to prevent melanoma. The methods used to find melanoma within a sea of many nevi relies on finding outlier lesions (the ugly duckling sign) and in identifying lesions that are new or have changed over time. Although change is a highly sensitive criterion for melanoma detection, it lacks specificity. Less than 10% of changing lesions identified during digital total-body photography and digital sequential dermoscopy prove to be melanoma. Several studies have demonstrated that focal dermoscopic structural changes are significantly associated with melanoma, however; it remains extremely difficult to differentiate changing atypical nevi from early melanoma via dermoscopy.


Approximately 8% to 10% of lesions monitored with dermoscopy, total-body photography, and sequential digital dermoscopy end up getting biopsied. If RCM is added as another investigative layer, approximately 70% of the excisions of changing or equivocal nevi could potentially be avoided without decreasing the sensitivity for melanoma detection. Table 1 summarizes the RCM features used to help differentiate melanoma from nevi. The main RCM features associated with melanoma include the presence of roundish pagetoid cells, atypical cells at the basal layer, nonedged papilla, and nucleated atypical cells within the dermis. Based on the aforementioned features, 2 algorithms have been published that are designed to help diagnose melanoma via RCM with high sensitivity and specificity ( Figs. 1 and 2 ). In addition, another algorithm was created ( Fig. 3 ) to assist in characterizing the degree of atypia present within melanocytic lesions. The combination of dermoscopy, digital follow-up, and RCM in the evaluation of equivocal melanocytic lesions has dramatically reduced the number of excisions of benign lesions in patients with the atypical mole syndrome while at the same time improved our ability to detect subtle melanomas. Figs. 4 and 5 showcase 2 lesions for which dermoscopy was unable to correctly diagnose the lesion as melanoma or dysplastic nevus; however, RCM was able to make the correct diagnosis.



Table 1

Confocal features used to differentiate melanoma from nevi








































































Evaluation Level Melanoma Atypical Nevus Nevus
Epidermis Marked or complete loss of honeycomb and/or cobblestone patterns Variably disarranged honeycomb and/or cobblestone patterns Well-conserved honeycombed pattern and/or
Widespread pagetoid cells and large atypical bright cells A few focal isolated atypical cells mostly located towards the center of the lesion Regular cobblestone pattern (when keratinocytes are pigmented)
Pagetoid spread is absent or very limited in its extent
DEJ Poorly demarcated lesion Well-circumscribed lesion Well-circumscribed lesion
Moderate to severe distortion of DEJ architecture Mild to severe distortion of DEJ architecture Well-defined DEJ architecture
Nonedged papillae (ill-defined dermal papillae) Focal areas of nonedged papillae and meshwork pattern Edged papillae (well-defined margins), regular in shape and distribution
Irregular, elongated, and fused interpapillary crests resulting in junctional thickenings Irregular, elongated, and fused interpapillary crests resulting in junctional thickenings Bright basal cells forming cobblestone and ringed pattern
Moderate to severe cellular atypia at the basal layer Junctional nests with irregular shape, size, and location Dense clusters and/or interpapillary processes forming meshwork and/or clod patterns
Variable atypia with hyper-refractile cells at basal layer
Dermis Atypical nucleated cells within dermal papillae Bright non-nucleated plump cells (melanophages) Regular dense and/or some sparse cells within the clods or clusters of cells at DEJ and superficial dermis
Cerebriform nests Bright triangle particles (inflammation) Uniform cellularity within the clods (occasionally bright roundish nucleated uniform cells can be seen in congenital nevi)
Bright non-nucleated plump cells (melanophages) Coarse collagen bundles forming networklike structure Regular vessels in the center of the papilla
Bright triangle particles (inflammation)
Prominent and atypical vessels
Coarse collagen bundles forming a gross networklike structure

Abbreviation: DEJ, dermoepidermal junction.



Fig. 1


The Barcelona algorithm is a 2-step process. The first step requires the clinician to differentiate melanocytic from nonmelanocytic tumors. A melanocytic lesion should be suspected if the dermal papillae are visualized and at least one of the following features is also seen: cobblestone pattern, pagetoid cells, and/or refractile nests. Lesions not displaying any of the aforementioned features are then evaluated to determine if they have any RCM features diagnostic of basal cell carcinoma, seborrheic keratosis, angioma, or dermatofibroma. If not, then by default the lesion is considered to be a melanocytic tumor. The second step of the Barcelona algorithm helps to differentiate melanoma from nevus . Lesions that are dermoscopically suspicious for melanoma are evaluated via RCM for the presence of 2 risk features (+1 point) and 2 protective features (−1 point) for melanoma diagnosis. The high-risk criteria include ( A ) pagetoid roundish cells in the superficial epidermal layers ( yellow arrows ) and ( B ) atypical nucleated cells in the papillary dermis ( yellow arrows ). The protective criteria include ( C ) presence of typical basal cells ( squares ) and ( D ) presence of edged papillae ( asterisks ) at the dermoepidermal junction. The points for the presence of any of the 4 features are summed together, and the final score is used to predict the probability of melanoma. Lesions with a score of −1 or greater have a sensitivity of 100% and specificity of 57.1% for melanoma. Lesions with scores of 0 or greater have a high probability of being melanoma with a sensitivity of 86.1% and specificity of 95.3%.



Fig. 2


Modena algorithm classifies lesions into melanoma or nevus based on the presence of 2 main risk features (+2 points for the presence of each, for a maximum of 4 points) observed at dermoepidermal (DE) junction: (1) cellular atypia at basal layer ( A , squares ); (2) nonedged papillae ( B , asterisks ). In addition, the lesion is evaluated for the presence of 4 minor risk features (+1 point for the presence of each, for a maximum of 4 points): (1) widespread pagetoid infiltration of spinous layer ( C , squares ); (2) roundish pagetoid cells ( D , arrows ); (3) cerebriform nests in dermis (E, squares); (4) nucleated atypical cells in upper dermis (F, arrows). Lesions with a total score of 3 or greater should be excised to rule out melanoma (sensitivity 91.9%, specificity 69.3%).



Fig. 3


Algorithm to distinguish dysplastic nevi from melanoma as described by Pellacani and colleagues. First step requires evaluation of the lesion for the presence or absence of cytologic and architectural atypia; junctional nests of different size or with sparse cells of differing refractility (nonhomogeneous) are considered irregular. Junctional thickenings or short interconnections are irregular refractile aggregates between papillae. Absence of atypical cells in epidermis and atypical junctional nests or aggregates is suggestive of a benign nondysplastic nevus. The second step quantifies the degree of atypia and allows differentiation of probable dysplastic nevi from melanoma. The presence of widespread pagetoid infiltration encompassing at least 50% of lesion, diffuse cytologic atypia at the dermoepidermal junction (DEJ) involving at least 50% of lesion, or nonedged papillae encompassing at least 10% of the lesion is suggestive of melanoma.



Fig. 4


( A ) Dermoscopy of a 3.0 × 3.5-mm new pigmented lesion noted on the forearm of a patient with a history of multiple primary melanomas. The lesion has a globular pattern with an irregular arrangement of globules at the periphery. ( B ) A 5 × 5-mm RCM mosaic at the dermoepidermal junction (DEJ) layer shows a well-demarcated lesion that is predominantly composed of a clod pattern at the periphery ( arrows ). A few small irregular nonaggregated clusters are visible ( red square ) and isolated eccentric pagetoid cells can also be seen ( yellow square ). ( C ) Single RCM image shows large atypical roundish and dendritic pagetoid cells within the superficial epidermis ( yellow arrows ). ( D ) Single RCM image at DEJ displaying nonedged papillae ( asterisks ) and junctional thickenings ( red arrow ) with dendritic cells along the basal layer. DIAGNOSIS: in situ melanoma.



Fig. 5


( A ) This patient has a history of multiple primary melanomas. Two changing lesions were detected on his back during digital surveillance. ( B ) Sequential digital dermoscopy of the 3-mm lesion on the central lumbar area disclosed progressive enlargement of the lesion with development of an atypical network and peripheral blotch. ( C , D ) RCM mosaics of epidermal layers show a well-conserved cobblestone pattern ( yellow squares ). One single isolated small dendritic cell was visible in the superficial layers ( arrow ). ( E ) Mosaic at dermoepidermal junction level shows typical architecture composed of edged papillae ( asterisks ) and meshwork pattern with typical basal cells between the dermal papillae. ( F ) A compound nevus with a lentiginous growth pattern, bridging of rete ridges, and moderate dysplasia (hematoxylin-eosin, original magnification ×20). These features correlate to the meshwork pattern and junctional thickening seen in ( E ). DIAGNOSIS: Compound melanocytic nevus with moderate atypia.

Only gold members can continue reading. Log In or Register to continue

Related posts:

Default ThumbnailBasics of Confocal Microscopy and the Complexity of Diagnosing Skin Tumors Default ThumbnailOpening a Window into Living Tissue Default ThumbnailShining into the White Default ThumbnailApplication of Handheld Confocal Microscopy for Skin Cancer Diagnosis Default ThumbnailThe Role of Reflectance Confocal Microscopy in Clinical Trials for Tumor Monitoring Default ThumbnailFluorescence (Multiwave) Confocal Microscopy

Stay updated, free articles. Join our Telegram channel
Tags:
Feb 11, 2018 | Posted by in Dermatology | Comments Off on Discriminating Nevi from Melanomas

Full access? Get Clinical Tree
Get Clinical Tree app for offline access