The gold standard for melanoma diagnosis is routine histopathology with hematoxylin and eosin (H&E) stains. For clear-cut cases of melanoma, this alone is sufficient. However, as in all things in medicine, there are cases that fall within the gray zone. Is it an atypical nevus or is it melanoma? This is a question encountered on a daily basis in the practice of dermatopathology, which is the study of cutaneous diseases at a microscopic and molecular level.
One reason for this challenge is that there are many benign diagnoses that mimic melanoma. For example, traumatized nevi and recurrent nevi may show pseudomelanomatous features overlying zones of dermal fibrosis. Another challenging group of melanocytic lesions includes “special site” nevi. These are benign nevi located on certain body sites that may have histologic features that mimic melanoma. Some of these “special sites” include the genital region, the breast/milk line, the scalp, palms and soles, and on and around the ear. Similarly, Spitz nevi were originally called “juvenile melanoma” because they histologically look so atypical. (This is not to be confused with spitzoid melanoma of childhood, which is a true melanoma.)
There is also a large category of atypical nevi, often called “dysplastic nevi,” which can demonstrate some overlapping histologic features with melanoma. On the other hand, there are also melanomas that may mimic a benign diagnosis. These lesions are aptly named “nevoid melanoma.”
While we are most comfortable with binary diagnoses of benign and malignant to guide our treatment decisions, as we learn more about the molecular basis of melanoma we must also acknowledge that melanoma, like many malignancies, often develops through sequential mutations, not just one molecular event. Pinning down a firm diagnosis in a lesion that has partially proceeded down the path of malignancy can be challenging, if not impossible.
In these gray zones, where there are overlapping histologic features of benign and malignant diagnoses, your dermatopathologist may reach into their toolbox for additional data points to support one diagnosis over another. Over the course of the history of dermatopathology, the array of tools available has expanded to aid in diagnosis. This article will highlight just a few of the tools available in the field of dermatopathology.
Clinical Information and Appearance
In addition to routine H&E staining, one of the most useful and readily available tools to the dermatopathologist is clinical information and the appearance of the lesion. Providing the dermatopathologist with a measurement and a clinical photograph or dermoscopic image can be immensely helpful, particularly if the image demonstrates a focal area of concern. This can prompt the dermatopathologist to obtain deeper sections to search for the area of concern, which may not be otherwise examined in initial cross-sections.
If a partial biopsy of a larger melanocytic lesion is obtained, providing the size of the lesion can also be a helpful clue to the dermatopathologist that the partial sample may not be fully representative of the lesion as a whole. Lastly, if there is a clear history of a prior biopsy, providing this information to the dermatopathologist can be helpful, as review of prior biopsy material can prevent an overdiagnosis of melanoma if recurrence of the lesion has led to “pseudomelanomatous” histologic features.
Tools to Aid in Diagnosis of Challenging Melanocytic Lesions
When the diagnosis of a melanocytic lesion is still not clear with the H&E stains and clinical information, a wide variety of immunohistochemical stains are available to your dermatopathologist to further sort things out. While beyond the scope of this article, there are also a number of molecular tests that may be useful in challenging cases, particularly pediatric melanoma.
Differentiation markers that highlight melanocytes include immunostains such as Melan-A/MART1, SOX10, MITF, S100, HMB-45 and Tyrosinase. While some of these stains are specific for melanocytes, others highlight other cell types in addition to melanocytes, such as those of neural lineage. In addition to confirming the lineage of the cells, using these stains allows the dermatopathologist to better evaluate the overall architecture of a melanocytic lesion. For example, some atypical architectural features such as poor circumscription, pagetoid scatter and adnexal extension may be better appreciated with the assistance of immunostains.
Biomarkers include immunostains such as Ki67, PHH3, p16, BAP1 and PRAME. A dual-labeling stain for Melan-A/Ki67 allows the dermatopathologist to visualize what proportion of the melanocytes are in a proliferation stage. Similarly, a dual stain for Melan-A/PHH3 highlights those melanocytes that are in or close to mitosis. If either of these stains demonstrates a high percentage of dual-labeled melanocytes, this could be a feature raising concern for a diagnosis of melanoma. However, these markers may also be elevated in the setting of pregnancy or early childhood, particularly in the setting of benign proliferative nodules arising within congenital nevi.
The immunostain for p16 can be used to determine if a lesion may have a mutation in the p16 tumor suppressor protein encoded by CDKN2A, located on chromosome 9p21.3. Complete loss of p16 expression is more commonly seen in melanomas, particularly in spitzoid melanomas. However, because some melanomas retain expression of p16 and some nevi demonstrate loss of p16, this stain must be interpreted with caution and in the context of the lesion as a whole.
There is a group of clinically bland melanocytic lesions termed BAP1 deficient epithelioid melanocytic nevus/tumor. Sometimes, these lesions are more casually referred to as “BAPomas.”
Classically, these lesions demonstrate two populations of melanocytes: a conventional component with small, bland melanocytes, and a spitzoid component with large, epithelioid, atypical melanocytes. This characteristic appearance on H&E may prompt the dermatopathologist to perform an immunostain for BAP1. Loss of nuclear expression of BAP1 within the atypical component confirms a diagnosis of a BAP1 deficient combined melanocytic nevus, which may be sporadic or may occur in the setting of a germline mutation.
If a patient has multiple such lesions, this will raise the possibility of the autosomal dominant BAP1 tumor predisposition syndrome. These patients and affected family members have an increased risk of uveal and cutaneous melanoma, basal cell carcinoma, renal cell carcinoma, lung adenocarcinoma and mesothelioma. In this instance, the BAP1 immunohistochemical stain can help uncover a potential genetic disorder. Loss of BAP1 expression has also been reported in blue nevi-like melanomas and melanoma arising within blue nevi.
One of the newest tools available to dermatopathologists is the immunohistochemical stain for PRAME (PReferentially expressed Antigen of MElanoma). Dermatologists may be familiar with this antigen as one of two genes detected in a noninvasive pigmented lesion assay used to guide clinicians on the need for biopsy of a pigmented lesion.
In contrast to the commercially available noninvasive assay, when PRAME is used as an immunohistochemical stain, the dermatopathologist is concurrently provided with the gold standard of H&E to evaluate architecture and cytology. When PRAME immunohistochemistry is diffusely positive within melanocytes, this provides an additional data point to support a diagnosis of melanoma while a negative result can be a reassuring finding.
In a study of 400 melanocytic lesions, PRAME was diffusely positive in 90.4% of primary cutaneous melanomas, excluding desmoplastic melanomas (which only demonstrated PRAME positivity in 35% of cases). Focal positivity with PRAME was also reported in 13.6% of cutaneous nevi. Thus, the results of this stain (as with all stains) needs to be interpreted cautiously.
While there are many cases in which a diagnosis of melanoma can be made simply with the gold standard of H&E alone, there are inevitable cases that will require additional investigation. In many cases, working closely with your dermatopathologist to provide clinical information and images may be one of the most affordable and useful tools available.
When necessary, there are many immunohistochemical stains available – each with their unique advantages and limitations. The number of stains available will continue to grow as new target antigens are discovered through investigational research. Ultimately, using all available information from the routine pathology, clinical information and immunohistochemical results, your dermatopathologist will be armed to provide more precise and accurate diagnoses to ensure the best possible care for our patient population.
Melinda R. Mohr, MD
Dr. Melinda Mohr is a dual board-certified dermatologist and dermatopathologist. Since 2013, she has worked as a dermatopathologist at Atlanta Dermatopathology, a PathGroup company. In 2019, she took on the leadership role as CLIA Clinical Consultant with Atlanta Dermatopathology. She completed her dermatopathology fellowship at Albert Einstein College of Medicine/Montefiore Medical Center. Her dermatology residency, internal medicine internship and medical school were all completed at Eastern Virginia Medical School in Norfolk, Va.
Lezcano C, Jungbluth AA, Nehal KS, et al. PRAME Expression in Melanocytic Tumors. Am J Surg Pathol 2018;42:1456-1465.
Busam KJ, Gerami P, Scolyer RA. (2019) Pathology of Melanocytic Tumors. Philadelphia, PA:Elsevier.