PRAME Expression and Melanoma In Situ Margin Assessment

Abstract:

Melanoma in situ on chronically sun-damaged skin may extend significantly beyond clinically visible borders. Immunohistochemistry, particularly PRAME (PReferentially expressed Antigen in MElanoma), has emerged as a useful adjunct in delineating subclinical spread of lentigo maligna-type melanoma in situ. We report a case illustrating the diagnostic challenge of determining margins for melanoma in situ of lentigo maligna type even with the benefit of PRAME immunohistochemistry.

CASE

An 80-year-old man with a history of multiple keratinocyte carcinomas underwent a routine full-body examination and was found to have a pigmented patch on the ear.^1,2 Two scouting biopsies were performed. Routine histologic evaluation revealed an atrophic epidermis without nest formation or pagetoid spread of atypical melanocytes (specimen 1). The other specimen demonstrated obvious melanoma in situ (specimen 2) (Figs. 1 and 2). SOX-10 and PRAME immunostains showed increased single melanocytes in both samples. Notably, specimen 1 exhibited a greater number of PRAME-positive melanocytes than expected given the histological appearance on routine sections. The surgical team opted to remove the entire pigmented lesion noted on clinical inspection (including the area of extension into the conchal bowl) and the patient has performed well.

FIGURE 1.

Left ear with an irregular, variably pigmented brown to tan patch with indistinct borders blending into surrounding photodamaged skin and the conchal bowl. The anterior circle marks the site of specimen #1 and the posterior circle marks the site of specimen #2.

FIGURE 2.

Biopsy of the area marked as #1 reveals pigmentation on routine hematoxylin and eosin-stained section but fails to reveal nests or pagetoid spread of melanocytes. Biopsy of the area marked as #2 reveals irregularly shaped nests and pagetoid spread of melanocytes (The images are from the same area with biopsy #1 on the left and biopsy #2 on the right. The first row is stained with hematoxylin and eosin, the second SOX-10, and the third PRAME; all are at original magnification ×200).

DISCUSSION

PRAME is widely used to distinguish melanoma from benign mimickers.³ A commonly used diagnostic cutoff for MIS—diffuse nuclear staining in ≥75% of melanocytes (4+ patterns) has reported sensitivity of 77% and specificity of 100% in high cumulative sun-damaged skin. By contrast, solar lentigo usually shows absent or minimal PRAME expression, with ≤10 positive melanocytes per millimeter and often none.⁴ Rare isolated PRAME-positive melanocytes may occur, but care should be taken to not overinterpret their significance. Diagnostic thresholds such as >10 PRAME-positive cells per millimeter or >50%–75% diffuse nuclear staining improve specificity, yet interpretation must always be integrated with morphologic and clinical features.⁴

Emerging biologic data suggest that extracellular vesicles (EV) containing PRAME may contribute to melanocyte proliferation and malignant transformation.⁵ Melanoma-derived EVs can transfer PRAME to benign melanocytes, leading to altered gene expression and protumorigenic behavior. PRAME knockout in melanoma lines reduces these effects, supporting PRAME's role in tumor-microenvironment modulation.⁵ Although these mechanisms may not directly affect histologic criteria, they could explain low-level PRAME expression in melanocytes near melanoma and could be a factor influencing the findings noted in our patient.

This case illustrates a notable discordance between light microscopic findings, clinical impression, and immunohistochemical results. Discussion with our surgical colleagues suggested a willingness to interpret a lesion with the density of PRAME-positive melanocytes observed in specimen 1 as melanoma in situ. By contrast, our dermatopathology team adopted a more cautious approach, recognizing that before the advent of PRAME and SOX-10 staining, such a lesion might have been considered negative for melanoma in situ. This underscores the importance of careful clinicopathologic correlation to avoid overdiagnosis and potential overtreatment. We highlight this scenario to encourage both dermatopathologists and clinicians to remain vigilant, as evolving ancillary techniques may challenge the traditional primacy of routine histology as the diagnostic gold standard. Unexpectedly increased densities of PRAME-positive melanocytes in chronically sun-damaged skin, especially when not accompanied by unequivocal cytologic atypia or architectural disorder—should prompt careful clinicopathologic correlation. In such settings, consideration of lesion size, evolution, anatomic site, and dermoscopic features becomes critical in determining the biologic significance of the PRAME findings. Leibing et al⁶ noted that retrospective PRAME staining of archival margins previously deemed negative revealed isolated and grouped PRAME-positive cells in some cases. None of their patients developed local recurrence despite limited follow-up and small sample size. Gradecki et al⁷ noted that PRAME immunohistochemistry is a highly sensitive (93.5%) and specific (94.7%) test for diagnosing lentigo maligna with strong correlation with conventional histologic methods but there may be cases where a nuanced approach is essential. A threshold-based and pattern-based approach, emphasizing diffuse confluent staining rather than scattered or patchy positivity, may help mitigate the risk of overdiagnosis. In our opinion, PRAME should be viewed as a powerful adjunct that refines but does not replace traditional histopathologic assessment in the diagnosis of melanoma in situ.