Phatcharamaneepakorn K.Sitthinamsuwan P.Pattanaprichakul P.Chairatchaneeboon M.Yodrabum N.Mahidol University2026-06-222026-06-222026-01-01Journal of Cutaneous Pathology (2026)03036987https://repository.li.mahidol.ac.th/handle/123456789/117470Introduction: Preferentially expressed antigen in melanoma (PRAME) immunohistochemistry helps distinguish non-acral melanocytic lesions, but few studies have addressed its value in acral lesions. We aimed to identify an optimal PRAME cut-off score to differentiate acral melanoma (AM) from acral melanocytic nevi (AMN). Methods: Formalin-fixed, paraffin-embedded specimens of AM, AMN, and small biopsy of atypical melanocytic lesions (AMLs) were included. PRAME staining was performed. Expression was categorized into five groups: 0 (0%), 1+ (1%–25%), 2+ (26%–50%), 3+ (51%–75%), and 4+ (> 75%). PRAME nuclear expression was compared between AM and AMN to identify the best cut-off score, which was then assessed in small-sized biopsies of AML specimens. Results: We analyzed 62 AMs, 62 AMNs, and 23 AMLs. A PRAME score of ≥ 3+ optimally differentiated AM from AMN (sensitivity 80.6%, specificity 98.4%, positive predictive value 98%, and negative predictive value 83.6%). PRAME expression score ≥ 3+ was strongly associated with AM, whereas AMN showed expression score < 3+ (p < 0.001). However, one AM was completely negative for PRAME, and one AMN exhibited strong PRAME staining. Conclusions: PRAME immunohistochemistry with a cut-off of ≥ 3+ is useful for differentiating AM from AMN. However, false positivity and negativity of PRAME expression must be considered.MedicineArticleSCOPUS10.1111/cup.701632-s2.0-10504211544616000560