Standardized Computer-Assisted Analysis of PRAME Immunoreactivity in Dysplastic Nevi and Superficial Spreading

Elias A T Koch1,2, Michael Erdmann1,2, Carola Berking1,2

  • 1Department of Dermatology, Uniklinikum Erlangen, Friedrich-Alexander-University Erlangen-Nürnberg (FAU), 91054 Erlangen, Germany.

Insights

PRAME immunostaining aids in distinguishing melanoma from nevi. Computer-aided analysis of PRAME-positive cells offers a reliable tool for dermatopathologists, improving diagnostic accuracy for melanoma.

Area of Science:

  • Dermatopathology
  • Oncology
  • Digital Pathology

Background:

  • PRAME (PReferentially expressed Antigen in MElanoma) is a cancer testis antigen.
  • PRAME is frequently expressed in melanoma but its immunostaining interpretation lacks standardization.
  • Distinguishing melanoma from nevi is critical for patient management.

Purpose of the Study:

  • To develop and validate a computer-aided method for quantifying PRAME expression.
  • To assess the utility of PRAME immunostaining in differentiating melanoma from dysplastic nevi.
  • To establish a standardized diagnostic approach for PRAME in histopathology.

Main Methods:

  • Digitization of PRAME-stained histological slides of melanoma and dysplastic nevi.
  • Analysis using the QuPath digital pathology platform.
  • Statistical analysis including t-tests and ROC AUCs with a defined cut-off score.

Main Results:

  • A significant difference in PRAME-positive cells/mm² was observed between melanoma (455.3) and dysplastic nevi (60.5; p < 0.001).
  • ROC AUC scores for PRAME-positive cells were 0.866 for the entire lesion and 0.901 for the epidermis.
  • A cut-off score of 97.67 cells/mm² correctly classified 94.1% of cases in a validation cohort.

Conclusions:

  • Computer-aided assessment of PRAME immunostaining is a valuable tool in dermatopathology.
  • This method aids in distinguishing dysplastic nevi from melanoma.
  • Challenging cases with moderate expression and ambiguous morphology require further investigation.

Related Concept Videos