Nathanael R. Fillmore
Publications / 2026
paper
JAAD international2026doi:10.1016/j.jdin.2026.04.021

Risk factors for the development of acral melanoma: A multi-institutional, retrospective cohort study.

Charles Lu†, Olivia Burke†, Christopher J. Thang, Sara Khattab, Nga Nguyen, Emma Beagles, Crystal Chang, Suzanne Xu, Jonathan C. Hwang, Linden Huhmann, Nicholas Starink, Martin A. Weinstock, Maryam M. Asgari, Marc S. Hurlbert, John M. Gaziano, Nathanael R. Fillmore, Theodore C. Feldman, Rebecca I. Hartman, Yevgeniy R. Semenov

† Equal first authorship

Summary

Acral melanoma is more prevalent among women and individuals of Asian or Black race, with additional risk associated with prior malignancies. These findings suggest that markers of ultraviolet exposure contribute to the development of this rare subtype, though to a lesser extent than in cutaneous melanoma.

Key result

Acral melanoma was significantly associated with female sex, Asian or Black race, a history of leukemia or lymphoma, nonmelanoma skin cancer, and actinic keratoses compared with both nonmelanoma and cutaneous melanoma controls.

Abstract

BACKGROUND: Acral melanoma (AM) is a rare but aggressive melanoma subtype. Understanding demographic and clinical risk factors is essential for advancing prevention and early detection. OBJECTIVES: To identify demographic and clinical risk factors for AM compared with nonmelanoma and cutaneous melanoma (CM) controls. METHODS: This multi-institutional retrospective cohort study of patients evaluated at the Mass General Brigham and Dana-Farber Cancer Institute. Three cohorts were created via manual curation of pathology reports: AM cases, nonmelanoma controls, and CM controls. AM patients were matched 1:4 to controls based on pre/postindex follow-up time and dermatologic care history. Multivariable logistic regression and structural equation modeling assessed direct and indirect pathways linking sex, actinic keratoses, and melanoma subtype. RESULTS: A total of 2115 patients were analyzed. AM patients were more often female and disproportionately Asian or Black compared with both controls. Relative to non-melanoma controls, AM was associated with female sex, Asian, or Black race, leukemia/lymphoma, nonmelanoma skin cancer, and actinic keratoses. Similar findings were present against CM controls. CONCLUSIONS: AM risk is associated with female sex, Asian or Black race, and select prior malignancies. Findings demonstrate an association between markers of UV exposure (actinic keratoses) and AM risk, though significantly weaker than in CM. These insights may guide prevention, risk stratification, and mechanistic research.