Comprehensive Genomic Profiling of Cutaneous Adnexal Carcinomas: A Genomic Landscape Study

Cutaneous adnexal carcinomas (CACs) comprise a diverse group of malignant tumors that show morphological differentiation toward one of the four main adnexal structures in normal skin: hair follicles, sebaceous glands, sweat-apocrine glands, and sweat-eccrine glands. These tumors can arise sporadically or may be associated with rare genetic syndromes. A total of 276 CACs cases underwent hybrid capture-based comprehensive genomic profiling (CGP) to assess all classes of genomic alterations (GA). Sequencing data were used to determine microsatellite instability (MSI) status, tumor mutational burden (TMB), genomic loss of heterozygosity (gLOH), genomic ancestry, and COSMIC mutational signatures. PD-L1 expression was evaluated by immunohistochemistry (TPS; Dako 22C3). Statistical analyses were performed using Fisher's exact test, with false discovery rate correction via the Benjamini-Hochberg method. Sequencing was performed on primary cutaneous tumors in 131 cases (47.4%) and on local recurrence or metastatic site biopsies in 145 cases (52.5%). Across all groups, there was a male predominance (64-81%) and similar mean ages (59-63 years), with apocrine (APO) tumors occurring in older patients than eccrine (ECC) tumors (72 vs. 62 years; p = 0.001). Histologically, 173 tumors (62.7%) were sweat gland-derived (SWT), 55 (19.9%) sebaceous gland-derived (SEB), 14 (5.1%) hair follicle-derived (HRF), and 34 (12.3%) unclassified (UNK). Among SWT tumors, 150 (86.7%) were eccrine and 23 (13.3%) apocrine. SWT tumors included digital papillary adenocarcinomas (DPA, 6.9%), mucinous carcinomas (MC, 6.3%), porocarcinomas (POR, 11.0%), spiradenocarcinomas (SPR, 8.1%), syringoadenocarcinomas (SRNG, 5.8%), and 77 (44.5%) unclassified cases. The number of GA per tumor was highest in SEB compared with SWT tumors (7.9 vs. 4.9; p = 0.005) and lowest in DPA (2.1 vs. 5.0 in non-DPA; p = 0.03). No differences in ancestry distribution were observed. Compared with SWT tumors, SEB tumors exhibited higher frequencies of RB1 (38.2% vs. 8.1%; p < 0.0001) and TP53 alterations (76.4% vs. 43.4%; p = 0.0002), suggesting potential neuroendocrine differentiation. MC tumors showed significantly higher PTCH1 alterations than non-MC tumors (36.4% vs. 1.8%; p = 0.044). MSI-high status was most frequent in SEB tumors compared with all other groups (15.7% vs. 1.2%; p = 0.005), and gLOH > 16% was also more common in SEB than SWT tumors (19.6% vs. 7.2%; p = 0.081). The MMR signature occurred more frequently in SEB than SWT tumors (32.0% vs. 2.1%; p = 0.005). Mean TMB was elevated across most CACs types, ranging from 10.4 mutations/Mb in HRF to 38.8 mutations/Mb in MC, with the exceptions of APO (2.7 mut/Mb; p = 0.001) and DPA (1.4 mut/Mb; p = 0.003). PD-L1 expression was generally low and did not differ significantly between SWT and SEB tumors (37.0% vs. 33.3%; NS). Given the limited data on CAC treatment, this study provides a catalog of commonly observed GA. SEB tumors exhibited the highest frequency of genomic alterations. Prospective clinical trials are needed to determine the prognostic and predictive value of CAC-specific biomarkers for immune checkpoint inhibitor (ICI) response, which is essential for integrating novel therapies into the evolving treatment landscape.