PD-L1 testing in cutaneous malignancies: interpretation, clinical utility, and unresolved challenges.

Immune checkpoint inhibitors have transformed the management of melanoma, cutaneous squamous cell carcinoma (cSCC), Merkel cell carcinoma (MCC), basal cell carcinoma (BCC), and selected rare cutaneous malignancies, yet the clinical value of programmed death-ligand 1 (PD-L1) immunohistochemistry has not kept pace with the therapeutic importance of the PD-1/PD-L1 pathway. This narrative review, prepared according to the SANRA quality criteria, evaluates PD-L1 testing at four linked levels: what PD-L1 expression represents biologically, how reliably it can be measured, whether it is prognostic or predictive, and whether the result changes clinical management. PD-L1 expression is inducible, spatially and temporally heterogeneous, and present on tumor and immune cells, and its measurement depends on specimen, clone, platform, scored compartment, and cutoff. In melanoma, tumor-cell PD-L1 enriches for response at the cohort level, but PD-L1-negative tumors respond to PD-1 blockade, and the only label that mandates testing is the European authorization of nivolumab-relatlimab for tumor-cell PD-L1 < 1%. In cSCC, MCC, and BCC, checkpoint inhibitors are approved without PD-L1 selection, and no PD-L1 companion diagnostic is approved for any cutaneous malignancy. PD-L1 should therefore be reported with the specimen, clone, platform, compartment, scoring system, and cutoff, and interpreted only within a defined tumor and therapeutic context. Composite approaches that integrate tumor immunogenicity, CD8-positive T cell spatial organization, interferon-gamma signaling, viral status, circulating tumor DNA, and multiplex or digital pathology are more likely than PD-L1 alone to improve treatment selection, but they must demonstrate incremental, reproducible clinical utility.