Agreement on the Clinical Diagnosis and Management of Cutaneous Squamous Neoplasms

Authors


Address correspondence and reprint requests to: Allan C. Halpern, MD, Memorial Sloan-Kettering Cancer Center, 160 East 53rd Street, New York, NY 10022, or e-mail: halperna@mskcc.org

Abstract

BACKGROUND Diagnostic accuracy and preferred therapeutic strategies for actinic keratoses (AKs) and squamous cell carcinoma (SCC) have significant public health implications.

OBJECTIVE To evaluate clinical–pathologic agreement on the diagnosis of AKs and early SCCs and to characterize the effect of diagnosis on therapeutic decisions.

METHODS & MATERIALS Nine dermatologists and two dermatopathologists reviewed an image-based dataset of AKs and early SCCs. Clinical–pathologic agreement, inter- and intraobserver reliability for clinical diagnosis, and frequencies of therapies according to pathologic diagnosis were assessed.

RESULTS Clinical–pathologic (κ=0.10) agreement was poor, whereas interobserver (κ=0.24) and intraobserver (κ=0.28) agreements were fair. Participants were more likely to treat AKs with cryotherapy (64.2%) and to manage SCCs with surgery (72.8%). Therapeutic choice rarely changed after participants were shown histological photomicrographs. Participating clinicians treated most lesions histologically diagnosed as SCC in situ arising within AK using surgery, whereas pathologists selected cryotherapy or curettage and electrodesiccation for these lesions.

CONCLUSION We found poor clinical–pathologic agreement and reproducibility for clinically distinguishing between AK and early SCC even between skin cancer specialists from a single academic group practice. Nomenclature used in the pathologic diagnosis of AK and SCC affects clinicians' therapeutic decisions.

The authors have indicated no significant interest with commercial supporters.

Ancillary