Expert Second Opinion Improves Reliability of Melanoma Diagnoses

Misclassification yielded the lowest rates when first, second and third reviewers were sub-specialty trained dermatopathologists.

Obtaining a second opinion from pathologists who are board certified or have fellowship training in dermatopathology may improve the accuracy and reliability of diagnosing melanoma, according to a new study in JAMA Network Open.

To evaluate the impact of obtaining second opinions, researchers used samples from the Melanoma Pathology Study, which comprises of 240 skin biopsy lesion samples. Among the 187 pathologists who examined the cases, 113 were general pathologists and 74 were dermatopathologists. The team studied misclassification rates, which is how often the diagnoses of practicing US pathologists disagreed with a consensus reference diagnosis of three pathologists who had extensive experience in evaluating melanocytic lesions. They found that the misclassification of these lesions yielded the lowest rates when first, second and third reviewers were sub-specialty trained dermatopathologists. Misclassification was the highest when reviewers were all general pathologists who lacked the subspecialty training. 

"Our results show having a second opinion by an expert with subspecialty training provides value in improving the accuracy of the diagnosis, which is imperative to help guide patients to the most effective treatments," says Joann Elmore, MD a professor of medicine at the David Geffen School of Medicine at UCLA and researcher at the UCLA Jonsson Comprehensive Cancer Center and the director of the UCLA National Clinician Scholars Program. ”While these findings suggest that second opinions rendered by dermatopathologists improve overall reliability of diagnosis of melanocytic lesions, they do not eliminate or substantially reduce misclassification.”

Dr. Elmore is now studying the potential impact of computer machine learning as a tool to improve diagnostic accuracy. She is partnering with computer scientists who specialize in computer visualization of complex image information, as well as leading pathologists around the globe to develop an artificial intelligence (AI)-based diagnostic system.