Clinical Images Enhance Pathologists' Accuracy in Diagnosing Skin Lesions

Credit: Adobe Stock/ blackday

A new study showed providing clinical information to pathologists may improve the histopathologic diagnosis of melanocytic skin lesions.¹

“Providing clinical images appears to result in more upgrades than downgrades in histopathologic diagnosis and more malignant diagnoses,” wrote investigators, led by Belinda Lai, MD, from The University of Sydney in Australia.

A histopathologic examination is the standard way to diagnose melanoma. However, a study back in 2012 found flaws in the gold standard, despite a high level of agreement when diagnosing clinically difficult melanocytic. The study revealed melanocytic skin lesions in younger patients and patients with atypical mole syndrome were harder to diagnose.²

Furthermore, a 2014 study found the opposite regarding the level of agreement, indicating a low interobserver agreement among the experts categorizing lesions as malignant or nonmalignant. The results showed there was a lack of consensus in the assessment of atypical Spitz tumors by expert dermatopathologists.³

Thus, earlier research suggests that the diagnoses of melanocytic skin lesions are inaccurate and not guaranteed to be reproducible. Investigators wanted to see if pathologists' having clinician information would improve their confidence and ability to diagnose these skin lesions.¹

After screening 2224 records from PubMed, Embase, Cochrane Library, and the references in the 2018 Cancer Council Australia evidence review, the team examined 162 full-text studies published between January 2018 and January 2024. The study included 7 studies, featuring pathologists from the US, Austria, Germany, Italy, the UK, and Australia, with a mean age of 43 to 55 years and female participants ranging from 23% to 63%. Most skin lesions were on the trunk (19% to 77%) or lower limbs (11% to 16%), and 8% to 45% of the cases were diagnosed as malignant melanoma.

The study revealed clinical images increased diagnostic certainty and agreement between pathologists. This ultimately led to diagnostic improvements in 7.6% to 16.7% of interpretations. One study showed agreement improved from a Fleiss K score of 0.57 (95% confidence interval [CI], 0.54 to 0.60) without clinical information to a score of 0.67 (95% CI, 0.64 to 0.70) with information on age, sex, location of the lesion, clinical diagnosis, clinical image, and dermoscopic image.

Furthermore, 1 study showed pathologists having access to a clinical diagnosis on the requisition form reduced the chances of missing a melanoma that could progress. Another study found including more clinical details, such as the lesion’s appearance, size, or specific concerns, on the forms was associated with greater rates of additional surgical procedures to remove melanoma or a suspicious lesion.

Among patients whose cancer spread to areas far from the original site, a melanoma diagnosis was based on histopathology alone.

The team explained that there were substantially more upgrades than downgrades at a population level due to the greater prevalence of benign lesions compared with malignant ones in the biopsies. In general, it was known whether the pathology upgrades due to clinical information lessened the underdiagnosis of clinically progressive lesions or the overdiagnosis of poor ones.

Investigators wrote how clinicians’ lack of time and reimbursement may pose a challenge for pathologists to have access to clinician information. They suggest creating better digital systems for sharing clinical information, such as a standardized templated requisition checklist. Pathologists may also be unfamiliar with interpreting clinical or dermoscopic images, and the team suggested this could be addressed by including dermoscopy training in dermatopathy training or providing a dermoscopy report that summarizes key features.

Despite these findings, the risk of bias assessment showed all studies had unclear or high risk of bias. The team also wrote the findings were limited by the small number of studies that were heterogenous in their design and the few data on clinical utility and patient health outcomes.

“This scoping review suggests that the provision of clinical information to pathologists interpreting melanocytic skin lesions may improve diagnostic confidence and interobserver agreement and change histopathologic diagnosis,” investigators concluded. “While the provision of key clinical information, such as suspected clinical diagnosis, can improve patient outcomes by reducing the odds of melanoma progression, the clinical value of diagnosis upgrading with the provision of clinical images remains uncertain.”

References

1. ^{Lai B, Soyer HP, Zhu L, et al. Impact of Clinical Information on Melanocytic Skin Lesion Pathology Diagnosis: A Scoping Review. JAMA Dermatol. Published online October 30, 2024. doi:10.1001/jamadermatol.2024.4281}
2. ^{Braun RP, Gutkowicz-Krusin D, Rabinovitz H, Cognetta A, Hofmann-Wellenhof R, Ahlgrimm-Siess V, Polsky D, Oliviero M, Kolm I, Googe P, King R, Prieto VG, French L, Marghoob A, Mihm M. Agreement of dermatopathologists in the evaluation of clinically difficult melanocytic lesions: how golden is the 'gold standard'? Dermatology. 2012;224(1):51-8. doi: 10.1159/000336886. Epub 2012 Mar 20. PMID: 22433231.}
3. ^{Gerami P, Busam K, Cochran A, Cook MG, Duncan LM, Elder DE, Fullen DR, Guitart J, LeBoit PE, Mihm MC Jr, Prieto VG, Rabkin MS, Scolyer RA, Xu X, Yun SJ, Obregon R, Yazdan P, Cooper C, Weitner BB, Rademaker A, Barnhill RL. Histomorphologic assessment and interobserver diagnostic reproducibility of atypical spitzoid melanocytic neoplasms with long-term follow-up. Am J Surg Pathol. 2014 Jul;38(7):934-40. doi: 10.1097/PAS.0000000000000198. PMID: 24618612.}