Global research team develops diagnostic model for iris melanoma

According to the model, features that favored a diagnosis of melanoma included size, pupillary distortion, peripheral extension, vascularity and others. Image credit: ©Santiago Cornejo – stock.adobe.com

Arun Singh, MD, and colleagues developed a model that provides direct diagnostic prediction of a lesion being iris melanoma.¹ He is Director of the Department of Ophthalmic Oncology, Cole Eye Institute at the Cleveland Clinic, Cleveland Clinic Main Campus Hospital, Cleveland.

The investigators conducted a retrospective consecutive case series that included 100 cases of pathologically confirmed iris melanoma and 112 cases of iris naevus, with either pathological confirmation or documented stability exceeding 1 year.

The investigators collected the patient demographic data, the features of the clinical presentation, tumor characteristics and follow-up. They excluded cases with iris melanoma with ciliary body extension.

The analysis showed “significant asymmetry in the location of both nevi and melanoma with preference for the inferior iris quadrants (83, 74%) and (79, 79%), respectively (P = 0.50).” Only melanoma cases included tumor seeding, glaucoma and hyphema.

“The features that favored the diagnosis of melanoma were size (increased height, odds ratio [OR], 3.35); increased largest basal diameter (OR, 1.64), pupillary distortion (ectropion uvea or corectopia, OR, 2.55), peripheral extension (angle or iris root involvement (OR, 2.83), secondary effects (pigment dispersion, OR, 1.12) and vascularity (OR, 6.79). The optimism-corrected area under the curve was 0.865. The calibration plot indicated good calibration with most of the points falling near the identity line and the confidence band containing the identity line through most of the range of probabilities,” they said.

Singh and colleagues believe that their predictive model provides direct diagnostic prediction of the lesion being iris melanoma expressed as the probability (%). The use of such a predictive model can enhance decision-making and patient counselling. Further refinements can be undertaken with additional datasets, forming the basis for automated diagnosis, they commented.

Singh was joined in this study by researchers from the Department of Clinical Medicine, Bergen University College, and the Department of Ophthalmology, Haukeland University Hospital, both in Bergen, Norway; and the Department of Quantitative Health Sciences & Taussig Cancer Institute, Cleveland Clinic Main Campus Hospital, Cleveland.

Reference

1. Singh A, Melendez-Moreno A, Krohn J, Zabor EC. Predictive model for iris melanoma. Br J Ophthalmol. 2024;108; https://doi.org/10.1136/bjo-2023-324558