The current American Joint Committee on Cancer (AJCC) classification (8th edition) relies on measurements of tumor thickness and largest basal diameter for prognostication of metastatic mortality in patients with ciliary body and choroidal melanoma. In this study, investigators used a simple formula for estimating primary tumor volume from these measurements. Their findings suggest that incorporating estimates of tumor volume may improve upon the already excellent prognostic utility of the current model, especially in later stages of disease.
Study Design
This retrospective, multicenter case series included 6528 patients with choroidal or ciliary body uveal melanoma managed between 1981 and 2022 in 3 ocular oncology centers including Wills Eye Hospital, Philadelphia (1192 patients), Liverpool Ocular Oncology Centre, Liverpool (3789 patients), and St. Erik Eye Hospital, Stockholm (1549 patients). Tumor characteristics evaluated included largest basal diameter (LBD), thickness, ciliary body involvement (CBI), and extrascleral extension (EXE). The smallest tumor diameter was assumed to be 85% of the LBD. Outcome of metastatic death was determined based on data from cause-of-death registries (United Kingdom; Sweden) and on retrospective review of medical charts (United States). Tumor volume was estimated using LBD and thickness, based on the assumption that most posterior uveal melanomas have a semi-ellipsoid, dome-shaped configuration. All patients were initially classified into AJCC-based categories and volume-based categories and then further distributed into 4 subcategories according to the presence or absence of CBI and EXE. Finally, corresponding stages were assigned (AJCC: I, IIA, IIB, IIIA, IIIB, IIIC; Volume: IA, IB, IIA, IIB, IIIA, IIIB, IIIC, IV).
Outcomes
Univariable Cox regression showed that tumor volume and all variables used in the current AJCC classification (LBD, thickness, CBI, and EXE) were significant predictors of metastatic death. Whereas thickness and EXE >5 mm were not independent predictors of survival in the AJCC classification, they became significant when LBD and thickness were replaced with tumor volume. The presence of monosomy 3, 8q gain, and epithelioid cell morphology showed significant positive correlation with tumor volume. Volume-based staging provided better incidence separation and cumulative incidences of metastatic death at 15 years compared with AJCC staging. Furthermore, patients classified by volume as stage IIIC showed significantly higher incidence of metastatic death than those in stage IIIB, whereas differences in disease-specific survival between patients with stage IIIC and IIIB as classified by AJCC staging were not significant.
Limitations
This is a retrospective study with different equipment and techniques used at each of the 3 study centers for measurement of tumor size and anatomical location. The tumor volume data used in this study are not based on actual volumetric measurement but rather an estimation of tumor volume based on tumor LBD and thickness. The volume estimation also assumed a tumor’s smallest diameter to be 85% of its largest diameter, which may not be accurate for many tumors depending on the tumor’s shape. Another potential limitation is the possibility of inaccuracies in determining the cause of death.
Clinical Significance
This study shows that staging ciliary body and choroidal melanomas based on estimation of tumor volume provides slightly better separation of survival curves when compared with the current AJCC classification methods—especially for patients with stage IIIB and IIIC—and improves prognostication of metastatic death.
Financial Disclosures: Dr. Arman Mashayekhi discloses no financial relationships.