Gene Expression Profiling Prognostication of Posterior Uveal Melanoma: Does Size Matter?

Purpose: Investigate the influence of tumor size by American Joint Committee on Cancer (AJCC) stage, Collaborative Ocular Melanoma Study (COMS) size, tumor largest basal diameter (LBD), and tumor thickness on prognostication by gene expression profiling (GEP) class.

Design: Two-center retrospective study.

Participants: Two hundred fifteen consecutive patients diagnosed with posterior uveal melanoma over a 5-year period who were evaluated with prognostic fine-needle aspiration biopsy at the time of primary treatment.

Methods: Patient demographics, tumor clinical size, AJCC stage, COMS size, GEP class, presence of metastasis, and mortality data were collected. Metastasis-free-survival (MFS) was defined as time to metastasis or death from any cause. Comparisons were made using Pearson chi-square tests or Fisher exact tests for categorical factors, and t tests or Kruskal-Wallis tests for continuous measures. Cox proportional hazards models were fit to identify whether size measurements increased the prognostic discrimination index (C-statistic).

Main outcome measures: Metastasis-free-survival.

Results: The average follow-up interval was 22.0 months (range, 12.0-37.0 months). Eighty-nine tumors were class 1A, 48 class 1B, and 78 class 2. Twenty-one patients developed metastatic disease detected by surveillance and confirmed by liver biopsy. Three-year MFS was 96% for class 1 and 63% for class 2. Five-year MFS was 96% for class 1 and 49% for class 2. All size measures significantly improved prognostic discrimination index by GEP class, as shown by increase in the C-statistic with addition of size variables (C-statistic 0.750 GEP alone, 0.830 GEP with AJCC [P = 0.016], 0.822 GEP with COMS [P < 0.001], 0.842 GEP with LBD [P <0.001], and 0.847 GEP with tumor thickness [P < 0.001]). Class 2 patients with metastasis had larger tumors compared with nonmetastatic class 2 tumors (AJCC class, P = 0.004; COMS class, P = 0.024; with metastasis mean thickness 6.5 mm [interquartile range (IQR), 3.8-9.5 mm]; without metastasis, 3.9 mm [SD, 3.1-6.0 mm]; P = 0.008), with metastasis mean LBD 14.9±2.8 mm, without metastasis, 12.3±2.7 mm (P < 0.001). All class 1 tumors with metastasis were large and required enucleation.

Conclusions: Incorporation of tumor size enhances the prognostic discrimination index of the GEP test in patients with posterior uveal melanoma. All size tumor parameters are equivalent in their ability to enhance GEP prognostication.