Purchase this article with an account.
Saaquib Bakhsh, Tatyana Spektor, Jack Cohen; Tumor thickness and gene expression profiling in estimating uveal melanoma prognosis. Invest. Ophthalmol. Vis. Sci. 2016;57(12):5899.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Over the last 35 years, clinicians have traditionally used tumor thickness of ocular melanomas as a prognostic predictor of metastatic potential. Recent innovations in gene expression profiling have revealed two classes of melanoma that can be used to estimate metastatic potential. Using retrospective observational data collected from biopsy-confirmed ocular melanoma subjects, we hypothesized that there is a positive correlation between tumor grade (based on gene expression profile class) and tumor thickness.
The medical records of 20 adult human subjects diagnosed with uveal melanoma were reviewed. Gene expression profile class and A-scan tumor thickness measurements were examined for an association. A one-way analysis of variance (ANOVA) study technique was used to determine significant differences between gene expression class and tumor size.
The one-way ANOVA with n=20 had a p-value of .0316, suggestive of a correlation between tumor class and tumor thickness. However, post-hoc analysis with Tukey’s HSD test revealed a significant difference only between class 1A and 1B mean thicknesses, and not class 2. In the preliminary analysis, tumors with a favorable gene expression profile of 1A had the smallest average mean thickness while class 1B profiles had a thickness average that was significantly higher than the 1A class (p=0.027). However, class 2 tumors were not associated with the thickest melanomas; instead, the mean tumor thickness was statistically similar to both the 1A (p= 0.24) and 1B (p= 0.41) classes, suggesting a lack of correlation.
In our study cohort, tumor thickness did not significantly increase as the gene profile escalated towards an unfavorable class. This would speak against the traditional use of tumor thickness as a prognostic indicator for uveal melanoma. As we continue to follow the subjects of this study to investigate a correlation between gene expression profile and clinical outcomes, we recommend continued study to further explore this association.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
A-scan tumor thickness percentile distributions in each gene expression profile class
This PDF is available to Subscribers Only