Proposed Prognostic Grading System for Uveal Melanoma on Fine Needle Aspiration Based on Cytomorphologic Features
OA Asojo, Z Correa, SN Masineni, JQ Mo, JJ Augsburger, H Yin. University of Cincinnati, Cincinnati, OH; Cincinnati Children's Hospital, Cincinnati, OH; University of Cincinnati Health Center, Cincinnati, OH
Background: Uveal melanoma is the most common primary ocular malignancy in adults. Currently, most patients are treated by eye-preserving therapy. Fine needle aspiration cytology (FNAC) is an important diagnostic tool, particularly in delineating simulating lesion. A small-scale study demonstrated excellent cytomorphologic correlation between FNAC and enucleation sections. However, no large studies have been published and no cytology grading system has been established. Such a system is useful in guiding or monitoring treatment and predicting outcome. We propose a grading system combining cytomorphologic features to derive a cytologic diagnosis. We further evaluate the association between cytologic diagnosis and clinical prognostic parameters.
Design: A retrospective review was conducted independently by 3 pathologists on eye fine needle aspirates from 62 cases of uveal melanoma. Cytomorphologic parameters were evaluated: cellularity (score 0-2), atypia (score 0-3), dyscohesion (score 0-1), pigmentation (score 0-2), necrosis (score 0-1) and cell morphology (epitheliod vs spindle). Clinical prognostic parameters include age, sex, ciliary body invasion (CB), tumor largest base diameter (LBD) and thickness of tumor. Spearman correlation analysis was performed to ascertain an association between the cytomorphologic grade and clinical parameters.
Results: Seven cytologic diagnosis were established: 1=Not sufficient for diagnosis (QNS), 2=Nevus, 3=Borderline/Indeterminate, 4= Suspicious for melanoma, 5= Melanoma, spindle type, 6= melanoma, mixed type, 7= Melanoma, epitheliod type. Cytologic grade and diagnosis correlates with tumor thickness (r=0.3556; p=0.0046), cellularity (r=0.7173; p<0.0001), dyscohesion (r=0.7522; p<0.0001), atypia (r=0.6951; p<0.0001), pigment (r=0.3992; p=0.0013) and necrosis (r=0.4482, p=0.003). There was no correlation between CB invasion and LBD.
Conclusions: The proposed grading system of FNAC shows a great promise as a cost-effective way to predict biological behavior of uveal melanoma. This grading method is easy to use and reproducible among pathologists. Adding this grading system to the cytology report may help clinicians patient management and predicting survival.
Wednesday, March 24, 2010 1:00 PM
Poster Session VI # 258, Wednesday Afternoon