Cutter- versus Needle-Based Biopsy for Choroidal Melanoma: Does Biopsy Technique Affect Molecular Yield?

Abstract

Purpose : Gene Expression Profiling (GEP) and Next Generation Sequencing (NGS) are performed on the majority of choroidal melanomas diagnosed in the US. Although these analyses were validated on fine needle aspirates, a substantial proportion of biopsies are currently performed using a non-validated vitrectomy Cutter-based Biopsy (Bx) technique. The purpose of this study was to validate the hypothesis that Cutter Bx is noninferior to Needle Bx terms of molecular yield.

Methods : Cutter Bx was performed with a 25g vitrectomy cutter inserted into the geometric center of the tumor and activated at a low mechanical cut rate (1000cpm) and maximum vacuum (600mm^Hg). All specimens were submitted to Castle Biosciences (Pheonix, AZ) for GEP Classification with or without NGS Analysis (Decision DX®-UM ±Seq). Primary outcome measures included: successful GEP tumor classification, GEP discriminant value, successful detection of a pathogenic Gq pathway mutation, and Gq allelic frequency.

Results : GEP Classification was attempted on 89 non-irradiated tumor samples and a valid result was achieved in 100% of both the Needle (50/50) and Cutter Bx (39/39) groups. There was no significant difference in mean discriminant value (0.81 vs. 0.87; P=0.3) indicating an equal strength of GEP classification. NGS Analysis was attempted on 59 tumor biopsies. There was no significant difference in the proportion of samples yielding a pathogenic Gq mutation (85%, 17/20 vs. 92%, 36/39; P=0.67), nor a difference in Gq mutation allelic frequency (44% vs. 41%; P=0.57) between Bx groups. Excessive specimen volume (>200µL) was observed more frequently with Cutter (85%) vs. Needle Bx (50%, p=0.003), however none of these specimens failed GEP classification, and there was no association between excess volume and failure to detect a Gq mutation (P=0.97). Low RNA concentration (<1.25ng/µL) was associated with an increased risk of NGS failure (P=0.0096), with this issue occurring at a similar frequency in both Bx groups (13% vs. 10%; P=0.92).

Conclusions : This study provides validation of a Cutter-based Bx approach for choroidal melanoma. Cutter Bx provided similar molecular yield for GEP Classification and NGS Analysis when compared to Needle Bx, the current standard of care. Low RNA concentration occurred rarely with both techniques. Cutter Bx was more prone to excess specimen volumes, but this was not associated with failed GEP or NGS.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.