The objective of this study was to establish that the denaturing fixative, formalin and acetic acid (AFA), is equivalent to neutral-buffered formalin (NBF) on the nCounter system using the Prosigna assay. Twenty-five previously resected tumors from breast cancer patients were apportioned and fixed with either NBF or AFA prior to embedding in paraffin. Differentially fixed and paraffin embedded tissue pairs were sectioned and pathological review was performed according to the Prosigna assay protocol. RNA was isolated using the Prosigna assay kit and analyzed using the NanoString nCounter Dx Analysis system. 47 of the 50 blocks (94%) passed RNA quality control (QC) criteria and were collectively designated as the analysis set. We found that RNA yield and purity (A260/A280) were not significantly different between the fixatives within the analysis set. In the analysis of ROR score, the tissue pairs had a Pearson Correlation Coefficient of 0.91, similar to the correlation observed between paired tissue blocks using a single fixative in previous studies. Subtype concordance between differentially fixed tissue pairs was high (K=0.80). No significant bias or variability in risk of recurrence (ROR) score attributable to fixative was detected in this study. Further analysis revealed that sample pairs with larger differences in ROR score were limited to nearly-depleted tissue blocks containing large amounts of normal tissue due to proximity of the tumor margin. The results of this study demonstrate that the fixative, AFA, does not contribute significantly to bias and variability of assay results.
Keywords: AFA; Breast cancer; Concordance validation; FFPE; Genomic testing; nCounter technology.
Copyright © 2017. Published by Elsevier B.V.