Comparative Evaluation of an Extensive Histopathologic Examination and a Real-Time Reverse-Transcription-Polymerase Chain Reaction Assay for Mammaglobin and Cytokeratin 19 on Axillary Sentinel Lymph Nodes of Breast Carcinoma Patients

Authors:
Publisher website
Google Scholar
Abstract
To assess the accuracy of a commercially available real-time reverse-transcription-polymerase chain reaction assay for mammaglobin and cytokeratin 19 mRNAs [GeneSearch Breast Lymph Node (BLN) Assay, Veridex LLC, Warren, NJ] in the detection of axillary sentinel lymph nodes (SLNs) metastases in patients with breast carcinoma. Because of the lack of standardized and widely accepted protocols for a truly accurate histopathologic examination of SLN, the relative merits of alternative assays based on the identification of tumor specific mRNA markers deserve further assessment. A prospective series of 293 consecutive SLNs from 293 patients was evaluated. The BLN assay results were compared with those of an extensive histopathologic examination of the entire SLNs performed on serial frozen sections cut at 40 to 50 μm intervals. The BLN assay correctly identified 51 of 52 macrometastatic and 5 of 20 micrometastatic SLNs, with a sensitivity of 98.1% to detect metastases larger than 2 mm, 94.7% for metastases larger than 1 mm, and 77.8% for metastases larger than 0.2 mm. The overall concordance with histopathology was 90.8%, with specificity of 95.0%, positive predictive value of 83.6%, and negative predictive value of 92.9%. When the results were evaluated according to the occurrence of additional metastases to non-SLN in patients with histologically positive SLNs, the assay was positive in 33 (91.7%) of the 36 patients with additional metastases and in 22 (66.6%) of the 33 patients without further echelon involvement. The sensitivity of the reverse-transcription -polymerase chain reaction assay is comparable to that of the histopathologic examination of the entire SLN by serial sectioning at 1.5 to 2 mm.

This publication has 27 references indexed in Scilit: