Utility of a One-Step Nucleic Acid Amplification (OSNA) Assay for Comprehensive Examination of Axillary Lymph Nodes in Breast Cancer.

Abstract

Abstract IntroductionMolecular testing for detection of epithelial and/or breast-specific mRNA has the potential to aid in the comprehensive examination of axillary lymph nodes (ALNs) in breast cancer. We evaluated the performance of a One-Step Nucleic Acid Amplification assay (OSNA; Sysmex), an automated assay for detection of cytokeratin 19 mRNA in chemotherapy(CT)-naive and -treated ALNs of breast cancer patients using conventional histopathologic examination (HE) as the gold standard.METHODSALNs obtained from complete axillary dissections were serially sectioned at 1-mm intervals using a proprietary lymph node cutter (Sysmex), and alternate slices were used for OSNA and HE, which included one hematoxylin and eosin stain and a pancytokeratin (CK) immunostain. Isolated tumor cells were classified as micrometastases (MiM) in CT-treated ALNs and as negative in CT-naive ALNs. The presence of tumor cells in either stained preparation was used for categorization of metastatic carcinoma. The agreement between OSNA and HE results for detection of metastatic carcinoma was estimated by using the kappa coefficient for CT-treated and -naive ALN groups separately.RESULTSWe studied 279 ALNs from 39 patients with breast cancer; these included 162 CT-treated and 117 CT-naive ALNs. In the CT-naive group, metastatic carcinoma was detected by HE in 15 of the 117 ALNs (12.8%), 7 classified as micrometastases and 8 as macrometastases. All but 3 of the 7 micrometastases were detected by OSNA. The kappa coefficient between HE and OSNA for this group was 0.87 (95% confidence interval [CI], 0.72–1.00) which implies near perfect agreement between the two methods. In the 162 CT-treated ALNs, metastatic carcinoma was detected in 44, 13 classified as micrometastases and 31 as macrometastases. The results for 17 ALNs in the CT-treated group were discordant between the two methods: 3 ALNs were considered positive by OSNA but negative by HE; 10 ALNs with a mean size of metastasis of 0.5 mm (MiM) and 4 with a mean size of metastasis of 2 mm (macrometastases) were considered positive by HE and negative by OSNA. The kappa coefficient between the two methods for this group was estimated to be 0.71 (95% CI, 0.58–0.84) which implies substantial agreement between the two methods for this group.CONCLUSIONS1. The detection of metastatic carcinoma was highly comparable between the OSNA assay and conventional HE in both CT-naive and -treated ALNs in breast cancer.2. The minimal discordance between HE and OSNA in the results for detection of very small metastases in ALNs in both CT-naïve and –treated groups was most likely the result of tissue-allocation bias.3. The OSNA assay has the potential to be used as either a stand-alone test or integrated with conventional HE for the comprehensive examination of ALNs in breast cancer, thus likely improving current methods of evaluation of ALNs for more accurate axillary staging. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 1012.