ABSTRACT
BACKGROUND: In the preoperative setting of breast cancer, fine-needle aspiration biopsy (FNAB) guided by ultrasonography can be used to detect axillary lymph node metastases. The aim of the current study was to determine whether measurement in needle washouts of CYFRA 21-1, a soluble fragment of cytokeratin 19 (CK19), can improve the diagnosis of lymph node tumor deposits. MATERIALS AND METHODS: Our prospective study included 329 breast cancer patients eligible for a sentinel node procedure. Eighty-two patients had a ultrasonography-guided FNAB of an abnormal or suspicious axillary node and were included in the current analysis. Samples were processed with smears and cell-block histology; afterwards, needles were washed with saline solution, and the CYFRA 21-1 content in the washouts was measured. The cutoff value for positive samples that expressed CK19, as shown by their immunohistochemistry, was determined by a receiver operating characteristic curve. On the basis of this value, the specificity and the sensitivity, as well as the positive and negative predictive values, were then calculated using cytology results as a reference standard. RESULTS: Of the 82 axillary lymph nodes sampled, 61 were positive on cytology and were metastatic for breast cancer, as confirmed by surgery. The median CYFRA 21-1 concentration of positive samples was 88.94 ng/mL, whereas in the 21 negative nodes, its median concentration was 1.02 ng/mL (P<0.0001). A CYFRA 21-1 cutoff value of 1.98 ng/mL, obtained by the receiver operating characteristic curve, was able to clearly separate negative from positive samples and gave the test a sensitivity of 0.98 and a specificity of 1.00. Positive and negative predictive values were 1.00 and 0.95, respectively. CONCLUSIONS: The measurement of CYFRA 21-1 levels in needle washouts after axillary lymph node FNABs represents an accurate, sensitive, and highly predictive procedure in detecting metastatic deposits of breast cancer and is fit to complement cytology and CK19 immunohistochemistry as a diagnostic tool.
