Original Article
Diagnostic performance of core needle biopsy in identifying breast phyllodes tumors
Abstract
Background: A retrospective analysis of diagnoses was performed in patients with phyllodes tumors of the breast (PTB) who received preoperative core needle biopsy (CNB) and had breast surgery at Fudan University Shanghai Cancer Center from January 1, 2002 to April 1, 2013. The resulting data allowed us to compare the accordance between CNB and excision diagnoses of PTB patients and evaluate the accuracy of CNB in preoperative diagnosis.
Methods: Data from 128 patients with PTB who had undergone preoperative CNB and breast surgery were retrospectively analyzed. We reviewed the medical history, clinical follow-up data, and CNB diagnostic data. A diagnostic test was used to evaluate the sensitivity and specificity of CNB in diagnosing benign, borderline, and malignant phyllodes tumors.
Results: The accuracy of CNB for diagnosing PTB was 13.3% (17/128). Of the remaining patients, 98 (75.5% of the PTB patients) were diagnosed with fibroadenoma or fibroepithelial lesions. The sensitivity of CNB at diagnosing benign, borderline, and malignant phyllodes tumors were 4.9% (2/41), 4.2% (3/71), and 25.0% (4/16), respectively, whereas the corresponding specificity were 92.0%, 98.2%, and 100%, respectively. Some clinical features, such as large tumor size, rapid growth, or surgical history of fibroadenomas, were indicative of an increased possibility of PTB.
Conclusions: CNB provides a pathological basis for the preoperative diagnosis of PTB, but it has a poor accuracy and offers limited guidance for surgical decisions. Considering CNB along with multiple histologic features may improve the ability to accurately diagnose PTB. An integrated assessment using CNBs in combination with clinical data and imaging features is suggested as a reliable strategy to assist PTB diagnosis.
Methods: Data from 128 patients with PTB who had undergone preoperative CNB and breast surgery were retrospectively analyzed. We reviewed the medical history, clinical follow-up data, and CNB diagnostic data. A diagnostic test was used to evaluate the sensitivity and specificity of CNB in diagnosing benign, borderline, and malignant phyllodes tumors.
Results: The accuracy of CNB for diagnosing PTB was 13.3% (17/128). Of the remaining patients, 98 (75.5% of the PTB patients) were diagnosed with fibroadenoma or fibroepithelial lesions. The sensitivity of CNB at diagnosing benign, borderline, and malignant phyllodes tumors were 4.9% (2/41), 4.2% (3/71), and 25.0% (4/16), respectively, whereas the corresponding specificity were 92.0%, 98.2%, and 100%, respectively. Some clinical features, such as large tumor size, rapid growth, or surgical history of fibroadenomas, were indicative of an increased possibility of PTB.
Conclusions: CNB provides a pathological basis for the preoperative diagnosis of PTB, but it has a poor accuracy and offers limited guidance for surgical decisions. Considering CNB along with multiple histologic features may improve the ability to accurately diagnose PTB. An integrated assessment using CNBs in combination with clinical data and imaging features is suggested as a reliable strategy to assist PTB diagnosis.
