Cancer size, histotype, and cellular grade may limit the success of fine‐needle aspiration cytology for screen‐detected breast carcinoma

2009 
BACKGROUND: Fine-needle aspiration cytology (FNAC) was adopted as the first-line method to assess breast lesions in the Verona Breast Cancer Screening Program. The radiological and pathological factors relating to the success of FNAC in breast cancer series were evaluated. METHODS: Between July 1999 and June 2004, 418 breast cancers were submitted to FNAC in the Verona Breast Cancer Screening Program. The results of FNAC diagnoses were compared with final histology. The FNAC sensitivity rate, underestimation of malignancy rate, and inadequacy rate were correlated with histotype, size, grading, and radiologic imaging. RESULTS: Of the 418 cancers, 95 were in situ, and 323 were invasive. The sensitivity rate was higher in invasive cancers (P < .001), and the underestimation of malignancy rate was greater in in situ cancers (P = .002). Lobular type cancers had a lower sensitivity rate in invasive and in situ cancers. The sensitivity rate was 100% in medullary, mucinous, and papillary cancers, and no case had inadequate sampling. The underestimation of malignancy rate was higher in tubular carcinoma (18.2%); lobular carcinoma showed a higher inadequacy rate (10.4%). The sensitivity rate was lower and the underestimation of malignancy rate was higher in low-grade carcinomas and in lesions <1 cm (P < .001). The performance of FNAC was not significantly influenced by mammographic imaging of lesions. CONCLUSIONS: Low-grade cancer histotype, cancer size <1 cm, and lobular and tubular histotypes limit the possibility of obtaining positive results by FNAC. Operator experience and multidisciplinary consultation may help in overcoming these limitations. Pathologists must be aware of the limits of FNAC; results must be critically evaluated in light of the triple assessment. Cancer (Cancer Cytopathol) 2009. © 2009 American Cancer Society.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    36
    References
    25
    Citations
    NaN
    KQI
    []