El-Sayed ME, Rakha EA, Reed J, et al.. Predictive value of needle core biopsy diagnoses of lesion of uncertain malignant potential (B3) in abnormalities detected by mammographic screening

Department of Histopathology, Nottingham University Hospitals NHS Trust, University of Nottingham, Nottingham, UK.
Histopathology (Impact Factor: 3.45). 01/2009; 53(6):650-7. DOI: 10.1111/j.1365-2559.2008.03158.x
Source: PubMed


Breast needle core biopsy (NCB) is now a commonplace diagnostic procedure in breast cancer screening, providing accurate diagnoses of both benign and malignant lesions. However, NCB may result in the borderline diagnoses of lesion of uncertain malignant potential (B3) or suspicious of malignancy (B4). The aim was to study a large series of B3 cases from population-based screening subjects in order to evaluate positive predictive values (PPVs) for malignancy.
The results of 523 NCBs of women screened over a 7-year period (1999-2006) in the East Midlands region, UK, with a B3 diagnosis who underwent surgical excision, were reviewed and compared with the final excision histology. Five percent of NCBs were reported as B3. The most frequent histological subtypes were atypical intraductal epithelial proliferation (AIDEP) and radial scar/complex sclerosing lesion (RS/CSL). Final excision histology was benign in 417 (80%) and malignant in 106 (20%) subjects (60 ductal carcinoma in situ and 46 invasive carcinoma). Lesion-specific PPVs were as follows: AIDEP 32%; lobular neoplasia (LN) 30%; RS/CSL with AIDEP or LN 24%; RS/CSL without atypia 9%; papillary lesion with AIDEP or LN 36%; and papillary lesion without atypia 4%. Five of the 32 fibroepithelial lesions with cellular stroma were phyllodes tumours (four benign and one borderline). None of the five mucinous lesions on NCB was malignant.
Our results show that approximately one-fifth of NCB of screen-detected breast lesions classified as B3 are malignant on excision, and the likelihood of malignancy varies substantially between different histological subtypes.

21 Reads
  • Source
    • "In the breast, a wide range of benign and malignant spindle cell lesions occur in the breast; for review see [10]. In this context, it is important to differentiate spindle cell metaplastic carcinoma-which can appear deceptively bland, from other mesenchymal lesions [11]. This current case shows no expression of epithelial markers. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Fibroepithelial stromal polyps (FESP) are benign lesions that typically occur in the genital area and are known to represent a diagnostic challenge for pathologists. Not only do they have a spectrum of morphological changes that ranges from bland morphology to rather atypical appearances, but they also share morphological features with a number of benign and malignant lesions. This is a report of a rare presentation of a FESP of the breast. Case presentation We describe an unusual case of a large polypoid mass arising from the nipple and connected to it by a long pedicle in a female of 45. The lesion comprised spindle and stellate shaped cells with bizarre stromal giant cells. The morphological and immunohistochemical diagnostic features are provided together with a discussion of possible mimics. Conclusion FESPs may occur in the female breast. It is important to differentiate the lesion from other benign and malignant spindle cell lesions particularly metaplastic carcinoma.
    Full-text · Article · Aug 2013 · BMC Research Notes
  • Source
    • "Moreover, when compared to thinner needle punction, it may distinguish benign and malignant tumors and in situ or invasive forms, mainly due to its ability to obtain malignant histological diagnosis. This method is able to provide important information as prognostic and predictive breast cancer such as histological type, nuclear grade and presence of lymphatic vascular embolization as well as information regarding the biological behavior of tumors where important prognostic factors and treatment are essential, such as the status of estrogen receptors, progesterone receptors, c-erbB-2 markers and also to new therapeutic targets (EGFR, Ki67, etc..), which may be analyzed with the diagnosis, hence, providing a better planning with respect to the therapy to be used [6,10-12]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background We evaluated the presence of ductal carcinoma in situ (DCIS) in core needle biopsies (CNB) from invasive ductal lesions. Methods Retrospective study, which analyzed 90 cases of invasive ductal carcinoma lesions. The percentage of DCIS was quantified in each specimens obtained from CNB, which were compared to the surgical specimens. CNB and surgical specimens were evaluated by the same pathologist, and the percentage of DCIS in CNB was evaluated (percentage) and divided into categories. We considered the following parameters regarding the amount of DCIS: 1 = 0; 2 = 1 for 5%; 3 = 6 for 24%; 4 = 25 for 50%; 5 = 51 for 75% and 6 = 76 for 99%. The number of fragments and the histological pattern of DCIS was found. Results We found the following results regarding the distribution of the percentage of DCIS in the CNB: 1 = 63.3%; 2 = 12.2%; 3 = 12.2%; 4 = 5.6%; 5 = 1.1% and 6 = 5.6%. The logistic regression analysis showed that CNB percentages above 45% reflected the presence of DCIS in the surgical specimen in 100% of the cases (p < 0.001), with a specificity of 100%, accuracy of 83.3% and false positive rate of 0% (p <0.001). Conclusion There is direct relationship between extensive intraductal component in the surgical specimen when the core biopsy shows 45% or more of the DCI or microinvasive in the material examined.
    Full-text · Article · Jun 2012 · International Archives of Medicine
  • [Show abstract] [Hide abstract]
    ABSTRACT: A second-generation nondestructive evaluation (NDE) system that discriminates between different types of chemical munitions is under development. The NDE system extracts features from the acoustic spectra of known munitions, builds templates from these features, and performs classification by comparing features extracted from an unknown munition to a template library. Improvements over first-generation feature extraction, template construction and classification algorithms are reported. Results are presented on the performance of the system on a large data set collected from surrogate-filled munitions
    No preview · Conference Paper · Jan 1994
Show more