The purpose of this paper was to determine whether diagnosis is influenced by the number and position of biopsies obtained from breast lesions and to assess intra- and interobserver diagnostic agreement.
The value of examination of multiple levels of mammary needle core biopsy specimens taken for investigation of lesions other than calcification.
Lee AH, Villena Salinas NM, Hodi Z, Rakha EA, Ellis IO
Purpose
Conclusion
An initial diagnosis based on a biopsy of non-calcified lesions or axillary nodes was only altered in 1.5% and 2.5% of lesions, respectively, after examination of additional biopsies sampled at other levels. In contrast, an initial diagnosis based on a single biopsy of a lesion with mammographic calcification was altered in 13% of lesions following examination of additional biopsies sampled at other levels, which included 12% of biopsies sampled by core needle biopsy (CNB) and 13% of those sampled by vacuum-assisted biopsy (VAB). Inter- and intraobserver agreements were comparable (κ=0.91 and κ=0.947, respectively) and the rate of diagnostic discrepancy was comparable for lesions identified due to symptoms and screening.
Studies (1)
Purpose
The purpose of this study was to determine whether diagnosis is influenced by the number and position of biopsies obtained from breast lesions, and to assess intra- and inter- observer diagnostic agreement. A minimum of three biopsies were sampled from each patient at different levels of the breast lesion. Biopsies were obtained from 272 lesions without mammographic calcification by CNB (266 lesions) or by VAB (6 lesions), which included those sampled by ultrasound (240 lesions), stereotactic (6 lesions), or unspecified (13 lesions) guidance or without guidance (13 lesions). Biopsies were also obtained from 103 lesions with mammographic calcification by CNB (40 lesions) or by VAB (63 lesions), which included those sampled by ultrasound (24 lesions), stereotactic (75 lesions), or unspecified (3 lesions) guidance or without guidance (1 lesion). Axillary node CNBs were also obtained from 40 patients using ultrasound guidance. An initial diagnosis was assigned based upon the first biopsy. A second diagnosis was then determined in consideration of all 3 biopsies. One pathologist examined all biopsies and each case was then blindly diagnosed by 1 of 4 pathologists.
Summary of Findings
Patient diagnosis, based on a single biopsy, was only altered in 1.5% of cases (4 of 272) following examination of an additional two biopsies collected from different levels of non-calcified lesions. These changes included a diagnostic upgrade from atypical intraductal proliferation to suspicion of ductal carcinoma in situ (DCIS) in one case. In another case, the initial biopsy was too shallow for diagnosis but biopsies sampled at additional levels displayed evidence of a fibroadenoma. The third case of discordance was upgraded from normal to a lesion of uncertain malignant potential with epithelial atypia that subsequently developed in to invasive carcinoma. The final case displayed evidence of DCIS in the initial biopsy but focal invasive carcinoma in the biopsy obtained from the third level.
In contrast, patient diagnosis based on a single biopsy of lesions with mammographic calcification was altered in 13% cases (13 of 103) following examination of an additional two levels. Cases that displayed discordance among calcified lesions included 5 of 40 (12%) CNB specimens and 8 of 63 (13%) VAB specimens. An additional 11 lesions required re-biopsy as no calcification or malignancy was observed in the first 3 planes. For 8 of these cases biopsying at additional levels resulted in a diagnostic change from normal to benign (2 cases) ductal carcinoma in situ (DCIS) (1 case), ora lesion of uncertain malignant potential with epithelial atypia (2 cases); from benign to a lesion of uncertain malignant potential with epithelial atypia (one case); and from a lesion of uncertain malignant potential with epithelial atypia to ductal carcinoma in situ (DCIS) (2 cases).
Examination of all three biopsies sampled from different levels of axillary nodes altered the diagnosis of only 2.5% (1 of 40) cases, in which the initial biopsy displayed small fragments of fibrofatty tissue while the additional biopsies sampled at different levels displayed evidence of metastasis.
Finally, inter- and intraobserver agreements were comparable (κ=0.91 and κ=0.947, respectively) and rate of diagnostic discrepancy was similar between lesions identified as a result of symptoms and those detected through screening.
Biospecimens
TissueBreastPreservation Types
FormalinDiagnoses
NormalNeoplasticBenignNeoplasticCarcinomaPlatforms
MorphologyLight microscopyMorphologyH-and-E microscopyPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Preaquisition Diagnosis/ patient condition BenignDCISInvasive carcinomaNormalUncertain malignant potential with epithelial atypiaPreaquisition Prognostic factor Mammographic calcification absentMammographic calcification presentScreening-detected lesionSymptomatic lesionBiospecimen Acquisition Method of tissue acquisition Core needle biopsyVacuum assisted biopsyBiospecimen Aliquots and Components Biospecimen heterogeneity Intratumoral sampling (exact positions not specified)