We performed a retrospective analysis in 100 consecutive women with invasive breast cancer who underwent US guided core needle biopsy and subsequent surgery in the breast diagnostic centre of Brotzu Hospital (Cagliari,
Italy).
Only the patients with definitive histological diagnosis obtained after surgical treatment were included.
DCIS cases were excluded.
In patients with multiple breast masses,
only the largest lesion was considered.
The ultrasound examination was performed with a Siemens Antares unit (Siemens Medical Solutions,
Sweden) using a high-resolution (10- to 13-MHz) linear array transducer.
We analyzed the hard copy of ultrasound examination performed during the US-guided biopsy procedure in order to study morphological features of the lesion.
Two experienced breast radiologists (MC and EF who had 15 and 20 years of breast imaging experience) independently evaluated the US findings.
These two investigators were blinded to the clinical history or pathologic results.
If there were disagreement in the interpretation of the images,
a final decision was reached using consensus evaluation.
Tumor characteristics were assessed using the ACR-BIRADS Ultrasound lexicon [3].
The characteristics considered were shape (round,
oval,
irregular),
margin (circumscribed,
indistinct,
angular,
microlobulated,
spiculated),
boundary (abrupt interface,
echogenic halo),
echo pattern (hypoechoic,
complex,
isoechoic,
hyperechoic),
posterior acoustic feature (no posterior acoustic feature,
shadowing or enhancement or combined pattern).
For echo patter descriptor we have divided the hypoechoic lesions into two groups: markedly hypoechoic lesion and inhomogeneous/middle hypoechoic lesions as Stavros suggests [2].
Each lesion was examined by the same pathologist who assessed histology and tumor grading.
For histological examination tissue were fixed in 10% buffered formalin and embedded in paraffin.
Section were stained with haematoxilin-eosin stain and if necessary or (indicated) with ck 14,
p63,
calponin,
a-sma and E-cadherin stain [13].
The grading was assessed by the Nottingham modification of the Bloom-Richardson grading system in which were evaluated three cancer’s features: the percentage of cancer composed of tubular structures,
nuclear pleomorphism and mitotic count [14].
Estrogen and progesterone receptor status was identified using immunoistochemical stains.
ER and PR were scored positive if more than 10% of tumor cells were immunreactive by evaluation of 10 random microscopic fields comprising at least 1000 number of cells [15].
The HER2 status was tested by immustoistochemical stains (Hercept Test,
Dako).
HER2 status was graded 0,
1+,
2+,
and 3+.
3+ was determined as positive [16].
Statistical analysis was made.