101 Background: Contemporary breast surgery is exploring less invasive techniques for the treatment of breast cancer and recent approaches such as ablative therapy have shown promise in early trials. Caution is required prior to adopting of these techniques, however, as ablative treatment may hinder interpretation of final pathology, which might impact adjuvant recommendations. To address this concern, we evaluated the concordance between preoperative core needle biopsy and postoperative final pathology, to determine whether the biopsy findings predicted final pathology. Methods: A retrospective chart review was performed of women diagnosed with breast cancers less than 2 cm on radiographic evaluation between 1997 and 2004. Concordance between preoperative assessment (by imaging and core needle biopsy) and final pathology results was analyzed based on tumor size, grade, lymphovascular space invasion, and in situ component. Kappa analysis was performed to characterize agreement. Results: We identified 56 women from our institution for analysis. The median age was 56 (range, 38-94) and the majority had presented with an abnormal mammogram (61%). The vast majority also underwent breast-conserving surgery (95%). Agreement was moderately accurate between preoperative and final assessment of grade (weighted κ 0.536) and size (weighted κ 0.464). It was not accurate in the assessment of lymphovascular invasion (weighted κ 0.123) and the presence of DCIS (weighted κ 0.197). Conclusions: Information from preoperative evaluation is insufficiently accurate for the evaluation of breast tumors, which may be important for adjuvant decision making. This information must be taken in to account when assessing new technologies, which may impact on the assessment of histologic criteria in excised breast cancers.
The impact of inter-observer variation in pathological assessment of node-negative breast cancer on clinical risk assessment and patient selection for adjuvant systemic treatment