The efficiency of MarginProbe in detecting positive resection margins in epithelial breast cancer following breast conserving surgery

Abstract

Introduction

Up to 30% of patients undergoing lumpectomy for breast cancer are in need for re-excision due to either close or involved margins. We evaluated the yield of Margin Probe© (MP; Dilon Technologies, USA) in a cohort of patients undergoing lumpectomy for both palpable and non-palpable tumors.

Methods

Following lumpectomy, margins were evaluated with MP and readings were compared to the lumpectomies' pathological evaluation irrespective of whether additional margins were removed during surgery. Involved margins or the presence of DCIS within 1 mm of the resection margins were considered as positive margins.

Results

48 patients with 51 tumors underwent lumpectomy. Thirteen of the 51 lumpectomies had pathological close or involved margins. MP identified 3 out of the 13 positive margins. False-positive readings were recorded in 97 out of 287 margins. The sensitivity, specificity, positive predictive value and negative predictive value were 23.1% (95%CI 5.0% 53.8%), 66.4% (95%CI 60.7%–71.9%), 3% (95%CI 0.6%–8.5%), and 95.1% (95%CI 91.1%–97.6%) respectively.

Conclusions

MP cancer detection rate is relatively low while high false-positive rate leads to unnecessary shavings in almost all patients. Evaluation of MP performance should be based on comparing MP read to pathology report.