Clinical Implications of Diagnosing Intraductal Papillary Lesions of the Breast by General Pathologists vs. Specialists in Breast Pathology
Joshua Warrick, Susan Holley, Catherine Appleton, Barbara Monsees, D Craig Allred. Washington University, St. Louis, MO
Background: There is considerable debate on the need for specialists in breast pathology. This study compared the accuracy and clinical implications of diagnosing intraductal papillary lesions on core needle biopsies (CNB) - a particularly challenging setting - between experienced general surgical pathologists (GPs) and a specialist in breast pathology. It also set out to determine the impact of providing specialized training to GPs.
Design: A breast specialist reviewed (blinded) a recent consecutive series (n=135) of intraductal papillomas (IDPs) diagnosed on CNBs by a group of GPs, as well as all follow-up excisional biopsies (FU EXBs; 78 cases = 59% of total). Diagnoses on CNBs by GPs included IDP (97%) and atypical IDP (3%). The primary study endpoints were concordance of diagnoses between GPs versus the specialist, and accuracy of CNB diagnoses made by GPs and the specialist for predicting EXB diagnosis (assuming the specialist's EXB diagnosis was correct).
Results: Diagnoses were concordant between GPs and the specialist in 87% of CNBs and 87% of EXBs. Relative to the specialist, GPs made 11% false-negative diagnoses on CNBs (IDP to DCIS), 3% false-negative diagnoses on EXBs (IDP to DCIS), and 10% false-positive diagnoses on EXBs (ADH/DCIS to IDP). The accuracy of CNB for predicting EXB diagnosis was 88% for GPs and 96% for the specialist (one IDP upgraded to DCIS on EXB). One GP (senior resident) who originally reviewed all cases with the specialist, is in the process of re-reviewing them alone in a blinded manner, and the impact of this training exercise is being evaluated.
Conclusions: Previous studies suggest that 10-15% of IDPs diagnosed on CNBs by general pathologists are up-graded to cancer (primarily DCIS) on FU EXBs (Cancer (2009)15:2837; Eur J Surg Oncol (2008)34:1304), which is upheld in this study. This study also shows that errors (primarily false-positives) of similar magnitude are being made by generalists on FU EXBs. All of these errors could lead to highly unfavorable clinical outcomes. Relying on specialists in breast pathology would largely eliminate these dangers, as well as dramatically reduce the need for FU EXBs. Giving GPs specialized training in the evaluation of papillary breast lesions could also potentially accomplish this goal.
Monday, March 19, 2012 9:30 AM
Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 34, Monday Morning