308 Consult Over-reads of Breast Pathology Uncover Major Discrepancies. (11th January 2018)

• Record Type: Journal Article
• Authors: Roman, Joy; Factor, Rachel
• Is Part Of: American journal of clinical pathology. Volume 149(2018)Supplement 1
• Page Start: S132
• Page End: S133
• Publication Date: 2018-01-11
• Subjects: Diagnosis, Laboratory -- Periodicals; Pathology -- Periodicals; 616.07
• DOI: 10.1093/ajcp/aqx125.307
• Languages: English
• ISSNs: 0002-9173

Abstract

Abstract: Introduction: Our institution is a tertiary care comprehensive cancer center. Many patients who present for oncologic care have had initial pathology performed at community hospitals. Our policy is to review outside pathology prior to treatment. In this study, we evaluated diagnostic agreement for breast pathology cases. Methods: A non-chartable comment is included in our pathology over-reads with a numeric score (1 = agreement; 2 = minor disagreement, no clinical significance; 3 = disagreement of uncertain clinical significance; and 4 = major disagreement, clinically significant). We reviewed the ratings for consecutive breast pathology over-reads from July 2013 to March 2017. In some cases, where a block was obtained for additional work-up, the agreement was re-scored by a QA medical director following the final results. Results: There were 1, 376 cases with rating scores. A proportion (20%) of these cases sent ancillary study slides for review. The ratings were 1 (73%), 2 (17%), 3 (8%), and 4 (2%). Reasons for disagreements in all categories included size (9%) and grade (25%) of invasive carcinoma, Her2 status (15%), ER/PR interpretation (10%), the presence of DCIS (13%), margin status (8%), the presence of lymphovascular invasion (8%), and lymph node status (5%). Most reasons for major discrepancies (25 cases) included Her2 interpretation (40%), the presence of invasion (32%), and the presence of DCIS (24%). We received slides to evaluate Her2 immunohistochemistry for 279 cases. For 61 cases (22%), we disagreed with the outside interpretation. For 31 cases Her2 was re-evaluated by FISH at our institution. The agreement score changed for 27 of the 31 cases. Of the changed ratings, 59% were re-rated 1, and 33% were re-rated 4. Conclusions: Over-reads show agreement with the majority of cases, but clinically significant discrepancies are found, including the presence of invasive carcinoma and differences in Her2 interpretation. These major discrepancies support the continued use of consult case over-reads at our institution.