Diagnostic discrepancies between second-opinion and referring pathology reports of neuroendocrine tumors.

Abstract

154 Background: Neuroendocrine tumors (NETs) are rare malignancies that can arise from neuroendocrine cells throughout the body, most commonly in the lungs, gastrointestinal tract, and pancreas. The nomenclature and histologic classification schemes for NETs have historically been heterogeneous and inconsistent. Klimstra, et al published a set of NET pathology guidelines in 2010. We undertook a systematic evaluation of discrepancies between referring versus second-opinion NET pathology reports using these guidelines as our reference. Methods: We developed a cohort ofall NET cases seen at Stanford University Hospital between April 1998 and February 2012 with available Stanford and referring pathology reports for the same specimen to identify the discrepancies between their clinical diagnoses. Results: Of 141 cases, primary sites were identified as 39 (28%) pancreas, 21 (15%) small intestine, 49 (35%) other NET sites, and 32 (22%) unknown or missing. Most reports agreed on a diagnosis of NET, although twelve different terms were used to describe the NET diagnosis. There were 24 cases (17%) with major discrepancies by histologic description (NET vs. not NET). Grade, mitotic index (MI), and Ki67 were among the variables examined; they were missing from the majority of cases in one or both reports, yet more likely to be included in the Stanford report. Grade was not reported (NR) in one or both reports in 124 cases (88%); of the 17 cases reporting grade in both reports, 3 had major discrepancies. MI was NR in one or both reports in 105 cases (74%); of the 36 cases reporting MI in both reports, 5 had major discrepancies. Ki67 was NR in one or both reports in 127 cases (90%); of the 14 cases reporting Ki67 in both reports, 3 had major discrepancies. Conclusions: Clinically relevant differences were frequently found between Stanford and referring pathology reports for NETs. Our results suggest that it is beneficial for NET cases to be reviewed by NET pathology experts given the potential to impact treatment. Future studies of NET pathology discrepancies are warranted to allow additional time for adoption of the 2010 guidelines.