Categorizing Renal Oncocytic Neoplasms on Core Needle Biopsy: a Morphologic and Immunophenotypic Study of 144 Cases with Clinical Follow-Up

Summary There is limited literature on renal oncocytic neoplasms diagnosed on core biopsy. All renal oncocytic neoplasm core biopsies from 2006-2013 were retrospectively reviewed. Morphologic features and an immunohistochemical panel of CK7, c-KIT, and S100A1 were assessed. Concordance with resectio...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Human pathology 2016-09, Vol.55, p.1-10
Hauptverfasser: Alderman, Megan A., MD, Daignault, Stephanie, MS, Wolf, J. Stuart, MD, Palapattu, Ganesh S., MD, Weizer, Alon Z., MD, Hafez, Khaled S., MB, BCh, Kunju, Lakshmi P., MD, Wu, Angela J., MD
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Summary There is limited literature on renal oncocytic neoplasms diagnosed on core biopsy. All renal oncocytic neoplasm core biopsies from 2006-2013 were retrospectively reviewed. Morphologic features and an immunohistochemical panel of CK7, c-KIT, and S100A1 were assessed. Concordance with resection diagnosis, statistical analysis including a random forest classification, and followup were recorded. The post immunohistochemical diagnoses of 144 renal oncocytic core biopsies were: favor oncocytoma (67%); favor renal cell carcinoma (RCC) (12%); cannot exclude RCC (21%). Diagnosis was revised following immunohistochemistry in 7% of cases. The most common features for oncocytoma (excluding dense granular cytoplasm) were nested architecture, edematous stroma, binucleation and tubular architecture; the most common features for favor RCC were sheet-like architecture, nuclear pleomorphism, papillary architecture, and prominent cell borders. High nuclear grade, necrosis, extensive papillary architecture , raisinoid nuclei , and frequent mitoses were not seen in oncocytomas. Comparing the pathologist and random forest classification, the overall out of bag estimate of classification error dropped from 23% to 13% when favor RCC and cannot exclude RCC were combined into one category. Resection was performed in 19% (28 cases) with a 94% concordance (100% of favor RCC biopsies and 90% of cannot exclude RCC biopsies confirmed as RCC; 83% of favor oncocytomas confirmed); ablation in 23%; surveillance in 46%. Follow-up was available in 92% (median follow-up 33 months) with no adverse outcomes. Renal oncocytic neoplasms comprise a significant subset (16%) of all core biopsies and the majority (78%) can be classified as favor oncocytoma or favor RCC.
ISSN:0046-8177
1532-8392
DOI:10.1016/j.humpath.2016.03.017