Challenges in pathologic staging of renal cell carcinoma: an interobserver variability study of urologic pathologists
Williamson SR, Rao P, Hes O, Epstein J, Picken M, Zhou M, Tretiakova M, Tickoo S, Chen YB, Reuter V, Fleming S, Maclean F, Gupta N, Kuroda N, Delahunt B, Mehra R, Przybycin C, Cheng L, Eble J, Grignon D, Moch H, Lopez J, Kunju LP, Tamboli P, Srigley J, Amin M, Martignoni G, Hirsch MS, and Trpkov K. Challenges in pathologic staging of renal cell carcinoma: an interobserver variability study of urologic pathologists. Lab Invest 2018; 98:399.
Background: Staging criteria for renal sinus, perinephric fat, and sinus vein invasion were defined more precisely at the ISUP Consensus Conference in 2012 (Am J Surg Pathol 2013;37:1505-17). However, renal cell carcinoma differs from many other cancers, in that tumors are often spherical with subtle tongue-like extensions into veins, renal sinus, or perinephric tissue, in contrast to infiltrative and desmoplastic growth prototypical of cancer. We sought to study whether urologic pathologists have uniform criteria for assigning pathologic stage categories in this setting. Design: An online survey was constructed and circulated to urologic pathologists with interest in kidney tumors, aiming to evaluate interobserver agreement, emphasizing challenging or borderline cases for renal staging. Response rate was 86% (30/35) of the invited pathologists with broad geographic distribution. Most questions included 1-4 images, divided into categories of: vascular and renal sinus invasion (n=24), perinephric tissue invasion (n=9), and gross pathology/specimen handling (n=17). Responses were collapsed for analysis into positive and negative for upstaging. Equivocal responses were included in the latter. Consensus was regarded as 67% (2/3) of participants. Results: Consensus was reached in 16/24 (67%) questions for sinus and vascular invasion (13 positive and 3 negative for pT3a, with 80% or greater consensus in 12), and 4/9 (44%, with 3 at 80% or greater consensus) for perinephric invasion. Lack of agreement was especially encountered regarding small tumor protrusions into possible vascular lumina, close to the tumor leading edge or within the tumor. For gross photographs, the most common response was that findings were “suspicious for venous invasion,” but required histologic confirmation. Most participants (60%) rarely used special stains to evaluate vascular invasion, usually endothelial markers (80%). Most agreed that a spherical mass bulging well beyond the kidney parenchyma into the renal sinus (70%) or perinephric fat (90%) did not necessarily indicate invasion. Criteria for assessment of biopsy site artifact were also incompletely agreed upon. Conclusions: Interobserver agreement in pathologic staging of renal cancer is relatively good among urologic pathologists interested in kidney tumors, even when selecting cases that test the earliest thresholds for extrarenal extension. Disagreements however remain, particularly for tumors with small, finger-like protrusions, closely juxtaposed to the main mass.