Objectives. To assess the usefulness of p53 staining of cytology speci
mens obtained ureteroscopically in the diagnosis of upper-tract transi
tional cell carcinoma (TCC). Methods. We collected specimens from 43 p
atients undergoing a total of 50 ureteroscopic procedures for a variet
y of indications, including the diagnosis of TCC. Specimens were obtai
ned by direct biopsy with forceps or basket whenever possible. We exam
ined specimens for evidence of TCC by cytospin as well as cell block f
or any visible fragments, Each specimen was then stained for overexpre
ssion of p53 by immunohistochemical staining, and the degree of staini
ng was graded. Eight patients subsequently underwent nephroureterectom
y; the pathologic specimens were stained for p53 and compared with the
cytology results. Results. Staining for p53 was positive in specimens
from 36 of 50 procedures, including all 28 with ureteroscopic or cyto
logic evidence of TCC (P < 0.0001). By contrast, cytology accurately d
iagnosed only 23 of the 28. Specimens from all 14 procedures that were
negative also stained negative for p53. Specimens from all 8 procedur
es with no tumor seen and atypical cytology stained positive for p53;
4 of 5 patients with adequate follow-up have had a tissue diagnosis of
TCC at that site. Overall, 35 of the 36 specimens that stained positi
ve for p53 were obtained from patients with some history of TCC (P < 0
.0001). No significant association could be found between degree of st
aining and grade (P = 0.3034). The degree of staining of ureteroscopic
biopsy specimens was identical to that of the nephroureterectomy spec
imen in 6 of 8 cases. Conclusions. p53 nuclear protein staining of cyt
ology specimens obtained ureteroscopically appears to correlate well w
ith the presence of upper-tract urinary TCC. Further study is needed t
o determine if it can provide a definitive diagnosis in cases with ind
eterminate cytologic findings. Copyright 1997 by Elsevier Science Inc.