Objectives. To study the diagnostic performance of the ImmunoCyt test in pa
tients in follow-up for superficial urothelial cell carcinoma (UCC) of the
bladder.
Methods. Voided urine samples were collected from all included patients. Sa
mples were processed with the ImmunoCyt test. The ImmunoCyt slides were sco
red under a fluorescence microscope by 3 observers. The ImmunoCyt test was
considered positive if one or more observers scored the test positive. Uret
hrocystoscopy (and additional histologic examination in the case of suspici
ous cystoscopic findings) was used as the reference standard. To investigat
e the validity of ImmunoCyt, sensitivity, specificity, positive predictive
value, negative predictive value, area under the receiver operating charact
eristic curve, and diagnostic odds ratios were determined. To investigate t
he reproducibility of ImmunoCyt, kappa values (measure of agreement) were c
omputed. The observers' findings were analyzed in pairs.
Results. One hundred four patients in follow-up after primary superficial U
CC of the bladder were included. Samples of 18 patients had to be excluded
because of low cellularity (ie, insufficient assessable urothelial cells).
Tumor recurrence was found in 22 of the remaining 86 patients (17 pTa, 3 pT
1, 1 carcinoma in situ, 1 pT2 or higher). The test had a sensitivity of 50%
, specificity of 73%, positive predictive value of 39%, and negative predic
tive value of 81%. The diagnostic odds ratio was 2.8 (95% confidence interv
al 1.0 to 7.5). The area under the curve for the different observers varied
between 0.54 and 0.60. The kappa values were low (0.05 to 0.45), represent
ing high interobserver variability.
Conclusions. The promising results from other studies could not be confirme
d in this specific group of patients in follow-up for superficial UCC of th
e bladder. The validity of ImmunoCyt was insufficient to justify the omissi
on of cystoscopy in patients in follow-up for superficial UCC. UROLOGY 58:
367-371, 2001. (C) 2001, Elsevier Science Inc.