Correlation of Cystoscopy With Histology of Recurrent Papillary Tumors of the Bladder

0022-5347/02/1683-0978/0 THE JOURNAL OF UROLOGY® Copyright © 2002 by AMERICAN UROLOGICAL ASSOCIATION, INC.®

Vol. 168, 978 –980, September 2002 Printed in U.S.A.

DOI: 10.1097/01.ju.0000025867.14412.2d

CORRELATION OF CYSTOSCOPY WITH HISTOLOGY OF RECURRENT PAPILLARY TUMORS OF THE BLADDER HARRY W. HERR, S. MACHELE DONAT

AND

GUIDO DALBAGNI

From the Department of Urology, Memorial Sloan-Kettering Cancer Center, New York, New York

ABSTRACT

Purpose: We correlated individual urologist impressions of tumor stage and grade of recurrent papillary bladder tumors at cystoscopy with histological findings after transurethral resection to determine whether cystoscopy can reliably identify low grade, noninvasive papillary tumor for outpatient fulguration. Materials and Methods: A total of 144 recurrent papillary bladder tumors identified on outpatient flexible cystoscopy were classified as low grade and noninvasive (stage Ta grade 1), high grade and noninvasive (stage Ta grade 3) or invasive (stage T1). Voided urine cytology was also performed. The cystoscopic impression of each tumor was correlated with the final histological findings of tumor stage and grade after transurethral resection. Results: Cystoscopy classified 97 tumors as stage Ta grade 1 and 47 as stage Ta grade 3 or stage T1. Cystoscopy correctly predicted the tumor stage and grade of 93% of stage Ta grade 1 and 99% of stage Ta grade 1 lesions associated with negative urine cytology. Conclusions: Urologists can usually identify noninvasive, low grade recurrent papillary tumors on followup cystoscopy that may be treated safely with outpatient fulguration. KEY WORDS: bladder; bladder neoplasms; cystoscopy; carcinoma, papillary; histology

Outpatient flexible cystoscopy is used to follow patients with superficial bladder tumors. Recurrent tumors are usually treated with transurethral resection, although fulguration alone may destroy small papillary tumors.1 In patients with known bladder tumors the American Urological Association Clinical Guidelines Panel on Nonmuscle Invasive Bladder Cancer recommends that endoscopic ablation of recurrent tumors should be done with electrocautery resection, fulguration or laser energy.2 Recurrent papillary tumors discovered during outpatient cystoscopy may not be biopsied before they are fulgurated.1 Treatment with fulguration demands that the urologist should visually distinguish low grade, noninvasive (low risk) tumors from high grade or invasive (high risk) tumors. The lack of histological findings raises concerns of tumor under staging and perhaps inadequate tumor ablation. Despite current practice guidelines and reports of successful outpatient fulguration of recurrent bladder tumors3–7 to our knowledge the cystoscopic impression of papillary tumors has not been correlated with histological findings after resection or biopsy in a prospective study. We correlated the cystoscopic appearance of recurrent papillary tumors of the bladder with histological results after transurethral resection to ascertain whether cystoscopy can reliably identify low grade, noninvasive papillary tumors suitable for outpatient fulguration. PATIENTS AND METHODS

A total of 144 recurrent papillary tumors of the bladder were evaluated in 125 patients undergoing surveillance outpatient cystoscopy between January and December 2001. All patients had a history of nonmuscle invasive papillary bladder tumors and had had at least 1 tumor recurrence. None had had carcinoma in situ previously or was being evaluated for the response to intravesical therapy. In the 19 patients with more than 1 tumor recurrent disease was evaluated at separate cystoscopy sessions at least 3 months apart. A voided urine specimen was obtained before cystoscopy for Accepted for publication April 19, 2002.

cytological examination. All recurrent tumors were subsequently excised completely by transurethral resection with the patient under general anesthesia using a cautery loop and video camera.8 The appearance of the recurrent papillary tumor at cystoscopy was correlated with the histological findings of tumor stage (noninvasive versus invasive), grade (low versus high) and the results of voided urine cytology for neoplastic cells (positive versus negative). In all patients in whom a stage T1 lesion was resected muscle was identified in the biopsy specimen. Patients with a history of muscle invasive tumors or flat carcinoma in situ of the bladder were excluded from analysis. Each recurrent tumor was classified independently by the urologist as papillary stage Ta grade 1, stage Ta grade 3 or stage T1 based on the cystoscopic impression. These data were recorded before histological findings in the transurethral resection specimen were available. Half of the cases was performed by 1 urologist (H. H.) and the other half was performed by 2 urologists (M. D. and G. D.). Transurethral resection specimens were evaluated by multiple experienced pathologists blinded to cystoscopic appearance and individual urologist impressions. Tumors were classified according to the recent WHO/International Society of Urological Pathology consensus classification of transitional cell neoplasms of the bladder.9 Superficial papillary urothelial lesions were diagnosed as papilloma, papillary urothelial neoplasms of low malignant potential, papillary carcinoma, low grade (stage Ta grade 1), papillary carcinoma, high grade (stage Ta grade 3) and papillary carcinoma with lamina propria invasion. Tumors were graded as low (grade 1) or high (grade 3). RESULTS

The table shows the results of urologist impressions of tumor stage and grade at cystoscopy correlated with the histology of recurrent papillary bladder neoplasms. Of the 144 recurrent papillary tumors identified by cystoscopy 97, 37 and 10 were classified via cystoscopy as stage Ta grade 1, stage Ta grade 3 and stage T1, respectively. Biopsy of the 97

978

CORRELATION OF CYSTOSCOPY WITH HISTOLOGY OF BLADDER TUMORS Correlation of cystoscopic appearance with histological findings in recurrent papillary bladder tumors Tumor on Cystoscopy

No. Tumors

Stage Ta grade 1: 97 Neg. cytology 86 Pos. cytology 11 Stage Ta grade 3: 37 Neg. cytology 3 Pos. cytology 34 Stage T1: 10 Neg. cytology 0 Pos. cytology 10 * Focal muscle invasion in 1 tumor.

No. Stage/Grade Ta/1

Ta/3

T1

90 85 5 7 1 6 0

6 1 5 22 1 21 4

1 0 1 8 1 7 6

0

4

6*

papillary tumors believed to be stage Ta grade 1 showed that 90 (93%) were noninvasive and low grade papillary neoplasms, including 3 papillomas, 9 papillary urothelial neoplasms of low malignant potential and 5 papillary hyperplasia lesions with atypia. Six (6%) of the tumors believed to be noninvasive and low grade were stage Ta grade 3 and 1 (1%) showed focal lamina propria invasion (stage T1). Of the 37 stage Ta grade 3 tumors at cystoscopy 22 (59%) were in fact noninvasive and high grade, whereas 7 (19%) were low grade and 8 (22%) were invasive. Of the 10 tumors believed at cystoscopy to be invasive 5 were confirmed to be stage T1, 1 involved focal muscle invasion and 4 were stage Ta grade 3. Papillary tumors classified at cystoscopy as stage Ta grade 1 had individually discrete papillary fronds of mucosa surrounding a clearly visible fibrovascular core and were generally less than 0.5 cm. Papillary lesions suspected of being stage Ta grade 3 had papillary fronds that were fused and less discrete than low grade tumors. On cystoscopy stage T1 neoplasms generally appeared papillary-nodular or solid and less well defined than purely papillary tumors. Urine cytology was positive in 11 of the 97 tumors considered at cystoscopy to be stage Ta grade 1, including all except 1 high grade or invasive lesion. Of the 86 recurrent stage Ta grade 1 tumors associated with negative urine cytology 85 (99%) proved to be low grade papillary tumors histologically, including 1 that was focal high grade. A total of 15 tumors were stage T1 at biopsy and all except 1 were associated with positive urine cytology. Of the 134 recurrent papillary tumors considered at cystoscopy to be noninvasive 9 (7%) showed invasion of the lamina propria, including 1 of the 97 (1%) stage Ta grade 1 and 8 (22%) of the 37 stage Ta grade 3 neoplasms. DISCUSSION

Given the prevalence of recurrent papillary bladder tumors there is a remarkable lack of data correlating their cystoscopic appearance with histological findings. A recent study showed that low grade neoplasms could not be distinguished from high grade carcinoma and invasion could not be accurately predicted at cystoscopy.10 Another study comparing the accuracy of transurethral resection for determining pathological tumor stage and grade in cystectomy specimens showed that more than half of tumors staged clinically as confined to the mucosa infiltrated the lamina propria.11 These groups concluded that the grade of a papillary urothelial neoplasm and the presence of invasion cannot be accurately predicted by the cystoscopic appearance of the lesion and they suggested that histological evaluation of tumor biopsies was essential. Based on the cystoscopic appearance of recurrent papillary tumors in a population of patients with known superficial bladder tumors our results show that 94% of noninvasive and 93% of low grade neoplasms were correctly identified. When considering negative urine cytology, 99% of low grade and noninvasive papillary tumors were correctly identified at followup outpatient cystoscopy.

979

What explains the excellent correlation of cystoscopic appearance with histological findings of papillary bladder tumors? Is the correlation good enough to justify outpatient fulguration only of recurrent tumors? All of our current patients had a history of superficial bladder tumors and were undergoing routine surveillance for recurrent disease. Recurrent low grade papillary tumors generally remain confined to the mucosa (stage Ta) and tend not to progress in stage or grade. Low grade recurrent papillary lesions appear relatively benign on followup cytoscopy and they are also usually small and delicate with clearly visible individual papillary fronds. Cystoscopy was also sensitive to change in the appearance of recurrent tumors. Of the 19 patients in whom more than 1 recurrent tumor was evaluated 14 had recurrent stage Ta grade 1 neoplasms that were similar to previous recurrences, while 5 who had had stage Ta grade 1 tumors showed recurrent neoplasms that appeared larger and more solid (less papillary) than the earlier tumors. All 5 lesions were stage Ta grade 3 on biopsy. How do we use the results of this study in clinical practice? Since we achieved excellent concordance of our cystoscopic impression and with histological findings of recurrent papillary bladder tumors, we believe that we can safely fulgurate select recurrent tumors when they are discovered during outpatient flexible cystoscopy. The fact that 3 urologists independently predicted final histological results based on their cystoscopic impression in almost all cases suggests that most urologists should be able to reproduce our results and apply them clinically. Our study is also strengthened by the fact that tumors were evaluated by different pathologists assigned to process and read submitted histological data on that day and not by a dedicated pathologist. We fulgurate recurrent tumors if they are papillary, small (usually less than 0.5 cm.), few in number (usually less than 5) and can be completely destroyed by electrocautery. If there is doubt as to the nature of the tumor, excisional biopsy may be performed before the base of the tumor is fulgurated.12 When urine cytology is positive, the patient usually undergoes formal transurethral resection of the tumor site as well as biopsy of any other suspicious mucosal areas. Alternatively in some cases we may repeat cystoscopy in 1 or 2 months as well as urine cytology. If the results suggest evidence of tumor recurrence or persistence, transurethral resection is repeated. In the majority of patients in whom fulguration is done to treat recurrent papillary bladder tumors it is successful and the patient can be followed thereafter every 3 to 6 months with outpatient flexible cystoscopy. In conclusion, the appearance of recurrent papillary bladder tumors identified during followup by flexible cystoscopy combined with urine cytology justifies the treatment of select papillary tumors that are small, few in number and capable of complete destruction by fulguration alone. REFERENCES

1. Herr, H. W.: Outpatient flexible cystoscopy with fulguration in the management of recurrent bladder tumors. J Urol, 144: 1365, 1990 2. Smith, J. A., Labasky, J. F., Cockett, A. T., Fracchia, J. A., Montie, J. E. and Rowland, R. G.: Bladder cancer clinical guidelines panel summary report on the management of nonmuscle invasive bladder cancer (stages Ta, T1 and TIS). The American Urological Association. J Urol, 162: 1697, 1999 3. German, K., Hasan, S. T. and Derry, C.: Cystodiathermy under local anesthesia using the flexible cystoscope. Br J Urol, 69: 518, 1992 4. Fowler, C. G. and Boorman, L. S.: Outpatient treatment of superficial bladder cancer. Lancet, 1: 38, 1986 5. Wedderburn, A. W., Ratan, P. and Birch, B. R.: A prospective trial of flexible cystodiathermy for recurrent transitional cell carcinoma of the bladder. J Urol, 161: 812, 1999 6. Kerbl, K. and Clayman, R. V.: Treatment of multifocal superficial transitional cell carcinoma of the bladder using roller ball

980

CORRELATION OF CYSTOSCOPY WITH HISTOLOGY OF BLADDER TUMORS

electrovaporization. J Urol, 165: 141: 2001 7. Dryhurst, D. J. and Fowler, C. G.: Flexible cystodiathermy can be rendered painless by using 2% lignocaine solution to provide intravesical anesthesia. BJU Int, 88: 437, 2001 8. Herr, H. W. and Reuter, V. E.: Evaluation of new resectoscope loop for transurethral resection of bladder tumors. J Urol, 159: 2067, 1998 9. Epstein, J. I., Amin, M. B., Reuter, V. E. and Mostofi, F. K.: The World Health Organization/International Society of Urological Pathology consensus classification of urothelial (transitional cell) neoplasms of the urinary bladder. Bladder Consensus Conference Committee. Am J Surg, 22: 1435, 1998 10. Endrizzi, J., Cina, S., Harmon, W., Seay, T., Schoenberg, M. and

Epstein, J.: Correlation of cystoscopic impression and histologic diagnosis on biopsy specimens of the bladder. J Urol, suppl., 163: 134, abstract 588, 2000 11. Cheng, L., Neumann, R. M., Weaver, A. L., Cheville, J. C., Leibovich, B. C., Ramnani, D. M. et al: Grading and staging bladder carcinoma in transurethral resection specimens: correlation with 105 matched cystectomy specimens. Am J Clin Pathol, 113: 275, 2000 12. Muraishi, O., Mitsu, S., Suzuki, K., Koshimizu, T. and Tukue, A.: A technique for resection of small bladder tumors using a flexible cystoscope on an outpatient basis: bladder tumor resection with newly designed hot cup forceps. J Urol, 166: 1817, 2001