Transurethral Resection For Bladder Cancer Using 5-Aminolevulinic Acid Induced Fluorescence Endoscopy Versus White Light Endoscopy

0022-5347/02/1682-0475/0 THE JOURNAL OF UROLOGY® Copyright © 2002 by AMERICAN UROLOGICAL ASSOCIATION, INC.®

Vol. 168, 475– 478, August 2002 Printed in U.S.A.

TRANSURETHRAL RESECTION FOR BLADDER CANCER USING 5-AMINOLEVULINIC ACID INDUCED FLUORESCENCE ENDOSCOPY VERSUS WHITE LIGHT ENDOSCOPY MARTIN KRIEGMAIR,* DIRK ZAAK,* KARL-HEINZ ROTHENBERGER, JENS RASSWEILER, DIETER JOCHAM, FERDINAND EISENBERGER, ROLAND TAUBER, ARNULF STENZL AND ALFONS HOFSTETTER From the Departments of Urology, Clinic Ebersberg and Klinikum Grosshadern, Ludwig-Maximilians-University of Munich, Munich, Klinikum Landshut, Landshut, Klinikum Heilbronn, Heilbronn, University of Luebeck, Lubeck, Katharinenhospital, Stuttgart, Barmbek General Hospital, Hamburg and University of Innsbruck, Innsbruck, Austria

ABSTRACT

Purpose: Endoscopy done under fluorescence induced by 5-aminolevulinic acid has proved to be a procedure with high sensitivity for detecting transitional cell carcinoma of the bladder. In this multicenter, parallel group, phase III study we compared 5-aminolevulinic acid fluorescence endoscopy guided transurethral bladder resection with transurethral bladder resection done using only white light endoscopy. The proportion of tumor-free resected cases in the 2 groups was evaluated. Materials and Methods: After patient stratification according to participating centers and European Organization for the Research and Treatment of Cancer risk score 65 and 64 were randomized to the 5-aminolevulinic acid fluorescence and white light endoscopy groups, respectively. Residual tumor was evaluated in the 2 groups by repeat transurethral resection 10 to 14 days later. Analysis was performed according to the intent to treat principle with all patients randomized, followed by per protocol analysis. Results: Intent to treat analysis revealed that in the white light endoscopy group 40.6% of cases were resected tumor-free at primary resection, whereas with 5-aminolevulinic acid fluorescence endoscopy guided transurethral resection 61.5% were resected tumor-free (p ⬍0014). On protocol analysis 46.9% patients in the white light and 67.3% in the 5-aminolevulinic acid fluorescence endoscopy groups were resected tumor-free (p ⬍0.031). No difference was noted in the 2 groups in regard to side effects or laboratory findings. Conclusions: The risk of residual tumor after transurethral resection of transitional cell carcinoma is significantly decreased by 5-aminolevulinic acid fluorescence endoscopy. KEY WORDS: bladder, bladder neoplasms, endoscopy, fluorescence, light

In 1999 approximately 54,000 new cases of bladder cancer were diagnosed in the United States, of which 70% involved superficial cancer at diagnosis.1 The main concern with superficial bladder cancer is the high recurrence rate, which is currently between 50% and 70%.1 The complete transurethral resection or destruction of all bladder tumors is supposed to be a crucial factor in preventing recurrent disease.2 However, flat urothelial lesions with no epithelial thickening, such as carcinoma in situ and dysplasia, and small papillary tumors are barely visible during conventional white light endoscopy and can easily be missed during transurethral resection.3 After transurethral resection of superficial bladder cancer residual tumor has been identified in up to 55% of cases at resection repeated 1 to 2 weeks later.4 – 6 Fluorescence endoscopy induced by 5-aminolevulinic acid has proved to have high sensitivity for detecting early stage bladder cancer. The results of additional evaluation by protoporphyrin IX fluorescence have shown a significant increase in diagnostic sensitivity for neoplastic urothelial lesions, such as dysplasia and carcinoma in situ, as well as for papillary tumors compared with conventional white light endoscopy.7, 8 Therefore, the question arises of whether 5-aminolevulinic acid induced fluorescence endoscopy guided transurethral resection would decrease the amount of residual tumor. We report the results of a multicenter phase III Accepted for publication March 8, 2002. * Equal study contribution.

trial of whether 5-aminolevulinic acid fluorescence endoscopy improves the ability to visualize tumor extension and increase the resection of malignant lesions compared with white light endoscopy only. MATERIALS AND METHODS

This multicenter, parallel group, phase III study was done at 8 centers in Germany and Austria from April 1997 to February 1998. The trial was performed in accordance with the Declaration of Helsinki, revised version of South Africa 1996, Principles for Carrying Out in the Prescribed Manner the Clinical Testing of Medicines, December 1987 and the German Law on Drugs. Patients. Entered into the study were 165 cases suspicious for primary bladder cancer or tumor recurrence. The randomization procedure was stratified according to participating centers and further by the potentially prognostically relevant risk score, defined according to the results of European Organization for the Research and Treatment of Cancer studies 30831, 30790 and 30782 as 1—recurrence, 2— early recurrence at less than 12 months, 3— bacillus Calmette-Guerin therapy less than 12 months in duration and 4 —a history of carcinoma in situ.9, 10 After stratification 83 and 82 patients were randomized to the 5-aminolevulinic acid fluorescence endoscopy and white light endoscopy groups, respectively. The 2-sided chi-square test for homogeneity showed no dif-

475

476

TRANSURETHRAL BLADDER CANCER RESECTION USING FLUORESCENCE ENDOSCOPY

ferences in the 2 groups in regard to demographic data or other baseline characteristics, such as patient age and sex (table 1). Treatment. Transurethral resection was performed under fluorescent blue light induced by 5-aminolevulinic acid or under white light. The method of transurethral resection was the same in the groups. It started with resection of the superficial layer of the tumor, followed by the deep portion with some underlying muscle. Resection was extended 1 sling within the peritumoral region, followed by complete fulguration, as previously described.11 Residual tumor was evaluated in the 2 groups by repeat transurethral resection under white light 10 to 14 days later. We determined whether resection under 5-aminolevulinic acid fluorescence endoscopic guidance improves the ability to visualize tumor extension and increases the resection of malignant lesions compared with white light endoscopy only. Therefore, resection was repeated at the initial resection margins and deeper tissue even in cases of no evident tumor. Fluorescence endoscopy induced by 5-aminolevulinic acid. Lyophilized 5-aminolevulinic acid hydrochloride (Medac GmbH, Wedel, Germany) (1.5 gm.) dissolved in 50 ml. 5.7% sodium monohydrogen phosphate dodecahydrate solution was instilled intravesically 2 hours before transurethral resection, as previously described.8 For fluorescence excitation of protoporphyrin IX a D-light (Storz GmbH, Tuttlingen, Germany) providing blue light at 375 to 440 nm. was used. A yellow long pass filter fit into the eyepiece of the endoscope reduced the blue excitation light and enhanced fluorescent contrast. Changing between blue and white light was possible. In the fluorescence endoscopy group the bladder and suspicious areas were judged with respect to malignancy under white light first, followed by careful inspection under blue light. Each suspicious tumor area detected by white or blue light was resected. It was also determined whether fluorescence positive findings had previously been missed under white light. Statistical methods. Analysis was performed according to the intent to treat principle, followed by analysis per protocol. The results of the 2 approaches were compared. General intent to treat analysis require that all patients included in the study should be analyzed for the primary end point. However, in contrast to the issue of patient noncompliance with the study regimen, determining patient eligibility for trial participation after randomization according to strict interpretation of the intent to treat principle can cause study results that are simply not credible. If patients with a presumptive diagnosis were assigned to the specific treatment arm by randomization, a portion in each treatment group would later prove not to have had a tumor. Strict application of the intent to treat principle requires that patients proved not to have bladder cancer should still be included in analysis in the treatment arm to which they were assigned. For valid case analysis a subgroup of patients per protocol set was formed from the full analysis set. This subgroup consisted of patients for whom the parameters of the primary efficacy

TABLE 1. Patient gender and sex, and tumor risk Intent to Treat Endoscopy Per Protocol Endoscopy Fluorescence White Light Fluorescence White Light No. randomized No. evaluable Mean age (range) No. women/No. men No. risk profile: I II III

83 82 83 82 65 64 52 49 69.3 (38–88) 69.6 (34–94) 70.1 (44–88) 70.7 (44–94) 12/53 37 16 12

19/45 40 11 13

11/41 31 12 9

13/36 30 8 11

variable were determined, that is repeat resection was performed in which a histological specimen was obtained and evaluated. For analysis of the primary end point treatment groups were statistically compared via Fisher’s exact test with 1-tailed significance considered at 5%. Data management and data set analysis were performed with commercially available, validated statistical software. RESULTS

Of the 165 randomized patients 129, including 65 in the fluorescence and 64 in the white light groups, had histological proof of transitional cell carcinoma after initial transurethral resection and were considered evaluable. The 2-sided chi-square test for the homogeneity of treatment groups showed no significant differences in the groups with respect to the risk profile (table 1). In 101 patients, including 52 in the fluorescence and 49 in the white light groups, transurethral resection was repeated. Thus, 13 patients in the fluorescence and 15 in the white light groups were excluded from the full analysis set. These 28 cases were not analyzed according to the intent to treat principle for several reasons. Five patients in the fluorescence and 3 in the white light groups who were excluded from study due to protocol violation were not excluded from the full analysis set according to the intent to treat principle, although there was no histological confirmation of tumor at initial resection. Nevertheless, they underwent repeat resection because of the surgeon decision to ensure that definite macroscopic evidence of a lesion of unclear status on primary resection was not cancer. Of the remaining patients excluded from analysis in the fluorescence and white light groups 2 and 5 underwent cystectomy, 4 and 2 refused treatment, and 1 and 5 were excluded for other reasons, respectively. In a single patient who underwent fluorescence endoscopy the histological data were lost. Table 2 lists stage and grade distributions of the study population after initial transurethral resection. After repeat transurethral resection intent to treat and per protocol analyses revealed a significant benefit for the 5-aminolevulinic acid fluorescence endoscopy procedure (p ⬍0014 versus ⬍0.031). On protocol analysis 52 (67.3%) but only 49 (46.9%) evaluable patients in the fluorescence and white light groups were tumor-free after repeat resection, while on intent to treat analysis 65 (61.5%) and 64 (40.6%) were tumor-free, respectively. The number of patients with residual tumor decreased about 40%. Tumor was identified at the previous resection margin in 12 of 17 and in 17 of 26 patients with residual disease in the fluorescence and white light groups,

TABLE 2. Stage and grade No. Intent to Treat Endoscopy

No. Per Protocol Endoscopy

Fluorescence White Light Fluorescence White Light Stage: No tumor Dysplasia II Ca in situ Ta T1 T2 Missing Totals Grade: I II III Missing No tumor Totals

5 2 3 36 12 5 1

3 2 4 30 13 10 1

— 2 2 33 11 4 —

— 2 5 25 11 6 —

65

64

52

49

21 21 6 12 5

8 27 8 18 3

19 19 6 8 —

6 24 6 13 —

65

64

52

49

TRANSURETHRAL BLADDER CANCER RESECTION USING FLUORESCENCE ENDOSCOPY

while disease was detected in deeper tissue in 5 and 7, respectively. In patients at low risk and those with stage Ta lesions the statistically significant superiority of the fluorescence endoscopy procedure was also documented (p ⬍0.05). The 2 groups showed no differences in regard to the number or regions of biopsies obtained during initial and repeat resection on intent to treat or on protocol analysis. There was no heterogeneous impact of the several physicians on outcome, as proved by exploration statistical testing using Zelen’s exact test. Furthermore, there was no difference in symptoms or in laboratory findings in the fluorescence and white light groups. DISCUSSION

Incomplete tumor resection is recognized as an important factor of the local recurrence and progression of superficial bladder cancer.12 After transurethral resection of superficial bladder cancer residual tumor has been detected in up to 55% of cases at repeat resection 1 to 2 weeks later.4 – 6 Even in cases of solitary superficial bladder neoplasms residual disease was identified in 24% at repeat transurethral resection 5 weeks later.4 In that study the quality of initial resection was emphasized by the fact that repeat resection revealed infiltrative growth in only 2% of cases. Therefore, most missed neoplasms were due to positive margins or heterotopic lesions. Klan et al reported fractional resection of stage T1 transitional cell cancer.5 In 28% of their patients positive tumor margins were noted, while residual disease from the tumor base was not observed. Routine resection repeated 8 to 14 days later revealed residual disease in 50% of patients despite the surgical report of complete resection. Of the missed lesions 76% were visible on repeat resection. Klan et al concluded that the extent of lesions can easily be misjudged even by experienced surgeons. Fitzpatrick at al reported on 414 newly diagnosed patients with stage pTa grades I to II cancer.13 Those who underwent intravesical therapy during followup were excluded from study. In a third of the cases recurrent or more likely residual disease was identified 3 months after initial resection. Of these patients 90% had further recurrences thereafter, whereas only 21% who were free of tumor 3 months after initial resection had tumor recurrence during further followup. The response to initial treatment for superficial bladder cancer14 as well as the initial number of tumors15 has been stressed as relevant to the outcome. When considering the clinical data described, more complete removal of bladder lesions seems advisable. Under white light endoscopy neoplasm detection is limited to morphological patterns. Since the 1960s, urologists have sought methods of labeling neoplasms in vivo to decrease the risk of missing cancer by means of additional color contrast. Diagnostic methods based on detecting the fluorescence of systemically administered tetracycline, systemic porphyrin mixtures and fluorescein or simple staining with methylene blue were only tested in a few cases and they have been abandoned.16 –19 After intravesical application of 5-aminolevulinic acid we noted selective accumulation of protoporphyrin IX in urothelial cancer, providing an intensive color contrast of red fluorescing malignant lesions and nonfluorescing normal blue mucosa.20 Spectral measurements in vivo showed more than 15-fold higher mean intensity of fluorescence of urothelial cancer compared with normal urothelium.21 On biopsy the procedure had 97% sensitivity and 65% specificity.22 The outstanding sensitivity of the procedure was confirmed in other series as 86% to 96% and the rate of enhanced detection of neoplastic disease was 18% to 76%.23–26 Performing biopsies after transurethral resection under white light in normal appearing mucosa does not contribute

477

to bladder cancer staging.27 In contrast, in another study 60% of the malignant lesions in cases of positive cytology results were detected only in 5-aminolevulinic acid fluorescence endoscopy biopsies of bladders with only normal or nonspecific, inflamed appearing mucosa.8 Almost half of these cases had previously been classified as carcinoma in situ or as high grade disease. The outcome of the current trial underlines the clinical impact of 5-aminolevulinic acid induced fluorescence endoscopy guided transurethral resection of bladder cancer. This clinical trial represents a multicenter, parallel group, phase III design. The study was controlled by reference pathological findings. After stratification according to participating centers and the European Organization for the Research and Treatment of Cancer risk score 83 and 82 patients were randomized to the fluorescence and white light groups, respectively. Residual tumor in the 2 groups was evaluated by repeat transurethral resection 10 to 14 days later. Analysis was performed according to the intent to treat principle with all patients randomized, followed by per protocol analysis. The intent to treat analysis set revealed that in the white light group 40.6% of cases were resected tumor-free at primary resection, whereas in the fluorescence group 61.5% were resected tumor-free (Fisher’s exact test p ⫽ 0.014). In the set per protocol 46.9% of cases in the white light arm and 67.3% in the fluorescence arm were resected tumor-free (Fisher’s exact test p ⫽ 0.031). Because most patients included in the study had superficial, low risk disease, the statistical power of the study outcome was based on these patients and was not determined in those with flat lesions, such as carcinoma in situ. Nevertheless, due to the enhanced detection of dysplasia and carcinoma in situ this difference could also be expected to be statistically significant with a sufficient number of evaluable patients. Our results are similar to those described by Riedl et al, who reported that the rate of missed tumors decreased by 59%.28 CONCLUSIONS

Fluorescence endoscopy based on 5-aminolevulinic acid facilitates the detection of neoplasms in the bladder during transurethral resection. The procedure is simple and easy to perform. In cases of complete negative fluorescence findings endoscopy can be done without biopsies due to the high sensitivity of the procedure. There was a significant decrease in the rate of missed tumors on transurethral resection done under 5-aminolevulinic acid induced fluorescence endoscopic guidance. A resulting, relatively decreased number of recurrences must be documented in prospective randomized trials. REFERENCES

1. Lee, R. and Droller, M. J.: The natural history of bladder cancer. Urol Clin North Am, 27: 1, 2000 2. Holzbeierlein, J. M. and Smith, J. A., Jr.: Surgical management of noninvasive bladder cancer (stages Ta/T1/CIS). Urol Clin North Am, 27: 15, 2000 3. Soloway, M. S., Murphy, W., Rao, M. K. and Cox, C.: Serial multiple-site biopsies in patients with bladder cancer. J Urol, 120: 57, 1978 4. Ko¨ hrmann, K. U., Woeste, M., Kappes, J., Rassweiler, J. and Alken, P.: Der Wert der transurethralen Nachresektion beim oberfla¨ chlichen Harnblasenkarzinom. Akt Urol, 25: 208, 1994 5. Klan, R., Loy, V. and Huland, H.: Residual tumor discovered in routine second transurethral resection in patients with stage T1 transitional cell carcinoma of the bladder. J Urol, 146: 316, 1991 6. Vo¨ geli, T. A., Grimm, M.-O. and Ackermann, R.: Prospective study for quality control of TUR of bladder tumors by routine 2nd TUR (ReTUR). J Urol, suppl., 159: 143, abstract 543, 1998 7. Kriegmair, M., Zaak, D., Stepp, H., Stepp, H., Baumgartner, R., Knuechel, R. et al: Transurethral resection and surveillance of bladder cancer supported by 5-aminolevulinic acid-induced fluorescence endoscopy. Eur Urol, 36: 386, 1999

478

TRANSURETHRAL BLADDER CANCER RESECTION USING FLUORESCENCE ENDOSCOPY

8. Zaak, D., Kriegmair, M., Stepp, H., Stepp, H., Baumgartner, R., Oberneder, R. et al: Endoscopic detection of transitional cell carcinoma with 5-aminolevulinic acid: results of 1012 fluorescence endoscopies. Urology, 57: 690, 2001 9. Bouffioux, Ch., Kurth, K. H., Bono, A., Oosterlinck, W., Kruger, C. B., De Pauw, M. et al: Intravesical adjuvant chemotherapy for superficial transitional cell bladder carcinoma: results of 2 European Organization for Research and Treatment of Cancer randomized trials with mitomycin C and doxorubicin comparing early versus delayed instillations and short-term versus long-term treatment. J Urol, 153: 934, 1995 10. Kurth, K. H., Denis, L., Bouffioux, C., Sylvester, R., Debruyne, F. M., Pavone-Macaluso, M. et al: Factors affecting recurrence and progression in superficial bladder tumors. Eur J Cancer, 31A: 1840, 1995 11. Soloway, M. S., Kurth, K. H., Herr, H., Huland, H., Denis, L., Suzuki, K. et al: Surgical techniques in the management of patients with superficial bladder cancer. Prog Clin Biol Res, 221: 123, 1986 12. Schellhammer, P. F.: Immediate adjuvant intravesical chemotherapy after transurethral resection of the bladder. Curr Opin Urol, 8: 407, 1998 13. Fitzpatrick, J. M., West, A. B., Butler, M. R., Lane, V. and O’Flynn, J. D.: Superficial bladder tumors (stage pTa grades 1 and 2): the importance of recurrence pattern following initial resection. J Urol, 135: 920, 1986 14. Pryor, J. P.: Factors influencing the survival of patients with transitional cell tumours of the urinary bladder. Br J Urol, 45: 586, 1973 15. Brausi, M., Kurth, K., van der Meijden, A. P., Sylvester, R. J. and Collette, L.: Variability in the three-months recurrence rate after TUR in superficial transitional cell carcinoma (TCC) of the bladder: a combined analysis of 8 EORTC studies. J Urol, suppl., 159: 143, abstract 541, 1998 16. Whitmore, W. F., Jr. and Bush, I. M.: Ultraviolet cystoscopy in patients with bladder cancer. Trans Am Ass Genitourin Surg, 57: 149, 1965 17. Jocham, D., Baumgartner, R., Fuchs, N., Lenz, H., Stepp, H. and Unso¨ ld, E.: Fluorescence diagnosis of porphyrin-marked urothelial tumors. Status of experimental development. Urologe A, 28: 59, 1989 18. Zimmern, P. E., Laub, D. and Leach, G. E.: Fluorescein angiography of the bladder: technique and relevance to bladder cancer and interstitial cystitis patients. J Urol, 154: 62, 1995 19. Vicente, J., Chechile, G. and Algaba, F.: Value of in vivo mucosa-

20.

21.

22.

23.

24.

25.

26.

27.

28.

staining test with methylene blue in the diagnosis of pretumoral and tumoral lesions of the bladder. Eur Urol, 13: 15, 1987 Kriegmair, M., Baumgartner, R., Knuechel, R., Steinbach, P., Ehsan, A., Lumper, W. et al: Fluorescence photodetection of neoplastic urothelial lesions following intravesical instillation of 5-aminolevulinic acid. Urology, 44: 836, 1994 Kriegmair, M., Stepp, H., Steinbach, P., Lumper, W., Ehsan, A., Stepp, H. G. et al: Fluorescence cystoscopy following intravesical instillation of 5-aminolevulinic acid: a new procedure with high sensitivity for detection of hardly visible urothelial neoplasias. Urol Int, 55: 190, 1995 Kriegmair, M., Baumgartner, R., Knu¨ chel, R., Stepp, H., Hofsta¨ dter, F. and Hofstetter, A.: Detection of early bladder cancer by 5-aminolevulinic acid induced porphyrin fluorescence. J Urol, 155: 105, 1996 Jichlinski, P., Forrer, M., Mizeret, J., Glanzmann, T., Braichotte, D., Wagnieres, G. et al: Clinical evaluation of a method for detecting superficial surgical transitional cell carcinoma of the bladder by light-induced fluorescence of protoporphyrin IX following the topical application of 5-aminolevulinic acid: preliminary results. Lasers Surg Med, 20: 402, 1997 Filbeck, T., Roessler, W., Knuechel, R., Straub, M., Kiel, H. J., Wieland, W. F. et al: Clinical results of the transurethral resection and evaluation of superficial bladder carcinomas by means of fluorescence diagnosis after intravesical instillation of 5-aminolaevulinic acid. J Endourol, 13: 117, 1999 Koenig, F., McGovern, F. J., Larne, R. Enquist, H., Schomacker, K. T. and Deutsch, T. F.: Diagnosis of bladder carcinoma using proroporhyrin IX fluoresence induced by 5-aminolaevulinic acid. BJU Int, 83: 129, 1999 De Dominicis, C., Liberti, M., Perugia, G., De Nunzio, C., Sciobica, F., Zuccala, A. et al: Role of 5-aminolevulinic acid in the diagnosis and treatment of superficial bladder cancer: improvement in diagnostic sensitivity. Urology, 57: 1059, 2001 Kiemeney, L. A., Witjes, J. A., Heijbroek, R. P., Koper, N. P., Verbeek, A. L. and Debruyne, F. M.: Should random urothelial biopsies be taken from patients with primary superficial bladder cancer? A decision analysis. Members of the Dutch SouthEast Co-Operative Urological Group. Br J Urol, 73: 164, 1994 Riedl, C. R., Daniltchenko, D., Koenig, F., Simak, R., Loening, S. A. and Pflueger, H.: Fluorescence endoscopy with 5-aminolevulinic acid reduces early recurrence rate in superficial bladder cancer. J Urol, 165: 1121, 2001