Do patients profit from 5-aminolevulinic acid-induced fluorescence diagnosis in transurethral resection of bladder carcinoma?

ADULT UROLOGY

DO PATIENTS PROFIT FROM 5-AMINOLEVULINIC ACIDINDUCED FLUORESCENCE DIAGNOSIS IN TRANSURETHRAL RESECTION OF BLADDER CARCINOMA? T. FILBECK, U. PICHLMEIER, R. KNUECHEL, W. F. WIELAND,

AND

W. ROESSLER

ABSTRACT Objectives. To evaluate in a prospective study the influence of fluorescence diagnosis (FD) controlled transurethral resection of bladder tumors on therapeutic consequences. The aim was to determine in how many patients FD led to a change in treatment strategy compared with conventional white light (WL) cystoscopy. Methods. A total of 279 patients with suspected bladder tumors underwent transurethral resection using FD in addition to WL cystoscopy. The number of additional tumor-positive patients, staging change, number of multilocular tumors exclusively detected by FD, and resulting therapeutic consequences compared with the results after WL cystoscopy were investigated. In addition a biopsy-based evaluation was performed. Results. Tumor or dysplasia II° (moderate dysplasia) was detected in 177 patients. In 168 patients, tumor was detected by WL cystoscopy, and in 9 (5.1%) of the patients, tumor was completely overlooked by WL cystoscopy and diagnosed exclusively by FD (n ⫽ 3 TaG1-G2, n ⫽ 2 carcinoma in situ, n ⫽ 1 greater than T1, and n ⫽ 3 dysplasia II°). Multilocular tumor involvement was detected in 10 cases using FD, and a change in the stage by detection of coexisting dysplasia II° and carcinoma in situ occurred in 8 patients. In 27 patients (15.3%), additional information was obtained by exclusive detection of tumors by FD. This resulted in a change in the treatment strategy for 16 patients (9%). Conclusions. FD leads to an improvement in the diagnosis of bladder carcinoma. It allows the early selection of the best treatment option and thus has a potentially positive effect on the prognosis of the affected patients. UROLOGY 60: 1025–1028, 2002. © 2002, Elsevier Science Inc.

S

ince 1995, 5-aminolevulinic acid (ALA)induced fluorescence diagnosis (FD) has been used increasingly in Europe, and particularly in Germany, Austria, and Switzerland, in the diagnosis and transurethral resection (TUR) of superficial bladder carcinoma. It is a highly sensitive procedure that facilitates or enables the detection of bladder cancer, particularly flat lesions of the bladder that are difficult or impossible to visualize using conventional white light (WL) endoscopy.1–5 5-ALA is a precursor in the biosynthetic pathway of heme. After intravesical instillation, the photoFrom the Department of Urology, St. Joseph’s Hospital, Regensburg; Institute of Mathematics and Computer Science in Medicine, University of Hamburg, Hamburg; Institute of Pathology, University of Regensburg, Regensburg, Germany Reprint requests: Thomas Filbeck, M.D., Department of Urology, St. Joseph Hospital, Regensburg, Landshuter Strasse 65, Regensburg 93053, Germany Submitted: March 7, 2002, accepted (with revisions): July 15, 2002 © 2002, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED

active metabolite protoporphyrin IX accumulates. A higher accumulation of protoporphyrin IX is found in urothelial tumor tissue compared with unaffected urothelium (ratio 17:1).6,7 Thus, tumors fluoresce red under excitation with violet light at a wavelength of 375 to 440 nm. This procedure allows more precise resection, showing the margins or extent of tumors considerably better. The results published hitherto show a clear advantage for FD. The results published to date refer to biopsy-based evaluations and comparative studies of the rates of residual tumor and recurrence.8 –10 The biopsy-based results gave no indication of the extent of the benefit for the individual patient provided by this method. Comparative studies, which showed markedly positive results in terms of a reduction in the rate of residual tumor and recurrence, did not answer the following questions: how many patients who are tumor free by conventional endoscopy are diagnosed with tumor after FD; how frequent are the changes in staging or detection of 0090-4295/02/$22.00 PII S0090-4295(02)01961-1 1025

TABLE I. Tumor-positive biopsies detected by conventional white light and excess detected exclusively by 5-ALA-induced fluorescence cystoscopy stratified by tumor stage TaG1-G2 Tumor-positive biopsies detected by conventional white light cystoscopy (n) Additional tumor-positive biopsies detected exclusively by FD (n) Total tumor-positive biopsies (n)

170 16 (8.6) 186

TaG3 T1G2 6 0 6

29 2 (7) 31

T1G3 37 5 (12) 42

CIS 13 17 (57) 30

>T1 22 3 (12) 25

DII° 9 7 (44) 16

Total 286 50 (14.9) 336

KEY: ALA ⫽ aminolevulinic acid; CIS ⫽ carcinoma in situ; DII ⫽ dysplasia II°; FD ⫽ fluorescence diagnosed. Numbers in parentheses are percentages.

multilocular tumors; and in how many patients does this have therapeutic consequences? These questions were investigated in a singleinstitution, prospective, nonrandomized study of the value of FD. The study consisted of both biopsy-based and patient-based evaluations. To our knowledge, this is the first study that deals with these questions. MATERIAL AND METHODS The study was designed as a within-patient evaluation to determine the additional number of patients in whom FDcontrolled TUR led to therapeutic consequences. From January 1997 to December 2000, 279 patients (age range 34 to 89 years) with suspected primary or recurrent bladder tumor underwent TUR with FD at our institution. Two hours before the planned TUR, all the patients were catheterized and 50 mL of a 3% 5-ALA solution (1.5 g ALA dry substance in 50 mL 5.4% sodium hydrogen carbonate solution, Medac, Hamburg, Germany) was instilled intravesically. On average, the duration of instillation was 142.6 minutes (range 60 to 240). A WL source with a band pass filter equipped with a xenon lamp was used as light source for the FD (Storz, Tuttlingen, Germany). This can produce a violet light of wavelength 345 to 440 nm to stimulate fluorescence. By means of a foot switch, it is possible to alternate between fluorescence and WL. The bladder was first inspected under WL. The inspection was repeated after switching to fluorescent light. All the fluorescing areas and all suspicious areas or tumors apparent under WL were resected. An area was regarded as suspicious under WL if it had the endoscopic appearance of nonspecific inflammation, papillary tumor, or solid tumor. After conclusion of the TUR, the extent of resection was again inspected under fluorescence control and any areas still fluorescing were resected. Using standardized case report forms, the fluorescence status, location, fluorescence quality, and histopathologic evaluation findings were recorded for each biopsy. The lesions or tumors resected had to be classified as fluorescence positive, as well as WL positive or negative. The histopathologic processing of all preparations took place in a single reference center (Department of Pathology, Regensburg University). The grading was carried out according to the World Health Organization classification of 199411 and pathologic staging according to the International Union Against Cancer criteria of 1997.12 Any biopsy judged by the pathologist to be dysplasia II° (DII) or worse was defined as a lesion with histologically verified tumor, otherwise the biopsy was judged to be tumor negative. A biopsy-based evaluation was performed that is widely accepted in published studies.1–5 This approach allowed the 1026

judging of the number of tumor lesions detected only under fluorescent light that would have been overlooked under conventional cystoscopy. Furthermore, a patient-based analysis was performed to quantify the number of patients diagnosed as tumor positive exclusively by FD. It was also possible to assess whether multilocular involvement was present in patients who had solitary tumors under WL and a change in staging by the detection of coexisting carcinoma in situ (CIS) and DII. These analyses demonstrated the clinical impact of 5-ALAinduced fluorescence-guided cystoscopy on therapeutic decisions and thus on the patient’s prognosis. Statistical analysis was performed by calculating the absolute and relative frequencies. In addition, 95% Clopper-Pearson confidence intervals (CIs) are presented.

RESULTS Tumor or DII was detected in 177 of 279 patients with suspected primary or recurrent tumor. A total of 636 biopsies were taken or tumors resected. Of 336 tumors, 50 (14.9%) were overlooked with WL and were only diagnosed by FD. The difference in the detection of flat urothelial lesions, CIS, and DII was particularly striking in the biopsy-based evaluation (Table I). Thus, 17 (57%) of cases of CIS and 7 (44%) of 16 cases of DII were diagnosed by FD, which is equivalent to an additional detection rate of 130% and 78%, respectively. Additional tumors with other stages were also detected but in a lower proportion. The patient-based evaluation resulted in the detection of tumor or DII in a total of 177 patients, and 102 were tumor free (Table II). Under WL, neoplasia was detected and resected in 168 patients. Accordingly, tumor or DII (n ⫽ 3 TaG1-G2, n ⫽ 2 CIS, n ⫽ 1 greater than T1, and n ⫽ 3 DII) was detected exclusively by FD in 9 patients (95% CI 2.4% to 9.4%). In the case of tumor invading muscle that was overlooked on WL, it was a very small flat recurrence of a muscle-invasive tumor in the region of a previous resection scar that did not appear either raised or solid and was apparent clearly only on FD. In the case of 3 patients with DII not visible conventionally, a TaG1-G2 tumor and CIS were diagnosed during follow-up after 6 and 9 months. UROLOGY 60 (6),

TABLE II. Tumor-positive patients detected by conventional white light and excess exclusively diagnosed with 5-ALA-induced fluorescence cystoscopy, stratified by tumor stage Tumor-positive patients detected by conventional white light TUR (n) Additional tumor-positive patients detected by FD (n) Staging of all tumor-positive patients (n)

TaG1-G2

TaG3

T1G2

T1G3

102

3

18

22

3 (2.9) 105

0 (0) 3

0 (0) 18

0 (0) 22

CIS

>T1

4

16

2 (33)

3

1 (6)

6

DII°

17

Total 168

3 (50) 6

9 (5.1) 177

KEY: TUR ⫽ transurethral resection; other abbreviations as in Table I. Numbers in parentheses are percentages.

TABLE III. Clinical usefulness quantified by excess number of FD-detected tumor patients, number of patients with any change in staging and detection of multilocular tumors by FD and resulting change of treatment strategy TaG1-G2 Additional tumor-positive patients detected by FD (n) Patients with additional detection of coexisting DII° and CIS using FD exclusively (n) Detection of multilocular tumors (same stage) by FD only (n) Total (n) Change in treatment strategy (n)

TaG3

T1G2

T1G3

CIS

>T1

DII°

Total

3

—

—

—

2

1

3

9

2*

—

—

2

2*

2

—

8

1

—

—

5

2

2

—

10

6 5

— —

— —

7 5

6 2

5 1

3 3

27 (15.3) 16 (9.0)

Abbreviations as in Table I. Numbers in parentheses are percentages (of 177 patients). * Coexisting DII° (n ⫽ 1 in case of TaG1-G2 and n ⫽ 2 in case of CIS).

FD resulted in a change in staging in 8 of 168 patients (95% CI 2.1% to 9.2%) identified under WL as tumor positive, either through the additional detection of coexisting CIS or DII in the case of papillary carcinoma (Table III). In 10 of 168 patients (95% CI 2.9% to 10.7%), FD showed multilocular tumor involvement when a solitary tumor was found initially under WL (Table III). Of the 177 patients in whom tumor or DII was diagnosed under WL, a change in the treatment strategy occurred in 16 cases (9%; associated 95% CI 5.3% to 14.3%; Table III). Initially, of course, a change occurred in the treatment strategy for all patients who were additionally identified as tumor positive (Table II). However, not every change in stage (n ⫽ 8) or the detection of multilocular tumor (n ⫽ 10) resulted in a change in treatment strategy. Changes in treatment were necessary in 7 cases. In 1 patient with an initially solitary TaG1-G2 tumor, intravesical prophylaxis with mitomycin was instituted because of the detection of multilocular involvement. Bacille CalmetteGue´ rin therapy was instituted instead of mitomycin in the case of coincidental CIS in patients with TaG1-G2 tumors. Early cystectomy was indicated in 5 patients with T1G3 tumors because of the adUROLOGY 60 (6), 2002

ditional detection of CIS or multilocular tumor involvement. Thus, 27 patients (15.3%, 95% CI 10.3% to 21.4%) were able to profit from FD: 18 patients with bladder carcinoma already diagnosed conventionally and an additional 9 patients who were tumor free under WL. A change in treatment strategy occurred in 9% of the patients, a small but significant number. COMMENT FD is characterized primarily by the possibility of diagnosing tumors that are difficult or impossible to visualize conventionally, thereby increasing the detection rate of individual tumors. In particular, flat lesions with high malignant potential such as CIS can be diagnosed markedly better and more often, as shown by the additional detection rate with FD of 57% in this study. Different investigators have confirmed this observation and have reported a markedly higher rate of detection of bladder tumors and DII using FD compared with WL cystoscopy, when exclusively biopsy-based investigations were used as a basis for calculation.1–5 However, the fundamental question arises as to 1027

whether these additionally detected tumors have consequences with regard to further treatment in the individual patients and how many patients actually profit from it. In our study, it appears that the additional detection rate of 14.9% by FD in the biopsy-based evaluation does not correlate to the same degree with the patient-based evaluation. In comparison, only 5.1% of the patients could additionally be classified as tumor positive. This does not argue against the method but does show that biopsy-based evaluation does not sufficiently show the benefit for the individual patient (Table I). However, the effect of a stage change (up-staging) and the detection of multilocular tumor must be taken into account in the patient-based evaluation, along with the resulting frequency of a change in treatment strategy. This occurred in 9% of the patients. Accordingly, FD offers a clear improvement in diagnostic possibilities. The detection of prognostically unfavorable tumor stages led to early cystectomy in 5 cases, and the prognosis could potentially be improved.13 Moreover, patients with CIS initially overlooked under WL could be treated early with bacille Calmette-Gue´ rin. Furthermore, additional detection of DII allowed us to follow-up the affected patients closely, and, as a result, 2 patients profited who later developed carcinoma (CIS and TaG2). Although DII is not classified as malignant, studies in small groups of patients have showed that the risk of developing an invasive tumor is 15% with DII,14 and recent genetic data confirm the potential role of DII as precursor of CIS.15 Close follow-up is therefore advisable in these patients. In the case of papillary stage, TaG1-G2, and T1G2 tumors, hardly any changes occurred in therapeutic strategy owing to FD. On the other hand, patients with these tumor stages profit from the possibility of more precise and more complete TUR and from the increased detection and resection of flat and papillary lesions, not visible conventionally. This should lead to a significant reduction in the rate of residual tumor and recurrence, because residual tumor after TUR and nonvisible or overlooked tumors are regarded as an important cause of the high rate of recurrence of superficial bladder cancers. This is also confirmed in early, randomized, prospective studies in which WL was compared with FD-assisted TUR. These showed that the rate of residual tumor can be reduced 50% to 80% and the rate of recurrence can be reduced by up to 67% for all tumor stages with a follow-up of on average of 21.2 months.8 –10 CONCLUSIONS FD leads to an improvement in the diagnosis in bladder carcinoma. Therefore, establishing this 1028

method as a standard procedure in the diagnosis and treatment of bladder tumors should be discussed. Multicenter clinical trials are necessary and have been ongoing for several years and will possibly strengthen our findings. REFERENCES 1. De Dominicis C, Liberti G, Perugia G, et al: Role of 5-aminolevulinic acid in the diagnosis and treatment of superficial bladder cancer: improvement in diagnostic sensitivity. Urology 57: 1059 –1062, 2001. 2. Filbeck T, Roessler W, Straub M, et al: Clinical results of the transurethral resection and evaluation of superficial bladder carcinomas by means of FD after intravesical instillation of 5-aminolevulinic acid. J Endourol 13: 117–121, 1999. 3. Kriegmair M, Baumgartner R, Knuechel R, et al: Detection of early bladder cancer by 5 aminolevulinic acid induced porphyrin fluorescence. J Urol 155: 105–109, 1996. 4. Jichlinski P, Forrer M, Mizeret J, 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: 402–408, 1997. 5. Zaak D, Kriegmair M, Stepp H, et al: Endoscopic detection of transitional cell carcinoma with 5-aminolevulinic acid: results of 1012 fluorescence endoscopies. Urology 57: 690 – 694, 2001. 6. Steinbach P, Weingandt H, Baumgartner R, et al: Cellular fluorescence of the endogenous photosensitizer protoporphyrin IX following exposure to 5-aminolaevulinic acid. Photochem Photobiol 62: 887–895, 1995. 7. Kriegmair M, Baumgartner R, Knuechel R, et al: Fluorescence photodetection of neoplastic urothelial lesions following intravesical instillation of 5-aminolevulinic acid. Urology 44: 836 –841, 1994. 8. Riedl CR, Daniltchenko D, Koenig F, et al: Fluorescence endoscopy with 5-aminolevulinic acid reduces early recurrence rate in superficial bladder cancer. J Urol 165: 1121– 1123, 2001. 9. Zaak D, Frimberger D, Muller-Lisse U, et al: Decrease of early recurrence by means of 5-aminolevulinic induced fluorescence endoscopy in patients with bladder cancer follow up of a randomized phase III trial. J Urol 165: 165, 2001. 10. Filbeck T, Pichlmeier U, Knuechel R, et al: Reduction of recurrence rate of superficial bladder carcinoma after transurethral resection with 5-aminolevulinic acid induced fluorescence diagnosis. J Urol 165: 189, 2001. 11. Murphy WM, Beckwith JB, and Farrow GE: Atlas of tumor pathology, in Rosai J, and Sobin LH (Eds): Tumors of the Kidney, Bladder and Related Urinary Structures. Washington, DC, Armed Forces Institute of Pathology, 1994, Third Series, Facsimile 11. 12. Sobin LH, and Wittekind C: TNM Classification of Malignant Tumours, 5th ed. New York, Wiley-Liss, 1997. 13. Herr HW, and Sognani PC: Does early cystectomy improve survival of patients with high-risk superficial bladder tumors? Aktuel Urol 32: 360 –363, 2001. 14. Cheng L, Cheville JC, Neumann RM, et al: Natural history of urothelial dysplasia of the bladder. Am J Surg Pathol 23: 443–447, 1999. 15. Hartmann A, Schlake G, Zaak D, et al: Occurrence of chromosome 9 and p53 alterations in multifocal dysplasia and carcinoma in situ of human urinary bladder. Cancer Res 62: 809 –818, 2002. UROLOGY 60 (6), 2002