Fax: (011) 0049-7071-295092
Detection and clinical outcome of urinary bladder cancer with 5-aminolevulinic acid-induced fluorescence cystoscopy†
A multicenter randomized, double-blind, placebo-controlled trial
Article first published online: 8 NOV 2010
Copyright © 2010 American Cancer Society
Volume 117, Issue 5, pages 938–947, 1 March 2011
How to Cite
Stenzl, A., Penkoff, H., Dajc-Sommerer, E., Zumbraegel, A., Hoeltl, L., Scholz, M., Riedl, C., Bugelnig, J., Hobisch, A., Burger, M., Mikuz, G. and Pichlmeier, U. (2011), Detection and clinical outcome of urinary bladder cancer with 5-aminolevulinic acid-induced fluorescence cystoscopy. Cancer, 117: 938–947. doi: 10.1002/cncr.25523
See editorial on pages 882-3, this issue.
- Issue published online: 18 FEB 2011
- Article first published online: 8 NOV 2010
- Manuscript Accepted: 12 APR 2010
- Manuscript Revised: 9 APR 2010
- Manuscript Received: 7 JAN 2010
- urinary bladder;
The medical community lacks results from prospective controlled multicenter studies of the diagnostic efficacy of 5-aminolevulinic acid (5-ALA) cystoscopy on tumor recurrence in patients with superficial bladder tumors.
A prospective randomized, double-blind, placebo-controlled study was conducted in 370 patients with nonmuscle-invasive urinary bladder carcinoma who received either 5-ALA (n = 187) or a placebo (n = 183) intravesically before cystoscopy. Each group underwent cystoscopy under visible white light and under fluorescent light followed by transurethral tumor resection. The primary study objective was to evaluate the 12-month recurrence-free survival.
Slightly more patients with tumors were detected by using 5-ALA than by using the placebo (88.5% vs 84.7%). The mean numbers of tumor specimens per patient were 1.8 (5-ALA) and 1.6 (placebo). Intrapatient comparison of fluorescent light versus white light cystoscopy in patients randomized to receive 5-ALA showed a higher tumor detection rate with fluorescent light than with white light cystoscopy. In patients receiving 5-ALA cystoscopy, the percentage of lesions that would not have been detected in these patients by white light cystoscopy ranged between 10.9% (pT1) and 55.9% (atypia). Progression-free survival was 89.4% (5-ALA) and 89.0% (placebo) (P = .9101), and recurrence-free survival 12 months after tumor resection was 64.0% (5-ALA) and 72.8% (placebo) (P = .2216).
In comparison to the placebo, 5-ALA cystoscopy did not increase the rates of recurrence-free or progression-free survival 12 months after tumor resection. Although more tumors per patient were detected in the 5-ALA group, the higher detection rate did not translate into differences in long-term outcome. Cancer 2011. © 2010 American Cancer Society.
An estimated 64,900 new cases of urinary bladder cancer will be diagnosed in the United States in 2010.1 Approximately 75% of all newly diagnosed bladder tumors are nonmuscle-invasive cancers characterized by low mortality but a recurrence rate that exceeds 50%, with the majority of these recurrences occurring within the first 12 months.2 Residual tumors are found in 38% to 63% of patients, leading to a second resection after 2 to 6 weeks and possibly contributing to the high recurrence rate within the first year.3 With a prevalence of currently >520,000 cases in the United States, for which lifelong surveillance is recommended, and which need numerous repeat transurethral surgical operations and at times even radical cystectomy, bladder cancer has evolved to be the most expensive cancer on a per-patient basis.4
Numerous attempts have been made to develop optical markers for better detection of urothelial tumors and thus improve the clinical results of cancer surgery. 5-Aminolevulinic acid (5-ALA) is a photosensitizer with a high selectivity in neoplastic tissues. It was the first topical photoactive substance that could be instilled into the urinary bladder. The optimum instillation time with 5-ALA to achieve the best visible difference has been found to be 2 to 4 hours. When 5-ALA–induced protoporphyrin IX is excited by blue light, red fluorescence is emitted, which allows the visualization of suspicious lesions. The effect of 5-ALA–induced fluorescence on tumor detection in the urinary bladder has been assessed in several clinical trials.5-7 5-ALA–induced fluorescence cystoscopy has been shown to be an efficient method of mapping the entire mucosa in a bladder carcinoma to detect urothelial tumors and flat carcinoma in situ lesions that would not have been detected by conventional white-light cystoscopy.6, 8 Transurethral surgery guided by 5-ALA cystoscopy has been shown to decrease the risk of postresection residual tumors,9 the latter being recognized as an important prognostic factor for local recurrence and progression of nonmuscle-invasive urinary bladder cancer.10 The use of 5-ALA cystoscopy has considerably increased in parts of Europe and, based on these results, working groups looking at the clinical application have recently been established.11
When this study was initiated, results of controlled multicenter studies assessing the diagnostic efficacy of 5-ALA cystoscopy and its impact on tumor recurrence in patients with superficial bladder tumors were lacking. Meanwhile, only 2 single-center studies have shown that 5-ALA cystoscopy leads to an increase in the number of patients with recurrence-free survival after 1 year when compared with placebo white-light cystoscopy.12, 13
MATERIALS AND METHODS
This randomized, double-blind, placebo-controlled study comprising 381 patients was conducted at 8 urology centers in Austria and Germany over a period of 27 months. The study objective was to evaluate the diagnostic efficacy of 5-ALA in comparison to a placebo under visible (white) light and fluorescent light during transurethral resection of nonmuscle-invasive carcinomas of the urinary bladder.
Randomization was stratified according to 2 categories of the potentially prognostic and relevant European Organization for Research and Treatment of Cancer risk score describing the probability of tumor recurrence and/or tumor progression.14 Any patient with either multifocal tumor recurrence, early tumor recurrence (ie, ≤12 months), Bacillus Calmette-Guérin (BCG) treatment within the last 12 months, or a history of carcinoma in situ (CIS) was assigned to the high-risk group. All other patients eligible for trial were assigned to the low-risk group.
Patients aged older than 19 years with suspected nonmuscle-invasive urothelial carcinoma based on at least 1 imaging procedure such as cystoscopy, sonography, x-ray, or cytology with Papanicolaou test >II were eligible for study participation. Patients were excluded from participation in the case of any of the following: general health status >2 according to the World Health Organization (WHO) score (Eastern Cooperative Oncology Group), porphyria, hypersensitivity to porphyrins, renal impairment (creatinine >2.5 mg/dL), hepatic impairment (bilirubin >3 mg/dL, Quick test <60%, gamma-glutamyl transpeptidase >70 U/L), leukocytes <3500/μL, platelets <100,000/μL, lymph node metastasis or other metastasis, other current malignancies (not including basalioma), pregnancy (planned or existent), breast feeding, no safety in contraception, simultaneous participation in other clinical trials, or mental disorders.
The study was performed in accordance with the Good Clinical Practice guidelines recommended by the International Conference on Harmonization of Technical Requirements. Ethics committees and institutional review boards relevant to the respective study sites approved the study protocol. Written informed consent was obtained from all patients.
The primary aim of the double-blind study was to demonstrate the superiority of 5-ALA–induced fluorescence cystoscopy compared with white light cystoscopy with regard to the period of time between randomization and the first tumor recurrence or the first tumor occurrence detected by means of control cystoscopies performed within 3 to 12 months after randomization. Secondary endpoints included the recurrence rate at 3, 6, 9, and 12 months; the number of patients with progressive disease during the first year of follow-up; the number of tumor lesions detected; and the safety of 5-ALA.
All patients who met the inclusion criteria and did not violate any exclusion criteria were randomized to receive, in a blinded manner, either 5-ALA or a placebo to be instilled into the urinary bladder before transurethral inspection and subsequent tumor resection.
Preparation of both the 5-ALA solution and the placebo solution before instillation was identical. One vial contained 1.5 g of 5-ALA or the placebo, each reconstituted by adding 1 vial of solvent containing 50 mL 0.9% sodium chloride solution. Not later than 2 hours after reconstitution, this solution was instilled into the bladder as a single dose before cystoscopy. The placebo consisted only of 50 mL of 0.9% sodium chloride and was instilled accordingly. Instillation times between 1 hour (minimum) and 4 hours (maximum) were to be achieved. An instillation time of 2 hours was considered optimal.
Transurethral inspection of the bladder with white light and fluorescent light was performed 2 hours (range, 1-4 hours) after instillation of 5-ALA or the placebo, followed by tumor resection. Only clearly identified tumor-suspicious tissue and tumor areas were resected without performing random biopsies, in accordance with the recommendations of the American and European Association of Urology.15, 16
Two to 4 weeks after the first resection, only those patients with reference pathology controlled pT1 G2-3 or T2 (with no indication for cystectomy) tumors underwent a second resection under white light. The first and second resections were performed by the same surgeon.
Adjuvant therapy using BCG instillation
Four weeks after the last resection, patients with confirmed high-risk nonmuscle-invasive tumors (CIS, pTa G3, or pT1 G2-3) received the first BCG instillation. The adjuvant therapy was to be continued over a period of 6 weeks with 1 instillation per week.
Generally, all patients within the full-analysis set were subject to control cystoscopies every 3 months. Consequently, control cystoscopies were to be performed in all patients with atypia, pTa, pT1, and/or CIS—except those who required immediate cystectomy based on the result of the primary resection—at 3, 6, 9, and 12 months after the date of randomization irrespective of any potential second resection. In patients diagnosed to be tumor free at the first resection, follow-up cystoscopies were performed on the same schedule. These cystoscopies were carried out under white light until the end of the study (12 months after randomization), or until tumor recurrence or first occurrence.
To maintain the double-blind design of this trial, all follow-up examinations were performed by a second individual trial team. Members of this team did not know the original procedure that had been carried out (placebo arm cystoscopy or 5-ALA cystoscopy). This was ensured and organized by using a second individual set of case report forms.
The study endpoint was reached either when a new pathologically confirmed tumor lesion was detected within the follow-up period or when the patient had passed the 1-year follow-up.
Handling of resected specimens
Biopsies obtained by endoscopy were sent to the reference pathologist within 24 hours after surgery for examination. Diagnostic investigations were carried out according to the international standard criteria of the Union Internationale Contre le Cancer and WHO.17, 18 The reference pathologists acted as blinded observers. Lacking any information on the diagnostic procedure previously performed, the reference pathologists evaluated the tissue specimens. Determination of the primary study endpoint and all subsequent therapy decisions were based solely on the reference pathologists' classification.
The study was designed as a double-blind trial. During the whole trial, neither the patient, the physicians involved in performing transurethral resections, nor the sponsor staff were aware of the study arm administered. This included anyone determining subject eligibility, evaluating endpoints, or assessing compliance with the protocol.
For this trial, 2 independent teams of investigators were required. The first team was responsible for patient registration as well as for the first and second transurethral resections. The second team was responsible for evaluating the primary study endpoint within the control cystoscopies and did not know whether the original procedure had been carried out under white light or fluorescent light.
All data were subjected to descriptive analyses. The confirmatory proof of efficacy was performed within the full-analysis set defined according to the intent-to-treat principle as all patients who 1) were randomized into 1 cystoscopy group; 2) actually underwent primary cystoscopy; and 3) were found to have no tumor, atypia, pTa, pT1, or CIS, except those patients who required immediate cystectomy based on the result of the primary resection. All randomized patients who had received study medication and had a documented cystoscopy were valid for safety evaluation.
The primary target criterion of time to tumor recurrence was analyzed using Kaplan-Meier techniques. A sample size of 170 patients per group was estimated with a 2-sided confirmatory statistical test to detect a clinically relevant difference of 15% between the 2 groups 12 months after tumor resection. This minimal relevant difference of 15% was also chosen in the light of the full analysis as the main confirmatory analysis of the data, which may have diluted the treatment effect compared with the per-protocol set. The null hypothesis of equality of the survival curves of both groups was tested against the superiority of the fluorescence procedure. A 2-tailed significance test with a probability of erroneously rejecting the null hypothesis of 5% was performed. The associated confirmatory test statistic was the log-rank statistic applied in its crude version in the full-analysis set.19 All secondary study endpoints were analyzed descriptively using tables, graphs, listings, and exploratory nonparametric statistical tests and confidence intervals, where appropriate.
Of the 381 patients randomized, 370 received 5-ALA or the placebo and were included in the safety analysis (Fig. 1).
Patient enrollment between study centers differed largely, and was between 2 and 136 patients per center. Of the 8 participating centers, 4 enrolled <40 patients each. A total of 359 patients were valid for efficacy analysis (placebo arm cystoscopy = 176; 5-ALA cystoscopy = 183). Twenty-two (5.8%) patients were excluded from the efficacy analysis because of a missing cystoscopy (n = 11, 2.9%), tumor ≥pT2 (n = 8, 2.1%), and indication for cystectomy after first resection (n = 3, 0.8%).
Of the 359 patients included in the efficacy analysis, 100 were women and 259 were men, with a mean age of 66 ± 12 years, a mean weight of 78 ± 15 kg, and a mean height of 171 ± 9 cm. Of all patients valid for efficacy analysis, 211 (58.8%) belonged to the low-risk group and 148 (41.2%) to the high-risk group. Six (1.7%) patients had received adjuvant instillation therapy with BCG within the 3 months before study enrollment. Median 5-ALA instillation duration was 2.2 hours (range, 0-5 hours). No relevant differences were observed between the 2 groups with regard to any baseline variables.
The results of tumor histology at first resection are shown in Table 1. The mean number of tumor specimens per patient was higher with 5-ALA cystoscopy (1.8) than with placebo arm cystoscopy (1.6). The difference was not significant (P = .1178). Furthermore, the mean number of resected specimens evaluated per patient was higher with 5-ALA cystoscopy (2.6) than with placebo arm cystoscopy (2.2), and the difference was statistically significant (P < .0091; Wilcoxon-Mann-Whitney test). Slightly more tumors were detected with 5-ALA cystoscopy than with placebo arm cystoscopy (88.5% vs 84.7%).
|Parameter||Placebo Cystoscopy, n=176||5-ALA Cystoscopy, n=183||Total, n=359|
|Number or resections, No.||381||475||856|
|Number of tumor specimens, No.||281||337||618|
|Number or resection specimens, mean (range)||2.2 (0-10)||2.6 (0-10)||2.4 (0-10)|
|Number or tumor specimens per patient, mean (range)||1.6 (0-10)||1.8 (0-8)||1.7 (0-10)|
|Biopsy histology, No. (%)|
|Any tumor||149 (84.7)||162 (88.5)||311 (86.6)|
|Carcinoma in situ, isolated||3 (1.7)||3 (1.6)||6 (1.7)|
|Carcinoma in situ, concomitant||19 (10.8)||22 (12.0)||41 (11.4)|
|pTaG1||50 (28.4)||61 (33.3)||111 (30.9)|
|pTaG2||35 (19.9)||35 (19.1)||70 (19.5)|
|pTaG3||0 (0.0)||2 (1.1)||2 (0.6)|
|pT1G1||1 (0.6)||2 (1.1)||3 (0.8)|
|pT1G2||15 (8.5)||16 (8.7)||31 (8.6)|
|pT1G3||23 (31.1)||19 (10.4)||42 (11.7)|
|pT2||8 (4.5)||10 (5.5)||18 (5.0)|
|Atypia I||5 (2.8)||3 (1.6)||8 (2.2)|
|Atypia III||1 (0.6)||3 (1.6)||4 (1.1)|
|pTxG1||0 (0.0)||1 (0.5)||1 (0.3)|
|pTxG2||2 (1.1)||1 (0.5)||3 (0.8)|
|pTxG3||2 (1.1)||2 (1.1)||4 (1.1)|
|No histology, no suspicious tumor||3 (1.7)||1 (0.5)||4 (1.1)|
|Not evaluable/missing||1 (0.6)||3 (1.6)||4 (1.1)|
|Multifocal tumors, No. (%)||64 (36.4)||79 (43.2)||143 (39.8)|
The percentages of diagnoses with isolated CIS were low (5-ALA n = 3, 1.6%; placebo arm n = 3, 1.7%); those with concomitant CIS were 10.8% (5-ALA) and 12.0% (placebo arm). Multifocal tumors were detected in 143 (39.8%) patients, and the diagnosis was made more frequently with 5-ALA cystoscopy (43.2%) than with placebo arm cystoscopy (36.4%).
Of all the patients, 13.1% had CIS lesions diagnosed at first resection. Neither the number of any CIS lesions diagnosed nor the different staging and grading categories differed between the 2 groups to any relevant degree.
Lesion detection rates comparing white light and fluorescent light cystoscopy in patients undergoing 5-ALA cystoscopy
Generally, intrapatient comparison showed that fluorescent light cystoscopy detected more lesions than white light cystoscopy (Table 2). In lesions occurring in at least 10 patients, fluorescent light cystoscopy had higher detections rates for CIS, pTaG1, pTaG2, pT1G3, pT2, atypia I, and atypia III. White light cystoscopy had a slightly higher detection rate compared with fluorescent light cystoscopy for pT1G2 lesions only.3
|Histology||White Light Cystoscopy, n/N (%)||Fluorescence Light Cystoscopy, n/N (%)|
|Carcinoma in situ||32/49 (65.3)||49/49 (100.0)|
|pTaG1||98/117 (83.8)||112/117 (95.7)|
|pTaG2||57/70 (81.4)||64/70 (91.4)|
|pTaG3||2/2 (100.0)||2/2 (100.0)|
|pT1G1||1/2 (50.0)||1/2 (50.0)|
|pT1G2||21/22 (95.5)||19/22 (86.4)|
|pT1G3||27/31 (87.1)||31/31 (100.0)|
|pT2||17/20 (85.0)||18/20 (90.0)|
|Atypia I||13/31 (41.9)||30/31 (96.8)|
|Atypia III||2/3 (66.7)||3/3 (100.0)|
|pTxG1||1/1 (100.0)||0/1 (0.0)|
|pTxG2||1/1 (100.0)||1/1 (100.0)|
|pTxG3||2/2 (100.0)||1/2 (50.0)|
|Parameter||Placebo Cystoscopy, n=36||5-ALA Cystoscopy, n=37||Total, n=73|
|Number or resections, No.||81||99||180|
|Number of tumor specimens, No.||36||53||89|
|Number or resection specimens, mean (range)||2.3 (0-6)||2.1 (1-10)||2.5 (0-10)|
|Number or tumor specimens per patient, mean (range)||1.0 (0-6)||1.4 (0-7)||1.2 (0-7)|
|Biopsy histology, No. (%)|
|Any tumor||17 (47.2)||24 (64.9)||41 (66.2)|
|Carcinoma in situ||6 (16.7)||4 (10.8)||10 (13.7)|
|pTaG1||0 (0.0)||1 (2.7)||1 (1.4)|
|pTaG2||2 (5.6)||2 (5.4)||4 (5.5)|
|pTaG3||0 (0.0)||2 (5.4)||2 (2.7)|
|pT1G2||0 (0.0)||3 (8.1)||3 (4.1)|
|pT1G3||1. (2.8)||5 (13.5)||6 (8.2)|
|pT2||3 (8.3)||0 (0.0)||3 (4.1)|
|>pT2||0 (0.0)||1 (2.7)||1 (1.4)|
|Atypia I||3 (8.3)||5 (13.5)||8 (11.0)|
|Atypia II||1 (2.8)||0 (0.0)||1 (1.4)|
|pTxG3||0 (0.0)||1 (2.7)||1 (1.4)|
|Not evaluable/missing||1 (2.8)||0 (0.0)||1 (1.4)|
|Carcinoma in situ, No. (%)|
|Isolated||6 (16.7)||4 (10.8)||10 (13.7)|
|Concomitant||4 (11.1)||7 (18.9)||11 (15.1)|
|Multifocal tumors, No. (%)||9 (25.0)||11 (29.7)||20 (27.4)|
Tumor histology after second resection
The mean number of tumor specimens per patient was higher with 5-ALA cystoscopy (1.4) than with placebo arm cystoscopy (1.0) (Table 3). Also, the mean number of resected specimens evaluated per patient was higher with 5-ALA cystoscopy (2.7) than with placebo arm cystoscopy (2.3); this difference was not statistically significant. Tumors were diagnosed more often with 5-ALA cystoscopy (64.9%) than with placebo arm cystoscopy (47.2%). Multifocal tumors were detected in 20 (27.4%) patients, and the percentage was slightly higher with 5-ALA cystoscopy (29.7%) than with placebo arm cystoscopy (25.0%).
Table 4 summarizes lesions that would otherwise not have been detected by white light cystoscopy. These include 55.9% of atypia, 34.7% of CIS, 16.9% of pTa, 15.0% of pT2, and 10.9% of pT1.
|Carcinoma in situ||17/49 (34.7)|
Recurrence-free survival rates at 12 months were 64.0% (5-ALA cystoscopy) and 72.8% (placebo arm cystoscopy); the difference was not statistically significant (P = .2216) (Fig. 2). Subgroup analyses by risk group, by presence of histologically verified high-risk or low-risk tumor at first resection, and by study center yielded similar results (data not shown). Generally, Kaplan-Meier curves separated slightly after 3 months in favor of placebo arm cystoscopy.
Rates of progression-free survival at 12 months were nearly identical: 89.4% for 5-ALA cystoscopy and 89.0% for placebo arm cystoscopy (P = .9101) (Fig. 3).
Of the 370 patients who underwent cystoscopy with 5-ALA or the corresponding placebo (safety population), 123 (33.2%) reported adverse events (white light, 33.9%; fluorescent light, 32.6%) (Table 5). Nearly all adverse events were related to the urogenital system and were reported in 121 (32.7%) patients. The most frequent adverse events were hematuria, dysuria, urinary frequency/urgency, and bladder spasms, which are all symptoms that occur frequently after transurethral tumor ablation.
|Adverse Events||Placebo Cystoscopy, n=183, No. (%)||5-ALA Cystoscopy, n=187, No. (%)||Total, n=370, No. (%)|
|Number of patients with adverse events||62 (33.9)||61 (32.6)||123 (33.2)|
|Urogenital systema||61 (33.3)||60 (32.1)||121 (32.7)|
|Hematuria||50 (27.3)||52 (27.8)||102 (27.6)|
|Dysuria||18 (9.8)||25 (13.4)||43 (11.6)|
|Urinary frequency/urgency||18 (9.8)||14 (7.5)||32 (8.6)|
|Bladder spasms||19 (10.4)||10 (5.3)||29 (7.8)|
|Urinary retention||4 (2.2)||7 (3.8)||11 (3.0)|
|Ureteral obstruction||2 (1.1)||2 (1.1)||4 (1.1)|
|Hemorrhagic cystitis||3 (1.6)||0 (0.0)||3 (0.8)|
|Incontinence||1 (0.5)||0 (0.0)||1 (0.3)|
|Gastrointestinal systema||0 (0.0)||2 (1.1)||2 (0.5)|
|Nausea||0 (0.0)||2 (1.1)||2 (0.5)|
|Vomiting||0 (0.0)||1 (0.5)||1 (0.3)|
|Pulmonary/respiratory systema||0 (0.0)||2 (1.1)||2 (0.5)|
|Dyspnea||0 (0.0)||1 (0.5)||1 (0.3)|
|Partial oxygen tension decreased||0 (0.0)||1 (0.5)||1 (0.3)|
|Cardiovascular systema||0 (0.0)||1 (0.5)||1 (0.3)|
|Hypertension||0 (0.0)||1 (0.5)||1 (0.3)|
|Nervous systema||0 (0.0)||1 (0.5)||1 (0.3)|
|Insomnia||0 (0.0)||1 (0.5)||1 (0.3)|
|Skina||1 (0.5)||0 (0.0)||1 (0.3)|
|Allergy||1 (0.5)||0 (0.0)||1 (0.3)|
|Fever/infection/influenzalike symptomsa||0 (0.0)||1 (0.5)||1 (0.3)|
|Chills||0 (0.0)||1 (0.5)||1 (0.3)|
|Sweating||0 (0.0)||1 (0.5)||1 (0.3)|
|General conditionsa||1 (0.5)||0 (0.0)||1 (0.3)|
|Not specified||1 (0.5)||0 (0.0)||1 (0.3)|
Six adverse events (5-ALA 4; placebo 2) were serious and included pulmonary embolism in 2 patients (5-ALA 1; placebo 1), angina pectoris in 1 (5-ALA), compression of the cervical vertebra in 1 (5-ALA), circulatory collapse in 1 (placebo arm), and apoplexy in 1 (5-ALA). None of them was related to the study drug. Four adverse events (placebo arm 2; 5-ALA 2) had a fatal outcome and occurred between 3.5 and 17 months after first surgery during follow-up: 2 cases of pulmonary embolism, 1 circulatory collapse, and 1 stroke. No death was drug related.
Changes in mean laboratory values assessed before surgery and on the second postoperative day were minor and without clinical relevance. In comparison to the placebo arm, the instillation of 5-ALA was safe and well tolerated.
5-ALA is the first topical photosensitizer that can be instilled into the bladder to improve optical tumor detection during transurethral resection of bladder tumors.5-9 Photodynamic diagnosis-supported transurethral resection of bladder tumors led to an improvement in the number of tumors found at the time of resection as well as in the number of patients being tumor free at the time of the second resection after 2 to 6 weeks.11 In recent years, several centers have reported improved results in long-term recurrence-free survival if transurethral resection of bladder tumors was combined with 5-ALA. Babjuk et al reported a statistically significant recurrence-free survival benefit in 122 patients at 1 year for photodynamic diagnosis (66%) versus conventional white light-guided transurethral resection of bladder tumors (39%).12 A similar benefit was seen by Filbeck et al in 191 patients (89.6 vs 73.8% in favor of photodynamic diagnosis).13 Extending the follow-up to 2, 4, 5, and 8 years in these and other studies has consistently shown the superiority of photodynamic diagnosis-guided cystoscopy during transurethral resection of bladder tumors with regard to recurrence-free survival.12, 13, 20, 21
The primary study endpoint, the superiority of 5-ALA cystoscopy versus placebo arm cystoscopy with regard to recurrence-free survival at 12 months, was not met in this study (5-ALA 64.0% vs placebo arm 72.8%). Although numerically in favor of placebo arm cystoscopy, the results did not show any statistically significant difference between the 2 groups, neither for the respective total patient populations nor for any subgroups analyzed (risk groups, or histologically verified tumor at first resection with either low or high risk). Also, progression-free survival after 12 months did not show any relevant difference between the groups.
This is the first controlled randomized multicenter study that does not show any benefit from photodynamic diagnosis with regard to 12-month recurrence-free survival, despite the finding that more tumors in general and multifocal tumors in particular were detected in the 5-ALA group (88.5% and 43.2% for 5-ALA vs 84.7% and 36.4% for the placebo arm, respectively). Several reasons may be given to help explain these unexpected findings that are contradictory to the findings of previous studies. All previous outcome studies were from single centers with a long experience of 5-ALA and photodynamic diagnosis. This study involved urologists and hospitals with previous exposure to photodynamic diagnosis ranging from no practice at all to up to 10 years of experience. Stratifying the statistical analysis of recurrence-free survival by center, study urologist, or amount of experience of the participating centers revealed no significant difference for any subgroup.
There was also a low incidence in both primary as well as concomitant CIS compared with other studies in the literature.22 Detection of CIS, which is a more aggressive disease that may not always be visible to the urologic surgeon intraoperatively, is regarded as an advantage of photodynamic diagnosis.23 In this study, CIS was also detected at second resection more often in the placebo group than in the 5-ALA group (16.7 vs 10.8%); however, because of the low overall number of CIS cases, this did not reach statistical significance.
The lack of difference observed in our study is difficult to explain. It may either be because of the biology of the tumors detected under fluorescent light or because of the efficacy of the adjuvant postoperative treatment. There were no relevant differences in baseline tumor and demographic characteristics between the placebo and 5-ALA cystoscopy groups, which could have explained the discrepancies observed in this study compared with other published data.
During the initial treatment, both study groups underwent a more thorough endoscopic inspection of the bladder with white light and fluorescent light, with an overall recurrence rate at 12 months of 20% (73 of 359 patients), which is lower than in some reports.2, 3 It may be that this study design led to a higher overall tumor detection than what would occur in the normal clinical situation and thus produced no statistically significant difference in the 12-month outcome data in both groups. Under this assumption, there should not be more tumors in the fluorescent light group after switching to 5-ALA fluorescence than in the white light group after switching to placebo arm blue light. Intrapatient comparison of fluorescent light versus white light cystoscopy revealed, however, that 5-ALA fluorescence-guided cystoscopy detected more lesions than white light cystoscopy. Thus the higher tumor detection rate of fluorescent light in comparison to white light could be demonstrated. The tumor detection rate in this study is similar to several other reports using 5-ALA fluorescence as well as hexaminolevulinate (Table 4).22
Does it have to be concluded from this study that photodynamic diagnosis is useless in reducing tumor recurrences? Photodynamic diagnosis has benefited from 5-ALA being the first topical photosensitizer that can be instilled into the bladder with practically no side effects from the substance, as confirmed by this study. However, to demonstrate an optimal difference in cellular uptake between normal and tumor urothelium, patients should retain the substance instilled into the bladder for 2 to 4 hours, with better results in those patients with a longer instillation time.24 For many patients with an irritated tumor-bearing bladder, who usually underwent cystoscopy a few days earlier and who may be nervous before surgery, withholding the instilled photosensitizer for such a long period may not be possible. Newly developed photosensitizers work at lower concentrations and show an accelerated cellular uptake, resulting in an approximately 50% shorter instillation time.25
It must therefore be concluded from this study that although intraoperative tumor detection was improved practically without side effects from photodynamic diagnosis, there was no benefit in the 12-month recurrence-free and progression-free survival using 5-ALA as a photosensitizer. It can be speculated that despite convincing results in single-center surgical studies, only multicenter studies might reveal the true benefit of a new method such as photodynamic diagnosis. However, other topical photosensitizers with better biological performance have recently been introduced and may still prove, when tested in a similar design to this study, that photodynamic diagnosis not only increases intraoperative tumor detection rate but also the recurrence-free survival of patients with nonmuscle-invasive bladder cancer.25-27
Twelve months after tumor resection, no differences with regard to recurrence-free or progression-free survival rates were observed between the 5-ALA and placebo groups using both white light and fluorescent light cystoscopy. Although more superficial urinary bladder tumors were detected in the 5-ALA group, the higher detection rate did not translate into differences in 12-month outcome.
CONFLICT OF INTEREST DISCLOSURES
This study was sponsored by medac Gesellschaft fur klinische Spezialpraparate mbH, Hamburg, Germany, which was involved in the design, conduct, monitoring, data management, and statistical analysis of the trial.
- 1Surveillance, Epidemiology, and End Result Program: SEER Cancer Statistics Review, 1975-2006. Bethesda, MD: National Cancer Institute. Available at:http://seer.cancer. gov/csr/1976_2006/ Accessed March 8, 2010.
- 6Clinical 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. 1997; 20: 402-408., , , et al.
- 14Intravesical 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. European Organization for Research and Treatment of Cancer Genitourinary Group. J Urol. 1995; 153(3 pt 2): 934-941., , , et al.
- 18TNM Classification of Malignant Tumors. 5thed. New York, NY: Wiley-Liss; 1997.,
- 19Modelling Survival Data in Medical Research. 2nd ed. London, UK: Chapman & Hall/CRC; 1994.