Comparison between intravesical and oral administration of 5-aminolevulinic acid in the clinical benefit of photodynamic diagnosis for nonmuscle invasive bladder cancer

Authors


Abstract

BACKGROUND:

This study was undertaken to evaluate the clinical value of photodynamic diagnosis (PDD) with intravesical and oral instillation of 5-aminolevulinic acid (ALA) (ALA-PDD), and transurethral resection of bladder tumor (TURBT) guided by ALA-PDD (PDD-TURBT) for nonmuscle invasive bladder cancer.

METHODS:

Of all 210 cases, 75 underwent PDD with intravesically applied ALA, and 135 cases underwent PDD with orally applied ALA. Diagnostic accuracy was evaluated by comparing the level on images of ALA-induced fluorescence with the pathological result. PDD-TURBT was performed in 99 completely resectable cases corresponding to 210 ALA-PDD cases. To evaluate the abilities of PDD-TURBT, survival analysis regarding intravesical recurrence was retrospectively compared with the historical control cases that underwent conventional TURBT.

RESULTS:

The diagnostic accuracy and capability of ALA-PDD were significantly superior to those of conventional endoscopic examination. Moreover, 72.1% of flat lesions, including dysplasia and carcinoma in situ, could be detected only by ALA-PDD. The recurrence-free survival rate in the cases that underwent PDD-TURBT was significantly higher than that of conventional TURBT. Moreover, multivariate analysis revealed that the only independent factor contributing to improving prognosis was PDD-TURBT (hazard ratio, 0.578; P = .012). Regardless of the ALA administration route, there was no significant difference in diagnostic accuracy, ability of PDD, or recurrence-free survival. All procedures were well tolerated by all patients without any severe adverse events.

CONCLUSIONS:

This multicenter study is likely to be biased, because it is limited by the retrospective analysis. This study suggests that regardless of the ALA administration route, ALA-PDD and PDD-TURBT are remarkably helpful in detection and intraoperative navigation programs. Cancer 2012;. © 2011 American Cancer Society.

INTRODUCTION

Bladder cancer is the second most common genitourinary neoplasm, with more than 60,000 and 120,000 new cases diagnosed each year in the United States and Europe, respectively.1, 2 In Japan, about 16,000 new cases are diagnosed and 50,000 endoscopic surgeries are performed each year.3 The standard therapy for nonmuscle invasive cancer, accounting for approximately 70% bladder cancer, is transurethral resection of bladder tumor (TURBT).4 TURBT enables a high quality of life, with preservation of the bladder and a good prognosis. However, TURBT results in frequent residual tumor, resulting in frequent subsequent intravesical recurrence in the early postoperative period. The high recurrence rate is attributed to residual lesions, such as minute lesions, flat lesions, and concomitant flat lesions with raised lesions. In particular, flat tumors, such as carcinoma in situ (CIS) and dysplasia, are difficult to detect accurately by cystoscopy; thus, it is no exaggeration to say that they are endoscopically invisible lesions.

5-Aminolevulinic acid (ALA) has received much attention as a new-generation photo-sensitive substance for photodynamic diagnosis (PDD) in recent years. ALA is an endogenous natural amino acid, and a common precursor of chlorophyll in plants and bilirubin in animals. The administered photosensitive substance, ALA, is incorporated by cells and synthesized into a fluorescent substance, protoporphyrin IX. In various cancer cells, this protoporphyrin IX biosynthesis pathway is promoted, whereas the protoporphyrin IX-metabolizing pathway is inhibited, resulting in the excess accumulation of protoporphyrin IX in cancer cells,5, 6 and the tumor selectivity is 17:1 in the urothelium, which is particularly high.7 Because protoporphyrin IX exhibits photoactivity, when protoporphyrin IX is excited by irradiation with a specific wavelength of light, mainly visible blue light (375-445 nm), it emits red fluorescence. Cancer cells can be accurately identified by detecting this fluorescence.8 This is the mechanism of PDD using ALA. This means that PDD mediated by ALA (ALA-PDD) is the most advanced photodynamic technology based on the fundamental biological profile of cancer cells, providing good visualization and precise detection of the lesions, leading to improved surgical curability and subsequent prognosis in various cancers, including bladder cancer.

Recently, orally applied ALA was approved as an optical imaging agent to enhance intraoperative detection of malignant glioma in Europe.9 Moreover, the hexyl ester derivative of 5-ALA (hexaminolevulinate), which was applied intravesically, was approved as an optical imaging agent to enhance intraoperative detection of bladder cancer, in particular CIS in Europe and the United States.9 Since then, excellent results have been reported for with hexaminolevulinate in the diagnostic accuracy10-17 and TURBT guided by PDD with hexaminolevulinate (PDD-TURBT) in the prognosis18-22 of nonmuscle invasive bladder cancer. Several prospective, randomized, multicenter studies have recently shown the contribution of PDD-TURBT to the improvement in outcome,23, 24 but there are still some points to discuss.25 Moreover, it was demonstrated in the retrospective series that, although PDD with ALA and hexaminolevulinate applied intravesically was demonstrated to be significantly superior to white light cystoscopy, there were no significant differences between ALA and hexaminolevulinate in clinical outcome such as residual tumor and recurrence-free survival.26 Thus, in this study, we retrospectively evaluated the value of ALA-PDD and also PDD-TURBT, and whether the differences depending on the ALA administration route affect the diagnostic accuracy, ability, and recurrence-free survival in nonmuscle invasive bladder cancer.

MATERIALS AND METHODS

Patients

PDD with intravesical instillation of ALA was approved by the ethics committees of Kochi Medical School in September 2004, and PDD with oral instillation of ALA was approved in January 2007.

In this study, all patients who were candidates for transurethral biopsy of the bladder or TURBT were enrolled after providing written informed consent in the Department of Urology of Kochi Medical School Hospital, Kochi National Hospital and Chikamori Hospital between September 2004 and August 2010. All patients were informed about the potential efficacy and also adverse events of ALA-PDD, for example, bladder irritability, such as urinary frequency and urgency, and systemic response, such as skin photosensitivity, transient elevation of serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT), nausea, and vomiting in conformity with the Common Terminology Criteria for Adverse Events version 3.0.27

ALA-PDD was performed in 210 patients, of whom 172 were men and 38 were women, with a median age of 70.6 (range, 44-90) years; 98 were primary cases, and 112 were recurrent cases of bladder cancer. There was no statistically significant difference in these background factors between 75 cases of intravesical administration of ALA and 135 cases of oral administration of ALA (Table 1).

Table 1. Patient Characteristics in Photodynamic Diagnosis
VariableAll CasesALA Administration
IntravesicalOralP
  • Abbreviations: ALA, 5-aminolevulinic acid; BCG, Bacillus Calmette-Guerin; CIS, carcinoma in situ; TURBT, transurethral resection of bladder tumor.

  • Patient characteristics, including number of patients, examination period, age, sex, past history, prior therapy and tumor stage, are shown in all 210 cases. There was no statistical significance in these background factors between 75 cases of intravesical administration of ALA and 135 cases of oral administration of ALA.

  • a

    Fisher exact test (2 × 2).

  • b

    Chi-square test.

Patients21075135 
Examination periodSeptember 2004-August 2010September 2004-August 2010January 2007-August 2010 
Age, y   .894a
 Mean70.671.270.3 
 Range44-9044-8845-90 
Sex   .979a
 Men17262110 
 Women381325 
Past history   .852a
 Primary case1495594 
 Recurrent case612041 
Prior therapy   .061a
 TURBT651847 
 TURBT+BCG27225 
Tumor stage   .154b
 Normal301911 
 pT    
 is471334 
 a (+CIS)101 (29)26 (5)75 (24) 
 1 (+CIS)65 (18)27 (8)28 (10) 

PDD-TURBT was also performed in 99 completely resectable cases corresponding to 210 PDD cases of bladder cancer, of whom 80 were men and 19 were women, with a median age of 69.8 (range, 46-90) years; 58 were primary cases, and 41 were recurrent cases of nonmuscle invasive bladder cancer. To evaluate the abilities of PDD-TURBT, survival analysis regarding vesical recurrence was retrospectively examined compared with historical control cases that had undergone conventional TURBT under white light guidance. There was no significant difference in any variables of patient characteristics between fluorescence TURBT and conventional TURBT (Table 2).

Table 2. Patient Characteristics in Fluorescence TURBT and Conventional TURBT
Variable32Fluorescence TURBTConventional TURBTP
  • Abbreviations: BCG, Bacillus Calmette-Guerin; CIS, carcinoma in situ; rec, recurrence; TURBT, transurethral resection of bladder tumor.

  • Patient characteristics including number of patients, examination period, age, sex, past history, prior therapy, adjuvant therapy, and the factors based on the European Organization for Research and Treatment of Cancer risk tables32 are shown for 99 cases of fluorescence TURBT and 99 cases of conventional TURBT. There was no significant difference in any variables of patient characteristics between fluorescence TURBT and conventional TURBT.

  • a

    Fisher exact test (2 × 2).

  • b

    Chi-square test.

Patients9999 
Examination periodSeptember 2004-August 2010May 1982-July 2008 
Age, y  .399a
 Mean69.871.7 
 Range46-9036-92 
Sex  .100a
 Men8069 
 Women1930 
Tumors, No.  .420b
 Single4652 
 2-74845 
 ≥852 
Tumor size  .510a
 <3 cm8985 
 ≥3 cm1014 
Prior recurrence  .160b
 Primary5858 
 ≤1 rec/y3224 
 >1 rec/y917 
T category  .100a
 Ta6957 
 T13042 
Concomitant CIS  .260a
 No7885 
 Yes2114 
Tumor grade  .400b
 G1510 
 G25955 
 G33534 
Prior therapy  .820a
 TURBT2625 
 TURBT+BCG1516 
Adjuvant therapy  .120a
 None7988 
 Intravesical BCG2011 

The patient characteristics of 32 cases of intravesical administration of ALA and 67 cases of oral administration of ALA are shown in Table 3. The presence of concomitant CIS and high tumor grade in fluorescence TURBT with oral administration of ALA was statistically greater than in fluorescence TURBT with intravesical administration of ALA.

Table 3. Patient Characteristics in Fluorescence TURBT With Intravesically Applied ALA and Orally Applied ALA
Variable32Intravesical ALAOral ALAP
  • Abbreviations: ALA, 5-aminolevulinic acid; BCG, Bacillus Calmette-Guerin; CIS, carcinoma in situ; rec, recurrence; TURBT, transurethral resection of bladder tumor.

  • Patient characteristics including number of patients, examination period, age, sex, past history, prior therapy, adjuvant therapy, and the factors based on the European Organization for Research and Treatment of Cancer risk tables32 are shown for 32 cases of intravesical administration of ALA and 67 cases of oral administration of ALA. The presence of concomitant CIS and high tumor grade in fluorescence TURBT with oral administration of ALA is statistically greater than that in fluorescence TURBT with intravesical administration of ALA.

  • a

    Fisher exact test (2 × 2).

  • b

    Chi-square test.

Patients3267 
Examination periodOctober 2004-December 2007March 2007-August 2010 
Age, y  .896a
 Mean73.368.1 
 Range46-8749-90 
Sex  .370a
 Male2753 
 Female514 
Tumors, No.  .454b
 Single1334 
 2-71829 
 ≥814 
Tumor size  .412a
 <3 cm2861 
 ≥3 cm46 
Prior recurrence  .160b
 Primary1939 
 ≤1 rec/y1121 
 >1 rec/y27 
T category  .199a
 Ta1049 
 T11218 
Concomitant CIS  .038a
 No2949 
 Yes318 
Tumor grade  .005b
 G141 
 G21940 
 G3926 
Prior therapy  .566a
 TURBT818 
 TURBT+BCG510 
Adjuvant therapy  .146a
 None2851 
 Intravesical BCG416 

Administration of ALA

For PDD, we used ALA as a photosensitizer. ALA hydrochloride (Cosmo Bio Co., Tokyo, Japan) was dissolved in 50 mL of 5% glucose solution, and 8.4% sodium hydrogen carbonate (NaHCO3) solution was added to adjust to pH 7.8 to 8.0. In 210 patients, 75 cases underwent PDD with intravesical instillation of 1.5 g ALA 1.5 hours before endoscopic examination, and 135 cases underwent PDD with oral instillation of 1.0 g ALA 3.0 hours before endoscopic examination.

PDD system

For ALA-PDD, a D-LIGHT System (Karl Storz GmbH & Co., Tuttlingen, Germany), including D-Light C, CCU Tricam SLII/3CCD CH Tricam-P PDD, and a HOPKINSII PDD telescope (30°), was used. The light source, D-Light C (300 W xenon arc lamp), is equipped with a band-pass filter that is designed to transmit blue light (excitation wavelength, 375-445 nm) (for excitation of fluorescence). The video camera system, CCU Tricam SLII/3CCD CH Tricam-P PDD, is equipped with a long-pass filter that is designed to cutoff blue light (for observation of fluorescence; fluorescence emission wavelength, 600-740 nm). This PDD system has the advantage that it can instantly switch between blue light mode for fluorescent observation and white light mode for conventional observation.

Examination Procedure

Under conventional white light and fluorescence light guidance, tumor locations were recorded and cold cup biopsies were taken. If cases were endoscopically completely resectable, tumorous lesions under white light guidance and lesions with fluorescent excitation under blue light (fluorescence) guidance were resected sequentially. First, biopsy using a cold cup was performed. After conventional systematic biopsy, specimens of the vesical mucosa emitting right fluorescence or with an abnormality under the white light source were collected from 8 vesical regions (neck of the urinary bladder, triangular region, posterior, left, and right walls, apex, anterior wall, and prostatic region of the urethra). When a tumor occupied these regions, the tumor tissue was collected. The specimens were categorized and recorded by fluorescence intensity-based evaluation using the blue light mode and macroscopic malignancy evaluation using the conventional white light mode. In the evaluation using blue light mode, the samples were evaluated by roughly dividing them into 3 categories following the semiquantitative macroscopic diagnostic method of red fluorescence emission used in a clinical study on brain tumors performed by Miyoshi et al28: none (no fluorescence emission), weak (weak fluorescence emission), and strong (strong fluorescence emission), based on the red fluorescence intensity. In the evaluation using the conventional white light mode, samples were evaluated by roughly dividing them into 3 categories based on comprehensive macroscopic malignancy in consideration of important features such as the mucosal properties and concentration of blood vessels: none (no abnormal finding), weak (mild abnormality with difficulty in judging benignity or malignancy), and strong (marked abnormality with a high possibility of malignancy). These evaluations were made by the 3 same instructors certified by the Japanese Urological Association in all examinations, and high-level reproducibility was demonstrated in the previous report.29

After hemostasis of the biopsied region, the tumor was resected using a resectoscope. In the test design, first the tumor was resected under the conventional white light source. The light source was then changed to the fluorescence, and the excited region was in addition resected. Operations were performed by 3 physicians certified by the Japanese Urological Association under instruction by the same 3 instructors certified by the Japanese Urological Association in all operations.30

Diagnostic accuracy based on semiquantitative evaluation was analyzed by comparing the level on images of ALA-induced fluorescence with the pathological diagnosis according to the General Rule for Clinical and Pathological Studies on Bladder Cancer, third edition.31 Diagnostic capability was assessed by the area under the receiver operative characteristic curve (AUC) in PDD compared with that in conventional white light endoscopic examination. These comparisons were analyzed using Fisher exact test (2 × 2), chi-square test, 2-sample test for equality of proportions, and the Wilcoxon rank sum test.

Moreover, in these cases, multivariate analysis using the Cox proportional-hazards model was performed to detect the clinicopathological factors including the factors based on the European Organization for Research and Treatment of Cancer risk tables32 that contribute independently to improving prognosis.

Routine Follow-up

Periodic tests were performed as postoperative follow-up using the conventional white light examination. Basically, cystoscopy was performed every 3 months for 1 year after the operation, every 6 months thereafter until 3 years, and then every year in all patients.

RESULTS

Pathological Evaluation

The diagnostic accuracy and also ability in blue light (fluorescence) mode was higher than in white light (conventional) mode in all 1372 specimens in 210 cases of PDD, including 534 specimens from 75 cases with intravesically applied ALA and also 838 specimens from 135 cases of PDD with orally applied ALA. Among the 1372 specimens from 210 cases obtained by transurethral biopsy, 485 specimens (35.3%) were pathologically diagnosed as malignant epithelium, including 106 specimens (7.7%) of CIS and 77 specimens (5.6%) of severe dysplasia detected pathologically (Tables 4 and 5). In semiquantitative analysis in conventional mode, macroscopic impression of malignancy was shown to be statistically significantly correlated with tumor grade, regardless of the method of ALA administration (P < .001) (Table 4). In semiquantitative analysis in fluorescence mode, fluorescence intensity was shown to be statistically significantly correlated with the tumor grade, regardless of the route of ALA administration (P < .001). Moreover, 132 samples (72.1%), including 44 dysplasia lesions and 88 CIS lesions, could be detected only in fluorescence mode in 183 flat lesions, 77 dysplasia lesions, and 106 CIS lesions. The percentage of flat lesions that could be detected only in fluorescence mode of PDD by orally applied ALA (74.3%) was higher than with PDD by intravesically applied ALA (68.6%; Table 5).

Table 4. Pathological Evaluation and Macroscopic Impression of Malignancy in the Examination in White Light Mode in All 1372 Biopsy Samples
White Light Mode (Conventional)Macroscopic Impression of MalignancySamples, Total No.P
NoneWeakStrong
  • Abbreviation: ALA, 5-aminolevulinic acid.

  • Correlation between pathological evaluation and macroscopic impression of malignancy in the examination in white light mode is shown. In semiquantitative analysis in conventional mode, macroscopic impression of malignancy was shown to be statistically significantly correlated with tumor grade, regardless of the method of ALA administration (P < .001).

  • a

    Fisher exact test (2 × 2).

All cases    <.001a
 Normal epithelium8352923887 
 Dysplasia6071077 
 UC G1521320 
 UC G2424112158 
 UC G360361124 
 UC G3-pTis9934106 
 Samples, total No.1101482231372 
Intravesically applied ALA    <.001a
 Normal epithelium3122414350 
 Dysplasia307340 
 UC G1521118 
 UC G2334450 
 UC G31922546 
 UC G3-pTis263130 
 Samples, total No.3954198534 
Orally applied ALA    <.001a
 Normal epithelium52359537 
 Dysplasia300737 
 UC G10022 
 UC G239168108 
 UC G34113678 
 UC G3-pTis730376 
 Samples, total No.7067125838 
Table 5. Pathological Evaluation and Fluorescence Intensity in the Examination in Blue Light Mode in All 1372 Biopsy Samples
Blue Light Mode (Fluorescence)Fluorescence IntensitySamples, Total No.P
NoneWeakStrong
  • Abbreviation: ALA, 5-aminolevulinic acid.

  • Correlation between pathological evaluation fluorescence intensity in the examination under blue light mode is shown. In semiquantitative analysis in fluorescence mode, fluorescence intensity was shown to be statistically significantly correlated with tumor grade, regardless of the method of ALA administration (P < .001). Moreover, 132 samples (72.1%), including 44 dysplasia lesions and 88 CIS lesions, could be detected only in fluorescence mode in 183 flat lesions, including 77 dysplasia lesions and 106 CIS lesions.

  • a

    Fisher exact test (2 × 2).

All cases    <.001a
 Normal epithelium55027958887 
 Dysplasia18451477 
 UC G1061420 
 UC G2117572158 
 UC G344575124 
 UC G3-pTis125242106 
 Samples, total No.5955022751372 
Intravesically applied ALA    <.001a
 Normal epithelium19513619350 
 Dysplasia825740 
 UC G1051318 
 UC G24232350 
 UC G32172746 
 UC G3-pTis0181230 
 Samples, total No.209224101534 
Orally applied ALA    <.001a
 Normal epithelium35514339537 
 Dysplasia1020737 
 UC G10112 
 UC G275249108 
 UC G32284878 
 UC G3-pTis12343076 
 Samples, total No.386278174838 

Diagnostic Accuracy and Capability

The diagnostic accuracy of ALA-PDD, including the positive rate, predictive accuracy, sensitivity, and specificity, was examined in all 1372 biopsy samples of all 210 cases. The sensitivity of PDD (93.4%) was significantly higher than the 44.7% sensitivity in white light mode, whereas the specificity of PDD (58.9%) was significantly lower than the 94.1% specificity in white light mode (P < .05). Regardless of the method of ALA administration, the sensitivity of ALA-PDD is significantly higher than that of conventional white light examination, whereas the specificity of ALA-PDD is low, which means that there are many false-positive findings in ALA-PDD. Both endoscopic examinations are equivalent in predictive accuracy (Table 6). The AUC in blue light (fluorescence) mode was greater than that in white light (conventional) mode in not only all PDD cases (P < .01) but also in PDD with intravesically applied ALA (P < .01) and PDD with orally applied ALA (P < .01) (Fig. 1).

Table 6. Diagnostic Accuracy of ALA-PDD
 Positive RateDiagnostic Accuracy, %
Predictive AccuracySensitivitySpecificity
  • Abbreviations: ALA, 5-aminolevulinic acid; ALA-PDD, photodynamic diagnosis mediated by ALA.

  • The diagnostic accuracy of ALA-PDD including the positive rate, predictive accuracy, sensitivity, and specificity regarding diagnostic accuracy was examined in all 1372 biopsy samples of all 210 cases. Regardless of the method of ALA administration, the sensitivity of ALA-PDD was significantly greater than that of conventional white light examination, whereas the specificity of ALA-PDD was low, which means that there were many false-positive findings in ALA-PDD. Both endoscopic examinations were equivalent in predictive accuracy.

  • a

    Two-sample test for equality of proportion.

All cases (1372 samples/210 cases)    
 Blue light mode (fluorescence)51.149.093.458.9
 White light mode (conventional)19.780.844.794.1
 P<.05a<.05a<.05a<.05a
Intravesically applied ALA (534 samples/75 cases)
 Blue light mode (fluorescence)60.353.892.053.8
 White light mode (conventional)26.072.754.989.1
 P<.05a<.05a<.05a<.05a
Orally applied ALA (838 samples/135 cases)
 Blue light mode (fluorescence)78.559.789.766.1
 White light mode (conventional)15.889.838.697.4
 P<.05a<.05a<.05a<.05a
Figure 1.

Diagnostic capability of photodynamic diagnosis (PDD) mediated by 5-aminolevulinic acid (ALA) (ALA-PDD) is defined by the area under the receiver operating characteristic curve (AUC). The AUC in blue light (fluorescence) mode (BLM) was more than that in white light (conventional) mode (WLM) in not only all PDD cases (P < .01) but also in PDD with intravesically applied ALA (P < .01) and PDD with orally applied ALA (P < .01).

Recurrence-Free Survival

The median follow-up period was 22.0 (range, 0.2-68.7) months in 99 patients who underwent PDD-TURBT. Thirty-three of 99 patients recurred, and the recurrence-free survival rate was 86.9% (at 12 months), 74.7% (24 months), 69.7% (36 months), 67.7% (48 months), and 66.7% (60 months). The median follow-up period was 21.5 (range, 0.2-204.1) months in 99 patients who underwent conventional TURBT. Sixty of 99 patients recurred, and the recurrence-free survival rate was 58.6% (at 12 months), 49.5% (24 months), 41.4% (36 months), 41.4% (48 months), and 40.4% (60 months). There was a statistically significant difference in the recurrence-free survival rate between these 2 therapeutic groups (P < .001) (Fig. 2).

Figure 2.

Recurrence-free survival in all cases is shown. Median follow-up period was 22.0 (range, 0.2-68.7) months in 99 patients who underwent transurethral resection of bladder tumor (TURBT) guided by photodynamic diagnosis mediated by 5-aminolevulinic acid. Thirty-three of 99 patients recurred, and recurrence-free survival rate was 86.9% (at 12 months), 74.7% (24 months), 69.7% (36 months), 67.7% (48 months), and 66.7% (60 months). Median follow-up period was 21.5 (range, 0.2-204.1) months in 99 patients who underwent conventional TURBT. Sixty of 99 patients recurred, and the recurrence-free survival rate was 58.6% (at 12 months), 49.5% (24 months), 41.4% (36 months), 41.4% (48 months), and 40.4% (60 months). There was a statistically significant difference in the recurrence-free survival rate between these 2 therapeutic groups (P < .001).

The median follow-up period was 32.4 (range, 0.2-68.7) months in 32 patients who underwent PDD-TURBT with intravesically applied ALA. Sixteen of 32 patients recurred, and the recurrence-free survival rate was 84.4% (at 12 months), 68.8% (24 months), 59.4% (36 months), 53.1% (48 months), and 50.0% (60 months). The median follow-up period was 17.1 (range, 1.9-40.8) months in 67 patients who underwent PDD-TURBT with orally applied ALA. Nineteen of 67 patients recurred, and the recurrence-free survival rate was 86.6% (12 months), 76.1% (24 months), and 71.6% (36 months). There was no statistically significant difference in the recurrence-free survival rate between these 2 therapeutic groups (P = .980) (Fig. 3).

Figure 3.

Recurrence-free survival stratified by 5-aminolevulinic acid (ALA) administration route is shown. Median follow-up period was 32.4 (range, 0.2-68.7) months in 32 patients who underwent transurethral resection of bladder tumor guided by photodynamic diagnosis mediated by ALA (PDD-TURBT) with intravesically applied ALA. Sixteen of 32 patients recurred, and recurrence-free survival rate was 84.4% (at 12 months), 68.8% (24 months), 59.4% (36 months), 53.1% (48 months), and 50.0% (60 months). Median follow-up period was 16.8 (range, 1.2-41.8) months in 70 patients who underwent PDD-TURBT with orally applied ALA. Nineteen of 67 patients recurred, and the recurrence-free survival rate was 86.6% (at 12 months), 76.1% (24 months), and 71.6% (36 months). There was no statistically significant difference in the recurrence-free survival rate between these 2 therapeutic groups (P = .980).

The median follow-up period was 18.1 (range, 3.7-38.8) months in 18 T1G3 patients who underwent PDD-TURBT. Nine of 18 patients recurred, and the recurrence-free survival rate was 72.2% (at 12 months), 55.6% (24 months), and 50.0% (36 months). The median follow-up period was 27.6 (range, 0.2-185.1) months in 18 T1G3 patients who underwent conventional TURBT. Thirteen of 18 patients recurred, and the recurrence-free survival rate was 47.3% (at 12 months), 33.3% (24 months), and 27.8% (36 months). There was no statistically significant difference in the recurrence-free survival rate between PDD-TURBT for T1G3 and conventional TURBT for T1G3 (P = .062) (Fig. 4).

Figure 4.

Recurrence-free survival is stratified by fluorescence transurethral resection of bladder tumor (TURBT) and conventional TURBT in T1G3. Median follow-up period was 18.1 (range, 3.7-38.8) months in 18 T1G3 patients who underwent TURBT guided by photodynamic diagnosis mediated by 5-aminolevulinic acid (PDD-TURBT). Nine of 18 patients recurred, and the recurrence-free survival rate was 72.2% (at 12 months), 55.6% (24 months), and 50.0% (36 months). Median follow-up period was 27.6 (range, 0.2-185.1) months in 18 T1G3 patients who underwent conventional TURBT. Thirteen of 18 patients recurred, and the recurrence-free survival rate was 47.3% (at 12 months), 33.3% (24 months), and 27.8% (36 months). There was no statistically significant difference in the recurrence-free survival rate between PDD-TURBT for T1G3 and conventional TURBT for T1G3 (P = .062). A deferred cystectomy because of recurrence and progression was performed in 1 case in the PDD-TURBT group and 2 cases in the conventional TURBT group. Median time to cystectomy was 7.9 months in the PDD-TURBT group, and 3.9 months and 10.4 months in the conventional TURBT group.

A deferred cystectomy because of recurrence and progression was performed for 1 patient in the PDD-TURBT group and 2 patients in the conventional TURBT group. Median time to cystectomy was 7.9 months in the PDD-TURBT group, and 3.9 months and 10.4 months in the conventional TURBT group (Fig. 4).

Moreover, multivariate analysis revealed that PDD-TURBT was the only independent factor that contributed to improving the intravesical recurrence rate (hazard ratio, 0.578; 95% confidence interval, 0.371-0.888; P = .012) (Table 7).

Table 7. Multivariate Analysis of All Cases
FactorsHazard Ratio95% Confidence IntervalP
  1. Abbreviations: BCG, Bacillus Calmette-Guerin; EORTC, European Organization for Research and Treatment of Cancer; PDD-TURBT, transurethral resection of bladder tumor guided by photodynamic diagnosis.

  2. Multivariate analysis revealed that the only independent factor contributing to improving prognosis was PDD-TURBT (hazard ratio, 0.578; 95% confidence interval, 0.371-0.888; P = .012).

EORTC recurrence score32 (0, 1-4, 5-9, 10-17)1.2400.482-3.234.657
BCG intravesical instillation0.8320.516-1.400.475
PDD-TURBT0.5780.371-0.888.012

Adverse Events

Although no special precaution, such as liver support and light shielding, was implemented throughout PDD, there were mild and transient adverse events in conformity with the Common Terminology Criteria for Adverse Events version 3.0.27 Urinary frequency and/or urgency occurred in 13 (17.3%) of 75 cases of PDD with intravesically applied ALA, photosensitivity reaction in 6 cases (4.4%), elevated level of AST and/or ALT in 4 cases (3.0%), and nausea and/or vomiting in 4 of 135 cases (3.0%) of PDD with orally applied ALA.

DISCUSSION

ALA-PDD is a cancer diagnostic method by fluorescence navigation, which has been clinically recognized as an effective procedure to detect various cancers, such as brain tumor33 and bladder cancer.10-17 In particular, PDD with hexaminolevulinate, the hexyl ester derivative of 5-ALA for bladder cancer, has already been approved and implemented as a legitimate medical practice in Europe and the United States, whereas in Japan, PDD with intravesically and orally applied ALA was performed for the first time in 2004,29 and has just been approved and implemented as advanced medical technology in registered institutions by the Ministry of Health, Labor, and Welfare in 2010. In this study, we demonstrated the clinical effectiveness and safety of ALA-PDD and also of TURBT guided by ALA-PDD with a large population in Japan.

At transurethral biopsy, intravesical observation using ALA-PDD was useful in detecting CIS, avoiding an incorrect decision on adjuvant therapy. The diagnostic accuracy of ALA-PDD was first reported with a sensitivity of 100.0% and specificity of 68.5% in 68 cases of bladder cancer by Kriegmair et al10 in 1994. Subsequently, many clinical trials of ALA-PDD have been performed, mainly in Europe. Hungerhuber et al17 reported 1713 procedures of ALA-PDD with a sensitivity of 92.0% and specificity of 55.6% in 875 cases of bladder cancer, the largest number of ALA-PDD cases. Including these reports, a summary of previous reports revealed that the sensitivity of PDD (94.6%; range, 77.8%-100%) was significantly higher than the 76.0% (range, 67.5%-84.0%) sensitivity of conventional white light examination, whereas the specificity of PDD (59.0%; range, 33.0%-87.1%) was significantly lower than the 68.5% (range, 66.4%-78.0%) specificity of conventional white light examination.10-17 In particular, the reports of flat lesions revealed that 34.2% (range, 14.9%-42.1%) of all lesions, 43.4%-57.0% of CIS, and 30.3%-44.0% of dysplasia could be detected by ALA-PDD, but not conventional white light examination.15-17, 34, 35 The results of this study support the results of these clinical studies of ALA-PDD. The AUC in PDD with a sensitivity of 93.4% and specificity of 58.9% was significantly greater than that in conventional white light examination. Only PDD could detect 72.1% of flat lesions including dysplasia and CIS. Moreover, it was demonstrated that regardless of the ALA administration route, there was no significant difference in the diagnostic accuracy and ability of PDD in this study.

Causal lesions of intravesical recurrence are endoscopic invisible lesions. In TURBT, additional resection under ALA-PDD could avoid insufficient resection of tumors, reducing the rate of intravesical recurrence. The rate of residual tumor in PDD-TURBT with a median rate of 8.0% (0%-32.7%) was significantly reduced compared with conventional TURBT under white light, with a median rate of 37.0% (19.2%-53.1%) in the summary of previous reports.14, 18, 21, 36, 37 Denzinger et al22 made a comparative review of intravesical recurrence-free survival between 88 cases of PDD-TURBT and 103 cases of conventional TURBT, with 96 months as the follow-up duration. They reported that the intravesical recurrence-free survival of 90.9% (at 12 months after PDD-TURBT), 90.9% (24 months), 85.0% (48 months), 79.0% (72 months), and 71.0% (96 months) in PDD-TURBT was significantly greater than that in conventional TURBT of 78.6% (at 12 months after conventional TURBT), 69.9% (24 months), 60.7% (48 months), 54.0% (72 months), and 45.0% (96 months) (P = .0003). Moreover, it was demonstrated that PDD-TURBT was an independent prognostic factor for improving the intravesical recurrence rate, with a hazard ratio of 0.29 (95% confidence interval, 0.15-0.56; P = .0002) by multivariate analysis using the Cox proportional hazards model. The results of this study support the results of previous clinical studies of PDD-TURBT.18-22 In this study, it was revealed that the intravesical recurrence-free survival was significantly greater than that in conventional TURBT at up to 60 months follow-up (P < .001). It was also revealed that PDD-TURBT was the only independent factor that contributed to improving the intravesical recurrence rate (hazard ratio, 0.578; 95% confidence interval, 0.371-0.888; P = .012) by multivariate analysis. Moreover, it was demonstrated that regardless of the ALA administration route, there was no significant difference in intravesical recurrence-free survival in this study.

In this study, only 18 cases of T1G3 were known to show highly aggressive behavior. PDD-TURBT for T1G3 had a greater tendency to reduce the intravesical recurrence rate, but with no statistically significant difference (P = .062) compared with conventional TURBT for T1G3 in this study population. However, more recently, the effectiveness of tumor control has been demonstrated in 77 cases of PDD-TURBT for T1G338; therefore, more data compilation may show the usefulness of PDD-TURBT for T1G3 in improving recurrence-free survival and also avoiding deferred cystectomy because of tumor recurrence and progression.

Currently ALA and hexaminolevulinate have been approved as photosensitizers of PDD by authorities around the world. In Europe, ALA was approved under the trade name Gliolan as an optical imaging agent to enhance intraoperative detection of malignant glioma (World Health Organization grade III and IV) by the European Medicines Evaluation Agency.9 Conversely, hexaminolevulinate was approved under the trade names Hexvix and Cysview as an optical imaging agent to enhance intraoperative detection of papillary bladder cancer by the European Medicines Evaluation Agency and the Food and Drug Administration in Europe and the United States, respectively.9 More accumulation of biosynthesized protoporphyrin IX was revealed by administration of hexaminolevulinate compared with ALA by in vivo spectrophotometric measurement39; therefore, it was reported that hexaminolevulinate yielded higher fluorescence intensity and contrast between normal and malignant urothelium, with less photobleaching. In 2002, Dalton et al investigated the pharmacokinetics of ALA administered intravenously, orally, and intravesically, in which <1% of ALA was absorbed through the urinary bladder, and the cumulative ALA amount in urine was about 20,000× higher than that in plasma.40 This finding indicates the low possibility of the occurrence of systemic adverse events induced by the intravesical administration of ALA, but does not demonstrate the usefulness for diagnosis compared with oral ALA administration, that is, differences in the fluorescence intensity between lesions and nonlesions (ie, contrast) and the fluorescence attenuation level with excitatory light irradiation (photobleaching) are parameters of diagnostic usefulness, and the diagnostic accuracy is based on these. We performed a clinical comparison of ALA administered intravesically and orally, and showed that the diagnostic accuracy of PDD was superior to that of white light examination excluding the predictive accuracy using intravesical ALA administration. In addition, there were no significant differences because of the variation in the route of administration (intravesical and oral ALA administrations) in diagnostic accuracy, diagnostic performance, or recurrence-free survival. To pharmacologically demonstrate this subjective evaluation of clinical data and fluorescence intensity, we are planning to compare intravesical and oral ALA administrations with regard to tissue ALA and protoporphyrin IX levels in lesions and nonlesions, and the time course changes of these. Moreover, it was demonstrated in the retrospective series that currently, although PDD with ALA and hexaminolevulinate applied intravesically were demonstrated to be significantly superior to white light cystoscopy, there were no significant differences between ALA and hexaminolevulinate in clinical outcome such as residual tumor and recurrence-free survival.26 Thus, in this study, we used ALA orally and intravesically to evaluate the clinical value in PDD and TURBT guided by ALA-PDD for bladder cancer. As described above, in this study, it was demonstrated that regardless of the ALA administration route, there was no significant difference in diagnostic accuracy, effectiveness of PDD, and recurrence-free survival, and all procedures were well tolerated by all patients without any severe adverse events. Although it is subjective, the intensity of ALA-induced fluorescence in PDD with orally applied ALA was higher than that in PDD with intravesically applied ALA. Moreover, PDD with orally applied ALA had less photobleaching. Hence, in the urological field, oral administration of ALA may be useful in PDD for malignancies other than bladder cancer. Further technical development of ALA-PDD is required to avoid photobleaching, improve diagnostic accuracy, and establish a standard intraoperative diagnostic system in the near future in Japan.

Conclusions

PDD with intravesically and also orally applied ALA is equally effective in detecting endoscopic invisible lesions such as dysplasia and carcinoma in situ, avoiding an incorrect decision on adjuvant therapy. In TURBT also, additional resection under ALA-PDD could avoid insufficient resection of tumors, reducing the rate of intravesical recurrence. It was suggested that regardless of the ALA administration route, ALA-PDD might be remarkably helpful in detection and intraoperative navigation programs.

FUNDING SOURCES

No specific funding was disclosed.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

Ancillary