CRD summary

This review concluded that photodynamic diagnosis detected more bladder tumour-positive patients and resulted in a greater proportion of patients with complete resection and long recurrence-free survival than white-light cystoscopy. These conclusions were supported by the results, but a lack of details on the included studies and their risk of bias mean that the conclusions should be interpreted with some caution.

Authors' objectives

To assess the effect of photodynamic diagnosis with 5-aminolevulinic acid or hexaminolevulinate in addition to white-light cystoscopy on the diagnosis and therapeutic outcome of primary or recurrent non-muscle-invasive bladder cancer investigated by cystoscopy or transurethral resection.

Searching

MEDLINE, EMBASE, The Cochrane Library, CancerLit and ClinicalTrials.gov were searched in January 2009. Four relevant conference proceedings and two relevant journals were handsearched from 1990 to 2008. The review was restricted to studies in English, German or French. Included studies had to be available as full text reports; abstracts were excluded. Where necessary, authors were contacted for further information.

Study selection

Prospective diagnostic cohort studies that applied photodynamic diagnosis following white-light cystoscopy in the same patients and that used histology of biopsies taken at detected locations as the reference standard were eligible for inclusion. The primary outcome for diagnostic accuracy studies was the additional detection rate (the proportion of patients additionally and correctly identified by photodynamic diagnosis to have bladder cancer). The additional false-positive rate was considered. To assess therapeutic outcome, randomised controlled trials (RCTs) that compared white-light cystoscopy alone to white-light cystoscopy combined with photodynamic diagnosis in patients who had undergone complete transurethral resection for primary or recurrent tumours were eligible for inclusion. The primary endpoints were residual tumour at second resection and recurrence-free survival.

Diagnostic accuracy studies included patients with known or suspected non-muscle-invasive bladder tumours. The studies of therapeutic impact all assessed photodynamic diagnosis or white-light cystoscopy plus photodynamic diagnosis at initial resection. Residual tumour rates were assessed at second resection 10 days to six weeks after the first transurethral resection.

Two reviewers independently assessed studies for inclusion.

Assessment of study quality

Two reviewers independently assessed the reporting quality of RCTs using criteria adapted from the CONSORT guidelines and the quality of diagnostic accuracy studies using criteria adapted from the STARD guidelines. Full details of the items assessed were reported in an appendix.

Data extraction

For diagnostic studies, data were extracted to calculate the additional detection rates and additional false positive rates associated with photodynamic diagnosis and white-light cystoscopy. For therapeutic studies, dichotomous data were extracted to calculate odds ratios and continuous data were extracted to calculate the median time to recurrence. 95% confidence intervals (CIs) were calculated for each outcome measure. Two reviewers independently performed data extraction.

Methods of synthesis

Additional detection rates were pooled using the Freeman-Tukey double arcsine transformation and inverse variance method. Odds ratios were pooled using the Mantel-Haenszel fixed-effect model. Residual free tumour survival data were pooled by combining p-values from the log-rank tests. Heterogeneity was assessed using I² and the Q-test. Subgroup analyses were performed for patients with carcinoma in situ.

Results of the review

Twenty-one publications that reported the results of 17 studies were included in the review. Twelve studies (n=1,449) assessed diagnostic accuracy and five RCTs assessed therapeutic outcome. Case number, definition of primary criterion and name of statistical procedure were reported most frequently. Rater agreement, sample selection and concealment of allocation were never reported.

Additional detection rates (12 studies, n=1,449): Additional detection rates ranged from 5% to 49% with a pooled estimate of 20% (95% CI 8% to 35%) based on seven studies. Additional detection rates in patients with carcinoma in situ ranged from 17% to 78% with a pooled estimate of 39% (95% CI 23% to 57%) based on seven studies. There was substantial heterogeneity for both analyses (I²=95% and 84%) . Exclusion of two studies with known selection bias from the analysis of patients with carcinoma in situ reduced the pooled estimate to 23% (95% CI 18% to 30%) with no evidence of heterogeneity (I²=0%).

Residual tumour (three RCTs, n=394 patients with tumours): Photodynamic diagnosis plus white-light cystoscopy was associated with a significant reduction in residual tumours compared to white-light cystoscopy alone (OR 0.28, 95% CI 0.15 to 0.52). There was evidence of heterogeneity (I²=23%).

Recurrence free survival (five RCTs): Photodynamic diagnosis was associated with fewer recurrences at 12 months (three RCTs, n=415) and 24 months (two RCTs, n unclear). The summary p-value was 0.00002 for recurrence at 12 months. Two trials reported recurrence free survival at 24 months. Two trials reported median time to recurrence and this ranged from five to eight months in the white-light cystoscopy groups and from 12 to 17 months in the photodynamic diagnosis groups.

Authors' conclusions

Photodynamic diagnosis detected more bladder tumour-positive patients (especially among patients with carcinoma in situ) than white-light cystoscopy. More patients have a complete resection and a longer recurrence-free survival when diagnosed with photodynamic diagnosis.

CRD commentary

The review addressed a focused objective and inclusion criteria were clearly defined. The literature search was extensive and included some attempts to locate unpublished studies. However, restriction of the review based on language of publication meant that some of the studies identified as potentially relevant were excluded. Appropriate steps were taken to minimise bias at all stages of the review process. Although a detailed checklist was used to assess the reporting quality of the included studies, that risk of bias was not considered and so the reliability of the included studies was unclear. Very few details of the included studies were reported, which made it difficult to determine the generalisability of results. Methods used to pool studies appeared appropriate. Heterogeneity was assessed. A lack of details on the additional false positive rates associated with photodynamic diagnosis made it difficult to draw conclusions regarding its overall accuracy.

The authors' conclusions were supported by the results of the review, but a lack of details on the included studies and their risk of bias mean that the conclusions should be interpreted with some caution.

Implications of the review for practice and research

Practice: The authors did not state any implications for practice.

Research: The authors stated that future studies needed to show wehther intervals to follow-up cystoscopies could be altered when photodynamic diagnosis was used and whether photodynamic diagnosis changed adjuvant treatment (for instance, due to the safe exclusion of carcinoma in situ). The necessity of re-transurethral resection after photodynamic diagnosis needed evaluation.

Funding

None stated.

Bibliographic details

Kausch I, Sommerauer M, Montorsi F, Stenzl A, Jacqmin D, Jichlinski P, Jocham D, Ziegler A, Vonthein R. Photodynamic diagnosis in non-muscle-invasive bladder cancer: a systematic review and cumulative analysis of prospective studies. European Urology 2010; 57(4): 595-606. [PubMed: 20004052]

Original Paper URL

http://dx.doi.org/10.1016/j.eururo.2009.11.041

Indexing Status

Subject indexing assigned by NLM

MeSH

Aminolevulinic Acid /analogs & derivatives /diagnostic use; Cystectomy; Cystoscopy; Disease-Free Survival; Evidence-Based Medicine; Humans; Neoplasm Invasiveness; Neoplasm Recurrence, Local; Neoplasm, Residual; Predictive Value of Tests; Prospective Studies; Randomized Controlled Trials as Topic; Time Factors; Treatment Outcome; Urinary Bladder Neoplasms /diagnosis /mortality /pathology /surgery

AccessionNumber

12010002040

Database entry date

05/01/2011

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.