• Bladder tumour;
  • superficial;
  • intravesical instillations;
  • prognosis

Objective To identify prognostic factors that enable patients with superficial bladder cancer to be divided into groups who will probably respond to therapy and those who will not.

Patients and methods In a prospective randomized study 38 7 patients with pTn-pT1 superficial bladder carcinoma received, after transurethral resection, intravesical instillations with immuno- or chemotherapy. A simultaneous computerized analysis of factors predicting the recurrence-free interval was performed, All these patients had negative random biopsies. Pretreatment factors analysed for recurrence were gender, age, history (primary or recurrent disease), location of tumour, number of tumours, pT-stage and grade.

Results Median follow-up was 2 7 months (range 12–56). During the follow-up period 37.2% of the patients had recurrence and eight patients (2.2%) had progression into muscle invasive disease. After a univariate analysis the number of tumours and location of the tumour in the bladder appeared to be a significant influence on the recurrence-free interval. Location of at least one of the tumours in the prostatic urethra, bladder neck, posterior wall, and trigone area was significantly related to a shorter recurrence-free interval: these areas were defined as high risk. Tumour stage showed borderline significance. By using multivariate methods to assess the relative importance of these factors, location of tumour in the high risk region was related especially to a short recurrence-free interval. The factor found to be of significance for the prognosis for recurrence was tumour multiplicity. Gender, age, history of recurrent disease, size of the largest tumour, tumour stage or grade gave no additional information about the risk of recurrence.

Conclusion Prognostic factor analysis, as an auxilliary study of trials of patients treated for superficial bladder tumours, is mandatory. The prognostic factors related to recurrence-free interval found in this study, location of the tumour and multiplicity, may be of use in the stratification necessary for current protocol design.