Patients choose certainty over burden in bladder cancer surveillance

Abstract

Background: Due to the high risk of recurrence of non-muscle invasive bladder cancer, all patients undergo regular cystoscopic surveillance for early detection. As cystoscopy is invasive, costly and increases the burden of the disease considerably, there is significant ongoing research and development into non-invasive urinary biomarker substitutes. This study aims to assess the level of sensitivity required before patients accept a new urinary biomarker.

Methods: We studied the preferences for a hypothetical diagnostic urinary biomarker and compared this to usual care (cystoscopy) at different levels of sensitivity among 437 patients with bladder cancer (354 men and 83 women) from the UK Bladder Cancer Prognosis Programme. A standard gamble approach was used to estimate the minimally acceptable sensitivity (MAS) of the new biomarker. Additionally, non-parametric statistical analyses were performed to investigate the association between surveillance preference and various patient characteristics.

Results: Almost half of patients (183, 43%) would not replace cystoscopy with a urinary biomarker unless it was 100% sensitive. The median MAS was 99.9999%, and nearly 85% of patients demanded a sensitivity of at least 99% before preferring a urinary biomarker test over cystoscopy. These results were consistent across all patient characteristics and demographic categories.

Conclusions: Our results indicate that patients demand urinary biomarkers as sensitive as cystoscopy before they would be willing to forego cystoscopy for bladder cancer surveillance.

Keywords: Non-invasive biomarkers; Non-muscle-invasive bladder cancer; Sensitivity and specificity; Standard gamble.

Fig. 1

Flow chart depicting patient recruitment in the Bladder Cancer Prognosis Programme (BCPP) cohort study and inclusion for the current study