Abstract 283P
Background
Non-muscle invasive bladder cancer (NMIBC) necessitates vigilant monitoring post-treatment due to its high recurrence and progression risk. Urine cystoscopy and cytology, the gold standard for diagnosis and surveillance, are invasive and exhibit low sensitivity with high inter-operator variability. Multiple urine DNA methylation-based tests, including Bladder EpiCheck (BE), have been developed for NMIBC detection. This review assesses studies analysing such tests' diagnostic accuracy in NMIBC surveillance.
Methods
Systematic searches were conducted in PubMed, EMBASE, Scopus, and ScienceDirect until March, 2024, using terms related to urine DNA methylation-based screening tests in NMIBC. RCTs and observational studies in English, focusing on adults with NMIBC history undergoing surveillance, were included. The diagnostic accuracy of such tests compared to standard methods was assessed. QUADAS-2 tool was used to assess the quality of studies. Due to incomplete data reporting, findings were qualitatively synthesised.
Results
30 studies with 7249 NMIBC patients were included. Recurrence was assessed over 9 months to >10 years. BE score of ≥60 was considered positive. For BE; sensitivity ranged from 62.3% (Trenti et. al.) to 100% (Cochetti et. al.) for all cancer grades; while Witjes et. al. showed lower sensitivity (33.3%) for low-grade cancers; specificity ranged from 79.6% (Pierconti et. al.) to 91.89% (Pena et. al.); PPV ranged from 44.8% (Witjes et. al.) to 78.57% (Pena et. al.); while NPV ranged from 78.6% (Trenti et. al.) to 97.14% (Pena et. al.); AUC values ranged from 0.74 (Trenti et. al.) to 0.95 (Cochetti et. al.). OncoUrine showed 100% sensitivity (Huang et. al.), while qMSP yielded 100% specificity (Friedrich et. al.); qMSP also showed a PPV of 99-100% (Reinert et. al.), while OncoUrine yielded 100% NPV (Huang et. al.); surpassing BE.
Conclusions
Bladder EpiCheck, though outperformed at times, remains preferred for NMIBC surveillance due to its broader study base and larger sample sizes, enhancing reliability. A DTA meta-analysis is needed to establish its validity, for which we urge the authors to provide complete data on the confusion matrix.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.