Diagnostic Test Accuracy of Urinary DNA Methylation-based Biomarkers for the Detection of Primary and Recurrent Bladder Cancer: A Systematic Review and Meta-analysis

Mariana Silva-Ferreira,João A Carvalho,Sofia Salta,Teresa S Henriques,Pedro Pereira Rodrigues,Sara Monteiro-Reis,Rui Henrique,Carmen Jerónimo
DOI: https://doi.org/10.1016/j.euf.2024.05.024
2024-06-19
Abstract:Background and objective: Diagnosis of primary and relapsed bladder carcinomas is accomplished by urethrocystoscopy, an invasive procedure, combined with urinary cytology, with limited sensitivity, resulting in a substantial burden. Thus, noninvasive biomarkers have been investigated, among which DNA methylation has shown promise. This systematic review and meta-analysis sought to assess the diagnostic accuracy of DNA methylation biomarkers reported in the literature for bladder cancer detection, pinpointing the most informative one. Methods: The search for this systematic review and meta-analysis was conducted on PubMed, Scopus, and Cochrane Library for relevant studies published until December 31, 2022. A meta-analysis was performed using a random-effect model, to compute the pooled sensitivity and specificity of the markers. PROSPERO's registration ID for the study is CRD42023397703. Key findings and limitations: Out of the 2297 studies retrieved, 68 were included in the final analysis, despite considerable heterogeneity. These involved 12 696 participants, of whom 5557 were diagnosed with bladder cancer. Using diagnostic odds ratio (DOR) as a comparative measure, the five most promising markers (pooled sensitivity, specificity, and DOR) were SALL3 (61%, 97%, and 55.67, respectively), PENK (77%, 93%, and 47.90, respectively), ZNF154 (87%, 90%, and 45.07, respectively), VIM (82%, 90%, and 44.81, respectively), and POU4F2 (81%, 89%, and 34.89, respectively). Urinary cytology identified bladder cancer with 55% sensitivity, 92% specificity, and 14.37 DOR. Conclusions and clinical implications: DNA methylation biomarkers disclose high accuracy for bladder cancer detection in urine. Nonetheless, validation studies in different clinical settings are scarce, hampering clinical use. The identified biomarkers should be prioritized in future validation studies. Patient summary: In this meta-analysis, we include previously published studies that used urine samples of bladder cancer patients' from all around the globe. We were able to compare the diagnostic accuracy of noninvasive markers across different populations. We were able to conclude on the most promising DNA methylation markers to detect bladder cancer using urine.
What problem does this paper attempt to address?