Cxbladder Triage vs Triple Workup in Haematuria: Key Findings

A prospective, single-center Australian cohort study of patients presenting with haematuria reports a head-to-head comparison of the urine genomic biomarker Cxbladder Triage versus the conventional “triple workup” of flexible cystoscopy, voided urine cytology, and upper tract imaging. The cohort included 258 consecutively recruited adults evaluated between 2020 and 2023, with histologically confirmed urothelial carcinoma (UC) identified in 14 patients (5.4%). Results are presented as a comparison of rule-out performance between Cxbladder Triage and cytology-based strategies (cytology alone and cytology plus upper-tract imaging), with the standard haematuria workup (including cystoscopy and histology where indicated) used as the reference standard. This report focuses on the study’s stated rule-out performance and the testing completeness observed in this real-world dataset.

For UC detection in the overall cohort, the study reports sensitivity and negative predictive value (NPV) estimates that were higher for the biomarker than for cytology-based strategies. Cxbladder Triage sensitivity was reported as 92.9% (95% CI, 66.0–99.8) with an NPV of 92.9% (95% CI, 66.0–99.8); these estimates were based on 14 UC cases, contributing to wide confidence intervals. By comparison, voided urine cytology alone had a reported sensitivity of 42.9% (95% CI, 9.9–81.6) and NPV of 78.9% (95% CI, 54.4–94.0). When cytology and upper tract imaging were combined (with positivity defined by either a positive cytology result or imaging suspicious for malignancy), the reported sensitivity was 75.0% (95% CI, 42.8–94.5) and the NPV was 80.0% (95% CI, 51.9–95.8). In this dataset, point estimates favored Cxbladder on sensitivity and NPV, alongside wide confidence intervals.

The authors also describe missed-case findings to contextualize the diagnostic summary. Cxbladder Triage missed one histologically confirmed UC case, and they note that this same case was also not correctly identified by cytology or imaging. Among the 14 UC diagnoses, tumor characteristics were reported as 1 carcinoma in situ, 4 low-grade tumors, and 9 high-grade tumors. The paper presents these distributions alongside the missed-case statement as part of its broader discussion of what a negative or non-contributory test result looked like in this cohort.

Operational details in the results section describe test completion and reporting. Diagnostic accuracy estimates were calculated in patients with complete Cxbladder Triage, cytology, and upper-tract imaging results, while missing results are described for the broader recruited cohort. The study reports that 58.1% of patients had a negative Cxbladder Triage result, while 3.5% recorded a Cxbladder failure result in which risk stratification could not be generated. Missing data were also reported for conventional tests: 15.5% of patients did not produce a cytology result, and 9.7% failed to complete upper tract imaging. The authors note that assay failures and incomplete conventional testing can affect how diagnostic estimates are interpreted and may influence generalizability.

In discussion, the authors describe the potential for biomarker-guided haematuria pathways to reduce invasive investigations and related resource use, and they mention prior decision analyses and modelling in that context without presenting new primary data from those publications. They also emphasize limitations they believe constrain inference from their findings, including the single-center design, the small number of UC events contributing to wide confidence intervals, and non-uniform cytology interpretation because samples were not read by a single cytopathologist. The paper closes by stating that external and ideally multicenter validation would be needed to further evaluate performance in other settings. Overall, the authors position their results as preliminary and contingent on validation in larger, more diverse cohorts.

Key Takeaways:

• The study reports higher sensitivity and NPV point estimates for Cxbladder Triage than for cytology alone or cytology combined with imaging in this haematuria cohort.
• One UC case was reported as missed by Cxbladder Triage, and the authors state this same case was not correctly identified by cytology or imaging.
• Assay failures and incomplete cytology/imaging results were reported, and the authors noted that larger, multicenter studies with more urothelial carcinoma cases may provide deeper insight given the study’s design and event-count limitations.