Advice
Clinical and technical evidence
Clinical and technical evidence
A literature search was carried out for this briefing in accordance with the interim process and methods statement. This briefing includes the most relevant or best available published evidence relating to the clinical effectiveness of the technology. Further information about how the evidence for this briefing was selected is available on request by contacting mibs@nice.org.uk.
Published evidence
One multicentre, prospective study involving 577 people presenting with haematuria or lower urinary tract symptoms with a suspicion of malignancy is summarised in this briefing. Results from this study were published in the form of a conference abstract only.
Two other studies evaluating the use of minichromosome maintenance complex component 5 (MCM5) as a biomarker in the detection of bladder cancer were identified (Stoeber et al. 2002; Kelly et al. 2012). Overall, they showed that immunofluorometric detection of MCM5 in urine sediment is a sensitive and specific diagnostic test for bladder cancer (with a sensitivity of 92% and specificity of 78%), with similar diagnostic accuracy to the nuclear matrix protein number 22 (NMP22) enzyme-linked immunosorbent assay (ELISA) test. These studies are not discussed further in this briefing because they are preliminary and do not fully reflect the technology in its current ELISA-based form.
ADXBLADDER can also be used for diagnosing bladder cancer recurrence within the urogenital tract during follow-up after transurethral resection of a bladder tumour but there are no published data to support this use.
Table 1 summarises the clinical evidence as well as its strengths and limitations.
Overall assessment of the evidence
Evidence from a multicentre prospective study showed that overall ADXBLADDER was able to detect bladder cancer with a sensitivity and specificity of 76% (95% confidence interval [CI]: 61% to 87%) and 69% (95% CI: 65% to 73%) respectively (using a diagnosis based on cystoscopy, imaging and biopsy as a reference standard). This contributed to a negative predictive value of 97% (95% CI: 95% to 98%). Despite reporting a high negative predictive value, based on the reported sensitivity around 24% of people with cancer would be classed as not having cancer by the test. Sensitivity of the test for the combined high-risk and muscle-invasive groups was 95% (95% CI: 75% to100%), suggesting the test may be better at detecting these tumour types over low-grade tumours. Results from the study also showed that ADXBLADDER was able to detect more cases of bladder cancer compared with cytology. However, this data came from a very small subset of patients (n=10).
The study was done in a relatively large number of patients across 6 UK centres, so the results are likely to be generalisable to NHS practice. Overall, the evidence for the technology is limited to 1 study which assessed the diagnostic ability of the test only. The study was funded by the company and data are available in abstract form only so additional details about the results could not be verified. The accuracy data are therefore subject to uncertainty and further studies will be needed to verify these results, as well as understand the effect that adopting this test will have on longer-term outcomes including the overall cost consequence for the healthcare system. Studies comparing the diagnostic accuracy of ADXBLADDER with that of cytology and other urinary biomarker tests (such as UroVysion fluorescence in-situ hybridisation, ImmunoCyt or NMP22 test) would also be helpful.
Table 1 Summary of selected studies
|
Study size, design and location |
Multicentre, prospective study involving 577 people presenting with haematuria or lower urinary tract symptoms between September 2016 and February 2017, with a suspicion of malignancy. The study was done across 6 UK sites. |
|
Intervention and comparator(s) |
Intervention: ADXBLADDER. A diagnosis obtained by cystoscopy, imaging, and resection biopsy of suspect lesions was used as a reference standard. |
|
Key outcomes |
Out of the 577 enrolled patients, 46 were diagnosed with cancer (7.96% prevalence). ADXBLADDER was able to detect 35 out of 46 of all tumours and 19 out of 20 high-risk muscle-invasive tumours. ADXBLADDER sensitivity was 76% for all tumours and 95% for high-risk and muscle-invasive tumours. The specificity and NPV of ADXBLADDER for all tumours were 69% and 97.1%, respectively. The NPV reported for high-risk tumours was 99.7%. In a subgroup analysis with cytology comparison data (n=10), ADXBLADDER detected 8 out of 10 tumours while cytology detected 2 out 10 tumours. All tumours detected by cytology were detected by ADXBLADDER. |
|
Strengths and limitations |
This was multicentre study involving a relatively large number of people and was done across 6 UK centres. Therefore, results are likely to be generalisable to NHS practice. The study was funded by the company and results were only published in the form of a conference abstract, full study details including exclusion/inclusion criteria and patient demographics were not available. |
|
Abbreviations: NPV, negative predictive value. |
|
Recent and ongoing studies
No ongoing or in-development trials were identified on any of the clinical trial registries searched. The company states that ADXBLADDER is currently being evaluated in 2 clinical studies. These are a prospective performance evaluation to determine the utility of ADXBLADDER in helping detect recurrent bladder cancer, and a single-centre Italian study looking at the optimisation of urine sample collection, processing and storage. The company also states that 5 additional clinical studies are planned, with the aim of providing further data on specific patient groups and to clarify product performance and specificity.