Login to Access Video or Poster Abstract: MP15-20
Sources of Funding: Cytosystems Ltd., Aberdeen, Scotland, UK

Introduction

Cystoscopy remains the gold standard in the investigation of haematuria and follow up of patients diagnosed with the urothelial carcinoma (UC) of the bladder. There has been extensive research into identifying urinary markers with the aim of improving diagnostic accuracy. Minichromosome maintenance 2 protein (MCM2) is a marker of cell proliferation and ectopic expression is a characteristic feature of malignancy and pre-malignancy._x000D_ Here we evaluate diagnostic accuracy of MCM2 in diagnosis and surveillance of UC._x000D_

Methods

A feasibility study was conducted on 176 patients and healthy volunteers from 3 centres in the UK from Cystoscopic Surveillance (CS) and Gross Haematuria (GH) clinics. The verification set comprised 149 volunteers. _x000D_ SurePath platform was used to process the urine cytology samples. Immunocytochemical analysis of MCM2 was performed as a marker for presence of UC. Feasibility data sets were used to determine MCM2 threshold for GH and CS counts using the optimised Youden’s Index (J) and optimal sensitivity, establishing conditions such that there was a zero false negative rate. Cut-off values were used to determine sensitivity, specificity, PPV and NPV._x000D_

Results

Using optimised J-approach, the feasibility sets for GH and CS samples yielded cut-offs of 130 and 81 MCM2 stained cells per slide respectively._x000D_ The zero false negative approach yielded cut-offs of 44 and 13 MCM2 stained cells for GH and CS patients respectively._x000D_ Optimised J cut-off yielded sensitivities of 81% in CS group in feasibility and 92.3% in validation with specificities of 80.3 and 94.1% respectively. Corresponding NPV and PPVs ranged from 94.2-96% and 73.9-75.0%._x000D_ GH sensitivity by optimised J was 87.5 in feasibility and 70.8% in validation sets. Specificities were 92.6 and 95.8% respectively. NPVs were 97.7 and 76.7%, PPVs were 77.8 and 94.4%._x000D_ Using zero false negative level for cut-off yielded sensitivities close to 100% across all sets, albeit with a concomitant loss of specificity._x000D_

Conclusions

MCM2 is a reliable non-invasive test, potentially useful in diagnosis of primary and recurrent UC. MCM2 may be used to stratify cases where there is no likelihood of disease i.e. zero false negatives, thus avoiding invasive intervention. In this setting false positives are acceptable as the intervention occurs anyway: the stratification aims to reduce intervention for a large subset of true negatives, resulting in huge health economic benefits plus reduction to morbidity and discomfort associated with flexible cystoscopy.

Funding

Cytosystems Ltd., Aberdeen, Scotland, UK