B-278 Pilot Clinical Study of the CarePoint Assay

Abstract Background An estimated >710 000 individuals live with bladder cancer in the United States, with approximately 82 290 new cases expected this year (2023). Recurrence rates for bladder cancer are high: approximately 65% of patients with non-invasive or in situ tumors and 73% of patients with more advanced disease experience recurrence within five years. Consequently, patients treated for bladder cancer require prolonged surveillance to safeguard against tumor recurrence. The gold standard of surveillance, cystoscopy—an uncomfortable, expensive, and invasive procedure—is recommended every three to 12 months. Consequently, there is a need for an inexpensive, effective, noninvasive diagnostic tool as an alternative to cystoscopy. The CarePoint Assay (CPA) is a 30-minute electrochemical immunoassay for the detection of disease-state biomarkers in human urine. The analytical performance of CPA has been evaluated for method comparison with a three-hour sandwich immunosorbent assay (ELISA) for bladder tumor antigen (BTA), a diagnostic biomarker for bladder cancer. Method Receiver operating characteristic (ROC) curves were calculated by simple logistic regression with Prism9 (GraphPad) using patient urine samples across multiple medical centers (n = 51), comparing BTA concentration (continuous variable) with high-grade tumor status (binary variable). Patient urine samples were evaluated by CPA using commercial potentiostats for comparison with a sandwich ELISA. The data was analyzed using simple linear regression. Results Patient urine evaluated by CPA was interpolated by simple linear regression using a three-point calibration with standard spiked into urine pooled from healthy donors. Data produced by sandwich ELISA was interpolated by five-parameter nonlinear regression using a 12-point standard curve. Both methods showed good correlation: (r) = 0.93. Simple logistic regression comparing analyte concentration to high-grade tumor status produced a sensitivity and specificity of 52% and 88% with an AUROC of 0.71 for the CPA, and a sensitivity and specificity of 72% and 77% with an AUROC of 0.77 for the sandwich ELISA. Sensitivities and specificities were calculated using Youden’s J statistic. Conclusions We established proof of concept for a rapid, electrochemical immunoassay for a diagnostic biomarker of bladder cancer. The CPA correlated well with sandwich ELISA, an established method for measuring biomarkers in bodily fluids. Future work will focus on refinement of the CPA architecture for point-of-care use and subsequent evaluation of analytical performance.