Development, Evaluation, and Multisite Deployment of a Machine Learning Decision Tree Algorithm To Optimize Urinalysis Parameters for Predicting Urine Culture Positivity.

PittUDT, a recursive partitioning decision tree algorithm for predicting urine culture (UC) positivity based on macroscopic and microscopic urinalysis (UA) parameters, was developed in support of a broader system-wide diagnostic stewardship initiative to increase appropriateness of UC testing. Reflex algorithm training utilized results from 19,511 paired UA and UC cases (26.8% UC positive); the average patient age was 57.4 years, and 70% of samples were from female patients. Receiver operating characteristic (ROC) analysis identified urine white blood cells (WBCs), leukocyte esterase, and bacteria as the best predictors of UC positivity, with areas under the ROC curve of 0.79, 0.78, and 0.77, respectively. Using the held-out test data set (9,773 cases; 26.3% UC positive), the PittUDT algorithm met the prespecified target of a negative predictive value above 90% and resulted in a 30 to 60% total negative proportion (true-negative plus false-negative predictions). These data show that a supervised rule-based machine learning algorithm trained on paired UA and UC data has adequate predictive ability for triaging urine specimens by identifying low-risk urine specimens, which are unlikely to grow pathogenic organisms, with a false-negative proportion under 5%. The decision tree approach also generates human-readable rules that can be easily implemented across multiple hospital sites and settings. Our work demonstrates how a data-driven approach can be used to optimize UA parameters for predicting UC positivity in a reflex protocol, with the intent of improving antimicrobial stewardship and UC utilization, a potential avenue for cost savings.

Incidence and Management of Therapeutic Monoclonal Antibody Interference in Monoclonal Gammopathy Monitoring.

BACKGROUND: The treatment of multiple myeloma (MM) has been revolutionized by the introduction of therapeutic monoclonal antibodies (tmAbs). Daratumumab, a human IgG1/κ tmAb against CD38 on plasma cells, has improved overall survival in refractory MM and was recently approved as a frontline therapy for MM. Work on tmAb interference with serum protein electrophoresis (SPE) during MM monitoring has failed to provide information for laboratories on incidence of interference and effective methods of managing the interference at a practicable level. We aimed to evaluate daratumumab and elotuzumab interference in a large academic hospital setting and implement immediate solutions. METHODS: We identified and chart reviewed all cases of possible daratumumab interference by electrophoretic pattern (120 of 1317 total cases over 3 months). We retrospectively reviewed SPE cases in our laboratory to assess clinical implications of tmAb interference before the laboratory was aware of tmAb treatment. We supplemented samples with daratumumab and elotuzumab to determine the limits of detection and run free light chain analysis. RESULTS: Approximately 9% (120 of 1317) of tested cases have an SPE and/or immunofixation electrophoresis (IFE) pattern consistent with daratumumab, but only approximately 47% (56) of these cases were associated with daratumumab therapy. Presence of daratumumab led to physician misinterpretation of SPE/IFE results. Limits of daratumumab detection varied with total serum gammaglobulin concentrations, but serum free light chain analysis was unaffected. CONCLUSIONS: Clinical laboratories currently rely on interference identification by electrophoretic pattern, which may be insufficient and is inefficient. Critical tools in preventing misinterpretation efficiently include physician education, pharmacy notifications, separate order codes, and interpretive comments.