RESUMEN
BACKGROUND AND OBJECTIVES: Blood products are scarce resources. Audits on the use of red blood cells (RBCs) in tertiary centers have repeatedly highlighted inappropriate use. Earlier retrospective audit at our local community hospitals has demonstrated that only 85% and 54% of all requests met Choosing Wisely Canada guidelines for pre-transfusion hemoglobin (Hb) of 80 g/L or less and single unit, respectively.We sought to improve RBC utilization by 15% over a period of 12 months (meeting Choosing Wisely Canada criteria of pre-transfusion Hb ≤80g/L by >80% and single-unit transfusion by >65%). METHODS: Following repeated PDSA (Plan-Do-Study-Act) cycles, we implemented educational strategies, prospective transfusion medicine (TM) technologist-led screening of orders, and an RBC order set. RESULTS: The 3-month median percentages of appropriate RBC use for pre-transfusion Hb and single unit (September-November 2021) across all 3 hospitals were 90% and 71%, respectively. Overall, the rate of appropriate RBCs based on pre-transfusion Hb remained above target (>80%), with minimal improvement across all hospitals (median percentage at pre- and post-technologist screening periods of 87% and 90%, respectively). The median percentage of appropriate RBCs based on single-unit transfusion orders has improved across all Niagara Health hospitals with sustained targets (3-month median percentage at pre- and post-technologist screening and most recent time periods of 54%, 56%, and 71%, respectively). CONCLUSIONS: We have taken a collaborative, multifaceted approach to optimizing utilization of RBCs across the Niagara Health hospitals. The rates of appropriate RBC use were comparable with the provincial and national accreditation benchmark standards. In particular, the TM technologist-led screening was effective in producing sustained improvement with respect to single-unit transfusion. One of the balancing outcomes was increasing workload on technologists. Local and provincial efforts are needed to facilitate recruitment and retention of laboratory technologists, especially in community hospitals.
RESUMEN
Cancer chemotherapy responses have been related to multiple pharmacogenetic biomarkers, often for the same drug. This study utilizes machine learning to derive multi-gene expression signatures that predict individual patient responses to specific tyrosine kinase inhibitors, including erlotinib, gefitinib, sorafenib, sunitinib, lapatinib and imatinib. Support vector machine (SVM) learning was used to train mathematical models that distinguished sensitivity from resistance to these drugs using a novel systems biology-based approach. This began with expression of genes previously implicated in specific drug responses, then expanded to evaluate genes whose products were related through biochemical pathways and interactions. Optimal pathway-extended SVMs predicted responses in patients at accuracies of 70% (imatinib), 71% (lapatinib), 83% (sunitinib), 83% (erlotinib), 88% (sorafenib) and 91% (gefitinib). These best performing pathway-extended models demonstrated improved balance predicting both sensitive and resistant patient categories, with many of these genes having a known role in cancer aetiology. Ensemble machine learning-based averaging of multiple pathway-extended models derived for an individual drug increased accuracy to >70% for erlotinib, gefitinib, lapatinib and sorafenib. Through incorporation of novel cancer biomarkers, machine learning-based pathway-extended signatures display strong efficacy predicting both sensitive and resistant patient responses to chemotherapy.
