P217 | AN AUTOMATED CARTRIDGE-BASED MICROFLUIDIC SYSTEM FOR REAL-TIME QUANTIFICATION OF BCR::ABL1 TRANSCRIPTS IN CHRONIC MYELOID LEUKEMIA: AN ITALIAN EXPERIENCE

Introduction: Chronic myeloid leukemia (CML) is a hematologic malignancy characterized by the BCR::ABL1 fusion gene. About 99% of cases express the e13a2 or e14a2 transcript, encoding the p210 protein with abnormal tyrosine kinase activity. The gold standard for CML monitoring relies on BCR::ABL1 transcript quantification by reverse transcription quantitative PCR (RT-qPCR). The Xpert^® BCR-ABL Ultra assay is an automated diagnostic test designed to quantify BCR::ABL1 and ABL1 mRNA in patients with the e13a2/e14a2 transcript, performing RNA extraction, cDNA synthesis, nested real-time PCR, and detection in a closed cartridge system in under 3 hours. This study assessed the Xpert^® BCR-ABL Ultra assay as a potential alternative to methods described in the Italian Laboratory Recommendations (RIL). Methods and Results: 129 leftover peripheral blood samples from CML patients at different disease stages were analyzed: 66 between MR 3.0 (0.1% IS) and MR 1.0 (10% IS), 27 at MR 4.0 (0.01% IS), 16 at MR 4.5 (0.0032% IS), and 20 undetectable or at MR 5 (<0.001% IS). Analyses were conducted in four Italian CML LabNet reference laboratories, each employing the Xpert^® BCR-ABL Ultra assay and their validated local method. Additionally, 20 healthy donor samples and two reference materials (ACROMETRIX BCR::ABL1 Panel and UK NEQAS controls) were tested. Repeatability for both methods was excellent, with coefficients of variation consistently <1%. Concordance was evaluated according to EUTOS criteria, with all laboratories meeting at least 2 out of 3 criteria. Bland-Altman analysis showed no significant bias between methods. For each method and disease level, coefficient of repeatability, standard deviation, mean %IS, coefficient of variation, bias, and 95% limits of agreement were calculated and reported in the Table. The overall concordance for MR values was 71.3% using the European LeukemiaNet guidelines, improving to 89.9% when applying the formula for calculating logarithmic reduction: 2-log10(%IS). The Xpert^® system also demonstrated good performance on negative samples and reference materials. Conclusion: Statistical analysis of data obtained with different systems and multiple types of material demonstrated a strong and consistent correlation across all disease levels. Xpert^® BCR-ABL Ultra proved to be a robust, rapid, and efficient tool, representing a valuable alternative to clinical methods currently validated within the Italian CML LabNet.