Analytical Performance of Two Flow Cytometry Systems for CD34⁺ Hematopoietic Stem Cell and CD3⁺ T-Cell Enumeration in Stem Cell Transplantation.

Background Accurate enumeration of hematopoietic stem cells (HSC) and T-lymphocytes is essential for stem cell transplantation (SCT), as engraftment success depends on the infusion of adequate CD34⁺ and CD3⁺ cell doses. Flow cytometry is the gold standard for quantifying these subsets; however, analytical performance may vary between platforms, necessitating formal validation before clinical implementation. Therefore, this study aimed to evaluate the analytical precision of the DxFLEX flow cytometer and to assess its correlation and agreement with the established FACSCalibur reference method for CD34⁺ HSC and CD3⁺ T-cell enumeration. Materials and methods Analytical precision and method-comparison studies were conducted in accordance with Clinical and Laboratory Standards Institute (CLSI) EP15-A3 guidelines. Two levels of third-party quality control (QC) materials were used for CD34⁺ and CD3⁺ assays. Within-run precision was assessed using ten replicates per QC level (n =10), while between-run precision was evaluated across five days with three runs per day (n = 15). Mean values, standard deviations (SD), and coefficients of variation (CV) were compared with QC performance specifications. Method correlation was performed using 40 specimens following CLSI H57-A guidelines. Correlation coefficients and Bland-Altman analyses were used to assess agreement between the evaluated system (DxFLEX) and the reference analyzer (FACSCalibur). Results Both within-run and between-run precision for CD34⁺ and CD3⁺ measurements fell within acceptable QC limits, demonstrating stable assay performance. Correlation analyses showed excellent agreement between the two flow cytometry platforms, with correlation coefficients (r = 0.999 for CD34⁺ absolute counts and r = 1.00 for CD3⁺ absolute counts). Agreement between instruments was further supported by Bland-Altman analysis. For CD34⁺ measurements, the relative mean bias was -1.2% (limits of agreement: -33.4% to 31.1%) for percentage values and -0.479 (limits of agreement: -20.415 to 19.457) for absolute counts. For CD3⁺ measurements, the mean bias was -0.3% (limits of agreement: -3.5% to 2.9%) for percentage values and -0.358 (limits of agreement: -14.129 to 13.413) for absolute counts. These findings indicate strong concordance between the evaluated system and the reference platform, with minimal systematic differences and clinically acceptable limits of agreement. Conclusion The evaluated flow cytometry platform demonstrated excellent analytical performance for CD34⁺ and CD3⁺ enumeration, with precision and correlation results meeting clinical laboratory standards. Strong agreement with the reference method supports its suitability for routine use in HSC monitoring and SCT-related immunophenotyping.