用于免疫治疗临床试验的稳健且精确的稀有事件分析

Modulating the human immune system has become a lifesaving option for selected malignancies. Biomedical approaches with both engineered cytotoxic T cells and cancer vaccines that boost antigen‑specific cytotoxic T cells (CTLs) continue to advance the field. Both rely on rare event analysis, detecting and quantifying interactions between uncommon biomarkers and common leukocyte antigens. Flow cytometry is essential for today's clinical research and is the gold standard for identifying and phenotyping very low‑frequency events in large mixed cell populations. Careful assay design is required for success, taking into consideration background, brightness, resolution, frequency, and donor to donor variation. We established a practical framework for robust, precise rare event analysis. First, define the rare events of interest-such as CAR T cells in peripheral whole blood, antigen‑specific CTLs, circulating tumor cells (CTC), minimal residual disease (MRD), or T lymphocyte stem cells. Sample volume to be tested is critical in order to achieve an appropriate level of precision. For example, achieving about 20% CV (coefficient of variation) at a target frequency of 1 in 100,000 typically requires roughly 2.5 million leukocytes. Inadequate sample volume may yield false negative results and often be a cause of a failed assay. Key performance limits-Lower Limit of Quantitation (LLOQ), Limit of Detection (LOD), and Limit of Blank (LOB)-must be part of the performance characterization of the method. In CAR T immunophenotyping, we assess LLOQ to find a linear, reliable range, then spike CAR T cells across serial dilutions in triplicate. The lowest dose with < 35% CV across donors defines the LLOQ. Because rare event work can be computationally heavy, concentrating the target population helps, for example, removing 80% of neutrophils during export of raw flow cytometry fcs data and using a dump channel containing with CD19, CD4, and CD56 focuses the analysis on antigen‑specific CTLs within the CD8 subpopulation cells. In summary, our approach improves precision of rare event analysis by addressing data imbalance, statistical bias, and sample inefficiency.