Background Killing of targeted tumors during adoptive cell transfer therapy is associated with cytotoxic T lymphocyte (CTL) numbers,immunophenotype,tumor-specificity,and in vivo residence time,migration,and distribution.Therefore,tracing in vivo persistence,migration,and distribution of CTLs is important for cancer immunotherapy.Methods Optimal staining concentration for CTL proliferation was determined by cell counting kit-8 (CCK-8) assay and killing efficiencies of CTLs or carboxyfluorescein diacetate succinimidyl ester (CFSE)-labeled melanoma antigen-specific cytotoxic T lymphocytes (CFSE-CTLs) for malignant melanoma cells in vitro were compared.Additionally,CFSE-CTLs were intravenously transfused to mice receiving B16 melanoma,and their residence time,migration,and distribution in vivo were observed by measuring fluorescence intensities of CFSE-CTLs per gram of tissue (％FI/g) in various tissues and analyzing tumor/non-tumor (T/NT) values.Anti-tumor effects of transferred CTLs and correlation between ％FI/g and D-value of tumor size were analyzed.Results Five-micromolar CFSE was optimal for labeling CTLs with minimal cytotoxicity.No significant difference occurred between CTLs and CFSE-CTLs for tumor cell killing (P=0.849) or interleukin-2 (P=0.318) and interferon-y (P=0.201)levels.Distribution of CTLs in vivo varied with time.A negative correlation between ％FI/g in tumors and D-value of tumor sizes by Spearman correlation analysis was observed.CTLs were recruited to and killed tumors from 6 hours to 3 days after cell infusion.CTLs were observed up to three weeks later in the tumor,liver,kidneys,and spleen; this was related to the abundant blood supply or the nature of immune organs.Conclusions CCK-8 assay is a novel method to select optimal CFSE staining concentrations.Fluorescence intensity of transferred CTLs reflects their killing efficiency of tumors.CFSE fluorescent markers can trace in vivo CTL persistence,migration,and distribution because of its stability,long half-life,and low toxicity.