Extracorporeal photopheresis (ECP) is a treatment modality that entails leukapheresis followed by mixing the buffy coat with methoxsalen (8MOP) and exposing it to UVA light. The buffy coat is then returned to the patient. The treatment is thought to have an immunomodulatory effect and is most commonly used to treat cutaneous T cell lymphoma, acute and chronic graft-versus-host disease (GVHD), and heart and lung allograft rejection. The exact mechanism and optimal dose of cells to be treated is unknown. The primary goal of this study is to examine the quantitative relationship between cell counts in peripheral blood (PBL) and buffy coat collected during ECP. Once this relationship is understood, the PBL cell count may serve as a surrogate marker for cell dose treated and help predicting the efficacy of ECP. The secondary goal of the study is to compare the cell collection efficiency of the two instruments currently used for ECP in the United States, UVAR XTS and CELLEX (Therakos, Exton, PA).
Twenty six patients with GVHD were prospectively recruited for this study which was approved by the institutional review board. ECP was performed with either UVAR XTS (5 cycles with small bowl) or CELLEX (1500 mL of whole blood processed). PBL was drawn within 24 hours prior to the procedure for complete cell count with automated differentials (CBC). At the end of cell collection but prior to 8MOP treatment, 1 mL of buffy coat was drawn from well-mixed treatment bag for CBC. Spearman correlation analysis was performed between cell counts in PBL and buffy coat, including leukocyte counts (WBC), differential counts (neutrophil, lymphocyte and monocyte), and red cell counts (RBC). Collection efficiency was assessed in two ways: (1) the ratio of lymphocyte count in buffy coat to that in PBL (fold enrichment), and (2) the percentage of total lymphocytes collected [(lymphocyte count in buffy coat × treatment volume × 100)/(lymphocyte count in PBL × total blood volume estimated based on height and weight)]. These two variables were compared between UVAR XTS and CELLEX by Mann Whitney test. A double-sided p value was set at 0.05. Data visualization and statistical analysis were performed using GraphPad Prism 5 and IBM SPSS 19.
A total of 22 patients were included in the final analysis after excluding 4 patients due to incorrect sampling of buffy coat. Seven patients were treated with UVAR XTS and 15 patients with CELLEX. The demographics (gender, race, age, weight, height, and estimated total blood volume), peripheral WBCs, Hemoglobin and hematocrits were not significantly different between the two groups.
Lymphocyte counts in PBL significantly correlated with those in buffy coat collected by UVAR XTS (r=0.79, p<0.05) and CELLEX (r=0.92, p<0.0001). Linear relationships were observed with a slope of 1.7 (95% CI, 1.3–2.2) for UVAR XTS and a slope of 5.1 (95% CI, 4.7–5.4) for CELLEX (Figure 1A). Monocyte counts in PBL and buffy coat collected by UVAR XTS (r=0.82, p<0.05) and CELLEX (r=0.84, p<0.0001) also had robust correlations. There were no significant correlations for RBCs or neutrophil counts between PBL and buffy coat collected by either instrument. For the overall WBCs, there was only a weak but significant correlation for CELLEX (r=0.53, p<0.05).
UVAR XTS enriched the lymphocyte counts by 2.2 ± 0.4 folds in buffy coat compared to PBL during ECP while CELLEX enriched the lymphocyte counts by 5.3 ± 0.4 folds (Mean ± SE, p<0.001, Figure 1B). UVAR XTS collected 9.9 ± 5.2% of total lymphocytes during each treatment while CELLEX collected 19.8 ± 8.6% of total lymphocytes (Mean ± SE, p<0.005, Figure 1C).
There is a robust linear relationship between PBL and buffy coat lymphocyte counts, which may allow PBL lymphocyte count to be used in future studies as a surrogate for number of cells treated. This relationship also ensures a practical way to correlate cell dose with outcome. CELLEX is significantly more efficient in collecting lymphocytes than UVAR XTS. Approximately twice as many lymphocytes are treated with CELLEX than with UVAR XTS during each procedure.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.