To the editor:
The costs of conducting a clinical trial from phase 1 through phase 3 completion are staggering at greater than $100 million,1 and the cost per patient in a trial has ballooned from ∼$25 000 in 2000 to ∼$100 000 in 2012.2 These high costs underscore the necessity for the information obtained from these trials to be evaluated with the utmost care. During clinical trials, blood from patients is often collected and analyzed for pharmacodynamic and biomarker changes to measure response or toxicity. Traditionally, blood collected into heparinized green-top tubes (GTTs) followed by Ficoll-Hypaque (Ficoll) density gradient separation methods has been used to isolate peripheral blood mononuclear cells (PBMCs).3,4 Recently, one-step closed systems, such as Vacutainer cell processing tubes (CPTs), have been used for PBMC isolation and shipment. We evaluated whether use of these tubes could have an impact on cellular proteins.
A 10-mL glass GTT contains 158 USP units of freeze-dried sodium heparin. The glass sodium heparin CPT used in this study has a 8-mL draw capacity and contains a minimum of 132 USP units of sodium heparin in 1 mL of phosphate-buffered saline solution, 3 g of polyester gel, and 2 mL of Ficoll. Blood from patients with chronic lymphocytic leukemia (CLL) was collected into a Vacutainer GTT or CPT, both containing sodium heparin, and processed immediately (Figure 1A). CPT- and Ficoll-isolated cells had similar levels of apoptosis as measured by flow cytometry using annexin V/propidium iodide staining; however, erythrocyte contamination was greater in CPT-isolated cells (28% vs 11%; n = 5; P = .0036). More strikingly, there was substantial decline in protein levels in CPT-isolated cells, and immunoblot analysis revealed that ZAP-70, Btk, PARP, STAT5, Akt, phospho-RNA Pol2(Ser2), RNA Pol2, Mcl-1, and β-tubulin protein levels were substantially lower in CPT-isolated CLL cells (Figure 1A-B). However, Bcl-2 family member proteins (Bcl-2 and Bak) remained stable as did loading controls glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and β-actin, and there were increased levels of hemoglobin β in CPT-isolated cells, consistent with erythrocyte contamination. When normalized to GAPDH, the decline in ZAP-70, Btk, PARP, STAT5, and Mcl-1 protein levels ranged from 0.1- to 0.5-fold of that of GTTs and was statistically significant, whereas Bcl-2 protein levels were not significantly affected (Figure 1C). In contrast, normalized hemoglobin β was increased to twofold to threefold and was statistically significant (Figure 1D).
To simulate conditions under which blood is drawn into CPTs and then shipped in CPTs overnight, we analyzed PBMCs from CPTs kept at 4°C overnight after processing (Figure 1E). The results were consistent with a substantial decline in proteins in CPT-isolated CLL cells as described for Figure 1A-B. Our data suggest that irrespective of processing time, collection of blood into CPTs may not be optimal for detection of proteins by immunoblot.
If CPTs are not used for blood collection, blood samples collected in GTTs are often shipped overnight to reference laboratories where CPTs may be used for processing, and the question remains whether this affects cellular protein integrity. To simulate this scenario, blood from 6 patients was collected into GTTs and stored at 4°C overnight; then, half the blood in the tube was processed by Ficoll and the remaining half was transferred into a CPT and processed (Figure 1F). Even using the CPT only for processing and not for overnight shipment resulted in greater erythrocyte contamination in the CPT-isolated cells as well as substantial decline in several CLL-relevant proteins, including ZAP-70, Btk, PARP, RNA Pol2, Mcl-1, and β-tubulin.
These results are alarming, and in the context of CLL, the decline in ZAP-70 is of particular relevance, given its application as a prognostic factor.6,7 Moreover, new kinase (Btk,8 PI3K,9 cyclin-dependent10) inhibitors and Bcl-2 antagonists are increasingly used in the clinic for patients with CLL, and loss of cellular proteins may have significant clinical implications (Figure 1G). Although the current work focused on CLL, we feel that these observations would apply to other hematologic and perhaps solid malignancies, and we propose immediate processing of blood collected in GTTs using the Ficoll method to better preserve cellular protein integrity. We hope that the data presented here will be taken into consideration during clinical trial design and decision making.
Acknowledgments: The authors thank Benjamin Hayes for obtaining blood samples and Susan C. Smith for providing information on patient characteristics. This work was supported in part by a grant from the Translational Research Program of the Leukemia and Lymphoma Society of America (LLS R6011-14), a Chronic Lymphocytic Leukemia (CLL) Research Consortium grant from the National Institutes of Health (PO1-CA081534), and a CLL-Global Research Foundation Alliance Grant.
Contribution: L.S.C. designed research, performed experiments, analyzed results, and wrote the manuscript; M.J.K. identified patients for this study, obtained consent for these studies, and reviewed the manuscript; and V.G. conceptualized and supervised research, analyzed data, and reviewed the manuscript.
Conflict-of-interest disclosure: The authors declare no competing financial interests.
Correspondence: Varsha Gandhi, Department of Experimental Therapeutics, The University of Texas MD Anderson Cancer Center, Houston, TX 77030; e-mail:.
- © 2014 by The American Society of Hematology