Hematopoietic stem cell transplant (HSCT), using cytokine‐mobilized peripheral blood (MPB) stem cells, harvested by apheresis, is well established as a potential curative therapy for many patients suffering with various hematologic disorders and malignancies, immune deficiencies, and congenital diseases. However, timely reconstitution of multilineage hematopoiesis in these patients requires infusion of at least 2 × 106 viable hematopoietic stem/progenitor cells (HSCs)/kg patient weight. Human HSCs responsible for engraftment are contained within a population of cells that express the cell surface antigen CD34. Accurate, cross‐platform methods of counting viable CD34+ cells by single platform flow cytometry based on the ISHAGE Guidelines (Keeney et al., 1998) are well established as the “Gold Standard” means to measure graft adequacy in the blood and marrow transplant setting.
The single platform ISHAGE protocol utilizes 1× NH4Cl as a red blood cell (RBC) lysis buffer. However, this buffer is pH dependent and therefore requires daily preparation from a 10× stock. Additionally, if lysed samples are not analyzed immediately (or within 1 hr if kept on melting ice), NH4Cl has been demonstrated to induce apoptosis/necrosis in CD34 expressing cells, thus potentially impacting the enumeration of viable CD34+ HSCs (Keeney et al., 1998). Furthermore, NH4Cl promotes further apoptosis and cell death in post‐thawed samples lysed with this agent and is not recommended for use thereon. Therefore, alternative lysis reagents which are more convenient for daily use and which may better preserve CD34+ HSC viability could impact the ability to accurately enumerate viable CD34+ cells in both the “fresh” and post‐thawed settings. Versalyse™ (ImmunoTech S.A.S., Marseille, France) is an alternative lysis buffer that, unlike NH4Cl, is available in a ready‐to‐use format, can be stored at room temperature, and has a 90‐day shelf‐life after opening.
In this brief communication, we provide a preliminary evaluation of Versalyse in comparison to NH4Cl for CD34+ HSC enumeration, as well as an assessment of post‐lysis CD34+ HSC viability.
2 MATERIALS AND METHODS
2.1 Flow cytometers
Sample analysis was performed on two flow cytometers, a 5‐color Cytomics FC500 flow cytometer (Beckman Coulter®, Miami, FL), equipped with a single laser (488 nm [blue]), and a 10‐color Navios flow cytometer (Beckman Coulter) equipped with three lasers (405 nm [violet], 488 nm [blue], and 638 nm [red]).
2.2 Cytomics FC500 setup and quality control
Instrument quality control was performed daily prior to sample acquisition using Flow‐Check™ and Flow‐Set™ Fluorospheres (Beckman Coulter). Instrument compensation was performed monthly, or as required (based on daily photomultiplier tube [PMT]) monitoring, using CYTO‐COMP cells (CYTO‐COMP™ Cell Kit; Beckman Coulter) and the QuickComp 2™ Kit (Beckman Coulter). Generated compensation matrixes were verified using normal peripheral blood (PB) samples spiked with Stem‐Trol™ Control Cells (Beckman Coulter), as outlined in the Stem‐Kit instructions for use (IFU; Beckman Coulter). Verification of instrument setup and compensation was performed using NH4Cl‐treated samples; however, modifications were performed, as required, for Versalyse‐treated samples to ensure specimens were run with optimal instrument settings.
2.3 Navios setup and quality control
Instrument quality control was performed daily prior to sample acquisition using Flow‐Check Pro and Flow‐Set Pro Fluorospheres (Beckman Coulter). Instrument compensation was performed as needed (based on daily PMT monitoring) using VersaComp Antibody Capture Beads (Beckman Coulter) and CYTO‐COMP cells (7‐AAD assessment). Compensation matrixes were generated using Kaluza software and verified using previously assessed MPB or apheresis samples with known viability and CD34+ HSC values. Verification of instrument setup (light scatter, PMT voltage, and compensation) was performed separately for NH4Cl‐lysed samples and Versalysed samples to ensure that data could be acquired under optimal conditions for each lysing agent.
2.4 Sample collection, storage, and preparation
A total of 59 specimens, including fresh MPB (n = 33), and fresh apheresis (n = 26) samples from both normal allogenic donors, as well as, malignant patients (e.g., myeloma, acute myeloid leukemia) undergoing autologous collection for subsequent transplant, were processed. Samples were collected, stored, and prepared following the modified ISHAGE guidelines for CD34+ HSC enumeration as outlined in the Stem‐Kit IFU. All reagents utilized for sample preparation, with the exception of Versalyse, are components of the commercially available FDA‐approved Stem‐Kit (Beckman Coulter). In brief, duplicate samples were incubated with CD45‐FITC and CD34‐PE antibodies, as well as, the viability dye 7‐AAD, and erythrocytes were lysed using either 1× NH4Cl or Versalyse. Samples were analyzed on both instruments (FC500 and Navios), within 1 hr of sample preparation. Between runs, samples were stored on melting ice. As per the Stem‐Kit IFU, 75,000 viable CD45+ events were collected per sample tube with a minimum of 100 CD34+ events collected per tube or until instrument time‐out (5 min [Navios]; 10 min [FC500]). CD34+ HSCs were identified as CD34+, CD45dim events, with low side scatter and low to intermediate forward scatter per ISHAGE guidelines. Samples were excluded if duplicate tube CD34+ HSC counts were >10% from the mean of the duplicate tubes (samples with working CD34+ HSC count of >20 cells/μl [n = 2]) or a >15% from the mean of duplicate tubes (samples with working CD34+ HSC count of ≤20 cells/μl [n = 2]). One sample was excluded due to high variation in the viability of the CD34+ HSC fraction.
2.5 Sample analysis and gating strategy
Automated sample gating and analysis was performed using stemCXP™ software for all samples run on the FC500 flow cytometer. The gating strategy utilized has been described in detail previously by Sutherland and Keeney (2016) in their report, Enumeration of CD34+ Cells by Flow Cytometry, published in Cellular Therapy: Principles, Methods, and Regulations, 2nd Edition. All samples were visually inspected for obvious errors and manual gating was performed only when obvious errors in automated gate placement were observed (n = 8; 3% of acquired data). For all samples run on the Navios flow cytometer, manual gating and analysis was performed offline on the FCS2 component of files after extraction with FlowJo software (v9.9.6, Treestar, Ashland, OR). The extracted FCS2 files were analyzed using ISHAGE analysis templates in Cellquest Pro 6 (BD Biosciences, San Jose, CA). For both FC500 and Navios platforms, average CD34+ HSC cells/μl was calculated by determination of the mean CD34+ HSC count across duplicate samples for each lysis buffer and subsequent multiplication by the appropriate dilution factor, when performed.
2.6 Statistical analysis
Statistical analysis was performed using Graph Pad Prism 6.0 (San Diego, CA). Correlation between matched NH4Cl‐ and Versalyse‐treated samples was assessed using linear regression analysis. Bias in CD34+ HSC enumeration between lysis buffers was assessed using Bland Altman analysis (percent difference). Differences in mean CD34+ HSC enumeration between lysis buffers was assessed using Wilcoxon matched pairs signed rank sum analysis, with p ≤ .05 considered statistically significant.
Data collected throughout this preliminary study demonstrated a strong correlation in CD34+ HSC enumeration between both lysis buffers in apheresis and MPB samples (Figure 1). In fact, compiled analysis demonstrated an r2 value of >0.98 across both sample types and both instruments. The mean number of CD34+ HSCs in NH4Cl‐ and Versalyse‐treated apheresis (1505 and 1493 cells/μl, respectively) and MPB (57 and 55 cells/μl, respectively) samples was not significantly different (p > .05) when assessed using the FC500 flow cytometer (Table 1). Similar results (p > .05) were also demonstrated for NH4Cl‐ and Versalyse‐treated apheresis (1496 and 1471 cells/μl, respectively) and MPB (55 and 54 cells/μl, respectively) samples assessed using the Navios flow cytometer. Bland Altman analysis for bias demonstrated a <2% positive bias toward NH4Cl‐treated samples (range: 0.49–1.63%), across all sample types and instruments utilized; data not shown). Additionally, the CD34+ HSC fraction demonstrated average cell viabilities of >95% in apheresis and MPB samples, across both lysis buffers, on either instrument (data not shown). Finally, CD45+ WBC enumeration demonstrated comparable mean WBC counts (<10% difference) between NH4Cl and Versalyse‐treated samples in >99% of investigated samples (data not shown).
|CD34+ HSC (cells/μl)||p value||CD34+ HSC (cells/μl)||p value|
|Mean/median (range)||Mean/median (range)||Mean/median (range)||Mean/median (range)|
|Apheresis||1505/1040 (408–5855)||1493/1015 (365–5705)||.249||1496/1030 (347–5906)||1471/1029 (350–5881)||.275|
|Mobilized peripheral blood||57/38 (10–207)||55/40 (10–176)||.140||55/37 (9–186)||54/35 (9–185)||.070|
In this study, we performed a preliminary comparison of CD34+ HSC enumeration using 1× NH4Cl versus Versalyse RBC lysing agents. In so doing, we demonstrated that there was a strong correlation in CD34+ HSC enumeration between NH4Cl‐ and Versalyse‐treated apheresis and MPB samples. Additionally, no significant difference in CD34+ HSC enumeration was noted between the two lysing agents, and CD34+ HSC cell viabilities were comparable (>95%) in investigated samples.
Samples were excluded if the determined CD34+ counts for each duplicate tube was not within 10% of the mean of the duplicate tubes, as per ISHAGE guidelines. However, this criterion was expanded to within 15% for samples with a CD34+ HSC working count of ≤20 cells/μl, due to significant limitation of the allowable variability between duplicate tubes at these low cell numbers (≤2 cells/μl). Overall, 1 apheresis (>20 cells/μl) and three MPB samples (>20 cells/μl [n = 1], ≤20 cells/μl [n = 2]) were excluded. Additionally, one MPB sample was excluded as there was a high degree of variability in the viability of the CD34+ HSC fraction. Although this is not an established ISHAGE guideline, and this sample passed the difference criteria outlined above, it was an outlier in the data set and the high variability suggested that enumeration may not be accurate in this sample. Of the remaining samples analyzed in this study, comparable results in terms of CD34+ HSC enumeration was demonstrated across both lysis agents.
Although comparable in terms of CD34+ HSC enumeration, slight differences in NH4Cl‐ and Versalyse‐treated samples were noted throughout the study. First, Versalyse‐treated samples demonstrated distinct light scatter changes, specifically, decreased median side scatter and increased median forward scatter compared to their NH4Cl‐treated counterparts, thus requiring flow cytometer settings to be optimized separately for NH4Cl‐ and Versalyse‐treated samples. Additionally, it is noteworthy, that while samples lysed in NH4Cl require storage on ice, preliminary analysis on apheresis samples, showed Versalyse‐treated samples stored at room temperature, for up to 2 hr, had comparable CD34+ HSC viability to those stored on ice (data not shown).
It is noteworthy, that alternative RBC lysis agents are commercially available, and comparisons of various lysis agents for CD34+ HSC enumeration have been previously performed (Menéndez et al., 1998). However, as this study was specifically designed to evaluate the utility of Versalyse as a RBC lysis agent, additional lysis agents were not included. Moving forward, users intending to investigate alternative lysis agents for use with the ISHAGE protocol, must also ensure compatibility with all reagents (e.g., 7‐AAD), as some may contain powerful fixatives, thus impacting results.
Overall, this preliminary study provides proof of principal that Versalyse may have potential as an alternative lysis agent to NH4Cl for CD34+ HSC enumeration, with the advantage of availability in a ready‐to‐use format, significantly improved reagent stability (>90 day), and the potential for greater flexibility in sample storage after preparation.
All reagents used for this study were kindly provided by ImmunoTech S.A.S, a Beckman Coulter company.
6 CONFLICT OF INTEREST
Michael Keeney and Benjamin D. Hedley hold consultancy positions with Beckman Coulter, Inc. D. Robert Sutherland and Lori E. Lowes have no conflicts of interest to disclose. Neither Immunotech S.A.S., nor Beckman Coulter, were involved in the design of this study, or in the subsequent manuscript preparation.