Cytometry Part B (Clinical Cytometry) 88B:361–370 (2015)
Original Article
Novel Lymphocyte Screening Tube Using Dried Monoclonal Antibody Reagents Benjamin D. Hedley,1* Michael Keeney,1 Janice Popma,1 and Ian Chin-Yee2 1
Pathology and Laboratory Medicine, Division of Hematology, London Health Sciences Centre, London, Ontario N6K 5W9, Canada 2 Department of Medicine, Division of Hematology, London Health Sciences Centre, Schulich School of Medicine Western University of Ontario, London, Ontario N6K 5W9, Canada
We previously developed a 10-color 11-antibody combination including a viability dye, to screen T-, B-, and natural killer (NK)-cell populations in blood, bone marrow, tissue, and body fluids. Recently, Beckman Coulter has introduced a line of dried reagents that, unlike liquid reagents and cocktails, require no refrigeration, titration, or manipulation before using. We evaluated custom tubes based on our standard lymphocyte screening panel, focusing on comparative analysis, ease of use, and advantages compared with our liquid reagent set. We tested 42 samples from blood (n 5 15), bone marrow (n 5 17), and tissue (n 5 10) with the combination CD4/CD8/KAPPA/LAMBDA/CD19/CD56/CD5/CD20/CD10/CD3/CD45 and a vital dye by both methods and compared positivity and staining intensity for each antigen. Of the 42 samples, 5 were normal samples, 3 were red cell disorders, 20 were B-cell malignancies, 5 T-cell malignancies, 4 myeloid malignancies, and the remaining 5 were other diagnoses. Dried reagents gave equivalent staining intensity results to our standard panel in a variety of sample types, with diagnoses including reactive lymphocytosis, chronic lymphocytic leukemia, and various lymphomas. Our standard panel for evaluation of mature lymphoid malignancies allows rapid assessment of any sample type while providing direct assessment of viability. The dried reagent tube reduces preanalytical work, with simple addition of sample and the viability dye to the tube, saving time, reducing potential errors, and obviating need to titrate and monitor individual antibodies. With a shelf life of at least 12 months, the reagents also offer potential savings in reagent costs by reducing wastage due to expiration C 2015 International Clinical Cytometry Society or tandem breakdown in standard liquid formulation. V Key terms: duraclone; flow cytometry; screening tube; acute leukemia; immunophenotyping
How to cite this article: Hedley BD, Keeney M, Popma J and Chin-Yee I. Novel Lymphocyte Screening Tube Using Dried Monoclonal Antibody Reagents. Cytometry Part B 2015; 88B: 361–370.
In recent years, the clinical flow cytometry laboratory has seen the introduction of highly sophisticated instrumentation that can detect up to 10 fluorochromes on FDA-approved cytometers (1). Major advantages of these instruments include the ability to analyze small tissue samples or body fluids, less redundancy in reagent use, and improved efficiency for the laboratory (2–7). Although the increase in the tube complexity has led to fast and, in some cases, more accurate analysis of patient samples, this has given rise to different challenges that have to be addressed in a high-volume clinical laboratory. Running a large standardized panel allows for greater reproducibility and efficiency but could potentially increase cost. Specifically, using a standardized panel developed by the Euroflow group (8,9), in our terti-
C 2015 International Clinical Cytometry Society V
ary care laboratory, with access to morphology, pathology, molecular, and cytogenetic results was not deemed cost-effective. Additional Supporting Information may be found in the online version of this article. This study was not funded by Beckman Coulter; however, MK is a consultant for Beckman Coulter. All authors contributed equally to the study design, analysis of samples, and manuscript review. *Correspondence to: B. D. Hedley, Pathology and Laboratory Medicine, Division of Special Hematology, London Health Sciences Centre, 800 Commissioners Road East, London, Ontario N6K 5W9, Canada. E-mail: [email protected] Received 19 December 2014; Revised 9 April 2015; Accepted 27 April 2015 Published online 17 July 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/cyto.b.21251
362
HEDLEY ET AL.
As an alternative approach, we developed a 10-color 11-antibody panel, which also includes a viability dye, to screen T-, B-, and natural killer (NK)-cell populations in a variety of blood, bone marrow, tissue, and body fluids. Morphology remains an essential part of the flow cytometry workup, further reducing the risk of missing other hematological neoplasms. Cases are not reported without morphological review. Reporting of results follows Bethesda guidelines, and although percentages of the abnormal population as a subset of the total nucleated cells or CD451 cells is given, percentage positive or negative for each cell marker is not reported (6,10). The screening panel used in our laboratory is a 10-color 11-antibody tube that also includes a viability dye. It is routinely used to screen blood, bone marrow, tissue, and body fluids for T-, B-, and NK-cell populations. This tube contains CD4/CD8/KAPPA/LAMBDA/ CD19/CD56/CD5/CD20/CD10/CD3/CD45 (Supporting Information Table S1) and has been made routinely into a cocktail in liquid format, from various suppliers, and prepared by technologists in large batches (150 tests). Prior to creation of a cocktail, each individual antibody must be titrated, validated, and then combined with each of the other antibodies. Once combined, individual antibodies titration volumes may have to be further adjusted prior to evaluation against the current reagents. Cocktail preparation is a complex time-consuming process that requires the technologist to individually pipette each of the antibodies into the batch. Although infrequent, errors in preparation can occur when pipetting large numbers of antibodies, resulting in wasted antibody incurring expense and potential misinterpretation of patient samples. Quality assurance of reagent cocktails requires rigorous documentation of each step and regular testing to assess the quality of each batch of cocktail (http://www. cap.org/ pages 9 and 10 of Flow Cytometry Checklist). This takes into account: variations from lot to lot in the intensity of the number of tests to be created, the number of tests in each vial, and the lot(s) of each individual antibody. It takes a significant amount of time performing and recording all of these steps in cocktail preparation, which could be spent analyzing and interpreting patient samples. Another long-standing issue that laboratories have had to be aware of is the instability of tandem dyes. Lot to lot variation has been reported requiring recalculation of compensation matrices when working with tandem dyes (11–14). The current generation of 10-color flow cytometers allows for a greater number of tandem dyes to be used. Some newer tandem dyes are more prone to breakdown than others, effectively reducing the shelflife of liquid format cocktails containing these tandem dyes. Lately, Beckman Coulter introduced a range of dried reagents that are stable at room temperature with much longer shelf life (including tandem dyes for phycoerythrin (PE) and allophycocyanin (APC)). In addition, they offer this product along with their custom design service, allowing flow cytometry laboratories to adapt
their own laboratory derived tests. Also supplied are single color dry format tubes for compensation, and a single batch may be purchased for 12 months requiring only once to be validated, eliminating the need to titrate, pipette, validate, and document individual batches of liquid format cocktail. The goal of this study was to demonstrate at least equivalence in performance of the dried format reagent when compared with the currently used liquid format. MATERIALS AND METHODS Flow Cytometer The instrument used was a Navios (Beckman Coulter, Miami, FL) and was maintained and quality controlled per the manufacturer’s instructions. Daily controls were run, including Flow Check Pro FluorospheresTM and Flow Set Pro FluorospheresTM fluorescent microspheres (Beckman Coulter) to monitor the instrument performance. As the fluorochromes differ between the control tube and the dried format tube (DuracloneTM, Beckman Coulter), a different compensation matrix was used. Compensation Target channels were established in our laboratory to achieve optimal signal for both dim and bright antigens and were the same for both tube types. Compensation was performed using VersaCompTM (Beckman Coulter) beads according to manufacturer’s instructions. The dried reagent kit came with single-color tubes for determining the compensation matrix; however, the liquid format reagents required the pipetting of each antibody into separate flow tubes. VersaComp beads were vortexed and added to the tubes that contained either the liquid or dried format color antibodies. Lyophilized human lymphocytes were used as a positive for the viability dye instead of the positive beads, and a drop of negative bead used for the negative peak. After 15 min incubation in the dark, each tube was washed once with 0.3% bovine serum albumin in phosphate-buffered saline. After centrifugation, the supernatant was decanted and samples resuspended in 1 mL of 0.3% bovine serum albumin in phosphate-buffered saline. Tubes were run with the voltages used to achieve the target values previously determined. The use of both positive and negative beads allowed for compensation to be calculated using the “positive and negative” option in the analysis software package KaluzaTM (Beckman Coulter). Verification of the compensation matrix was performed by analyzing a normal bone marrow sample stained with either liquid or dried format reagents. If required, any small adjustment was made to the matrix; this was then saved as the final compensation matrix. Measurement of Performance Performance was measured first by determining the overall ability to interpret each specimen and report the same features in the new dried format when compared with the current format. Secondly and thirdly, to show
Cytometry Part B: Clinical Cytometry
NOVEL LYMPHOCYTE SCREENING TUBE USING DRIED MONOCLONAL ANTIBODY REAGENTS
at least equivalence between the dried and liquid formats when either enumerating percentages of major populations or overall staining intensity. Antigen and Fluorochrome Selection This custom screening tube contains 11 antibodies using 9 out of the 10 detectors on the Navios flow cytometer (Supporting Information Table S1). Multiplexing kappa and lambda with CD8 and CD4 on fluorescein isothiocyanate (FITC) and PE, respectively, permits gating of mutually exclusive populations, CD3 gating for CD8 and CD4; and CD19 for kappa and lambda (Fig. 1). CD3 and CD5 antigens were moved from liquid to dry format from e450 to PE-Cy7 and PE-Cy7 to Pacific Blue, respectively, to allow for all antibodies to be supplied by one manufacturer. The addition of a near infra-red live/dead (NIR) viability dye (Molecular Probes, Eugene, OR) in substitution for an antibody such as APC-H7 (which emits at a wavelength >750 nm with excitation from the 635 nm laser) allows for easy selection of only viable cells and is particularly useful in tissue samples, as nonviable cells have a tendency to nonspecifically bind antibodies. The NIR dye is added to the tube at the same time as the patient sample to either the liquid or dried format. Addition of the NIR dye at this point reduces the time required for sample staining over dyes that are added after incubation with antibodies and red cell lysis. For suspected Tcell malignancies, absence or decreased expression of CD3, altered or co-expression of CD4 and CD8, or altered expression of CD5 will lead to a follow-up tube(s) to assess other mature and immature T-cell markers. Additionally, morphological examination is performed on all samples at time of analysis. Dried Format Reagent Using Beckman Coulters’ custom design service and its proprietary dry coating technology, our custom tube was made using all Beckman Coulter reagents (see Supporting Information Table S1). The sampler kit with our custom tube contained five light-impermeable pouches containing 10 tubes each and had a shelf life of 12 months from the date of manufacturer and a 1-month open pouch stability. Sample Preparation As the lymphoid screen tube contains antibodies for the detection of surface kappa and lambda light chains, sample washing prior to staining is required. Two aliquots of a normalized sample, of approximately 10 3 109/L were washed twice using 0.3% bovine serum albumin in phosphate-buffered saline, followed by a 10-min block at 378C using 5% bovine serum albumin to minimize nonspecific binding when staining for immunoglobulins. A third wash was performed after the blocking step, and one aliquot stained with a pre-made liquid cocktail or the other sample added to the dried format (Duraclone) tube. To each tube, the viability dye was added, and both tubes incubated for 20 min at room
Cytometry Part B: Clinical Cytometry
363
temperature protected from light. Lysis was carried out using ammonium chloride with 0.25% formaldehyde for 10 min at room temperature, with the tubes protected from light. Subsequent to red cell lysis, samples are spun, and one more wash was performed, and each tube resuspended in 0.5 mL of Dulbecco’s phosphatebuffered saline. All samples were run on a Navios 3 laser 10 color flow cytometer (Beckman Coulter). Quality Control Testing Evaluation of each antibody and validation of each batch are critical processes in determining whether the batch is acceptable for use with patient samples. To perform titration studies for a specific antigen, a sample must have an adequate amount of the target cells (typically > 20%), which can become challenging. Quality control (QC) tests that are performed in our laboratory for new lot numbers of already titrated antibodies are to stain a sample (e.g., peripheral blood) with the currently used antibody (e.g., CD8 FITC) and then the same sample with the new lot. An acceptable difference is allowed between the current lot and new lot for the QC to pass. For rarely expressed antigens, positive beads (available from many manufacturers) may be used to parallel test new lots of antibody. If this QC tests fail, then the process of titration of the antibody must start again. Sample Analysis and Comparison Listmode data for forward and side light scatter, all fluorescence parameters, and time was collected on the flow cytometer with a minimum of 100,000 viable, CD451 events. Listmode data files were analyzed using Kaluza analysis software (Beckman Coulter). A common analysis template was utilized to visualize all major cell populations in both the liquid and dried format tube (Figs. 1A and 1B). Normal T-cells can be characterized by CD3, CD4, CD8, and CD5 expression. Normal B-cell expression of CD19, CD20, and CD10 antigens, coupled with CD45 and kappa or lambda light chains allow for detection of both mature and immature B-cells. NK cells are characterized by negativity for CD3 and CD5 and positivity for CD56. By multiplexing several markers on the same fluorochrome, the maximum amount of information can be realized from a single tube (2,4). In cases other than lymphoma screening/staging, the screen tube is accompanied by a limited precursor myelo/mono panel, which includes CD45, CD64, CD13, CD16, CD34, and CD38. Subsequent testing to further characterize acute leukemia, myelodysplastic syndrome, multiple myeloma, or suspected T-cell neoplasms is performed as required. The statistics export function in KaluzaTM was used to export median channel numbers for populations of interest between tubes to quantitate expression levels of certain key markers. It should be noted that cross fluorochrome fluorescent intensity comparison for CD3 and CD5 was not performed, as these antigens moved from
FIG. 1. First row: Normal bone marrow sample stained with (A) liquid format cocktail or (B) dried format reagents. The lysed sample is first gated on “time” to allow exclusion of any data bursts or air bubbles in the acquisition. A single parameter histogram of the viability dye is then displayed, gated on time, which allows selection of viable only events (negative for near-IR viability stain). From there, a dotplot of FS peak versus FS area (FS INT) is displayed to allow identification and exclusion of doublets. A CD45 versus SS dotplot allows visualization of multiple populations. A SS versus CD19 dotplot gated on singlets only displays all CD191 B-cells. From there, the surface immunoglobulins kappa and lambda are shown for the entire population. CD31 cells are gated similarly. A SS versus CD3 dotplot displays all CD31 T-cells; subsequently, a CD8 versus CD4 dotplot shows the relative proportions of subsets. A lymphoid region (gated on a CD45 versus SS dotplot) is used as the gate for a dotplot of CD19 versus CD3, showing any possible dual positive CD31CD191. Dotplots of CD19 versus either CD5 or CD10 show populations of B-cells that express CD5 or CD10. In this figure, kappa and lambda light chains are only shown gated on CD191CD101, but for all samples CD191CD51 was evaluated. The CD3 versus CD56 dotplot also gated on the lymphoid region is used in the evaluation of NK cells (CD32 CD561) and NK T-cells (CD31 CD561). B-cell maturation is shown on the dotplot of CD20 versus CD10 gated on CD191 cells; early B-cells have the brightest CD10 and no CD20. When maturing, B-cells initially lose some CD10 and gain CD20. The brightest CD20 cells then start to lose the CD10 until they become mature B-cells that lack CD10. The final two dotplots show singlet gated plots of CD20 and CD10 versus SS, showing the intensity of CD20 and CD10 on all the populations that have been previously gated.
365
NOVEL LYMPHOCYTE SCREENING TUBE USING DRIED MONOCLONAL ANTIBODY REAGENTS
Table 1 Patient Diagnoses and Specimen Types for All 42 Samples Used for Evaluation Diagnosis
Peripheral blood
Bone marrow
Tissue
Total
4
–
1
5
– 1 3 1 – –
5 3 – 1 – 2
– – – 1 3 –
5 4 3 3 3 2
– – 1 – 2 2 1 – – – – 15
2 – 1 1 – – – 1 – 1 – 17
– 2 – 1 – – – – 1 – 1 10
2 2 2 2 2 2 1 1 1 1 1 42
Chronic lymphocytic leukemia/small lymphocytic lymphoma Multiple myeloma B-cell acute lymphoblastic leukemia Hairy cell leukemia Normal Diffuse large B-cell lymphoma Myelodysplasia with refractory anemia with ring sideroblasts Acute myeloid leukemia Follicular lymphoma T-cell acute lymphoblastic leukemia Idiopathic thrombocytopenic purpura Monocytosis Monoclonal T-cell population Monoclonal B-cell population Lymphoplasmacytic lymphoma Marginal cell lymphoma Refractory anemia with unilineage dysplasia Peripheral T-cell lymphoma Total
Table 2 Summary of the Samples Used in the Comparison Study Normal
Red cell disorder
B-cell malignancy
T-cell malignancy
Myeloid malignancy
Other
Total
1 2 2 5
0 3 0 3
9 4 7 20
3 1 1 5
2 2 0 4
0 5 0 5
15 17 10 42
Peripheral blood Bone marrow Tissue Total
Normal samples were not found to have any malignancy after chart review.
liquid to dry format from e450 to PE-Cy7 and PE-Cy7 to Pacific Blue, respectively. RESULTS In total, 42 samples were split and stained with liquid and dried reagents (Table 1). Antigens were selected to identify most lymphoid malignancies. In addition, some abnormal nonlymphoid samples were run, and while a diagnosis could not be made from this screening tube, the abnormal population was readily identified to allow follow-up testing (Supporting Information Fig. 1). SamTable 3 Results of Cell Population Analysis Between the Liquid and Dried Formats Test tube (Duraclone) Positive Negative Total
Comparison tube (liquid cocktail) Positive Negative Total 32 0 32
0 10 10
32 10 42
Negative samples were confirmed samples to have absence of malignancy at the time of analysis. Positive samples were confirmed samples to have malignancy at the time of analysis. Morphological, pathological, and clinical information were also used to determine the presence or absence of malignancy.
Cytometry Part B: Clinical Cytometry
ples were divided into broad categories of malignancy; normal, B-cell, T-cell, myeloid, and “other” (e.g., multiple myeloma) sections. Sample types included peripheral blood, bone marrow, and tissues specimens (which include soft tissues, cerebral spinal fluids, nodes, and other fluids) (Table 2). Comparison of the liquid to dried reagents demonstrated equivalency of interpretation between the formats for all 42 samples (Table 3). A number of normal samples were run, Figure 1 shows an example of some of the analysis performed in KaluzaTM for a normal bone marrow sample. Time is used to gate out any variations in flow rate; subsequently, the viability dye is used to exclude any cells that may have been damaged during transportation or storage since collection. A dotplot of Forward Scatter (FS) - Peak versus FS-area is used to remove doublets. A region named “singlets only” is created on this histogram. A dotplot of CD45 versus SSC is gated on the singlets only region. On this dotplot, a lymph gate is drawn and used in future dotplots. A dotplot of CD19 versus SSC is gated on the above singlets gate, and a region created that allows for surface immunoglobulin staining to be investigated on CD191 cells. Similarly CD4 and CD8 are viewed from a region drawn on a dotplot of CD3 versus SSC. In Figure 1, a selection of the total number of dotplot are shown, The colors for each population
366
HEDLEY ET AL.
FIG. 2. Examples of positive samples for dried or liquid format tubes. Dotplots of dried reagent tubes are shown in panels A, C, and E. Comparison dotplots of liquid format tubes are shown in panels B, D, and F. Monoclonal populations shown in samples from: (A, B) Peripheral blood, two malignant populations: CD191CD102CD201Kappa1 variable CD5 expression (red) and a CD191CD51Lambda1 (blue), and polyclonal B-cells (black). (C, D) Bone marrow, CD191CD52CD102Kappa1. (E, F) Lymph node tissue, CD191CD52CD10 partial, CD201, Lambda1.
are: CD191 red, CD31 blue, CD191CD101 orange, CD32CD561 dark green, CD32CD562 light green, granulocytes purple, and monocytes beige.
Figure 2 shows three examples of positive samples for both liquid and dried formats for a number of different positive samples. As this is a lymphocyte screening tool,
Cytometry Part B: Clinical Cytometry
NOVEL LYMPHOCYTE SCREENING TUBE USING DRIED MONOCLONAL ANTIBODY REAGENTS
367
FIG. 3. Comparison of gating percentages from liquid and dried format reagents for different antigens. For all plots, x-axis is percentage of cells with liquid format and y-axis is percentage gated with dried format reagents. A. CD31 cells gated from a dotplot of CD3 versus SS gated on singlets. B. CD191 cells gated from a dotplot of CD19 versus SS gated on singlets. C. CD191CD51 cells gated from a dotplot of CD19 versus CD5 gated on all lymphocytes. D. CD191CD101 cells gated from a dotplot of CD19 versus SS gated on all lymphocytes. E. All kappa and lambda cells gated from a dotplot of kappa versus lambda gated on all CD191 cells. F. CD201 cells gated from a dotplot of CD20 versus SS gated on all lymphocytes.
representative samples for peripheral blood (A), bone marrow (B), and lymph node tissue (C) are shown. There is excellent agreement between the dried and liquid reagents with all the antigens. Comparison of CD3 and CD5 staining showed dimmer staining when comparing Pacific Blue to PE-Cy7 fluorochromes and e450 to PE-Cy7; however, this did not alter sample interpretation between the dried and liquid formats. Figure 2A shows
Cytometry Part B: Clinical Cytometry
a population of B-cells, gated on CD19 that seem to be polyclonal with slight excess of lambda light chains (top row Dotplot 2). When CD19 versus CD5 is viewed, it can be seen that there are three separate expressions of CD52CD191CD52, CD191CD51, and CD191CD5 variable (top row Dotplot 3). The B-cells can also be separated by their expression of CD20 (dim to bright; bottom row Dotplot 3) When each population is
368
HEDLEY ET AL.
FIG. 4. Comparison of gating percentages from liquid and dried format reagents for different antigens. For all plots, x-axis is percentage of cells with liquid format and y-axis is percentage gated with dried format reagents. A. CD41 cells gated from a dotplot of CD8 versus CD4 gated on CD31 cells. B. CD81 cells gated from a dotplot of CD8 versus CD4 gated on CD31. C. CD41CD81 cells gated from a dotplot of CD8 versus CD4 gated on CD31 cells. D. CD32CD561 cells (NK cells) gated from a dotplot of CD3 versus CD56 gated on all lymphocytes.
analyzed separately, it can be seen that there are three distinct light chain patterns: (1) CD191CD5 variableexpressing kappa light chains (red), (2) CD191CD51 expressing lambda light chains (blue), and (3) a normal population of polyclonal B-cells (black). This sample contains two monoclonal populations with differing light chain expression as well as some normal polyclonal B-cells. Statistics for the gating percentages and staining intensity (median channel number) were exported using Kaluza software and graphed in Excel for some key antigens. Figure 3 shows the gating percentages for: CD31, CD191, CD191CD51, CD191CD101, kappa and lambda, and CD201. The hashed lines in each graph are the 2SD boundaries based on the population data within that graph. Figure 4 shows data similar to that in Figure 3 for CD41, CD81, CD41CD81, and CD32CD561. Both Figures 3 and 4 show excellent agreement between the liquid format (x-axes) and the dried format (y-axes). For all panels in both Figures 3 and 4, [mt]95% of all the samples lie within the 2SD boundaries. Figure 5 shows median channel for a selected number of antigens (CD3 and CD5 were not considered as the fluorochrome changed between formats). It should be noted that analysis of the expression of surface immunoglobu-
lins is extremely sensitive to washing, and we hypothesize that the outliers (one higher and one lower for the dried reagent) for both kappa and lambda light chains seen in Figures 5B and 5D are due to differences in sample washing. Figures 5A and 5F (CD8 FITC and CD32CD56 PE-Cy5.5, respectively) showed similar signals when comparing dried with liquid format reagents; however, on average, the other antigens analyzed in panels B, C, D, E, G, H (Kappa FITC, CD4PE, Lambda PE, CD19 ECD, CD20 APC, and CD101 APC Alexa700) all showed brighter signal in the dried format. Equivalence between formats was seen with all antigens, with the exception of CD4, which gave a brighter signal for all samples in the dried format over the liquid format. DISCUSSION Over 50% of samples suspected of a hematological malignancy received in our flow cytometry laboratory are from patients with mature lymphoid malignancies. Additionally, lymph node fine-needle aspirates, small tissue samples, and body fluids with low cell counts are sent for evaluation by flow cytometry. These samples often yield only enough cells for single tube analysis. The inclusion of a viability dye has proven extremely useful, particularly in tissue samples where nonspecific
Cytometry Part B: Clinical Cytometry
NOVEL LYMPHOCYTE SCREENING TUBE USING DRIED MONOCLONAL ANTIBODY REAGENTS
369
FIG. 5. Comparisons of different antigen intensity from both liquid and dried format reagents. For all plots, x-axis is percentage of cells with liquid format and y-axis is percentage gated with dried Duraclone reagents. A. CD8 FITC1 cells gated from a dotplot of CD8 versus CD4 gated on CD31 cells. B. Kappa FITC1 cells gated from a dotplot of CD19 versus SS gated on singlets. C. CD4 PE1 cells gated from a dotplot of CD8 versus CD4 gated on CD31 cells. D. Lambda PE1 cells gated from a dotplot of CD19 versus SS gated on singlets. E. CD19 ECD1 cells gated from a dotplot of CD19 versus SS gated on singlets. F. CD32CD56 PE2Cy5.51 cells (NK cells) gated from a dotplot of CD3 versus CD56 gated on all lymphocytes. G. CD20 APC1 cells gated from a dotplot of CD20 versus SS gated on singlets. H. CD101 APC Alexa 700 gated from a dotplot of CD19 versus CD10 gated on all lymphocytes.
staining with tandem dyes can be a significant issue (Supporting Information Fig. 2). Morphology is always available to the laboratory, and clinical findings are usually, but not always, provided, aiding in interpretation of samples. Pattern recognition with the use of a standardized analysis template allows for quick learning by laboratory staff and clinicians and allowed for comparative
Cytometry Part B: Clinical Cytometry
analysis between formats in this study. The goal of this study was to determine whether substitution of liquid with dried format reagent would alter sample interpretation, percentage of subset gated, and the signal intensity. Kaluza software was used with our customized standardized templates and allowed for comparative analysis of both data from liquid and dried formats for the goals
370
HEDLEY ET AL.
of this study. The exported numerical data from these templates showed that there was excellent agreement between formats as measured by sample interpretation, equivalence of enumerated percentage of major populations, or overall staining intensity. Therefore, we believe we have demonstrated that dry coated reagents do not hinder our ability to detect multiple neoplasms with our initial screening tube. We have previously reported that a standardized Kaluza report template has streamlined result interpretation and has already improved efficiency in our laboratory (15). The substitution of dried format tubes over in house prepared liquid combinations will allow for the significant reductions in sample handling, instrument setup and analysis time can now be coupled with substantial reduction in the amount of quality assurance for batches of liquid reagents. The substitution of dry coated (DuraCloneTM) reagents for liquid cocktails will reduce; the time spent making reagent cocktails, time validating reagent, wastage of left over antibodies, and the potential for errors when making expensive cocktails. CONCLUSION As early as 2006, Wood et al. (6) have shown the feasibility of routine ten color flow cytometry in the diagnosis of leukemia and lymphoma in the clinical laboratory. In our laboratory, using a single screening tube for lymphoid malignancies has allowed standardized evaluation of a large percentage of samples received for flow cytometry. Multiplexing with a viability dye has increased the usability of this approach to a wide range of samples, including those with low cell counts and/or poor viability (often seen with tissue samples). Other groups have reported adding even more antibodies to take advantage of multiplexing (4), and it should be possible to have a comprehensive screening panel that assesses lymphoid, myelomonocytic cells and viability in a single tube. However, increased complexity in single tube multicolor analysis is accompanied by challenges in reagent cocktail preparation, a time consuming, error-prone process in busy clinical laboratories. In this study, we have demonstrated equivalency between a novel dried reagent combination and liquid format tubes in sample interpretation, staining percentages, and intensity in a variety of sample types with diagnoses including normal, reactive lymphocytosis, chronic lymphocytic leukemia, and lymphoma. Dried, room-temperature tubes offer advantages over liquid format, as they are stable for at least 12 months and, unlike liquid reagents, are made with a single lot for each antigen. In our laboratory, using dried format tubes effectively eliminates the need to individually titrate, pipette, create liquid cocktails, and documentation. In the previous year, [more than ]50 batches of our lymphocyte screening tube were prepared, validated, and documented. Removal of these duties allows significant time savings in our laboratory while eliminating multiple sources of error. The relative instability of APC tandem dyes limits the outdate for liquid
format cocktails containing them, which is not the case with the dried reagent. Finally, we are confident that the use of dried format tubes in our laboratory will translate to improved efficiencies and reduce the risk of errors in preparation of complex liquid reagent cocktails.
ACKNOWLEDGMENTS The authors thank the following people in the laboratory: Samantha Vandevyvere, Dominka Benjamins, and Erin Bourguignon. LITERATURE CITED 1. Craig FE, Foon KA. Flow cytometric immunophenotyping for hematologic neoplasms. Blood 2008; 111:3941–3967. 2. Preijers FW, Huys E, Moshaver B. OMIP-010: A new 10-color monoclonal antibody panel for polychromatic immunophenotyping of small hematopoietic cell samples. Cytometry Part A 2012; 81A:453–455. 3. Mittag A, Lenz D, Gerstner AO, Sack U, Steinbrecher M, Koksch M, Raffael A, Bocsi J, Tarnok A. Polychromatic (eight-color) slide-based cytometry for the phenotyping of leukocyte, NK, and NKT subsets. Cytometry Part A 2005; 65A:103–115. 4. Bradford JA, Buller G, Suter M, Ignatius M, Beechem JM. Fluorescence-intensity multiplexing: Simultaneous seven-marker, two-color immunophenotyping using flow cytometry. Cytometry Part A 2004; 61A:142–152. 5. Preijers FW, Huys E, Favre C, Moshaver B. Establishment of harmonization in immunophenotyping: A comparative study of a standardized one-tube lymphocyte-screening panel. Cytometry Part B 2014; 86B:418–425. 6. Wood BL, Arroz M, Barnett D, DiGiuseppe J, Greig B, Kussick SJ, Oldaker T, Shenkin M, Stone E, Wallace P. 2006 Bethesda international consensus recommendations on the immunophenotypic analysis of hematolymphoid neoplasia by flow cytometry: Optimal reagents and reporting for the flow cytometric diagnosis of hematopoietic neoplasia. Cytometry Part B 2007; 72B: S14–S22. 7. Preijers FW, Huys E, Leenders M, Nieto L, Gautherot E, Moshaver B. The new violet laser dye, krome orange, allows an optimal polychromatic immunophenotyping based on CD45-KO gating. J Immunol Methods 2011; 372:42–51. 8. Kalina T, Flores-Montero J, van der Velden VH, Martin-Ayuso M, Bottcher S, Ritgen M, Almeida J, Lhermitte L, Asnafi V, Mendonca A, et al. EuroFlow standardization of flow cytometer instrument settings and immunophenotyping protocols. Leukemia 2012; 26:1986–2010. 9. van Dongen JJ, Lhermitte L, Bottcher S, Almeida J, van der Velden VH, Flores-Montero J, Rawstron A, Asnafi V, Lecrevisse Q, Lucio P, et al. EuroFlow antibody panels for standardized n-dimensional flow cytometric immunophenotyping of normal, reactive and malignant leukocytes. Leukemia 2012; 26:1908–1975. 10. Greig B, Oldaker T, Warzynski M, Wood B. 2006 Bethesda international consensus recommendations on the immunophenotypic analysis of hematolymphoid neoplasia by flow cytometry: Recommendations for training and education to perform clinical flow cytometry. Cytometry Part B 2007; 72B: S23–S33. 11. Biancotto A, Fuchs JC, Williams A, Dagur PK, McCoy JP Jr. High dimensional flow cytometry for comprehensive leukocyte immunophenotyping (CLIP) in translational research. J Immunol Methods 2011; 363:245–261. 12. Johansson U, Macey M. Tandem dyes: Stability in cocktails and compensation considerations. Cytometry Part B 2014; 86B:164–174. 13. Le Roy C, Varin-Blank N, Ajchenbaum-Cymbalista F, Letestu R. Flow cytometry APC-tandem dyes are degraded through a cell-dependent mechanism. Cytometry Part A 2009; 75A:882–890. 14. Rawstron AC, Bottcher S, Letestu R, Villamor N, Fazi C, Kartsios H, de Tute RM, Shingles J, Ritgen M, Moreno C, et al. European Research Initiative in CLL. Improving efficiency and sensitivity: European Research Initiative in CLL (ERIC) update on the international harmonised approach for flow cytometric residual disease monitoring in CLL. Leukemia 2013; 27:142–149. 15. Keeney M, Benjamins D. A 10 color 11 antibody and viability lymphocyte screening tube. ICCS e-Newsletter 2014;5. Available at: http://cytometry.org/public/newsletters/eICCS-5-1/index.php
Cytometry Part B: Clinical Cytometry
Original Article
Novel Lymphocyte Screening Tube Using Dried Monoclonal Antibody Reagents Benjamin D. Hedley,1* Michael Keeney,1 Janice Popma,1 and Ian Chin-Yee2 1
Pathology and Laboratory Medicine, Division of Hematology, London Health Sciences Centre, London, Ontario N6K 5W9, Canada 2 Department of Medicine, Division of Hematology, London Health Sciences Centre, Schulich School of Medicine Western University of Ontario, London, Ontario N6K 5W9, Canada
We previously developed a 10-color 11-antibody combination including a viability dye, to screen T-, B-, and natural killer (NK)-cell populations in blood, bone marrow, tissue, and body fluids. Recently, Beckman Coulter has introduced a line of dried reagents that, unlike liquid reagents and cocktails, require no refrigeration, titration, or manipulation before using. We evaluated custom tubes based on our standard lymphocyte screening panel, focusing on comparative analysis, ease of use, and advantages compared with our liquid reagent set. We tested 42 samples from blood (n 5 15), bone marrow (n 5 17), and tissue (n 5 10) with the combination CD4/CD8/KAPPA/LAMBDA/CD19/CD56/CD5/CD20/CD10/CD3/CD45 and a vital dye by both methods and compared positivity and staining intensity for each antigen. Of the 42 samples, 5 were normal samples, 3 were red cell disorders, 20 were B-cell malignancies, 5 T-cell malignancies, 4 myeloid malignancies, and the remaining 5 were other diagnoses. Dried reagents gave equivalent staining intensity results to our standard panel in a variety of sample types, with diagnoses including reactive lymphocytosis, chronic lymphocytic leukemia, and various lymphomas. Our standard panel for evaluation of mature lymphoid malignancies allows rapid assessment of any sample type while providing direct assessment of viability. The dried reagent tube reduces preanalytical work, with simple addition of sample and the viability dye to the tube, saving time, reducing potential errors, and obviating need to titrate and monitor individual antibodies. With a shelf life of at least 12 months, the reagents also offer potential savings in reagent costs by reducing wastage due to expiration C 2015 International Clinical Cytometry Society or tandem breakdown in standard liquid formulation. V Key terms: duraclone; flow cytometry; screening tube; acute leukemia; immunophenotyping
How to cite this article: Hedley BD, Keeney M, Popma J and Chin-Yee I. Novel Lymphocyte Screening Tube Using Dried Monoclonal Antibody Reagents. Cytometry Part B 2015; 88B: 361–370.
In recent years, the clinical flow cytometry laboratory has seen the introduction of highly sophisticated instrumentation that can detect up to 10 fluorochromes on FDA-approved cytometers (1). Major advantages of these instruments include the ability to analyze small tissue samples or body fluids, less redundancy in reagent use, and improved efficiency for the laboratory (2–7). Although the increase in the tube complexity has led to fast and, in some cases, more accurate analysis of patient samples, this has given rise to different challenges that have to be addressed in a high-volume clinical laboratory. Running a large standardized panel allows for greater reproducibility and efficiency but could potentially increase cost. Specifically, using a standardized panel developed by the Euroflow group (8,9), in our terti-
C 2015 International Clinical Cytometry Society V
ary care laboratory, with access to morphology, pathology, molecular, and cytogenetic results was not deemed cost-effective. Additional Supporting Information may be found in the online version of this article. This study was not funded by Beckman Coulter; however, MK is a consultant for Beckman Coulter. All authors contributed equally to the study design, analysis of samples, and manuscript review. *Correspondence to: B. D. Hedley, Pathology and Laboratory Medicine, Division of Special Hematology, London Health Sciences Centre, 800 Commissioners Road East, London, Ontario N6K 5W9, Canada. E-mail: [email protected] Received 19 December 2014; Revised 9 April 2015; Accepted 27 April 2015 Published online 17 July 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/cyto.b.21251
362
HEDLEY ET AL.
As an alternative approach, we developed a 10-color 11-antibody panel, which also includes a viability dye, to screen T-, B-, and natural killer (NK)-cell populations in a variety of blood, bone marrow, tissue, and body fluids. Morphology remains an essential part of the flow cytometry workup, further reducing the risk of missing other hematological neoplasms. Cases are not reported without morphological review. Reporting of results follows Bethesda guidelines, and although percentages of the abnormal population as a subset of the total nucleated cells or CD451 cells is given, percentage positive or negative for each cell marker is not reported (6,10). The screening panel used in our laboratory is a 10-color 11-antibody tube that also includes a viability dye. It is routinely used to screen blood, bone marrow, tissue, and body fluids for T-, B-, and NK-cell populations. This tube contains CD4/CD8/KAPPA/LAMBDA/ CD19/CD56/CD5/CD20/CD10/CD3/CD45 (Supporting Information Table S1) and has been made routinely into a cocktail in liquid format, from various suppliers, and prepared by technologists in large batches (150 tests). Prior to creation of a cocktail, each individual antibody must be titrated, validated, and then combined with each of the other antibodies. Once combined, individual antibodies titration volumes may have to be further adjusted prior to evaluation against the current reagents. Cocktail preparation is a complex time-consuming process that requires the technologist to individually pipette each of the antibodies into the batch. Although infrequent, errors in preparation can occur when pipetting large numbers of antibodies, resulting in wasted antibody incurring expense and potential misinterpretation of patient samples. Quality assurance of reagent cocktails requires rigorous documentation of each step and regular testing to assess the quality of each batch of cocktail (http://www. cap.org/ pages 9 and 10 of Flow Cytometry Checklist). This takes into account: variations from lot to lot in the intensity of the number of tests to be created, the number of tests in each vial, and the lot(s) of each individual antibody. It takes a significant amount of time performing and recording all of these steps in cocktail preparation, which could be spent analyzing and interpreting patient samples. Another long-standing issue that laboratories have had to be aware of is the instability of tandem dyes. Lot to lot variation has been reported requiring recalculation of compensation matrices when working with tandem dyes (11–14). The current generation of 10-color flow cytometers allows for a greater number of tandem dyes to be used. Some newer tandem dyes are more prone to breakdown than others, effectively reducing the shelflife of liquid format cocktails containing these tandem dyes. Lately, Beckman Coulter introduced a range of dried reagents that are stable at room temperature with much longer shelf life (including tandem dyes for phycoerythrin (PE) and allophycocyanin (APC)). In addition, they offer this product along with their custom design service, allowing flow cytometry laboratories to adapt
their own laboratory derived tests. Also supplied are single color dry format tubes for compensation, and a single batch may be purchased for 12 months requiring only once to be validated, eliminating the need to titrate, pipette, validate, and document individual batches of liquid format cocktail. The goal of this study was to demonstrate at least equivalence in performance of the dried format reagent when compared with the currently used liquid format. MATERIALS AND METHODS Flow Cytometer The instrument used was a Navios (Beckman Coulter, Miami, FL) and was maintained and quality controlled per the manufacturer’s instructions. Daily controls were run, including Flow Check Pro FluorospheresTM and Flow Set Pro FluorospheresTM fluorescent microspheres (Beckman Coulter) to monitor the instrument performance. As the fluorochromes differ between the control tube and the dried format tube (DuracloneTM, Beckman Coulter), a different compensation matrix was used. Compensation Target channels were established in our laboratory to achieve optimal signal for both dim and bright antigens and were the same for both tube types. Compensation was performed using VersaCompTM (Beckman Coulter) beads according to manufacturer’s instructions. The dried reagent kit came with single-color tubes for determining the compensation matrix; however, the liquid format reagents required the pipetting of each antibody into separate flow tubes. VersaComp beads were vortexed and added to the tubes that contained either the liquid or dried format color antibodies. Lyophilized human lymphocytes were used as a positive for the viability dye instead of the positive beads, and a drop of negative bead used for the negative peak. After 15 min incubation in the dark, each tube was washed once with 0.3% bovine serum albumin in phosphate-buffered saline. After centrifugation, the supernatant was decanted and samples resuspended in 1 mL of 0.3% bovine serum albumin in phosphate-buffered saline. Tubes were run with the voltages used to achieve the target values previously determined. The use of both positive and negative beads allowed for compensation to be calculated using the “positive and negative” option in the analysis software package KaluzaTM (Beckman Coulter). Verification of the compensation matrix was performed by analyzing a normal bone marrow sample stained with either liquid or dried format reagents. If required, any small adjustment was made to the matrix; this was then saved as the final compensation matrix. Measurement of Performance Performance was measured first by determining the overall ability to interpret each specimen and report the same features in the new dried format when compared with the current format. Secondly and thirdly, to show
Cytometry Part B: Clinical Cytometry
NOVEL LYMPHOCYTE SCREENING TUBE USING DRIED MONOCLONAL ANTIBODY REAGENTS
at least equivalence between the dried and liquid formats when either enumerating percentages of major populations or overall staining intensity. Antigen and Fluorochrome Selection This custom screening tube contains 11 antibodies using 9 out of the 10 detectors on the Navios flow cytometer (Supporting Information Table S1). Multiplexing kappa and lambda with CD8 and CD4 on fluorescein isothiocyanate (FITC) and PE, respectively, permits gating of mutually exclusive populations, CD3 gating for CD8 and CD4; and CD19 for kappa and lambda (Fig. 1). CD3 and CD5 antigens were moved from liquid to dry format from e450 to PE-Cy7 and PE-Cy7 to Pacific Blue, respectively, to allow for all antibodies to be supplied by one manufacturer. The addition of a near infra-red live/dead (NIR) viability dye (Molecular Probes, Eugene, OR) in substitution for an antibody such as APC-H7 (which emits at a wavelength >750 nm with excitation from the 635 nm laser) allows for easy selection of only viable cells and is particularly useful in tissue samples, as nonviable cells have a tendency to nonspecifically bind antibodies. The NIR dye is added to the tube at the same time as the patient sample to either the liquid or dried format. Addition of the NIR dye at this point reduces the time required for sample staining over dyes that are added after incubation with antibodies and red cell lysis. For suspected Tcell malignancies, absence or decreased expression of CD3, altered or co-expression of CD4 and CD8, or altered expression of CD5 will lead to a follow-up tube(s) to assess other mature and immature T-cell markers. Additionally, morphological examination is performed on all samples at time of analysis. Dried Format Reagent Using Beckman Coulters’ custom design service and its proprietary dry coating technology, our custom tube was made using all Beckman Coulter reagents (see Supporting Information Table S1). The sampler kit with our custom tube contained five light-impermeable pouches containing 10 tubes each and had a shelf life of 12 months from the date of manufacturer and a 1-month open pouch stability. Sample Preparation As the lymphoid screen tube contains antibodies for the detection of surface kappa and lambda light chains, sample washing prior to staining is required. Two aliquots of a normalized sample, of approximately 10 3 109/L were washed twice using 0.3% bovine serum albumin in phosphate-buffered saline, followed by a 10-min block at 378C using 5% bovine serum albumin to minimize nonspecific binding when staining for immunoglobulins. A third wash was performed after the blocking step, and one aliquot stained with a pre-made liquid cocktail or the other sample added to the dried format (Duraclone) tube. To each tube, the viability dye was added, and both tubes incubated for 20 min at room
Cytometry Part B: Clinical Cytometry
363
temperature protected from light. Lysis was carried out using ammonium chloride with 0.25% formaldehyde for 10 min at room temperature, with the tubes protected from light. Subsequent to red cell lysis, samples are spun, and one more wash was performed, and each tube resuspended in 0.5 mL of Dulbecco’s phosphatebuffered saline. All samples were run on a Navios 3 laser 10 color flow cytometer (Beckman Coulter). Quality Control Testing Evaluation of each antibody and validation of each batch are critical processes in determining whether the batch is acceptable for use with patient samples. To perform titration studies for a specific antigen, a sample must have an adequate amount of the target cells (typically > 20%), which can become challenging. Quality control (QC) tests that are performed in our laboratory for new lot numbers of already titrated antibodies are to stain a sample (e.g., peripheral blood) with the currently used antibody (e.g., CD8 FITC) and then the same sample with the new lot. An acceptable difference is allowed between the current lot and new lot for the QC to pass. For rarely expressed antigens, positive beads (available from many manufacturers) may be used to parallel test new lots of antibody. If this QC tests fail, then the process of titration of the antibody must start again. Sample Analysis and Comparison Listmode data for forward and side light scatter, all fluorescence parameters, and time was collected on the flow cytometer with a minimum of 100,000 viable, CD451 events. Listmode data files were analyzed using Kaluza analysis software (Beckman Coulter). A common analysis template was utilized to visualize all major cell populations in both the liquid and dried format tube (Figs. 1A and 1B). Normal T-cells can be characterized by CD3, CD4, CD8, and CD5 expression. Normal B-cell expression of CD19, CD20, and CD10 antigens, coupled with CD45 and kappa or lambda light chains allow for detection of both mature and immature B-cells. NK cells are characterized by negativity for CD3 and CD5 and positivity for CD56. By multiplexing several markers on the same fluorochrome, the maximum amount of information can be realized from a single tube (2,4). In cases other than lymphoma screening/staging, the screen tube is accompanied by a limited precursor myelo/mono panel, which includes CD45, CD64, CD13, CD16, CD34, and CD38. Subsequent testing to further characterize acute leukemia, myelodysplastic syndrome, multiple myeloma, or suspected T-cell neoplasms is performed as required. The statistics export function in KaluzaTM was used to export median channel numbers for populations of interest between tubes to quantitate expression levels of certain key markers. It should be noted that cross fluorochrome fluorescent intensity comparison for CD3 and CD5 was not performed, as these antigens moved from
FIG. 1. First row: Normal bone marrow sample stained with (A) liquid format cocktail or (B) dried format reagents. The lysed sample is first gated on “time” to allow exclusion of any data bursts or air bubbles in the acquisition. A single parameter histogram of the viability dye is then displayed, gated on time, which allows selection of viable only events (negative for near-IR viability stain). From there, a dotplot of FS peak versus FS area (FS INT) is displayed to allow identification and exclusion of doublets. A CD45 versus SS dotplot allows visualization of multiple populations. A SS versus CD19 dotplot gated on singlets only displays all CD191 B-cells. From there, the surface immunoglobulins kappa and lambda are shown for the entire population. CD31 cells are gated similarly. A SS versus CD3 dotplot displays all CD31 T-cells; subsequently, a CD8 versus CD4 dotplot shows the relative proportions of subsets. A lymphoid region (gated on a CD45 versus SS dotplot) is used as the gate for a dotplot of CD19 versus CD3, showing any possible dual positive CD31CD191. Dotplots of CD19 versus either CD5 or CD10 show populations of B-cells that express CD5 or CD10. In this figure, kappa and lambda light chains are only shown gated on CD191CD101, but for all samples CD191CD51 was evaluated. The CD3 versus CD56 dotplot also gated on the lymphoid region is used in the evaluation of NK cells (CD32 CD561) and NK T-cells (CD31 CD561). B-cell maturation is shown on the dotplot of CD20 versus CD10 gated on CD191 cells; early B-cells have the brightest CD10 and no CD20. When maturing, B-cells initially lose some CD10 and gain CD20. The brightest CD20 cells then start to lose the CD10 until they become mature B-cells that lack CD10. The final two dotplots show singlet gated plots of CD20 and CD10 versus SS, showing the intensity of CD20 and CD10 on all the populations that have been previously gated.
365
NOVEL LYMPHOCYTE SCREENING TUBE USING DRIED MONOCLONAL ANTIBODY REAGENTS
Table 1 Patient Diagnoses and Specimen Types for All 42 Samples Used for Evaluation Diagnosis
Peripheral blood
Bone marrow
Tissue
Total
4
–
1
5
– 1 3 1 – –
5 3 – 1 – 2
– – – 1 3 –
5 4 3 3 3 2
– – 1 – 2 2 1 – – – – 15
2 – 1 1 – – – 1 – 1 – 17
– 2 – 1 – – – – 1 – 1 10
2 2 2 2 2 2 1 1 1 1 1 42
Chronic lymphocytic leukemia/small lymphocytic lymphoma Multiple myeloma B-cell acute lymphoblastic leukemia Hairy cell leukemia Normal Diffuse large B-cell lymphoma Myelodysplasia with refractory anemia with ring sideroblasts Acute myeloid leukemia Follicular lymphoma T-cell acute lymphoblastic leukemia Idiopathic thrombocytopenic purpura Monocytosis Monoclonal T-cell population Monoclonal B-cell population Lymphoplasmacytic lymphoma Marginal cell lymphoma Refractory anemia with unilineage dysplasia Peripheral T-cell lymphoma Total
Table 2 Summary of the Samples Used in the Comparison Study Normal
Red cell disorder
B-cell malignancy
T-cell malignancy
Myeloid malignancy
Other
Total
1 2 2 5
0 3 0 3
9 4 7 20
3 1 1 5
2 2 0 4
0 5 0 5
15 17 10 42
Peripheral blood Bone marrow Tissue Total
Normal samples were not found to have any malignancy after chart review.
liquid to dry format from e450 to PE-Cy7 and PE-Cy7 to Pacific Blue, respectively. RESULTS In total, 42 samples were split and stained with liquid and dried reagents (Table 1). Antigens were selected to identify most lymphoid malignancies. In addition, some abnormal nonlymphoid samples were run, and while a diagnosis could not be made from this screening tube, the abnormal population was readily identified to allow follow-up testing (Supporting Information Fig. 1). SamTable 3 Results of Cell Population Analysis Between the Liquid and Dried Formats Test tube (Duraclone) Positive Negative Total
Comparison tube (liquid cocktail) Positive Negative Total 32 0 32
0 10 10
32 10 42
Negative samples were confirmed samples to have absence of malignancy at the time of analysis. Positive samples were confirmed samples to have malignancy at the time of analysis. Morphological, pathological, and clinical information were also used to determine the presence or absence of malignancy.
Cytometry Part B: Clinical Cytometry
ples were divided into broad categories of malignancy; normal, B-cell, T-cell, myeloid, and “other” (e.g., multiple myeloma) sections. Sample types included peripheral blood, bone marrow, and tissues specimens (which include soft tissues, cerebral spinal fluids, nodes, and other fluids) (Table 2). Comparison of the liquid to dried reagents demonstrated equivalency of interpretation between the formats for all 42 samples (Table 3). A number of normal samples were run, Figure 1 shows an example of some of the analysis performed in KaluzaTM for a normal bone marrow sample. Time is used to gate out any variations in flow rate; subsequently, the viability dye is used to exclude any cells that may have been damaged during transportation or storage since collection. A dotplot of Forward Scatter (FS) - Peak versus FS-area is used to remove doublets. A region named “singlets only” is created on this histogram. A dotplot of CD45 versus SSC is gated on the singlets only region. On this dotplot, a lymph gate is drawn and used in future dotplots. A dotplot of CD19 versus SSC is gated on the above singlets gate, and a region created that allows for surface immunoglobulin staining to be investigated on CD191 cells. Similarly CD4 and CD8 are viewed from a region drawn on a dotplot of CD3 versus SSC. In Figure 1, a selection of the total number of dotplot are shown, The colors for each population
366
HEDLEY ET AL.
FIG. 2. Examples of positive samples for dried or liquid format tubes. Dotplots of dried reagent tubes are shown in panels A, C, and E. Comparison dotplots of liquid format tubes are shown in panels B, D, and F. Monoclonal populations shown in samples from: (A, B) Peripheral blood, two malignant populations: CD191CD102CD201Kappa1 variable CD5 expression (red) and a CD191CD51Lambda1 (blue), and polyclonal B-cells (black). (C, D) Bone marrow, CD191CD52CD102Kappa1. (E, F) Lymph node tissue, CD191CD52CD10 partial, CD201, Lambda1.
are: CD191 red, CD31 blue, CD191CD101 orange, CD32CD561 dark green, CD32CD562 light green, granulocytes purple, and monocytes beige.
Figure 2 shows three examples of positive samples for both liquid and dried formats for a number of different positive samples. As this is a lymphocyte screening tool,
Cytometry Part B: Clinical Cytometry
NOVEL LYMPHOCYTE SCREENING TUBE USING DRIED MONOCLONAL ANTIBODY REAGENTS
367
FIG. 3. Comparison of gating percentages from liquid and dried format reagents for different antigens. For all plots, x-axis is percentage of cells with liquid format and y-axis is percentage gated with dried format reagents. A. CD31 cells gated from a dotplot of CD3 versus SS gated on singlets. B. CD191 cells gated from a dotplot of CD19 versus SS gated on singlets. C. CD191CD51 cells gated from a dotplot of CD19 versus CD5 gated on all lymphocytes. D. CD191CD101 cells gated from a dotplot of CD19 versus SS gated on all lymphocytes. E. All kappa and lambda cells gated from a dotplot of kappa versus lambda gated on all CD191 cells. F. CD201 cells gated from a dotplot of CD20 versus SS gated on all lymphocytes.
representative samples for peripheral blood (A), bone marrow (B), and lymph node tissue (C) are shown. There is excellent agreement between the dried and liquid reagents with all the antigens. Comparison of CD3 and CD5 staining showed dimmer staining when comparing Pacific Blue to PE-Cy7 fluorochromes and e450 to PE-Cy7; however, this did not alter sample interpretation between the dried and liquid formats. Figure 2A shows
Cytometry Part B: Clinical Cytometry
a population of B-cells, gated on CD19 that seem to be polyclonal with slight excess of lambda light chains (top row Dotplot 2). When CD19 versus CD5 is viewed, it can be seen that there are three separate expressions of CD52CD191CD52, CD191CD51, and CD191CD5 variable (top row Dotplot 3). The B-cells can also be separated by their expression of CD20 (dim to bright; bottom row Dotplot 3) When each population is
368
HEDLEY ET AL.
FIG. 4. Comparison of gating percentages from liquid and dried format reagents for different antigens. For all plots, x-axis is percentage of cells with liquid format and y-axis is percentage gated with dried format reagents. A. CD41 cells gated from a dotplot of CD8 versus CD4 gated on CD31 cells. B. CD81 cells gated from a dotplot of CD8 versus CD4 gated on CD31. C. CD41CD81 cells gated from a dotplot of CD8 versus CD4 gated on CD31 cells. D. CD32CD561 cells (NK cells) gated from a dotplot of CD3 versus CD56 gated on all lymphocytes.
analyzed separately, it can be seen that there are three distinct light chain patterns: (1) CD191CD5 variableexpressing kappa light chains (red), (2) CD191CD51 expressing lambda light chains (blue), and (3) a normal population of polyclonal B-cells (black). This sample contains two monoclonal populations with differing light chain expression as well as some normal polyclonal B-cells. Statistics for the gating percentages and staining intensity (median channel number) were exported using Kaluza software and graphed in Excel for some key antigens. Figure 3 shows the gating percentages for: CD31, CD191, CD191CD51, CD191CD101, kappa and lambda, and CD201. The hashed lines in each graph are the 2SD boundaries based on the population data within that graph. Figure 4 shows data similar to that in Figure 3 for CD41, CD81, CD41CD81, and CD32CD561. Both Figures 3 and 4 show excellent agreement between the liquid format (x-axes) and the dried format (y-axes). For all panels in both Figures 3 and 4, [mt]95% of all the samples lie within the 2SD boundaries. Figure 5 shows median channel for a selected number of antigens (CD3 and CD5 were not considered as the fluorochrome changed between formats). It should be noted that analysis of the expression of surface immunoglobu-
lins is extremely sensitive to washing, and we hypothesize that the outliers (one higher and one lower for the dried reagent) for both kappa and lambda light chains seen in Figures 5B and 5D are due to differences in sample washing. Figures 5A and 5F (CD8 FITC and CD32CD56 PE-Cy5.5, respectively) showed similar signals when comparing dried with liquid format reagents; however, on average, the other antigens analyzed in panels B, C, D, E, G, H (Kappa FITC, CD4PE, Lambda PE, CD19 ECD, CD20 APC, and CD101 APC Alexa700) all showed brighter signal in the dried format. Equivalence between formats was seen with all antigens, with the exception of CD4, which gave a brighter signal for all samples in the dried format over the liquid format. DISCUSSION Over 50% of samples suspected of a hematological malignancy received in our flow cytometry laboratory are from patients with mature lymphoid malignancies. Additionally, lymph node fine-needle aspirates, small tissue samples, and body fluids with low cell counts are sent for evaluation by flow cytometry. These samples often yield only enough cells for single tube analysis. The inclusion of a viability dye has proven extremely useful, particularly in tissue samples where nonspecific
Cytometry Part B: Clinical Cytometry
NOVEL LYMPHOCYTE SCREENING TUBE USING DRIED MONOCLONAL ANTIBODY REAGENTS
369
FIG. 5. Comparisons of different antigen intensity from both liquid and dried format reagents. For all plots, x-axis is percentage of cells with liquid format and y-axis is percentage gated with dried Duraclone reagents. A. CD8 FITC1 cells gated from a dotplot of CD8 versus CD4 gated on CD31 cells. B. Kappa FITC1 cells gated from a dotplot of CD19 versus SS gated on singlets. C. CD4 PE1 cells gated from a dotplot of CD8 versus CD4 gated on CD31 cells. D. Lambda PE1 cells gated from a dotplot of CD19 versus SS gated on singlets. E. CD19 ECD1 cells gated from a dotplot of CD19 versus SS gated on singlets. F. CD32CD56 PE2Cy5.51 cells (NK cells) gated from a dotplot of CD3 versus CD56 gated on all lymphocytes. G. CD20 APC1 cells gated from a dotplot of CD20 versus SS gated on singlets. H. CD101 APC Alexa 700 gated from a dotplot of CD19 versus CD10 gated on all lymphocytes.
staining with tandem dyes can be a significant issue (Supporting Information Fig. 2). Morphology is always available to the laboratory, and clinical findings are usually, but not always, provided, aiding in interpretation of samples. Pattern recognition with the use of a standardized analysis template allows for quick learning by laboratory staff and clinicians and allowed for comparative
Cytometry Part B: Clinical Cytometry
analysis between formats in this study. The goal of this study was to determine whether substitution of liquid with dried format reagent would alter sample interpretation, percentage of subset gated, and the signal intensity. Kaluza software was used with our customized standardized templates and allowed for comparative analysis of both data from liquid and dried formats for the goals
370
HEDLEY ET AL.
of this study. The exported numerical data from these templates showed that there was excellent agreement between formats as measured by sample interpretation, equivalence of enumerated percentage of major populations, or overall staining intensity. Therefore, we believe we have demonstrated that dry coated reagents do not hinder our ability to detect multiple neoplasms with our initial screening tube. We have previously reported that a standardized Kaluza report template has streamlined result interpretation and has already improved efficiency in our laboratory (15). The substitution of dried format tubes over in house prepared liquid combinations will allow for the significant reductions in sample handling, instrument setup and analysis time can now be coupled with substantial reduction in the amount of quality assurance for batches of liquid reagents. The substitution of dry coated (DuraCloneTM) reagents for liquid cocktails will reduce; the time spent making reagent cocktails, time validating reagent, wastage of left over antibodies, and the potential for errors when making expensive cocktails. CONCLUSION As early as 2006, Wood et al. (6) have shown the feasibility of routine ten color flow cytometry in the diagnosis of leukemia and lymphoma in the clinical laboratory. In our laboratory, using a single screening tube for lymphoid malignancies has allowed standardized evaluation of a large percentage of samples received for flow cytometry. Multiplexing with a viability dye has increased the usability of this approach to a wide range of samples, including those with low cell counts and/or poor viability (often seen with tissue samples). Other groups have reported adding even more antibodies to take advantage of multiplexing (4), and it should be possible to have a comprehensive screening panel that assesses lymphoid, myelomonocytic cells and viability in a single tube. However, increased complexity in single tube multicolor analysis is accompanied by challenges in reagent cocktail preparation, a time consuming, error-prone process in busy clinical laboratories. In this study, we have demonstrated equivalency between a novel dried reagent combination and liquid format tubes in sample interpretation, staining percentages, and intensity in a variety of sample types with diagnoses including normal, reactive lymphocytosis, chronic lymphocytic leukemia, and lymphoma. Dried, room-temperature tubes offer advantages over liquid format, as they are stable for at least 12 months and, unlike liquid reagents, are made with a single lot for each antigen. In our laboratory, using dried format tubes effectively eliminates the need to individually titrate, pipette, create liquid cocktails, and documentation. In the previous year, [more than ]50 batches of our lymphocyte screening tube were prepared, validated, and documented. Removal of these duties allows significant time savings in our laboratory while eliminating multiple sources of error. The relative instability of APC tandem dyes limits the outdate for liquid
format cocktails containing them, which is not the case with the dried reagent. Finally, we are confident that the use of dried format tubes in our laboratory will translate to improved efficiencies and reduce the risk of errors in preparation of complex liquid reagent cocktails.
ACKNOWLEDGMENTS The authors thank the following people in the laboratory: Samantha Vandevyvere, Dominka Benjamins, and Erin Bourguignon. LITERATURE CITED 1. Craig FE, Foon KA. Flow cytometric immunophenotyping for hematologic neoplasms. Blood 2008; 111:3941–3967. 2. Preijers FW, Huys E, Moshaver B. OMIP-010: A new 10-color monoclonal antibody panel for polychromatic immunophenotyping of small hematopoietic cell samples. Cytometry Part A 2012; 81A:453–455. 3. Mittag A, Lenz D, Gerstner AO, Sack U, Steinbrecher M, Koksch M, Raffael A, Bocsi J, Tarnok A. Polychromatic (eight-color) slide-based cytometry for the phenotyping of leukocyte, NK, and NKT subsets. Cytometry Part A 2005; 65A:103–115. 4. Bradford JA, Buller G, Suter M, Ignatius M, Beechem JM. Fluorescence-intensity multiplexing: Simultaneous seven-marker, two-color immunophenotyping using flow cytometry. Cytometry Part A 2004; 61A:142–152. 5. Preijers FW, Huys E, Favre C, Moshaver B. Establishment of harmonization in immunophenotyping: A comparative study of a standardized one-tube lymphocyte-screening panel. Cytometry Part B 2014; 86B:418–425. 6. Wood BL, Arroz M, Barnett D, DiGiuseppe J, Greig B, Kussick SJ, Oldaker T, Shenkin M, Stone E, Wallace P. 2006 Bethesda international consensus recommendations on the immunophenotypic analysis of hematolymphoid neoplasia by flow cytometry: Optimal reagents and reporting for the flow cytometric diagnosis of hematopoietic neoplasia. Cytometry Part B 2007; 72B: S14–S22. 7. Preijers FW, Huys E, Leenders M, Nieto L, Gautherot E, Moshaver B. The new violet laser dye, krome orange, allows an optimal polychromatic immunophenotyping based on CD45-KO gating. J Immunol Methods 2011; 372:42–51. 8. Kalina T, Flores-Montero J, van der Velden VH, Martin-Ayuso M, Bottcher S, Ritgen M, Almeida J, Lhermitte L, Asnafi V, Mendonca A, et al. EuroFlow standardization of flow cytometer instrument settings and immunophenotyping protocols. Leukemia 2012; 26:1986–2010. 9. van Dongen JJ, Lhermitte L, Bottcher S, Almeida J, van der Velden VH, Flores-Montero J, Rawstron A, Asnafi V, Lecrevisse Q, Lucio P, et al. EuroFlow antibody panels for standardized n-dimensional flow cytometric immunophenotyping of normal, reactive and malignant leukocytes. Leukemia 2012; 26:1908–1975. 10. Greig B, Oldaker T, Warzynski M, Wood B. 2006 Bethesda international consensus recommendations on the immunophenotypic analysis of hematolymphoid neoplasia by flow cytometry: Recommendations for training and education to perform clinical flow cytometry. Cytometry Part B 2007; 72B: S23–S33. 11. Biancotto A, Fuchs JC, Williams A, Dagur PK, McCoy JP Jr. High dimensional flow cytometry for comprehensive leukocyte immunophenotyping (CLIP) in translational research. J Immunol Methods 2011; 363:245–261. 12. Johansson U, Macey M. Tandem dyes: Stability in cocktails and compensation considerations. Cytometry Part B 2014; 86B:164–174. 13. Le Roy C, Varin-Blank N, Ajchenbaum-Cymbalista F, Letestu R. Flow cytometry APC-tandem dyes are degraded through a cell-dependent mechanism. Cytometry Part A 2009; 75A:882–890. 14. Rawstron AC, Bottcher S, Letestu R, Villamor N, Fazi C, Kartsios H, de Tute RM, Shingles J, Ritgen M, Moreno C, et al. European Research Initiative in CLL. Improving efficiency and sensitivity: European Research Initiative in CLL (ERIC) update on the international harmonised approach for flow cytometric residual disease monitoring in CLL. Leukemia 2013; 27:142–149. 15. Keeney M, Benjamins D. A 10 color 11 antibody and viability lymphocyte screening tube. ICCS e-Newsletter 2014;5. Available at: http://cytometry.org/public/newsletters/eICCS-5-1/index.php
Cytometry Part B: Clinical Cytometry
