Precision performance of the DxFLEX ClearLLab 10C application system

Sample Preparation

ClearLLab 10C System

ClearLLab 10C System

Integrated FDA cleared and CE-marked IVD leukemia and lymphoma* (L&L) immunophenotyping solution for lymphoid and myeloid lineages

*For Non-Hodgkin’s lymphoma only

Sample Acquisition

DxFLEX flow cytometer with tube auto loader

DxFLEX Flow Cytometer

The DxFLEX Flow Cytometer streamlines high-complexity testing with a 13-color** capability and new detector technology for easier compensation.

**In the US, the DxFLEX flow cytometer is cleared for 10-color in vitro diagnostic use with the ClearLLab 10C Reagent System. Fluorescence channels FL11-FL13 and all other applications are for Research Use Only.

Sample Analysis

Kaluza C Analysis Software 10 Color Data

Kaluza C

Flow Cytometry Analysis Software

Purpose-built for clinical laboratories, Kaluza C Flow Cytometry Analysis Software helps you turn critical data into meaningful patient reports.

Introduction

Flow cytometry immunophenotyping (FCI) is a valuable diagnostics tool for hematolymphoid malignancies.1

The DxFLEX* flow cytometer is the clinical version of the CytoFLEX flow cytometer. Derived from the successful research platform, the DxFLEX flow cytometer has the advantage of using Avalanche Photodiode (APD) technology, which has higher sensitivity than typical Photomultiplier Tube (PMT)-based cytometers and makes high-color flow cytometry testing more accessible because there is a linear correlation between gain setting/voltages and measured fluorescence intensities.2, 3

By leveraging the ClearLLab 10C** application, a multisite clinical study was conducted by using DxFLEX flow cytometers to demonstrate the precision performance of the DxFLEX ClearLLab 10C as a complete system.

* In the US, the DxFLEX flow cytometer is cleared for 10-color in vitro diagnostic use with the ClearLLab 10C Reagent System. Fluorescence channels FL11-FL13 and all other applications are for Research Use Only.

** The ClearLLab 10C Reagent System on the DxFLEX flow cytometer has received 510k clearance and IVDR certification.

Results

Precision using ClearLLab Control Cells

Targeted populations as those defined in the assay sheet for ClearLLab Control Cells were evaluated for precision performance on the DxFLEX ClearLLab10C system. Both ClearLLab Control Cells, Normal and Abnormal, were tested using the ClearLLab 10C B, T, M1 and M2 panels. A total of 140 replicates per each panel for each level of control cells were collected and analyzed using Kaluza C software. The overall mean percent recovery of the targeted populations in the B panel ranged from 7.99% to 97.92%, in the T panel ranged from 2.77% to 97.83%, in the M1 panel ranged from 7.70% to 97.88%, and in the M2 panel ranged from 7.96% to 98.03%. For each targeted population, the overall repeatability from both levels ranged from 0.24% CV to 5.56% CV, and the reproducibility ranged from 1.00% CV to 16.93% CV. The representative results from using the Abnormal Control Cells are shown in Table 1 for B panel, Table 2 for T panel, Table 3 for M1 panel, and Table 4 for M2 panel. Although data is not shown, Normal ClearLLab Control Cells demonstrated similar performance. In addition, each individual site had satisfying precision performance on each individual DxFLEX instrument, with repeatability from both levels ranging from 0.01 % CV to 9.49 % CV and variation within instrument ranging from 0.02% CV to 12.16% CV.

Name N Mean Repeatability Between runs Between days Between instruments Reproducibility
SD CV% SD CV% SD CV% SD CV% SD CV%
% Total CD45 140 95.56 0.65 0.68 0.71 0.74 3.83 4.01 1.51 1.58 4.23 4.43
% Gated CD45+ 140 97.92 0.61 0.62 0.58 0.59 2.49 2.54 1.73 7.76 3.14 3.21
% Gated Lymphs 140 30.34 0.88 2.89 0.70 2.31 0.27 0.88 1.30 4.27 1.73 5.72
% Gated Monos 140 9.09 0.35 3.85 0.19 2.04 0.25 2.72 0.87 9.56 0.99 10.85
% Gated Grans 140 57.80 1.15 1.98 0.96 1.66 0.38 0.66 2.45 4.24 2.90 5.01
% Gated CD5+ 140 74.89 0.48 0.64 0.36 0.48 0.41 0.55 0.16 0.21 0.75 1.00
% Gated CD10+ 140 56.99 1.19 2.09 0.89 1.57 0.37 0.66 2.71 4.75 3.11 5.46
% Gated CD19+CD20+ 140 10.05 0.32 3.20 0.16 1.61 0.00 0.00 1.17 11.62 1.22 12.16
% Gated CD34+ 140 7.99 0.26 3.23 0.20 2.55 0.35 4.43 0.50 6.25 0.69 8.70
% Gated CD38+ 140 8.65 0.31 3.56 0.28 3.25 0.29 3.33 1.15 13.29 1.26 14.52
% Gated CD200+ 140 9.16 0.31 3.42 0.16 1.72 0.09 0.99 1.06 11.59 1.12 12.25
% Gated Kappa 140 59.51 1.38 2.32 0.47 0.80 0.54 0.91 0.39 0.65 1.61 2.70
% Gated Lambda 140 40.45 1.37 3.38 0.48 1.19 0.56 1.39 0.23 0.56 1.57 3.88
Name N Mean Repeatability Between runs Between days Between instruments Reproducibility
SD CV% SD CV% SD CV% SD CV% SD CV%
% Total CD45 140 95.34 1.38 1.45 0.92 0.97 3.59 3.76 1.75 1.84 4.32 4.53
% Gated CD45+ 140 97.83 1.22 1.25 0.62 0.64 2.31 2.36 1.90 1.94 3.29 3.36
% Gated Lymphs 140 30.34 0.86 2.89 1.10 3.62 0.41 1.34 2.02 6.65 2.49 8.20
% Gated Monos 140 8.91 0.34 3.85 0.33 3.66 0.29 3.27 0.98 10.95 1.12 12.60
% Gated Grans 140 58.08 1.09 1.88 1.38 2.38 0.51 0.88 3.25 5.59 3.73 6.42
% Gated CD5+ 140 75.50 0.55 0.73 0.19 0.25 0.58 0.76 0.94 1.25 1.25 1.65
% Gated CD34+ 140 8.15 0.44 5.46 0.00 0.00 0.00 0.00 0.50 6.11 0.67 8.19
% Gated CD2+ 140 84.61 0.48 0.57 0.13 0.15 0.48 0.56 1.47 1.73 1.62 1.92
% Gated CD3+ 140 76.08 0.51 0.67 0.29 0.38 0.44 0.58 0.33 0.43 0.81 1.06
% Gated CD4+CD8- 140 65.11 0.70 1.08 0.67 1.03 0.00 0.00 1.22 1.87 1.56 2.39
% Gated CD4-CD8+ 140 29.10 0.63 2.15 0.67 2.31 0.00 0.00 2.39 8.20 2.56 8.78
% Gated CD7+ 140 79.02 0.50 0.63 0.55 0.69 0.77 0.98 3.92 4.96 4.06 5.14
% Gated CD3-CD56+ 140 11.16 0.33 2.95 0.18 1.57 0.22 1.93 1.81 16.17 1.86 16.62
% Gated TCRgd+ 140 2.79 0.11 4.09 0.09 3.22 0.04 1.44 0.15 5.30 0.21 7.57
Name N Mean Repeatability Between runs Between days Between instruments Reproducibility
SD CV% SD CV% SD CV% SD CV% SD CV%
% Total CD45 140 95.52 1.05 1.10 0.77 0.81 3.62 3.79 1.66 1.74 4.19 4.39
% Gated CD45+ 140 97.88 0.95 0.98 0.71 0.72 2.38 2.43 1.88 1.92 3.26 3.33
% Gated Lymphs 140 30.12 1.04 3.44 0.75 2.49 0.42 1.39 1.97 6.53 2.38 7.91
% Gated Monos 140 8.89 0.30 3.37 0.17 1.86 0.37 4.20 0.55 6.19 0.75 8.42
% Gated Grans 140 58.38 1.27 2.18 0.94 1.61 0.63 1.08 2.98 5.11 3.43 5.88
% Gated CD10+ 140 57.62 1.29 2.24 1.00 1.74 0.72 1.25 3.20 5.55 3.67 6.36
% Gated CD34+ 140 8.04 0.28 3.51 0.23 2.92 0.28 3.44 0.71 8.84 0.85 10.53
% Gated CD7+ 140 80.95 0.44 0.55 0.34 0.42 0.52 0.64 4.00 4.94 4.07 5.03
% Gated CD14+CD11b+ 140 7.77 0.36 4.61 0.20 2.54 0.20 2.58 0.94 12.04 1.04 13.40
% Gated CD13+ 140 58.14 1.31 2.25 0.90 1.54 0.80 1.38 2.85 4.91 3.36 5.79
% Gated CD16+ 140 54.33 1.22 2.25 0.96 1.76 0.65 1.20 3.19 5.87 3.61 6.64
% Gated CD14+CD64+ 140 7.70 0.35 4.58 0.21 2.72 0.23 3.00 0.89 11.51 1.00 13.03
% Gated HLA-DR+ 140 8.45 0.35 4.08 0.20 2.36 0.39 4.67 0.37 4.40 0.67 7.96
Name N Mean Repeatability Between runs Between days Between instruments Reproducibility
SD CV% SD CV% SD CV% SD CV% SD CV%
% Total CD45 140 95.53 0.77 0.81 0.35 0.37 3.51 3.67 1.74 1.82 4.00 4.19
% Gated CD45+ 140 98.03 0.65 0.66 0.30 0.31 2.19 2.23 1.74 1.78 2.89 2.95
% Gated Lymphs 140 29.97 0.83 2.76 0.83 2.76 0.63 2.09 2.48 8.26 2.81 9.37
% Gated Monos 140 8.78 0.35 3.97 0.14 1.65 0.25 2.82 0.64 7.32 0.79 8.95
% Gated Grans 140 58.61 1.03 1.76 0.98 1.67 0.81 1.38 3.44 5.87 3.81 6.50
% Gated CD34+ 140 7.96 0.31 3.85 0.27 3.34 0.32 4.07 0.63 7.93 0.82 10.26
% Gated CD38+ 140 8.31 0.32 3.81 0.25 3.03 0.23 2.81 0.85 10.28 0.97 11.72
% Gated CD13+ 140 58.01 1.05 1.80 0.90 1.55 0.86 1.48 3.78 6.52 4.11 7.09
% Gated HLADR+ 140 8.35 0.34 4.02 0.21 2.51 0.23 2.76 0.60 7.21 0.76 9.06
% Gated CD15+ 140 57.80 1.04 1.80 1.05 1.82 0.82 1.42 3.62 6.26 3.99 6.91
% Gated CD19+ 140 9.69 0.33 3.45 0.06 0.63 0.18 1.88 0.65 6.74 0.76 7.83
% Gated CD33+ 140 8.62 0.36 4.19 0.19 2.26 0.25 2.94 1.11 12.88 1.21 14.04
% Gated CD117+ 140 8.06 0.31 3.84 0.25 3.14 0.34 3.00 0.48 5.90 0.71 8.79
% Gated CD123+ 140 8.16 0.31 3.79 0.28 3.46 0.34 4.12 0.47 5.81 0.72 8.78

Precision using clinical specimens

All four (4) ClearLLab 10C panels were evaluated by targeting both phenotypically normal and phenotypically abnormal whole blood specimens per intended use. The targeted populations summarized in Table 5 were evaluated according to the phenotype assessment.

Phenotype Assessment B PANEL T PANEL M1 PANEL M2 PANEL
Normal Phenotype
  • Mature Kappa+
  • Mature B Lambda+
  • NK cells
  • TCRgd cells
  • T Helper cells
  • T Suppressor cells
  • CD16+ Neutrophils
  • CD64+ Monocytes
  • CD15+ Neutrophils
  • CD33+ Monocytes
Abnormal Phenotype
  • CD19+CD5+
N/A
  • CD34-CD7+
  • CD117+
  • CD34+CD123+

Table 5. Evaluated Targeted Populations.

Both qualitative and quantitative results were assessed in this study. For quantitative evaluation with data combined from three (3) operators across two (2) DxFLEX instruments, the mean percent recovery of the targeted populations from the ClearLLab 10C B panel ranged from 8.84%-91.55%, the T panel ranged from 2.20%-80.62%, the M1 panel ranged from 6.22%-57.46%, and the M2 panel ranged from 2.85%-62.79%. Representative results from the ClearLLab 10C B panel are summarized in Table 6, and the statistics for repeatability such as mean, SD, %CV are provided for each targeted population. Also included are SD and %CV calculations between runs, between instruments, between operators and within sample.

Similarly, the repeatability results using the ClearLLab 10C T, M1 and M2 panels were overall below 8% CV, and the CVs between runs, between instruments, between operators and within sample were below 13% (Data not shown).

Specimen ID Population N Mean Repeatability Between runs Between instruments
SD CV% SD CV% SD CV%
MLL-TC7-002 CD19+CD5+ %Gated 36 91.55 0.67 0.73 0.40 0.44 0.64 0.70
MLL-TC7-004 Mature B Kappa+ %Gated 36 90.62 0.68 0.75 0.17 0.19 0.00 0.00
Mature B Lambda+ %Gated 36 8.84 0.58 6.56 0.19 2.20 0.00 0.00
MLL-TC7-006 Mature B Kappa+ %Gated 36 63.62 1.99 3.12 0.00 0.00 0.00 0.00
Mature B Lambda+ %Gated 36 35.75 1.87 5.22 0.26 0.73 0.00 0.00
Specimen ID Population N Mean Between operators Within sample
SD CV% SD CV%
MLL-TC7-002 CD19+CD5+ %Gated 36 91.55 0.00 0.00 1.01 1.10
MLL-TC7-004 Mature B Kappa+ %Gated 36 90.62 0.00 0.00 0.70 0.77
Mature B Lambda+ %Gated 36 8.84 0.00 0.00 0.61 6.92
MLL-TC7-006 Mature B Kappa+ %Gated 36 63.62 0.41 0.65 2.03 3.19
Mature B Lambda+ %Gated 36 35.75 0.77 2.16 2.04 5.70

For qualitative evaluation, the DxFLEX 10C system demonstrated 100% agreement across the two (2) DxFLEX instruments and three (3) operators for both phenotypically normal and abnormal specimens. Table 7 through Table 10 summarize the positive (abnormal), negative (normal), and overall agreement for the ClearLLab 10C B, T, M1 and M2 panels, respectively.

Phenotype Total Actual Total Expected Agreement Estimate 95% Confidence Limits
Lower Upper
Abnormal 72 72 100% 0.949 1.000
Normal 36 36 100% 0.904 1.000
Both 108 108 100% 0.966 1.000

Table 7. ClearLLab 10C B Panel Qualitative Results.

Phenotype Total Actual Total Expected Agreement Estimate 95% Confidence Limits
Lower Upper
Normal 36 36 100% 0.949 1.000

Table 8. ClearLLab 10C T Panel Qualitative Results.
*Note: Only normal specimens were collected; therefore, overall agreement is the same as negative agreement.

Phenotype Total Actual Total Expected Agreement Estimate 95% Confidence Limits
Lower Upper
Abnormal 36 36 100% 0.904 1.000
Normal 36 36 100% 0.904 1.000
Both 72 72 100% 0.949 1.000

Table 9. ClearLLab 10C M1 Panel Qualitative Results.

Phenotype Total Actual Total Expected Agreement Estimate 95% Confidence Limits
Lower Upper
Abnormal 36 36 100% 0.904 1.000
Normal 36 36 100% 0.904 1.000
Both 72 72 100% 0.949 1.000

Table 10. ClearLLab 10C M2 Panel Qualitative Results.

Discussion and conclusion

The precision performance of the DxFLEX ClearLLab 10C system using the control material of ClearLLab Control Cells (normal and abnormal) was evaluated across multiple external sites. Quantitative results of each targeted population from all four (4) panels of ClearLLab 10C application demonstrated well acceptable6 imprecision performance within each site (< 10% CV for repeatability and <13% CV for within instrument variation) and from site to site (< 6% CV for repeatability and < 17% CV for reproducibility).

The study using clinical specimens also demonstrated satisfying precision performance quantitatively (< 8% CV for repeatability and < 13 % CV within sample) and qualitatively (100% agreement) for the clinical application per its intended use.

Considering all the variations associated with the DxFLEX 10C application system from lab to lab, such as manual sample preparation, data analysis, operator to operator, and instrument to instrument, the precision performance summarized above shall be able to meet the customer need in clinical flow labs.

Materials

  1. DxFLEX Flow Cytometer with CytExpert for DxFLEX software and DxFLEX AutoLoader.
  2. DxFLEX Daily QC Fluorospheres
  3. ClearLLab Compensation Kit and Compensation Beads
  4. ClearLLab Control Cells (Normal and Abnormal)
  5. ClearLLab 10C Panels (B, T, M1 and M2 Cell Tubes)
  6. IOTest 3 Fixative Solution and Lysing Solution
  7. CytoFLEX and DxFLEX Sheath Fluid
  8. PBS
  9. Heat inactivated FBS
  10. FlowClean Cleaning Agent
  11. Contrad 70 Reagent
  12. De-identified residual clinical specimens per intended use of the ClearLLab 10C application
  13. Kaluza C Analysis software

Methods

The study design when using ClearLLab Control Cells to evaluate the precision performance across multiple clinical sites is shown in Table 11. The same operator prepared all replicates for the morning run as well as for the afternoon run. Both levels of control cells (Normal and Abnormal levels), which express all surface markers corresponding to the antibodies in the ClearLLab 10C Panels, were used.

Site Site 1 Site 2 & Site 3
Design 20 days x 2 runs x 2 replicates 5 days x 2 runs x 3 replicates

Table 11. Precision using ClearLLab 10C Control Cells study design.

When using de-identified residual clinical specimens to demonstrate the application’s precision performance across multiple operators and multiple DxFLEX flow cytometers, the following study was designed at a clinical site: after screening a residual sample and confirming the eligibility for clinical study use per intended use of the application, specimen was enrolled and prepared by three (3) individual operators. Three (3) individual replicates were prepared and acquired by each operator once in the morning and once in the afternoon. Each replicate tube was acquired once on two (2) DxFLEX flow cytometers separately at a random order.

All products were handled based on the instructions for use or associated System Guide. All tested residual specimens were approved under WCG IRB. All acquired data from DxFLEX flow cytometers were analyzed using Kaluza C Analysis software.

SAS statistical software was used. For analysis of quantitative results, per CLSI EP05-A3, 4 variance components were estimated for all sources of variability. Standard deviation (SD) and coefficients of variations (%CV) were calculated for all sources of variability from variance component estimates. For analysis of qualitative results for specimen’s evaluation, per CLSI EP12-Ed3, 5 both positive (phenotypically abnormal) and negative (phenotypically normal) agreement, as well as overall agreement, were calculated along with its 95% confidence intervals (2-sided).

References

  1. Craig, F. E. and K. A. Foon (2008). "Flow cytometric immunophenotyping for hematologic neoplasms." Blood 111(8): 3941-3967.
  2. Hedley BD, Cheng G, Keeney M, et al. A multicenter study evaluation of the ClearLLab 10C panels. Cytometry Part B: Clinical Cytometry. Published online 2020.
  3. Lawrence WG, Varadi G, Entine G, Podniesinski E, Wallace PK. Enhanced Red and Near Infrared Detection in Flow Cytometry Using Avalanche Photodiodes. Cytometry A. 2008;73(8):767-776. doi:10.1002/cyto.a.20595
  4. CLSI. Evaluation of Precision of Quantitative Measurement Procedure; Approved Guideline-Third Edition. CLSI document EP05-A3. Wayne, PA: Clinical Laboratory Standard Institute; 2014.
  5. CLSI. User Protocol for Evaluation of Qualitative Test Performance; Approved Guideline – Third Edition. CLSI document EP12-Ed3. Wayne, PA: Clinical and Laboratory Standard Institute; 2023.
  6. Wood, B., Jevremovic, D., Béné, M. C., Yan, M., Jacobs, P., Litwin, V., & on behalf of ICSH/ICCS Working Group. (2013). Validation of cell-based fluorescence assays: Practice guidelines from the ICSH and ICCS – part V – assay performance criteria. Cytometry Part B: Clinical Cytometry, 84(5), 315-323. doi:https://doi.org/10.1002/cyto.b.21108.

Talk To An Expert