The Cytometry Part B: Clinical Cytometry supplement (74B, Supplement 1, 2008) titled “The Global Health and Diagnostic (Flow) Cytometry—Breakthrough in HIV and Tuberculosis' is sponsored by Beckman Coulter, BD Biosciences and Caltag-Medisystems.
The contributing authors to this article have declared no conflict of interest.
How to cite this article: Pattanapanyasat K, Sukapirom K, Kowawisatsut L, Thepthai C. New BD FACScount™ CD4 reagent system for simultaneous enumeration of percent and absolute CD4 T-lymphocytes in HIV-1-infected pediatric patients. Cytometry Part B 2008; 74B (Suppl. 1): S98–S106, 2008.
Absolute CD4+ T-lymphocyte counts are used in the initiation and monitoring of antiretroviral therapy in HIV-infected patients. Becton Dickinson's (BD) FACSCount™ system was introduced 12 years ago as a dedicated instrument for enumeration of absolute CD4+ T-lymphocytes. However, this system does not provide percent CD4+ T-lymphocyte that is the required monitoring parameter in pediatric patients. We evaluated a new BD FACSCount CD4 software and reagents for simultaneous percent and absolute CD4+ T-lymphocytes in HIV-infected blood.
Percent and absolute CD4+ T-lymphocytes in 149 HIV-infected blood samples were determined using a new FACSCount system. Results of percent and absolute CD4+ T-lymphocytes were compared between the dual-platform (DP) method, using BD FACScan™ flow cytometer plus hematology analyzer and the standard FACSCount system. Correlation and agreement were analyzed using linear regression and Bland–Altman analysis.
Percent CD4+ T-lymphocyte values obtained from the new FACSCount system correlated well with DP FACScan method (r2 = 0.977, P < 0.0001). Mean bias was only −0.36% [limit of agreement (LOA): −2.52% to +1.80%] and percent similarity was 101.36%. For absolute CD4+ T-lymphocyte, the new system correlated highly with standard FACSCount system (r2 = 0.986, P < 0.0001), with a percent similarity of 98.2. Mean bias was +3.39 cells/μl with LOA of −52.53 cells/μl to +59.31 cells/μl.
Human immunodeficiency virus (HIV) and acquired immunodeficiency syndrome (AIDS) pandemic have severely affected health development and deteriorated improvements in life expectancy, particularly in countries with a high prevalence of infection. According to data from the 2007 report on the Global AIDS Epidemic by the joint United Nations Program on HIV/AIDS and World Health Organization (WHO) (1), the estimated number of persons living with HIV worldwide in 2007 was 33.2 million (30.6–36.1 million), with sub-Saharan Africa carrying the highest burden. Of this estimated number, there were 2.5 million (2.2–2.6 million) children under 15 years of age living with HIV. During 2007 alone, an estimated 420,000 (310,000–380,000) children were newly infected with HIV.
Measurement of CD4+ T-lymphocyte counts is particularly relevant in dealing with HIV epidemic throughout the world, as appropriate utilization of antiretroviral therapy (ART) and success of HIV vaccine trials are critically dependent on the availability of CD4 counts. For HIV-infected adults, the absolute number of CD4+ T-lymphocyte counts is a useful guide for clinical decision-making throughout the course of HIV/AIDS from disease progression assessment to ART initiation, and for monitoring ART effectiveness in guiding regiment changes and initiation of opportunistic infection prophylaxis (2–4). In pediatric patients, percent CD4+ T-lymphocyte value is a more useful parameter for monitoring HIV infection as the value varies significantly less than the absolute CD4+ T-lymphocyte count (5). Because standard virological tests for definitive diagnosis of HIV infection in infants are expensive and/or not available in resource-constrained countries, WHO has recommended that percent CD4+ T-lymphocytes of total lymphocytes be used in the decision making to initiate ART in infants under 5 years of age (6, 7). WHO also encourages approaches for presumptive clinical diagnosis of HIV infection in children younger than 18 months, including studies on percent CD4+ T-lymphocytes combined with clinical signs and symptoms (7).
Flow cytometry is the most accepted standard for enumeration of CD4+ T-lymphocytes because of its accuracy, precision, and reproducibility (8, 9). Flow cytometry can be performed using either a dual-platform (DP) or single-platform (SP) method (8–12). DP approach uses two instruments: a flow cytometer (FCM) for generating percent CD4+ T-lymphocyte among total lymphocytes and a hematology analyzer to enumerate absolute lymphocyte count obtained from percent lymphocyte in total whole blood cell count. An absolute CD4+ T-lymphocyte count is then derived by multiplying percent CD4+ T-lymphocyte by absolute lymphocyte count. On the other hand, the SP approach produces absolute CD4+ T-lymphocyte counts without the need of a hematological cell analyzer. These counts are derived from the ratio of CD4+ T-lymphocyte events to a known number of fluorescent microbeads admixed to a known volume of CD4+-stained whole blood.
FACSCount™ system, the first SP CD4+ T-lymphocyte-enumerating FCM, was launched by Becton Dickinson Biosciences (BDB)/Immunocytometry Systems over 12 years ago (13), and represented a remarkable achievement at a time when the international guidelines still called for 12 monoclonal antibodies in six tubes using dual-color strategy (8). This system was the first to count CD4+ and CD8+ T-lymphocytes in a twin tube using a whole blood, no lysis, SP absolute cell count method. Recently, a single tube for CD4+ T-lymphocyte counts was introduced. The tube contains a mixture of monoclonal antibody reagents for CD4/CD3 and a known density of fluorescent microbeads for counting absolute CD4+ T-lymphocytes. However, this standard FACSCount FCM is not suitable for pediatric blood samples, as the system provides absolute CD3+ and CD4+ T-lymphocyte counts and does not provide percent CD4+ T-lymphocyte in total lymphocytes. Ideally, it is preferable to use a single system that would give absolute as well as percent CD4+ T-lymphocyte counts that can be used both in adult and pediatric patients. Recently, a new FACSCount reagent and software system has been introduced, which uses a monoclonal antibody to CD4 T-lymphocytes, a nucleic acid fluorochrome and a combination of two additional monoclonal antibodies to CD14 and CD15. This system allows measurement of both absolute and percent CD4+ T-lymphocytes. It will eventually increase access to percent CD4+ T-lymphocyte determinations for pediatric patients as well as absolute CD4+ T-lymphocyte counts for adult patients in laboratories that are already equipped with a FACSCount FCM. There are presently more than 1,500 BD FACSCount instruments in resource-limited countries, of which about 1,000 instruments are in sub-Saharan Africa countries, 200 in Latin American countries, and more than 250 in Asia-Pacific countries.
The purpose of this study was to evaluate the new FACSCount CD4 reagent and software system for enumerating absolute count and percent CD4+ T-lymphocyte compared to values obtained by the standardFACSCount system and the DP method employing a FACScan™ FCM, TriTEST™ reagents, and hematology analyzer.
MATERIALS AND METHODS
Patients and Blood Samples
One hundred forty-nine HIV-1 seropositive blood samples from adult and pediatric patients ranging in age from 5 months to 77 years were evaluated. Venous blood samples were collected by venipuncture into K3EDTA-containing tubes, kept at room temperature, and processed for immunophenotyping within 6 h. Blood samples, unlinked and anonymous, used in this study were residuals of routine clinical specimens obtained from the Department of Immunology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand. HIV-1 infection was diagnosed serologically (AxSYM HIV-1/HIV-2, Abbott, Germany) with confirmation by two other different serologic tests (1 + 2 VITROS, Ortho-Clinical Diagnostics and SERODIA HIV, Fuji Rebio, Japan). An aliquot of EDTA blood from each sample was also sent for complete blood count using Sysmex DF3000 hematology analyzer (TOA Medical Electronics, Kobe, Japan). The study was approved by the Ethical Committee of the Faculty of Medicine Siriraj Hospital, Mahidol University.
Methods and Equipment
FACScan FCM (BDB, San Jose, CA) is a three-color bench-top instrument equipped with a 15-mW argon ion laser that operates at 488 nm for excitation of three fluorescent parameters: fluorescein isothiocyanate (FITC), phycoerythrin (PE), and peridinin chlorophyll protein (PerCP). When this system is combined with three fluorescence (FITC/PE/PerCP)-conjugated TriTEST monoclonal antibody reagents for CD3/CD4/CD45 (BDB), it is able to generate percent CD4+ T-lymphocyte values. To use this FACScan FCM as a DP method in order to obtain an absolute CD4+ T-lymphocyte count, percent CD4+ T-lymphocyte value is multiplied by absolute lymphocyte count obtained from the hematology analyzer.
FACSCount system (BDB) is a SP bench-top FCM employing standard reagent kit. The instrument is equipped with a green laser for excitation of two fluorescent parameters: PE and tandem fluorochrome composed of PE and Cy™5 cyanine dye (PE-Cy5). In this study, two different FACSCount reagents were tested: (1) standard FACSCount CD4/CD3 reagent kit, consisting of the monoclonal antibody pair, CD4/CD3 (conjugated with PE and PE-Cy5 fluorescence respectively) and a known number of fluorescent microbeads; and (2) the new FACSCount CD4 reagent kit, consisting of a single tube containing a mixture of three monoclonal antibodies, CD4/CD14/CD15 (conjugated with PE/PE-Cy5/PE-Cy5, respectively), a nuclear DNA fluorescent dye, and a known number of fluorescent microbeads. Monoclonal antibody to CD14 recognizes a human monocyte/macrophage antigen, whereas monoclonal antibody to CD15 recognizes a human myelomonocytic antigen.
To assess the accuracy of the FACSCount system, quality control reagents consisting of three known concentrations of fluorescent beads, namely, low (50 beads/μl), medium (250 beads/μl), and high (1,000 beads/μl), were analyzed prior to measuring each batch of stained whole blood samples.
Immunophenotypic Staining of Peripheral Blood
For DP FACScan method, 20 μl of TriTEST three-color monoclonal antibody reagent and 50 μl of EDTA-uncoagulated whole blood were added to a polystyrene tube. The mixture was vortexed and incubated in the dark for 20 min at room temperature before 450 μl of FACS lysing solution (BDB) were added. After 15 min of incubation, the lysed no-wash stained sample was vortexed and analyzed by FACScan FCM.
In standard SP FACSCount method, 50 μl of EDTA-uncoagulated whole blood was added to the CD4/CD3 reagent tube (containing monoclonal antibodies and microbeads) using an electronic pipette (BDB). The tube was vortexed for 5 s and incubated in the dark at room temperature for 60 min. Then, 50 μl of a fixative solution provided with the reagent kit was added to the tube. The tube was vortexed, and the nonlysed stained sample was analyzed in FACSCount FCM using an automated FACSCount software.
Staining with the new FACSCount, CD4 reagent was performed by adding 50 μl of EDTA-uncoagulated whole blood to the CD4 reagent tube using an electronic pipette. The mixture was vortexed for 6 s and incubated in the dark for 30 min at room temperature. Then 50 μl of fixative solution provided in the reagent kit was added to the tube. The tube was vortexed, and the nonlysed stained sample was analyzed in FACSCount FCM using the new FACSCount CD4 software.
Flow Cytometric Analysis
Each TriTEST reagent-stained blood sample was analyzed using MultiSET™ software (BDB) of FACScan FCM. In brief, blood-samples stained with FITC-, PE- and PerCP-conjugated monoclonal antibodies were detected using the logarithmic amplification of fluorescence 1 (FL1), FL2, and FL3 parameters, respectively. Forward scatter and side scatter (SSC-H) signals were measured using a linear scale. After acquiring data on 15,000 cells, a region was automatically set on the lymphocyte cluster (SSC-Hlow/CD45 PerCPhigh+ cells) (R1). Cells outside this gate were considered to be monocytes (SSC-Hmedium/CD45 PerCPintermediate+ cells) and granulocytes (SSC-Hhigh/CD45low+; Fig. 1A). Once the lymphocyte gate was established, percent double-positive CD3+/CD4+ T-lymphocyte in the upper right quadrant of a two-parameter dot-plot (Fig. 1B) was then automatically generated. For absolute CD4+ T-lymphocyte count, percent CD4+ T-lymphocyte is then multiplied by absolute lymphocyte count obtained from the hematology analyzer.
In the standard FACSCount absolute CD4 counting method, following introduction of a stained sample into the FACSCount system, CD3+ T-lymphocytes and reference beads in the threshold gate were automatically generated by the built-in FACSCount software (Fig. 1C). Two elliptical regions and one rectangular gate were automatically set around CD3+/CD4+ T-lymphocytes, reference beads, and CD3+/CD4- T-lymphocytes, respectively (Fig. 1D). Events were collected for up to 30,000 per sample. The absolute number of CD3+/CD4+ and CD3+ T lymphocytes was calculated automatically by multiplying the ratio of number CD3+/CD4+ T-lymphocyte events to that of beads by the known concentration of reference microbeads in the tube.
When a stained sample was acquired by the new FACSCount system, the software algorithm automatically identifies total lymphocytes, CD4+ T-lymphocytes, and reference beads (Fig. 2). The new CD4 reagent identifies lymphocytes by their DNA fluorescence and size while excluding nonlymphocytes (monocytes and granulocytes) from their CD14+/CD15+ expression. Subsequently, percent CD4+ T-lymphocyte was obtained automatically as a percentage of total lymphocytes. The absolute number of CD4+ T-lymphocytes was calculated automatically by multiplying the ratio of fluorescent CD4+ T-lymphocyte events to bright fluorescent microbead events by the known number of reference microbeads in the tube.
To ensure quality control of FCM instrument performance, the same lot of reagents was used throughout the study. FCM photomultiplier tube voltage, sensitivity, and fluorescence compensation settings were optimized prior to sample acquisition and analysis using Calibrite™ beads (BDB) and FACSCount controls (BDB) for FACScan and FACSCount, respectively.
Absolute CD4+ T-lymphocyte counts and percent CD4+ T-lymphocytes obtained by the new FACSCount system were compared to the other two predicate systems by linear regression analysis and coefficient of determination (r2) using StatView™ software (Brainpower, Calabasas, CA). Bland–Altman statistical bias method (14) was used to determine the level of agreement between the results obtained using the new system and those obtained by the other two systems. Bland–Altman statistics were created by plotting the average of CD4+ T-lymphocyte measurements obtained by the two systems [(a + b)/2] on the horizontal axis and differences between each pair of data measured by the two systems (a − b) on the vertical axis. In the plot (a) refers to the new system and (b) to the standard system. Mean percent difference between the two systems (the bias) and the limit of agreement (LOA; equivalent to the mean difference ± 1.96 SD) were then calculated. To examine possible differences and potential clinical impact of the percent CD4+ T-lymphocytes obtained by the new FACSCount system, significance tests for percent CD4+ T-lymphocytes in samples with less than 20% CD4+ T-lymphocytes were also determined.
Percent similarity between results obtained by the new FACSCount system and the other two predicate systems was also performed by taking the average between the new system and each predicate system divided by the predicate system and multiplying by 100 (15). Coefficient of variation (CV) was calculated and used to define agreement between each method pair.
The mean percent CD4+ T-lymphocytes from all 149 HIV-infected blood samples obtained from FACSCount FCM using the new CD4 reagent and from DP FACScan FCM using TriTEST reagent is shown in Table 1. Both systems gave similar mean percent CD4+ T-lymphocyte values. When a correlation plot of percent CD4+ T-lymphocyte values obtained from these two systems was performed (Fig. 3A), the new CD4 reagent analyzed by FACSCount FCM correlated well with the standard DP FACScan method (r2 = 0.977, y = 0.995x − 0.25, P < 0.0001). Figure 3B shows Bland–Altman plot comparing percent CD4+ T-lymphocyte values from these two methods, giving a minimum overall bias of −0.36% and LOA from −2.52% to +1.80% (new FACSCount CD4 reagent vs. DP TriTEST/FACScan reagent), with a similarity of 101.36% and CV of 4.27% (Table 2). Bland–Altman analysis of the results of percent CD4+ T-lymphocytes, when subdivided into 0–20% and >20% groups, also showed low bias; the mean percentage of bias of the two systems was −0.44 (LOA from −2.27% to +1.39%; percent similarity of 101.91 with CV of 4.86%) and −0.18 (LOA from −2.86% to +2.51%; percent similarity of 100.33 with CV of 2.76%), respectively (Figs. 3C and 3D). These data indicate that the two systems yield comparable values with excellent agreement.
Table 1. Mean ± SD and Median of Percent and Absolute CD4+ T-Lymphocytes in 149 HIV-Infected Blood Samples Determined by the New FACSCount™, Standard FACSCount, and DP TriTEST™/FACScan™ Systems
Table 2. Comparison of Percent Similarity for Enumeration of CD4+ T-Lymphocytes from Data of Each Flow Cytometric Method Pair
Mean similarity (%)
CV, coefficient of variation.
New FACSCount system vs. DP TriTEST/FACScan system
New FACSCount system vs. DP TriTEST/FACScan system
New FACSCount system vs. DP TriTEST/FACScan system
New FACSCount system vs. DP TriTEST/FACScan system
New FACSCount system vs. Standard FACSCount system
For absolute CD4+ T-lymphocyte counts, the new FACSCount and the standard FACSCount systems showed similar results (Table 1). However, both FACSCount systems showed higher mean absolute CD4+ T-lymphocyte counts than those from DP TriTEST/FACScan system. When comparisons were made between the absolute CD4+ T-lymphocyte values obtained from the new FACSCount system with those from the standard FACSCount system and from the DP TriTEST/FACScan system, the overall correlation coefficient for absolute CD4+ T-lymphocyte values from the entire set of samples was highly significant (r2 = 0.986, y = 0.989x + 8.255, P < 0.0001 and r2 = 0.974, y = 1.061x − 5.143, P < 0.0001 respectively; Figs. 4A and 4B). Absolute CD4+ T-lymphocyte counts obtained from standard FACSCount CD4 system and DP TriTEST/FACScan system were also well correlated (r2 = 0.978, y = 1.058x − 8.37, P < 0.0001; Fig. 4C). Bland–Altman plots of the comparison between the new FACSCount and the standard FACSCount system gave a low bias of +3.39 cells/μl (LOA from −52.53 cells/μl to +59.31 cells/μl). In contrast, there was a higher bias of +20.12 cells/μl (LOA from −57.92 cells/μl to +98.14 cells/μl; similarity = 98.2% and CV = 5.28%) for the new FACSCount system and the DP TriTEST/FACScan system (Figs. 4D and 4E). Similarly, a higher bias was also obtained when the standard FACSCount system was compared with the DP TriTEST/FACScan system (mean bias of +15.63 cells/μl and LOA from −56.04 cells/μl to +87.23 cells/μl; Fig. 4F). Nevertheless, percent similarity was very high, with 99.48 (CV = 2.75%) and 98.71 (CV = 4.51%) when the new and the standard FACSCount system was compared to the DP TriTEST/FACScan system, respectively. These results indicate that the absolute CD4+ T-lymphocyte counts analyzed by the new FACSCount system are in good agreement agree with the two predicate systems.
For monitoring HIV infection in pediatric patients, percent CD4+ T-lymphocyte value is a more useful parameter as it shows less age-related variability than the absolute CD4+ T-lymphocyte count (5, 16). In 2002, WHO recommended percent CD4+ T-lymphocyte values to be used for decision making in initiating ART for HIV-1-infected infants and children under the age of 5 years (6). More recently, WHO has recommended initiation of ART for HIV-positive infants and children with WHO Pediatric Stage III and IV disease irrespective of CD4+T-lymphocyte count, and for those with WHO Pediatric Stage I and II disease with absolute or percent CD4+T-lymphocytes at the age-specific cutoff levels (≤11 months: < 1,500 cells/μl or < 25%; 12–35 months: <750 cells/μl or < 20%; 36–59 months: < 350 cells/μl or < 15%; ≥ 5 years: < 200 cells/μl or < 15%) (7). The recommendation for percent CD4+ T-lymphocyte determination in HIV-infected pediatric patients means that the current version of FACSCount absolute CD4+ T-lymphocyte count system is no longer useful. A new FACSCount percent CD4 reagent system capable of being performed in the current FACSCount FCM was developed to allow determination of percent CD4+ T-lymphocyte and thus provide valuable parameters for both pediatric and adult patients.
Validation of the new FACSCount CD4 reagent and software system showed excellent correlation with both the original standard FACSCount system and the predicate DP FACScan system for both percent and absolute CD4+ T-lymphocytes throughout the CD4+ T-lymphocyte range. The overall bias for percent CD4+ values was −0.36%, with 98–99% similarity when the new FACSCount system was compared with the DP method using TriTEST reagent. When one considers the situation of HIV-infected infants who need ART because their percent CD4+ T-lymphocyte values are less than 20% or 25% (7), the bias was only 0.44% or less, with a degree of similarity of the new FACSCount system and the DP method of 98%. While these biases may influence clinical decision making, they are unlikely to affect monitoring as long as the two methods are not used interchangeably. For absolute CD4+ T-lymphocyte counts, the new FACSCount CD4 reagent and software system also gave a good correlation with more than 98% similarity to those of the standard FACSCount system and the DP method. As in previous studies, there were higher biases when SP FACSCount system was compared with DP method (17–19). Systems using SP method require a high level of precision in the dispensing of reagents in all the pipetting steps (17, 20). The DP TriTEST/FACScan system generates absolute CD4+ T-lymphocytes by the multiplication of percent CD4 T-lymphocyte value with absolute white blood cell count and percent lymphocyte from the hematology analyzer. Given the similarity of percent CD4+ T-lymphocyte values between the two systems, the bias in the absolute CD4+ T-lymphocyte counts is due to the technical differences used to derive these absolute values (17, 19).
For enumeration of absolute CD4+ T-lymphocyte values, the new FACSCount system showed a very small bias of no more than 3.39 cells/μl when compared to the standard FACSCount system, but higher absolute CD4+ T-lymphocyte values (mean bias = +20.12 cells/μl) were observed when this new system was compared with the standard DP TriTEST/FACScan system. A similar bias of 15.63 cells/μl was also observed when the standard FACSCount system was compared with the standard DP TriTEST/FACScan system. Such biases have been observed in other studies comparing SP and DP methods (17–19). The tendency toward higher CD4+ T-lymphocyte counts in the bead-based system than with the DP method may be a reproducible characteristic of bead-based technology.
The new FACSCount CD4 system is a valid method for enumeration of both absolute and percent CD4+ T-lymphocytes when compared with both DP FCM method and standard FACSCount system. The method is a dedicated stand-alone system and yet is very simple to use. The new FACSCount CD4 software requires only one day of training. The software is user friendly and can easily replace any existing FACSCount software system, thus leading to savings in new instrument investment in laboratories that are already equipped with FACSCount system. Given the high numbers of FACSCount instruments in the resource-limited settings, the availability of the new FACSCount percent CD4 reagent and software system should allow better and more cost-effective monitoring of HIV infection in pediatric patients in such countries. Moreover, this new FACSCount system is capable of a high throughput of more than 100 blood samples per day, comparable with DP TriTEST/FACScan system or standard SP bead-based TriTEST method. Interestingly, this new FACSCount system has a higher throughput than the original FACSCount system as it requires a shorter incubation time (30 min vs. 60 min) and shorter sample running time (1.5–4 min vs. 3–8 min). Similar to the standard FACSCount system, the new FACSCount system employs minimal sample handling, lysis-free cell preparation, automated gating of fluorescent events using built-in calibrated reference microbeads, and error-code reporting (13). With the burden of high numbers of HIV-infected individuals, particularly in the resource-constrained countries, health care system is under increasing pressure to operate with more cost effectiveness, and it is difficult to justify having two separate instruments for CD4+ T-lymphocyte enumeration of adult and pediatric patients, given the perishable nature and relatively high cost of reagents. It is therefore more desirable and cost-effective to have this new FACSCount system that provides both absolute and percent CD4 + T-lymphocyte measurements.
In conclusion, the new FACSCount CD4 reagent in combination with new FACSCount software system compares favorably with established systems for enumeration of both absolute and percent CD4+ T-lymphocytes.
We thank Dr. Prapon Wilairat for assistance in the writing of the paper. K.P. is a Senior Research Scholar of the Thailand Research Fund.