Bi, Xiuqiong*†; Gatanaga, Hiroyuki*; Tanaka, Mari*; Honda, Miwako*; Ida, Setsuko*; Kimura, Satoshi*; Oka, Shinichi*
From the *AIDS Clinical Center, International Medical Center of Japan, Tokyo, Japan; and †Graduate School of Medicine, University of Tokyo, Tokyo, Japan.
Received for publication May 31, 2004; accepted September 14, 2004.
Supported in part by a grant-in-aid for AIDS Research from the Ministry of Health, Labor, and Welfare of Japan (H15-AIDS-001, H15-International Medical Cooperative Study-03), by the Organization of Pharmaceutical Safety and Research (01-4), and by the Japanese Foundation for AIDS Prevention (X.B.).
Reprints: Shinichi Oka, AIDS Clinical Center, International Medical Center of Japan, 1-21-1 Toyama, Shinjuku-ku, Tokyo 162-8655, Japan (e-mail: email@example.com).
The CD4+ T-lymphocyte count (CD4 count) is an important surrogate marker for the clinical course of HIV infection, such as initiation of prophylactic treatment of opportunistic infections, initiation of antiretroviral therapy (ART), and monitoring the response to ART.1-4 In developed countries, the CD4 count is usually measured by flow cytometry, which is considered to be the standard reference method.3,4 In resource-limited areas, however, flow cytometry is available only in limited settings such as tertiary medical centers because it requires expensive reagents and well-trained technicians. Furthermore, equipment maintenance is another difficult issue, because a technical support system is needed in areas afflicted with frequent electrical power failures, which could potentially cause machine-related problems.
In recent years, lower cost and less technically demanding methods for enumerating CD4 cells have been tried but have not been used widely even in resource-limited settings for various reasons.4,5 In the World Health Organization (WHO) guidelines for treatment of HIV-infected individuals in resource-limited environments, a total lymphocyte count (TLC) of 1200 cells/μL is recommended to represent a CD4 count threshold of 200 cells/μL in making a decision regarding therapy when the CD4 count is unavailable.1 In addition, various research groups have recommended the use of a TLC,5 absolute lymphocyte count or TLC,6 and TLC combined with hemoglobin measurement7 as surrogate markers for monitoring ART. These studies suggested that the lymphocyte count might have some value in monitoring ART. The lymphocyte count is readily available and inexpensive, but it is not sufficiently adequate to predict the absolute CD4 count in many settings.4
Among several low-cost and less technically demanding methods,8-15 the Dynabeads assay, which uses magnetic particles coated with a monoclonal antibody to CD4 to capture CD4+ cells, seems to be a good candidate as an alternative to flow cytometry based on its good correlation with the results of flow cytometry.8,10,11,13 According to the protocol recommended by the manufacturer, however, CD4 and CD8 cells are enumerated at the same time using a large volume of Dynabeads. The large volume of Dynabeads used in each assay is also relatively expensive (approximately $5), particularly for poor settings. In addition, division of the samples into 2 aliquots during the procedure might jeopardize the accuracy of the results. Moreover, in this assay, the cells are lysed and nuclei are stained to count them, which makes the operation complex.
For monitoring ART in HIV infections, only the CD4 component is necessary and only the CD4 count (not CD8 count) is mentioned in ART guidelines.1,2 For this reason, further modifications are needed for the expanded use of the CD4 count in resource-limited areas. In the present study, we modified the protocol to make it simple and inexpensive so that it could be applied widely in resource-limited facilities.
MATERIALS AND METHODS
Study Population and CD4 Enumeration
This study included 242 adult patients infected with HIV-1 who regularly consulted the AIDS Clinical Center of the International Medical Center of Japan between June and October 2003. The inclusion criteria were a CD4 count less than 1000 cells/μL and consent granted to participate in the study. Patient age ranged from 20 to 78 years (mean ± SD: 40 ± 11.5). A total of 315 blood samples were collected using EDTA-containing tubes and tested for CD4 count within 4 hours by flow cytometry (FlowcytoCD4; Coulter-EPICS XL-MCL, Beckman-Coulter, Fullerton, CA) with CD45-fluorescein isothiocyanate (FITC)/CD4-phycoerythrin (RD1)/CD8-pheycoerythrin-Texas Red (ECD)/CD3-pheycoerythrin-cyanin 5.1 (PC5) (Beckman-Coulter). The CD4 cell count in the rest of the blood sample was enumerated using Dynabeads (Dynabeads CD4; Dynal Biotech ASA, Oslo, Norway) within 24 hours. When different protocols such as 25-μL and 5-μL volumes of CD4 Dynabeads or 10 and 30 minutes of incubation time were compared, the same sample was used in each experiment.
Modified Protocol (Original Protocol) of the Dynabeads Method
A well-mixed whole-blood sample (125 μL) was placed into a 1.5-mL microtube containing 375 μL (350 μL) of buffer (0.1% bovine serum albumin in phosphate-buffered saline [PBS]). CD14 Dynabeads were suspended with buffer (1:1 diluted buffer); 5 μL (25 μL) of CD14 Dynabeads was then added to the microtube containing the blood sample and the tube was inverted several times and then incubated in Dynal MX-1 for 10 minutes. The tube was spun down in a microcentrifuge and then placed in magnetic particle concentration for microcentrifuge tubes (Dynal MPC-S) (6 tubes per batch) for 2 minutes, followed by transfer of the entire volume (division into 2 200-μL aliquots) of monocyte-depleted blood into a new microtube. In the next step, 5 μL (25 μL) of CD4 Dynabeads was added to the tube and incubated in Dynal MX-1 for 30 minutes (10 minutes). The cells were washed with 500 μL of buffer, vortexed gently, and spun down; the tube was then placed in Dynal MPC-S for 2 minutes, the wash buffer was discarded, and the tube was removed from the Dynal MPC-S. The cells were washed 3 more times (once), resuspended by adding 125 μL of buffer, and kept at 4°C until counting (50 μL of lysing solution was added, followed by thorough vortexing for resuspension, and the cells were allowed to stand for 5 minutes, after which 50 μL of acridine orange staining solution was added and the sample was kept in darkness until counting). Finally, the sample was vortexed well, 10 μL of cells was applied to a hemocytometer, and the mononuclear cells with attached Dynabeads were counted as CD4+ cells under a light microscope (the number of nuclei was determined under a fluorescence microscope). After reducing the volume of CD4 Dynabeads, it was not difficult to count the CD4 cells under an optical microscope, even without staining. All procedures were performed at room temperature at approximately 23°C. All Dynabeads-related equipments and reagents were products of Dynal Biotech ASA.
All data are expressed as mean ± SD. StatView v5.0 software was used to analyze the correlation and single linear regression between DynabeadsCD4 and FlowcytoCD4. P values were calculated by 2-sided test and considered as significant if at a level less than 5%. All confidence intervals were 2-sided, with a significant level of 5%.
First, we examined the influence of a reduced volume of CD14 (from 12.5 μL to 5 μL) Dynabeads on monocyte depletion. The percentage of monocytes in 5 blood samples was analyzed by flow cytometry before and after treatment with 5 μL of CD14 Dynabeads. The result showed that 5 μL of CD14 Dynabeads deleted 92.4% to 97.5% (average = 95.6%) of monocytes from 125 μL of whole blood. The remaining experiments were performed using 5 μL of CD14 Dynabeads. Next, we examined the influence of a reduced volume of CD4 Dynabeads on the CD4 count in 23 samples. The volume of CD4 Dynabeads was reduced from 25 μL to 5 μL, but the incubation time was still 10 minutes (like that of original protocol), which we called modified protocol 1. CD4 counts by the original protocol and modified protocol 1 correlated significantly with those determined by flow cytometry: DynabeadsCD4 by the original protocol (r = 0.90 [P < 0.0001]; slope = 1.05, intercept = −32) and DynabeadsCD4 by modified protocol 1 (r = 0.92 [P < 0.0001]; slope = 1.05, intercept = 26). These results indicated that DynabeadsCD4 obtained by using the reduced volume of CD4 Dynabeads with a 10-minute CD4 separation correlated well with FlowcytoCD4. When the number of samples was increased to 56, however, the mean DynabeadsCD4 of 56 samples by modified protocol 1 was 269 ± 140 cells/μL compared with a mean FlowcytoCD4 of 336 ± 178 cells/μL (Table 1). The difference was −67 cells/μL (P < 0.0001). This result suggested that the 10-minute CD4 separation time was too short. We then examined the effect using a reduced volume of Dynabeads and a different incubation time.
Next, with 5 μL of CD4 Dynabeads, we lengthened the CD4 separation time from 10 minutes to 30 minutes in 34 samples. The correlations between DynabeadsCD4 and FlowcytoCD4 were r = 0.91 (P < 0.0001) and r = 0.94 (P < 0.0001), with slopes of 1.05 and 1.0 and intercepts of 22 and 8, for 10 and 30 minutes of incubation time, respectively. The mean difference with flowcytoCD4 was −32 cells/μL (P = 0.008) and −8 cells/μL (P = 0.42), respectively. According to these data, the 30-minute incubation time for CD4 separation yielded a better result than that of the 10-minute incubation time. We then fixed the protocol as 5 μL of CD14 Dynabeads with 10 minutes of incubation time and 5 μL of CD4 Dynabeads with 30 minutes incubation time (which we called modified protocol 2) and tested 246 samples. DynabeadsCD4 showed a significant correlation with FlowcytoCD4 (r = 0.91 [P < 0.0001]; slope = 1.03, intercept = −16; Fig. 1). At less than 200 cells/μL, the sensitivity and specificity of DynabeadsCD4 compared with FlowcytoCD4 were 79% and 94%, respectively, and at less than 350 cells/μL, the sensitivity and specificity were 95% and 88%, respectively. The mean DynabeadsCD4 was 262 ± 135 cells/μL and that of FlowcytoCD4 was 254 ± 154 cells/μL (see Table 1). The difference in the mean values was 8 cells/μL (95% confidence interval [CI]: 0.4-16; P = 0.04), with a random error of 64 cells/μL. The positive and negative predictive values of DynabeadsCD4 and FlowcytoCD4 for less than 200 cells/μL and less than 350 cells/μL were 90% and 87% and 97% and 83%, respectively. Other factors (eg, on therapy vs. off therapy, male vs. female) had no influence on DynabeadsCD4 (data not shown).
Table 2 shows the results of a comparison between the original protocol and our modified protocol. In our modified protocol, volumes of CD14 and CD4 Dynabeads were reduced from 12.5 μL and 25 μL, respectively, to 5 μL each against 125 μL of whole blood. Accordingly, the cost of the Dynabeads test decreased from $2.84 to $0.89. The incubation time for CD4 separation was prolonged to 30 minutes to obtain a better yield. In our protocol, after monocyte depletion, we transferred all treated blood to a new microtube for CD4 cell separation because we did not consider the CD8 count. We also skipped over lysis and nuclear staining steps so as to simplify the procedure.
To attain the "3 by 5" goal of effective ART promoted by the WHO, precise monitoring of ART is indispensable. Low cost, in addition to good accuracy, is thus an important issue. In this regard, maintenance of a "high-tech" machine for long-term monitoring may be impossible. The Dynabeads method is currently used as an alternative method to flow cytometry for CD4 count in a number of countries. In this study, we successfully modified the protocol of the Dynabeads method to make it more suitable in resource-limited areas with 2 goals in mind: reasonable cost and sufficient accuracy.
In the present study, DynabeadsCD4 obtained by using the original protocol also showed a good result (see Table 1). During the operation, we found 2 problems with the original protocol, however. One was the transfer of 200 μL of blood from 500 μL of blood to a new tube after monocyte depletion. This step might lead to inaccurate results because we could not mix the blood well while the tube was on the Dynal MPC-S. The other was that too many free Dynabeads (which did not attach to CD4 cells) and red blood cells were identified when the cells were counted under a light microscope. This might be the reason for recommending lysis of the cells, staining the nuclei, and using a fluorescent microscope in the last step of the original protocol. In our modified protocol, the entire sample was transferred to a new tube after monocyte depletion. The number of free Dynabeads decreased after the volume of CD4 Dynabeads was reduced. Furthermore, we washed the sample 4 times after CD4 cell separation in spite of the original protocol recommending washing only twice. The red blood cells could be almost completely removed by 4 washes, especially when the washing buffer had been discarded completely at each wash. These modifications made a direct count under a light microscope possible.
After reduction of the volume of Dynabeads used in the assay, the cost of reagents used for analysis of 1 sample decreased to less than $1.00. Thus, the total cost of 1 CD4 count, including other disposable materials such as syringes, tubes, and tips, could be less than $3.00.
In conclusion, the present study demonstrated that our final modified protocol of Dynabeads assay could be used as a good alternative to flow cytometry with sufficient accuracy, reliability, and simplicity at a reasonable cost. Therefore, the assay could be suitable for monitoring ART in resource-limited settings.
The authors thank Naomi Wakasugi (International Medical Center of Japan) and Kenji Tamura (WHO) for their helpful suggestions and encouragement during the study.
1. World Health Organization. Scaling up antiretroviral therapy in resource limited settings: guidelines for a public health approach. Available at: http://www.who.int/hiv/
. Accessed April 2002.
2. Department of Health and Human Services. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Available at: http://AIDsinfo.nih.gov
. Accessed July 14, 2003.
3. Brando B, Barnett D, Janossy G, et al. Cytofluorimetric methods for assessing absolute numbers of cell subsets in blood. Cytometry.
4. Crowe S, Turnbull S, Oelrichs R, et al. Monitoring of human immunodeficiency virus infection in resource-constrained countries. Clin Infect Dis.
5. Badri M, Wood R. Usefulness of total lymphocyte count in monitoring highly active antiretroviral therapy in resource-limited settings. AIDS.
6. Jacobson AM, Liu L, Khayam-Bashi H, et al. Absolute or total lymphocyte count as a marker for the CD4 T lymphocyte criterion for initiating antiretroviral therapy. AIDS.
7. Spacek AL, Griswold M, Quinn CT, et al. Total lymphocyte count and hemoglobin combined in an algorithm to initiate the use of highly active antiretroviral therapy in resource-limited settings. AIDS.
8. Diagbouga S, Chazallon C, Kazatchkine MD, et al. Successful implementation of a low-cost method for enumerating CD4+ T lymphocytes in resource-limited settings: the ANRS 12-26 study. AIDS.
9. Brinchmann JE, Leivestad T, Vartdal F. Quantification of lymphocyte subsets based on cells directly from blood. J Immunogenet.
10. Didier J, Kazatchkine M, Demouchy C, et al. Comparative assessment of five alternative methods for CD4 T-lymphocyte enumeration for implementation in developing countries. J Acquir Immune Defic Syndr.
11. Lyamuya EF, Kagoma C, Mbena EC, et al. Evaluation of the FACScount, TRAx CD4 and Dynabeads methods for CD4 lymphocyte determination. J Immunol Methods.
12. Nicholson JK, Velleca W, Jubert S, et al. Evaluation of alternative CD4 technologies for the enumeration of CD4 lymphocyte. J Immunol Methods.
13. Lyamuya EF, Schechter M, Echeverria de Perez G, et al. Multicentre evaluation of alternative methods for CD4 lymphocyte determination [abstract Mo.A.513]. Presented at the International AIDS Conference, Vancouver, July 1996.
14. Carella AV, Moss MW, Provost V, et al. A manual bead assay for the determination of absolute CD4+ and CD8+ lymphocyte counts in human immunodeficiency virus-infected individuals. Clin Diagn Lab Immunol.
15. Gaudernack G, Lundin KE. Rapid immunomagnetic phenotyping of cells. J Immunogenet.
© 2005 Lippincott Williams & Wilkins, Inc.