EasyCD4 low cost CD4 cell assay as accurate as current gold standard

A cheap, easy-to-perform method of counting CD4 T-cells is just as accurate as the current gold standard test, according to a study conducted in the United States and Uganda, published in the April 15^th edition of the Journal of Acquired Immune Deficiency Syndromes. The investigators believe that the low-cost Guava EasyCD4 assay will therefore be of particular value and importance to HIV care providers in resource-limited settings.

CD4 cell count measurements are very important for HIV treatment and care. They are used to make decisions on clinical diagnosis, the use of preventive treatment against opportunistic infections and the initiation of anti-HIV treatment.

The current gold standard for measuring CD4 cell counts is called flow cytometry, which is expensive and requires specialised equipment and staff. As the provision of antiretroviral therapy expands in resource-limited countries, cheaper methods of measuring CD4 cell counts are needed, ideally using technologies that can be successfully used in these areas.

Guava EasyCD4 is a new method of measuring CD4 cell counts that uses a technology called microcapillary cytometry. At a cost of only $1 per test it is significantly cheaper than the $25 per count involved in the current testing method. In addition, EasyCD4 has other potential advantages, requiring a smaller volume of blood and generating a lower amount of waste material. The cost for EasyCD4 testing equipment is $35,000, similar to the cost for flow cytometry equipment.

Investigators evaluated the precision of the EasyCD4 assay in the United States and Uganda, comparing CD4 cell counts with those obtained using the flow cytometry.

In October 2004, blood samples were obtained from 77 individuals receiving HIV care at John Hopkins University in Baltimore and in February 2005 samples were obtained from 142 HIV-positive patients attending the Infectious Diseases Institute in Kampala. Flow cytometry was conducted on these samples within 24 hours and EasyCD4 within 48 hours.

Mean CD4 cell count amongst patients in the United States was 697 cells/mm³ using flow cytometry and 688 cells/mm³ using EasyCD4. For patients in Uganda, the mean CD4 cell count was 335 cells/mm³ by flow cytometry and 340 cells/mm³ using EasyCD4. These differences were not statistically significant.

Because mean CD4 cell count was higher amongst patients in the United States than those in Uganda, the investigators performed a subanalysis restricted to United States patients with a CD4 cell count below 500 cells/mm³. Once again, no significant difference was seen between results obtained using standard flow cytometry and EasyCD4.

The investigators then looked at the ability of EasyCD4 to predict a CD4 cell count below 200 cells/mm³, the level at which antiretroviral therapy should be started according to World Health Organization guidelines. In samples obtained from Ugandan patients, 59 (42%) were below 200 cells/mm³. The EasyCD4 cell count had a sensitivity of 90% and a specificity of 98%. The positive predictive value was 96% and the negative predictive value was 98%. Flow cytometry and EasyCD4 agreed on 94% of the samples.

Only a small number of patients in the United States (9; 12%) had a CD4 cell count below 200 cells/mm³. The EasyCD4 assay was shown to have a sensitivity of 89% and a specificity of 99%, with a positive predictive value of 89% and a negative predictive value of 99%. The two testing assays agreed on 97% of the United States samples with a low CD4 cell count.

“Because access to antiretroviral therapy increases world-wide, inexpensive laboratory tests are need to monitor responses to therapy”, write the investigators, adding, “in identifying eligibility for therapy at CD4 cell counts below 200 cells/mm³, CD4 cell counts by flow cytometry and EasyCD4 were very comparable.”

They conclude, “our results suggest that with training of laboratory technicians and appropriate technical support, the EasyCD4 assay may be used with high positive and negative predictive value in resource-limited settings.”