Access to HAART for the developing world The next hurdle: affordable lab monitoring (part 1)

From 1990 to 1993 Tobias Rinke de Wit (1961) performed post-doctoral research as a molecular immunologist at the Armauer Hansen Research Institute (AHRI) in Addis Ababa, Ethiopia. This was primarily in the field of identifying antigens for a vaccine against leprosy. From 1994 to 2000 he was Laboratory Manager of the Ethio-Netherlands AIDS Research Project (ENARP) at the Ethiopian Health and Nutrition Research Institute (EHNRI). He established a state-of-the-art laboratory, initiated prospective cohort studies on the natural history of HIV/AIDS, performed sentinel surveillances and supervised PhD and MSc students in a multitude of research topics on AIDS and in the field of immunology and virology. Tobias Rinke de Wit was also actively involved in AIDS information campaigns by promoting African educational theatre groups, writing pop songs about AIDS, performing with Ethiopian musicians and producing the first ¡§soap¡¨ about HIV/AIDS on Ethiopian Television. Since June 2001 he is heading the Research & Development department of PharmAccess International (PAI), the foundation that aims to provide access to HAART to more people in developing countries. Tobias Rinke de Wit published 65 scientific papers in peer-reviewed journals. In this issue of IATEC Update, the recent developments in the field of affordable CD4 T-cell enumeration for HAART monitoring are discussed; the next issue will deal with viral load.

Recent developments in the field of HIV/AIDS therapy in resource-poor settings are spectacular and raise hope that Highly Active Anti-Retroviral Therapy (HAART) will soon become a reality for substantially more HIV-infected people. The very significant drawback to HAART is its cost; however, this impediment is quickly disappearing: prices of drugs have decreased considerably. The Accelerated Access Initiative (AAI), a collaborative effort of UNAIDS and the major pharmaceutical industries, has achieved important price reductions of branded HAART drugs (down to 5-10%). Manufacturers of generic drugs are continuously offering further price reductions, the latest breakthrough being the one provided by the Thai Government Pharmaceutical Organization (GPO), which will allow drugs to be sold at $0.92 per day. In March 2002, the World Health Organization (WHO) officially released a list of manufacturers of safe HAART drugs, approved for UN purchase, which is further strengthening the option of generic HAART drugs for resource-poor countries. Meanwhile, the multimillion Global Fund to fight AIDS, Tuberculosis and Malaria is processing its first round of funding applications, which will release substantial amounts for fighting HIV/AIDS.

Although these developments are promising, there is no room for complacency, since a substantial gap remains between ¡§solving the money issue¡¨ and actual large-scale implementation of HAART in the developing world. Moreover, as long as there are countries that have an annual budget of $2 per capita for health care, it remains questionable whether HAART that costs anything at all is a realistic option. It goes beyond the scope of this article to discuss in detail the multitude of issues that still hamper access to HAART for the developing world. Instead, this paper concentrates on one of the next hurdles to be taken: simplifying and reducing the costs of laboratory monitoring of HAART.

Today, the costs of HAART laboratory monitoring equal or even surpass the costs of treatment. Prices for measuring the most important HAART monitoring markers are high: $25 for a CD4 T-cell test and $150 for a viral load measurement are rather rule than exception in Africa. This is partly because the manufacturers of laboratory equipment and assays for HAART monitoring have not yet been pressured to reduce prices and partly because the market for these industries is too small, mostly since there are still very few people in developing countries actually receiving HAART and thus in need of lab monitoring. With small markets, there is obviously little interest to invest in the development of more affordable technology.

In light of this reality, various initiatives have been launched to come up with recommendations for affordable lab technologies for the developing world. The WHO organized technical working groups on diagnostic support of HAART in August and November 2001 and in February 2002, specifically on CD4 T-cell enumeration. A WHO working group on viral load determination is planned for September 2002. Simultaneously, there was an initiative on these topics by the Gay Men¡¦s Health Crisis and Project Inform, sponsored by Rockefeller Foundation and NIH (November 2001, Bethesda, USA). A follow-up is planned by the Forum for Collaborative HIV Research (April 2002, Washington, USA). This paper summarizes the most important observations.

In general, it should be mentioned that the absence of laboratory facilities should not be a reason to refrain from implementing HAART in a resource-poor setting. In principle, good clinical follow-up is sufficient, provided it is combined with an extra effort in guaranteeing compliance, for example through direct observed therapy (DOT), as applied in tuberculosis control programs. It should be emphasized that HAART does not cure the patient and constant alertness is required from both the patient (compliance) and the clinician (monitoring) to avoid sub-optimal use of the medication from resulting in potentially catastrophic, massive drug resistance in the developing world.

However, if a laboratory is at hand, HAART monitoring should ideally consist of measuring hematology, white blood cell count (WBC), blood chemistry, CD4 T-cells and virology at regular intervals. In addition, some incidental tests should be performed at enrollment or if indicated, like a pregnancy test, hepatitis B/C serology, or an X-ray for tuberculosis. Hematology and WBC is quite often possible in resource-poor settings, since this minimally requires a microscope or a small hematocrit centrifuge. Quality control could be organized through central laboratories, which quite often are equipped with automatic hematological counters. Blood chemistry is a bit more expensive, since it needs a spectrophotometer, which involves maintenance and expensive kits with reagents that often need ¡§cold chains¡¨ during transport. However, quality control of blood chemistry can be done and in general hematology and some essential blood chemistry (liver enzymes) can be performed for less than $5 per patient.

Quantification of CD4 T-cells is especially important to help clinicians in making a decision on the initiation of HAART. Although there is no consensus for application in the developing world, the start of HAART is usually recommended when CD4 T-cell counts are between 200-350 per microliter (ƒÝl) whole blood. The cut-off point for starting HAART is an important issue, not only because it has implications for the incidence of opportunistic infections, but also regarding costs. Preliminary data suggest that, as a rule of thumb, 50% of all HIV-infected patients qualify for HAART when 350 cells/ƒÝl is taken as a cut-off, but only 25% qualifies with the lower cut-off (200/ƒÝl).

The most affordable way of quantifying CD4 T-cells is by manual technologies, using synthetic beads and microscopy for read-out. Currently, the most efficient methods for low-volume applications are Dynabeads and Cytospheres. Dynabeads cost $3 per test and need an investment of $750 for a magnet and mixer. The original technology needs an additional expensive immuno-fluorescence microscope ($7500), but there are ways to get around this, either by using alternative staining techniques that are compatible with light microscopy or purchasing a device, which converts a light microscope into a fluorescent microscope ($1800). Cytospheres are easier to use but more expensive ($9/test), although investments are low: only a light microscope and a hemacytometer. Both Dynabeads and Cytospheres are widely used in laboratories in the developing world. However, there are important disadvantages when using these techniques. First, microscope reading of the results is difficult, needs trained technicians and is prone to subjective interpretation. As a result of this, only limited numbers of blood samples can be tested on a daily basis (5-10 tests per technician), as there is a serious ¡§fatigue factor¡¨ associated with such manual technology. Both tests have some systemic bias, either underestimation (Dynabeads) or overestimation (Cytospheres) of the actual number of CD4 T-cells in whole blood (up to 30%). It is hoped that the systemic bias can be resolved with a correction factor. However, this correction factor will probably be unique to each specific laboratory and needs regular validation. There is some encouraging news about a stabilized blood product helping with this challenge. Also, both Dynabeads and Cytospheres need fresh blood for proper performance, which significantly limits the action radius of a lab, since no blood preservatives can be used. In the absence of convincing comparative data, the current WHO standpoint is that both methods should be evaluated in a multi-center study, before they can be officially recommended to labs in the developing world. All other manual CD4 technologies have been declared obsolete (TRAx, Capcellia, Zymmune, Immunoalkaline phosphatase assay).

The gold standard for CD4 enumeration remains flow cytometry. The easiest instrument to use is a single-platform volumetric flow cytometer, like the B&D FACScount. However, this equipment is very expensive ($35.000) and the maintenance contracts ($6000/year) and reagents ($12/test, cold-chain dependent) are not affordable for most settings. There are other flow cytometers, like the B&D FACScan and FACScalibur, the Coulter XL and the Ortho Cytoron, which are even more expensive and either need an additional blood cell counter (dual platform technology) or very expensive beads for CD4 T-cell quantification.

Alternative flow cytometry-based CD4 T-cell enumeration methods are all in various stages of development. First of all, there is the possibility of using simpler protocols on already existing flow cytometers. These protocols use innovative ways of gating (CD45 pan-leukogating, CD4-gating: see the very interesting site http://www.affordcd4.com) and as a consequence need only one or two generic monoclonal antibodies, which are substantially less expensive than commercial kits. However, these protocols cannot be performed on a FACScount, which excludes the far majority of those privileged labs in Africa that do have a flow cytometer. Currently, these pan-leukogating and CD4-gating technologies are under multi-center evaluation.

Another alternative is the use of flow cytometers with low energy red-diode lasers. These machines are smaller, less expensive and need considerably less electrical power (a car battery of 12V is sufficient). A new instrument has been launched by Partec (the CyFlow), which is basically a half-price FACScount with half-price kits, but needs multi-center validation before it can be recommended. The alternative approach of ¡§undressing¡¨ the complex Luminex 100 to a simpler red diode laser-based cytometer has met little enthusiasm from industry, even though multiplex technology will be the future in laboratory diagnostics.

The advantages of using flow cytometry for CD4 T-cell enumeration are clear: the automated methodology is easier to use, more reproducible, statistically more reliable and features a higher throughput (up to 300 samples/day). Using pan-leukogating, blood samples of more than one week old can be analyzed and blood fixatives (like the revolutionary Transfix) can be applied without problems significantly increasing the action radius of central laboratories. Data can be stored on computers, reanalyzed and a good quality control system for flow cytometry is in place, including the NEQAS standards with 200, 400 and 800 CD4 cells/ƒÝl blood.

Finally, there are developments on non-flow cytometry-based devices, like the Easy Count (developed at Twente University) based on magnetic sorting and optical image analysis of CD4 T-cells. Major advantages are: compact stand-alone, volumetric device, short sample measurement time, good linearity, low power consumption (10W, can be run on a laptop battery) and potentially low costs. However, the instrument still has a long way to go: it is in its ¡§proof of concept¡¨ phase and a second prototype is just being constructed.

In conclusion, the FACScount is unfortunately the only option that currently can be recommended for CD4 T-cell enumeration. Dynabeads, Cytospheres, the CyFlow and even alternative technologies like pan-leukogating need multi-center evaluation. The most important drawback of the FACScount remains the price. One way of dealing with this is to reduce the frequency of CD4 determination during HAART monitoring to 3x or even 2x per year (usually recommended at 4x per year). Another quite frequently used option is to dilute the commercial reagents 1-2 times. However, the best approach would probably be to organize direct negotiations between the Global Fund and the manufacturers of the FACScount, to come up with a mega deal for the developing world, including an unlimited supply of affordable reagents.