How best to deploy point-of-care CD4 cell testing in resource-limited settings?

Widespread introduction of point-of-care (POC) CD4 cell tests that don’t require a laboratory technician but can be performed onsite by a nurse, providing results in less than an hour – while the patient waits) is expected in the next 12 to 24 months, said Dr Steven Reid of Imperial College, London, during a session on advances in diagnostics on Monday at the International AIDS Society conference (IAS 2011) in Rome, Italy.

In fact, one such test is already being rolled out in some settings. Speakers during the session described how POC CD4 tests might improve service delivery for people living with HIV, their potential cost-effectiveness and where best to deploy them.

For instance, the availability of POC CD4 cell tests – paired with a mobile HIV counselling and testing service – significantly increased linkage to care for people testing positive for HIV, according to one study.

Dr Reid presented a modelling study suggesting that starting treatment on the basis of POC results would be more cost-effective than with traditional CD4 cell tests, and better than initiating treatment on the basis of clinical (syndromic) management of people living with HIV — potentially adding years to their lives.

However, another study, presented by Dr Ilesh V Jani of the Instituto Nacional de Saúde reported that the cost of implementing POC CD4 would depend upon the clinic and volume of tests it would perform – and that there should be “a more judicial deployment of the technology", prioritising “higher volume sites and clinics that cannot refer samples to laboratories”.

CD4 cell tests

CD4 depletion is the hallmark of acquired immune deficiency, and CD4 cell measurements are the best indication of when a person should start antiretroviral treatment – preferably before there is a seriously increased risk of opportunistic infections and tuberculosis. The World Health Organization’s 2010 antiretroviral treatment (ART) guidelines recommend that treatment begin at a CD4 count of 350 cells/mm³ – although not all countries follow this recommendation yet.

"The aim was to develop tests which require limited or no infrastructure – no electronics, simple to use, and cheap.” Dr Steven Reid

But in a number of resource-constrained settings, it is difficult to do CD4 cell measurements to start treatment – and these have instead relied on clinical staging (based upon symptoms and signs of immune deficiency). But a number of studies have shown that clinical staging is inadequate since it misses large numbers of people with very low CD4 cell counts, who are at great risk of catastrophic illnesses.

The gold standard for CD4 measurement is flow cytometry (Becton Dickinson or Beckman), but this equipment is relatively complex and it requires skilled laboratory technicians to perform. Many resource-poor settings simply don’t have the infrastructure, human resources or funds to provide convenient access to CD4 cell testing in more remote or rural settings. People either have to be referred to a site that provides CD4 cell testing, or have their specimens sent to the site and return to the clinic to get their results.

Many people simply don’t bother because of time, distance or money, and are often lost to follow-up, until they either fall ill, and come in for care again, or die. Researchers have therefore been working to make CD4 cell testing cheaper, more accessible and easier to use.

“The CD4 Initiative was established in 2005 to develop rapid, economical, point-of-care tests for CD4,” said Dr Reid. “The aim was to develop tests which require limited or no infrastructure – no electronics, simple to use, and cheap.”

Such CD4 cell results could then be introduced into the most remote rural settings to put people onto ART on the basis of CD4 counts rather than clinical staging. The new tests would also facilitate the delivery of ART in decentralised settings (including the primary health care clinic) and possibly improve retention into ART programmes, by speeding the time it takes between when a person gets tested and when he or she goes onto ART, and by reducing loss to follow-up.

The new generation of CD4 cell tests

Dr Reid listed three of the leading POC CD4 products.

Alere’s PIMA is already in use (in both the mobile clinic study and in Mozambique). It consists of a disposable CD4 cartridge containing sealed reagents and a portable analyser to produce a CD4 cell test from either a fingerstick or venous whole blood sample within 20 minutes.

A finger prick is made with a sterile lancet to collect 25μL of capillary whole blood into the cartridge, without manual sample handling or processing. The reagents (all dried and requiring no refrigeration) are sealed inside the cartridge. The cartridge is then inserted into the battery (or A/C) powered portable analyser, which automatically begins the testing process, providing a direct CD4 measurement in 20 minutes.

Daktari CD4 is another POC test, that requires no pipetting, labels or reagents. Its cartridge is then placed into an analyser that reads electrical signals, and reports the CD4 count within minutes, according to the manufacturer’s site.

The third POC test, from CD4 Initiative working with Zyomyx, Inc., is a single disposable unit that can deliver an absolute CD4 count without complex electronics and instrumentation. It is the first readerless point-of-care, that is, without an analyzer. Instead, the healthcare worker reads the CD4 cell count by a simple visual inspection of the test tube-like unit.

Modelling

Despite the promise of POC CD4 cell tests, “the impact of introduction is not obvious”, said Dr Reid. So he adapted a published model of ART initiation, to compare the impact on life-years saved (LYS) of syndromic management and the two CD4 counting strategies (flow cytometry and POC CD4 tests), adding in costs for the CD4 cell monitoring technologies. He adjusted a number of parameters in the model (based upon the POC test’s expected impact on retention), and used two different CD4 cell initiation thresholds (250 and 350 cells/mm³) and different estimates for the CD4 cell costs.

He calculated high and low costs for the CD4 cell technologies based upon costs of reagents and the test price, the staffing, personnel costs needed to perform the test, infrastructure, laboratory costs (when needed), the overhead for hospital or laboratory (if needed), referring to data on costs from Zimbabwe and Uganda. The unit costs were higher on the POC CD4 test, but reagents and other expenses made the centralised flow cytometry test more expensive overall.

In the model, initiation based upon syndromatic management clearly cost the least (almost half of POC), but has the poorest outcomes in terms of life-years saved. Initiation on the basis of flow cytometry and POC CD4 would have similar overall costs, but POC CD4 cell tests would result in more life-years saved and thus be more cost-effective.

However, these results are entirely dependent upon the cost calculations and assumptions about how the availability of point-of-care testing would affect retention and the timely initiation of ART.  Other reports presented at the conference were conflicting on some of these points.

Increased linkage to care

For instance, there was a poster presentation of data from Themba Lethu Clinic in Johannesburg, previously reported at the 5^th South African AIDS Conference, showing that having a POC CD4 test available didn’t dramatically shorten the time between testing positive and going onto ART. The lack of impact was primarily due to the fact that 39% of the people rejected the offer of POC CD4.

However, another study from the same researchers in Johannesburg, presented by Dr Bruce Larson, reported much greater acceptance – 90% – of the Pima POC CD4 cell test when it was piloted to a few hundred people testing positive at a mobile HIV counselling and testing service. The study involved phone interviews of people who had been offered the test versus those who had not. There was a 26% increase in referral uptake (those making their first visit to the referral site within eight weeks of testing positive).

“The increase in referral completion went from 38.5% to 64.7% (a 68% improvement),” said Dr Larson.

He said the difference in the uptake of the offer for POC CD4 between the mobile clinic and Themba Lethu clinic was puzzling, but he noted that the Pima test provided results within about 20 minutes, while the test used at Themba Lethu took much longer (closer to an hour). “Once it was explained how long it would take, many of the patients at Themba Lethu may simply not have wanted to wait around that long,” he said.

Likewise, Dr Jani mentioned similar data from a study he conducted in Mozambique, which found that nurses in primary health clinics can accurately perform CD4 counting using point-of-care devices. Comparing percentages of patients getting and returning to the clinic at baseline to post-piloting of the POC CD4, there was a clear reduction in pre-treatment loss to follow-up. At baseline, only 55.2% got their CD4 cell results, compared to 92.9% afterwards, while only 28.4% returned to the clinic for care, versus 79.4% after the introduction of POC CD4.

Costs depends on throughput needed — deploy carefully

However, Dr Jani presented data on indicating that the cost of implementing POC CD4 could be significant, and suggested that programmes may want to consider carefully where they want to install POC CD4 cell monitoring.

The traditional CD4 test has several things going for it: a higher throughput (50 to 75 tests per day), a large installed base (>1000 instruments across Africa), already existing infrastructure in some settings, and it may be more efficient and cheaper due to economy of scale. POC CD4 however can only perform 5 to 20 tests per day, and is new technology – although it does not require significant infrastructure and can be performed by non-specialised personnel.

He developed a different costing model based upon data gathered from health facilities across 13 countries in sub-Saharan Africa, looking at reagent, control, and consumable costs, equipment and maintenance costs, lab infrastructure and overhead costs, sample transport costs, human resource salaries, and site patient volumes.

These data were used to calculate a total cost per test for a site with a known testing volume to determine if it would be less costly to refer samples to an existing CD4 laboratory or to implement POC CD4 testing on-site (with the PIMA test).

The cost of conventional CD4 cell testing actually varied significantly by site (due to reagent costs), while POC CD4 costs were more stable. At sites with average throughput, the costs of testing were similar: $10.50 for the conventional CD4 and $11.78 for the POC CD4. However, for increasing volume, the cost per test for POC CD4 drops dramatically (though it will increase again if more than 5000 tests are performed a year — because a second analyser would be needed).

The cost of POC CD4 versus conventional CD4 is about equal if ~2,900 tests per year are performed by a site.

“Sites above 2900 tests per year comprise more than 90% of CD4 test demand, but it is more intuitive that POC would belong at small, remote sites,” said Dr. Jani.

But even though it may be less costly to put the POC CD4 tests at a number of higher volume sites, Dr Jani stressed site selection should depend on other factors than simply cost, including universal access, equity, distance to laboratories, patient loss-to-follow-up, size of catchment area, ART coverage, HIV prevalence, and PMTCT services.