MICA: Performance evaluation of a point-of-care whole blood viral load test (SAMBA) to optimize HIV treatment in resource-limited settings

As of 2012, 9.7 million individuals were receiving antiretroviral therapy (ART) for treatment of HIV-1 infection in low- and middle-income countries. However, among patients receiving first-line ART, 20-30% will have a detectable viral load after 12 months. The long-term success of ART programmes is heavily dependent on accurate, available and timely viral load monitoring to document viral suppression or indicate the need for second line ART switch.

Existing HIV viral load assays require significant infrastructure and highly trained personnel to perform, thus only a few centralised laboratories are capable of viral load testing in most developing countries. Centralised testing leads to poor access to those in need, long turn-around times, delay of results and subsequent loss of patients, all of which lead to sub-optimal ART treatment outcomes, which can cause significant morbidity and mortality.

The University of Cambridge, in partnership with Diagnostics for the Real World, has developed an innovative point-of-care nucleic acid amplification test for the diagnosis monitoring the efficacy of HIV treatment by determining of the amount of circulating virus is above or below the threshold for treatment failure. The SAMBA HIV-1 Semi-Q Test allows the patient to be tested and treated in the same visit, thus improving access and treatment outcomes. The plasma-based semi-automated (SAMBA I) assay has received product approval in Malawi, Kenya, Zimbabwe and Uganda and has been used for testing of over 35,500 patients at Médecins sans Frontières (MSF) sites in Malawi and Uganda to date.

The SAMBA HIV-1 Semi-Q Test has recently been improved to enable the use of small volume whole blood, using a patented method of filtration. This removes the need for plasma separation, thus increasing utility at lower cadre health clinics and reduces the time to result. In addition a fully automated SAMBA instrument system has been developed, called SAMBA II. The key aims of this proposal are: 1) integrate the whole blood SAMBA HIV-1 Semi-Q Test onto the fully automated SAMBA II instrument, 2) perform an evaluation in house, in Africa and in Europe to evaluate its perforce compared to another CE-marked test and 3) submit for CE-mark. Although the SAMBA test will most likely be used in Africa, a CE-mark is required to allow it to be procured by NGO's.

Access to antiretroviral therapy (ART) for HIV-1 infected individuals in developing countries is expanding rapidly, improving survival and preventing transmission. Routine viral load (VL) monitoring of HIV-infected individuals on ART is necessary to assess treatment efficacy and adherence. Currently the monitoring of HIV-1 VL is performed largely in centralised laboratories using plasma. Therefore, patients living outside of the capital city in developing countries often cannot access VL testing and if they can there are regularly unacceptable delays in obtaining test results, with consequent loss to follow-up (LTFU). These issues could be overcome by point of care (POC) VL testing.
The Diagnostics Development Unit (DDU) and its spin out company, Diagnostics for the Real World (DRW), have developed the SAMBA HIV-1 Semi-Q Test for the diagnosis of ART failure (VL >1,000 copies/ml). The SAMBA HIV-1 Semi-Q Test uses plasma as has been used for VL testing of >37,500 patients at Médecins sans Frontières sites in Malawi and Uganda since August 2013 using the semi-automated CE-marked SAMBA I instrument. DDU have recently improved the SAMBA chemistry for use with small volume whole blood (WB) samples by incorporating a leucodepletion step to remove CD4+ cells containing integrated HIV RNA and DNA. This integrated DNA and RNA interferes with the VL quantification. A preliminary study assessed the feasibility of using venous WB on the SAMBA I Semi-Q Test. The leucodepleted WB had a 3-log reduction in CD4+ cells without significantly affecting plasma VL. The clinical performance of the assay was evaluated in Kenya compared to the Roche Cobas TaqMan HIV-1 test. A concordance of 96.5% was obtained between the Roche (plasma) and SAMBA (whole blood).
The key aims of this proposal are to integrate whole blood SAMBA chemistry onto the fully-automated CE-marked SAMBA II POC instrument, perform a prospective, cross-sectional evaluation in Africa and Europe and submit for CE-marking

The need for viral load testing at POC in resource-limited settings
As of 2014, an estimated 36.9 million people were living with HIV worldwide. Of those infected, approximately 12.9 million were receiving antiretroviral therapy (ART) at the end of 2013, 11.7 million of them in low- and middle-income countries. With an increase of 2 million people receiving ART, the growth in HIV treatment provision in 2013 was bigger than any previous year. However, only 36% of the estimated 32.6 million people living with HIV in low- and middle- income countries are receiving ART, when 85% are eligible.

Effective ART not only improves the survival of individuals infected with HIV but also prevents transmission. Routine monitoring of HIV-infected individuals on ART using viral-load (VL) testing is necessary to assess treatment efficacy and adherence, to identify treatment failure and initiate a timely switch to second-line therapy. Currently the monitoring of HIV-1 VL testing in resource-limited settings is performed largely in centralised laboratories by PCR-based methods using plasma or dry blood spots (DBS) samples. To access VL testing, peripheral HIV treatment facilities must transport patient specimens to central laboratories under optimal conditions within a limited period of time, followed by testing and return of results. This results in increased costs of service delivery and unacceptable delays in obtaining test results, with consequent losses to follow-up (LTFU). These issues could be overcome by POC VL testing.

Dried blood spots (DBS) have in many instances been employed as an alternative sample type to plasma to expand VL testing to peripheral sites. DBS are cheap, require minimal training for collection, can be stored easily, and are readily transportable, even in harsh environmental conditions. However, the appropriateness of DBS for VL testing has been questioned. Depending on the assay used, the proviral DNA present in whole blood DBS samples interferes with RNA quantification and RNA in DBS might not be sufficiently stable. In addition, DBS being tested centrally do little to improve LFTU rates.

The proposed solution for isolated sites in resource-limited settings is to conduct the testing at the patient's Point of Care (POC) with low throughput, high level of automation and limited requirements on staff competence. Although centralized testing is cost-effective, in areas with a high prevalence of HIV infections, decentralized testing is required to care for the greatest number of patients and to limit LTFU.12

SAMBA Whole Blood Semi-Q Test for POC VL testing
The Diagnostics Development Unit (DDU) and its spin out company, Diagnostics for the Real World (DRW), have developed the SAMBA HIV-1 Semi-Q Test. This is an innovative point-of-care (POC) nucleic acid amplification test (NAAT) for the diagnosis of ART failure for use with small volume liquid whole blood samples (e.g. finger prick whole blood) by incorporating a leucodepletion step, to remove CD4+ cells containing integrated HIV RNA and DNA. They have also developed and CE-marked a fully automated POC platform (SAMBA II) for use in resource-limited setting without the need for cold storage or shipment. This proposal is to integrate the SAMBA chemistry onto the SAMBA II instrument system, evaluate its performance in Africa and Europe and submit for CE-mark. The CE-mark is required to allow procurement of the SAMBA test into the developing world by NGO's.

The benefit of the Whole Blood Semi-Q Test for POC VL testing
The SAMBA Semi-Q Test using whole blood will be the only POC test using whole blood samples at POC. This will allow patients at decentralised sites to benefit from viral load monitoring. Furthermore they can be tested in the same visit (within 1.5 hours) so they can wait for their results, which should significantly reduce the loss to follow up rate.