Statistics from Altmetric.com
Developing countries face many challenges in achieving the UNAIDS 90:90:90 targets for HIV testing, treatment and viral load (VL) undetectability by 2020.1 Even if a high proportion of patients are diagnosed and put on antiretroviral therapy (ART), the third target requires access to VL testing. The fact is that this is not available to a high proportion of patients in resource-poor countries and most are monitored using clinical assessment or CD4 counts (if available).2 Even when the VL machinery is introduced, lack of maintenance, reagents and skilled staff means that testing is still not done.3 Some WHO member states report that 10% of VL analysers are out of action and only 36% of available capacity is being used.3
The most likely way that these problems will be overcome is by the use of HIV viral load point-of-care testing (VL-POCT).4 If this becomes available at a reasonable cost and the technology proves to be simple and robust, then VL results could be routinely available on the day of testing. This would allow clinical decisions to be made and treatment given before the patient leaves the clinic, as opposed to the several weeks’ gap that is usual for most patients in developing countries, if they get a test at all.2–4 The three situations where this would be most useful are in monitoring for HIV treatment failure, early infant diagnosis (EID) and confirmation of a positive HIV antibody test in pregnancy or acute infection.2–4
WHO define treatment failure as a VL >1000 copies/mL5 and so VL-POCT technology has been developed to simplify the results output to a binary ‘>1000’ or ‘<1000’ copies/mL. Decisions can then be made about the need for ART switch and/or adherence support.5–7 When used for EID, the sensitivity of the test is set to detect a VL as low as is possible and a positive result allows the infant to be placed on ART quickly to prevent disease progression.6 7 Using the same sensitivity setting, the VL-POCT can also be used in settings where rapid confirmation of infection is important, such as women in labour or acute infection.6 7
The Cepheid Xpert HIV-1 viral load point-of-care assay has already received WHO approval6 and produces a result in about 90 min from input of the sample although should be considered as a ‘Near-POCT’ due to additional steps required for sample taking and preparation.6 7 When monitoring for ART treatment failure at >1000 copies/mL, the performance of the Xpert system is similar to most current laboratory HIV VL assays with a 98% concordance, 94% sensitivity and 99% specificity6–8 using plasma samples. Using a different (‘qualitative’) module, it has a published sensitivity and specificity 93% and 100% for EID, which can be performed on dried blood spot or whole blood samples.6–8 The machine is reported to be relatively easy to use and therefore is not necessarily fixed to major city hospital laboratories, enabling it to be available in smaller towns and therefore accessible to more patients. A further advantage is that Cepheid Xpert technology is already being used in many developing countries for the diagnosis of tuberculosis.7
There are some downsides to the Cepheid assay however. First, the cost per test is similar to most other commercial VL assays.6 Unless the system and reagents are sold at very discounted prices, it will still be too expensive for many or even most patients in resource-poor settings.6 It also requires a relatively sophisticated laboratory setting.7 The quantitative assay for ART treatment monitoring only works with plasma samples obtained by phlebotomy or capillary tube samples followed by centrifugation, and therefore this requires additional time and resources, extending the time from patient arrival to result by several hours.6–8
There are two other plasma-based HIV VL-POCT assays in development: the Alere Q NAT POC and the Iquum/Roche LIAT HIV Quant.9 10 If these are developed further for use in developing countries, the advantages and disadvantages will be similar to the Cepheid Xpert. The LIAT assay has a 100% sensitivity but only 88% specificity using plasma, falling to 41% specificity for whole blood samples.10 The Alere assay also has very poor specificity when used with whole blood,9 and so further developments are awaited for both assays.
At the moment, there is only one HIV VL-POCT assay that performs well with whole blood samples—the SAMBA-II Semi Q assay (Diagnostics for the Real World, Cambridge).11 12 This has an in-built leukodepletion step that removes the white cells, and therefore the intracellular RNA and pro-viral DNA, that gives overquantification in other assays.11 12 That being so, the test can be performed on finger-prick whole blood samples and thus avoiding phelobotomy and centrifugation.11 12 It is fully automated, allowing sample-in/result-out operation, is reported to require very little training to use, has a high temperature and humidity tolerance and less reliance on a stable power supply.11 It therefore has the potential to be used in very remote areas and each machine is about the size of a domestic microwave and therefore is easily transported. Studies of its performance at the 1000 copies/mL cut-off report a 96%–99% concordance with commercial VL assays and a sensitivity of 96% and specificity of 99%–100%.11 12 SAMBA can also be used for EID using heel-prick specimens.11 12 By avoiding the phlebotomy and centrifugation stages of other assays, the time to result should be quicker than with other assays. SAMBA-II can test about 12 samples per day, although multiple machines would increase throughput. There is a semiautomated SAMBA-I assay that allows a higher sample throughput of 24–48 samples per day. It has not received WHO approval yet, although this is anticipated soon and the manufacturer hopes that the cost per test (including all consumables) may be as little as US$15, although the cost of the machine has also to be taken into account.
As these technologies are rolled out, it can be envisaged that the SAMBA-II could be used in more remote areas and for hard-to-reach communities (including those in developed countries).11 12 Other assays that are more laboratory-based, such as Xpert, SAMBA-I and the Alere and Iquum assays, may be more suitable for use in larger towns and cities, providing for higher patient volumes.8–11
Other possible technologies in development for rapid and cheap HIV-VL testing include the lateral flow assay and RT-LAMP (Loop Isothermal Amplification) assay.13 14 The latter is described as a ‘lab on a chip’ using a smartphone and requiring a single drop of blood. These appear to be a long way from commercialisation yet, but would be a huge leap forward if they come to fruition.
We appear to be on the cusp of a major step forward in making HIV-VL testing more available to developing countries and in remote communities. Providing the HIV VL-POCT assays can be offered at an affordable price, prove to be robust and are less hampered by breakdown and lack of reagents, widespread HIV-VL testing should become more readily achievable to the benefit of the many patients who receive suboptimal ART monitoring currently.
Funding None declared.
Competing interests The author has been a lead investigator in a study of the SAMBA II assay
Patient consent Not required.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.