In spite of increased publicity in the Western world about malaria and drives to provide mosquito nets, the disease is still endemic in a large part of the world. This article discusses different methods of malaria diagnosis and the role that point-of-care tests can play in the ultimate goal of malaria elimination.

by Dr Jackie Cook

Finding the balance: over- and under-diagnosing malaria
Malaria remains a huge burden in many parts of the world, particularly in sub-Saharan Africa. Despite increased availability of effective treatments and interventions, malaria elimination is still out of reach for many countries. Whilst availability of effective interventions that reduce contact with infected mosquitoes, such as insecticide treated bed-nets or indoor spraying with insecticide are key to reducing malaria prevalence, case management also plays a key role. Many who need treatment are unable to get it, either through lack of access to healthcare, or because infections remain undiagnosed. Conversely, some studies suggest that many patients are receiving anti-malarials unnecessarily due to a tendency to diagnose based solely on clinical symptoms, many of which are similar to other infections, rather than using a diagnostic. In under-resourced settings, this can result in any child presenting with a fever being prescribed malaria drugs. This simultaneously means non-malaria fevers remain undiagnosed and untreated, as well as a large proportion of unnecessary prescriptions for malaria drugs, which increases healthcare costs and the risk of drug resistance, a very potent threat. In order to counteract this, the last few decades have brought a push from health officials, researchers, donors and governments alike to confirm every suspected case of malaria before prescribing treatment.

Microscopy
For many, malaria diagnosis is performed using microscopy, a procedure that is relatively cheap but requires a skilful operator. Malaria is caused by the plasmodium parasite and it undergoes several developmental and replication stages in the human. These stages can be seen through a microscope when blood is prepared on either a thin or thick film and stained, normally, with Giemsa or Wright’s stains. Experienced microscopists can detect down to 1 parasite per microlitre of blood, although the typically quoted sensitivity for microscopy is approximately 100 parasites per microlitre. In reality, the sensitivity of the test depends greatly on the microscopist. In areas where malaria transmission is declining, microscopists can go months without seeing a positive slide, and as such, skills may begin to decline. In addition, the need for well-maintained microscopes and access to slides and stain can mean microscopy is not always available.

Rapid diagnostic tests
The first malaria rapid diagnostic test (RDT) was developed in 1993 and in the decades since many variations have proliferated on the market. RDTs are typically immunochromatograhic tests that use monoclonal antibodies to detect the presence of plasmodium antigens (proteins produced by the parasite) which are present in the blood of infected, or recently infected, individuals. They are generally stable at a range of temperatures and do not require special storage conditions. RDTs require significantly less training for use than microscopy and a positive infection is easy to identify by visualization of a ‘positive’ line, meaning the results are much less subjective. Most RDTs require 15–20 minutes for development, meaning treatment can be given while patients wait at health facilities.

However, there are a few downsides to the use of RDTs. The presence of parasite antigen doesn’t always equate with a current infection, but can signify a recently cleared infection from within the previous two weeks. In addition, several studies have reported the deletion of certain antigens detected by RDTs in plasmodium parasites, meaning false-negative results may be obtained in areas using these types of RDTs [1]. The World Health Organization (WHO), in collaboration with the Foundation for Innovative New Diagnostics (FIND), has set up an RDT product testing programme, an essential quality assurance component considering the huge influx of RDT brands that have popped up in the past 20 years [2]. The reports from the programme make worrying reading with very low sensitivity for some brands, differences between batches of RDT and a general lower sensitivity for non-falciparum infections for nearly all brands.

The hidden reservoir: asymptomatic, low-density infections
In general, the limited sensitivity of both microscopy and RDT (unreliable detection in infections with a parasite density less than 100 parasites per microlitre) is not an issue for symptomatic malaria infections, the majority of which will consist of high parasite densities. However, asymptomatic infections are numerous, in high and low transmission settings. These asymptomatic infections pose a problem for control programmes. The carriers do not feel unwell so have no reason to present to a health facility for testing and yet, they may be infectious to mosquitoes, meaning they pose a risk for onward transmission. In order to detect and treat these asymptomatic infections, malaria programmes are now taking their diagnostics into the community in a strategy termed Mass Screening and Treatment (MSAT). This involves testing everyone within a community regardless of whether they have symptoms. Many of these infections are asymptomatic and therefore also likely to be low-density; hence which test you use can mean the difference between detecting 10 infections or 100 infections. Whilst RDT is ideal for field conditions, studies have shown that they can miss a large proportion of infections that are present [3].

Molecular tests
Polymerase chain reaction
More sensitive diagnostics are available in the terms of molecular tests. The most commonly used is polymerase chain reaction (PCR). Numerous PCR assays have been developed, many based on amplifying the 18S ribosomal RNA (18SrRNA), first published by Snounou and colleagues in 1993 [4]. PCR detects parasite nucleic acids and can detect much lower parasite densities than RDT or microscopy, with tests reportedly able to detect down to 1 parasite per microlitre of blood, as well as being able to accurately distinguish between plasmodium species. However, the number of assays available has resulted in calls for a standardized test so results can be compared across the world. PCR tests are generally performed on blood collected on filter paper but the equipment required for PCR and the expense of maintaining a sterile lab environment precludes PCR from being available in many health facilities. This means that samples need to be sent away, with an often long wait for results. Although more field-friendly PCR methods are in the pipeline, currently, PCR is not generally considered suitable for a point-of-care test, although it’s use in epidemiological studies is undisputed.

Loop-mediated isothermal amplification
Loop-mediated isothermal amplification (LAMP) was first developed in 2000, with the aim to amplify DNA in a sensitive, specific and speedy manner (Figs 1, 2). One of the main advantages is the fact it can be performed under isothermal conditions, and thus averting the need for a thermocycler. LAMP can be thought of as a ‘rough-and-ready’ PCR, as it is also less sensitive to inhibitors present in biological samples, and therefore allows the use of simple and cheap DNA extraction methods. The fast time-to-results and the minimal equipment required make LAMP an attractive option for field diagnosis. In order to make this a viable option, FIND and partners Eiken Chemical Ltd, Japan, and the Hospital for Tropical Diseases (HTD), London, UK have developed a field-stable kit with all reagents freeze-dried into the lid of the reaction tube, which means minimal processing is required. Although still in the development and testing stage, current results of the use of the kit are promising, with strong agreement with PCR results and a considerably higher sensitivity than RDT [5-8]. Whilst seemingly the most sensitive of the point-of-care tests available, there are some downsides to LAMP. Results still take considerably longer than RDT, requiring patients to wait at clinics for 2 hours for results, or leaving the health facility staff with the complicated task of contacting and following up any positive patients. In addition, electricity is required for the processing of samples, making it not practical for many places.

Future for point-of-care diagnostics for malaria
These advances in molecular diagnostics mean infections that would previously have remained undetected can now be confirmed, treated and cleared. Identifying and treating all infections becomes a greater priority as transmission reduces and the possibility of elimination comes into focus. This is occurring in areas around the world such as Swaziland and Zanzibar in Africa and in South East Asia, where the need to eliminate has become ever more important with the emergence of drug-resistant parasites. In these areas, identification of every last parasite is the aim and development of a quick, sensitive and reliable diagnostic is key to that.

As more studies reveal the extent of the low-density parasite reservoir, there is a sense of ‘the more we look the more malaria we will find’. But do we need to find all these infections in order to eliminate malaria? It should be noted that these ‘super-sensitive’ tests are a relatively recent phenomena and that countries have succeeded in malaria elimination without them. The role these low-density parasitemias play in transmission is not fully understood but for now the aim remains to clear the last parasite standing.

References
1. Houze S, Hubert V, Le Pessec G, Le Bras J, Clain J. Combined deletions of pfhrp2 and pfhrp3 genes result in Plasmodium falciparum malaria false-negative rapid diagnostic test. J Clin Microbiol. 2011; 49(7): 2694–2696.
2. WHO, FIND, CDC. Malaria rapid diagnostic test performance: Results of WHO product testing of malaria RDTs: Round 5. 2013; http://www.who.int/malaria/publications/atoz/9789241507554/en/.
3. Cook J, Xu W, Msellem M, Vonk M, Bergström B, Gosling R, Al-Mafazy AW, McElroy P, Molteni F, Abass AK, Garimo I, Ramsan M, Ali A, Mårtensson A, Björkman A. Mass screening and treatment on the basis of results of a plasmodium falciparum-specific rapid diagnostic test did not reduce malaria incidence in Zanzibar. J Infect Dis. 2015; 211(9): 1476–1483.
4. Snounou G, Viriyakosol S, Zhu XP, Jarra W, Pinheiro L, do Rosario VE, Thaithong S, Brown KN. High sensitivity of detection of human malaria parasites by the use of nested polymerase chain reaction. Mol Biochem Parasitol. 1993; 61(2): 315–320.
5. Hopkins H, González IJ, Polley SD, Angutoko P, Ategeka J, Asiimwe C, Agaba B, Kyabayinze DJ, Sutherland CJ, Perkins MD, Bell D. Highly sensitive detection of malaria parasitemia in a malaria-endemic setting: performance of a new loop-mediated isothermal amplification kit in a remote clinic in Uganda. J Infect Dis. 2013; 208(4): 645–652.
6. Polley SD, González IJ, Mohamed D, Daly R, Bowers K, Watson J, Mewse E, Armstrong M, Gray C, Perkins MD, Bell D, Kanda H, Tomita N, Kubota Y, Mori Y, Chiodini PL, Sutherland CJ. Clinical evaluation of a loop-mediated amplification kit for diagnosis of imported malaria. J Infect Dis. 2013; 208(4): 637–644.
7. Aydin-Schmidt B, Xu W, González IJ, Polley SD, Bell D, Shakely D, Msellem MI, Björkman A, Mårtensson A. Loop mediated isothermal amplification (LAMP) accurately detects malaria DNA from filter paper blood samples of low density parasitaemias. PLoS One 2014; 9(8): e103905.
8. Cook J, Aydin-Schmidt B, González IJ, Bell D, Edlund E, Nassor MH, Msellem M, Ali A, Abass AK, Mårtensson A, Björkman A. Loop-mediated isothermal amplification (LAMP) for point-of-care detection of asymptomatic low-density malaria parasite carriers in Zanzibar. Malar J. 2015; 14(1): 43.

The author
Jackie Cook PhD
London School of Hygiene and Tropical Medicine, London, UK

E-mail: Jackie.cook@lshtm.ac.uk

To reduce the window period for HIV-1 infection, a method for detecting trace amounts of HIV-1 p24 in blood is needed. We developed a simple de novo ultrasensitive colorimetric ELISA by adding a thio-NAD cycling solution to the standard ELISA. The limit of detection for p24 was 0.005 IU (i.e. attomoles) per assay by the ultrasensitive colorimetric ELISA.

by Dr A. Nakatsuma, M. Kaneda, H. Kodama, M. Morikawa, S. Watabe, et al.

Background
During the window period between infection with human immunodeficiency virus type 1 (HIV-1) and the appearance of detectable antibodies to HIV-1, the infection cannot be diagnosed. Attempts to shorten this period have been made using a fourth-generation immunoassay that detects both HIV-1/2 IgG/M and HIV-1 p24 antigens [1]. However, most of the commercially available detection systems for fourth-generation immunoassays use chemiluminescent measurement and thus require specialized, highly expensive automated measurement equipment. For this reason, fourth-generation immunoassays are performed only at diagnostics companies and hub hospitals. To overcome this limitation and to test many samples simultaneously, there is need of an immunoassay with increased sensitivity for the HIV-1 p24 antigen that nonetheless uses a common enzyme and does not require any specialized instruments.

In 2010, French health authorities mandated a limit of detection of at least 2 IU/mL of HIV-1 p24 antigen for a Conformité Européenne (CE)-marked HIV antigen/antibody assay [2]. According to this mandate, commercially available assay kits were manufactured to detect p24 antigen with limits of detection ranging from 0.505 to 1.901 IU/mL and from 11.9 to 33.5 pg/mL [2]. Units of pg/mL are used for the Société Française de Tranfusion Sanguine (SFTS) standard (i.e. recombinant proteins), versus IU/mL for the WHO (World Health Organization) standard. As 1 IU/mL is estimated to be equivalent to 10 pg/mL and MW = 24 000 for p24, the best sensitivity in these kits is 0.505 IU/mL, which is ~2 × 10⁻¹⁶ moles/mL.

To date, numerous methods have been proposed for the detection of p24 antigen. However, the limit of detection of p24 antigen is not expected to overcome the sensitivity of 10−17 to 10⁻¹⁸ moles/mL. In addition, we have to note that HIV testing of many samples requires not only ultrasensitive HIV-1 p24 detection but also rapidity, a reasonable cost, and a simple protocol without the requirement of special equipment. In the present review, we introduce a de novo ultrasensitive colorimetric enzyme-linked immunosorbent assay (ELISA) for HIV-1 p24 [3].

Mechanism of ultrasensitive colorimetric ELISA
Watabe and colleagues developed an ultrasensitive ELISA to measure trace amounts of proteins by combining a conventional ELISA with thionicotinamide-adenine dinucleotide (thio-NAD) cycling [4]. Their rationale was that although proteins cannot be amplified by polymerase chain reaction (PCR) in the manner of nucleic acids, a detectable signal for proteins can be amplified. Thus, their ultrasensitive ELISA (Fig. 1) employs a sandwich method using a primary and a secondary antibody for antigens. An androsterone derivative, 3α-hydroxysteroid, is produced by the hydrolysis of 3α-hydroxysteroid 3-phosphate with alkaline phosphatase linked to the secondary antibody. This 3α-hydroxysteroid is oxidized to a 3-ketosteroid by 3α-hydroxysteroid dehydrogenase (3αHSD) with a cofactor thio-NAD. By the opposite reaction, the 3-ketosteroid is reduced to a 3α-hydroxysteroid by 3α-HSD with a cofactor NADH. During this cycling reaction, thio-NADH accumulates in a quadratic function-like fashion. Accumulated thio-NADH can be measured directly at an absorbance of 400 nm without any interference from other cofactors.

This method enables the detection of a target protein with ultrasensitivity (10⁻¹⁹ moles/assay) by measuring the cumulative quantity of thio-NADH by a colorimetric method without the use of any special instruments for the measurements of fluorescence, luminescence or radio isotopes [4]. Further, we should note that this ultrasensitive method will allow a technician to detect trace amounts of proteins simply by applying thio-NAD cycling reagents to the conventional ELISA system. We therefore applied this ultrasensitive ELISA to the detection of HIV-1 p24 antigen in blood [3].

Sensitivity and stability of the ultrasensitive colorimetric ELISA for HIV-1 p24
A typical linear calibration curve for HIV-1 p24 antigen provided by the ultrasensitive ELISA coupled with thio-NAD cycling was y = 0.27x + 0.019, R² = 0.99 in the range of 0.1‒1.0 IU/mL. The limit of detection of p24 was 0.0055 IU/assay (i.e. ~2 × 10−18 moles/assay). These findings indicate that the ultrasensitive colorimetric ELISA succeeds in detecting p24 at the attomole level [3]. Because this measurement system employs a 50 µL solution for each assay, the detection limit corresponded to 0.1 IU/mL, or 10⁻¹⁷ moles/mL. Therefore, even in terms of the concentration per mL, our detection limit is less than one-tenth of that required by the French health authorities [2]. The coefficient of variation was 8% for 1 IU/mL.

Spike-and-recovery test using serum
We attempted to perform spike-and-recovery tests in which the HIV-1 p24 antigen was added to the control serum. Because our results demonstrated that the ratio was about 100% for 0.5 IU/mL of HIV-1 p24, which was less than the value (2 IU/mL) required for a CE-marked HIV antigen/antibody assay (see Background), the ultrasensitive method was judged to sufficiently detect IV-1 p24 antigen in human blood obtained from patients in the very early period after infection.

Detection of HIV-1 p24 in the early stages of infection
It is important to diagnose primary HIV-1 infection and begin antiretroviral treatment as early as possible. Most HIV-1/2 antibody diagnostic tests detect the antibodies for the antigens of HIV-1 gp41 and HIV-2 gp36, which are highly conservative transmembrane proteins. These tests are quick and easy, and thus have been widely used in many clinics and public health centres. However, when only the antibody diagnostic tests are used, there is a long delay (generally a 28-day window period) before diagnosis is possible [5]. Further, HIV-1/2 antibody tests in children younger than 18 months tend to be especially inaccurate as a result of the continued presence of maternal antibodies [6]. To shorten the delay and to validate HIV tests, the HIV-1 p24 antigen, the concentration of which is expected to increase before antibodies emerge, should be detectable in trace amounts. HIV-1 p24 in blood emerges transiently in the very early period after infection, and then its concentration quickly returns to the basal level [5]. An HIV-1 p24 test is, therefore, very useful as a screening test in the early stage of infection.

Closing the gap on PCR-based nucleic acid testing (NAT)
Generally, the gold standard for diagnosing HIV-1 is PCR-based nucleic acid testing (NAT) [7], but this method is expensive and has infrastructure requirements, a long measuring time, and high complexity, thereby limiting its usefulness for large numbers of samples. There is also the issue that much of the world lacks access to reliable NAT, and thus in many geographic regions the policy is to simply wait until symptoms develop. Use of ultrasensitive detection of HIV-1 p24 antigen for early diagnosis would be a simple and reasonable alternative to NAT, such as for monitoring HIV treatment and protecting the blood supply. Accordingly, it is time to reconsider whether NAT should be the gold standard for diagnosing HIV-1. Barletta et al. claimed that the target protein (i.e. HIV-1 p24 antigen) is present in the virion in much higher numbers than viral RNA copies (approximately 3000 HIV-1 p24 antigen molecules versus 2 RNA copies per virion) [8]. The 10⁻¹⁸ moles/assay value in our present results corresponds to 10⁶ protein molecules/assay, or ~10³ RNA copies/assay. Although under laboratory conditions a real-time PCR (i.e. NAT) can detect on the order of 10¹ RNA copies/assay, the limitation of detection is usually in the order of 10² RNA copies/assay [9]. Hence, the ultrasensitive ELISA coupled with thio-NAD cycling for HIV-1 p24 is closing in on the detection limit obtained by NAT, with a margin of difference of only one order of magnitude.

Conclusion
The ultrasensitive ELISA coupled with thio-NAD cycling is a very convenient method for the early testing of HIV-1 infection because it requires only the addition of a thio-NAD cycling solution to the usual ELISA without the use of any specialized measuring equipment. Consequently, the present method could be widely used as a powerful tool to test many samples simultaneously.

References
1. George CRR, Robertson PW, Lusk MJ, Whybin R, Rawlinson W. Prolonged second diagnostic window for human immunodeficiency virus type 1 in a fourth-generation immunoassay: Are alternative testing strategies required? J Clin Microbiol. 2014; 52: 4105–4108.
2. Ly TD, Plantier JC, Leballais L, Gonzalo S, Lemée V, Laperche S. The variable sensitivity of HIV Ag/Ab combination assays in the detection of p24Ag according to genotype could compromise the diagnosis of early HIV infection. J Clin Virol. 2012; 55: 121–127.
3. Nakatsuma A, Kaneda M, Kodama H, Morikawa M, Watabe S, Nakaishi K, Yamashita M, Yoshimura T, Miura T, Ninomiya M, Ito E. Detection of HIV-1 p24 at attomole level by ultrasensitive ELISA with thio-NAD cycling. PLoS One 2015; 10: e0131319.
4. Watabe S, Kodama H, Kaneda M, Morikawa M, Nakaishi K, Yoshimura T. Ultrasensitive enzyme-linked immunosorbent assay (ELISA) of proteins by combination with the thio-NAD cycling method. BIOPHYSICS. 2014; 10: 49–54.
5. World Health Organization (WHO). HIV/AIDS Fact sheet No 360. WHO 2015; http://www.who.int/mediacentre/factsheets/fs360/en/
6. Zijenah LS, Tobaiwa O, Rusakaniko S, Nathoo KJ, Nhembe M, Matibe P, Katzenstein DA. Signal-boosted qualitative ultrasensitive p24 antigen assay for diagnosis of subtype C HIV-1 infection in infants under the age of 2 years. J Acquir Immune Defic Syndr. 2005; 39: 391–394.
7. Patel P, Mackellar D, Simmons P, Uniyal A, Gallagher K, Bennett B, Sullivan TJ, Kowalski A, Parker MM, LaLota M, Kerndt P, Sullivan PS; Centers for Disease Control and Prevention Acute HIV Infection Study Group. Detecting acute human immunodeficiency virus infection using 3 different screening immunoassays and nucleic acid amplification testing for human immunodeficiency virus RNA, 2006-2008. Arch Intern Med. 2010; 170: 66–74.
8. Barletta JM, Edelman DC, Constantine NT. Lowering the detection limits of HIV-1 viral load using real-time immuno-PCR for HIV-1 p24 antigen. Am J Clin Pathol. 2004; 122: 20–27.
9. Wagatsuma A, Sadamoto H, Kitahashi T, Lukowiak K, Urano A, Ito E. Determination of the exact copy numbers of particular mRNAs in a single cell by quantitative real-time RT-PCR. J Exp Biol. 2005; 208: 2389–2398.

The authors
Akira Nakatsuma¹ PhD, PhC; Mugiho Kaneda¹ BAgr; Hiromi Kodama¹ MAgr; Mika Morikawa^1,2 BASc; Satoshi Watabe³ BPha; Kazunari Nakaishi²; Masakane Yamashita⁴ PhD; Teruki Yoshimura⁵ PhD, PhC; Toshiaki Miura⁶ PhD, PhC; Masaki Ninomiya¹ PhD, PhC; Etsuro Ito*¹ PhD

¹ Kagawa School of Pharmaceutical Sciences, Tokushima Bunri University, Sanuki, Japan
² TAUNS Laboratories, Inc., Izunokuni, Japan
³ BL Co., Ltd., Numazu, Japan
⁴ Faculty of Science, Hokkaido University, Sapporo, Japan
⁵ Faculty of Pharmaceutical Sciences, Health Sciences University of Hokkaido, Ishikari-Tobetsu, Japan
⁶ Graduate School of Pharmaceutical Sciences, Hokkaido University, Sapporo, Japan

*Corresponding author
E-mail: eito@kph.bunri-u.ac.jp