INTRODUCTION
The London Declaration of 2012, endorsed by the World Health Organization (WHO), the Gates Foundation and 13 leading pharmaceutical companies, resolved to sustain, expand and extend programmes that ensure the necessary supply of drugs and other interventions to help eradicate Guinea worm disease, and help eliminate by 2020 lymphatic filariasis (LF), leprosy, human African trypanosomiasis (HAT) and blinding trachoma; and to help control by 2020 schistosomiasis, soil-transmitted helminths (STH), Chagas disease, visceral leishmaniasis (VL) and onchocerciasis [www.who.int/ntd]. This was to be done through partnerships and provision of funding to find next-generation treatments and interventions for neglected tropical diseases (NTDs).
The role of diagnostics in the control and elimination of NTDs was not specifically addressed at the London meeting, but was raised as a high priority at a follow-up meeting held at the Institut Pasteur in Paris in April 2014. Experience with disease elimination and eradication programmes has shown that adequate tools for surveillance are critical determinants of success. The report of the 1997 Dahlem workshop on the eradication of infectious diseases identified ‘practical diagnostic tools of sufficient sensitivity and specificity to detect levels of infection that can lead to transmission’ as one of the three essential requirements for disease elimination or eradication (Dowdle, Reference Dowdle1998). In the case of Guinea worm there is no need for a diagnostic test, since the clinical features are unmistakable (Al-Awadi et al. Reference Al-Awadi, Al-Kuhlani, Breman, Doumbo, Eberhard, Guiguemde, Magnussen, Molyneux and Nadim2014). The other NTDs mentioned in the declaration can be divided into those for which mass drug administration (MDA) is recommended, and those for which individual case management is required. The diagnostic needs for these two groups are clearly different.
MDA is recommended for the control and elimination of LF, onchocerciasis, schistosomiasis, STH and trachoma. Traditionally, with the exception of trachoma which is diagnosed clinically, these diseases have been diagnosed microscopically, using day or night bloods, urine or stool samples or skin snips. Microscopy can be a sensitive technique when parasite burdens are high, but requires trained technicians and a laboratory with a functioning microscope; quality assurance is often lacking. As the prevalence and intensity of infection falls following repeated rounds of MDA, a more sensitive test is required. Microscopy, despite concentration techniques, may no longer be sufficiently sensitive to detect the remaining cases. As elimination targets are approached, large numbers of samples need to be tested to be sure that transmission has been interrupted, so high-throughput tests are needed. Moreover, as the prevalence of infection falls, so does the positive predictive value of a diagnostic test, and highly specific tests are required to certify elimination.
Antigen detection assays either in the form of rapid immunochromatographic tests (ICTs) or enzyme linked immunoassays (ELISAs) offer alternative means of detecting infection. While ICTs offer ease of use and on-site results, ELISAs offer higher sensitivity and high throughput capability. As communities achieve interruption of transmission, it may be necessary to use molecular assays such as real-time PCR to detect the few remaining cases. However, for NTDs, these techniques remain laboratory-based, as they are expensive and require highly trained personnel and sophisticated equipment. For monitoring interruption of transmission and for certifying elimination, the consensus is that detection of specific antibodies to a pathogen may be the most sensitive marker of exposure. Detection of antibodies to immunodominant epitopes of a pathogen in sentinel populations (e.g. young children) is the ideal tool for certifying elimination, and for post-elimination surveillance. It may be necessary to use a combination of tests to achieve the required sensitivity and specificity. The role of diagnostics as a surveillance tool to support policy decisions related to treatment and control strategies for Schistosomiasis japonica towards elimination in China is an example of effective use of different types of diagnostics in disease control and elimination programmes (Zhou et al. Reference Zhou, Xu, Chen, Wang, Huang, Lin, Wang, Tang, Guo, Wu, Feng, Chen, Guo, Chen, Li, Wu and Peeling2011).
In the case of HAT and VL, for which treatment is difficult to administer, toxic, or both, it is necessary to diagnose and treat individual cases, and the diagnostic needs are different. Point-of-care (POC) tests are needed for these diseases, which are often found in remote rural areas far from a laboratory (Mabey et al. Reference Mabey, Peeling, Ustianowski and Perkins2004; Peeling and Mabey, Reference Peeling and Mabey2010).
In the following sections we describe the tools currently available for the diagnosis of the major NTDs mentioned in the London Declaration, the need for new diagnostics, and emerging diagnostic technologies that could meet these needs. Due to space constraints this is not a comprehensive review of diagnostics for NTDs; our aim was to focus on the key issues.
CURRENT DIAGNOSTICS TOOLS AND UNMET NEEDS
Lymphatic filariasis
Antigen detection tests adapted to ICT or ELISA formats are now used routinely by control programmes. These assays can be adapted to a bead or microfluidics-based assay to allow multiplexing. The ICT antigen detection test has been widely used for monitoring the impact of MDA programmes on the prevalence of Wuchereria bancrofti infection and for transmission assessment surveys. Tests for antigen are preferable to tests for antibody, which cannot distinguish between past, treated and current infection. Antibody detection assays using the BmR1 antigen have been used to monitor the impact of MDA on Brugia malayi infection. A recent multi-centre study showed that transmission assessment surveys, in which children aged 6–7 years were tested using these two assays, were an effective and practical tool for deciding when to discontinue MDA (Chu et al. Reference Chu, Deming, Biritwum, Bougma, Dorkenoo, El-Setouhy, Fischer, Gass, Gonzalez de Peña, Mercado-Hernandez, Kyelem, Lammie, Flueckiger, Mwingira, Noordin, Offei Owusu, Ottesen, Pavluck, Pilotte, Rao, Samarasekera, Schmaedick, Settinayake, Simonsen, Supali, Taleo, Torres, Weil and Won2013). A backlog of disabled patients with lymphoedema caused by LF will remain after transmission of the parasite has been interrupted, and it will be important for elimination programmes to address the needs of these patients.
Onchocerciasis
In 2001 the WHO published criteria for the certification of interruption of transmission/elimination of human onchocerciasis (WHO, 2001). The two key criteria were absence, or near absence, of infective stage larvae in the vector population, and infection rates of <0·1% in children living in the endemic area. Both require testing of large numbers of samples, using a diagnostic test with a high negative predictive value. Testing of vectors in pools of up to 200 for Onchocerca volvulus DNA by PCR has been shown to be effective in ruling out infection (Gopal et al. Reference Gopal, Hassan, Rodrıguez-Perez, Toe, Lustigman and Unnasch2012), and an ELISA to detect antibodies to a parasite-specific 16 kDa antigen (Ov16) in children has been used to certify interruption of transmission in foci in South America and Uganda (Oguttu et al. Reference Oguttu, Byamukama, Katholi, Habomugisha, Nahabwe, Ngabirano, Hassan, Lakwo, Katabarwa, Richards and Unnasch2014). Since there is a risk of encephalopathy in patients with Loa loa who are treated with ivermectin, a rapid diagnostic test for L. loa would be helpful in expanding coverage of MDA with ivermectin for both onchocerciasis and LF.
Schistosomiasis
In communities with a high prevalence and intensity of Schistosoma haematobium infection, simple dipstick tests for blood or protein in urine can identify communities requiring MDA; but as the prevalence and intensity declines following MDA, more sensitive tests are needed. Recombinant or soluble egg antigens with diagnostic potential for antibody detection assays have been characterized by a number of groups, and are both sensitive and specific (Tsang and Wilkins, Reference Tsang and Wilkins1991); but they are unable to distinguish present from past infection. A point of care ICT for the circulating cathodic antigen of Schistosoma mansoni has been shown to be more sensitive than microscopy of Kato-Katz smears in lower transmission settings (Sousa-Figueiredo et al. Reference Sousa-Figueiredo, Betson, Kabatereine and Stothard2013). Rapid immunoassays have also been developed and validated for Schistosoma japonicum (Xu et al. Reference Xu, Feng, Lin, Wang, Tang, Wu, Guo, Peeling and Zhou2011). A sensitive PCR test has been developed for the diagnosis of S. japonicum infection, which is able to detect 0·5 eggs per gram of stool (Fung et al. Reference Fung, Xiao, Wang and Carlton2012).
Trachoma
WHO recommends that MDA with azithromycin should be given annually for at least 3 years to communities where the prevalence of follicular trachoma (TF) in children aged 1–9 years exceeds 10%, and continued until the prevalence falls below 10%. However, it has been shown that follicular trachoma can persist for months or years after transmission of the causative agent, Chlamydia trachomatis, has been interrupted (Solomon et al. Reference Solomon, Foster and Mabey2006), suggesting that a test for infection could allow programmes to stop MDA sooner (Harding-Esch et al. Reference Harding-Esch, Sillah, Edwards, Burr, Hart, Joof, Laye, Makalo, Manjang, Molina, Sarr-Sissoho, Quinn, Lietman, Holland, Mabey, West and Bailey2013). A variety of nucleic acid amplification tests (NAATs) that can reliably detect 10–100 copies of C. trachomatis are widely used for the diagnosis of genital chlamydial infection, and have been used in trachoma research programmes. Tests that can quantify the ocular bacterial load are particularly useful in monitoring progress towards the interruption of transmission (Solomon et al. Reference Solomon, Holland, Burton, West, Alexander, Aguirre, Massae, Mkocha, Munoz, Johnson, Peeling, Bailey, Foster and Mabey2003; Roberts et al. Reference Roberts, Last, Molina-Gonzalez, Cassama, Butcher, Nabicassa, McCarthy, Burr, Mabey, Bailey and Holland2013). A recent publication demonstrated the potential of a bead-based serological assay for antibodies to two conserved antigens of C. trachomatis in certifying elimination and for post-elimination surveillance (Goodhew et al. Reference Goodhew, Priest, Moss, Zhong, Munoz, Mkocha, Martin, West, Gaydos and Lammie2012).
Visceral leishmaniasis
Microscopy of a splenic aspirate is both sensitive and specific, but splenic aspiration is not without risk. A simple, point of care serological test which detects antibodies to the rk39 protein has been widely used to diagnose VL, but it does not distinguish between subclinical infection, past treated infection and active disease requiring treatment (Chappuis et al. 2010). A comparison of five commercial VL rapid tests detecting antibodies to rK39 or rKE16 antigen showed sensitivities ranging from 92·8 to 100·0% and specificities ranging from 96·0 to 100·0%. However, sensitivity was lower (range 61·5–91·0%; 36·8–87·2%) against Brazil and East African panels, respectively. Specificity was consistently >95% in Brazil and ranged from 90·8 to 98·0% in East Africa (Cunningham et al. Reference Cunningham, Hasker, Das, El Safi, Goto, Mondal, Mbuchi, Mukhtar, Rabello, Rijal, Sundar, Wasunna, Adams, Menten, Peeling and Boelaert2012). The lower sensitivity may be due to antigenic differences between parasite strains (Bhattacharyya et al. Reference Bhattacharyya, Boelaert and Miles2013).
Human African trypanosomiasis
Because microscopic examination of a blood film is an insensitive test for the diagnosis of Trypanosoma brucei gambiense infection, a serological test (the card agglutination test, CATT) is used to screen at-risk populations. If low titres are considered positive this is a sensitive test, but lacks specificity (Inojosa et al. Reference Inojosa, Augusto, Bisoffi, Josenado, Abel, Stich and Whitty2006). Treatment is not recommended unless the infection can be confirmed parasitologically on a blood film or lymph node aspirate. Various techniques have been developed to increase the sensitivity of microscopy, such as centrifugation of blood and examination of the buffy coat, and the mini anion exchange column technique, but these are not always available in the remote rural areas where most cases of HAT are found (Chappuis et al. Reference Chappuis, Loutan, Simarro, Lejon and Büscher2005). The management of subjects who are CATT positive but parasitologically negative is problematic (Wastling et al. Reference Wastling, Picozzi, Wamboga, von Wissmann, Amongi-Accup, Wardrop, Stothard, Kakembo and Welburn2011). The recommendation is that they should be followed up and the tests repeated, but this is not always possible. Since the recommended treatment depends on whether the central nervous system (CNS) is involved, a lumbar puncture and cerebrospinal fluid (CSF) microscopy for staging the disease is also required. This is not always culturally acceptable, and the criteria for diagnosing stage 2 when parasites cannot be found in the CSF are not entirely clear. The presence of 5 or more white blood cells per mm3 of CSF is generally taken to indicate CNS involvement (Chappuis et al. Reference Chappuis, Loutan, Simarro, Lejon and Büscher2005).
EMERGING DIAGNOSTIC TECHNOLOGIES TO TRANSFORM DISEASE SURVEILLANCE
A new generation of immunoassays and molecular technologies has been developed that can be combined with readers or mobile phones to provide real-time NTD surveillance data with improved accuracy to inform treatment and disease-control strategies and monitor progress towards elimination. These technology platforms can be used to detect multiple targets from a single specimen. Many can yield quantitative results to give estimates of transmission intensity linked to precise geographical locations using GPS.
Bead-based immunoassays
Bead-based technologies are more versatile and potentially offer increased sensitivity over ELISAs. Instead of using enzymatic amplification of a colorimetric substrate, bead-based assays can use a range of labels including laser or fluorescence to detect the binding of the secondary antibody to the antigen or antibody target. The beads are in solution offering more sites for ligand binding compared with solid phase support used for ELISAs. Fluorescent intensities can be quantitative. Protein arrays offer the potential of quantification of the antigen target as a surrogate measure of parasite load remaining in the community to inform treatment strategies. Detection of multiple targets from the same pathogen can lead to increased sensitivity over single-target assays. Detection of targets from multiple pathogens using a single specimen can offer substantial cost and sample savings over traditional ELISA measurements.
Microfluidic immunoassays
Immunoassays developed on a microfluidic platform that miniaturizes but reproduces all the steps of an ELISA have been developed (Chin et al. Reference Chin, Lakasanasopin, Cheung, Steinmiller, Linder, Parsa, Wang, Moore, Rouse, Umviligihozo, Karita, Mwambarangwe, Braunstein, van de Wijgert, Sahabo, Justman, El-Sadr and Sia2011). The microfluidic disc can fit into the palm of a hand and is rapid and inexpensive to manufacture. These platforms allow multiplexing but they work best for antibody detection in blood samples. Antigen detection from urine or stool samples will require multistep specimen processing and/or purification and concentration before initiating the antigen-antibody binding and detection steps within the microfluidic channels.
Molecular assays
Bead-based molecular assays offer a promising approach to increase the performance of diagnostics beyond limit of detection of immunoassays. There are a number of exciting lab-on-a-chip developments in the pipeline with nanowires and quantum dot barcodes using microfluidics and magnetism (Giri et al. Reference Giri, Sykes, Jennings and Chan2011). These are simple to use and give answers in less than 30 min. The convergence of these nanotechnologies with rapidly evolving smart phone technology will lead to simple, rapid, low-cost multiplex DNA hybridization assays that can also transmit data to a central database, yielding real-time surveillance data from the field (Gao et al. Reference Gao, Lam and Chan2013). These lab-on-a-chip technologies offer an integrated approach combining pathogen detection and speciation with genetic analysis such as detection of single nucleotide polymorphisms. Hence they can be used for high-throughput screening of drug mutation targets. Advocacy to apply these novel technologies to NTDs is an urgent priority (Jani and Peter 2013).
POC MOLECULAR ASSAYS
A number of nucleic acid amplification tests (NAATs) have been developed in the last 20 years and have been used as the gold standard to which the performance of other diagnostic tests is compared. They have largely remained laboratory based but, in recent years, a number of POC NAATs have been developed. The first sample in–answer out real-time PCR platform that can be used wherever there is a source of electricity has been introduced for the diagnosis of tuberculosis and detection of rifampicin resistance (Boehme et al. Reference Boehme, Nicol, Nabeta, Michael, Gotuzzo, Tahirli, Gler, Blakemore, Worodria, Gray, Huang, Caceres, Mehdiyev, Raymond, Whitelaw, Sagadevan, Alexander, Albert, Cobelens, Cox, Alland and Perkins2011). The GeneXpert real-time PCR platform allows random access so that different tests can be initiated as required. It requires minimal onsite expertise and has remote quality control.
Isothermal amplification platforms such as helicase-dependent amplification (HDA), cross-priming amplification (CPA), recombinase polymerase amplification (RPA) and loop-mediated amplification (LAMP) assays can be made into POC NAATs as there is no requirement for sophisticated equipment to perform thermal cycling. The details of these technologies are described in an excellent review article (Niemz et al. Reference Niemz, Ferguson and Boyle2011). POC NAATs based on these technologies have largely been applied to high-burden diseases such as HIV and tuberculosis, but LAMP assays have been developed and evaluated for the diagnosis of VL and HAT. In the case of VL, the LAMP assay for Leishmania donovani is highly sensitive and specific for the diagnosis of VL (using a blood sample), and for the diagnosis of post kala-azar dermal leishmaniasis (PKDL) using a skin biopsy (Verma et al. Reference Verma, Avishek, Sharma, Negi, Ramesh and Salotra2013). Diagnosis of PKDL will be important for the elimination agenda, since patients with this condition are an important source of infection. In the case of HAT, the LAMP assay was shown to have a sensitivity of 87·3% (95% CI 80·9–91·8%, and a specificity of 92·8% (95% CI 86·4–96·3%) against a PCR gold standard (Mitashi et al. Reference Mitashi, Hasker, Ngoyi, Pyana, Lejon, Van der Veken, Lutumba, Büscher, Boelaert and Deborggraeve2013). Advocacy and investments are needed to apply these technologies to the control and elimination of NTDs.
MOVING FORWARD
The development of appropriate diagnostics starts with defining target product profiles. At a WHO-sponsored meeting held at the London School of Hygiene & Tropical Medicine in 2010, the first steps towards defining TPPs for diagnostic tools for measuring the impact of MDA, and for post-elimination surveillance for these diseases were taken (Solomon et al. Reference Solomon, Engels, Bailey, Blake, Brooker, Chen, Chen, Churcher, Drakeley, Edwards, Fenwick, French, Gabrielli, Grassly, Harding-Esch, Holland, Koukounari, Lammie, Leslie, Mabey, Rhajaoui, Secor, Stothard, Wei, Willingham, Zhou and Peeling2012). Further consultations were organized by the Task Force for Global Health with key stakeholders to define target product profiles for diagnostic tools needed in two phases of disease control: pre-elimination and post-elimination. An important issue that needs consensus is the sampling strategy as the throughput needs to be matched to appropriate technologies. The sampling strategy depends on a number of factors, including the prevalence threshold above which MDA is implemented, the inherent spatial heterogeneity of the individual infections, and how this varies for different NTDs and different stages of control; cost of conducting surveys in the field and cost and accuracy of diagnostic tests in classifying populations, which could be termed efficiency. This will require modelling the cost-effectiveness of using different tools for different phases of control and needs to take into account spatial characteristics for the distribution of infections and differing diagnostic accuracies and costs. The models can be used to investigate the required sensitivity and specificity within given control programme contexts (Drain et al. 2013). Consideration should be given the use of multiplex testing for more than one NTD, which could prove cost-effective for post-MDA surveillance in some settings.
Once a test is developed and standardized, its performance and utility need to be validated. A bank of well-characterized specimens is critical to the evaluation of the analytical performance of these tests. POC tests will also need to be validated in settings of intended use, covering the full spectrum of populations in which the test will be used (Banoo et al. Reference Banoo, Bell, Bossuyt, Herring, Mabey, Poole, Smith, Sriram, Wongsrichanalai, Linke, O'Brien, Perkins, Cunningham, Matsoso, Nathanson, Olliaro, Peeling and Ramsay2006). Standardized protocols for laboratory and field evaluation and a network of competent evaluation sites that can perform evaluations and pilot studies to implement new technologies are needed to accelerate the adoption and introduction of new diagnostic tools for NTDs.
CONCLUSIONS
Appropriate diagnostics to monitor disease trends and assess the impact of interventions are essential for guiding treatment strategies for NTDs at different thresholds of control, interruption of transmission, elimination and post-elimination surveillance. While there remains a need for standardization and evaluation of new tests in a variety of settings before they are widely adopted, it is clear that recent technical advances have led to improved diagnostic tests for NTDs, which are sensitive, specific, able to diagnose multiple infections using a single specimen, and can be used at the point of care. Since there is extensive geographical overlap between different NTDs targeted by MDA, multiplex surveillance platforms are likely to prove cost-effective. However, barriers remain to the rapid deployment of these new technologies for the control and elimination of NTDs. The importance of better diagnostics for NTD control and elimination must be brought to the attention of the policymakers and funding agencies.
ACKNOWLEDGEMENT
This paper is based on a plenary lecture delivered by Professor Peeling at a meeting of the British Society of Parasitology in Ness Gardens, Liverpool in 2013.
