INTRODUCTION
The population at risk for lymphatic filariasis (LF) is 1.4 billion people in 73 countries (WHO, 2013). It is a major cause of severe morbidity and a significant impediment to socioeconomic development (Taylor et al. Reference Taylor, Hoerauf and Bockarie2010). The causative agents of LF are Wuchereria bancrofti, Brugia malayi and Brugia timori with W. bancrofti accounting for 91% of infections worldwide and all infections in Africa and the Americas (Bockarie et al. Reference Bockarie, Pedersen, White and Michael2009a). These parasites cause damage to the lymphatic system and produce lymphoedema, hydrocoeles and elephantiasis in 44 million people worldwide (Taylor et al. Reference Taylor, Hoerauf and Bockarie2010). The life cycle of all the parasites involve a definitive mammalian host and various species of mosquito vectors belonging to the Anopheles, Aedes, Culex, Mansonia and Ochlerotatus genera (Bockarie et al. Reference Bockarie, Pedersen, White and Michael2009a). Wuchereria bancrofti seems to be exclusively a human parasite, whereas Brugia spp. are zoonotic and anthroponotic in many situations. The infective third stage larvae (L3) of the parasite penetrate the skin of the definitive human host, from the bite of a mosquito, and develop into adult worms in the lymphatic system. Following sexual reproduction, the adult worms produce microfilariae (MFs) in the host bloodstream. MFs are ingested with the blood meal taken by female mosquitoes of a susceptible vector species.
The availability of safe, single-dose, two-drug treatment regimens capable of reducing microfilaraemia to near zero levels for 1 year or more, along with remarkable improvement in techniques for diagnosing the infection resulted in a World Health Assembly in 1997 calling for the elimination of LF as a public health problem globally and eventually the launch of the Global Programme to Eliminate Lymphatic Filariasis (GPELF) in 2000 (Bockarie et al. Reference Bockarie, Taylor and Gyapong2009b).
The London declaration in 2012 saw a seismic shift in the momentum for the fight against neglected tropical diseases (NTDs) with increased commitment from the global health community to complete mapping in all endemic countries; increase drug donation and support mass-drug administration (MDA) implementation in line with the targets set by the World Health Organization (WHO) (Lancet, 2014). By the end of 2013, pharmaceutical partners supplied more than 2 billion treatments to meet increased requests from endemic countries scaling up MDA to meet the 2020 elimination targets for 10 NTDs including LF. Better diagnostic tools are now required to accelerate the mapping process that will enable MDA scale up. The Bill and Melinda Gates Foundation (BMGF) has provided additional funds to improve on point of care diagnostics (Weil et al. Reference Weil, Curtis, Fakoli, Fischer, Gankpala, Lammie, Majewski, Pelletreau, Won, Bolay and Fischer2013) and enhance capacity and logistical support for the WHO African regional team to accelerate the mapping process. At the same time countries that have completed 5 or more rounds of MDA and entering the surveillance phase will require more sensitive diagnostic tools for the verification of cessation of transmission, the so-called endgame scenario (Rebollo and Bockarie, Reference Rebollo and Bockarie2013a). This paper provides an update on the human diagnostic tests available for LF and their current applications as tools in mapping, transmission assessment surveys (TAS) and post-MDA surveillance. The values of entomological indicators and diagnostic methods applied to vector populations are also discussed.
PROGRAMMATIC STEPS FOR ELIMINATION OF LF
WHO recommends 4 sequential programmatic steps for the elimination of LF through MDA (Fig. 1): (1) mapping of the distribution of the disease to determine endemic implementation units for MDA; (2) performing MDA for at least 5 years to reduce prevalence and intensity of MFs infections so that the blood feeding mosquitoes are no longer able to ingest parasites that can develop to infective L3; (3) implementing post-MDA surveillance activities and (4) confirming interruption of transmission at the national level conducting TAS.
The 4 programmatic steps recommended for elimination of lymphatic filariasis.
CURRENT STATUS OF ENDEMICITY OF LF
By the end of 2012, 59 countries had completed mapping for LF and Eritrea was the only country yet to start the process. Mapping is ongoing in 13 countries including the large and highly populated nations of Nigeria, Democratic Republic of Congo (DRC) and Ethiopia. Completing mapping in Africa's most populous countries and confirming the cessation of LF transmission in countries undertaking TAS will significantly alter the current estimates for the number of people at risk for LF in the region. In 2002, 39 African countries were endemic for LF and no fewer than 477 million people were estimated to be at risk of the disease (WHO, 2003). However, according to GPELF's progress report for 2000–2009, the evidence for active transmission of LF in many of the 39 endemic countries was weak and some probably did not require MDA (WHO, 2011b). The statuses of 5 countries (Burundi, Cape Verde, Mauritius, Rwanda and Seychelles) were reviewed in 2011 and reclassified as non-endemic, reducing the number of endemic countries in Africa to 34 including South Sudan (WHO, 2012). In the same year, the International Task Force for Disease Eradication reviewed the eradicability of LF and onchocerciasis in Africa for the first time and concluded that LF could be eradicated in the region by 2020 if the mapping of LF distribution was accelerated and implementation units accurately delineated to enable MDA scale up (WHO, 2014). This necessitated the ‘Shrinking the Map’ project funded by the BMGF as a supplement to the ‘Filling the Gaps grant’ awarded to the NTD Support Centre in the Task Force for Global Health in Atlanta, USA. The objective of the project is to fill the current mapping gaps in the African region and complete mapping of 5 NTDs by 2015 (onchocerciasis, LF, trachoma, schistosomiasis and soil-transmitted helminthiasis) and serve as an operational research platform to roll out new diagnostic tools and algorithms and provide the necessary evidence for the utilization of those tools to accelerate the mapping and transmission monitoring of LF and other NTDs.
CURRENT DIAGNOSTIC TOOLS
When GPELF was launched in 2000, diagnostic tools for LF were limited to clinical examination, detection of antibodies to native-antigen preparations, observation of MF in venous night blood samples by microscopic examination of thick blood films or membranes resulting filtration of blood (Ramzy, Reference Ramzy2002).There has been a significant improvement in the traditional LF diagnostic tools in recent years and some new tools have been developed (Weil and Ramzy, Reference Weil and Ramzy2007; Molyneux, Reference Molyneux2009). Diagnostic tools currently available for mapping and impact assessment for LF have been described previously (Weil and Ramzy, Reference Weil and Ramzy2007). The tools currently used include blood films to detect the presence of microfilaraemia by microscopy; rapid diagnostic tests to detect circulating antigen to W. bancrofti adult worm; laboratory-based antigen enzyme-linked immunosorbent assay (ELISA); filarial antibody detection tests for Brugian filariasis and polymerase chain reaction (PCR) methods for detection of parasite DNA in humans and mosquitoes.
Blood films
Examining a stained blood slide for microfilaraemia can reveal whether a person has LF parasites in the peripheral blood. In areas with nocturnally periodic LF, a blood sample must be taken in the middle of the night, with a maximum density in blood between 10 pm and 2 am, and prepared as a thick blood smear. It is not a very sensitive method for detecting active infections because it misses low MF carriers that may be infectious to competent mosquito vectors. Night blood collection is also labour intensive and impractical in some areas. It is however specific, inexpensive and can be carried out in laboratories with limited facilities.
Immunochromatographic tests
Filarial antigen tests are available for detecting filarial antigens in human blood (Weil and Ramzy, Reference Weil and Ramzy2007). They can be used to test samples collected during the day or night and are more sensitive than the MF detection technique. Antigen to W. bancrofti can be detected in the laboratory using the antigen enzyme-linked immunosorbent assay known as Og4C3 ELISA. Nevertheless, a rapid diagnostic format in the form of an immunochromatography card test (ICT) is the main tool for mapping in areas endemic for W. bancrofti (WHO, 2011a). The ICT, which is as sensitive as the ELISA, is a quick (10 min) and easy to perform test requiring only 100 μL blood from a finger prick. However, it presents several problems like the requirement of cold chain (2–8 °C), the instability of the results if the card is not read at 10 min exactly, the short life of the cards (maximum 9 months if kept in optimal conditions of cold chain) and the fact that people may remain antigen positive for several months after treatment for LF as the adult worms and MF die and disintegrate (Bockarie et al. Reference Bockarie, Tisch, Kastens, Alexander, Dimber, Bockarie, Ibam, Alpers and Kazura2002; Reimer et al. Reference Reimer, Thomsen, Tisch, Henry-Halldin, Zimmerman, Baea, Dagoro, Susapu, Hetzel, Bockarie, Michael, Siba and Kazura2013).
Antibody tests
Early antibody diagnostic tests for LF based on detection of immunoglobulin G (IgG) antibodies to native parasite antigens were not highly specific (Weil and Ramzy, Reference Weil and Ramzy2007). Recently developed assays detecting IgG4 antibodies to recombinant filarial antigens, such as Bm14 and BmR1, are more specific (Weil et al. Reference Weil, Curtis, Fischer, Won, Lammie, Joseph, Melrose and Brattig2011; Wongkamchai et al. Reference Wongkamchai, Chiangjong, Sinchaikul, Chen, Choochote and Thongboonkerd2011; Makhsin et al. Reference Makhsin, Razak, Noordin, Zakaria and Chun2012; Noordin et al. Reference Noordin, Muhi, Md Idris, Arifin and Kiyu2012). The Bm14 antibody test is sensitive for infection with B. malayi and W. bancrofti, whereas the BmR1 antibody test is sensitive for Brugia infections only. Antibody prevalence rates tend to be higher than antigen or MF rates in young children in LF-endemic areas. Field studies have shown that antibody rates in young children decrease rapidly in the years following initiation of effective MDA (Weil et al. Reference Weil, Kastens, Susapu, Laney, Williams, King, Kazura and Bockarie2008). Testing for Brugian filariasis is based on recombinant antigen BmR1 that is marketed as the Brugia Rapid™ test that detects antibodies to B. malayi and B. timori (Fischer et al. Reference Fischer, Bonow, Supali, Ruckert and Rahmah2005; Weil and Ramzy, Reference Weil and Ramzy2007). However, the results of antibody testing do not distinguish between current and past infection. Urinary antibody tests have also been used to determine LF endemicity and assess exposure of children to infection with W. bancrofti but the tests are laboratory based and the antibodies are not available commercially (Samad et al. Reference Samad, Itoh, Moji, Hossain, Mondal, Alam and Kimura2013; Yahathugoda et al. Reference Yahathugoda, Weerasooriya, Sunahara, Kimura, Samarawickrema and Itoh2014).
Polymerase chain reaction
PCR assays are now commonly used to detect LF DNA in humans and mosquitoes (Ramzy et al. Reference Ramzy, Farid, Kamal, Ibrahim, Morsy, Faris, Weil, Williams and Gad1997; Bockarie et al. Reference Bockarie, Fischer, Williams, Zimmerman, Griffin, Alpers and Kazura2000; Ramzy, Reference Ramzy2002) but not used routinely in endemic countries as they require more advanced laboratory facilities for DNA extraction and processing using relatively expensive thermocyclers. Recent studies have shown that LF DNA can be detected in both vector and non-vector mosquitoes for 2 weeks or longer after they ingest MF positive blood (Fischer et al. Reference Fischer, Erickson, Fischer, Fuchs, Rao, Christensen and Weil2007). Thus, although molecular xenomonitoring (MX) may be sensitive for indirectly detecting LF in human populations, positive test results for parasite DNA in mosquitoes do not necessarily prove that transmission is ongoing in the area. New rapid, sensitive and specific methods are required for detection of mosquitoes that harbour infective larvae.
A large multi-country study was conducted during 2007–2008 to compare the effectiveness of the currently available human diagnostic tests described above (Gass et al. Reference Gass, Beau de Rochars, Boakye, Bradley, Fischer, Gyapong, Itoh, Ituaso-Conway, Joseph, Kyelem, Laney, Legrand, Liyanage, Melrose, Mohammed, Pilotte, Ottesen, Plichart, Ramaiah, Rao, Talbot, Weil, Williams, Won and Lammie2012). A total of 8513 people from the 6 countries implementing MDA participated in the study and specimens from these participants were used to conduct 49 585 diagnostic tests. Overall, the ICT test was found to be 76% sensitive at detecting and 93% specific at identifying individuals negative for both MF and antifilarial antibody; the Og4C3 test was 87% sensitive and 95% specific. The value of a semi-quantitative scoring of the ICT card test was recently evaluated during a field survey in the Republic of Congo and shown to be useful for grading the intensity of filarial infections in individuals and populations (Chesnais et al. Reference Chesnais, Missamou, Pion, Bopda, Louya, Majewski, Weil and Boussinesq2013).
NEW DIAGNOSTIC TOOLS
Immunochromatographic strip test
A new filarial antigen test (the Alere Filariasis Test Strip) was compared with the ICT in laboratory and field surveys (Weil et al. Reference Weil, Curtis, Fakoli, Fischer, Gankpala, Lammie, Majewski, Pelletreau, Won, Bolay and Fischer2013). The 2 tests were shown to have similar high rates of sensitivity and specificity and >99% agreement based on laboratory testing of 227 archived serum or plasma samples. However, the test strip was more stable and more sensitive detecting 26.5% more people with filarial antigenemia than the ICT in a field study conducted in Liberia where LF is highly endemic. The authors acknowledged that additional studies are needed to compare the performance of the strip and card tests in areas with low residual LF endemicity rates after multiple rounds of MDA.
Improved antibody tests
Earlier antibody tests have suffered from poor sensitivity and/or specificity or from a lack of standardization. In a multi-centre evaluation of a new commercial ELISA that detects IgG4 antibodies to the recombinant filarial antigen Bm14, a shared panel of coded serum or plasma samples including those from MF positive individuals was evaluated and shown to be useful not only for diagnosing LF in patients with clinical signs of filariasis but also for monitoring transmission intensity (Weil et al. Reference Weil, Curtis, Fischer, Won, Lammie, Joseph, Melrose and Brattig2011). Recombinant antigens Bm14 and Bm33 from B. malayi were coupled to carboxylated polystyrene microspheres as part of a multiplex bead assay (MBA) used to analyse serum samples collected longitudinally from children enrolled in a drug trial for treatment of filariasis in Leogane, Haiti (Moss et al. Reference Moss, Priest, Boyd, Weinkopff, Kucerova, Beach and Lammie2011). They showed that the MBA is an excellent serologic technique for multiple antigens that offers substantial advantages over single-antigen-based ELISAs in studies monitoring changes in antibody levels after MDA. Monitoring of LF transmission in Samoan villages was performed by measuring MF circulating filarial antigen (CFA) and antibody prevalence using the Filariasis Cellabs ELISA kit (CELISA), based on recombinant antigen Bm14, to detect antibodies. CELISA shown to be a promising diagnostic tool for inclusion in future surveillance in the South Pacific (Joseph et al. Reference Joseph, Maiava, Naseri, Silva, Lammie and Melrose2011). Hamlin et al. (Reference Hamlin, Moss, Priest, Roberts, Kubofcik, Gass, Streit, Nutman, Eberhard and Lammie2012) performed a longitudinal monitoring of the development of antifilarial antibodies and acquisition of W. bancrofti in a highly endemic area of Haiti measuring antibody responses to Wb123, a W. bancrofti L3 antigen, Bm14 and Bm33. Their results show that antifilarial antibody responses can serve as an important epidemiological indicator in a sentinel population of young children.
Steel and co-workers (Steel et al. Reference Steel, Golden, Kubofcik, LaRue, de Los Santos, Domingo and Nutman2013) recently developed a rapid IgG4 ELISA and a lateral-flow strip immunoassay based on Wb123 and was shown to have high sensitivities and specificities (93 and 97% [ELISA] and 92 and 96% [strips], respectively). The specificities of the assays, with regard to other filarial parasites (onchocerciasis, loiasis and strongyloidiasis) varied between 91 and 100% and both assays had high positive and negative predictive values.
Loop-mediated isothermal amplification (LAMP)
A relatively new single-tube technique for the amplification of DNA, the LAMP, assay has been used for the rapid detection of filarial genomic DNA (Aonuma et al. Reference Aonuma, Yoshimura, Perera, Shinzawa, Bando, Oshiro, Nelson, Fukumoto and Kanuka2009, Reference Aonuma, Yoshimura, Kobayashi, Okado, Badolo, Nelson, Kanuka and Fukumoto2010; Takagi et al. Reference Takagi, Itoh, Kasai, Yahathugoda, Weerasooriya and Kimura2011; Poole et al. Reference Poole, Tanner, Zhang, Evans and Carlow2012; Drame et al. Reference Drame, Fink, Kamgno, Herrick and Nutman2014; Fernandez-Soto et al. Reference Fernandez-Soto, Mvoulouga, Akue, Aban, Santiago, Sanchez and Muro2014). Amplification is detected using turbidity or fluorescence as readouts and results are generally obtained within 30 min. LAMP is an isothermal method that does not require expensive thermocyclers used in conventional PCR. It is a rapid, sensitive and highly specific method that has the potential for field application. It is also less sensitive than PCR to inhibitors in blood and insects. This robustness has been attributed to use of a different DNA polymerase (typically BstDNA polymerase rather than Taq polymerase as in PCR). Lamp assays have been developed for detecting filarial DNA in human blood and vector mosquitoes (Takagi et al. Reference Takagi, Itoh, Kasai, Yahathugoda, Weerasooriya and Kimura2011). Poole and co-workers (Reference Poole, Tanner, Zhang, Evans and Carlow2012) recently developed and evaluated a Brugia Hha I repeat LAMP assay for the rapid detection of Brugia genomic DNA. They demonstrated that the Brugia Hha I repeat LAMP assay is rapid, sensitive and Brugia specific with the potential to be developed further as a field tool for diagnosis and mapping of brugian filariasis. The LAMP assay was also used to confirm positive mosquito samples that were tested for infection with W. bancrofti using standard PCR techniques (de Souza et al. Reference de Souza, Sesay, Moore, Ansumana, Narh, Kollie, Rebollo, Koudou, Koroma, Bolay, Boakye and Bockarie2014).
Molecular xenomonitoring
The detection of filarial DNA in mosquitoes (MX) has been proposed as a sensitive marker for assessing the endemicity of LF and a useful tool for evaluating the success and progress of control programmes (Ramzy et al. Reference Ramzy, Farid, Kamal, Ibrahim, Morsy, Faris, Weil, Williams and Gad1997; Bockarie et al. Reference Bockarie, Fischer, Williams, Zimmerman, Griffin, Alpers and Kazura2000). Work in Egypt, Papua New Guinea and recently in Sierra Leone have shown that that a PCR-based method for detecting LF in mosquitoes can be used successfully to determine endemicity of LF and impact of interventions on transmission intensity (Ramzy et al. Reference Ramzy, El Setouhy, Helmy, Ahmed, Abd Elaziz, Farid, Shannon and Weil2006). Filarial DNA is detectable in both vector and non-vector mosquitoes and can persist for weeks after feeding on a MF positive individual (Fischer et al. Reference Fischer, Erickson, Fischer, Fuchs, Rao, Christensen and Weil2007). Therefore, finding positive mosquitoes by MX does not necessarily indicate ongoing transmission. Stage-specific assays have now been developed for the detection of mosquitoes that harbour infective larvae. The assay detects an L3-activated mRNA transcript by reverse-transcriptase PCR (RT-PCR) and RNA Later has been used to preserve specimens that can be analysed later in the laboratory (Laney et al. Reference Laney, Buttaro, Visconti, Pilotte, Ramzy, Weil and Williams2008, Reference Laney, Ramzy, Helmy, Farid, Ashour, Weil and Williams2010).
CONCLUSION
New and improved rapid diagnostic tools are central to achieving the objective of shrinking the LF map to zero endemic countries by 2020 (Rebollo and Bockarie, Reference Rebollo and Bockarie2013a). They are essential for the mapping and impact assessment processes that inform programme managers about where drugs should be targeted to initiate MDA, to determine when MDA can be stopped after transmission has been interrupted and ensure strong surveillance procedures post-MDA for early detection and action in the event that transmission resumes as we move towards global elimination (Rebollo and Bockarie, Reference Rebollo and Bockarie2013b). However, some of these new tests, including antigen detector, will be of limited value in determining when transmission has been interrupted because antigens from adult worms will persist in previously infected people long after transmission has been interrupted. In areas where Lf transmission has been interrupted, antibody tests could be useful in detecting early exposure before patent infection is established. This will allow measures that can prevent recrudescence. Moreover, many molecular assays currently used for LF diagnostics, including LAMP and PRC, are laboratory based and lack some elements that can ensure quality assurance in resource poor settings.
FINANCIAL SUPPORT
Studies carried out by both authors are supported by the Centre for Neglected Tropical Diseases, Liverpool School of Tropical Medicine, through funding from DFID, UK. DFID had no role in the design, data collection and analysis, decision to publish or preparation of the manuscript.
