Utilizing the ultrasensitive Schistosoma up-converting phosphor lateral flow circulating anodic antigen (UCP-LF CAA) assay for sample pooling-strategies

Schistosomiasis is a chronic and debilitating disease, most prevalent in Africa but also in other parts of the tropics and subtropics [1]. The disease is caused by an infection with the helminth Schistosoma. Generally, different species infecting humans are geographically localized depending on the presence of species-specific freshwater snail vectors. The main human-targeting schistosome species of concern are S. haematobium and S. mansoni in Africa, S. japonicum and S. mekongi in Asia [2, 3]. Control strategies, such as preventive chemotherapy, mass drug administration, MDA, with the anti-schistosomal drug praziquantel (PZQ) treating school children, are being used in high Schistosoma endemic settings to reduce the burden of infection, prevent development of severe morbidity, and decrease prevalence [4–6]. This approach is usually accompanied by improved access to clean water, sanitation and hygiene (WASH; [7]), as well as snail control and community education. Accurate mapping of the disease and monitoring efficiency of different approaches to reduce morbidity and to break transmission still mostly depend on determination of eggs in urine (urine filtration technique [8]; for urogenital schistosomiasis: S. haematobium) or stool (Kato-Katz technique [9]; for intestinal schistosomiasis: S. mansoni, S. japonicum and S. mekongi) by microscopy. These assays have a high specificity, but results are often variable and lack sensitivity in areas with low worm burden and near-elimination, transmission-stop, settings [10–13]. Moreover, in cases where drug treatment only temporarily suppresses egg production infections are easily missed; in this context, recovery of affected adult worms or maturation of young worms that are less susceptible to the drug [14, 15] as well as rapid reinfections, are known concerns [16]. As deposition of the eggs in either stool or urine depends upon the Schistosoma species, in areas of mixed infections or when dealing with unknown species, both stool and urine need to be examined making mapping of low endemic areas a laborious and time-consuming enterprise. Therefore there have been repeated calls for highly accurate genus specific diagnostics with all species detectable in the same non-invasively collectable biological matrix and preferably deployable in the field [12, 17–19].

For a number of years, a well validated laboratory-based test with excellent accuracy using urine has been available for detection of active infections of all different Schistosoma species [20, 21]. This test, an Up-Converting Phosphor Lateral Flow (UCP-LF) based assay [22], detects the Schistosoma Circulating Anodic Antigen (CAA) which is a glycosaminoglycan-like carbohydrate with a unique structure [23]. The CAA test should not be confused with the commercially available, field-friendly, Point-Of-Care Circulating Cathodic Antigen (POC-CCA) test which detects a different Schistosoma antigen [24]. The POC-CCA detects a ‘cathodic’ antigen [25] in urine whereas the UCP-LF CAA test detects an ‘anodic’ antigen in blood and urine and can be used in any other body fluid. The indication ‘cathodic’ and ‘anodic’ relates to their migration behaviour in an electric field [26]. Both antigens are regurgitated by Schistosoma worms in the blood stream of the human circulation [27, 28] and end up in urine; the rapid clearing mechanisms have not been determined in full detail. The POC-CCA test was developed as a rapid, field-friendly assay for detection of S. mansoni infections in urine and is widely used for mapping and monitoring as an excellent alternative to the Kato-Katz (KK) stool analysis [29, 30]. POC-CCA is a qualitative assay allowing visual interpretation of the test results. CCA structurally has some homology with the class of Lewis X antigens which may limit the specificity of the POC-CCA. In contrast, the UCP-LF CAA test was developed as a sensitive quantitative test for detection of all Schistosoma species, and current test formats allow detection of a single worm pair, or single sex worm infections [31–33]. The UCP-LF CAA test includes a sample preparation step and a UCP-LF strip reader to analyse the test result. The pooling applications described here are only applicable for CAA, as its unique biochemical structure allows almost unlimited concentration from urine samples.

The sample preparation step comprises an acidic treatment of the urine sample with trichloroacetic acid (TCA), leaving CAA in a protein-free supernatant after centrifugation. The resulting clear solution can be concentrated to increase sensitivity using dedicated molecular weight cut-off filtration devices. When using urine as a clinical sample, high concentration factors can be achieved [33]. Presence of CAA in urine is indicative of an active ongoing infection [34]. Measurements correlate qualitatively with eggs detected in stool for intestinal schistosomiasis (S. mansoni and S. japonicum; [34–36] or eggs detected in urine samples for urogenital schistosomiasis (S. haematobium, [27]). Importantly, CAA concentrations can be used as an accurate and direct measure of worm burden [35]. The correlation may vary somewhat between different species, variability in urine production or factors such as differences in immune status of the host, chronicity and intensity of the infection, or reduced metabolic activity i.e. worms recovering from drug treatment. Considering that urine can be obtained non-invasively and in large volume, it is a very practical clinical sample also in terms of sample pooling. An additional advantage is that CAA in urine is stable for several days or weeks at ambient temperature, thus not requiring a cold-chain when testing is performed away from the sample collection site, e.g. in a centralized laboratory. The test has been validated in several scientific studies, worldwide under different conditions [31, 37–39]. Based on experimental data, the assumption is that analysis of the concentrate of 2 mL of urine with the UCP-LF CAA assay with an analytical sensitivity < 0.1 pg/ml will identify the majority of active infections.

In this manuscript large scale urine pooling is considered and potential applications are discussed that can be relevant to mapping, hot spot detection, mass drug administration (MDA), efficiency of drug treatment, re-emergence of infection, monitoring transmission stop, surveillance, and reduction of the number of diagnostic tests when requiring identification of infected individuals in low prevalence settings.

The current manuscript describes a methodological approach for using different formats of the UCP-LF CAA test in ‘diagnostic’ applications with pooled urine samples. This methods section describes the relevant and practical details required to perform the test. Source information regarding required materials and equipment is given at the end of the result section. Actual practical data as determined from a random available sample set before and after drug treatment serves the purpose of demonstrating validity of accurate testing urine pools with the UCP-LF CAA test.

Procedure of the Schistosoma genus specific CAA test

General description of the UCP-LF CAA strip test

The assay can be performed with all body fluids including dried blood samples [40] and is optimized for samples extracted with trichloroacetic acid (TCA) [22]. The TCA extraction separates antigen-antibody immune complexes and precipitates protein material while leaving CAA and possibly other carbohydrate structures in solution. Following extraction, centrifugation-clarified material (TCA-sup) is mixed and incubated (1 h, 37 °C, 900 rpm) with the luminescent UCP reporter coated with mouse monoclonal anti-CAA antibodies (UCP-MαCAA). This pre-flow incubation improves sensitivity when using ‘large’ (400 nm) YOS reporter particles [41], but may be omitted upon implementation of novel smaller (< 100 nm) NaYF particles [42]. The amount of sample input can be increased by concentrating TCA-sup using Amicon filtration devices [33]. UCP-MαCAA/TCA-sup mixtures are examined by immunochromatography on LF strips with a test line comprised of mouse monoclonal anti-CAA antibodies [22]. Test results are analysed quantitatively by scanning LF strips for UCP reporter signals with dedicated strip readers. A summary of the different steps is shown in Fig. 1. A more detailed description of the different formats of the assay can be found in previous publications [20, 33, 37]. The description below is focused on testing urine samples, in particular the TCA extraction and sequential concentration steps to increase sensitivity as these form the basis of the pooling applications.

TCA extraction of urine samples

Extraction of the clinical samples is performed at a final concentration of 2% (w/v) TCA. Protocols optimized for serum analysis require mixing of equal volumes of serum and 4% (w/v) TCA [22]. The same protocol is applicable for urine, but extraction of a low viscosity sample allows mixing with a higher percentage TCA solution. Up to 40% (w/v) TCA can be used, although care should be taken as it may lead to some gas formation when patients are burdened with urolithiasis. After thorough mixing with TCA, samples are left at ambient temperature for 5 min for efficient denaturation of proteins after which the mixture is clarified by centrifugation; the carbohydrate structure CAA remains in the soluble fraction, TCA-sup. As an alternative for centrifugation, a combination of disposable glass- and paper-filters could be applied for removal of the TCA-precipitate.

Concentration of urine TCA-sup

To improve sensitivity TCA-sup can be concentrated using 10 kDa molecular weight cut-off membranes, available as Amicon ultra-centrifugal filtration devices up to 15 ml or larger cell pressure-based filtration devices (> 50 ml). The maximum level of concentration is depending on the final viscosity of the concentrated TCA-sup. The TCA-sup from urine samples does allow rigorous concentration; e.g. using the 15 ml Amicon centrifugal device a 1400-fold concentration can be obtained (see below UCP-LF CAA assay formats: UCAA1400).

The UCP-LF CAA assay for serum and urine has been reported in many scientific studies and has demonstrated superior diagnostic performance allowing accurate determination of true prevalences with close to 100% specificity and sensitivity [31, 33]. The main factors accountable for its high (analytical) sensitivity are: i) the high affinity of the anti-CAA mouse monoclonal IgG antibody; ii) the unique structure and stability of the CAA biomolecule and its solubility in trichloroacetic acid allowing concentration, and; iii) the ultra-sensitive background-free UCP reporter particle technology. Ongoing efforts are focussing on adapting the current user-friendly assay format, which requires some basic laboratory equipment to allow rapid point-of-care applications as needed for test-and-treat approaches in low endemicity and post MDA settings. Here we report and discuss another aspect, the potential of the current UCP-LF CAA assay format with pooled urine samples to obtain population-based test results and to identify individual low grade infections.

Several studies have indicated the applicability of the ultra-sensitive UCP-LF CAA test platform to identify individuals with low Schistosoma infection levels using non-invasive urine samples. In the majority of the cases, the measured CAA concentration can be used as a reliable guide for worm burden and correlates to egg production and subsequently to transmission. Although it must be recognized that CAA levels cannot yet be directly related to numbers of worms per infected individual, hence they do not always reveal the actual parasitology and clinical situation at the individual level, the presence of CAA always indicates an ongoing/active infection. Renal clearance of CAA from human circulation is only a matter of hours to few days [36, 49]. Taking into account differences between individuals and worm species, CAA concentrations before and after PZQ administration, longitudinal studies can provide relevant information about the short- and long-term drug efficacy, as well as other mechanism leading to continuing worm infections. Two formats of the UCP-LF CAA test, the SCAA500 (for serum) and the UCAA2000 assay (for urine), have been applied in several studies [31–33, 38, 44] and may detect a single worm pair. Although desired, the assays (yet) don’t support actual field POC testing mainly because of the requirement of centrifuges in the sample preparation step. The current format, developed with the goal to make the assay as sensitive as possible, also includes the use of centrifugal filtration devices. These devices are convenient for use in the laboratory, but are costly and difficult to implement in real field applications. Ongoing efforts are now focussing on less expensive and more field-friendly alternatives for concentration and/or extraction. Meanwhile, and in particular for urine testing, pooling strategies using the current formats to reduce the number of tests and associated costs are definitely feasible.

Although the theoretical examples of potential pooling approaches described here are unpretentious representations of the real situation, they indicate suitability of the UCP-LF CAA test for applications involving pooling approaches in high as well as low endemic settings (including the transmission-stop and elimination areas). For stratified mapping, monitoring of MDA efficiency and re-infection, the pooling approach can provide a rapid quantitative value on which actions can be based. Clearly, studies are needed to investigate optimal pool-sizes in relation to expected endemic levels, as well as the effect of hot-spots, incidental super-excretors, etc. In high endemic settings this type of pooling approach is straightforward and probably well acceptable by the ‘diagnosed community’ even if no individual test results are provided. However, additional studies demonstrating that average CAA levels determined from pools correlate with the average CAA levels determined from the individuals are needed; the data presented here for the Senegal selection successfully demonstrated pooling competence. When moving towards low endemicity, transmission stop and even elimination settings, it is obvious that the amount of each individual sample (urine volume) added to the pool needs to be increased with evident increase on the total volume of the urine pool. Where in high endemicity settings individual urine volumes of 10 μl (or even less) can be sufficient, in elimination settings individual samples of 2000 μl may be required to identify a single individual with a low infection level (single worm) in the pool. Pooling in such cases will still be efficient and save diagnostic costs, but it needs to be considered that in these situations identification of the infected individual is likely a requirement for individual drug treatment.

We propose that at the population level, average CAA concentrations in urine can be an appropriate measure to indicate worm burden. Urine pooling strategies allowing this type of large scale testing are feasible with the various UCP-LF CAA test formats. Overall mapping, stratified testing of high risk groups (e.g. based on age, profession or habitat), and identification of hot spots, can be done easily with only a limited number of diagnostic tests. The determined population-based CAA level can provide essential data and impact treatment considerations. Moreover, as drug efficacy can be monitored through a decrease in CAA levels measured before and shortly (within days/weeks) after PZQ treatment, rapid identification of areas (becoming) insensitive to the applied MDA approach (including development of potential drug resistance) is practicably feasible. Finally, urine pooling protocols while maintaining the highest sensitivity (required in low endemicity setting as transmission-stop and elimination sites) and at the same time allowing identification of the infected individual(s) can be designed and relevantly reduce the number of diagnostic tests and related costs.