Sensitivity and specificity of human point-of-care circulating cathodic antigen (POC-CCA) test in African livestock for rapid diagnosis of schistosomiasis: A Bayesian latent class analysis

JournalFeeds

Authors:
Beatriz Calvo-Urbano , et al.

Abstract

Schistosomiasis is a major neglected tropical disease (NTD) affecting both humans and animals. The morbidity and mortality inflicted upon livestock in the Afrotropical region has been largely overlooked, in part due to a lack of validated sensitive and specific tests, which do not require specialist training or equipment to deliver and interpret. As stressed within the recent WHO NTD 2021–2030 Roadmap and Revised Guideline for schistosomiasis, inexpensive, non-invasive, and sensitive diagnostic tests for livestock-use would also facilitate both prevalence mapping and appropriate intervention programmes. The aim of this study was to assess the sensitivity and specificity of the currently available point-of-care circulating cathodic antigen test (POC-CCA), designed for Schistosoma mansoni detection in humans, for the detection of intestinal livestock schistosomiasis caused by Schistosoma bovis and Schistosoma curassoni. POC-CCA, together with the circulating anodic antigen (CAA) test, miracidial hatching technique (MHT) and organ and mesentery inspection (for animals from abattoirs only), were applied to samples collected from 195 animals (56 cattle and 139 small ruminants (goats and sheep) from abattoirs and living populations) from Senegal. POC-CCA sensitivity was greater in the S. curassoni-dominated Barkedji livestock, both for cattle (median 81%; 95% credible interval (CrI): 55%-98%) and small ruminants (49%; CrI: 29%-87%), than in S. bovis-dominated Richard Toll ruminants (cattle: 62%; CrI: 41%-84%; small ruminants: 12%, CrI: 1%-37%). Overall, sensitivity was greater in cattle than in small ruminants. Small ruminants POC-CCA specificity was similar in both locations (91%; CrI: 77%-99%), whilst cattle POC-CCA specificity could not be assessed owing to the low number of uninfected cattle surveyed. Our results indicate that, whilst the current POC-CCA does represent a potential diagnostic tool for cattle and possibly for predominantly S. curassoni-infected livestock, future work is needed to develop parasite- and/or livestock-specific affordable and field-applicable diagnostic tests to enable determination of the true extent of livestock schistosomiasis.

Author summary

Schistosomiasis is a neglected tropical and zoonotic disease, infecting over 230 million people and millions of animals worldwide. The potential contribution of livestock schistosomiasis to disease transmission in human populations has implications for the design of effective disease management and elimination programmes. However, our understanding of the true prevalence and impact of animal schistosomiasis is severely limited, in part due to a lack of accessible and accurate diagnostic tools. This need for sensitive and specific tools for animal schistosomiasis diagnosis has been recognised in the most recent WHO Guideline and NTD roadmap. As a point-of-care circulating cathodic antigen (POC-CCA) diagnostic test is now available to assess intestinal schistosomiasis caused by Schistosoma mansoni in humans, we hypothesised that POC-CCA could be used to detect livestock intestinal schistosomiasis caused by Schistosoma bovis and Schistosoma curassoni. The aim of this study was to evaluate the sensitivity and specificity of POC-CCA for the detection of intestinal livestock schistosomiasis in Senegal. POC-CCA sensitivity varied by ruminant group and by parasite species/location, while POC-CCA specificity in small ruminants, at least, did not vary across populations. We conclude that, whilst the currently-available POC-CCA does represent a potential diagnostic tool for schistosomiasis in cattle, the factors determining test performance warrant further investigation and further livestock-specific assays would be ideal.

Introduction

The development and application of sensitive and specific diagnostic techniques for detection of infectious diseases is vital for the monitoring and evaluation of all treatment programs in endemic areas, especially whenever considering elimination and/or drug-resistance pharmacovigilance. Within this, point-of-care diagnostic testing is particularly needed wherever there is a necessity for a fast diagnostic outcome that is independent from sophisticated, time-consuming, labour-intense and/or expensive laboratory procedures [1]. This need may be most exemplified for the neglected tropical and zoonotic diseases (NTDs/NZDs), as clearly stressed within the recent World Health Organization’s (WHO) NTD Roadmap for 2021–2030 [2] and revised WHO Guideline for the control and elimination of human schistosomiasis, most notably that of recommendation 6 [3, 4].

Schistosomiasis is one of the major debilitating NTDs/NZDs, caused by snail-borne dioecious Schistosoma spp. blood-flukes. Approximately 90% of the 230 million people infected worldwide live in the Afrotropical region [2]. Animal schistosomiasis is also of major veterinary and socio-economic importance, causing significant mortality and morbidity to livestock, as well as reduced productivity for their owners, although the contribution and consequences of this within the Afrotropical region are only just beginning to be realised [2, 58]. The main Schistosoma species found amongst livestock in Africa are the intestinal Schistosoma bovis and Schistosoma curassoni in West Africa and Schistosoma mattheei in East Africa. These species are phylogenetically close to the human urogenital parasite Schistosoma haematobium and have been found to regularly form viable hybrids within humans, both across Africa [813], and even within its recent expanse to Europe [14, 15]. Outside of Asia, the contribution of zoonotic schistosomes to human schistosomiasis cases have been largely ignored, despite prevalence levels in humans often remaining unacceptably high following high coverage mass drug administration (MDA) programmes across much of West Africa in particular [8, 16, 17]. Furthermore, recent work combining epidemiological, molecular and mathematical modelling work has demonstrated that the relative role of zoonotic transmission from livestock in Africa is likely to increase as disease control efforts move towards elimination [18].

The recently launched WHO NTD 2021–2030 Roadmap and Guideline for the control and elimination of human schistosomiasis therefore poses the question of anti-schistosomiasis treatment of livestock across Africa in order to achieve the new targets of Elimination as a Public Health Problem (EPHP) in all 78 currently-endemic counties and Interruption of Transmission (IoT) in selected African regions by 2030 [2, 3]. Widespread, indiscriminate use of anthelmintics in livestock have increased drug-resistance [19, 20]. This is particularly pertinent for schistosomiasis as there is currently only one efficacious drug, praziquantel (PZQ), available for both humans and livestock. Recent work has highlighted examples of use and misuse of PZQ in livestock, including, but not exclusive to, PZQ tablets donated free via the WHO MDA programmes for school-aged children being used instead for infected livestock, with little knowledge of application or dosage requirements [5, 7]. Furthermore, recent surveys and socio-economic analyses have revealed highly important, if also often overlooked, animal welfare, productivity, and financial impacts of animal schistosomiasis, which further jeopardize livelihoods, food security and nutrition amongst neglected communities [5]. Indeed the financial costs incurred to subsistence farmers of infected animals was projected to be significantly greater than would be those of a theoretical test and treat programme [5].

To promote sustainable livestock schistosomiasis prevention and treatment in the Afrotropical region, there is therefore a need to both easily and inexpensively diagnose animal schistosomiasis, and subsequently effectively treat only infected individuals and/or herds, such as through a targeted test-and-treat (TnT) or T3: Test, Treat, Track design for livestock schistosomiasis in the Afrotropical region, which anchors the key recent WHO policy recommendations on diagnostic testing, treatment and surveillance in general.

There are, however, currently limited diagnostic techniques that can detect schistosomiasis in livestock with sufficient levels of sensitivity and specificity, as well as logistical ease, that can fully inform disease management decisions [21]. Of those available, a recent systematic review (although data analysed were exclusive to Schistosoma mansoni and Schistosoma japonicum infections) has recommended the formalin-ethyl acetate sedimentation-digestion with quantitative polymerase chain reaction, as the most promising field-applicable techniques in non-human animal hosts [22], although both are time consuming and/or expensive. A recent extensive field survey of both abattoir and live-sampled livestock within Senegal found some utility with both the Kato-Katz technique and miracidial hatching technique (MHT), but the latter was again labour-intensive and showed significant differences in sensitivities by host and/or parasite species [8].

Tests for antigens, rather than antibodies, are preferred in endemic areas due to the lag in clearance of parasite-specific antibodies after infection subsides [23] and therefore immunochromatographic circulating cathodic antigen (CCA) and circulating anodic antigen (CAA) tests for worm antigens in urine (or serum) were developed specifically for current Schistosoma spp. infections amongst humans [24]. Whilst a high sensitivity laboratory-based lateral flow (LF) test platform utilizing luminescent up-converting reporter particles (UCP) comprises assay formats for quantitative detection of circulating cathodic and anodic antigen detection in urine (respectively, UCCA and UCAA assays) is available [25, 26], rapid point-of-care testing with visual detection is currently only available for the CCA (POC-CCA, Rapid Medical Diagnostics, Pretoria, South Africa) [27, 28]. Given its utility and field-friendly application, this semi-quantitative (negative, trace, “single positive” +, “double positive” ++ or “triple positive” +++) POC-CCA is now recommended by the WHO for mapping human S. mansoni prevalence in endemic areas [29, 30]. The application for detection of human urogenital S. haematobium infections with POC-CCA is less efficient i.e. to find the lower trematode burden infections [3134].

The POC-CCA has been used to assess S. mansoni infection in one non-human-primate study [35]). Given that S. bovis and S. curassoni (as well as S. mattheei) are intestinal schistosomes of livestock, and taking into account our recent social surveys which showed a demand amongst livestock owners within Western Africa [5], we predicted that the POC-CCA, when used in partnership with local veterinary technicians, could be a cost-effective tool for intestinal livestock schistosomiasis control. The aim of this study was therefore to evaluate the sensitivity and specificity of the currently available, human-focused, POC-CCA for the detection of intestinal livestock schistosomiasis caused by S. bovis, S. curassoni and their hybrids in Senegal, by host species, in relation to traditional and novel alternative parasitological and immunological diagnostic methods currently available, employing Bayesian latent class models.

Methods

Study sites and diagnostic methods

The study was conducted in Senegal, West Africa. Cross-sectional livestock parasitological surveys were conducted from May to August 2016 and from October 2017 to January 2018 in two areas located in the north of the country; the town of Richard Toll, in the Senegal River Basin, and villages around the Lac de Guiers (area hereafter referred to as Richard Toll), and the town of Linguere and villages around Barkedji along the Vallée du Ferlo (area hereafter referred to as Barkedji). Following the construction of the Diama Dam in 1986, situated on the Senegal River over 100 km downstream from Richard Toll, the environment surrounding this location has undergone important permanent alterations. Desalination, creation of irrigation canals and creation of permanent freshwater bodies have facilitated the expansion of Schistosoma snail intermediate host and human-livestock water contact points throughout the year, supporting the co-occurrence and interspecific interactions between S. haematobium, S. mansoni, and other Schistosoma spp. of veterinary importance [3638]. The main livestock schistosome species circulating in Richard Toll is S. bovis and cattle is the most affected host species. The Barkedji area presents temporary water sources that disappear during the dry season, leading to important seasonal migration of livestock-keeping communities, and annual interruption in schistosomiasis transmission. Small ruminants are the most important livestock host species, infected by S. curassoni. However, both schistosome species and hybrids are present in the two areas and all livestock species can be found infected to a lesser extent. Full details on the infection prevalence amongst definitive and intermediate hosts, together with the habitat of two areas, can be found in [8].

Animal sampling was part of a larger survey conducted in these two areas [8]. All animals (cattle, sheep, and goats) routinely slaughtered as part of the normal work of the abattoirs and available for inspection at the time of the surveys were examined post-mortem. Living animals were randomly selected in each village with the initial randomisation carried out at the unit level (owner in this case) and a maximum of five animals of each species (cattle, sheep and goats) then randomly sampled from each selected owner (fewer if the owner had less than five animals). Randomisation was carried out using random number generators. The mesenteric vessels of slaughtered animals were visually inspected for Schistosoma spp. adults (single males, single females, and paired worms) and those found were stored in RNA-later for molecular analysis (see Sections 1 and 2 in S1 Appendix). Faecal, urine, lung, and liver samples were collected post-mortem and were examined for infection using the miracidial hatching technique (MHT) (see Section 1 in S1 Appendix). Free-swimming miracidia were individually pipetted onto Whatman FTA cards (GE Healthcare Life Sciences, UK) for deoxyribonucleic acid (DNA) storage and subsequent genotyping (see Section 2 in S1 Appendix). Only faecal and urine samples were obtained from live animals. Faecal samples were assessed for the presence of Schistosoma spp. eggs via two Kato-Katz (KK) slides and MHT (see Sections 1 and 2 in S1 Appendix). Animals for which a sufficient volume of urine (≥15 mL) was collected were tested on-site for schistosomiasis with a single POC-CCA cassette (Rapid Medical Diagnostics, Pretoria, South Africa) and for haematuria with a single Hemastix strip (Siemens Healthcare Diagnostics, Surrey, UK) (see Section 1 in S1 Appendix). The remaining urine was frozen and transported to Leiden Medical University Centre (LUMC) in the Netherlands for application with the up-converting phosphor-lateral flow (UCP-LF) based assays formats (UCCA and UCAA) for the detection and quantitation of circulating cathodic and anodic antigen in urine (see Section 1 in S1 Appendix) [26]. Definitions of positive results are described in Section 1, Table A in S1 Appendix.

Statistical analyses

Association between haematuria and POC-CCA results.

When extracting urine from abattoir animals, some samples became contaminated with blood. As false-positive POC-CCA tests in humans have been associated with haematuria [39], it was of interest to assess whether POC-CCA results were affected by the blood in urine. Hemastix results were dichotomised as negative (score 0) or positive (scores 1 and 2) and by means of logistic regression we tested the hypothesis that POC-CCA results were independent of blood-contamination status.

Composite reference standard and Bayesian latent class model specification.

Due to the absence of a gold standard for the diagnosis of livestock schistosomiasis against which we could evaluate the performance of POC-CCA, we derived a composite reference standard (CRS) that was based on three additional diagnostic methods, namely MHT, KK and UCAA. CRS is based on a combination of tests with moderate sensitivity and high specificity [40]. CRS results were considered positive if animals tested positive for either UCAA, MHT or KK. CRS results were assumed to be negative if animals tested negative for all three tests.

We then developed a Bayesian latent class model (BLCM) to assess the sensitivity and specificity of POC-CCA for the detection of active schistosomiasis in both large (cattle) and small (sheep and goats) ruminants [4143] that took account of the imperfect reference diagnostic tests employed. Two latent (i.e. non-observed) classes were assumed, corresponding to either an infected or non-infected status, and these classes were related to the outcomes of POC-CCA, CRS and UCCA by means of a multinomial distribution [41, 44, 45]. We developed a two-test model and a three-test model (see Section 3, Table B in S1 Appendix). The diagnostic tests included in the two-test model were POC-CCA and CRS. The three-test model made use of the UCCA results and comprised POC-CCA, CRS and UCCA outcomes, albeit at the expense of lower sample sizes in each combination of diagnostic tests.

Accuracy model assumptions.

Latent class models (LCM), as proposed by Hui and Walter [42], involve three assumptions. Firstly, more than one population needs to be assessed, each with distinct disease prevalence. Secondly, diagnostic test accuracy must remain constant across populations. Thirdly, the accuracy of the tests must be conditionally independent, so that the sensitivity or specificity of one test is independent of the results of a second test [41, 44, 45]. In the present study we were planning to model abattoir and live animals as two distinct populations. However, as schistosomiasis observed prevalence in these two populations were similar (see Section 4, Table D in S1 Appendix), it was likely that the assumption of different prevalence levels amongst populations was not satisfied. Hence, abattoir and live data were combined and a one population approach was adopted for each site and host species group [43]. However, and in order to assess whether test accuracy differed in these two populations, independent abattoir and live estimates were subsequently derived and compared. To help overcome potential identifiability problems that a one population approach could entail and taking into account the assumed high specificity and medium sensitivity of CRS, we employed moderately informative priors for the CRS (see Section 3, Table C in S1 Appendix). Equally, prevalence priors were moderately informative and their values were based on results from a study carried out in the same area [8]. The impact that these moderately informative priors had on the accuracy estimates (sensitivity/specificity) was assessed through a sensitivity analysis (see Section 6 in S1 Appendix).

Conditional independence of POC-CCA and CRS in the two-test model was assessed in two ways. Firstly, the Deviance Information Criterion (DIC) [46] of model variants that included (dependence) and excluded (independence) covariance terms were estimated and compared. Models with lower DIC were preferred over models with higher DIC. Secondly, 95% credible interval (CrI) of covariance parameters were assessed [45], with CrIs intersecting zero indicative of conditional independence. Three-test models included dependence terms for POC-CCA and UCCA only as the two tests measure the same antigen (CCA) and their outcomes may not be independent of each other. Furthermore, the covariances between POC-CCA and CRS had been found not to be relevant (see results section “POC-CCA accuracy”).

Selection of priors and priors’ sensitivity analyses.

The beta distributions used to model the multinomial distribution parameters (Section 3, Table C in S1 Appendix) were parameterized by specifying the mode and the minimum/maximum accuracy value that was believed to be true for each variable, with 95% certainty. Priors for UCAA and UCCA sensitivity and specificity were established based on expert knowledge and published records of high performance [4750]. Priors for POC-CCA accuracy were non-informative beta (1,1) distributions. Prevalence priors were based on post-mortem examination of abattoir specimens from a previous study [8], where 81% and 82% of cattle were found to be infected in Barkedji and Richard Toll, respectively, and 26% and 16% of small ruminants were found to be infected in Barkedji and Richard Toll, respectively. As the proportion of infected animals were similar at the two sites, only one prior distribution per ruminant group was defined (Section 3, Table C in S1 Appendix). The sensitivity of the results to the prior parameterisations was undertaken by re-estimating POC-CCA accuracy assuming more diffuse priors for prevalence and CRS accuracy (see Section 6, Table F in S1 Appendix).

Results

Descriptive statistics

Abattoir surveys were carried out on 89 animals routinely slaughtered, whilst live animal samples were obtained from 106 cattle, sheep and goats from communities within the study area. Given the low number of goats and sheep that were infected in the present study, their data were grouped and analysed as “small ruminants” (see Table 1). A total of 195 animals were surveyed (56 cattle and 139 small ruminants). POC-CCA results were missing from four cattle and one sheep. The distributions of samples across sites and sources of animals (live or abattoir) are shown in Table 1. Barkedji livestock was primarily infected by S. curassoni, whilst animals from Richard Toll shed mainly S. bovis eggs (see Section 5, Table E in S1 Appendix).

Cross-tabulated results

The cross-tabulated results for CRS and POC-CCA are shown in Table 2. The number of small ruminants’ discordant pairs (POC-CCA -, CRS +) was relatively large, suggesting that POC-CCA sensitivity in small ruminants might be low. Cattle were CRS negative in 8 out of 56 animals, showing that most animals were infected, which limited our ability to estimate POC-CCA specificity in cattle. The number of (+, +) concordant pairs was greater in cattle than in small ruminants and the number of (-,-) concordant pairs were greater in small ruminants than in cattle, suggesting that the test behaved differently in each ruminant group.

POC-CCA accuracy

Conditional dependence between tests and selection of the best fitting models.

BLCM results from the two-test dependence and two-test independence models, by site and ruminant group (Table 3), indicate that POC-CCA and CRS are conditionally independent as: 1) the 95% CrIs of the covariance terms in the two-test model intersected zero; and 2) independence models had lower DIC values than the dependence ones. Results from the three-test model indicate that POC-CCA and UCCA sensitivities were conditionally dependent (lower 95% CrI limit was greater than zero) in all cases with the exception of the small ruminants from Richard Toll (where the CrI contained zero) (Tables 3 and 4). By contrast, POC-CCA and UCCA specificities were conditionally independent, their 95% CrI intersecting zero (Tables 3 and 4). Given the DIC of the two-test independence models were the lowest, we adopted these as the best fitting models and the following conclusions on POC-CCA accuracy were derived from them.

Direct diagnostic and POC-CCA accuracy.

Table 5 shows that the sensitivity and specificity results derived from BLCM that included POC-CCA and CRS (live and abattoir animals) were comparable to those obtained from BLCM that comprised POC-CCA and direct diagnostic (abattoir animals only). The main difference between the two approaches being that sensitivity in small ruminants in Richard Toll could not be assessed with precision when using direct diagnostic.

Visual comparisons (Fig 1) of the accuracy results obtained in this study with those of the routinely-employed diagnostic techniques, MHT and KK, reported from Senegal [8], suggest that POC-CCA sensitivity was better than that of MHT and KK in Barkedji cattle and small ruminants, where S. curassoni was the main parasite species, and better than KK but worse than MHT in Richard Toll cattle, where S. bovis was the main parasite. POC-CCA sensitivity performed worse than MHT and KK in Richard Toll small ruminants.

thumbnail

Fig 1. POC-CCA sensitivity (% median and 95% CrI).

POC-CCA sensitivity based on composite reference standard (CRS), POC-CCA based on direct diagnostic (Direct), miracidial hatching technique (MHT) [8] and Kato Katz (KK) [8], by ruminant group and location.


https://doi.org/10.1371/journal.pntd.0010739.g001

Discussion

In multi-host, multi-parasite systems, all reservoir hosts must be considered in order to achieve disease elimination. Accurate detection of Schistosoma spp. infection in animals would provide not only critical information to guide surveillance and inform control [24], but also help to improve livestock-keeping communities wellbeing, finances, and animal welfare [5, 7]. As WHO therefore calls for consideration of the need to treat livestock within Africa in order to minimize zoonotic transmission to humans, as well as for improved diagnostics in general [24], accurate diagnosis of schistosomiasis at both the individual and population levels is required for sustainable control programmes as well as assessing, and mitigating against, changes in drug efficacy. This study thereby evaluated the clinical performance of the commercially-available POC-CCA, a diagnostic test routinely used for the detection of the human intestinal parasite, S. mansoni, for intestinal S. bovis, S. curassoni and hybridized schistosomiasis infections within ruminant livestock of Senegal, West Africa.

Visual assessments of POC-CCA, MHT and KK sensitivity (see Fig 1) suggested that POC-CCA performance might have been associated both with host group and with parasite species. POC-CCA specificity in small ruminants was high (91%; 95% CrI: 77%, 99%) and did not seem to depend on parasite species. We are not aware of other studies in the literature reporting accuracy results for the detection of CCA in S. bovis or S. curassoni infections in livestock. However, CCA has been reported as being strongly associated with S. bovis worm and faecal egg counts in goats [54, 55] and CAA were detected in S. mattheei infected cows [56].

The test manufacturers have found POC-CCA accuracy to vary by parasite species, indicating that the test was particularly useful for the detection of human intestinal schistosomiasis caused by S. mansoni, and less so in the diagnosis of S. haematobium [57]. Kittur et al. suggested that S. haematobium and S. mansoni worms may produce different amounts of CCA or that S. haematobium may metabolise it more efficiently [58]. It is plausible that S. bovis and S. curassoni similarly have different patterns of CCA excretion and metabolic pathways, and thus respond differently in POC-CCA accuracy. For instance, if the maximum number of adult worms a host can carry falls beneath the test’s minimum level of detection, infected hosts would be misdiagnosed as uninfected. These maxima may differ with parasite species, resulting in differential misclassification. To our knowledge, the respective maximum number of worms each host can carry, by parasite species and ruminant group, has not been determined. The disparity between each host species’ test performance might be attributable to the factors such as consistency of urine, rate of metabolising CCA, or cross-reaction with other infections.

Our study also found that there were no differences in POC-CCA specificity in small ruminants across sites and parasite species, and that POC-CCA specificity was not affected by haematuria. This contrasts with results in humans, where POC-CCA specificity has been found to be affected by the host’s age (in pre-school aged children < 5 years), their pregnancy status, and whether or not they have haematuria or a urogenital infection [39, 59].

POC-CCA accuracy has been found to vary depending on production batch, raising questions regarding production quality control and calls for the optimisation and standardisation of production [6062]. This is of particular relevance if POC-CCA is to become reliable a tool with which to determine whether to treat human and livestock populations [24].

Limitations

Due to the absence of a gold-stand diagnostic test for livestock schistosomiasis, this study derived a composite reference standard (CRS) that was based on KK, MHT and UCAA results. The sensitivities of KK and MHT are relatively low, partly due to the technical difficulties associated with the management of large volumes of faecal material. These methods are highly specific, although misclassification of eggs and miracidia can occur. On the other hand, UCAA assays had been labelled as ultrasensitive for human schistosomiasis detection [50, 63], but have not been optimised for livestock. By combining these tests into a CRS we aimed to obtain a composite measure that was highly specific and moderately sensitive. However, we found eight cases where UCAA was negative and KK or MHT were positive, indicating that the high sensitivity UCAA test may not detect all the infections despite its lower limit of detection of 0.6 pg/mL CAA [63]. Given the variable quality of the urine samples, it was unfortunate that paired serum samples were not available to validate urine testing. CRS in diagnostic accuracy assessment can lead to biased estimates when CRSs are assumed to be error-free [64]. We addressed this potential source of bias by introducing uncertainty in the CRS classification through the adoption of CRS sensitivity and CRS specificity prior distributions, and by carrying out prior sensitivity analyses, that led to the conclusion that the estimates were stable. Our POC-CCA (CRS) results were similar to those obtained by direct diagnostic of abattoir ruminants, suggesting that they were robust with respect to CRS-associated bias. There is no consensus as to whether POC-CCA trace readings should be considered as representing a positive result or as a lack of infection. By considering trace as negative result, one would reduce the test sensitivity and increase its specificity, as fewer positive cases would be detected but more true negative cases would be identified. In this study we followed the recommendations in the literature for human schistosomiasis [65], and we considered trace results as positive infections. However, this may have resulted in an overestimation of sensitivity and underestimation of specificity of POC-CCA. The relatively low sample sizes for goats and sheep limited our ability to analyse goat and sheep data separately. The half-width of the credible intervals reported in this study was on average 20% from the median, except for specificity in cattle, where the width was greater. Lager sample sizes and samples from populations with a wider range of prevalence would have enabled us to calculate more precise estimates both for sensitivity and specificity. It is likely that POC-CCA accuracy in Barkedji and Richard Toll can be assimilated to accuracy for S. curassoni and S. bovis detection, respectively. However, this need to be further investigated. The observed high probabilities of sensitivity differences point to novel evidence of variation between host and parasite species that can inform the design and analyses of future diagnostic evaluations.

Supporting information

References

  1. 1.
    Bharadwaj M, Bengtson M, Golverdingen M, Waling L, Dekker C. Diagnosing point-of-care diagnostics for neglected tropical diseases. PLoS Negl Trop Dis. 2021;15(6):e0009405. pmid:34138846
  2. 2.
    World Health Organization. Ending the neglect to attain the sustainable development goals: a road map for neglected tropical diseases 2021–2030: overview. World Health Organization, 2020.
  3. 3.
    World Health Organization. WHO guideline on control and elimination of human schistosomiasis. World Health Organization, 2022.
  4. 4.
    Lo NC, Bezerra FSM, Colley DG, Fleming FM, Homeida M, Kabatereine N, et al. Review of 2022 WHO guidelines on the control and elimination of schistosomiasis. Lancet Infect Dis. 2022;22(11):e327–e35. pmid:35594896
  5. 5.
    Adeyemo P, Léger E, Hollenberg E, Diouf N, Sène M, Webster JP, et al. Estimating the financial impact of livestock schistosomiasis on traditional subsistence and transhumance farmers keeping cattle, sheep and goats in northern Senegal. Parasit Vectors. 2022;15(1). pmid:35317827
  6. 6.
    De Bont J, Vercruysse J. The epidemiology and control of cattle schistosomiasis. Parasitol Today. 1997;13(7):255–62. pmid:15275062
  7. 7.
    Gower CM, Vince L, Webster JP, Hygiene. Should we be treating animal schistosomiasis in Africa? The need for a One Health economic evaluation of schistosomiasis control in people and their livestock. Trans R Soc Trop Med Hyg. 2017;111(6):244–7. pmid:29044372
  8. 8.
    Léger E, Borlase A, Fall CB, Diouf ND, Diop SD, Yasenev L, et al. Prevalence and distribution of schistosomiasis in human, livestock, and snail populations in northern Senegal: a One Health epidemiological study of a multi-host system. Lancet Planet Health. 2020;4(8):e330–e42. pmid:32800151
  9. 9.
    Boon NA, Van Den Broeck F, Faye D, Volckaert FA, Mboup S, Polman K, et al. Barcoding hybrids: heterogeneous distribution of Schistosoma haematobium× Schistosoma bovis hybrids across the Senegal River Basin. Parasitology. 2018;145(5):634–45. pmid:29667570
  10. 10.
    Huyse T, Webster BL, Geldof S, Stothard JR, Diaw OT, Polman K, et al. Bidirectional introgressive hybridization between a cattle and human schistosome species. PLoS Pathog. 2009;5(9):e1000571. pmid:19730700
  11. 11.
    Leger E, Webster JP. Hybridizations within the Genus Schistosoma: implications for evolution, epidemiology and control. Parasitology. 2017;144(1):65–80. pmid:27572906.
  12. 12.
    Webster BL, Diaw OT, Seye MM, Webster JP, Rollinson D. Introgressive hybridization of Schistosoma haematobium group species in Senegal: species barrier break down between ruminant and human schistosomes. PLoS Negl Trop Dis. 2013;7(4):e2110. pmid:23593513
  13. 13.
    Webster BL, Alharbi MH, Kayuni S, Makaula P, Halstead F, Christiansen R, et al. Schistosome Interactions within the Schistosoma haematobium Group, Malawi. Emerg Infect Dis. 2019;25(6):1245. pmid:31107237
  14. 14.
    Boissier J, Grech-Angelini S, Webster BL, Allienne J-F, Huyse T, Mas-Coma S, et al. Outbreak of urogenital schistosomiasis in Corsica (France): an epidemiological case study. Lancet Infect Dis. 2016;16(8):971–9. pmid:27197551
  15. 15.
    Oleaga A, Rey O, Polack B, Grech-Angelini S, Quilichini Y, Pérez-Sánchez R, et al. Epidemiological surveillance of schistosomiasis outbreak in Corsica (France): Are animal reservoir hosts implicated in local transmission? PLoS Negl Trop Dis. 2019;13(6):e0007543. pmid:31233502
  16. 16.
    Deol AK, Fleming FM, Calvo-Urbano B, Walker M, Bucumi V, Gnandou I, et al. Schistosomiasis—Assessing Progress toward the 2020 and 2025 Global Goals. N Engl J Med. 2019;381(26):2519–28. pmid:31881138
  17. 17.
    Webster JP, Molyneux DH, Hotez PJ, Fenwick A. The contribution of mass drug administration to global health: past, present and future. Philos Trans R Soc Lond B Biol Sci. 2014;369(1645):20130434. pmid:24821920
  18. 18.
    Borlase A, Rudge JW, Léger E, Diouf ND, Fall CB, Diop SD, et al. Spillover, hybridization, and persistence in schistosome transmission dynamics at the human–animal interface. Proc Natl Acad Sci USA. 2021;118(41):e2110711118. pmid:34615712
  19. 19.
    Kaplan RM. Drug resistance in nematodes of veterinary importance: a status report. Trends Parasitol. 2004;20(10):477–81. pmid:15363441
  20. 20.
    Kaplan RM, Vidyashankar AN. An inconvenient truth: global worming and anthelmintic resistance. Vet Parasitol. 2012;186(1–2):70–8. pmid:22154968
  21. 21.
    Weerakoon KG, Gordon CA, McManus DP. DNA Diagnostics for Schistosomiasis Control. Trop Med Infect Dis. 2018;3(3). pmid:30274477
  22. 22.
    Liang S, Ponpetch K, Zhou Y-B, Guo J, Erko B, Stothard JR, et al. Diagnosis of Schistosoma infection in non-human animal hosts: A systematic review and meta-analysis. PLoS Negl Trop Dis. 2022;16(5):e0010389. pmid:35522699
  23. 23.
    Doenhoff MJ, Chiodini PL, Hamilton JV. Specific and sensitive diagnosis of schistosome infection: can it be done with antibodies? Trends Parasitol. 2004;20(1):35–9. pmid:14700588
  24. 24.
    van Lieshout L, Polderman AM, Deelder AM. Immunodiagnosis of schistosomiasis by determination of the circulating antigens CAA and CCA, in particular in individuals with recent or light infections. Acta Trop. 2000;77(1):69–80. pmid:10996122
  25. 25.
    Corstjens PLAM, Van Lieshout L, Zuiderwijk M, Kornelis D, Tanke HJ, Deelder AM, et al. Up-Converting Phosphor Technology-Based Lateral Flow Assay for Detection of Schistosoma Circulating Anodic Antigen in Serum. J Clin Microbiol. 2008;46(1):171–6. pmid:17942645
  26. 26.
    Corstjens PLAM De Dood CJ, Kornelis D Tjon Kon Fat EM, Wilson RA Kariuki TM, et al. Tools for diagnosis, monitoring and screening of Schistosoma infections utilizing lateral-flow based assays and upconverting phosphor labels. Parasitology. 2014;141(14):1841–55. pmid:24932595
  27. 27.
    Coulibaly JT, Knopp S, N’Guessan NA, Silué KD, Fürst T, Lohourignon LK, et al. Accuracy of Urine Circulating Cathodic Antigen (CCA) Test for Schistosoma mansoni Diagnosis in Different Settings of Côte d’Ivoire. PLoS Negl Trop Dis. 2011;5(11):e1384. pmid:22132246
  28. 28.
    Van Dam GJ, Wichers JH, Ferreira TMF, Ghati D, Van Amerongen A, Deelder AM. Diagnosis of Schistosomiasis by Reagent Strip Test for Detection of Circulating Cathodic Antigen. J Clin Microbiol. 2004;42(12):5458–61. pmid:15583265
  29. 29.
    Colley DG, Binder S, Campbell C, King CH, Tchuem Tchuenté L-A, N’Goran EK, et al. A Five-Country Evaluation of a Point-of-Care Circulating Cathodic Antigen Urine Assay for the Prevalence of Schistosoma mansoni. Am J Trop Med Hyg. 2013;88(3):426–32. pmid:23339198
  30. 30.
    World Health Organization. Schistosomiasis [online], Available at: https://www.who.int/news-room/fact-sheets/detail/schistosomiasis 2021.
  31. 31.
    Midzi N, Butterworth AE, Mduluza T, Munyati S, Deelder AM, van Dam GJ. Use of circulating cathodic antigen strips for the diagnosis of urinary schistosomiasis. Trans R Soc Trop Med Hyg. 2009;103(1):45–51. pmid:18951599
  32. 32.
    Rubaba O, Chimbari MJ, Soko W, Manyangadze T, Mukaratirwa S. Validation of a urine circulating cathodic antigen cassette test for detection of Schistosoma haematobiumin uMkhanyakude district of South Africa. Acta Trop. 2018;182:161–5. pmid:29486172
  33. 33.
    Sanneh B, Joof E, Sanyang AM, Renneker K, Camara Y, Sey AP, et al. Field evaluation of a schistosome circulating cathodic antigen rapid test kit at point-of-care for mapping of schistosomiasis endemic districts in The Gambia. PLoS One. 2017;12(8):e0182003. pmid:28797128
  34. 34.
    Stothard JR, Kabatereine NB, Tukahebwa EM, Kazibwe F, Rollinson D, Mathieson W, et al. Use of circulating cathodic antigen (CCA) dipsticks for detection of intestinal and urinary schistosomiasis. Acta Trop. 2006;97(2):219–28. pmid:16386231
  35. 35.
    Standley CJ, Mugisha L, Dobson AP, Stothard JR. Zoonotic schistosomiasis in non-human primates: past, present and future activities at the human–wildlife interface in Africa. J Helminthol. 2012;86(2):131–40. pmid:22269859
  36. 36.
    Picquet M, Ernould JC, Vercruysse J, Southgate VR, Mbaye A, Sambou B, et al. The epidemiology of human schistosomiasis in the Senegal river basin. Trans R Soc Trop Med Hyg. 1996;90(4):340–6. pmid:8882173
  37. 37.
    Van den Broeck F, Maes GE, Larmuseau MHD, Rollinson D, Sy I, Faye D, et al. Reconstructing Colonization Dynamics of the Human Parasite Schistosoma mansoni following Anthropogenic Environmental Changes in Northwest Senegal. PLoS Negl Trop Dis. 2015;9(8):e0003998. pmid:26275049
  38. 38.
    Vercruysse J, Southgate VR, Rollinson D. The epidemiology of human and animal schistosomiasis in the Senegal River Basin. Acta Trop. 1985;42(3):249–59. pmid:2865881
  39. 39.
    Greter H, Utzinger J, Ngandolo BNR, Krauth SJ, Alfaroukh IO, Zinsstag J. Validation of a Point-of-Care Circulating Cathodic Antigen Urine Cassette Test for Schistosoma mansoni Diagnosis in the Sahel, and Potential Cross-Reaction in Pregnancy. Am J Trop Med Hyg. 2016;94(2):361–4. pmid:26556831
  40. 40.
    Banoo S, Bell D, Bossuyt P, Herring A, Mabey D, Poole F, et al. Evaluation of diagnostic tests for infectious diseases: general principles. Nat Rev Microbiol. 2010;8(12):S16–S28. pmid:21548184
  41. 41.
    Branscum AJ, Gardner IA, Johnson WO. Estimation of diagnostic-test sensitivity and specificity through Bayesian modeling. Prev Vet Med. 2005;68(2–4):145–63. pmid:15820113
  42. 42.
    Hui SL, Walter SD. Estimating the Error Rates of Diagnostic Tests. Biometrics. 1980;36(1):167–71. pmid:7370371
  43. 43.
    Walter SD, Irwig LM. Estimation of test error rates, disease prevalence and relative risk from misclassified data: a review. J Clin Epidemiol. 1988;41(9):923–37. pmid:3054000
  44. 44.
    Dendukuri N, Joseph L. Bayesian Approaches to Modeling the Conditional Dependence Between Multiple Diagnostic Tests. Biometrics. 2001;57(1):158–67. pmid:11252592
  45. 45.
    Gardner IA, Stryhn H, Lind P, Collins MT. Conditional dependence between tests affects the diagnosis and surveillance of animal diseases. Prev Vet Med. 2000;45(1–2):107–22. pmid:10802336
  46. 46.
    Spiegelhalter DJ, Best NG, Carlin BP, Van Der Linde A. Bayesian measures of model complexity and fit. J Roy Stat Soc Ser B (Stat Method). 2002;64(4):583–639.
  47. 47.
    Corstjens PL, Nyakundi RK, de Dood CJ, Kariuki TM, Ochola EA, Karanja DM, et al. Improved sensitivity of the urine CAA lateral-flow assay for diagnosing active Schistosoma infections by using larger sample volumes. Parasit Vectors. 2015;8:241. pmid:25896512
  48. 48.
    de Dood CJ, Hoekstra PT, Mngara J, Kalluvya SE, van Dam GJ, Downs JA, et al. Refining Diagnosis of Schistosoma haematobium Infections: Antigen and Antibody Detection in Urine. Front Immunol. 2018;9. pmid:30487796
  49. 49.
    Sousa MS, van Dam GJ, Pinheiro MCC, de Dood CJ, Peralta JM, Peralta RHS, et al. Performance of an Ultra-Sensitive Assay Targeting the Circulating Anodic Antigen (CAA) for Detection of Schistosoma mansoni Infection in a Low Endemic Area in Brazil. Front Immunol. 2019;10:682. pmid:31019510
  50. 50.
    van Dam GJ, Xu J, Bergquist R, de Dood CJ, Utzinger J, Qin ZQ, et al. An ultra-sensitive assay targeting the circulating anodic antigen for the diagnosis of Schistosoma japonicum in a low-endemic area, People’s Republic of China. Acta Trop. 2015;141(Pt B):190–7. pmid:25128703
  51. 51.
    Kostoulas P, Leontides L, Enoe C, Billinis C, Florou M, Sofia M. Bayesian estimation of sensitivity and specificity of serum ELISA and faecal culture for diagnosis of paratuberculosis in Greek dairy sheep and goats. Prev Vet Med. 2006;76(1–2):56–73. pmid:16806541
  52. 52.
    Mathevon Y, Foucras G, Falguières R, Corbiere F. Estimation of the sensitivity and specificity of two serum ELISAs and one fecal qPCR for diagnosis of paratuberculosis in sub-clinically infected young-adult French sheep using latent class Bayesian modeling. BMC Vet Res. 2017;13(1):230. pmid:28774299
  53. 53.
    R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2022.
  54. 54.
    Agnew AM, Murare HM, Sandoval SN, Jong ND, Krijer FW, Deelder AM, et al. The susceptibility of adult schistosomes to immune attrition. Mem Inst Oswaldo Cruz. 1992;87(suppl 4):87–93. pmid:1343931
  55. 55.
    Johansen MV, Fillié Y, Monrad J, Christensen NO, Deelder A. Experimental Schistosoma bovis infection in goats. Circulating antigen and antibody responses to egg and adult worm antigens during infection and following treatment with praziquantel. Parasitology. 1996;113 (Pt 4):367–75. pmid:8873476.
  56. 56.
    Gabriël S, De Bont J, Phiri IK, Masuku M, Riveau G, Schacht AM, et al. Transplacental transfer of schistosomal circulating anodic antigens in cows. Parasite Immunol. 2002;24(11–12):521–5. pmid:12694602
  57. 57.
    Rapid Medical Diagnostics. Schisto POC-CCA Rapid test for qualitative detection of Bilharzia (Schistosomiasis) 2018 [02/03/2022]. Available from: https://www.rapid-diagnostics.com/.
  58. 58.
    Kittur N, Castleman JD, Campbell CH Jr, King CH, Colley DG. Comparison of Schistosoma mansoni prevalence and intensity of infection, as determined by the circulating cathodic antigen urine assay or by the Kato-Katz fecal assay: a systematic review. Am J Trop Med Hyg. 2016;94(3):605.
  59. 59.
    Casacuberta-Partal M, Beenakker M, De Dood CJ, Hoekstra PT, Kroon L, Kornelis D, et al. Specificity of the Point-of-Care Urine Strip Test for Schistosoma Circulating Cathodic Antigen (POC-CCA) Tested in Non-Endemic Pregnant Women and Young Children. Am J Trop Med Hyg. 2021;104(4):1412–7. pmid:33534739
  60. 60.
    Graeff-Teixeira C, Favero V, Pascoal VF, de Souza RP, de Vargas Rigo F, Agnese LHD, et al. Low specificity of point-of-care circulating cathodic antigen (POCCCA) diagnostic test in a non-endemic area for schistosomiasis mansoni in Brazil. Acta Trop. 2021;217:105863. pmid:33587944
  61. 61.
    Peralta JM, Cavalcanti MG. Is POC-CCA a truly reliable test for schistosomiasis diagnosis in low endemic areas? The trace results controversy. PLoS Negl Trop Dis. 2018;12(11):e0006813. pmid:30408030
  62. 62.
    Viana AG, Gazzinelli-Guimaraes PH, Castro VN, Santos Y, Ruas ACL, Bezerra FSM, et al. Discrepancy between batches and impact on the sensitivity of point-of-care circulating cathodic antigen tests for Schistosoma mansoni infection. Acta Trop. 2019;197:105049. pmid:31158344
  63. 63.
    Corstjens PLAM De Dood CJ, Knopp S Clements MN, Ortu G, Umulisa I, et al. Circulating Anodic Antigen (CAA): A Highly Sensitive Diagnostic Biomarker to Detect Active Schistosoma Infections—Improvement and Use during SCORE. Am J Trop Med Hyg. 2020;103(1_Suppl):50–7. pmid:32400344
  64. 64.
    Dendukuri N, Schiller I, De Groot J, Libman M, Moons K, Reitsma J, et al. Concerns about composite reference standards in diagnostic research. BMJ. 2018:j5779. pmid:29348126
  65. 65.
    Clements MN, Corstjens PLAM, Binder S, Campbell CH, De Dood CJ, Fenwick A, et al. Latent class analysis to evaluate performance of point-of-care CCA for low-intensity Schistosoma mansoni infections in Burundi. Parasit Vectors. 2018;11(1). pmid:29475457

Source link