Prevalence of Strongyloidiasis in a Cohort of Migrants in Italy and Accuracy of a Novel ELISA Assay for S. stercoralis Infection, a Cross-Sectional Study

Strongyloides stercoralis infection is a life-threatening neglected tropical disease. Diagnostic issues have caused an underestimation of its global burden. The choice of appropriate diagnostic tests for the screening of populations at risk of the infection, such as migrants from endemic countries, is of paramount importance. From November 2017 to July 2018, all migrants presenting to the National Institute for Health Migration and Poverty (INMP) in Rome, Italy were offered screening tests for S. stercoralis infection. The study objective was to estimate the prevalence of strongyloidiasis in the study population and the accuracy of a novel ELISA assay. The following tests were carried out at the IRCCS Sacro Cuore Don Calabria hospital in Negrar, Verona: stool microscopy, real-time PCR for S. stercoralis, in-house immunofluorescence test (IFAT), a commercial ELISA assay (Bordier ELISA), and a novel ELISA assay (Euroimmun ELISA). A latent class analysis (LCA) model set up with test results, clinical variables, and eosinophilia indicated a prevalence around 7.5%, in line with previous findings. The sensitivity and the specificity of Euroimmun ELISA were 90.6% (95% CI 80.5–100) and 87.7% (95CI 84.5–91.0); these results indicate that the novel ELISA assay would be suitable for screening of migrants from endemic countries.


Introduction
Strongyloides stercoralis infection is a neglected tropical disease (NTD) affecting an estimated 640 million people worldwide [1]. The infection is transmitted through direct penetration of the skin by infective (filariform) larvae free-living in soil contaminated by human faeces. Thus, the transmission occurs in areas with inadequate sewage systems and hygienic conditions [2]. The larvae moult while migrating in the human body, and the adult female worm settles in the intestine, where it produces eggs. The newborn rhabditoid larvae hatch out of the eggs while still in the bowel and are shed with faeces. However, some of them moult into the filariform stage before leaving the body and thus can re-infect the host through rectal mucosa or perianal skin. This auto-infective cycle leads to a chronic infection, which can perpetuate through the years even in absence of re-exposure to an external infective source [2]. The infected individual can be either asymptomatic or complain of a series of unspecific symptoms, mainly involving skin and respiratory and gastrointestinal tracts [3]. Symptoms can be mild and intermittent but under some circumstances can worsen and manifest as severe conditions. This mainly happens in immunosuppressed individuals, for whom the disseminated infection, characterized by all parasite stages migrating all over the body, represents a therapeutic challenge and is life-threatening [2]. To prevent the risk of dissemination, screening and individual diagnosis of cases followed by proper treatment of the infection, which is based on a single dose of ivermectin in case of uncomplicated disease [4], are of paramount importance, even in asymptomatic individuals. Unfortunately, there is no gold standard for the diagnosis of strongyloidiasis, and this causes an underestimation of the burden of the infection and poses infected individuals at risk of misdiagnosis [3]. Indeed, microscopy examination of stool samples has very low sensitivity for the detection of S. stercoralis. Although its sensitivity was demonstrated to improve with the examination of multiple samples, it still remains disappointing [5]. Baermann method and agar plate culture have better performance but are seldom done because they are time-consuming and require parasitological skills that are not widely available in non-endemic settings [6]. Polymerase chain reaction (PCR) for S. stercoralis demonstrated good sensitivity and specificity, though it is not yet widely available outside referral laboratories [7]. At the moment, serological tests show the best sensitivity, although cross-reactions, in particular with filarial nematodes, can cause false positive results [8,9]. However, due to the potential harm posed by the infection and the good tolerability of ivermectin, overtreatment of possible false-positive cases should be preferred to missed diagnosis and treatment.
This study was part of the Tropical Neglected Diseases Project, "Strengthening the fight against Neglected Tropical Diseases in the migrant population through the use of medical devices" that was carried out from August 2016 to July 2018. The project was led by the Italian National Institute for Health Migration and Poverty (INMP) with the collaboration of the Department of Infectious Tropical Diseases and Microbiology (DITM) of the IRCCS Sacro Cuore Don Calabria hospital, as regards the diagnostic aspects. The general objective of the project was to strengthen the fight against NTDs in migrant populations from endemic areas living in Rome through an estimate of the prevalence and an evaluation of the epidemiological characteristics of some major NTDs in order to achieve early diagnosis and care of affected migrants.
In this work, we report the results of the screening for S. stercoralis. Primary objective was to estimate the prevalence of strongyloidiasis in this population, based on a combination of diagnostic tests. Secondary objective was to estimate the accuracy of a novel ELISA test (Euroimmun ELISA).

Materials and Methods
This was a cross-sectional study, which took place from November 2017 to July 2018. Campaigns for promotion of the screening activities were conducted in order to disseminate information about selected NTDs (in addition to strongyloidiasis, targets of the campaign were schistosomiasis and Chagas disease) and about the free access to a dedicated INMP outpatient clinic.

Participants
Migrants of any age were offered the extended screening upon spontaneous presentation to the INMP outpatient clinic. Demographic information such as data concerning migration route, personal habits, and living conditions (economic and hygienic aspects) were collected. An infectious diseases consultant sought written informed consent, provided medical visit, and asked the participants to supply faecal and blood samples for S. stercoralis (and other NTDs based on epidemiological risk factors) testing. Information sheets and informed consent forms were available in Italian, Spanish, English, and French. Transcultural mediators were available for illiterate people and for those speaking other languages (Arabic, for instance). Parents' or legal guardians' consent was sought for minors. All consenting consecutive participants underwent the screening and were included in the analysis. Participants received a copy of the results of the test and were treated accordingly at the INMP clinic. Any positive test (including serology) constituted an eligibility criterion for treatment.

Test Methods
All biological samples were collected at the INMP. Aliquots were sent to the laboratory of San Camillo hospital, Rome for the full blood count. Other samples were sent to the DITM, where the following tests were carried out: microscopy examination of faeces concentrated by Ritchie's modified method, a commercial ELISA for S. stercoralis, an in-house immunofluorescence test (IFAT), a novel ELISA kit, and real-time PCR for S. stercoralis. An ELISA test for Schistosoma mansoni (Bordier Affinity Products SA, Crissier, Switzerland) was also done at DITM. The laboratory staff performing the serological tests were blinded towards the results of the other tests.
The ELISA commercial kit (Bordier Affinity Products SA, Crissier, Switzerland) is based on somatic antigens from larvae of Strongyloides ratti [10]. A previous retrospective study estimated its sensitivity and specificity at 90.8% (95% CI 85.8-95.7) and 94.0% (95% CI 91.2-96.9), respectively [9]. The test was performed as per manufacturer's instructions. A normalized optical density (OD) ratio was used to compare the results obtained in different sessions. A ratio ≥1 defined positive results.
The ELISA performed at the DITM was a novel kit from Euroimmun. The assay is based on antigens of Strongyloides papillosus [12]. The test was performed per the manufacturer's instructions. Results were classified in accordance with the package leaflet: ratio <0.8 = negative; ratio ≥ 0.8 < 1.1 = borderline; ratio ≥1 = positive.
The real-time PCR is based on Verweji's method [13] and is used routinely at the DITM. Briefly, for DNA extraction, about 200 mg of faeces were suspended in 200 µL of phosphate-buffered saline containing 2% polyvinylpolypyrolidone (Sigma-Aldrich, Milan, Italy) and frozen overnight at −20 • C until the extraction. After thawing and boiling, the samples were run by an automated extractor instrument (Magnapure LC.2, Roche Diagnostics, Monza, Italy). The real-time assay was performed as described previously [13]. The amplification target was the small-subunit rRNA gene sequence for S. stercoralis. Appropriate positive and negative controls were included in all the experiments. As control for PCR inhibitors and amplification quality, the PhHV-1 control DNA was amplified with the appropriate primers/probe mix in the same reaction as S. stercoralis in multiplex PCR. The reactions, detection, and data analysis were performed with the CFX96 detection system (Bio-Rad Laboratories, Milan, Italy). In a previous retrospective study, the method demonstrated a sensitivity of 56.8% (95% CI 41.0-71.6) [14]. Specificity was considered virtually 100%.

Analysis
The sample size was based on a convenience sample constituted by all eligible participants who were enrolled during the study period. This was aimed to maximize the power and the generalizability of the study results.
Demographic and clinical data were summarized using descriptive statistics and measures of variability and precision. All parameters were reported with 95% confidence intervals (CI). For proportions, the exact Clopper-Pearson CI was computed.
Diagnostic test results were presented in contingency tables where patient's disease status was inferred based on results of each single test, the composite reference standard, and also on probabilistic models using latent class analysis (LCA) [15].
Composite reference standard (CRS) is an alternative method used for assessing test accuracy using a combination of tests and was obtained by applying the following rule: if microscopy or PCR were positive or both IFAT and Bordier ELISA were positive, then CRS = positive; otherwise CRS = negative.
Data analysis was performed using SAS software, version 9.4 (SAS Institute, Inc., Cary, NC, USA). Statistical significance level was fixed at 0.05.

Results
In total, 650 participants received screening for S. stercoralis. The overall numbers of samples positive to Bordier ELISA, Euroimmun ELISA, and IFAT were 60/646 (9.29%), 113/648 (17.44%), and 118/646 (18.27%), respectively. The agreement between Bordier ELISA and Euroimmun, Bordier ELISA and IFAT, and Euroimmun and IFAT was moderate in all cases, k coefficients being 0.5198, 0.4106, and 0.4887, respectively. The proportion of participants with positive serological tests was similar between the two subgroups: 20.3%, 15.8%, and 9.05% participants of African origin were positive to IFAT, Euroimmun ELISA, and Bordier ELISA, respectively. The proportions of participants from Latin America with a positive result to IFAT, Euroimmun ELISA, and Bordier ELISA were 14.9%, 20%, and 9.7%, respectively. All 23 samples with positive PCR and/or microscopy were positive to all serological methods. Additional positive samples were found by the serological tests, as displayed in Figure 1.
According to the CRS, 43 samples were classified as positive and 605 as negative. Table 1 reports the number of positive and negative Euroimmun samples against this classification.  According to the CRS, 43 samples were classified as positive and 605 as negative. Table 1 reports the number of positive and negative Euroimmun samples against this classification.  1-90.4). Among the 73 false positive results (i.e., positive to Euroimmun ELISA but negative to the CRS), nine were found positive to hookworm and/or Schistosoma spp. Specifically, seven patients had positive ELISA, and three of them had also S. mansoni eggs in stools. Moreover, three patients had hookworm, one of them had also positive S. mansoni ELISA.
An LCA model using data from 425 participants for whom data on symptoms were available was set by adding to the results of all diagnostic tests (but Euroimmun ELISA), the clinical variables eosinophilia (p-value < 0.0001), and itching (p-value = 0.0072). Abdominal pain was excluded as it resulted statistically significant only in the univariable model (p-value = 0.0042). According to the LCA, 32 (7.5%) samples were classified as positive. Table 2 reports the number of positive and negative Euroimmun ELISA samples against this classification.  1-90.4). Among the 73 false positive results (i.e., positive to Euroimmun ELISA but negative to the CRS), nine were found positive to hookworm and/or Schistosoma spp. Specifically, seven patients had positive ELISA, and three of them had also S. mansoni eggs in stools. Moreover, three patients had hookworm, one of them had also positive S. mansoni ELISA.
An LCA model using data from 425 participants for whom data on symptoms were available was set by adding to the results of all diagnostic tests (but Euroimmun ELISA), the clinical variables eosinophilia (p-value < 0.0001), and itching (p-value = 0.0072). Abdominal pain was excluded as it resulted statistically significant only in the univariable model (p-value = 0.0042). According to the LCA, 32 (7.5%) samples were classified as positive. Table 2 reports the number of positive and negative Euroimmun ELISA samples against this classification. Based on these results, the sensitivity and the specificity of the Euroimmun ELISA were 90.6% (95%CI 80.5-100) and 87.7% (95%CI 84.5-91.0), respectively. The same helminthic co-infections (in terms of both type of helminths and number of cases) detected for the 73 false positives with the CRS were found among the 48 cases classified as false positives with the LCA.

Discussion
We collected data on screening for strongyloidiasis in a large cohort of immigrants from Africa (61.5%) and Latin America (38.5%). About one quarter of them presented eosinophilia, and 38% reported any symptom, mostly itching (31.7%) and abdominal pain (27.7%). About 3% of participants had a positive faecal test for S. stercoralis (PCR and/or stool microscopy), while the proportion of positive serological tests ranged from 9.3% of Bordier ELISA to 18.3% of IFAT.
A systematic review [16] reported a 12.2% (95%CI 9.9-15.9) seroprevalence of strongyloidiasis in migrants living in the USA, Canada, Australia, New Zealand, Israel, and Western Europe. Our findings are in line with the review, although figures vary slightly based on the serological assay considered.
It should be considered that serology tends to overestimate the prevalence of strongyloidiasis based on the lower specificity compared to faecal-based tests caused by possible cross-reactions [9]. Here, we report a few co-helminthic infections that could be the cause of false positive results but, for instance, filarial nematodes were not thoroughly investigated. Although lower specificity might be acceptable in specific contexts (e.g., screen and treat strategies), a classification based on a combination of diagnostic tests can be more appropriate in others. As an example, surveys of prevalence in endemic areas can benefit from the addition of a more specific test, as co-infections with helminths that can cause cross-reactions are of higher concern in that setting. Here, we adopted two different approaches to the classification of positive/negative cases, which are CRS and LCA. With these methods, we classified as positive 6.6% and 7.5% cases, respectively. These figures, which are between the 3% found with faecal tests and a maximum of 18% with IFAT serology, can presumably give a better idea of the real prevalence of strongyloidiasis found in our cohort.
Eosinophilia and itching were included in the LCA on the basis of significant association with S. stercoralis infection in a multivariate analysis. Abdominal pain was significantly associated with the infection only in the univariate model, and other symptoms (for instance respiratory symptoms, weakness) did not show any association. Eosinophilia has repeatedly been reported as a frequent feature of strongyloidiasis, while clinical presentation is more debated, as symptoms are often intermittent and aspecific [17,18]. A systematic review of studies carried out in Strongyloides endemic areas [19] showed association with urticaria, while itching was frequent but the association was uncertain (though we might debate that itching is presumably associated with urticaria).
An unexpected result of our analysis is the excellent agreement between stool microscopy and PCR. Indeed, stool microscopy is usually considered to have a sensitivity exceedingly low compared to PCR [8]. The results of microscopy here (3.2% positive participants) are in line with a previous study reporting data on screening of 462 asylum seekers in Northern Italy [20], where 3.3% positive individuals were found with stool microscopy (with modified Ritchie's concentration method). In other studies, the screening of populations at risk resulted in a higher proportion of individuals positive to S. stercoralis detected by PCR than other parasitological methods [6,21,22].
Here, it should be considered that laboratory staff performing microscopy was not blinded to the results of PCR, thus this might have caused a more intense examination of stool samples when PCR was positive. Conversely, agreement between the pair of serological assays was found good in all cases. In literature, we could find only a few conference proceedings reporting data on accuracy of Euroimmun ELISA, which was deemed good in comparison with other ELISAs [12,23]. Hence, to our knowledge, this is the first work reporting full data on the evaluation of this assay, which demonstrated good performance for the screening of individuals at risk of strongyloidiasis.
Finally, another limitation of the study is that prevalence of the infection was assessed in a group of migrants who spontaneously presented to the clinic. Thus, they might have been motivated by presence of symptoms or other health concerns, reducing the presence of asymptomatic infections in the cohort.

Conclusions
In a cohort of 650 individuals from Africa (61.5%) and Latin America (38.5%), screening with different diagnostic methods for S. stercoralis showed a prevalence of strongyloidiasis of about 7%. As expected, a higher proportion of positive results was found with serological tests than with faecal-based tests. Unexpectedly, Ritchie's concentration stool microscopy showed excellent agreement with PCR for S. stercoralis. The accuracy of a novel ELISA (Euroimmun) kit was found good, indicating that the assay would be useful for screening activities.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The study database will be available in Mendeley Data upon acceptance for publication.