Species specificity preliminary evaluation of an IL‐4‐based test for the differential diagnosis of human echinococcosis

Abstract The diagnosis of cystic echinococcosis (CE) is based on imaging, while serology is a complementary test of particular use when imaging is inconclusive. Serology has several limitations. Among them, false‐positive results are often obtained in subjects with alveolar echinococcosis (AE), rendering difficult the differential diagnosis. We set up an immune assay based on IL‐4‐specific production after stimulating whole blood with an antigen B (AgB)‐enriched fraction from E granulosus that associates with CE and CE cysts in active stage. We aimed to evaluate potential cross‐reactivity of this test using samples from patients with AE. Twelve patients with AE were recruited; IL‐4 levels ranged from 0 to 0.07 pg/mL. Based on the previously identified cut‐off of 0.39 pg/mL using samples from patients with CE, none of samples from AE patients scored positive. In contrast, almost 80% of samples from AE patients scored positive in serology tests based on different E granulosus‐derived antigenic preparations. Our preliminary data show that this experimental whole‐blood assay has no cross‐reactivity in our cohort of patients with AE, in turn indicating a high specificity of the assay for CE diagnosis. This result supports further work towards the development of improved diagnostic tests for CE.

CE greatly differ in terms of disease progression, management and prognosis. [4][5][6] The geographical distribution of CE and AE greatly overlaps in central Asia and China, making the differential diagnosis of these two conditions particularly challenging. This is further complicated by the high rate of cross-reactivity of serological tests (50%-100%). [7][8][9] Several approaches, such as the use of different sources of antigens and different readout systems, have been attempted to overcome the serology shortfalls. [9][10][11][12][13][14] The most commonly used serological tests for human CE are based on the detection of IgG antibodies against E granulosus hydatid cyst fluid (HCF). These tests show an 80%-99% sensitivity with variable specificity (60%-97%), while tests detecting IgG against purified or recombinant antigens show a better specificity (80%-100%) but lower sensitivity (38%-93%). 9,15 The discrimination between CE and AE on serology may not be clear in up to about 25% of cases even using specific tests, such as band-pattern evaluation of HCF-based immunoblotting and E multilocularis-specific Em2plus-or Em18-based serological assays, 14,16 making the results of these tests not always reliable to distinguish these two aetiologies in clinical practice.
Cytokines have been considered the basis of potential tools for the diagnosis and clinical management of CE, and associations between cytokine responses and some clinical features have been reported. [17][18][19][20][21][22] In this context, we recently set up an immune-based test measuring the IL-4 production after stimulating whole blood with an enriched fraction of antigen B (AgB-the most abundant antigen of HCF) or AgB peptides. 11,12 In this whole-blood test, higher levels of IL-4 associated with CE infection (71.4% sensitivity, 93.3% specificity) and with the presence of active cysts (84.6% sensitivity, 92% specificity). 12 However, cross-reactivity with AE has not been tested. Therefore, the aim of this study was to evaluate the specificity of the experimental IL-4 test based on AgB-enriched fraction for CE diagnosis, analysing samples from patients with AE from Eastern Switzerland, where AE is endemic. 23  Blood samples and clinical information were collected from each patient. Serology was carried out at the time of performing the experimental whole-blood test (or within 6 months) as part of clinical routine at the Institute of Parasitology of the University of Zürich.

| MATERIAL S AND ME THODS
Serological tests were performed according to manufacturer instructions or as previously reported, 24 and were as follows: HCF-enzyme immunotest (EIA), EgP-EIA, Em2plus-EIA, Em18-EIA; AgB-electroimmuno-transfer blot. All serological tests were in-house assays 24 with the exception of the commercial Em2plus-EIA (Bordier Affinity Products). Unfortunately, the leftover serum was not available for further analyses.
For the whole-blood test, AgB-enriched fraction was purified from HCF obtained aseptically from echinococcal cysts of sheep from Sardinia (Italy). A purified AgB-enriched fraction preparation was obtained after boiling HCF for 15 min; the sample was then centrifuged at 50 000 g at 4°C for 1 h. Protein content was determined by protein assay (Bio-Rad). Protein integrity and analysis of AgB major bands have been performed through a 4%-20% gradient gel ( Figure 1, left part) followed by Western blot using as primary antibody the serum from a patient with CE ( Figure 1, right part). In the gel, as well as in the Western blot analysis, bands corresponding to AgB are evident as they appear as molecular weight multiples of 8kDa (i.e. 8, 16, 24, 32 kDa) ( Figure 1C). Whole blood was stimulated or not (negative control) with AgB-enriched fraction and staphylococcal enterotoxin B (SEB, positive control); supernatants were sent to INMI for batch-wise IL-4 determination by ELISA, as previously described. 12 Laboratory personnel was blinded to the patient diagnosis.
F I G U R E 1 Analysis of AgB preparation. AgB-enriched fraction was subjected to 4%-20% SDS-PAGE under reducing conditions and stained with Coomassie blue (left part) or transferred to nitrocellulose paper and incubated with a serum from a person with cystic echinococcosis (right part). AgB: antigen B-enriched fraction; Mw M, prestained standard of molecular weight markers The whole-blood test cut-off for positivity of ≥ 0.39 pg/mL (IL-4 concentration upon stimulation with AgB-enriched fraction minus IL-4 concentration in negative control) was previously established through receiver operator characteristic curve analysis of data from healthy donors and well-characterized patients with CE. 12 An indeterminate result was defined as an IL-4 level < 4pg/mL in response to the SEB independently of the response to AgB of the same sample. IL-4 results were expressed as medians and interquartile ranges. The AgBenriched fraction used in Petrone et al 12 was purified as described above, and antigen yield was 570 µg/mL. The gel images of the AgBenriched fraction batches were compared and showed that the AgBenriched fraction is a well-purified preparation (data not shown).

| RE SULTS
We enrolled 12 patients with AE. Their demographic and clinical characteristics at the time of performing the whole-blood test are shown in Tables 1 and 2. All but one patient with AE had received medical therapy with albendazole (median: 2 years, range 10 days to 8 years), and 10/11 were still receiving albendazole at the time of performing the whole-blood test, whereas in one patient the treatment was discontinued 3 years before performing the whole-blood test; 4 had undergone nonradical surgical resection of liver lesions. Unfortunately, the activity of these infections at the time of the study could not be defined using positron emission tomography with fluorodeoxyglucose integrated with computed tomography (FDG-PET/CT); however, all but 2 patients could be considered having active infection from the results of serology followed over time (AE1 and AE6; Table 2). 25 Serology results at the time of performing the whole-blood test were as follows: 9/12 AE subjects were tested with HCF-EIA and EgP-EIA, and among them 7/9 had positive results in both tests; a positive AgB-EITB result was found in 3/3 patients; positive results to Em2plus and to Em18 were found in 7/9 and 4/5 AE patients, respectively. The results of the whole-blood IL-4 test in comparison with serology are shown in Table 2. Whole-blood test indeterminate results were obtained in 2/12 patients with AE; no obvious explanations were found for such results in the clinical records of the patients. These patients were excluded from further evaluations. Among the 10 AE mitogen-responding patients, the IL-4 levels ranged from 0 pg/mL to 0.07 pg/mL [median 0, interquartile range (IQR): 0-0.03 pg/mL] (Figure 2). When we compared the robustness of the results obtained in this study with those previously published by Petrone et al 12 by comparing the IL-4 levels in response to the SEB-positive control, no significant differences were found (P = .91) (Figure 2). Based on the IL-4 cut-off in response to the AgB-enriched fraction of 0.39 pg/mL, 12 none of the 10 AE patients scored positive to the whole-blood test. Therefore, as shown in Table 3, the previously found overall sensitivity for CE diagnosis was

| D ISCUSS I ON
The cross-reactivity between AE and CE is an important limitation of the serological tests used to complement the diagnosis of these two infections, with serious consequences in clinical practice. 4 In our work, 78% of patients with AE had positive results in one or more tests based on E granulosus antigens; in contrast, our preliminary results interestingly showed that none of AE patients in our cohort had a positive whole-blood IL-4 test. These preliminary data suggest a high specificity of the whole-blood test based on an AgBenriched fraction, which may be able to allow ruling out AE in case of a positive result. Strikingly, although AgB from E granulosus and E multilocularis shows a high amino acid homology, 26 the wholeblood test, based on a T-cell response, may allow the identification of E granulosus-specific epitope regions not recognized by IgG antibodies. These deserve future identification to develop recombinant antigens and therefore a standardized whole-blood assay.
This preliminary study has several limitations. The patients number was limited, although it must be taken into account that AE is a rare condition and only 104 new cases were reported throughout In conclusion, our preliminary data show that the IL-4 whole-blood test is a promising assay to rule out AE in case of test positivity, in the context of a differential diagnosis with CE. This finding, combined with its potential usefulness in discriminating active from inactive CE stages 11,12 and the possibility to standardize the test, underscores this assay's potential to further improve the diagnosis of CE.

ACK N OWLED G EM ENTS
The authors are grateful to all the patients, nurses (in particular

TA B L E 3
Comparison of the wholeblood test sensitivity for CE or AE diagnosis participated in the interpretation of data; FT, WCA and MH participated in the interpretation of data and contributed to the writing of the manuscript; DG designed the study, coordinated and supervised the project, contributed to the interpretation of the results, contributed to the writing of the manuscript. All authors discussed the results and approved the final version of the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.