Accuracy of an experimental whole-blood test for detecting reactivation of echinococcal cysts

Background Cystic echinococcosis (CE) is a complex disease for which clear understanding of clinical manifestations is needed to avoid misdiagnosis, inappropriate treatment, and severe complications. We evaluated the accuracy of a whole-blood stimulation test based on Interleukin (IL)-4 detection in response to Antigen B (AgB) of Echinococcus granulosus sensu lato to discriminate cyst viability and detect cyst reactivation in patients with hepatic CE. Methodology/Principal findings Thirty patients with CE3b cysts and 37 patients with spontaneously-inactivated CE4-CE5 cysts were recruited (T0). After enrollment, 5 patients with CE3b cysts received albendazole, resulting in cyst solidification (CE4) in 4/5. Within a two-year follow-up, the whole-blood test was repeated at two time-points, in ≥14 (T1) and in ≥4 (T2) patients per group. IL-4 and a panel of other soluble factors were measured in the stimulated plasma. Baseline IL-4 levels were significantly higher in patients with CE3b compared to those with CE4 cysts (p = 0.006). Test accuracy for CE3b diagnosis had a sensitivity of 33–60% and a specificity of 76–95%, depending on the cut-off applied. Overall, IL-4 levels did not change significantly over time in either group; however, patients within the CE3b group showed a significant decrease of IL-1ra, IL-6, IL-8, G-CSF, IFN-γ, IP-10, MCP-1, MIP-1α, FGF at T1 compared to T0 (p≤0.042). Conclusions/Significance Whole-blood IL-4-response to AgB is significantly higher in patients with active compared to inactive CE but apparently not modulated over time after treatment. On the contrary, the levels of IL-1ra, IL-6, IL-8, G-CSF, IFN-γ, IP-10, MCP-1, MIP-1α, FGF significantly decreased in active CE during follow-up. Additional studies are needed to understand whether these findings might have a clinical significance for patients’ follow-up.

Introduction Cystic echinococcosis (CE) is a parasitic zoonosis globally distributed, caused by infection with the larval stage (cysts) of Echinococcus granulosus sensu lato, endemic in rural livestockraising areas, including Italy [1]. Human CE has considerable medical and economic impact, causing the loss of estimated >1 million disability adjusted life years (DALYs) every year [2]. CE is a complex disease with clinical presentations ranging from asymptomatic infection to severe, even fatal, disease [3]. CE cysts develop most commonly in the liver, followed by lungs, but all organs/tissues may be affected. Furthermore, CE cysts may present in different stages, each with a peculiar morphology and different recommended clinical management approach [4]. Inaccurate diagnosis and staging contribute to the health and economic burden that CE causes on patients and health systems in endemic areas. Currently, the gold standard for abdominal CE diagnosis and staging is ultrasound (US) [4], whereas serology, due to its limitations [5,6], can only support the diagnosis in cases where imaging is inconclusive [4].
US allows cyst classification into CE1, CE2, CE3a, CE3b, CE4, and CE5 stages [4]. This classification reflects the different "activity" of cyst stages, with cysts in CE1, CE2 and CE3b stages being active, CE3a transitional, and CE4-CE5 inactive [7]. Cysts that reach the CE4 stage as the consequence of therapy have a higher reactivation rate compared to those that became inactive spontaneously. Thus, inactive cysts can reactivate [8][9][10][11], calling for a long-term (5-10 years) US follow-up. However, US requires specific expertise, not always available in rural areas where the disease is endemic. The development of a laboratory test able to assess cyst viability would facilitate the classification of patients and cysts, and their rational assignment to a clinical management option, reducing risks and costs of overtreatment of inactive cysts [12]. It would also improve the implementation of a systematic patient's follow-up in areas where access to imaging and specific expertise is unavailable, allowing a "triage" of patients requiring referral to specialized centers.
Cytokine-release tests based on the stimulation of whole-blood in vitro are currently used for the diagnosis of latent tuberculosis [13][14][15] and have been explored in several infectious diseases such as hepatitis B [16], toxoplasmosis [17], cytomegalovirus infection [18], and COVID-19 [19]. An interleukin (IL)-4-release assay based on the in vitro stimulation of whole-blood with an enriched fraction of Antigen B (AgB) of E. granulosus or its peptides, has been set-up and preliminarily evaluated for CE diagnosis, showing potential to distinguish patients with active and inactive CE cysts [20][21][22]. Furthermore, several works described associations between cytokine profiles and CE clinical features [23][24][25][26][27][28]. In this prospective study, we explored the accuracy of the cytokine-release whole-blood test in discriminating cyst viability and in detecting cyst reactivation over time in patients with active CE3b and spontaneously-inactivated cysts.

Study population
Patients with hepatic CE cysts in care at the outpatient clinic of San Matteo Hospital, who signed the informed written consent, were prospectively enrolled from September 2015 to September 2017. CE cysts were diagnosed and staged based on the presence of pathognomonic features on US, according to the WHO-Informal Working Group on Echinococcosis classification [4]. Two study groups were defined: the "CE3b-group" and the "CE4-group".
Inclusion criteria for the "CE3b-group" were presence of �1 hepatic CE cyst in CE3b stage, regardless of previous treatment, or presence of �1 hepatic CE cyst in CE4 stage resulting from the albendazole treatment of a CE3b cyst. Patients with cysts in this latter condition were included in the "CE3b-group" because, in the vast majority of cases, CE4 cysts resulting from albendazole treatment of CE3b cysts are only temporarily inactive, and reactivate to CE3b on average 6-12 months after inactivation [9]. The exclusion criterion for the CE3b-group was the presence of concomitant (hepatic or extrahepatic) cyst in CE1, CE2, or CE3a stage.
The inclusion criterion for the "CE4-group" was the presence of �1 hepatic CE cyst in CE4 or CE5 stage, spontaneously-inactivated. Exclusion criteria for CE4-group were the presence of concomitant (hepatic or extrahepatic) cysts in CE1, CE2, CE3a, or CE3b stages and having received previous treatment.
Demographic and clinical information were collected at enrollment (baseline-T0); blood samples for routine serology and for the experimental whole-blood test were collected at baseline and at two consecutive follow-up time points (T1 and T2) over two years. Routine serology tests performed at San Matteo Hospital Foundation were RIDASCREEN Echinococcus IgG ELISA (R-Biopharm, Darmstadt, Germany), Cellognost Echinococcosis indirect hemagglutination (IHA; Siemens, Erlangen, Germany), Western Blot Echinococcus WB IgG (LDBIO Diagnostics, Lyon, France), and VIRapid HYDATIDOSIS immunochromatographic test (ICT; Vircell, Granada, Spain).

Statistical analysis
The analysis was carried out with Prism 8 software (Graphpad Software 8.0, San Diego, MO, USA). Medians and interquartile ranges (IQR) were calculated for continuous measures. Groups were compared using Mann-Whitney U, Wilcoxon, and Chi-square tests, as appropriate. Receiver Operator Characteristic (ROC) analysis was used to define cut-off values and sensitivity/specificity of the test; Spearman Rank Correlation was applied for correlations (r s >0.7 was considered high correlation, 0.7<r s >0.5 moderate correlation and r s <0.5 low correlation).

Study population
We enrolled 30 patients within the CE3b-group and 37 within the CE4-group. Their demographic and clinical information are shown in Table 1.
Within the CE3b-group (n = 30), at enrollment 26 patients had CE3b cysts and 4 had treatment-inactivated CE4 cysts deriving from CE3b cysts, with the last treatment dating back 18 months before enrollment (median: 18 months; IQR: 12-33 months). Fifteen patients were evaluated at T1 (median time after enrollment: 6 months; IQR: 6-12 months). After enrollment, albendazole was administered to 5 patients with CE3b cysts, which induced cysts solidification (CE4 stage) in 4 of them, and therapy started before enrollment was continued in one patient who showed treatment-induced CE4 at baseline. This patient showed reactivation (to CE3b) of the cyst at T1. The patients not receiving albendazole at enrollment did not show any change of the cyst stage at T1. The second follow-up (T2) was performed in 5 patients (median time after enrollment: 24 months; IQR: 15-24 months). None of them received albendazole between T1 and T2. The patient showing the transition from albendazole-induced CE4 at T0 to CE3b at T1 received surgical resection of the cyst between T1 and T2, and was therefore excluded from the T2 analysis. At T2, 3/5 patients showed a reactivation of the cysts compared to T1.

The AgB-specific IL-4-response is significantly higher in the presence of CE3b cysts
IL-4 levels at baseline were significantly higher in the CE3b-group compared to the CE4-group (p = 0.006) ( Fig 1A).
Moreover, a low but significant positive correlation (r s = 0.43, p = 0.02) was found in patients in the CE3b-group between the IL-4 levels and the total number of CE cysts of each patient (S1 Fig). IL-4 levels were compared at baseline among patients with CE3b cysts (n = 26 in the CE3b-group), with treatment-induced CE4 cysts (n = 4 in the CE3b-group) and with spontaneously-inactivated cysts (n = 37; CE4-group). The highest IL-4 levels were found in patients with CE3b cysts compared to the other groups (IL-4 medians: CE3b 0.54 pg/mL> treatment-induced CE4 0.33 pg/mL> spontaneously-inactivated cysts 0.09 pg/mL) and the difference was significant comparing patients with CE3b cysts and patients with spontaneously-inactivated cysts (p = 0.014) (S2 Fig).
To explore the correlation between IL-4-response and the humoral response, patients were stratified based on serology (excluding patients with discordant serology results, n = 4). In the CE3b-group, a higher IL-4-response to AgB was observed in sero-positive compared to seronegative patients (S3A Fig); no differences were found between patients in the CE4-group with positive or negative serology (S3B Fig).

In the CE3b-group, the AgB-specific IL-4-response changed over time compared to baseline independently of therapy
After enrollment, the whole-blood test was repeated at T1 and T2 in 15 and 5 patients in the CE3b-group and in 14 and 4 patients in the CE4-group, respectively. As expected, we observed no change in IL-4 levels in the CE4-group at T1 compared to T0 (p = 0.38) with the exception of 4 patients, 3 of whom showed a �34% decrease and 1 an increase of 118% (Fig 2A). IL4-specific response is significantly associated to CE3b cysts. A. IL-4 levels are significantly increased in the CE3b-group compared to CE4-group. B. AUC analysis for CE3b-group diagnosis; the 0.25 pg/mL cut-off maximizing sensitivity for CE3b-group diagnosis (panel A, dashed line) showed a 60% sensitivity and 76% specificity whereas the 1.39 pg/mL cut-off to maximize specificity (panel A, dotted line) showed 33% sensitivity and 95% specificity. C. The IL-4 levels were significantly increased in patients with CE3b cysts compared to the CE4-group. D. AUC analysis for CE3b diagnosis; the cut-off of 0.34 pg/mL maximizing sensitivity for CE3b cysts (panel C, dashed line) predicted CE3b with a 57.7% sensitivity and 78.4% specificity; the cut-off of 1.39 pg/mL cut-off to maximize specificity (panel C, dotted line) predicted CE3b with 38.5% sensitivity and 95% specificity. Footnotes: Horizontal bars represent medians. IL-4 concentrations were determined by ELISA. Responses in panels A and C were compared using the Mann-Whitney test; ROC-analysis was applied in panels B and D; differences were considered significant at p-values of �0. 05

PLOS NEGLECTED TROPICAL DISEASES
IL-4 levels at T2 did not significantly change (p = 0.43) with the exception of 1 patient who showed a 66% decrease from T0 to T2 (Fig 2B).
Regarding the CE3b-group, the IL-4 levels at T1 compared to T0 decreased in 6 patients, increased in 2, and were unchanged in 7 patients. Overall, no significant differences were found between the two time points (p = 0.2). Five patients with CE3b cysts (red diamonds in Fig 2C and 2D) and one patient with treatment-induced CE4 stage (blue diamonds in Fig 2C  and 2D) were receiving albendazole at enrollment (dotted lines in Fig 2C). The treatment led to cyst solidification (CE4 at T1) in 4/5 of patients with CE3b at T0. Within these 4 subjects with therapy-induced CE4 cysts at T1, compared to T0 (Fig 2C), 2 patients showed a �33% IL-4 decline; one patient showed almost no change (7.6% decrease) and one patient showed an increase from 0 to 1.16 pg/mL. At T2, 3 of the 4 patients with CE4 cysts at T1 had their cysts reactivated to CE3b stage (red dots in Fig 2D) and the analysis of the IL-4 levels showed an increase only in 1 of them (Fig 2D). We also evaluated the IL-4-response stratifying patients in the CE3b-group at T1 as follows: patients not receiving albendazole at T0 (n = 7), still showing CE3b cysts at T1 after receiving albendazole at T0 (n = 2), and showing treatment-induced CE4 cysts at T1 after receiving albendazole at T0 (n = 4). The two patients with treatmentinduced CE4 cysts at T0 in the CE3b-group were excluded. Although no significant differences were found among the three subgroups, the highest IL-4 levels were found in patients with treatment-induced CE4 cysts (S4 Fig). To assess if other factors may be useful to monitor cysts reactivation/inactivation, we evaluated a panel of immune/growth factors in the AgB-stimulated plasma. The analysis was performed in 12 patients per study group, from whom plasma was available after the IL-4 analysis. In the CE3b-group, 10 patients had CE3b cysts at T0 and 4/10 received albendazole at enrollment that induced solidification of the cysts in 3/4 patients at T1; 2 patients had treatment induced-CE4 cysts at T0 and in 1/2 the cyst reactivated at T1.
No other factors were modulated at T1 compared to baseline ( S6 Fig and Table 2). The same analysis was conducted in patients in the CE4-group. No significant differences were found between T1 and T0, with the exception of IL-1ra that decreased in 9/12 patients evaluated ( S7 Fig and Table 2, p = 0.0273). A heatmap of all the factors evaluated in the CE3b and in the CE4 groups at T0 and at T1 is shown in the S8 Fig.   Fig 3. The AgB-specific IL-1ra, IL-6, IL-8, FGF, G-CFS, IFN-γ, MIP-1α, IP-10

Discussion
CE is a complex disease, although mostly benign, for which a clear understanding of clinical manifestations is needed to avoid misdiagnosis, inappropriate treatment, and severe complications [29]. The development of a laboratory test able to assess cyst viability and reactivation over time would facilitate the implementation of a regular follow-up of patients in areas where access to imaging and specific expertise is unavailable, allowing a "triage" of patients requiring referral to specialized centres. Serology has proven so far of limited use for this purpose [6,[30][31][32]. In this study, we evaluated the accuracy of an experimental whole-blood stimulation test based on the detection of cytokines release, to reflect cyst viability and detect reactivation during a two-year follow-up. The results showed that IL-4 measurement is not informative enough of cyst viability and reactivation to be used in clinical practice. However, other soluble factors such as IL-1ra, IL-6, IL-8, FGF, G-CFS, IFN-γ, MIP-1α, IP-10 and MCP-1 are modulated during follow-up, suggesting their potential for monitoring cysts viability. No other cytokines of the large panel tested was found significantly different between groups and time points.
CE is characterized by the coexistence of Th1 and Th2 responses; the Th2 response has been correlated with disease susceptibility/chronicity and the Th1 response with protective immunity [33]. E. granulosus-specific T-cell clones from patients with active cysts showed a mixed Th1/Th2 profile whereas clones from patients with inactive cysts a Th1 profile [34]. Recently, the proteomic analysis of circulating exosomes from CE patients confirmed the involvement of the Th1-Th2 and regulatory T-cell compartments in active CE [35]. Moreover, we showed that AgB-specific cells producing IL-2/TNF-α/Th2-cytokines were associated with presence of active cysts [36]. Studies evaluating the ex-vivo levels of circulating cytokines in patients with hepatic CE gave conflicting results [37,38]. However, when explored after invitro stimulation, we showed that the IL-4-specific-response using a whole-blood platform associated with cyst activity [20,21]. In agreement with these studies, here we confirmed that higher levels of IL-4 were associated with the presence of active CE3b cysts compared to inactive CE4 cysts. Test accuracy for CE3b diagnosis had a sensitivity of 60% with a specificity of 76%, or a sensitivity of 33% with a specificity of 95%, depending on the cut-off applied. The application of a lab-based test even with such moderate accuracy may still have a role in rural endemic areas where access to US diagnosis may be difficult, to refer for US evaluation those patients who might have active cysts, because CE3b cysts tend to remain stable over time [9] and do not generally need emergency treatment but rather careful planning of the best clinical management. Unfortunately, a stable trend of the cytokine response over time according to cyst evolution could not be detected, possibly due to a too short follow-up, and too few patients presented at T1 and T2 to properly estimate the usefulness of the test for follow-up.
When we explored other soluble factors, we found that their levels were not different at T0 between patients in the CE3b-and in the CE4-groups. However, in patients in the CE3bgroup, IL-1ra, IL-6, IL-8, FGF, G-CFS, IFN-γ, MIP-1α, IP-10 and MCP-1 significantly decreased during follow-up.
Interestingly, some of these factors, as IL-8, MCP-1, IL-6, MIP-1α, G-CFS, are linked to neutrophils recruitment and differentiation. Neutrophils are key actors during infections as they mediate the resolution of the inflammation. Their role in CE is intriguing. AgB inhibits neutrophils chemotaxis [39]; moreover, protoscoleces, but not AgB, have been reported to induce neutrophils activation and increase IL-8 production [40]. If protoscoleces are accidentally released from cysts, their mediated activation of neutrophils is modulated by AgB, possibly allowing the development of secondary cysts [39]. Therefore, the reduction of neutrophilassociated factors during follow-up in patients with CE3b cysts may have the dual interpretation of deriving from inflammation resolution or, more likely, being the result of a negative immunomodulatory effect mediated by AgB. These findings need to be confirmed in larger studies to establish associations with inactivation/reactivation. This study had several limitations, including the low number of patients fulfilling the inclusion criteria, the short period of the study, and the small number of patients who attended the follow-up visits. A future study should envisage a longer recruitment period and follow-up length to guarantee the inclusion of an adequate sample size to confirm our findings. On the other hand, this study had the strength of the prospective design and of the stringent characterization of the enrolled population.
To conclude, we showed that the in-vitro IL-4-specific-response measurement in wholeblood, although associated with cyst viability, is probably not informative enough of the viability and reactivation of the cyst to be used Although patients in the CE3b-group showed higher levels compared to patients in the CE4-group for the majority of the analytes considered, no significant differences were found between the two groups at baseline. Footnotes: Horizontal bars represent medians. Analytes concentrations were determined by luminex. Responses in all the panels were compared using the Mann-Whitney; differences were considered significant at p-values of �0.05. IL: Interleukin; ra: receptor antagonist; FGF: fibroblast growth factor; G-CSF: granulocyte-colony stimulating factor; GM-CSF: granulocyte-macrophage colony-stimulating factor; IFN: interferon; IP: IFN-γ-induced protein; MCP: monocyte chemoattractant protein; MIP: macrophage inflammatory protein; PDGF: Platelet-derived growth factor; RANTES: regulated on activation, normal T cell expressed and secreted; TNF: tumour necrosis factor; VEGF: vascular endothelial growth factor. (TIF) No factors, excepted IL-1ra, were modulated at T1 compared to baseline. Footnotes: Analytes concentrations were determined by luminex. Responses in all the panels were compared using the Wilcoxon test; differences were considered significant at p-values of �0.05. IL: Interleukin; ra: receptor antagonist; FGF: fibroblast growth factor; G-CSF: granulocyte-colony stimulating factor; GM-CSF: granulocyte-macrophage colony-stimulating factor; IFN: interferon; IP: IFN-induced protein; MCP: monocyte chemoattractant protein; MIP: macrophage inflammatory protein; PDGF: Platelet-derived growth factor; RANTES: regulated on activation, normal T cell expressed and secreted; TNF: tumour necrosis factor; VEGF: vascular endothelial growth factor. (TIF) S8 Fig. Heatmap showing soluble factors concentrations in the CE3b-group and in the CE4-group at T0 and at T1. Each cytokine value was normalized by subtracting the mean cytokine value calculated for each specific cytokine within each patient-group (CE3b or CE4 groups). Subsequently, this value was divided by the standard deviation calculated for the specific cytokine within each patient-group (CE3b or CE4 groups). Colour codes refer to "red" for the highest expression and "green" for the lowest expression levels. Footnotes: CE: cystic echinococcosis; pt: patient; IL: Interleukin; ra: receptor antagonist; FGF: fibroblast growth factor; G-CSF: granulocyte-colony stimulating factor; GM-CSF: granulocyte-macrophage colonystimulating factor; IFN: interferon; IP: IFN-γ-induced protein; MCP: monocyte chemoattractant protein; MIP: macrophage inflammatory protein; PDGF: Platelet-derived growth factor; RANTES: regulated on activation, normal T cell expressed and secreted; TNF: tumour necrosis factor; VEGF: vascular endothelial growth factor. (TIF)