Evaluation of the LDBio Aspergillus ICT lateral flow assay for serodiagnosis of allergic bronchopulmonary aspergillosis

Background Early recognition and diagnosis of allergic bronchopulmonary aspergillosis (ABPA) is critical to improve patient symptoms, and antifungal therapy may prevent or delay progression of bronchiectasis and development of chronic pulmonary aspergillosis. Objective A recently commercialized lateral flow assay (Aspergillus ICT) (LDBio Diagnostics, Lyons, France) detects Aspergillus-specific antibodies in <30 minutes, requiring minimal laboratory equipment. We evaluated this assay for diagnosis of ABPA compared to diseased (asthma and/or bronchiectasis) controls. Methods ABPA and control sera collected at the National Aspergillosis Centre (Manchester, UK) and/or from the Manchester Allergy, Respiratory and Thoracic Surgery research biobank were evaluated using the Aspergillus ICT assay. Results were read both visually and digitally (using a lateral flow reader). Serological Aspergillus-specific IgG and IgE, and total IgE titres were measured by ImmunoCAP. Results For 106 cases of ABPA versus all diseased controls, sensitivity and specificity for the Aspergillus ICT were 90.6% and 87.2%, respectively. Sensitivity for ‘proven’ ABPA alone (n = 96) was 89.8%, and 94.4% for ‘presumed’ ABPA (n = 18). ‘Asthma only’ controls (no bronchiectasis) and ‘bronchiectasis controls’ exhibited 91.4% and 81.7% specificity, respectively. Comparison of Aspergillus ICT result with Aspergillus-specific IgG and IgE titres showed no evident immunoglobulin isotype bias. Digital measurements displayed no correlation between ImmunoCAP Aspergillus-specific IgE level and ICT test line intensity. Conclusions The Aspergillus ICT assay exhibits good sensitivity for ABPA serological screening. It is easy to perform and interpret, using minimal equipment and resources; and provides a valuable simple screening resource to rapidly distinguish more serious respiratory conditions from Aspergillus sensitization alone.


Introduction
Centre (NAC) (Manchester, UK) (n = 67) and samples archived in the Manchester Allergy, Respiratory and Thoracic Surgery (ManARTS) (Manchester, UK) research tissue bank (n = 46). The NAC is a nationally commissioned service providing long-term specialist care for patients with aspergillosis throughout the UK. For convenience sampling, residual sera from patient samples collected as part of routine care between 06/2018 and 01/2019 were identified. Control serum samples obtained from ManARTS were collected under ethical consent between 2011-2016 approved by the Local Research Ethics Committee (LREC, ethics approval 15/NW.0409). All sera were stored at -80˚C until use. This study was assessed through the NHS Health Research Authority system (HRA) (http://www.hra-decisiontools.org.uk/ research/) and was found to meet the UK NHS definition of a retrospective service evaluation for which formal ethical review was therefore not required.
Diseased controls were obtained from the ManARTS biobank, a research tissue bank comprised of samples from consenting patients with respiratory and allergic diseases (and other related conditions) and patients undergoing thoracic surgery, at the Northwest Lung Research Centre at Wythenshawe Hospital (Manchester University NHS Trust Foundation). Serum samples from 164 patients with bronchiectasis and/or asthma were identified using two sets of inclusion criteria. 'Asthma' control cases (n = 93) were defined as asthmatic patients with a normal chest X-ray and total IgE <1000 IU/ml within one year of sample collection; and with no clinical or diagnostic evidence of CPA, ABPA, bronchiectasis or any other lung diseases that might predispose to CPA including COPD, sarcoidosis, prior tuberculosis, cancer, and rheumatoid arthritis. Patients with Aspergillus sensitivity, allergic rhinitis/nasal polyps, severe food or drug allergies, or anaphylaxis were also excluded from this control group. 'Bronchiectasis' control cases (n = 71) included patients with computed tomography evidence of bronchiectasis. Patients with a diagnosis of ABPA, CPA, or Aspergillus bronchitis, or signs of ABPA or CPA on radiology were excluded. Bronchiectasis cases with underlying asthma (n = 48) and/ or COPD (n = 10) were included, as well as patients with microbiological (respiratory culture) evidence of Aspergillus colonization (n = 3) or raised Aspergillus-specific IgE (> 0.35 kU A /ml IgE) (n = 5).

Serological analysis
Each sample was tested using the Aspergillus ICT IgG-IgM (LDBio Diagnostics, Lyon, France) lateral flow assay. Test kits were shipped at ambient temperature and stored at 4˚C upon receipt. Each batch (10/pack) of ICT cartridges were equilibrated to ambient laboratory temperature before use and were run according to the manufacturer's instructions. Briefly, 15μl of sera was dispensed onto the ICT sample application pad, followed by four drops of eluting solution (provided with kit). The test was read at 30 minutes and results were interpreted both visually (by 'eye') and digitally using the Qiagen ESEQuant LR3 (Lake Constance, Germany) lateral flow reader. Both reads were conducted by the same user, with the visual reading conducted first to eliminate bias resulting from the digital reading. For visual reads, the test was determined positive by the appearance of two lines: a blue positive control ("C") line, and a black positive test ("T") line. The appearance of any black line at the "T" marker was considered positive, as recommended in the manufacturer's guidelines. Using the LR3 lateral flow reader, a positive test was defined by detection of peaks (any height) between 46.0-48.0mm and between 53.5-55.5mm for control and test lines, respectively.

Routine diagnostics
In many cases, Aspergillus-specific IgG and IgE, and total IgE levels were measured on ABPA patient samples as part of routine clinical care. Testing was carried out by the Manchester University NHS Foundation Trust, Department of Immunology, using the automated Immuno-CAP Phadia 1000 System.

Statistical analyses
To assess diagnostic performance, we calculated Youden's J statistic (sensitivity + specificity-1), likelihood ratios and the diagnostic odds ratio (DOR) [17]. Binomial confidence intervals (95%) were calculated for sensitivity, specificity, and DOR. Pearson's chi-square statistic was used for comparisons between subgroups. Spearman's rank correlation coefficient (ρ) was used for correlation between ImmunoCAP titers and ICT results. For all results, a two-tailed P-value < 0.05 was considered statistically significant.

Patients and sera
Patient characteristics for 106 ABPA patients and 141 diseased controls are shown in Table 1. Within the ABPA patient group, two subgroups were defined as 'proven' or 'presumed' according to total serum IgE level ('proven': >1000 IU/ml, 'presumed': 113-999 IU/ml). Diseased controls consisted of patients diagnosed with asthma and/or bronchiectasis. Subgroups for analysis were defined as (1) ABPA-S: serological ABPA (no bronchiectasis evident, ± asthma), and (2) ABPA-B: ABPA or control with bronchiectasis ± any other conditions (including asthma and COPD). Additionally, as part of routine testing, 64 of the 106 ABPA patients in this study had microbiological culture performed on respiratory samples (sputum), usually multiple specimens. High volume fungal culture technique was usually used, which has a higher yield than routine culture [18]. Seven (7) cases yielded no growth. Aspergillus fumigatus was the main pathogen isolated in the remaining 57 cases, and 21 of these also grew other Aspergillus species in culture. The following data was not available for 'asthma only' control subjects: (1) gender, (2) whether sputum culture was performed (and results).
Analysis of subgroups based on underlying conditions demonstrated a significant difference in sensitivity between the ABPA-B and ABPA-S subgroups (93.9% and 79.2% sensitivity, respectively; Pearson's chi-square statistic = 4.719, P = 0.03). No significant differences in specificity were observed. For three (3) cases of bronchiectasis with evidence of Aspergillus colonization by positive sputum culture results, there were no ICT positive results (100% specificity) and for five (5) cases of bronchiectasis with raised Asp IgE, two (2) cases were ICT positive (60% specificity). We also evaluated serum from seven (7) cystic fibrosis patients with 'proven' ABPA and found 100% sensitivity using the ICT for this group. Youden's J statistic calculated for overall ICT test results indicated a good balance between sensitivity and specificity. The test had an overall positive likelihood ratio of 7.07 (95% CI, 4.72 to 10.59) and negative likelihood ratio of 0.11 (95% CI, 0.06 to 0.20), with a high DOR of 65 (95% CI, 30 to 145). All tests were read both visually (i.e., by eye) and digitally (using the Qiagen ESEQuant LR3 lateral flow reader), with 100% agreement between methods.  All control: asthma ± bronchiectasis c n/a = data not available d COPD = chronic obstructive pulmonary disease. e A. niger [17], A. nidulans [3], A. fischeri [1], A. montevidensis [1], A. terreus [1], A. versicolor [1], A. pallidoflavus [1].
https://doi.org/10.1371/journal.pone.0238855.t001 In relation to Aspergillus species detected in sputum from ABPA patients (Table 3), the ICT performed well in detecting Asp antibodies for cases with A. fumigatus isolated alone (97.2% sensitivity) as well as for cases where non-A. fumigatus species were isolated in addition (95.2% sensitivity), with no significant difference (Pearson's chi-square statistic = 0.154; P = 0.69). There was no significant difference in sensitivity between samples from patients with at least one Aspergillus species isolated by culture ('culture positive', 96.5% sensitivity) and patients whose sputum yielded no culture growth ('culture negative', 100% sensitivity). Of the 42 patients for whom no sputum culture was conducted, 34 were positive by ICT (81.0% sensitivity; 95% CI, 65.9% to 91.4%).
Finally, we assessed three (3) different production lots of the LDBio Aspergillus ICT assay kits using 30 ABPA serum samples and observed 100% agreement of results across all lots tested (data not shown).

ICT detection of immunoglobulin isotypes
The Aspergillus ICT uses a homogeneous antigen sandwich format to detect Asp antibody in patient sera (Fig 1A), and may theoretically detect immunoglobulin isotypes other than (as claimed by the manufacturer) IgG and IgM. For 52 ABPA samples in this study, ImmunoCAP Asp IgE and IgG results were available and compared to the ICT result. We observed no evident immunoglobulin isotype bias for the ICT assay (Fig 1B), but did observe a significant correlation between Asp IgG and IgE titers (Spearman's rank correlation coefficient ρ = 0.3524, P = 0.01).
Using data generated by digitally reading the ICT on the Qiagen LR3 reader, we found a weak but significant correlation between endpoint (30 minutes) test line peak height and ImmunoCAP Asp IgG titer (n = 46, Spearman's rank correlation coefficient ρ = 0.2961, P = 0.046), supporting our previous findings [16]. However, as with the previous findings 16 , the correlation was not sufficient for quantification of Asp IgG. We found no significant correlation between peak height and ImmunoCAP total IgE (n = 48, Spearman's rank correlation coefficient ρ = 0.2101, P = 0.15) or Asp IgE (n = 48, Spearman's rank correlation coefficient ρ = 0.0295, P = 0.84) titers.

Discussion
Global prevalence of ABPA is estimated at 1-4% of adult asthma patients (approximately 4.8 million people) worldwide [3], but regional estimates may be higher. A study in Rio de Janiero

PLOS ONE
found 20% prevalence among Brazilian asthmatic patients [19], and several studies in India found ABPA in approximately 5% [20] to 8% [5,21], and up to 20% [22] of asthmatics, with up to half of patients being misdiagnosed as pulmonary tuberculosis [9]. The incidence of ABPA is also higher in patients with severe asthma and/or corticosteroid-dependent asthma (7-14%), and in patients with atopy [23]. Diagnosis of ABPA relies on a body of clinical, radiological, and immunological (and/or mycological) evidence; and the degree of difficulty in diagnosing ABPA largely depends on the staging and severity of disease. Diagnosis is relatively straightforward when a patient exhibits all clinical symptoms including bronchiectasis, and most patients are diagnosed at this stage [24]. Earlier diagnosis, however, is ideal to prevent development of bronchiectasis and irreversible tissue damage [25]. Serological tests are a useful tool for early detection of ABPA and comprise a significant portion of the recently suggested guidelines for ABPA diagnosis [10], with raised total IgE and Asp IgE (or positive skin prick test) being 'major' criteria and raised Asp IgG or positive precipitins test included under 'minor' criteria. The LDBio Aspergillus ICT lateral flow assay is a new diagnostic test requiring minimal time and resources. Our recent evaluation in CPA patients (vs. healthy controls) determined the test to have good sensitivity (91.6%) and specificity (98.0%), and a significant improvement in performance over our current workhorse assay (ImmunoCAP EIA) to detect raised Asp IgG [16]. The current evaluation in ABPA patients and diseased controls from the United Kingdom has shown the Aspergillus ICT to have good overall sensitivity (90.6%) and specificity (87.2%) in distinguishing patients with ABPA from those with underlying respiratory disease. Not surprisingly, the test exhibited significantly better sensitivity for ABPA with bronchiectasis (ABPA-B) (93.9%) than for serological ABPA (ABPA-S) ± asthma (no evidence of bronchiectasis) (79.2%). Bronchiectasis is associated with progression of ABPA and a more severe form of disease [1]. Persistent Aspergillus infection may lead directly to bronchiectasis or drive recurrent exacerbations and progression of bronchiectasis already present due to ABPA or severe asthma [26,27]. Once present, the pathophysiology of bronchiectasis can facilitate persistent infection through impairment in mucociliary clearance [28] and structural/tissue damage that creates a permissive environment for establishment of infection [26,29,30]. Patients with bronchiectasis (with or without ABPA) may develop Aspergillus bronchitis, and this entity is also associated with raised Asp IgG antibodies [31]. We did not explicitly assess a group of asthmatic patients with Aspergillus sensitisation for Aspergillus IgG with the ICT or ImmunoCap, and given the likely interaction with Aspergillus bronchitis, we would need to do so knowing if Aspergillus bronchitis was or was not present.
All cases in this study with Aspergillus species growth in high-volume sputum culture were positive for Aspergillus fumigatus and approximately one-third (37%) were also positive for other Aspergillus species. There was no significant difference in ICT performance between cases with A. fumigatus growth alone versus those with other species present. There were no cases in this study with non-A. fumigatus ABPA, however, a limited number of non-A. fumigatus CPA cases were evaluated in a previous study and we found no significant difference in ICT performance between A. fumigatus and non-A. fumigatus CPA cases [16]. The ability to detect antibody to non-A. fumigatus may be of particular importance in regions where ABPA due to non-A. fumigatus species is more prevalent [32]. For the diagnosis of ABPA, sputum culture is considered a supportive (not diagnostic) test [10]. In Indian patients thought to have ABPA, sputum tested by microscopy and culture was positive for Aspergillus species in only 32.5% of 203 cases [8]. Higher volume cultures yielded a positive in 56% of 75 cases (similar to this series), some of whom were on antifungal therapy compared with conventional culture (12.5%) [18]. Aspergillus ICT was more sensitive than this. Culture has the advantage of being able to check antifungal susceptibility, if positive, although direct detection of resistance is now possible [33][34][35][36].
Given the 'major' and 'minor' requirements of both raised Asp IgE and IgG for ABPA diagnosis, we sought to determine if the homogeneous antigen-sandwich format of the Aspergillus ICT rendered it capable of detecting IgE (in addition to IgG and IgM as claimed) and whether the test displayed an isotype bias. Although we found a significant correlation between levels of Asp IgG and Asp IgE, we did not observe any such isotype bias using the ICT. As expected based on our previous findings [16], we found the test to perform well at increasing levels of Asp IgG (ImmunoCAP 'positive', >40mgA/L) regardless of Asp IgE titer. The test also performed well at raised levels of Asp IgE paired with values of Asp IgG considered 'negative' by the recommended ImmunoCAP cut-off (� 40mgA/L). While this may possibly indicate the

PLOS ONE
detection of Asp IgE by the ICT test format, we have previously found this assay to detect Asp antibody in patients with clinically confirmed CPA who have tested 'negative' by ImmunoCAP Asp IgG [16,37]. Further studies would be necessary to evaluate the individual contributions of IgG and IgE from patient sera and the antibody class-specific performance of the Aspergillus ICT.
Individual serological tests alone for total IgE or Asp IgE or IgG are not specific for ABPA. It is important to note that routine tests used for ABPA diagnosis-raised Asp IgE, IgG and/or positive precipitins test-are also used in the diagnostic pathways for Aspergillus diseases other than ABPA including Aspergillus bronchitis (AB), acute and sub-acute invasive aspergillosis (IA), CPA, severe asthma with fungal sensitization (SAFS), and chronic or granulomatous Aspergillus rhinosinusitis (RS) [12]. The diagnostic interpretation of these tests, alone or in combination with each other and/or clinical presentation is summarized in Fig 2. Under generally accepted diagnostic criteria for Aspergillus disease, ABPA is the only condition using the combined requirements of raised Asp IgG and IgE. In practice however, it is not uncommon to see raised total and/or Asp IgE in CPA patients [38], especially those that have developed CPA as a result of untreated ABPA progression [39]. Additionally, although a raised level of Asp IgG is considered an exclusion criteria for SAFS, approximately 10% of all asthmatics present with raised Asp IgG [1] and there are likely to be cases of SAFS with increased levels of both Asp IgG and IgE, but lacking the clinical and radiological features of ABPA [12]. High titers of total serum IgE are encountered not only in ABPA, but also in cases of SAFS, Aspergillus rhinosinusitis, and in atopic asthmatics [40,41]. Furthermore, cut-off values to distinguish these conditions are speculative and may even be different in ABPA complicating asthma versus ABPA complicating CF [4,42].
In this study of clinically confirmed cases of ABPA compared to diseased controls, we found the LDBio Aspergillus ICT to have good sensitivity and specificity. The test effectively distinguished between Aspergillus-sensitization complicating asthma and/or bronchiectasis, and underlying conditions. It is rapid (result in <30 minutes) and easy to perform, with simple result interpretation by visible inspection. Overall, the LDBio Aspergillus ICT exhibits excellent performance as a screening tool in the ABPA diagnostic pathway.