Introduction: The use of biological therapy in eosinophilic and atopic asthma has grown exponentially over the last 14years. Treatment was initially hospital-based, but the COVID-19 pandemic has accelerated the implementation of patient homecare self-administration (HSA) of biologics. Aim: To assess the stability of patients transfered on HSA by comparing data from biologic initiation to HSA discharge and annual review. Results: This report includes 56 patients who attended for annual review between April and December 2021 (60% female, mean age 54 [SD13.4]) for Benralizumab (42.9%), Mepolizumab (35.7%) and Omalizumab (21.4%). The time on biologic when commencing HSA was 19 (IQR 21.8) months, with an annual review 12 (IQR 8.3) months later. Previously obtained improvements in asthma control, lung function, eosinophil suppression and oral corticosteroid use, were maintained in 53(95%) of the patients (Table 1). HSA was stopped in 3(5%) patients due to deterioration in asthma control. Conclusion: The vast majority of patients recieving HSA of biologics maintained previous improvements across asthma outcomes, thereby strongly supporting the use of HSA in the correctly identified patient, consequently optimising service capacity. Appropriate monitoring arrangements are still needed to promptly identify any deterioration.

Introduction Asthma is common and can complicate pregnancy. Uncontrolled asthma is associated with adverse outcomes for both mother and baby. National and regional guidelines advocate good asthma control through continuous monitoring, patient education and stepwise approach to treatment. Objectives To assess if the Greater Manchester and Eastern Cheshire (GMEC) guidelines are being followed and identify where care can improve. Methods A prospective service evaluation of 40 women attending Wythenshawe hospital's general asthma clinic (March 2019 – March 2022) and severe asthma clinic (November 2020 – March 2022) Results Patients attending the general asthma clinic were commonly referred through A&E after an exacerbation which required oral corticosteroids (OCS) figure 1. These patients often required escalation of treatment, multiple courses of OCS and more frequent follow-up. In contrast, patients attending the severe asthma clinic were already known to the clinic prior to pregnancy, required fewer courses of OCS as well as – figure 1; fewer clinic appointments during their pregnancy. Women in the severe asthma clinic on new biologic treatment still required OCS. Overall, adherence to medication was good, however, novel asthma treatments on occasion were stopped by primary care. Furthermore, conversations regarding patients' asthma triggers, flu vaccination and personal asthma action plan were poorly executed, and monitoring wasn't comprehensive. This may have been influenced by a high incidence of telephone consultations, COVID-19 or these conversations occurring verbally. A well-known risk of uncontrolled asthma is delivering via caesarean section, we recorded that many patients in the severe cohort had elective c-sections, however in the general asthma cohort over 50% of our patients required an emergency c-section. Discussion The high number of emergency c-sections highlights the need to work closer with obstetrics in those patients with asthma, as well as the need for further research to support our findings. GMEC guidelines were poorly adhered to, due to their nonspecific nature. Consequently, a facilitated template is required to provide a simple reminder of the key discussion points and a means to record the important data collected, to optimise asthma care.

Background Anti-IL-5 monoclonal antibodies reduce systemic corticosteroid use and exacerbation rate in severe eosinophilic asthma. Although their specific mechanisms of action differ, there is as yet no guidance on whether to choose mepolizumab or benralizumab for any given patient. Aims To investigate clinical outcomes of patients switching to benralizumab after non-response to mepolizumab associated with persistent sputum eosinophilia. Methods We recruited patients who failed to respond to mepolizumab over 12 months (no meaningful reduction in maintenance steroid use and/or exacerbation rate), with sputum eosinophils above the threshold of 2%. Over the subsequent 12 months of benralizumab treatment we recorded daily prednisolone dose, ACQ6, AQLQ, and sputum & blood eosinophils. Lung function and FeNO testing was severely impacted by the COVID pandemic. Results Fifty-three of 165 (32%) mepolizumab patients were identified as candidates for a switch of therapy with residual positive sputum eosinophils, of which 36 have completed 12 months of treatment, 20 with available sputum eosinophil results. Five patients stopped therapy before a full year (two for failure to respond to treatment at 6–9 months, one allergic reaction, two unspecified). After 12 months on benralizumab, all patients had negative sputum eosinophils as well as improvements in ACQ6, AQLQ, blood eosinophils and reduced exacerbation rate compared to before switching (table 1). Conclusions Around a third of patients failed to respond to mepolizumab and displayed persistent airway eosinophilia. All of these patients had complete suppression of sputum eosinophils on benralizumab with significant clinical improvements in exacerbation rate, blood eosinophils, asthma control and quality of life. These real-life observations in our severe asthma patient population confirm the results of the SIROCCO and Calima trials.

Introduction Severe asthma patients were assumed to be at greater risk of morbidity from infection with the novel severe acute respiratory syndrome coronavirus (COVID-19), hence, in the UK, were advised to shield. Community data on COVID-19 infection in severe asthmatics is lacking. We assessed the burden of shielding, the impact of COVID-19 and the effect of asthma medication on the UK severe asthma population. Methods Adults previously consented to inclusion in the UK Severe Asthma Registry (UKSAR) across 14 centres were contacted in June 2020 to collect data on potential COVID-19 infection, asthma control and shielding. Electronic records, where available, were reviewed for confirmation. Data was combined with clinical data from the UKSAR. Univariate and multivariate logistic regression analyses were performed to identify risk factors for COVID-19 infection. Results 1365 patients were included. 1268 (93%) were advised to shield, 1131 (89%) patients who received shielding advice followed it. Men (OR 0.4, p=0.045) and those in non-shielding households (OR 0.27, p=0.001) were less likely to follow shielding advice. 544 (47%) of patients advised to shield reported worsening of mental health; females (OR 1.59, p=0.001) and those with history of anxiety or depression (OR 2.12 p=0.001) were at greater risk. 97 (7.1%) patients had suspected/confirmed COVID-19 infection, 19 (1.39%) PCR/serology confirmed infection, 13(0.95%) were hospitalised and 2 patients (0.15%) died (table 1). 918 (67%) were on biologic therapy, 515 (37%) maintenance oral corticosteroid (mOCS). Multivariate analysis showed neither biologic therapy (OR 0.73, p=0.165) nor mOCS (OR 1.18, p=0.427) increased the risk of COVID-19 infection. Patients on biologics were less likely to require an acute course of corticosteroids for asthma symptoms (OR 0.6, p=0.002) while patients on mOCS were more likely (OR 1.96 p≤0.001). Inhaled corticosteroids (ICS) were not associated with COVID-19 infection, including high dose (2000 mcg BDP equivalent) (OR 0.64, p=0.234). Hospitalised patients were on lower median doses of ICS vs non-hospitalised patients (1000 vs 2000 mcg BDP equivalent, p=0.002). Conclusion Hospitalisation and death occurred in small numbers in our severe asthma population. From this observational data, biologic agents for asthma were not associated with increased risk of COVID-19 infection or hospitalisation.

Background Anti-IL-5 monoclonal antibodies reduce systemic corticosteroid use, exacerbation rate, and airway inflammation in severe asthma. Different effectors of the eosinophil IL-5 pathway are targeted, but algorithms to choose the most appropriate biologic for each patient are still being refined. Our team regularly monitors severe asthma patients' airway inflammation using sputum eosinophils. Aims To investigate clinical outcomes of patients that switched to benralizumab after non-response to mepolizumab associated with persistent sputum eosinophilia. Methods We prospectively monitored sputum cell counts of patients starting MDT-approved benralizumab after: failing to respond to mepolizumab, and whose positive sputum eosinophils remained above the threshold of 3%. Additional clinical parameters were recorded including corticosteroid dose, ACQ, AQLQ, and blood eosinophils. Lung function testing was severely impacted by the COVID pandemic. Results Fifty-one of 183 (27.9%) mepolizumab patients to date have been identified as candidates for a switch of therapy with residual positive sputum eosinophils, of which 44 have received their first doses of benralizumab and 16 have completed 6 months of treatment with available sputum eosinophil results. After 6 months on benralizumab therapy, clinically and statistically significant improvements in ACQ and mAQLQ were observed compared to baseline. All but two patients had negative sputum eosinophils (< 3%) compared to levels on mepolizumab (Table). Both also had elevated blood eosinophils: one patient was sampled during an exacerbation, the other is being investigated for parasitic infestation due to frequent foreign travels. At baseline 81% of patients were taking maintenance oral corticosteroids compared to 44% at 6 months (NS). Conclusions More than a quarter of patients failed to respond to mepolizumab and displayed persistent airway eosinophilia. Of these, 88% achieved negative airway inflammation on benralizumab with significant clinical improvement (ACQ, AQLQ) and clinically (though not statistically on this small sample size) significant fall of OCS. Benralizumab appears to prove effective at reducing inflammation at tissue level in patients previously unable to achieve this on a different anti-IL-5 therapy. As more patients progress through their treatment, repeat monitoring at 12 months will establish if this benefit is maintained and is associated with long term steroid dose reduction, and exacerbation rate.

Introduction 250,000 individuals are affected by severe asthma, which can cause a huge physical and psychological burden. Severe asthma requires a comprehensive and systematic assessment, to confirm diagnosis, distinguish the correct phenotype, identify co-existing conditions and tailor therapy. Historically our initial assessment of the referred patient was dictated by a traditional medical model. We have redeveloped our processes to facilitate a daycase multi-disciplinary (MDT) systematic assessment for all patients referred to our service. Methods A retrospective review of patient records from baseline assessment and their 12-week follow-up was performed, to assess their initial outcomes after attending the MDT assessment. Results In the first 6 months 100 patients were referred 94% had a pre-existing diagnosis of asthma and 63% were female. Day-case assessment identified a primary diagnosis of atopic asthma (25%), eosinophilic asthma (34%), neutrophilic asthma (6%), occupational asthma (1%), and mixed phenotype or differential diagnosis (24%/10%). By second visit we had confirmed co-diagnosis of Tracheo-bronchomalacia n=20(20.4%), inducible laryngeal obstruction n=13(13.1%) and breathing pattern disorder n=20 (20.6%). Comparison of Asthma Control (ACQ) at baseline to 12 weeks review shows an overall improvement of 0.75 (m=0.75, SD 1.5 t (78)4.43, p<0.001) and a 0.76 (z=-2.7, p=0.005) improvement in Asthma quality of life (AQLQ). Poor inhaler technique was demonstrated by 48 (62%), fair technique by 17(22%) and good technique by 12(15%). Only 8.1% had an asthma action plan on referral. Through delivery of educational intervention at baseline, the ACQ at 12 weeks has shown the highest improvement in the group with the poorest technique dropping by 0.96(CI 0.45–1.4). The fair technique group dropped by 0.5 (CI 0.08–0.92) and good technique dropped by 0.4 (CI 0.11–0.93). Similarly, the non-adherent group (collection of <80% prescription refills) at baseline n=18 (25.3%) showed an improvement in their ACQ of 1.1 (C I 0.28–1.9), when compared to the adherent groups n=53(74%) ACQ of 0.67(CI 0.28–1.07). Conclusion Our results indicate that poor control may in part be due to poor adherence and inaccurate diagnosis. Through adopting an MDT systematic assessment, we can consider differential diagnosis and demonstrate an improvement in ACQ and AQLQ with a significant positive patient feedback.

In 2018, the Severe Asthma team at Wythenshawe Hospital assessed adherence to preventer inhalers in long-tern omalizumab patients and found that 50.6% of patients were non-adherent (1). In response, service improvements focussed on patient education and closer monitoring of adherence were implemented. All patients initiating on biologic therapies now have an education session with the Severe Asthma Pharmacist where the importance of adherence is highlighted. The Severe Asthma nurses reiterate this at each visit and adherence is reassessed after 12 months on biologics. To assess the impact of these service improvements we reviewed the adherence of 50 patients who had received 12 months of mepolizumab. In line with the North West Severe Asthma Network criteria for biologic approval, a patient was classified as adherent if they had collected over 80% (10/12) of prescriptions for their preventer inhaler, from their GP. In addition, ACQ-7 scores, forced exhaled nitric oxide (FENO) and forced expiratory volume (FEV1) at baseline and 12 months in the adherent vs non-adherent groups were compared. 10% (5/50) of patients had collected less than 80% of their preventer inhaler prescriptions in the last 12 months. In the non-adherent group ACQ-7 scores had risen at 12 months, indicating poorer asthma control. FENO increased in both groups. FEV1 had fallen in the adherent group. In conclusion, the results indicate that the service improvements we have implemented have led to improved collection of prescriptions for preventer inhalers. Non-adherent patients demonstrated a decrease in asthma control through a rise in their ACQ-7 scores. However, the median ACQ-7 and FENO score are better in the non-adherent group compared to the adherent group. These results compare to those seen in the omalizumab adherence review. This raises the question of whether non-adherence is related to patients' illness perception and a belief that they are better controlled and therefore do not require their preventer inhalers. In these patients the perceptions and practicalities approach (PAPA®) can be used to make targeted interventions to improve adherence. Reference Allen DJ, et al. Non-adherence with inhaled preventer therapy in severe asthmatic patients on long term Omalizumab. ERJ 2018;54:1.

Background Pseudomonas and other Gram-negative infections in chronic lung disease are associated with high levels of morbidity and mortality and are difficult to treat. Omalizumab has been available for more than 10 years and mepolizumab for the last 2 years as commonly used monoclonal antibodies in the treatment of severe asthma. Aim To evaluate a clinical suspicion of an increased prevalence of Pseudomonas spp.(and other Gram-negative bacteria) seen among our patients on long term omalizumab compared to a comparable population (those treated with mepolizumab). Methods The patient cohort was being treated at a large severe asthma service with either omalizumab (n=179) or mepolizumab (n=209). Sputum sample results from 01/01/17 to 11/02/19 on the electronic patient records for both groups were compared, along with demographic and disease severity data. Statistical analyses were performed using either Pearson's Chi-squared test or Student's t-test. Results Comparing the demographic features of the two treatment groups, there was no significant difference in the FEV1 (1.94L in the omalizumab group vs 2.02L, p=0.39) or the proportion of patients with evidence of bronchiectasis on CT (22.5% vs 19.8%, p=0.21). However, the mepolizumab cohort was older (52.2 vs 48.8 years old, p<0.05). There were 15.8 and 7.9 sputum samples/100 patients/year positive for potentially pathogenic bacteria in the omalizumab and mepolizumab groups respectively (p<0.05), and 3.7 versus 3.0 sputum samples/100 patients/year positive for Gram-negative bacteria (p=0.67). Of the omalizumab patients, 3.4% and grew Gram-negative bacteria in their sputum over the study period compared to 1.4% of the mepolizumab patients. Three patients (1.7%) grew Pseudomonas spp. in the omalizumab group, compared to none in the mepolizumab group. Conclusion This retrospective study lends support to a clinical suspicion of an excess of the Gram-negative bacteria and Pseudomonas spp.in patients on long term omalizumab. The cause could be a sampling bias, with more samples being performed on omalizumab patients, longer duration of follow up in a severe asthma service, or a possible switch to an infection phenotype in patients on long term anti-IgE therapy. This clinical concern needs further evaluation in a multi-centre longitudinal real life study.

Introduction Reslizumab is the second biologic available targeting Interleukin 5 (IL5), for the management of severe eosinophilic asthma. We use it in patients who have failed to respond to mepolizumab or with suspected vasculitis. Aims To determine the efficacy of reslizumab in reducing steroid dose and exacerbation rate (as well as e.g.: blood eosinophils, FEV1, weight) in patients previously unresponsive to mepolizumab (with proven persistent airway eosinophilia), or in selected anti-IL5-naive patients with suspected vasculitis. Methods Maintenance prednisolone dose and exacerbation history were prospectively recorded at baseline (Bas) and at six months (6M), along with weight, blood eosinophils, lung function, Asthma Quality of Life Questionnaire (AQLQ), Asthma Control Questionnaire (ACQ), fractional concentration of expired nitric oxide (FeNO). The data were analysed retrospectively. In view of the small number of patients descriptive statistics only are reported. Results Between January 2018 and April 2019, 17 severe asthma patients had received at least one reslizumab infusion: nine of these were anti-IL5-naive and eight had previously been unresponsive to mepolizumab. This latter group had worse AQLQ (2.6), exacerbations (4.5) and lung function (53% predicted) at baseline compared to the naive group (3.2, 1 and 59% respectively). Sixteen patients completed at least six months treatment, with the remaining patient stopping due to an adverse reaction (rash). In previous mepolizumab-failed patients, reslizumab reduced the median daily prednisolone dose from 20 to 15 mg over 6-months, and from 17.5 to 15 mg in the biologic-naive group. Summary Reslizumab appears to be an effective medication in reducing prednisolone use in patients who had failed mepolizumab, or who were considered too severe for mepolizumab at the UK licensed dose (patients with suspected of vasculitis). A larger population and for a longer duration of follow-up are needed to confirm these real-life patient findings.

Background The 2013 NHS England service specifications for severe asthma aimed to develop a limited number of high volume specialist hub centres. These centres would have a multi-disciplinary team (MDT) working to improve patient outcomes and reduce healthcare costs. In the North West we have developed a networked approach to specialised severe asthma services; the first Operation Delivery Network (ODN) for a chronic disease. Representatives from 11 NHS Trusts and a central hub undertake a monthly virtual MDT meeting with physicians, nurses, pharmacists, physiotherapists, clinical psychologists, speech and language therapists, and radiologists represented. All patients being considered for specialised treatments undergo MDT discussion for consensus approval of treatment. Aim To summarise the experience and case-mix encountered during the first 4 years of operation of our regional virtual severe asthma MDT. Methods We reviewed all cases discussed at the MDT between January 2015 and June 2018. Proformas are submitted via nhs.net accounts and data entered into a central database by MDT coordinator. All biologic proformas are pre-assessed by the ODN Pharmacist and Specialist nurse to ensure compliance with the NICE guidelines. Results During this period 41 meetings were held, with 933 cases discussed. Omalizumab was approved in 78% of cases submitted, BT in 39%, mepolizumab in 84% and reslizumab in 94%. The most common reasons for non-approval of omalizumab were insufficient steroid requirement, poor adherence and lack of allergy to perennial allergen. The most common reasons for non-approval to mepolizumab and reslizumab were poor adherence and lack of evidence of raised eosinophil levels within the stated 12 month period. The potential 5 year cost saving of high cost therapies not approved at the Central hub is estimated to be £7.5 million. Conclusion A multi-site virtual severe asthma MDT meeting facilitates expert care across a wide geographical area, ensures governance in the use of novel and expensive therapies, strengthens collaboration and aims to improve patient care. Reference Ryan D, et al. The UK's largest severe asthma multidisciplinary team meeting; experience from the first 18 months. ThoraxDec 2016:71(3);A5.

Background Omalizumab (Xolair) is licensed for the treatment of severe allergic asthma patients with IgE mediated disease. Within the UK NICE guidance includes criteria for documented adherence to Inhaled Corticosteroids (ICS) and other asthma related medications. Within the UK there is no clear definition of ‘documented compliance’ Within the Severe Asthma North West MDT this is defined as a≥80% collection of ICS based on GP data. Objective To assess adherence to ICS treatment for patients on Omalizumab therapy and to explore causal relationships and outcomes between the adherent and non-adherent groups. Methods Over a 2 week period in Feb 2017 patients attending for routine Omalizumab care (post 16 week assessment) underwent standard observations (FEV1, FeNO and ACQ-7) additionally each patient was asked to clarify the type and frequency of current ICS and adherence data was obtained from the patients GP for the last 6 months. Results Of 79 patients (67.1%) were female and (32.9%) were male, with a mean age of 50.7 (SD 14.12). ICS adherence was observed in 39 (49.4%) of patients. Comparison between the two groups (Table 1), demonstrates little difference between both groups. The adherent group have a lower baseline FEV1 (55%) and FeNO (15.5 ppb) compared to the non-adherent group (61.9% and 22.0 ppb), and demonstrate a higher percentage change in FEV1 from baseline (8.3% compared to 5.67%). Both groups demonstrate a significant improvement in ACQ-7 from baseline, and there was a trend (p=0.067) for better control (lower ACQ-7 score) in the non-adherent group. Conclusion Non adherence to ICS in patients on Omalizumab therapy is a significant issue. There were no significant differences between the 2 groups in terms of exacerbations or ICS treatment regimes. There was a statistical, but not clinically significant rise in FeNO in the non-adherent group. There was a trend towards better perceived asthma control as measured by the ACQ-7 in the non-adherent group which may influence ongoing adherence to ICS. Reference Asthma Control Questionnaire. https://www.qoltech.co.uk/acq.html

ORAL ABSTRACT SESSION 1—Asthma: from mechanisms to managementO01 Serum IL-1RL1-A levels predict an eosinophilic subtype of asthma in preschool wheezing childrenM. E. Ketelaar1, K. Van De Kant2, F. N. Dijk1, E. M. M. Klaassen3, N. Grotenboer4, M. C. Nawijn4, E. Dompeling2, G. H. Koppelman1 1University Medical Center Groningen, Beatrix Children's Hospital, Groningen Research Institute for Asthma and COPD (GRIAC), Groningen, The Netherlands; 2Department of Pediatric Pulmonology, School for Public Health and Primary Care (CAPHRI), Maastricht University Medical Center (MUMC+), Maastricht, The Netherlands; 3Department of General Practice, School for Public Health and Primary Care (CAPHRI), Maastricht University Medical Center (MUMC+), Maastricht, The Netherlands; 4University Medical Center Groningen, Department of Pathology and Medical Biology, Laboratory of Experimental Pulmonology and Inflammation Research (EXPIRE), Groningen Research Institute for Asthma and COPD (GRIAC), Groningen, The Netherlands Correspondence: Maria Elizabeth Ketelaar - m.e.ketelaar@student.rug.nl Clinical and Translational Allergy 2017, 7(Suppl 2):O01 Introduction: Respiratory symptoms are common in preschool children. However, which of these wheezers will develop asthma at school age, and what phenotype they will develop remains difficult to predict. Current models such as the asthma prediction index (API) are based on clinical parameters and have only modest predictive accuracy. Expression levels of well replicated asthma genes could potentially form novel biomarkers for asthma prediction. IL1RL1 is an asthma susceptibility gene, and has also been linked to eosinophilia. Therefore, we hypothesized that expression levels of IL1RL1 in the form of soluble IL-1RL1-a measured in serum from wheezing preschool children contribute to the prediction of asthma at school age. Moreover, since IL1RL1 was previously associated with blood eosinophilia, our second aim was to determine whether serum IL-1RL1-a levels predict eosinophilic asthma. Method: We used logistic predictive modeling in a prospective Dutch birth cohort (n = 202 wheezers), and calculated the area under the curve (AUC) of the sensitivity/1-specificity curves of potential models. Results: Neither IL-1RL1-a serum levels at age 2–3 years alone nor its combination with the API had predictive value for doctors' diagnosed asthma at age 6y (IL-1RL1-a alone: AUC = 0.50 [95 CI 0.41–0.59, P = 0.98], API + IL-1RL1-a: AUC = 0.57 [95 CI 0.49–0.66, P = 0.12]).However, IL-1RL1-a serum levels at age 2–3 years correlated with the severity of airway eosinophilia (determined by levels of exhaled fraction of NO, [FeNO]) in children who had developed asthma at age 6y (Pearson's R = −0.24, P = 0.046, N = 59). Logistic predictive modeling of eosinophilic asthma at age 6y (asthma with FeNO ≥ 20 ppb) showed that IL-1RL1-a serum levels itself and in combination with the API could predict this eosinophilic subphenotype of asthma (IL-1RL1-a alone: AUC = 0.65 [95 CI 0.52–0.79, P = 0.04], API + IL-1RL1-a: AUC = 0.70 [95 CI 0.56–0.84, P = 0.01]). Interestingly, IL-1RL1-a levels had a negative direction of effect. Conclusion: Our study shows that serum IL-1RL1–a levels measured in wheezing children at age 2–3 years do not predict doctors' diagnosed asthma as general phenotype at age 6 years, but negatively predict an eosinophilic subphenotype of asthma.This suggests that IL-1RL1 might play a protective role in the development of eosinophilia in children who experience asthma at school age and implies that IL-1RL1 targeted therapy could rather be further explored in the subphenotype of asthmatic children with predominant eosinophilic inflammation. Keywords: Childhood Asthma, Eosinophilic Asthma, Prediction, IL-1RL1, SerumFigure 1 Prediction of eosinophilic childhood using IL-1 RL1-a serum levels and the API O03 Diagnosing asthma in symptomatic children using lung function: evidence from a birth cohort studyClare Murray1, Philip Foden1, Lesley Lowe1, Hannah Durrington1, Adnan Custovic2, Angela Simpson1 1University of Manchester, Manchester, United Kingdom; 2Imperial College, London, United Kingdom Correspondence: Angela Simpson - angela.simpson@manchester.ac.uk Clinical and Translational Allergy 2017, 7(Suppl 2):O03 Introduction: In the UK, new national draft guidance for the diagnosis of childhood asthma proposes algorithms based on four tests of lung function, each used as a dichotomous variable (FEV1/FVC ratio less than the lower limit of normal [LLN], bronchodilator reversibility [BDR] ≥12%, FeNO ≥ 35 ppb and PEFR variability). However, accuracy of these tests in diagnosing asthma in children is unknown, as the evidence is largely derived from studies of adults. Within the setting of a population-based birth cohort (Manchester Asthma and Allergy Study—MAAS), we investigated the value of FEV1/FVC, BDR and FeNO in diagnosing asthma in children. Method: Using validated questionnaires we assessed study participants at age 16 years. Current asthma was defined as all three of: (1) doctor-diagnosed asthma ever, (2) wheezing in the previous 12 months and (3) current use of asthma treatment. We assigned children negative to all three features as non-asthmatic controls. Using ATS/ERS guidelines, we measured spirometry and FeNO (NIOX chemiluminescence analyser; Sweden). BDR was considered positive if FEV1 increased by ≥12% following administration of 400 mg of salbutamol. PEFR variability was not measured. To test the diagnostic algorithms simulating the clinic situation, we selected only children reporting recent symptoms of wheeze, cough or breathlessness who were not on regular inhaled corticosteroids (ICS). Results: Of the 630 MAAS children with full data available, 163 reported recent symptoms, but were not using regular ICS; 34 of these met our definition of current asthma, with 55 as non-asthmatic controls. In the multivariable logistic regression analysis, increasing FeNO was associated with an increased risk of asthma (OR 1.02, 95% CI 1.01–1.04, p = 0.006), with a trend for FEV1/FVC ratio (OR 0.95, 95% CI 0.87–1.02, p = 0.17), and no association for BDR (p = 0.94). The proportion of those with each combination of positive tests is show as a Venn diagram (Figure 1). Of 58 children with three negative tests, 29.3% had current asthma, accounting for 50% of those with asthma. Only 5.9% of those with asthma were positive to all three tests. Conclusion: Applying 3 tests of lung function to children with symptoms and a diagnosis of asthma failed to detect 50% of asthma cases. Proposed algorithms for the diagnosis of asthma in symptomatic children need to be tested prospectively. Keywords: Asthma, Diagnosis, FeNO, Lung Function, ChildrenFigure 2 Venn diagram showing number of children with symptoms (n163) who were positive for each combination of tests O04 Treatable traits in the European U-BIOPRED adult severe asthma cohortAndrew J. Simpson1, Dominick E. Shaw2, Ana R. Sousa3, Louise J. Fleming4, Graham Roberts5, Ioannis Pandis6, Aruna T. Bansal7, Julie Corfield8, Scott Wagers9, Ratko Djukanovic5, Kian Fan Chung4, Peter J. Sterk10, Jorgen Vestbo1, Stephen J. Fowler1 1Division of Infection, Immunity and Respiratory Medicine, School of Biological Sciences, The University of Manchester and University Hospital of South Manchester, NHS Foundation Trust, Manchester, United Kingdom; 2Respiratory Research Unit, University of Nottingham, Nottingham, United Kingdom; 3Respiratory Therapeutic Unit, GSK, Stockley Park, London, United Kingdom; 4National Heart and Lung Institute, Imperial College, London, United Kingdom; 5NIHR Southampton RespiratoryBiomedical Research Unit, Clinical and Experimental Sciences and Human Development and Health, Southampton, United Kingdom; 6Data Science Institute, South Kensington Campus, Imperial College London, London, United Kingdom; 7Acclarogen Ltd, St John's Innovation Centre, Cambridge, United Kingdom; 8AstraZeneca R&D, Mölndal, Sweden; 9BioSci Consulting, Maasmechelen, Belgium; 10Dept of Respiratory Medicine, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands Correspondence: Andrew J. Simpson - Andrew.Simpson-2@Manchester.ac.uk Clinical and Translational Allergy 2017, 7(Suppl 2):O04 Introduction: Individuals with severe asthma may remain uncontrolled and exacerbation-prone despite intensive guideline-directed treatment, and management options are limited. The concept of treatable traits, based on the identification of treatable disease-associated characteristics, may thus be a particularly useful framework in this context. In the Unbiased Biomarkers for the Prediction of Respiratory Disease Outcomes (U-BIOPRED) project we have recruited individuals with severe asthma (SA) under specialist care, and controls with mild to moderate asthma (MMA). Aim: To identify and quantify treatable traits within the U-BIOPRED adult asthma cohorts. Method: We defined criteria for treatable traits based on Agusti (Eur Respir J, 2016) and identified prevalence rates within the U-BIOPRED database. Chi Square tests were used to examine differences in frequency between individuals with SA and MMA Results: Data from 509 individuals with asthma were included in the analysis; 421 with SA and 88 with MMA. Twenty-nine treatable traits were identified, including 13 pulmonary, 13 extra-pulmonary and three behavioral traits. Pulmonary treatable traits such as airflow limitation (SA 50% vs. MMA 6%, P < 0.001), reversibility (SA 58% vs. MMA 39%, P = 0.002), eosinophilia (SA 54% vs. MMA 43%, P = 0.067), exercise-induced asthma (SA 77% vs. MMA 54%, P = 0.002), allergic rhinitis (SA 47% vs. MMA 44%, P = 0.641), cough (SA 63% vs. MMA 19%, P < 0.001) and bronchitis (SA 51% Vs. MMA 16%, P = 0.000) were highly prevalent in the asthma cohorts, and typically more common in SA versus MMA. The most common extra-pulmonary treatable traits were; atopy (SA 74% vs. MMA 92%, P < 0.001), obesity (SA 30% vs. MMA 38%, P = 0.178), reflux (SA 36% vs. MMA 11%, P < 0.001), hypertension (SA 25% vs. MMA 9%, P = 0.001) and obstructive sleep apnea (SA 26% vs. MMA 11%, P = 0.003). Behavioral traits included low medication adherence (SA 39% vs. MMA 52%, P = 0.031) and smoking (SA 8%, smokers not eligible in MMA cohort). Participants with SA had mean (SD) 8 ± (2), and MMA 5 ± (2) treatable traits. Conclusion: We have applied a new approach for the characterisation of severe asthma, based on treatable traits. In general these traits were found more commonly in severe than non-severe asthma; the very high prevalence of many of these traits suggests that there are potential targets for treatment even in such severe patients, and supports the need for specialist management. Keywords: Phenotypes, Treatable Traits

Objective To identify barriers which prevent patients meeting the NICE criteria for mepolizumab. Background Mepolizumab is a steroid sparing agent for patients with severe eosinophilic asthma. All patients must meet the NICE1 criteria: Eosinophil count≥300 cells/microlitre in past 12 months Adherence to optimised therapy Four or more courses of oral corticosteroids (OCS) in past 12 months or continuous OCS equivalent to at least 5 mg of for 6 months Methods Adherence data was obtained from the patient’s GP and eosinophil counts recorded from the hospital electronic patient records. If a patient did not meet the approval criteria the following data was collected: their current OCS dose; number of years under the severe asthma service; last eosinophil count ≥300 cells/ microlitre. Results Of the 269 patients identified as potential mepolizumab candidates 133 have been assessed and 32 (24%) have so far been approved, 38 (28%) have not been approved due to non-adherence to inhaled corticosteroid (ICS) therapy. In this cohort the average ICS adherence was 47% (±SD 0.23) and time under severe asthma specialist care equates to 6.1 years (±3.6). 63 (51%) patients did not have an eosinophil count ≥300 cells/microlitre in the last 12 months. In this group the mean OCS dose is equivalent to 11.5 mg (±7.6) of prednisolone. The mean of the highest historical eosinophil level was 700 cells/ microlitre (±360). Conclusions The following barriers to mepolizumab therapy have been identified: Poor adherence to ICS therapy in patients believed to be on optimal therapy. Long-term OCS treatment is suppressing the eosinophil count below 300 cells/microlitre Firstly, this highlight the need for service improvement to ensure that adherence is regularly monitored. Those who are non-adherent can be referred to the RASP study or for inhaled nitric oxide (FeNO) monitoring. Secondly, patients on long-term OCS may be prevented from benefiting from the steroid sparing effects of mepolizumab. To achieve the necessary eosinophil counts steroids must be progressively reduced risking destabilisation of asthma control. Reference National Institute for Health and Clinical Excellence. Mepolizumab for treating severe refractory eosinophilic asthma. Nice Technology Appraisal Guidance 2017;431.

Introduction Severe Asthma, characterised by persistent symptoms despite maximal medical therapy, represents 5% of asthma cases. Bronchial Thermoplasty (BT) is a novel therapy where radiofrequency thermal energy is applied to airways distal to the main-stem bronchi, permanently reducing airway smooth muscle mass. It is unknown whether treatment of smooth muscle hypertrophy impacts persistently upon systemic signs of allergic inflammation. Peripheral blood eosinophils (PBEs) are a marker of inflammation in asthma. We have previously shown that BT modifies signs of Th2 inflammation as measured by PBEs and therefore wanted to ascertain whether this persisted over time? Method We reviewed the 13 consecutive Severe Asthma cases treated with BT who were included in the initial analysis. Serial PBEs measured up to 3 years post BT were compared to those in the year proceeding BT. Blood eosinophil levels taken peri-procedure were excluded from analysis due to escalated steroid therapy at this time. Results figure 1 demonstrates the absolute values of PBE for each patient during the 3 time periods (year before BT, Year immediately after BT and third year post BT). The group mean in year 3 has remained suppressed compared to baseline (year 3; 0.17 baseline p=0.0035). Conclusion Previous findings showed that Severe Asthma patients undergoing BT had a significant reduction in average peripheral blood eosinophil levels from baseline. The data shows overall the group of 13 patients continue to remain eosinophil supressed supporting the possibility of a long term eosinophil suppressive impact of BT.

The Manchester Severe Asthma Service had planned 37 patients for Bronchial Thermoplasty (BT), under conscious sedation to complete three therapy sessions. Of these 2 (5%) were identified at first session, as part of a routine diagnostic assessment before active therapy is commenced, as having signficant tracheo-bronchial malacia and treatment with BT was not perfomed. One (2.5%) patient had severe bronchospasm, before the onset of active therapy precluding progression to treatment. Five (14.5%) patients experienced issues with pain management during the procedures using conscious sedation, despite routine use of midazolam, alfentanyl and iv Paracetemol. All except one of these patients completed the session at which pain was reported. These 5 patients subsequently completed their therapy under a planned general anaesthetic for their remaining procedures. In total 96/111 (86%) were completed as planned under conscious sedation. Excluding those patients already identified as unsuitable at baseline procedure (no events for 2nd or 3rd treatments), conscious sedation was succesfully used in 96/106 (90.6%), 6 were perfomed (5.7%) under general anaesthetic and 4 (3.8%)procedures were abandoned pre-active treatment or incomplete. We describe the baseline characteristics of all of these patients, in order to consider if any baseline features were predictive of a failure to complete therapy as planned.

Background Omalizumab is an anti-IgE monoclonal antibody therapy used in patients with inadequately controlled persistent allergic IgE mediated asthma who require continuous or frequent treatment with oral corticosteroids. Previous studies have tried to predict a patient’s response to omalizumab based on pre-treatment baseline characteristics. Most recent data has suggested that baseline blood eosinophils, serum periostin or FeNO may be predictive of response to omalizumab in the TH2 phenotype. Aims This study will attempt to identify a characteristic that may explain why some patients suffering with severe asthma in a single severe asthma centre do not achieve a response when treated with the anti-IgE monoclonal antibody, omalizumab. Methods The target population was represented by all patients previously treated or undergoing treatment with omalizumab at the Severe Asthma Service at University Hospital South Manchester (n = 185). The study population was those for whom records could be found within the study period (n = 154). Demographic and clinical data was collected retrospectively from patient medical records. Results 16.2% of patients at UHSM did not show response to omalizumab at 16 weeks. Baseline serum IgE levels in the non-response group were on average 77.28 kU/L lower than those in the response group, statistical analysis of the two groups show that this difference was significant (P = 0.04). Mean eosinophils in the true non-responder group were actually higher than those in the true responder group, however this difference was not statistically significant. No other demographic or disease specific measures predicted a lack of response to omalizumab. Discussion The results from the study indicate that a lower baseline serum IgE may predict non-response to treatment with omalizumab. The results also show that non-response rates at the NWLC were lower than those demonstrated in clinical trials (INNOVATE), were consistent with other real life studies (PERSIST/APEX I and APEX II) but markedly lower than those quoted in the eXpeRience registry.

Background The monoclonal anti-IgE agent Omalizumab holds an established place in the management of severe allergic asthma patients (GINA Step 5). Fungal allergic asthma possesses added complexity as fungi are ubiquitous in our environment and are capable of not only triggering asthma, but may grow, colonise and infect host tissue. Current treatment approaches include: Allergen avoidance, mucus reduction, control of bacterial infection, control of inflammation, reducing fungal burden and recently blockade of allergy using Omalizumab. Aims/purpose Investigate the response to Omalizumab in severe asthma patients who are sensitised to fungal allergens compared to those who are non-fungal allergic. Current literature describes the use of Omalizumab in fungal allergic airways disease, though published data takes the form of case reports/series with limited total population. Methods Retrospective cohort study of severe asthma patients treated with Omalizumab (n = 168). Patients were grouped into fungal or non-fungal allergy status, followed by a comparison of the change in a variety of clinical and physiological outcomes at 16 weeks and 52 weeks from baseline between these two groups. The change in Asthma Quality of Life Questionnaire (AQLQ) between baseline and 16 weeks was utilised as the primary outcome. Groups will be compared using an unpaired t-test or Chi-squared test, as appropriate, to test for non-inferiority (threshold −0.25) in the fungal allergic group compared to the non-fungal allergic group. Results The fungal allergic group (n = 76) was found to have a mean AQLQ difference between baseline and 16 weeks of +1.34 ( ± 1.25). When compared to the non-fungal allergic cohort, the fungal allergic group was found to have a statistically significant mean AQLQ difference between baseline and 16 weeks of −0.41 (95% CI: −0.14 – −0.81). See Figure 1. Conclusion Fungal allergic patients have a less profound AQLQ response to Omalizumab than non-fungal allergic although the benefit is still clinically significant in the majority of cases. The reduced benefit is statistically different in terms of change, though it does not fulfil the a-priori threshold for non-inferiority.

Intro: Omalizumab (Xolair) is licensed for the treatment of severe allergic asthma patients with IgE mediated disease. Dose calculation is governed by total IgE and weight. Stringent controls in prescribing mean patients deemed clinically appropriate to commence treatment are illegible as consequence of IgE levels outside of the dosing table. Our aim was to identify the fluctuation of IgE, to ascertain if a degree of flexibility should be considered when prescribing Omalizumab. Methods: Retrospectively the total serum IgEs of patients whom were referred to potentially start therapy were reviewed. Patients with 2 or more IgEs within a six month period were included. The range of variability of IgE and percentage of variability between the two IgEs were recorded. Results: n=29. IgE 1 median=460 (IQR 155.5-870.0), IgE 2 median=340 (IQR 140.0-885). All IgE9s varied with 14 increasing and 15 decreasing. The average IgE variation was 89 IU/ML (IQR 75-123, median 98). This equated to an average percentage variation of 36.6% (IQR 13.5%-40.7%, median 23.1%). There was no statistically significant difference between IgE 1 and 2 yet, the variation in IgE 2 resulted in an additional 4 of the 29 patients becoming eligible for treatment and the dose could have been altered in 8 of the 19 patients whose IgE 1 was in the dosing range. This would mean that overall 79% of patients were eligible for Omalizumab instead of 66%. Conclusion: Our data collected shows that Serum IgE has significant variability, and therefore frequent testing of IgE in patients whom are potential for Omalizumab therapy and do not meet dosing criteria on initial review should have repeat measures of serum IgE.