Real word evidence studies: Is it the way forward?

Abstract

Seasonal allergic rhinitis (SAR) is a type I hypersensitivity-mediated chronic inflammatory disease of the nasal mucosa resulting from exposure to prevalent common allergens such as grass, tree and weed pollen during the summer months. Typical symptoms include sneezing, rhinorrhoea, nasal congestion, and nasal itching.1, 2 SAR is often associated with ocular (eye) symptoms of allergic conjunctivitis. Pharmacotherapy strategies include antihistamines and intranasal corticosteroids which only provide temporary symptomatic relief and fail to control symptoms in 30%–60% of patients. In these patients, allergen immunotherapy (AIT) is indicated. AIT can be administered by either the subcutaneous or the sublingual route. AIT is associated with the reduction of seasonal symptoms and the need for rescue medication. It is the only disease-modifying treatment when given for a period of three years or more.3 Patients with severe allergic rhinitis treated with AIT have improved quality of life.4 The clinical benefit of AIT depends on the accurate identification of the symptom-triggering allergen. However, this is particularly challenging in patients with several co-sensitisations as the cross-reactivity in pollens can elicit symptoms with overlapping pollen seasons, masking the identity of the main allergen. Component resolved diagnostics (CRD) can help identify the allergens eliciting the symptoms.5 Algorithms on the molecular diagnosis of allergies have been published but remain infrequently used. Moreover, Mobile Health (mHealth) technology can also be used as part of a clinical decision support systems (CDSS) assisting patients, clinicians, and pharmacists at the point of care.6 In this issue of Clinical & Experimental Allergy, Arasi et al.1 have evaluated the utility and impact of a recently established algorithm (@IT2020) for a clinical decision support system (CDSS) for the diagnostic workup of seasonal allergic rhinitis using skin prick test or serum IgE, CRD, and real-time digital symptom recording (eDiary) on physician's AIT prescription decisions. The study interestingly found that the combined use of CRD and an eDiary increased the hypothetical AIT prescriptions, both among 18 allergy specialists and 28 general practitioners. Moreover, relying on amnesis and skin prick test or serum IgE readout, AIT prescription for pollen and alternaria allergy was found to be heterogeneous but converged towards a consensus when integrating CRD and eDiary information. Physicians considered the algorithm useful and recognized its potential in enhancing traditional diagnostics. The implementation of CRD and eDiary in the CDSS algorithm improved consensus on AIT prescription for seasonal allergic rhinitis among allergy specialists and General Practitioner (Figure 1). These findings warrant further evaluation of the potential utility of a CDSS for aetiological diagnosis of SAR and AIT prescription in clinical practice. The importance of CRD in allergy diagnosis and AIT has slowly been unravelled, with the concept being accepted by World Allergy Organisation (WAO). Studies to investigate diagnostic approaches for component testing have therefore been of huge interest.7 In a systematic review published within the current issue by Maesa et al.,8 the diagnostic accuracy of component testing within the ImmunoCAP ISAC panel was thoroughly investigated. The review particularly investigated their reliability in food allergy diagnosis of component-specific IgE immunoassays and whether they are comparable to the current gold standard of oral provocation tests, open food challenge (OFC) or double-blind placebo-controlled food challenge. Whilst the current gold standard approach is essential in the diagnosis of food allergy, OFC is time-consuming and has been associated with severe reactions such as anaphylaxis. The systematic review identified contradicting published evidence towards four components (Gald1 for egg allergy, Bosd5 for milk allergy, and Ara h1 or h2 for peanut allergy) where the risk of bias in the studies resulted in a reduction of quality of the results. Peanut components (Ara h1 and Ara h2) were analysed in a larger number of studies and high variability in the diagnostic performance was found with Ara h1 but better performance for Ara h2. The results compiled from various studies highlighted that diagnostic test accuracy information for ISAC components is highly specific but not very sensitive, in contrast to other diagnostics for food allergy such as skin prick test and specific IgE to whole allergens, where sensitivity tends to be high but false positive readings are very common. With this thorough systematic review, we can gather that there is a lack of evidence for most components and the quality of evidence analysed is insufficient to confirm that ImmunoCAP ISAC can be used as a diagnostic test in place of the gold standard approach Component testing may, however, provide useful additional information above standard allergy tests, taking into consideration the lower sensitivity and higher specificity than other forms of allergy testing. Further studies on diagnostic tests are warranted to obtain higher quality information and better define the clinical utility of ImmunoCAP ISAC for food allergy diagnosis. The use of novel algorithms such as @IT2020 to support AIT prescription decisions has highlighted their potential importance in making a difference towards real-world clinical practice. In another study relying on real-world evidence, Azim et al. focused on the relevance of using electronic healthcare records of blood eosinophil count in a 10-year retrospective period involving patients in an asthma cohort of difficult asthma (WATCH) study.9 Measurement of airway eosinophilia in induced sputum is well regarded as a predictive biomarker for asthma exacerbation but can be challenging for routine clinical practice due to the limitation of undertaking sputum in a clinical setting.10 Instead, blood eosinophils which are much more easily accessible in clinical practice are considered as a biomarker that can be used in place of airway eosinophils, though studies have revealed these two factors are not always associated. The study aimed to interrogate the phenotypic characterisation of patients with difficult asthma and their suitability in receiving biological therapies. The study found that within the study population, a large proportion of 40.3% was eosinophilic at the study enrolment and that an additional 43.1% demonstrated eosinophilia at least once in the preceding decade, despite not displaying this phenotype during the enrolment period. These phenotypic findings are extremely important as variability in eosinophilia has been associated with poor asthma control and worse lung function. Patients who displayed eosinophilic phenotype were further stratified into four groups based on the frequency of the eosinophilia and a close association was found between the frequency of eosinophilia and lung function (Figure 2). Finally, the authors also identified that 16.6% of patients who do not display any eosinophilia showed preserved lung function and lower markers of type 2 inflammation. The findings from this study highlight the role that type 2 inflammation plays in difficult asthma and underscores the importance of longitudinal electronic healthcare record analysis to better phenotype clinical asthma. The application of this electronic healthcare record may allow a better understanding of disease progression and a more focused treatment approach. These articles within the current issue have provided substantial information to highlight the importance of and verify the use of novel approaches for component resolved diagnosis to assist allergy diagnosis and allergen immunotherapy prescription. Whilst further studies are undoubtedly required, these articles are highly relevant especially as they involve real-world evidence with direct translatability towards clinical settings in the context of allergy, asthma and AIT.