The overall objective of the CASCADE program was to allow physicians to assess their perceptions of their own clinical practice and then compare those perceptions with their clinical management of their own patients as well with the perception and clinical practice of their peers from across Canada. As part of the program, a series of meetings between the physicians who participated in the program were held, moderated by the program faculty, to identify barriers and gaps in patient management and to develop strategies to improve patient outcomes. The consensuses from these meetings will be published subsequently.

The physician practices included in the program were not specifically selected using any specific characteristic but through normal interactions with representatives of the sponsor based on an expressed interest in participation; sponsor representatives did not approach all physicians but did invite physicians with a wide variety of practice patterns. The practice profile data that were collected in the program support that the practices represented a diverse variety of clinical practices. Geospatial analysis of a variety of population characteristics from Canadian Census data (data not shown) suggested diverse patient populations for urban–rural ratio, large metropolitan centres versus smaller regional centres, a range of reported incomes and ethnic composition. Physician provided data further supports the diversity of the clinical populations: for example, physicians reported a varied proportion of children, youth or adult patients, with some practices reported a high number of children with severe asthma (up to 40%) and others reported no children (Table 1). Likewise, the number of patients seen in a typical week, practice type (academic versus community) and other data are consistent with a varied selection of physicians.

The program did not collect data on the total number of patients with any condition in the practice, nor did the program ask clinicians to specific any specific clinical interest. However, the program did ask physicians to estimate the proportion of their severe asthma patients who were adult, adolescent, or pediatric. It should be noted that this does not represent the distribution of patients that were included in the patient reviews as data collection was limited to adolescent and adult patients only. Overall, in these practices, an average estimate of 91% of severe asthma patients were adult (over the age of 18 years); however, there was variation between practices with some reporting higher estimates of younger patients. Considering the patients that were actually reviewed for this program, 97.4% where adult (over the age of 18 years), with only 2.6% adolescence patients.

This is the first Canadian assessment of specialist management of patients with severe asthma, particularly focusing on the use of biologic add-on therapies. Physician management of mild to moderate asthma [9] and in moderate to severe asthma [10], has been assessed following the release of GINA recommendations in 2021 [11], however, these were not focused on specialist care for severe asthma and biologic use. Perception of asthma control and management of patients from the Carenity asthma community was assessed in a study including 200 patients from 7 countries. However, this may not fully reflect Canadian patients as few patients from Canada were included [12].

Several key observations arose from this program. It identified that there is a difference in perception of the distribution of asthma phenotypes in specialist practice and the phenotype reported for individual patients, with patients of mixed phenotype representing a majority of patients (53%), while they are perceived to represent a much lower proportion (26%). The reason for this difference is unclear but may reflect either a selection of patients of mixed phenotype for inclusion in the program or an incorrect perception of the composition of their practice. The demographics of the patients included in this program were however similar to recent clinical trials in severe asthma (e.g., ASTHMA QUEST [13], NAVIGATOR [14], and others) suggesting that the included patients are representative of a severe asthma population.

Asthma phenotyping is considered an essential step in management of severe asthma [2], and is included in both the Asthma Canada patient charter [15, 16] and in the recommendations from the Canadian Delphi consensus for severe asthma [17]. However, the identification of phenotypes is complicated by the complexity of the underlying disease immunopathology, with current biomarkers representing an incomplete surrogate measure of underlying processes. Assessment of the biomarker data that was collected during the patient reviews shows that 83% of the patients had at least one biomarker associated with type 2 inflammation (FeNO above 20ppb, blood eosinophils 300 cells/µL or greater, sputum eosinophils 3% or greater, IgE of 30 or greater); when using a lower eosinophils threshold of 150cells/uL, the number of patients with at least one biomarker increases to 89%. This suggests that type 2 inflammation is the predominant mechanism for asthma pathogenesis in severe asthma and a common finding [18]. However, in this program, this may be due to a selection bias of the participating physicians who may have chosen patients on biologics as they are more easily identified but are also more likely to have a type 2 phenotype. Highly variable and mixed biomarkers are a reflect of real-world patient populations when compared with selected clinical trial populations. Physicians may perceive a limited number of biomarker driven phenotypes, but, in reality, the majority of patients will likely have characteristics found in many phenotypes. Analysis of the International Severe Asthma Registry identified that the majority of patients in the registry had multiple elevated biomarkers, with 59% having 2 or more [19]. An issue for effectively phenotyping asthma that has been confirmed in this study is poor access to specialized testing such as sputum analysis and FeNO testing [3, 17]. While FeNO was reported in about one quarter of the patients (24%), sputum eosinophils were infrequently available (6%). The expertise currently required to perform induced sputum analysis prevent its widespread use but point-of-care alternatives are being develop and may assist in future patient assessment [20].

In the Canadian context, there are a number of drivers that may impact specific assessments of asthma severity and phenotype. Typically, FeNO testing is not reimbursed by provincial medical systems and patients may not wish to pay for testing. FEV1 testing is essential, but in certain conditions it may also not be reimbursed (for example, when performed by allergists in Quebec). The underuse of testing for asthma represents a specific concern for ensuring that patients are offered the most effective treatment for their severe asthma.

A high proportion of patients in this study were experienced with biologic (65% currently using a biologic and 5% used a biologic in the past) which is consistent with studies performed in other jurisdictions. In the US, the CHRONICLE study identified that 66% of severe asthma subspecialist-treated US adults where using biologics [21]. However, data for proportion of patients using biologic treatments is scarce, and none could be identified for Canada. The data collected here offers an estimate of biologic use in Canada, but this must be tempered by the realization that the patients reviewed in this program were not rigorously selected and may therefore represent either an over- or under-representation of patients with severe asthma treated with biologics.

While biologic use was similar in Quebec versus the rest of Canada, there were different perceived and reported burdens for using biologics in the different jurisdictions. In Quebec, there was less consideration of cost or access for the use of biologics, but eligibility was a major barrier; in other provinces, the concern over cost and access was greater, although on the individual patient basis, that was not a limiting factor.

This program sought to assess treatment satisfaction and treatment goals, a different perspective from disease control that is assessed in clinical trials. It is clear that, for this group of specialists, there is a specific concern for overuse of systemic corticosteroids and the adverse outcomes associated with them [22]. These specialists, as a group, perceived that limiting OCS use was an important goal for patients with severe asthma, in general. However, when assessing patients on an individual basis, reduction in the OCS dose was infrequently cited, suggesting that this goal was already achieved for most patients entered in the program. While symptom and disease control is a well-established benefit of biologics in clinical trials, in real-world clinical settings, patient response to a specific biologic may be suboptimal and clinicians should be willing to explore other biologic therapies in their patients.

The program represents a point in time assessment of the patients included by the physician. It was clear that while patients using biologics were less likely to have frequent exacerbations (41% vs 57% of patients never using a biologic), the outcomes over time cannot be assessed. However, physicians indicated their intention to start a biologic for about half of the patients that were not currently using a biologic, most likely to address the high frequency of exacerbations. Indeed, 47% of the patients included in the program reported 2 or more exacerbations in the past year, a proportion that increased to 84% in those prescribed a biologic after the visit with the healthcare provider. Similarly, 68% of patients who were reported to be switching to a different biologic had experienced 2 or more exacerbations. It is however unclear if the starting or switching of biologic were triggered by these exacerbations or other factors of the clinical evaluation.

In the patients included in this program, when an ACQ value was available, it was generally indicating poor control. Over three quarters of patients had ACQ values that suggested poor or borderline control of asthma symptoms. However, it is understood, particularly by specialists treating severe asthma, that ACQ assessment may also capture non-asthma respiratory symptoms, leading to higher ACQ scores than those from asthma alone [23]. In some cases, the treating physician may consider that a patient is well controlled by current treatment despite an elevated ACQ value, particularly in those cases where the value is only slightly elevated. Hence, the high proportion of patient with ACQ above 1.5 in the program may be an overestimation of the real proportion of patients with symptomatic asthma.

Symptoms of airway hyperresponsiveness (AHR) were noted in the majority of patients assessed in this program. Furthermore, physicians reported that management of AHR should be considered a general goal of treatment. However, even though AHR is an important characteristic of the disease that contributes to disease pathology, this appears to be under-appreciated or disregarded in the individual patient assessments. This may be due to a relative lack of data regarding the impact of biologic treatment on AHR and a need to further investigate how to target this trait in clinical practice [24].

Patients were not deemed fully satisfied with their treatments, including those that were treated with biologics. This may reflect a biologic use that in poorly tailored to the patient’s disease. Biologic selection is an art based on the available evidence and the reimbursement requirements. However, the levels of individual biomarkers don’t always predict the response to the biologic treatment, and patients with variable levels may have stronger or weaker response [3]. Physicians should be cognizant of this factor and accept that patients with complex diseases may not respond to treatment in expected ways. Compounding this, biologics that have been commercially available for a longer time may be more frequently used, either because patients have been on these therapies for several years and are sufficiently satisfied or because physician are most comfortable with prescribing a drug they are familiar with. At the time of data collection for this program, tezepelumab had been available for prescription for less than one year, accounting for the limited use in this patient population.

While patients were not universally satisfied with their current biologic treatment, it was apparent from the program that patients currently using biologics were deemed much more satisfied with their treatment than patients who were not currently using a biologic (see Fig. 7). This suggests that biologics do improve patient goals. Due to limited sample size, it is unclear why the patients who used biologics in the past were no longer using them, but if those patients were achieving similar satisfaction when using a biologic, it is a wonder that they are not agitating for that treatment again.

In this program, more than 20% of patients reviewed (of any phenotype) where not using biologics, with 70% of patients with type 2 low phenotype not treated with biologics. These patients experienced much poorer satisfaction with treatment than those patients that were treated with biologics. In half of these cases, the major barrier for patients to biologic access was eligibility—likely referring to the specific criteria required for biologic prescription (exacerbations, biomarkers, OCS, etc.). The results from this program clearly show that patients not eligible to current biologics have an unmet need that should be addressed in future biologic development. Biologics effective in several phenotypes or in non-exacerbating patients may provide an avenue for these patients.

The CASCADE program was not comprehensive in the selection of practices to be included. Practice selection could be skewed by bias of both the recruiter and the physicians as not all physicians who were invited to participate in the program chose to do so. Further, the patient selection was not random or sequential and no specific criteria were given to the participating physicians to guide their selection. Selection of patients was rather left to the discretion of the physicians and their personal biases may have intruded, leading to a population that do not fully represent the makeup of patients in the practice. Because of these limitations, the findings of this program should be extended to other practices with care and these results should only be considered qualitatively.

The practice reflective program was carried out in the context of the Canadian single payer health care system and negotiated drug reimbursement criteria by Health Technology Access (HTA) organizations. We believe these approaches to align more closely to European Union models and less to the United States models of care. We hope to see similar programs carried out in other jurisdictions such as the European Union, the United States of America, or Asian Pacific countries.