A prospective observational study to evaluate the prevalence of severe asthma among subjects with difficult-to-treat asthma at a tertiary care center in North India

Abstract

Background:

Severe asthma accounts for disproportionate asthma-related morbidity, mortality, and healthcare costs, yet prospective data on its prevalence remain limited, particularly for developing countries. We aimed to determine the prevalence of severe asthma among subjects with difficult-to-treat asthma at a tertiary care center in North India.

Methods:

Between January and December 2023, we prospectively enrolled consecutive adults with difficult-to-treat asthma that remained uncontrolled despite treatment with medium- or high-dose inhaled corticosteroids-long-acting beta-agonist (ICS-LABA) or required high-dose ICS-LABA to maintain good control. All subjects underwent systematic evaluation, including diagnosis confirmation, assessment of inhaler technique and adherence, comorbidity screening, and treatment optimization, with reassessment after 3–6 months. We diagnosed severe asthma (uncontrolled despite high-dose ICS-LABA or required high-dose ICS-LABA to maintain good control) after addressing all modifiable factors and assessed type 2 (T2)-targeted biological therapy eligibility.

Results:

Among 240 subjects (mean age 47.4 ± 14.9 years; 68.8% female), we identified modifiable factors in 218 (90.8%), including suboptimal therapy (74.2%), poor adherence (64.2%), and incorrect technique (58.8%). Following interventions, 154 completed reassessment; 102 (66.2%) no longer met severe asthma criteria. Overall, 67 subjects had severe asthma (27.9%; 95% CI: 22.4-33.9%). Among them, 66.7% had blood eosinophils ≥150 cells/µL, and 68.7% met criteria for T2-targeted biological therapy.

Conclusion:

Approximately one-quarter of subjects with difficult-to-treat asthma have confirmed severe asthma after systematic assessment. A significant proportion improve with targeted interventions, emphasizing the importance of comprehensive evaluation. Most patients with confirmed severe asthma are candidates for T2-targeted biologicals, highlighting the need for structured assessment protocols and improved access to biological therapies.

INTRODUCTION

Asthma is a common chronic respiratory illness affecting about 2% of the adult population.[1] However, most of the asthma-related morbidity, mortality, and socioeconomic burden is attributable to its most serious form, referred to as severe asthma.[2] Epidemiological studies examining medical records have estimated the prevalence of severe asthma to be approximately 5–10% of all patients with asthma.[3,4,5] However, database-driven studies may be limited by several methodological concerns, including misclassification of asthma diagnoses, failure to systematically evaluate and address modifiable factors contributing to poor control (such as medication non-adherence, incorrect inhaler technique, untreated comorbidities, and persistent environmental exposures), and lack of standardized assessment protocols. Consequently, the true prevalence of severe asthma that does not incorporate rigorous, prospective phenotyping by specialists remains unclear.

The landscape of severe asthma management has been transformed by the advent of targeted biological therapies, particularly those directed against type 2 (T2) inflammatory pathways. Currently available T2-targeted biologicals include anti-IgE therapy (omalizumab), anti-IL-4 therapy (dupilumab), and anti-IL-5 therapies (mepolizumab, reslizumab, benralizumab). More recently, a monoclonal antibody targeting thymic stromal lymphopoietin (TSLP), an upstream epithelial cytokine that orchestrates both T2 and non-T2 inflammatory responses, has demonstrated efficacy across a broader range of asthma phenotypes, including those with low T2 inflammation. Given the high cost and specific eligibility criteria for these biologicals, understanding the proportion of patients with severe asthma who are potential candidates for T2-targeted therapy is essential for healthcare planning, resource allocation, and optimizing patient outcomes.

There is a paucity of prospective data from South Asia, including India, on the prevalence of severe asthma when evaluated using systematic, guideline-based assessment protocols. Such data are crucial for understanding the local disease burden, identifying gaps in care, and informing health policy decisions regarding access to biological therapies. Herein, we systematically evaluated subjects presenting with difficult-to-treat asthma at a tertiary care center in North India. Our primary objective was to determine the prevalence of severe asthma after comprehensive assessment and targeted management of modifiable factors. Our secondary objective was to characterize the T2 inflammatory phenotype and determine the proportion of subjects with severe asthma who met eligibility criteria for T2-targeted biological therapies.

METHODS

We conducted a prospective observational study at the Department of Pulmonary Medicine, a tertiary care referral center in North India, between January 2023 and December 2023. The Institute Ethics Committee approved the study protocol (INT/IEC/2023/SPL-620). We obtained written informed consent from all study participants before enrolment. We report the study according to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.[6]

Study participants

We enrolled consecutive subjects ≥18 years with difficult-to-treat asthma. We excluded subjects who were aged <18 years, were pregnant, had received asthma treatment for <3 months, or declined consent.

Study definitions

We diagnosed asthma based on: (1) recurrent episodes of wheezing, cough, chest tightness, or breathlessness (with no alternative explanation for these symptoms), (2) wheeze on auscultation or expiratory airflow limitation in spirometry, and (3) documented variability in lung function or symptomatic improvement with inhaled corticosteroid (ICS)-based therapy.[7]

We considered asthma uncontrolled when there was poor symptom control (nocturnal awakening due to asthma, any activity limitation due to asthma, or daytime symptoms or reliever use >2/week) or frequent exacerbations (≥2 exacerbations requiring oral corticosteroid [OCS] in the preceding year or ≥1 exacerbation in the preceding year requiring hospitalization).[7]

We diagnosed difficult-to-treat asthma when asthma remained uncontrolled despite treatment with medium-/high-dose ICS-long-acting beta-agonist (LABA), or required high-dose ICS-LABA to maintain good control.[7] We defined a daily dose of inhaled beclomethasone dipropionate ≥500 to <1000 µg and ≥1000 µg (or equivalent) as medium- and high-dose ICS, respectively.[7]

We diagnosed severe asthma when asthma was uncontrolled despite high-dose ICS-LABA or required high-dose ICS-LABA to maintain good control, after systematically ruling out the following: incorrect diagnosis, suboptimal adherence, incorrect inhaler technique, inadequate treatment of comorbidities that could contribute to asthma symptoms, insufficient control of modifiable risk factors, and suboptimal pharmacological treatment.[7]

Study procedures

At enrolment, we recorded demographic data (age, gender), current treatment regimen, exacerbation history in the preceding 12 months, and presence of comorbidities. All ICS doses were converted to beclomethasone dipropionate-equivalent doses. The most commonly used ICS preparations in our setting were budesonide- and fluticasone-based inhalers. The following doses of ICS were considered equivalent: beclomethasone dipropionate (standard particle) 1000 mcg, fluticasone propionate 500 mcg, and budesonide 800 mcg. We defined regular OCS use as the use of OCS for ≥90 days in the preceding year. All subjects underwent spirometry using the Easy On-PC spirometer (NDD Medical Technologies, Andover, MA, USA) according to American Thoracic Society/European Respiratory Society guidelines.[8,9,10] We obtained at least three acceptable spirograms from each subject and interpreted the results using locally derived reference values previously developed at our center.[11]

Evaluation and management of subjects with difficult-to-treat asthma

All subjects with difficult-to-treat asthma underwent structured stepwise evaluation over 3–6 months to identify and address modifiable factors that could contribute to poor asthma control. This protocol included the following components:

1. Confirmation of asthma diagnosis: We verified asthma diagnosis through detailed clinical history and objective testing. We carefully assessed and excluded alternative diagnoses that could mimic asthma, including cardiac dysfunction, inducible laryngeal obstruction, upper airway cough syndrome, gastroesophageal reflux disease (GERD), and others.

2. Assessment of inhaler technique and medication adherence: We evaluated inhaler technique using a standardized checklist,[1] and provided targeted corrective training. For subjects unable to master their current device despite adequate instruction, we prescribed spacers (for metered dose inhalers [MDI] users) or alternate delivery devices (breath-actuated MDI or dry powder inhaler [DPI]). We assessed medication adherence by questioning subjects regarding the frequency of medication use compared to the prescribed regimen, and by estimating the number of MDI canisters or DPI bottles used over the preceding three months.

3. Screening and management of comorbidities: We screened all subjects for allergic bronchopulmonary aspergillosis (ABPA) using established diagnostic criteria.[12,13] We evaluated subjects with hypereosinophilia (blood eosinophil count >1500 cells/µL) for alternative diagnoses, including antineutrophil cytoplasmic antibody-associated vasculitis, chronic eosinophilic pneumonia, hypereosinophilic syndrome, and tropical pulmonary eosinophilia. We also systematically assessed for other comorbidities that could contribute to respiratory symptoms, including GERD, allergic rhinitis, obstructive sleep apnoea (OSA), and chronic rhinosinusitis, and instituted appropriate treatment when identified. Subjects were clinically evaluated for symptoms suggestive of OSA, including snoring, witnessed apnoeas, nocturnal choking, excessive daytime sleepiness, and obesity.[14] Polysomnography was performed when clinically indicated.

4. Identification and control of modifiable exposures: We obtained detailed environmental and occupational histories to identify potential triggers, including tobacco smoke (active or passive), biomass fuel exposure, allergens (dust mites, moulds, pet dander, cockroaches, etc.), occupational sensitizers (flour, grain dust, wood dust, chemicals, metals, etc.), and medications that could aggravate asthma (beta-blockers, non-steroidal anti-inflammatory drugs, etc.). Wherever feasible, we implemented measures to control or eliminate these exposures. General environmental control strategies comprised avoidance of smoking and biomass fuel exposure. Measures to reduce indoor humidity consisted of use of windows and exhaust fans to improve indoor sunlight and ventilation, prompt repair of water leaks, avoidance of indoor drying of clothes, and appropriate use of air-conditioning systems. Additional measures included regular cleaning and vacuuming, avoidance of strong odours and perfumes, and use of indoor air purifiers. In subjects with specific allergen sensitization, we employed targeted allergen avoidance strategies, including the use of dust mite–impermeable mattress and pillow covers, removal of carpets and upholstered furnishings, avoidance of pets, and cockroach control measures. Use of personal protective equipment and occupational modification was also advised wherever appropriate.

5. Optimization of asthma pharmacotherapy: We optimized asthma therapy per the Global Initiative for Asthma (GINA) guidelines.[7] For subjects not already receiving these medications, we added long-acting muscarinic antagonist (LAMA) or leukotriene modifiers as appropriate. Wherever appropriate, we attempted to de-escalate the ICS dose while maintaining asthma control.

Following these interventions, we reassessed subjects after 3 months to determine whether asthma control had improved or whether they met criteria for severe asthma.

Evaluation and management of severe asthma

Subjects who met the definition of severe asthma even after the systematic assessment and optimization described above were classified as having severe asthma. In these patients, we performed additional investigations to characterize asthma phenotype and identify potential candidates for targeted biological therapy. These investigations included: blood eosinophil count, serum total IgE, Aspergillus fumigatus-specific IgE, screening for aeroallergen sensitization, and HRCT of the chest. We defined Aspergillus sensitization as Aspergillus fumigatus-specific IgE >0.35 kUA/L.[15] We screened for aeroallergen sensitization using a commercially available allergy screening test for adults called Phadiatop™.[16] The assay includes a mixture of common airborne allergens, including house dust mites (Dermatophagoides pteronyssinus, Dermatophagoides farinae), grasses (timothy grass, Bermuda grass, rye grass), trees (birch, oak, elm, olive, cedar), weeds (ragweed, mugwort, plantain weeds), animal dander (cat, dog, and horse), moulds (Alternaria alternata, Cladosporium herbarum, Aspergillus, Penicillium), and insects (cockroach). The test was performed on an automated platform (Phadia250, Thermo Fisher Scientific, Uppsala, Sweden). Test results were interpreted qualitatively as positive or negative. The management of severe asthma was individualized and conducted at the discretion of the treating physician, consistent with international guidelines.[7,17] The level of asthma control was assessed using the seven-item Asthma Control Questionnaire (ACQ-7).[18] We categorized subjects with severe asthma as potential candidates for T2-targeted biological therapy if they demonstrated either aeroallergen sensitization with serum total IgE ≥30 IU/mL or blood eosinophil counts ≥150 cells/µL.[19] We also calculated the eligibility for T2-targeted biological therapy at a higher blood eosinophil count of ≥300 cells/µL.

Sample size calculation

Assuming a prevalence of severe asthma of 25% among individuals with difficult-to-treat asthma,[3,17] and targeting an absolute precision of 6.25% (one-fourth of the assumed prevalence) with 5% type 1 error rate and an anticipated 30% loss to follow-up, we calculated a required sample size of 239 participants.

Statistical analysis

We performed all statistical analyses using Statistical Package for the Social Sciences version 22.0 (IBM Corp., Armonk, NY, USA). We present continuous variables as mean ± standard deviation (SD) and categorical variables as frequencies and percentages.

RESULTS

Baseline characteristics

We included 240 subjects with difficult-to-treat asthma during the study period (mean age 47.4 ± 14.9 years; 165 women [68.8%]) [Table 1]. Allergic rhinitis (49.2%), GERD (28.3%), and ABPA (12.9%) were the most common comorbidities. All subjects were receiving ICS-LABA therapy with a mean ICS dose of 1185.4 ± 409.0 µg beclomethasone dipropionate-equivalent per day. Other controller medications included LAMA (22.9%), leukotriene receptor antagonist (20.4%), methylxanthines (n = 22, 9.2%), and regular OCS (n = 15, 6.3%). Despite treatment, asthma remained poorly controlled in most subjects. Only 40 subjects (16.7%) had well-controlled asthma according to GINA criteria, while 91 (37.9%) had partly controlled asthma and 109 (45.4%) had uncontrolled asthma. In the preceding year, 111 subjects (46.3%) had experienced at least one severe exacerbation requiring systemic corticosteroids or hospitalization. Spirometry was available for 195 subjects (81.3%). The mean pre-bronchodilator FEV₁ was 68.3 ± 23.4% predicted, and mean FVC was 81.9 ± 18.1% predicted. Obstructive spirometry pattern was observed in 156 subjects (80.0%), restrictive pattern in 7 (3.6%), and normal spirometry in 32 (16.4%). A positive bronchodilator response was documented in 49 of 187 subjects (26.2%) in whom post-bronchodilator spirometry was performed.

Table 1.

Baseline characteristics of the subjects with difficult-to-treat asthma (n=240)

Characteristic	Value
Demographics
Age (years)	47.4±14.9
Female sex	165 (68.8)
Comorbidities
Allergic rhinitis	118 (49.2)
GERD	68 (28.3)
ABPA	25/194 (12.9)*
Obesity	12 (5.0)
Cardiac diseases	7 (2.9)
Psychiatric illness	7 (2.9)
OSA	1 (0.4)
Current asthma treatment
ICS	240 (100.0)
ICS dose in BDP-equivalents (µg)	1185.4±409.0
ICS-LABA combination	240 (100)
LAMA	55 (22.9)
Leukotriene receptor antagonist	49 (20.4)
Methylxanthines	22 (9.2)
Regular OCS therapy	15 (6.3)
Asthma control (preceding 4 weeks)
Day time symptoms >2/week	125 (52.1)
Reliever medication use >2/week	105 (43.8)
Nocturnal awakening due to asthma	92 (38.3)
Activity limitation due to asthma	180 (75.0)
Overall asthma control
Well controlled	40 (16.7)
Partly controlled	91 (37.9)
Uncontrolled	109 (45.4)
Exacerbation history (preceding 12 months)
≥1 severe exacerbation (requiring hospitalization or OCS)	111 (46.3)
≥1 exacerbation requiring hospitalization	62 (25.8)
≥1 exacerbation requiring OCS	77 (32.1)
≥2 exacerbations requiring OCS	41 (17.1)
Spirometry
Spirometry available	195 (81.3)
Spirometry pattern
Obstructive	156/195 (80)
Restrictive	7/195 (3.6)
Normal	32/195 (16.4)
Pre-bronchodilator FEV1 (litres)	1.56±0.62
Pre-bronchodilator FEV1 (% predicted)	68.3±23.4
Pre-bronchodilator FVC (litres)	2.39±0.75
Pre-bronchodilator FVC (% predicted)	81.9±18.1
FEV1/FVC ratio	0.63±0.15
Positive bronchodilator response	49/187 (26.2)

All values are provided as number (percentage) or mean±standard deviation. BDP: beclomethasone dipropionate, FEV1: forced expiratory volume in the first second, FVC: forced vital capacity, ICS: inhaled corticosteroid, LABA: long-acting beta agonist, LAMA: long-acting muscarinic antagonist, OCS: oral corticosteroid. *Evaluation for ABPA could be completed in only 194 subjects

Identification and classification of modifiable factors

We identified at least one modifiable factor contributing to poor asthma control or requirement for high-intensity asthma therapy in 218 of the 240 subjects (90.8%) [Figure 1]. Only 15 subjects (6.3%) had no identifiable modifiable factors and were classified directly as having severe asthma. Seven subjects (2.9%) were found to have an incorrect diagnosis of asthma on detailed evaluation. Among the 218 subjects with modifiable factors, the most frequently identified issues were suboptimal pharmacotherapy (n = 178, 74.2%), inadequate medication adherence (n = 154, 64.2%), and incorrect inhaler technique (n = 141, 58.8%). Other modifiable factors included inadequately managed comorbidities (n = 95, 39.6%) and modifiable environmental or occupational exposures (n = 7, 2.9%).

Figure 1.

Flow diagram depicting evaluation and follow-up of subjects with difficult-to-treat asthma. *Numbers do not add up as subjects may have had more than one modifiable factor

Outcomes following targeted interventions

Among the 218 subjects with modifiable factors, 205 (94.0%) underwent intervention, while 13 subjects were lost to follow-up before intervention. Many subjects had multiple concurrent modifiable factors and therefore received more than one intervention during follow-up. Among the 205 subjects who underwent intervention, 154 (75.1%) completed reassessment after 3-6 months, whereas 51 (24.9%) were lost to follow-up before reassessment [Figure 1]. Among the 154 subjects who completed reassessment, 102 (66.2%) no longer met criteria for severe asthma after interventions. Specifically, 66 subjects (42.9%) demonstrated significant improvement in asthma control following the targeted interventions, and 36 (23.4%) had their respiratory symptoms primarily attributed to non-asthma comorbidities after appropriate treatment of those conditions.

Prevalence of severe asthma among subjects with difficult-to-treat asthma

Overall, 67 subjects had severe asthma. Fifteen subjects fulfilled severe asthma criteria at baseline, while 52 additional subjects fulfilled criteria after reassessment following systematic evaluation and intervention. Thus, the overall prevalence of severe asthma in our cohort was 27.9% (67/240; 95% CI: 22.4-33.9%). Because 64 subjects did not complete the full assessment pathway (13 lost before intervention and 51 lost after intervention), we performed sensitivity analyses for attrition [Figure 1, Table 2]. Under the best-case assumption that all subjects lost to follow-up improved after intervention, the prevalence of severe asthma remained 27.9%. Under the worst-case assumption that all subjects lost to follow-up had persistent severe asthma, the prevalence increased to 54.6%. Assuming outcomes among those lost to follow-up were similar to reassessed subjects, the estimated prevalence of severe asthma was 37.1%.

Table 2.

Sensitivity analysis for the primary objective accounting for the 64 subjects lost to follow-up

Scenario		Assumption		Prevalence of severe asthma, n/N (%)
Primary analysis		Confirmed severe asthma in full cohort		67/240 (27.9)
Best-case		All lost subjects improved and did not have severe asthma		67/240 (27.9)
Missing-as-observed*		Lost subjects had same severe asthma rate as reassessed subjects with modifiable factors		89/240 (37.1)
Worst-case		All lost subjects had severe asthma		131/240 (54.6)

*Missing-as-observed assumes that subjects lost to follow-up despite having at least one modifiable factor had the same severe asthma rate as subjects with modifiable factors who completed reassessment (52/154=33.8%), yielding approximately 22 additional severe asthma cases

Characteristics and comorbidities in severe asthma

Among the 67 subjects with severe asthma, detailed phenotypic characterization was performed [Table 3]. The most prevalent comorbidities were allergic rhinitis (n = 31, 46.3%), GERD (n = 22, 32.8%), obesity (n = 9, 13.4%), bronchiectasis (n = 5, 7.5%), and cardiac diseases (n = 5, 7.5%). ABPA was diagnosed in 4 subjects (6.0%), chronic rhinosinusitis in 4 (6.0%), and nasal polyposis in 3 (4.5%). Psychological comorbidities, including anxiety (n = 3, 4.5%) and depression (n = 1, 1.5%), were also identified.

Table 3.

Characteristics of the subjects with severe asthma (n=67)

Characteristic		Value
Comorbidities
Allergic rhinitis		31 (46.3)
GERD		22 (32.8)
Obesity		9 (13.4)
Bronchiectasis		5 (7.5)
Cardiac diseases		5 (7.5)
ABPA		4 (6)
Chronic rhinosinusitis		4 (6)
Nasal polyposis		3 (4.5)
Anxiety		3 (4.5)
Chronic obstructive pulmonary disease		1 (1.5)
Depression		1 (1.5)
Asthma control
ACQ-7 score		1.8±0.9
Lab investigations
Blood eosinophils (cells/µL)		344.18±334.87
Blood eosinophils ≥150 cells/µL		40/60 (66.7)
Blood eosinophils ≥300 cells/µL		25/60 (41.7)
Aspergillus fumigatus specific IgE (kUA/L)		3.42±14.57
Aspergillus sensitization		10 (14.9)
Serum total IgE (IU/mL)		682.61±1088.52
Sensitization to aeroallergen		20/41 (48.8)
Spirometry
Spirometry available		64 (95.5)
Spirometry pattern
Obstructive		55/64 (85.9)
Restrictive		2/64 (3.1)
Normal		7/64 (10.9)
Pre-bronchodilator FEV1 (litres)		1.35±0.56
Pre-bronchodilator FEV1 (% predicted)		62.86±26.13
Pre-bronchodilator FVC (litres)		2.26±0.60
Pre-bronchodilator FVC (% predicted)		79.56±20.21
FEV1/FVC ratio		0.58±0.15
Positive bronchodilator response		15/63 (23.8)
High-resolution chest computed tomography
Mosaic attenuation		9/54 (16.6)
Centrilobular nodules		8/54 (14.8)
Bronchiectasis		6/54 (11.1)
Emphysema		4/54 (7.4)
Eligibility for type 2 inflammation-targeted biological therapy
Anti-IL5, anti-IL4, or anti-IgE (aeroallergen sensitization with serum total IgE ≥30 IU/mL or blood eosinophils ≥150 cells/µL)		46/67 (68.7)
Anti-IL5, anti-IL4, or anti-IgE (aeroallergen sensitization with serum total IgE ≥30 IU/mL or blood eosinophils ≥300 cells/µL)		33/67 (49.3)
Anti-IL5 or anti-IL4 (blood eosinophils ≥150 cells/µL)		40/60 (66.7)
Anti-IL5 or anti-IL4 (blood eosinophils ≥300 cells/µL)		25/60 (41.7)
Anti-IgE (aeroallergen sensitization with serum total IgE ≥30 IU/mL)		18/41 (43.9)

All values are provided as number (percentage) or mean±standard deviation. ABPA: Allergic bronchopulmonary aspergillosis, ACQ-7: Asthma Control Questionnaire-7, FEV1: forced expiratory volume in the first second, FVC: forced vital capacity, GERD: gastroesophageal reflux disease

Type 2 inflammatory biomarkers and radiological findings

Blood eosinophil counts were available for 60 of 67 subjects with severe asthma (89.6%). The mean blood eosinophil count was 344.2 ± 334.9 cells/µL, with 40/60 subjects (66.7%) having eosinophil counts ≥150 cells/µL and 25/60 (41.7%) having counts ≥300 cells/µL. Serum total IgE was measured in all 67 subjects, with a mean level of 682.6 ± 1088.5 IU/mL. Aspergillus fumigatus-specific IgE was elevated (>0.35 kUA/L) in 10 subjects (14.9%), indicating Aspergillus sensitization. Aeroallergen sensitization testing using Phadiatop™ was performed in 41 subjects (61.2%), with 20 (48.8%) demonstrating sensitization to common aeroallergens.

HRCT chest was performed in 54 of 67 subjects (80.6%) with severe asthma. The most common radiological abnormalities were mosaic attenuation (n = 9, 16.7%), centrilobular nodules (n = 8, 14.8%), bronchiectasis (n = 6, 11.1%), and emphysema (n = 4, 7.4%). Some subjects had multiple radiological findings.

Eligibility for type 2-targeted biological therapy

Overall, 46 of 67 subjects (68.7%) met eligibility criteria for at least one biological agent [Table 3]. When the eosinophil threshold was increased to ≥300 cells/µL, 33 of 67 subjects (49.3%) remained eligible for at least one biological therapy. Based on blood eosinophil criteria alone, 66.7% and 41.7% of subjects qualified for anti-IL-5 or anti-IL-4 receptor-α targeted therapy using thresholds of ≥150 cells/µL and ≥300 cells/µL, respectively. Among the 41 subjects who underwent both serum total IgE measurement and aeroallergen testing, 18 (43.9%) demonstrated aeroallergen sensitization with total IgE ≥30 IU/mL, thereby fulfilling eligibility criteria for anti-IgE therapy with omalizumab.

DISCUSSION

In this prospective observational study, we systematically evaluated 240 subjects with difficult-to-treat asthma using a comprehensive, guideline-based assessment protocol. After targeted interventions to address modifiable factors, we identified severe asthma in 67 subjects, representing 27.9% of the entire cohort. Importantly, among the 154 subjects who completed follow-up after intervention, 102 (66.2%) demonstrated improvement in asthma control or had symptoms attributable to non-asthma conditions, highlighting the substantial impact of systematic evaluation and management of modifiable factors. Furthermore, 68.7% of subjects with confirmed severe asthma were potential candidates for T2-targeted biological therapy, underscoring the importance of phenotypic characterization in this population.

The prevalence of severe asthma observed in our study (27.9% of all subjects with difficult-to-treat asthma) is substantially higher than the 5-10% prevalence reported in population-based registries.[3,4,5] This difference is expected and reflects referral bias, as our tertiary care center receives patients with more refractory disease from a wide catchment area. More critically, our study demonstrates that nearly half of subjects with initially difficult-to-treat asthma showed significant improvement after systematic identification and management of modifiable factors.[20] We identified at least one modifiable factor in 90.8% of subjects, with the most common being suboptimal pharmacotherapy, inadequate medication adherence, and incorrect inhaler technique. Further, a significant proportion achieved improved control through targeted interventions, and almost one-fourth had symptoms primarily attributable to comorbidities rather than asthma. Because many subjects had multiple concurrent modifiable factors, the relative contribution of individual interventions to subsequent improvement in clinical status could not be determined, and the observed benefits were likely multifactorial. Our results underscore a crucial clinical message: the label of severe asthma should only be applied after comprehensive evaluation by specialists, confirmation of the asthma diagnosis, optimization of pharmacotherapy, assessment and correction of inhaler technique and adherence, identification and treatment of comorbidities, and control of environmental exposures. Misclassification of difficult-to-treat asthma as severe asthma has significant implications, potentially subjecting patients to unnecessary expensive biological therapies, while missing opportunities for simpler, more effective interventions. Therefore, all asthma clinics managing patients with poor asthma control should incorporate systematic assessment protocols like ours before classifying patients as having severe asthma.

A substantial proportion of patients with severe asthma (68.7%) met criteria for at least one T2-targeted biological therapy based on current guidelines.[21] Specifically, 66.7% and 41.7% had blood eosinophil counts ≥150 cells/µL and ≥300 cells/µL, respectively, making them eligible for anti-IL5 (mepolizumab, benralizumab, reslizumab) or anti-IL4 receptor-α (dupilumab) therapy. Among subjects with complete testing, 43.9% met criteria for anti-IgE therapy (omalizumab) based on elevated total IgE (≥30 IU/mL) and documented aeroallergen sensitization. The higher eligibility for eosinophil-based therapies compared to omalizumab reflects the broader applicability of blood eosinophil count as a biomarker and the more restrictive criteria for anti-IgE therapy, which requires both elevated IgE and confirmed allergen sensitization.[19] Notably, about one-third of our severe asthma subjects did not meet criteria for any T2-targeted biological therapy based on the biomarkers we assessed. These patients may represent non-T2 or T2-low phenotypes, which historically had limited treatment options. The recent development of tezepelumab, which targets TSLP, has shown efficacy across a broader range of asthma phenotypes, including those with low eosinophil counts.

One notable observation was the low prevalence of ABPA (6%) in our severe asthma cohort. Previous studies have reported a prevalence of approximately 13% among asthmatics attending specialized clinics, with potentially even higher prevalence in severe asthma populations.[22] Even in our study, the prevalence of ABPA among subjects with difficult-to-treat asthma was 12.9%. However, asthma control improved in many of these subjects after initiation of appropriate treatment for ABPA, and they therefore no longer met criteria for severe asthma during follow-up. Thus, the apparently low prevalence of ABPA observed in our severe asthma cohort may reflect our institutional practice of routinely screening all asthma patients for ABPA regardless of asthma severity. This observation may not be generalizable to centres with different referral patterns and screening practices.

Our study has some limitations. First, this was a single-center study conducted at a tertiary referral center and is therefore subject to substantial referral bias. Accordingly, the prevalence estimate reported in this study should not be extrapolated to the general asthma population, particularly to patients managed in primary-care or community-based settings, where the prevalence of severe asthma is likely to be substantially lower. Second, despite our extensive efforts to ensure completeness of follow-up, 26.7% of subjects were lost to follow-up during the assessment process. Nevertheless, because our initial sample size calculation accounted for 30% attrition, the study remained adequately powered for the primary objective. However, loss to follow-up may have introduced selection bias if subjects who were lost differed systematically from those who completed reassessment. Although we could not definitively determine outcomes in these subjects, sensitivity analyses suggested that the prevalence of severe asthma remained substantial even under conservative assumptions, supporting the robustness of our overall findings. Third, we assessed adherence using self-report and counting of dispensing units rather than more objective measures like electronic monitoring devices which could have overestimated adherence. Fourth, we did not measure fractional exhaled nitric oxide (FeNO) or sputum eosinophils, which are additional biomarkers used to identify T2 inflammation and guide biological therapy selection. Fifth, subjects were evaluated clinically for symptoms suggestive of OSA and a formal validated screening questionnaire was not systematically used in all subjects. Consequently, OSA may have been under-recognized in our cohort. Finally, our study did not include long-term follow-up to assess the stability of severe asthma classification or their phenotypes over time or the real-world effectiveness of biological therapies in our population. The strengths of our study include its prospective design with systematic, protocol-driven evaluation by pulmonary specialists that minimizes misclassification bias inherent in retrospective studies.

In conclusion, we found that severe asthma was present in approximately one-quarter of patients presenting with difficult-to-treat asthma at our tertiary care center. Importantly, nearly half of patients with initially difficult-to-treat asthma demonstrated improvement after systematic evaluation and targeted management of modifiable factors, emphasizing the critical importance of comprehensive assessment before labelling patients as having severe asthma. Most patients with confirmed severe asthma are potential candidates for T2-targeted biological therapies, highlighting the need for improved access to these treatments in India. Given the limitations of a single-center tertiary-care cohort, the prevalence estimate requires confirmation in multicenter studies with broader geographic representation to obtain nationally relevant estimates. Nevertheless, the observations from our study underscore the importance of implementing structured assessment protocols in asthma clinics to optimize patient outcomes and healthcare resource utilization.

Authors’ contributions

SS: involved in patient management, data analysis, initial drafting, and correction of the manuscript, RA: involved in conception of study, patient management, and review of the manuscript, VM: involved in patient management and review of the manuscript, SMR: involved in patient management and review of the manuscript, YK: involved in patient management and review of the manuscript, NP: involved in patient management and review of the manuscript, IPS: involved in patient management and review of the manuscript, SD: involved in patient management and review of the manuscript, ANA: involved in patient management and review of the manuscript, KTP: conception of the study, planning, execution, patient management, and final preparation of the manuscript.

Ethical approval statement

The Institutional Ethics Committee of the Postgraduate Institute of Medical Education and Research approved the study protocol (No. INT/IEC/2023/SPL-620).

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.