Challenges in Evaluating Clinical Remission in Severe Asthma: Insights from the Mayo Clinic

Abstract

Introduction

Clinical remission (CR) is an ambitious and achievable treatment goal for many patients with severe asthma. This study evaluated real-life care of patients in the U.S. using CR criteria defined by the American Thoracic Society; American College of Allergy, Asthma, and Immunology; and American Academy of Allergy, Asthma & Immunology.

Methods

This retrospective cohort study (GSK ID: 219744) utilized data from the Mayo Clinic’s electronic health record database (January 1, 2014–March 31, 2023). Eligible adults had severe asthma, ≥ 1 respiratory biologic initiated, and ≥ 12 months of clinical activity post-index date. The primary objective quantified the proportion of patients with documented CR component criteria 12-months post-biologic initiation. Criteria included asthma exacerbations, systemic corticosteroid use for asthma, missed work/school due to asthma, ≥ 2 pulmonary function tests, controller medication use for asthma, ≥ 2 asthma control tests, and rescue medication use for asthma.

Results

Of 4623 patients receiving respiratory biologics, 707 were eligible. Documentation was available for ≥ 1 component in 94.2% of patients; none had all criteria documented. Overall, 90.2%, 83.2%, 55.4%, and 33.0% of patients had documented controller medication use, rescue medication use, systemic corticosteroid use, and asthma exacerbations, respectively. For patients with documentation, 91.2% achieved ≥ 1 criterion. However, the proportion achieving remission decreased with the number of components; 0.6% of patients achieved ≥ 5 criteria. Of 141 (19.9%) patients receiving mepolizumab, documentation was available for ≥ 1 component in all patients; none had all criteria documented. The proportion of patients with documentation, and who achieved ≥ 1 to ≥ 4 criteria, was higher versus the overall population.

Conclusion

This study demonstrated infrequent documentation of the CR components in routine practice, thereby limiting the comprehensive evaluation of CR. Standardized assessment protocols encompassing all domains are needed to enable accurate assessment of CR, and for treatment targets to provide clear goals for clinicians and patients.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12325-025-03433-y.

Key Summary Points

Why carry out this study?

Severe asthma often remains uncontrolled or requires the use of biologics to achieve disease control; clinical remission (CR) has also become possible with such therapies.

This study evaluated the documentation and achievement of CR components in real-life care of patients in the U.S. using CR criteria defined by the American Thoracic Society, American College of Allergy, Asthma, and Immunology, and American Academy of Allergy, Asthma & Immunology.

What was learned from this study?

Data from the Mayo Clinic’s electronic health record database demonstrated that, while 94.2% of patients receiving mepolizumab had documentation of ≥1 CR component, none had the required data available for all 7 criteria. Nearly half of patients receiving mepolizumab achieved ≥3 CR components.

Overall, infrequent documentation of the different CR components was observed in routine practice; standardized assessment protocols encompassing all domains would aid a more comprehensive evaluation of CR and enable a more accurate assessment of asthma remission.

Introduction

Severe asthma is a chronic, inflammatory condition, characterized by persistent symptoms and frequent exacerbations despite high doses of controller medications [1, 2]. Despite optimized treatment with high-dose inhaled corticosteroid (ICS)/long-acting β₂-agonist (LABA), severe asthma often remains uncontrolled or requires use of biologics to achieve control [2]. Clinical remission, an ambitious and achievable treatment goal for many patients with severe asthma, has become possible with precision therapies such as biologics that target specific inflammatory pathways; early targeted treatment may aid in achieving clinical remission [3–6]. On-treatment clinical remission aligns with the Global Initiative for Asthma’s goal of optimal long-term asthma outcomes [7], which includes symptom control, maintained productivity, and minimization of long-term asthma risk; these goals also include no systemic corticosteroid (SCS) courses, an absence of exacerbations, and stable lung function outcomes [2]. Using clinical remission as a treat-to-target goal may also be a useful approach for simplifying complex medication sequences, facilitate shared decision-making, and provide the clinician with explicitly specified and sequentially measured goals [8].

There is currently no global consensus on the definition of asthma remission. Various European and Asian guidelines generally agree on no or minimal symptoms, absence of exacerbations, and no SCS use for asthma for at least 12 months [4, 9]. However, definitions continue to evolve, with consensus guidance from the American Thoracic Society (ATS), American College of Allergy, Asthma and Immunology (ACAAI), and American Academy of Allergy, Asthma and Immunology (AAAAI) also including stable lung function (see Table 1 for a summary of these criteria, which also proposed the inclusion of no missed work and/or school due to asthma, and the ability to maintain disease control with continued use of low-to-medium dose controller medication) [10]. However, availability of real-world data on applicability in clinical practice is currently limited [10].

Table 1.

Clinical remission criteria from the ATS/ACAAI/AAAAI [10] and populations for the current data analyses

ATS/ACAAI/AAAAI criteria for on-treatment asthma clinical remission [10]	Populations for data extraction and analyses
No exacerbations requiring a physician visit, emergency care, hospitalization, and/or no SCS for asthma (oral or injectable)	SCS use Number and proportion of patients with ≥ 1 SCS order/administration during the 12-month follow-up period (extraction from structured data) Asthma exacerbations Exacerbations were considered a single event if ≥ 2 were observed within 14 days of each other. Exacerbations at index date were defined as follows:  Inpatient and emergency exacerbation:   Asthma-related hospitalization or ED visit that resulted in hospitalization within + 1 day   Asthma-related ED visit  SCS exacerbation:   Asthma-related ED visit or outpatient visit with an SCS administration/order at Mayo within ± 5 days, in addition to explicit mention in the patient’s relevant clinical note
No missed work or school over 12 months due to asthma-related symptoms	Missed work/school due to asthma Number and proportion of patients with asthma-related symptoms assessed during the 12-month follow-up period (extraction from unstructured clinical notes)  Symptoms included wheezing, shortness of breath, chest tightness, and cough
Stable and optimized lung function over 12 months, with a minimum of two measurements during the year	Stable PFTs Number and proportion of patients with ≥ 2 stable PFT measurements (FEV1, FVC and FEV1/FVC) > 14 days apart during the 12-month follow-up period (extraction from structured data)
Continued use of controller therapies (ICS, ICS/LABA, LTRA) only at low or medium ICS dose (or lower), as defined by the most recent GINA guidelines	Controller medication use Number and proportion of patients with ≥ 1 ICS, ICS/LABA and/or LTRA order/administration at low or medium dose during the 12-month follow-up period initiation (extraction from structured data)
An ACT > 20, AirQ < 2, ACQ < 0.75 on all occasions measured over 12 months, with a minimum of two measurements during the year	Asthma control Number and proportion of patients with ≥ 2 ACT > 20, AirQ < 2, or ACQ < 0.75 measurements assessed over the 12-month follow-up period (extraction from structured data)
Asthma-related symptoms requiring one-time rescue therapy (SABA, SABA/SAMA, ICS/LABA) no more than once a month	Rescue medication use Number and proportion of patients with ≥ 1 SABA and/or SABA/SAMA order during the 12-month follow-up period (extraction from structured data)  ICS/SABA regimens were not found in the study population and were excluded from the analyses

AAAAI American Academy of Allergy, Asthma & Immunology, ACAAI American College of Allergy, Asthma and Immunology, ACQ Asthma Control Questionnaire, ACT Asthma control test, AirQ Asthma Impairment and Risk Questionnaire, ATS American Thoracic Society, ED emergency department, FEV₁ forced expiratory volume in 1 s, FVC forced vital capacity, GINA Global Initiative for Asthma, ICS inhaled corticosteroid, LABA long-acting β₂-agonist, LTRA leukotriene receptor antagonist, PFT pulmonary function test, SABA short-acting β₂-agonist, SAMA short-acting muscarinic antagonist, SCS systemic corticosteroid

Several biologics have been approved by the U.S. Food and Drug Administration for asthma treatment, including omalizumab (2003), mepolizumab (2015), reslizumab (2016), benralizumab (2017), dupilumab (2017), and tezepelumab (2021) [11–16]. Substantial benefits for patients with severe asthma, which may include reduced exacerbations and glucocorticoid use and improved symptoms, have been demonstrated for these therapies [17–19]. A number of studies to date have indicated that clinical remission may be an achievable goal for some patients treated with mepolizumab; however, the available data are mainly limited to smaller studies (n ≤ 71)[20, 21] or post hoc analyses of larger real-world evidence studies [6], and as such, additional data from larger real-world studies are required. A recent study of 300 patients with severe asthma with an eosinophilic phenotype treated with mepolizumab in Southern Italy (REMI-M study) reported clinical remission rates at 24 months ranging from 26.8% to 52.9%, dependent on the definition used [22]. A real-world cohort study of patients treated with any biologic showed that the proportion of patients achieving individual components of clinical remission ranged from 39% to 80% [23]. These findings indicate that clinical remission is influenced by the number and type of criteria that must be met, highlighting the need for standardized tools and definitions to guide research and clinical practice [22, 23]. Tackling the lack of consensus on remission criteria and variability in treatment goals across different clinical settings is a key barrier to the widespread adoption of clinical remission as an achievable treatment goal in severe asthma [3].

Another substantial barrier is the fragmented care pathways and poor interconnectedness among multidisciplinary teams [3]. Centralized healthcare information systems may be useful in providing consistent documentation, and enable multispecialist teams involved in disease management to track and assess patient outcomes and treatment adherence [3]. In Canada, centralized healthcare information systems are in place via the Canadian universal healthcare system and capture Canadians’ day-to-day interactions with the healthcare system, including physician claims and medical drug claims based on prescriptions [24]. In a real-world study in Canada, mepolizumab was associated with significant reductions in key clinical remission criteria in patients with severe asthma with an eosinophilic phenotype [mean number of asthma exacerbations per patient, and number of oral corticosteroid (OCS) and short-acting β₂-agonist (SABA) [24]].

In the absence of an integrated healthcare system in the U.S., the aim of this study was to utilize data from the Mayo Clinic’s electronic health record (EHR) database to evaluate the real-world documentation and achievement of clinical remission criteria in the first year after initiation of any respiratory biologic, or specifically mepolizumab, using the clinical remission criteria defined by the ATS/ACAAI/AAAAI.

Methods

Study Design

This was a real-world, retrospective cohort study (GSK ID: 219,744) of patients with severe asthma initiated on biologics (mepolizumab, reslizumab, benralizumab, dupilumab, tezepelumab, or omalizumab) using data from Mayo Clinic’s EHR database from January 1, 2014, to March 31, 2023 (Fig. 1). The index date was defined as the date of initiation of any one of the respiratory biologics (the first structured order/administration recorded per patient) at Mayo Clinic between January 1, 2015, and March 31, 2022. The baseline period was defined as the 12-month pre-index, non-inclusive of the index date, and was used to evaluate patient demographics and characteristics. The follow-up period was defined as the 12 months on and after the index date or end of data availability (i.e., end of care at Mayo Clinic or death).

Fig. 1.

Study design. ^aMepolizumab, dupilumab, benralizumab, omalizumab, reslizumab, tezepelumab. ^bIdentified using International Classification of Diseases codes and a natural language processing model that extracts diagnoses from clinical notes. ^cInpatient, outpatient, telemedicine, or emergency department visit. EGPA eosinophilic granulomatosis with polyangiitis, HES hypereosinophilic syndrome

The Mayo Clinic EHR database contains only de-identified and anonymized patient data (i.e., does not include names, addresses, social security or medical record numbers, or other obvious identifiers), and is fully compliant with the Health Insurance Portability and Accountability Act (HIPAA) regulations. No direct patient contact or primary collection of individual human patient data occurred. Confidentiality of patient records was maintained throughout the study and access was limited to investigators participating in data analysis, in line with Mayo Clinic privacy policies. Study results are in tabular form and aggregate analyses that omit patient identification, therefore informed consent, ethics committee or Institutional Review Board authorization were not required. The publication and reports do not include patient identifiers.

Participants

Eligible patients were aged ≥ 18 years at index date, with ≥ 1 respiratory biologic initiated (mepolizumab only for the exploratory objective) between January 1, 2015, and March 31, 2022 (the patient identification period), and ≥ 12 months of clinical activity after the index date. Patients required an asthma diagnosis during the baseline period, extracted via International Classification of Diseases (ICD)-9/10 codes or using nference's augmented curation natural language processing disease diagnosis model [25, 26]. To ensure longitudinality in Mayo Clinic’s health system and continued use of Mayo Clinic as a primary care destination for asthma management, patients were also required to have ≥ 2 refills/administrations of the biologic and ≥ 1 clinical encounter (inpatient, outpatient, telemedicine, or emergency department visit) within the 12-month follow-up period, in addition to ≥ 1 hospitalization encounter after the 12-month follow-up period. Patients were excluded if they had evidence of respiratory biologic use prior to coming to Mayo Clinic for care, or a diagnosis of eosinophilic granulomatosis with polyangiitis (EGPA) or hypereosinophilic syndrome (HES) during baseline.

EHR Data Abstraction and Processing

This analysis used retrospective EHR de-identified patient data from Mayo Clinic, including structured (demographics, vitals, laboratory tests, diagnosis codes, medications, etc.) and unstructured (physicians notes, procedural reports, etc.) data for over 7 million patients across U.S.-based sites. To identify whether patients were on respiratory biologics for asthma, or to identify other comorbid conditions or clinical manifestations, nference's augmented curation disease diagnosis model was utilized [26–28]. This Bidirectional Encoder Representations from Transformers-based model was trained to classify the context of a mentioned disease or symptom in a patient’s clinical note as Yes (confirmed disease), No (ruled out disease), Maybe (possibility of disease), or Other (alternative context, such as family history) [25, 26, 29, 30]. The model was trained on over 25,000 sentences, labeled, and agreed on by at least three clinical scientists. To capture biologic usage otherwise not found in structured data tables, nference’s augmented curation natural language processing medication administration model was used to find positive mentions in unstructured clinical notes [28]. This model was trained on over 13,000 sentences sampled from a variety of therapeutics. It achieved a classification accuracy, sensitivity, and specificity of > 0.88. Manual review of de-identified clinical notes by a clinical scientist was done as needed to confirm findings. All other variables were extracted from structured data tables in the EHR (including ICD codes and medication orders). Definitions for data analyses are presented alongside the ATS/ACAAI/AAAAI clinical remission criteria in Table 1.

Outcomes

The primary objective was to quantify the proportion of patients for whom clinical remission component criteria were documented during the first year after initiation of a respiratory biologic. Clinical remission criteria set out by the ATS/ACAAI/AAAAI consensus [10] were used, which included criteria for asthma exacerbations, SCS use for asthma (oral, injectables), missed work/school due to asthma, ≥ 2 pulmonary function tests (PFT), controller medication use for asthma (ICS, ICS/LABA, leukotriene receptor antagonist [LTRA]), ≥ 2 Asthma Control Tests (ACT), and rescue medication use for asthma (SABA, SABA/short-acting muscarinic antagonists [SAMA], ICS/SABA) (Table 1). The secondary objective was to describe baseline demographics and clinical characteristics among patients with severe asthma receiving respiratory biologics. The exploratory objective was to quantify the proportion of patients initiating mepolizumab that achieved each clinical remission component criterion (and combinations of these) during the follow-up period, specifically in those patients who had individual clinical remission component endpoints documented.

Data Analyses

All study variables, including baseline and outcome measures, were analyzed descriptively. Counts and percentages were provided for categorical variables. Means and standard deviations (SD), medians, and interquartile ranges were provided for continuous variables, and 95% confidence intervals were calculated for proportions. Patients with missing or invalid age, sex, or enrollment dates were excluded from the study sample. There was no data manipulation or imputation for missing values.

Results

Study Population

Of the 4623 patients in Mayo Clinic’s database identified as being on respiratory biologics, 44 were excluded because they were not biologic-naïve at baseline, 2153 were excluded because they did not have an asthma diagnosis at baseline, 116 were excluded as they were not ≥ 18 years of age, 1534 were excluded due to insufficient follow-up, and a further 69 were excluded because they had a diagnosis of HES or EGPA during baseline, resulting in 707 patients considered eligible for inclusion (Fig. 2). Of these, 141 (19.9%) patients were receiving mepolizumab (Supplementary Fig. S1), 198 (28.0%) omalizumab, 77 (10.9%) benralizumab, 298 (42.1%) dupilumab, and 2 (0.3%) reslizumab. Overall, mean (SD) patient age at baseline was 52.1 (15.9) years, and most patients were female, Caucasian, and classified as overweight or obese. The mean (SD) Quan–Charlson comorbidity index score was 1.1 (1.1), corresponding to mild comorbidity (Table 2; Supplementary Table S1).

Fig. 2.

Patient disposition (overall population). Percentages refer to the proportion of the initial population (n = 4623). ^aInpatient, outpatient, or emergency department visit. EGPA eosinophilic granulomatosis with polyangiitis, HES hypereosinophilic syndrome, ICD International Classification of Diseases

Table 2.

Baseline patient demographics and clinical characteristics

	Asthma population on respiratory biologics (N = 707)
Age, years, mean (SD)a	52.1 (15.9)
Sex, n (%)a
Female	431 (61.0)
Male	276 (39.0)
Race/ethnicity, n (%)a
Non-Hispanic White	615 (87.0)
Non-Hispanic Black	29 (4.1)
Hispanic	26 (3.7)
Non-Hispanic Asian	11 (1.6)
Non-Hispanic Native American or Pacific Islander	7 (1.0)
Unknown	19 (2.7)
BMI (kg/m2)b
Mean (SD)	29.7 (7.2)
Median (Q1, Q3)	28.5 (24.1, 33.5)
Quan–CCI score, mean (SD)b	1.13 (1.1)
Quan–CCI score, n (%)b
0	138 (19.5)
1–2	497 (70.3)
3–4	61 (8.6)
≥ 5	11 (1.6)
All-cause HCRUb
Patients with ≥ 1 visit during the 12-month baseline period excluding the index date, n (%) [events, mean (SD)]
IP visit	92 (13.0) [1.6 (1.1)]
ED visit	217 (30.7) [2.2 (1.9)]
OP visit	683 (96.6) [23.0 (25.4)]
Virtual visit	8 (1.1) [1.3 (0.7)]
Asthma-related HCRUb,c
Patients with ≥ 1 visit during the 12-month baseline period excluding the index date, n (%) [events, mean (SD)]
IP visit	70 (9.9) [1.6 (1.0)]
ED visit	155 (21.9) [1.8 (1.5)]
OP visit	568 (80.3) [6.8 (7.6)]
Virtual visit	0 (0.0) [0.0 (0.0)]
Asthma exacerbations, n (%)b
IP/ED-defined exacerbationd	171 (24.2)
SCS-defined exacerbatione	33 (4.7)
IP/ED or SCS-defined exacerbation	176 (24.9)
FEV1 (% predicted)
Patients with ≥ 1 assessment, n (%)	237 (33.5)
Number of assessments per patient
Mean (SD)	1.5 (0.9)
Median (Q1, Q3)	1.0 (1.0, 2.0)
Results
Mean (SD)	65.3 (30.1)
Median (Q1, Q3)	72.0 (50.0, 87.8)
ACT scoreb,f
Patients with ≥ 1 assessment, n (%)	203 (28.7)
Number of assessments per patient
Mean (SD)	1.9 (1.3)
Median (Q1, Q3)	1 (1, 2)
Resultsg
≤ 15, n (%)	86 (42.4)
16–19, n (%)	28 (13.8)
20–25, n (%)	89 (43.8)
Mean (SD)	16.8 (6.3)
Median (Q1, Q3)	18 (11, 22)
Missed school or work due to asthma or asthma-related symptoms, n (%)	9 (1.27)

ACQ Asthma Control Questionnaire, ACT asthma control test, AirQ Asthma Impairment and Risk Questionnaire, BMI body mass index, ED emergency department, FEV₁ forced expiratory volume in 1 s, HCRU healthcare resource utilization, IP inpatient, OCS oral corticosteroid, OP outpatient, Q1 first quartile, Q3 third quartile, Quan–CCI Quan–Charlson comorbidity index, SCS systemic corticosteroid, SD standard deviation

^aEvaluated at the index date

^bEvaluated during the 12-month baseline period, not including the index date

^cAsthma-related HCRU episodes were identified as any clinical encounter with a primary diagnosis of asthma

Type of asthma exacerbation at index (i.e., IP/ED-defined exacerbation, SCS-defined exacerbation), which was defined as: ^dIP/ED-defined exacerbation: includes hospitalization-defined and ED-defined exacerbations. Hospitalization-defined exacerbation: defined as an asthma-related hospitalization or asthma-related

^eSCS-defined exacerbation: defined as an asthma-related ED visit or asthma-related OP visit with an SCS (i.e., OCS or parenteral corticosteroid) administration or order at Mayo within ± 5 days and explicit mention in the clinical note of an asthma exacerbation or flare

^fACQ score was also collected in 2 patients and both had a score ≥ 1.5; AirQ score was not collected in any patients

^gProportions are out of the total number of patients with ≥ 1 measurement

Documentation of Clinical Remission Components

For the overall population of patients receiving respiratory biologics, documentation during the 12-month follow-up period was available for ≥ 1 clinical remission component in 94.2% (n = 666) of patients, but none had the required data available for all 7 criteria (Fig. 3A). In total, 90.2%, 83.2%, 55.4%, and 33.0% of patients had documentation of controller medication use, rescue medication use, SCS use, and asthma exacerbations, respectively. Less than 15% of patients had documentation of asthma control and PFT stability, and < 1% of patients had documentation of missed work/school due to asthma.

Fig. 3.

Patients with documentation (A) and achievement (B) of clinical remission criteria combinations during follow-up^a. ^aIncludes patients meeting any combination of clinical remission components; no patients achieved ≥ 6 clinical remission components. ^bNumber of patients who met clinical remission component criteria during follow-up/number of patients with that component captured during follow-up. CI confidence interval

The proportion of patients with documentation of clinical remission components was greater for the subgroup of patients receiving mepolizumab than for the overall population, in terms of number of clinical remission components documented (Fig. 3A) and for each individual component, except missed work/school (Fig. 4A). For the 141 patients with asthma receiving mepolizumab, documentation during the 12-month follow-up period was available for ≥ 1 clinical remission component in all patients, although none had documentation of all 7 criteria. In total, 99.3%, 98.6%, 83.7%, and 44.0% of patients had documentation of controller medication use, rescue medication use, SCS use, and asthma exacerbations, respectively.

Fig. 4.

Patients with documentation (A) and achievement (B) of individual clinical remission criteria during follow-up^a. ^aProportion of patients who achieved the missed work/school due to asthma component was not assessed due to its infrequent documentation. ^bNumber of patients who met clinical remission component criteria during follow-up/number of patients with that component captured during follow-up. CI confidence interval, N/A not applicable, PFT pulmonary function test, SCS systemic corticosteroid

Achievement of Clinical Remission Components

For the overall population of patients receiving respiratory biologics who had clinical remission components documented during the 12-month follow-up period, most patients (91.2%; n = 645) achieved ≥ 1 clinical remission component criterion during the same period (Fig. 3B). The proportion of patients achieving multiple clinical remission component criteria decreased inversely with the number of criteria, with 6.2% of patients achieving ≥ 4 and 0.6% of patients achieving ≥ 5 clinical remission components. For the individual clinical remission component criterion, 85.0% of patients with ≥ 2 PFT assessments documented during the 12-month follow-up period had stable lung function, 82.9% of patients with ≥ 1 low- or medium-dose ICS, ICS/LABA and/or LTRA order/administration documented required only low- or medium-dose controller medication, and 75.2% of patients with ≥ 1 SABA and/or SABA/SAMA order had asthma-related symptoms requiring rescue therapy no more than once a month (Fig. 4B). Among patients whose records documented exacerbations and SCS use, 44.2% experienced no exacerbations and 44.1% did not use SCS during the 12-month follow-up period, respectively.

For the subgroup of patients receiving mepolizumab who had clinical remission components documented during the 12-month follow-up period, almost all patients (98.6%; n = 139) achieved ≥ 1 clinical remission component criterion during the same period (Fig. 3B). A greater proportion of patients receiving mepolizumab achieved ≥ 1, ≥ 2, ≥ 3, and ≥ 4 clinical remission component criteria than the overall biologics population. For the individual clinical remission component criteria, 85.6% of patients with ≥ 1 SABA and/or SABA/SAMA order had asthma-related symptoms requiring rescue therapy no more than once a month, 84.4% of patients with ≥ 2 PFT assessments documented during the 12-month follow-up period had stable lung function, and 79.3% of patients with ≥ 1 low- or medium-dose ICS, ICS/LABA, and/or LTRA order/administration documented required only low- or medium-dose controller medication (Fig. 4B). In patients with documented prior exacerbations or SCS use, 45.2% experienced no exacerbations and 33.9% did not use SCS during follow-up, respectively.

When investigating clinical remission component criteria that were achieved in combination in patients with the relevant criteria documented, 62.1% met the criteria for controller medication use at low or medium doses only plus rescue medication use less than once per month; 53.4% met the criteria for controller medication use at low or medium doses only plus no SCS use (Table 3; Supplementary Fig. S2).

Table 3.

Combinations of clinical remission component criteria achieved by each patient during follow-up

Combinations of clinical remission component criteria	Patients with asthma who received respiratory biologics during follow-up
	Number of patients with documentation of all clinical remission component criteria listed	Of those with documentation, number (%) of patients who achieved all clinical remission component criteria listed
2: Controller medication usea  + Rescue medication useb	564	350 (62.1)
2: Controller medication usea  + No SCS use	251	134 (53.4)
2: Rescue medication useb  + No SCS use	208	119 (57.2)
3: Controller medication usea  + Rescue medication useb + no SCS use	187	97 (51.9)
2: Controller medication usea  + No asthma exacerbations	415	77 (18.6)
2: Rescue medication useb  + No asthma exacerbations	364	77 (21.2)

SCS systemic corticosteroid

^aController medication use at low or medium doses

^bRescue medication use due to asthma less than once per month

Discussion

In this real-world analysis of chronic disease care in patients with severe asthma in the U.S. who initiated biologics (using the Mayo Clinic’s EHR database and clinical remission component criteria agreed upon by the ATS/ACAAI/AAAAI [10]), over 90% of patients receiving respiratory biologics had documentation of at least one clinical remission component, but no patients had all ATS/ACAAI/AAAAI criteria documented. The proportion of patients achieving multiple criteria decreased inversely with the number of criteria, with < 1% of patients achieving > 5 and no patients achieving 6. These results are as expected; the ATS/ACAAI/AAAAI working group acknowledged that the standards for clinical remission are high and highlighted that, in the absence of an available and acceptable standard definition, there was a need to increase the rigor of existing clinical remission criteria. These criteria can provide an ambitious treatment goal, but the low rate of documentation and attainment reported here may also partially reflect the challenges encountered in collecting these data in routine real-world practice in the U.S. Additionally some criteria may be more relevant to real-world practice than others; for instance, clinicians may wish to prioritize meeting certain selected outcomes over others when making treatment decisions in real-world practice, for example, the reduction of exacerbation rates may be of particular importance for those with severe asthma, given that this is the primary outcome driving biologic use in this patient population [1]. Treating to a specific target, in this case clinical remission, is recommended in other chronic diseases and has been demonstrated to be beneficial by providing physicians with clearly defined and sequentially measured goals [31, 32]. Physicians should help patients recognize the level of control achievable with current asthma treatments and work collaboratively to set ambitious goals through shared decision-making [3].

The results of this study also highlight the difficulty of assessing clinical remission in real-life clinical practice, even when using assessments considered part of usual care, in part due to the absence of a structured approach with standardized assessments for patients with severe asthma. Some of the ATS/ACAAI/AAAAI criteria were infrequently recorded in patients’ EHRs, particularly absence from work or school due to asthma, which was documented in < 1% of records. Structured care models, such as the Chronic Care Model and the Innovative Care for Chronic Conditions Framework, emphasize evidence-based decisions, routine surveillance, and adherence support in the management of chronic conditions, promoting strategies such as coordinated care teams, standardization of care process, patient education, and self-management [33, 34]. These strategies may further enhance care and outcomes for patients with severe asthma. While incorporating these elements into asthma management will require additional resources, it may facilitate more comprehensive documentation of clinical remission criteria, ultimately contributing to a more holistic approach and better care and outcomes for these patients. Another potential solution to the lack of comprehensive data documentation is the use of standardized assessment protocols across healthcare settings and centralized healthcare information systems, which would improve documentation of critical metrics and facilitate the achievement of important patient outcomes. Substantial variabilities in clinical priorities and documentation practices were evident in the current study, with only 33% of patients having exacerbation rates documented, although this may also have been due to medication being used as a proxy for exacerbation rates. Documentation rates also varied for SCS use (55% of patients) and asthma symptom control or lung function stability (< 15%), despite the latter two assessments being recommended for asthma control at clinical visits for adults, adolescents and children, and part of clinical remission criteria defined in 2023 national guidelines [2, 4, 9].

Compared with the overall population, a greater proportion of patients receiving mepolizumab had clinical remission components documented during the 12-month follow-up period. This may be because patients receiving mepolizumab had higher baseline disease severity, resulting in more frequent clinic visits and testing, which may have increased documentation opportunities. A greater proportion of patients receiving mepolizumab achieved ≥ 1, ≥ 2, ≥ 3, and ≥ 4 clinical remission component criteria compared with the overall population. In particular, less than one-third (28%) of patients in the overall population achieved ≥ 3 criteria compared with nearly half (45%) in the mepolizumab group. While these data point towards a higher likelihood of achieving clinical remission with mepolizumab, they may also be explained, at least in part, by the higher proportion of patients with documentation of clinical remission components in that subgroup.

Regarding individual clinical remission component criteria, there was a trend towards a lower proportion of patients in the mepolizumab subgroup achieving the criterion of no SCS use compared with the overall population. The absence of any SCS use is a stringent criterion. In routine practice, mepolizumab is often initiated in patients with a high baseline corticosteroid burden. Such patients may demonstrate clinically meaningful SCS reductions but still require one or more short bursts, which means they fail the binary “zero use” definition despite improvement. Even under this stringent threshold, approximately one third (33.9%) of patients receiving mepolizumab had no SCS use during follow-up. This is aligned with the recognized OCS sparing effect of mepolizumab in severe asthma with an eosinophilic phenotype.

A previous post hoc analysis of a retrospective, real-world, Phase IV, observational cohort study enrolling patients with severe eosinophilic asthma across 24 Spanish hospitals (REDES) investigated asthma clinical remission with mepolizumab [6]. In that study, data for 3- and 4-component clinical remission definitions (3-component: OCS-free, exacerbation-free, and an ACT score of ≥ 20 at 12 months; 4-component: 3-component with post-bronchodilator forced expiratory volume in 1 s [FEV₁] ≥ 80% at 12 months) were available for 82% (3-component group, n = 260/318) and 45% (4-component group, n = 144/318) of patients, whereas in the current study 10–15% of patients receiving mepolizumab had 3- and 4-component clinical remission criteria available [6]. In both groups in REDES, 37–38% of patients achieved the 3-component definition, and in the group with 4-component clinical remission definitions (n = 144), 30% achieved the 4-component definition, demonstrating that clinical remission is an achievable outcome with mepolizumab [6]. Recently, a retrospective, real-world, multicenter study in Southern Italy (REMI-M study) assessed clinical remission using five different definitions in 303 patients with severe asthma with an eosinophilic phenotype. The study included one 3-component definition (no exacerbations, no OCS use, and ACT score ≥ 20) along with four different 4-component definitions (all three criteria from the 3-component definition plus one of four different lung function criteria). In this study, clinical remission appeared to be an achievable outcome for patients receiving mepolizumab, with 43.2% and 52.9% of patients achieving clinical remission after 12 and 24 months, respectively. Among the 4-component definitions, the greatest proportion of patients achieving remission was observed using FEV₁ decline ≤ 5% from baseline (35.8% and 46.9% after 12 and 24 months, respectively), and the lowest using < 100 mL from the best value of the first 12 months (26.8% after 24 months) [22]. Another retrospective, real-world study of 71 patients with severe asthma who continued mepolizumab for 3 years assessed achievement of clinical remission using six different published remission criteria (including the 4-component definition described above) [21]. Again, the primary distinction among the definitions was for lung function, which was defined as either FEV₁ ≥ 80% or, more generally, ‘stabilization’. Overall, 39–52% of patients achieved complete clinical remission in the first year, which increased to 51–73% at 3 years [21]. The data suggested that factors negatively affecting remission delayed rather than prevented remission, and indicated that earlier treatment may increase the chances of remission [21].

While these studies, and our own, indicate that clinical remission can be achieved in a proportion of patients receiving biologics for severe asthma, the reported achievement rates must be viewed as exploratory considering the inherent limitations of observational research [35], many of which highlight the challenges of estimating clinical remission among real-world patients with asthma. The data used in the current study came exclusively from the Mayo Clinic and represented patients who were predominantly Caucasian and mainly resided in Minnesota, Arizona, and Florida. These data are subject to selection bias, including referral bias (since the Mayo Clinic is a tertiary care facility), such that the patient cohort is limited to individuals whose insurance status, policy terms, or socioeconomic position allowed them access to the Mayo Clinic. Therefore, the results may have limited generalizability to the broader U.S. population and other healthcare systems or hospitals.

Our study, like all observational research, was affected by missing data [36]. Indeed, a key goal of our study was to determine the extent to which data relating to clinical remission criteria are missing from patient health records, within a particular health system. Therefore, we did not utilize methods to account for missing data such as imputation or sensitivity analyses. The extent to which data are missing may be determined by the documentation requirements of the health system itself, and by State regulations or guidelines. For example, many Mayo Clinic sites are in Minnesota, where the documentation of ACT or other acceptable screening tools for asthma control are suggested by Minnesota Community Measurements (mncm.org) and have been adopted by the Minnesota Department of Health. Therefore, it is plausible that the study may have overestimated the extent of real-world ACT data capture in other states. Nonetheless, the ACT is a recommended tool for assessing asthma symptom control, so documentation of ACT data could be considered part of best practice.

The data recorded in EHRs are often imperfect or incomplete [37]. We used natural language processing models to limit the number of data points that may be missed in the unstructured data fields of a patient’s EHR. However, this cannot overcome the possibility that physicians do not record every element of the patient encounter. As noted, absences from work or school as a result of asthma were rarely recorded, but a lack of documentation of remission endpoints does not necessarily mean that physicians did not assess them. Similarly, physicians may be more likely to record the presence of a sign or symptom (e.g., exacerbations requiring SCS or emergency department visits) than their absence, which can affect data collection on clinical remission in conditions such as asthma.

Medication utilization was assessed from structured data tables in the patient’s EHR. Thus, medications were not recorded in the EHR. For example, over-the-counter medications, drug samples, or some medications received during an inpatient stay, were not accounted for in the analysis. Similarly, the analysis does not capture care received by patients at other centers, and may therefore underestimate exacerbations, assessments, or medication use, if patients attended a non-Mayo Clinic facility for emergency or primary care. Had these data been available, the proportion of patients with ≥ 1 or more clinical remission criteria documented would likely be different.

Finally, as described above, the clinical remission component criteria used in this study were defined by American societies, in the absence of a global definition encompassing criteria that are applicable worldwide. Arguably, real-world research based on patients’ EHRs is particularly challenging in the U.S., where the healthcare system is fragmented between multiple payers and providers, compared with countries or regions that have centralized healthcare systems (e.g., Canada, Taiwan, UK). The development of a global definition of clinical remission in asthma would facilitate research that may be compared across countries with different healthcare systems.

Overall, and taken together with these previous findings, our study suggests that the documentation and achievement of ATS/ACAAI/AAAAI-defined clinical remission component criteria may be an attainable goal in the future of severe asthma management. However, the full value of these data can only be realized with global consensus on remission criteria (which should take into account the feasibility of collection in medical records), standardization of patient assessments between organizations and healthcare professionals, a more structured approach to data documentation, and more clearly defined roles for patients and healthcare professionals regarding outcome documentation and data sharing.

Conclusion

This study has revealed infrequent documentation of different clinical remission component criteria in routine clinical practice, thereby limiting the evaluation of whether patients with severe asthma can achieve clinical remission with biologics. Thus, the low coverage observed indicates a need for standardized assessment protocols that encompass all components to accurately assess efficacy of respiratory biologics in the real-world setting. Further evaluation of clinical remission will be required as definitions evolve and, in more diverse patient populations, to determine if the trends observed here are representative of all patients with severe asthma receiving respiratory biologics.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Arijita Deb, Lydia Lee, and Judy Kelloway contributed to the study conception and design. Kaiser Lim, Arijita Deb, Lydia Lee, Hannah Barman, and Tyler Wagner contributed to the acquisition of data. Safak Simsek, Mithun Manoharan, Hannah Barman, and Tyler Wagner contributed to data analysis. Kaiser Lim, Arijita Deb, Thomas Corbridge, Lydia Lee, Judy Kelloway, Hannah Barman, and Tyler Wagner contributed to data interpretation. All authors reviewed and revised the manuscript critically for important intellectual content, agreed to submit to the current journal, gave final approval of the version to be published, and agreed to be accountable for all aspects of the work.

Funding

This study was funded by GSK (219744), who was involved in study design and implementation, as well as data collection, analysis, interpretation, writing the study report and reviewing this manuscript. The sponsor did not place any restrictions on access to data or statements made in the manuscript. All authors had full access to the data upon request and had final responsibility for the decision to submit for publication. The journal’s Rapid Service Fee was also paid by the sponsor.

Data Availability

The summary statistics derived from the Mayo Clinic EHRs are enclosed within the manuscript. The EHR data cannot be shared or released due to HIPAA regulations. For more information on the analyses and how nference use de-identified patient data, please contact Hannah Barman (hannah@nference.net).

Declarations

Conflict of Interest

Kaiser Lim was a consultant for nference, which received payment from GSK to conduct this study. Arijita Deb, Thomas Corbridge, and Judy Kelloway are employed by GSK and hold financial equities in GSK. Lydia Lee was a fellow at GSK at the time of study and is currently employed by Boehringer Ingelheim (Boehringer Ingelheim has no connection to this study). Safak Simsek was employed by nference when this study was conducted, which received payment from GSK to conduct this study. Mithun Manoharan, Hannah Barman, and Tyler Wagner are employees of nference, which received payment from GSK to conduct this study.

Ethical Approval

Aggregated data that omit patient identification were used, informed consent and ethics committee approval was not required, and authorization by the Institutional Review Board was waived. Conference had access to patient data through a partnership with the Mayo Clinic; therefore, permission was not required to access the database.