In-clinic spirometry measurements were taken prior to the administration of bronchodilators. Baseline was the last measurement prior to first injection of investigational product (IP).

The least squares (LS) means, LS mean differences and 80% confidence intervals (CIs), and one-sided p-value results were based on a mixed model repeated measures (MMRM). The model included fixed effects for baseline, background medication, geographic region, baseline inhaled corticosteroids (ICS) total daily dose, visit, treatment, and the baseline by visit and treatment by visit interactions. Visits within participant were considered as repeated measurements.

The ITT population included participants who were randomised and received any study intervention. Participants with data available are included.

In-clinic spirometry measurements were taken following the use of bronchodilators. Bronchodilatation was induced using albuterol (90 µg metered dose), salbutamol (100 µg metered dose), or levalbuterol (45 µg metered dose), and measurements were taken after up to a maximum of 4 inhalations. Baseline was the last measurement prior to first injection of IP.

The LS means, LS mean differences and 80% CIs, and one-sided p-value results were based on MMRM. The model included fixed effects for baseline, background medication, geographic region, baseline ICS total daily dose, visit, treatment, and the baseline by visit and treatment by visit interactions. Visits within participant were considered as repeated measurements.

The ITT population included participants who were randomised and received any study intervention. Participants with data available are included.

Tozorakimab serum concentrations were measured using a validated assay method.

The PK population included participants who received at least one dose of tozorakimab and had at least one detectable serum concentration measurement post-first dose of study intervention. Participants with data available at each time point are presented.

ADA prevalence is the number of participants ADA positive (ADA+) at baseline and/or post-baseline. Treatment-emergent ADA+ (TE-ADA +) positive is defined as being either of treatment-induced ADA+ (ADA negative [ADA-] at baseline and at least one post-baseline ADA+) and treatment-boosted ADA+ (ADA+ at baseline and baseline titre is boosted by ≥ 4-fold increase at ≥ 1 post-baseline time point). Treatment-emergent ADA- (TE-ADA-) is defined as ADA+ but not fulfilling the definition of TE-ADA+. ADA persistently positive is defined as ADA- at baseline and ADA+ at ≥ 2 post-baseline assessment with ≥ 16 weeks between first and last positive assessments, or ADA+ at the last post-baseline assessment. ADA transiently positive is defined as ADA- at baseline, having at least one post-baseline ADA+ assessment and not fulfilling the conditions of ADA persistently positive. Baseline is defined as the last ADA assessment prior to first injection of IP.

The as-treated population included participants who were randomised and received any study intervention. Participants with data available are included.

In the ACQ-6, participants were asked to recall how their asthma has been during the previous week by responding to one BD-use question and 5 symptom questions. Questions were weighted equally and scored from 0 (totally controlled) to 6 (severely uncontrolled). Mean scores of ≤ 0.75 indicate well-controlled asthma, scores between >0.75 and <1.5 indicate partly controlled asthma, and scores ≥1.5 indicate not well-controlled asthma. Results were based on an MMRM which included fixed effects for baseline, background medication, geographic region, baseline ICS total daily dose, visit, treatment and the baseline by visit and treatment by visit interactions. Visits within participant were considered as repeated measurements. A negative change from baseline indicates an improvement in asthma control.

The ITT population included participants who were randomised and received any study intervention. Participants with data available are included.

In the ACQ-6, participants were asked to recall how their asthma has been during the previous week by responding to one BD-use question and 5 symptom questions. Questions were weighted equally and scored from 0 (totally controlled) to 6 (severely uncontrolled). Mean scores of ≤ 0.75 indicate well-controlled asthma, scores between >0.75 and <1.5 indicate partly controlled asthma, and scores ≥1.5 indicate not well-controlled asthma. A decrease in ACQ-6 score baseline indicates an improvement in asthma control, and individual changes of at least 0.5 are considered clinically meaningful.

The ITT population included participants who were randomised and received any study intervention. Participants with evaluable ACQ-6 scores were included in the analysis.

In the ACQ-6, participants were asked to recall how their asthma has been during the previous week by responding to one BD-use question and 5 symptom questions. Questions were weighted equally and scored from 0 (totally controlled) to 6 (severely uncontrolled). Mean scores of ≤ 0.75 indicate well-controlled asthma, scores between >0.75 and <1.5 indicate partly controlled asthma, and scores ≥1.5 indicate not well-controlled asthma.

The ITT population included participants who were randomised and received any study intervention. Participants with evaluable ACQ-6 scores were included in the analysis.

The SGRQ is a 50-item patient-reported outcome instrument to measure the health status of participants with airway obstruction diseases, giving a total score and 3 domain scores (symptoms, activity, and impacts). The total score is expressed as a percentage of overall impairment, with 100 representing the worst possible health status and 0 the best possible health status. Each domain score ranges from 0 to 100, with higher scores indicating greater impairment. A negative change from baseline indicates an improvement in impairments. Results were based on an MMRM which included fixed effects for baseline, background medication, geographic region, baseline ICS total daily dose, visit, treatment and the baseline by visit and treatment by visit interactions. Visits within participant were considered as repeated measurements.

The ITT population included participants who were randomised and received any study intervention. Participants with data available are included.

The SGRQ is a 50-item patient-reported outcome instrument to measure the health status of participants with airway obstruction diseases, giving a total score and 3 domain scores (symptoms, activity, and impacts). The total score is expressed as a percentage of overall impairment, with 100 representing the worst possible health status and 0 the best possible health status. A decrease in the SGRQ total score indicates an improvement in overall impairment.

The ITT population included participants who were randomised and received any study intervention. Participants with evaluable SGRQ scores were included in the analysis.

Asthma CompEx is a combination of exacerbations of asthma and diary events (i.e., a combination of electronic diary [eDiary] variables). eDiary events are defined by criteria using morning/evening diary variables of PEF, symptoms, and use of rescue medication. A participant was considered to have a CompEx event if they had one or both of an asthma exacerbation or diary event. For participants who did not experience an on-treatment CompEx event, date of censoring was the minimum between the date of last dose + 28 days, and the last day of eDiary recording during the on-treatment period.

The ITT population included participant who were randomised and received any study intervention.

The annualised rate of asthma CompEx events was calculated as the total number of asthma CompEx events / (date of last dose of IP + 28 - date of first dose of IP - recovery time + 1) / 365.25.

The rates, rate ratios, and one-sided p-values were estimated from a negative binomial regression, with the log(follow up time) included as an offset term. The dependent variable will be the number of CompEx events during the on-treatment period (i.e., from baseline to last dose date +28 days), and the model will include treatment group, background medication, geographic region and baseline ICS total daily dose as covariates.

The ITT population included participants who were randomised and received any study intervention.

A standardised single-breath FeNO test was performed to evaluate airway inflammation. Results were based on MMRM on log-transformed change from baseline. Log-transformed change from baseline is calculated as the visit value in log minus the baseline value in log. The results from the model were then back transformed. The model included fixed effects for baseline (in log), background medication, geographic region, baseline ICS total daily dose, visit, treatment and the baseline by visit and treatment by visit interactions. Visits within subject were considered as repeated measurements.

The ITT population included participants who were randomised and received any study intervention. Participants with data available are included.

In-clinic spirometry measurements were taken prior to the administration of bronchodilators. Baseline was the last measurement prior to first injection of IP.

The LS means, LS mean differences and 80% CIs, and one-sided p-value results were based on MMRM. The model included fixed effects for baseline, visit, treatment, and the baseline by visit and treatment by visit interactions. Visits within participant were considered as repeated measurements. Analysis is presented by the number of asthma exacerbations experienced within the 12 months prior to baseline (1 or ≥ 2 exacerbations in the previous 12 months).

Participants in the ITT population with 1 or ≥ 2 exacerbations in the previous 12 months are included in the analysis. The ITT population included participants who were randomised and received any study intervention. Participants with data available are included.

The eosinophil count at baseline and week 16 are presented. Baseline was defined as the last measurement prior to first injection of IP.

The ITT population included participants who were randomised and received any study intervention. Participants with data available are included.