Inhaled Placebo administered at the clinic as once weekly inhalations with the I-neb® device. Start dose is Placebo 4 mg and dose adaptation is applied (downtitration to 1mg or uptitration to 8 mg or remain on 4 mg) based on appearance of flu like adverse events

In the ACQ-5 questionnaire, participants were asked to recall the status of their asthma during the previous week with regards to symptoms. The questionnaire included the items:

• Awoken at night by asthma symptoms.
• Severity of asthma symptoms in the morning.
• Limitation of daily activities due to asthma.
• Shortness of breath.
• Wheeze. The ACQ-5 score was computed as the un-weighted mean of responses to the 5 items, measured on a 7-point scale from 0 (totally controlled) to 6 (severely uncontrolled). A lower score indicated a better outcome. If ACQ-5 reached a value of 1.5 or more, the participant was reported as having LOAC. Estimates of the LS mean over 52 weeks were analyzed using a repeated measures analysis with treatment, baseline ACQ-5, week and treatment-by-week with participant as random effects, and age and gender as covariates. Baseline was the average of non-missing daily measures/scores over the last 5 days prior to and including the morning of randomization.

Asthma symptoms during night-time and daytime were recorded by the participant each morning and evening in the Asthma Daily Diary. Symptoms were recorded using a 4-point response scale, which ranged from 0 to 3, where 0 indicated no asthma symptoms. Asthma symptom daytime score (recorded in the evening), night-time score (recorded in the morning), and total score were calculated separately. The daily asthma symptom total score was calculated by taking the sum of the night-time and daytime asthma symptom scores recorded each day, ranging from 0 to 6. A lower symptom score indicated a better outcome. Estimates of the LS mean over 52 weeks were analyzed using a repeated measures analysis with treatment, baseline asthma daily diary weekly average, week and treatment-by-week with participant as random effects, and age and gender as covariates. Baseline was the average of non-missing daily measures/scores over the last 5 days prior to and including the morning of randomization.

Moderate exacerbation was defined as a temporary increase in maintenance therapy to prevent a severe event supported by sustained (≥ 2 day) worsening in at least 1 key control metric ie, asthma score, reliever medication use, night time awakening or morning PEF.

Severe exacerbation was defined as a worsening in asthma symptoms and:

• Use of systemic corticosteroids for at least 3 days and/or
• An unscheduled or emergency room visit due to asthma symptoms requiring systemic corticosteroids and/or
• An in-patient hospitalization due to asthma requiring systemic corticosteroids. Time to moderate or severe asthma exacerbation was calculated as start date of first moderate or severe exacerbation - date of randomization + 1. Time to moderate or severe asthma exacerbation was displayed using a Kaplan-Meier plot and the outcome measure is presented as number of participants with events.

The use of SABAs was allowed as rescue medication (reliever bronchodilator) throughout the study. Reliever medication use was captured in the Asthma Daily Diary twice daily (morning and evening), recorded as the number of inhaler puffs. The number of inhalations (puffs) per day was calculated as: number of night inhaler puffs + number of day inhaler puffs. Percentage of participants using reliever medication (SABA) up to Week 52 is presented.

Lung function was assessed by pre- and post-BD FEV1 which was measured by spirometry. To ensure quality control, all spirometry measurements were reviewed to ensure that they met American Thoracic Society / European Respiratory Society criteria for acceptability. Estimates of the LS mean over 52 weeks were analyzed using a repeated measures analysis with treatment, baseline FEV1 (pre- or post-BD, as applicable), visit and treatment-by-visit with participant as random effects, and age and gender as covariates. Baseline was the last non-missing measurement recorded prior to randomization.

Morning and evening PEF measurements were recorded by the participant on a daily basis and then averaged over the week. The weekly average total PEF was calculated by taking the sum of the average of the weekly morning mean and weekly evening mean. Estimates of the LS mean over 52 weeks were analyzed using a repeated measures analysis with treatment, baseline PEF, week and treatment-by-week with participant as random effects, and age and gender as covariates. Baseline was the average of non-missing daily measures/scores over the last 5 days prior to and including the morning of randomization.

FeNO measurements were taken at home by participants every second day. The weekly average FeNO was based on the average of measurements taken at home for a specific week. Estimates of the LS mean over 52 weeks were analyzed using a repeated measures analysis with treatment, baseline FeNO, week and treatment-by-week with participant as random effects, and age and gender as covariates. Baseline was the average of non-missing daily measures/scores over the last 5 days prior to and including the morning of randomization.

Participants received placebo for inhalation via nebuliser solution (up to 13 doses). All participants received placebo to match 4 mg AZD1419 once weekly for 4 doses. The following 9 doses were adapted based on tolerability (severity of flu-like symptoms). Participants were prescribed their usual controller medication (ICS + LABA) in separate inhalers. During the dosing period, controller medications were gradually reduced and discontinued. Treatment was initiated with 6 doses of placebo on top of the participant's controller medication, LABA was then stopped on the evening prior to receiving dose 7, and ICS was gradually tapered down and discontinued during the next 3 weeks before being discontinued in the evening prior to receiving placebo dose 10.

Total of all reporting groups

LOAC was defined as any of the following:

• Increase of ACQ-5 to ≥ 1.5.
• ≥ 30% reduction in morning PEF from baseline on 2 consecutive days.
• ≥ 6 additional reliever inhalations of SABA in a 24-hour period relative to baseline on 2 consecutive days.
• Exacerbation requiring systemic corticosteroids. Number of participants experiencing LOAC up to Week 52 is presented. A generalized linear model based on a generalized estimating equation was used to compare treatments.

The FAS included all randomized participants who received any IP, irrespective of their protocol adherence and continued participation in the study.

In the ACQ-5 questionnaire, participants were asked to recall the status of their asthma during the previous week with regards to symptoms. The questionnaire included the items:

• Awoken at night by asthma symptoms.
• Severity of asthma symptoms in the morning.
• Limitation of daily activities due to asthma.
• Shortness of breath.
• Wheeze. The ACQ-5 score was computed as the un-weighted mean of responses to the 5 items, measured on a 7-point scale from 0 (totally controlled) to 6 (severely uncontrolled). A lower score indicated a better outcome. If ACQ-5 reached a value of 1.5 or more, the participant was reported as having LOAC. Estimates of the LS mean over 52 weeks were analyzed using a repeated measures analysis with treatment, baseline ACQ-5, week and treatment-by-week with participant as random effects, and age and gender as covariates. Baseline was the average of non-missing daily measures/scores over the last 5 days prior to and including the morning of randomization.

The FAS included all randomized participants who received any IP, irrespective of their protocol adherence and continued participation in the study. Only participants with data available for analysis are presented.

Asthma symptoms during night-time and daytime were recorded by the participant each morning and evening in the Asthma Daily Diary. Symptoms were recorded using a 4-point response scale, which ranged from 0 to 3, where 0 indicated no asthma symptoms. Asthma symptom daytime score (recorded in the evening), night-time score (recorded in the morning), and total score were calculated separately. The daily asthma symptom total score was calculated by taking the sum of the night-time and daytime asthma symptom scores recorded each day, ranging from 0 to 6. A lower symptom score indicated a better outcome. Estimates of the LS mean over 52 weeks were analyzed using a repeated measures analysis with treatment, baseline asthma daily diary weekly average, week and treatment-by-week with participant as random effects, and age and gender as covariates. Baseline was the average of non-missing daily measures/scores over the last 5 days prior to and including the morning of randomization.

The FAS included all randomized participants who received any IP, irrespective of their protocol adherence and continued participation in the study. Only participants with data available for analysis are presented.

Moderate exacerbation was defined as a temporary increase in maintenance therapy to prevent a severe event supported by sustained (≥ 2 day) worsening in at least 1 key control metric ie, asthma score, reliever medication use, night time awakening or morning PEF.

Severe exacerbation was defined as a worsening in asthma symptoms and:

• Use of systemic corticosteroids for at least 3 days and/or
• An unscheduled or emergency room visit due to asthma symptoms requiring systemic corticosteroids and/or
• An in-patient hospitalization due to asthma requiring systemic corticosteroids. Time to moderate or severe asthma exacerbation was calculated as start date of first moderate or severe exacerbation - date of randomization + 1. Time to moderate or severe asthma exacerbation was displayed using a Kaplan-Meier plot and the outcome measure is presented as number of participants with events.

The FAS included all randomized participants who received any IP, irrespective of their protocol adherence and continued participation in the study.

The use of SABAs was allowed as rescue medication (reliever bronchodilator) throughout the study. Reliever medication use was captured in the Asthma Daily Diary twice daily (morning and evening), recorded as the number of inhaler puffs. The number of inhalations (puffs) per day was calculated as: number of night inhaler puffs + number of day inhaler puffs. Percentage of participants using reliever medication (SABA) up to Week 52 is presented.

The FAS included all randomized participants who received any IP, irrespective of their protocol adherence and continued participation in the study.

Lung function was assessed by pre- and post-BD FEV1 which was measured by spirometry. To ensure quality control, all spirometry measurements were reviewed to ensure that they met American Thoracic Society / European Respiratory Society criteria for acceptability. Estimates of the LS mean over 52 weeks were analyzed using a repeated measures analysis with treatment, baseline FEV1 (pre- or post-BD, as applicable), visit and treatment-by-visit with participant as random effects, and age and gender as covariates. Baseline was the last non-missing measurement recorded prior to randomization.

The FAS included all randomized participants who received any IP, irrespective of their protocol adherence and continued participation in the study. Only participants with data available for analysis are presented.

Morning and evening PEF measurements were recorded by the participant on a daily basis and then averaged over the week. The weekly average total PEF was calculated by taking the sum of the average of the weekly morning mean and weekly evening mean. Estimates of the LS mean over 52 weeks were analyzed using a repeated measures analysis with treatment, baseline PEF, week and treatment-by-week with participant as random effects, and age and gender as covariates. Baseline was the average of non-missing daily measures/scores over the last 5 days prior to and including the morning of randomization.

The FAS included all randomized participants who received any IP, irrespective of their protocol adherence and continued participation in the study. Only participants with data available for analysis are presented.

FeNO measurements were taken at home by participants every second day. The weekly average FeNO was based on the average of measurements taken at home for a specific week. Estimates of the LS mean over 52 weeks were analyzed using a repeated measures analysis with treatment, baseline FeNO, week and treatment-by-week with participant as random effects, and age and gender as covariates. Baseline was the average of non-missing daily measures/scores over the last 5 days prior to and including the morning of randomization.

The FAS included all randomized participants who received any IP, irrespective of their protocol adherence and continued participation in the study. Only participants with data available for analysis are presented.

Participants received placebo for inhalation via nebuliser solution (up to 13 doses). All participants received placebo to match 4 mg AZD1419 once weekly for 4 doses. The following 9 doses were adapted based on tolerability (severity of flu-like symptoms). Participants were prescribed their usual controller medication (ICS + LABA) in separate inhalers. During the dosing period, controller medications were gradually reduced and discontinued. Treatment was initiated with 6 doses of placebo on top of the participant's controller medication, LABA was then stopped on the evening prior to receiving dose 7, and ICS was gradually tapered down and discontinued during the next 3 weeks before being discontinued in the evening prior to receiving placebo dose 10.

Comparison between groups for participants with moderate or severe asthma exacerbation. Odds ratio was estimated using a generalized linear model based on a generalized estimating equation approach fitting treatment and visit as covariates.

2-sided p-value