Safety and Tolerability of Nintedanib in Patients with Fibrosing Interstitial Lung Diseases: Post-marketing Data

Abstract

Introduction

Nintedanib is approved for the treatment of idiopathic pulmonary fibrosis (IPF), other forms of progressive pulmonary fibrosis (PPF), and systemic sclerosis-associated interstitial lung disease (ILD). We present global post-marketing safety data for nintedanib in these fibrosing ILDs.

Methods

Data on adverse events in patients with fibrosing ILDs who were treated with nintedanib were collected via spontaneous reporting and solicited reporting in various studies (excluding clinical trials). Data were collected from 15 October 2014 (first regulatory approval) to 15 October 2023. Adverse events were coded using the Medical Dictionary for Regulatory Activities. Cumulative exposure to nintedanib was estimated using sales data.

Results

Cumulative exposure to nintedanib was 380,557 patient–years. Diarrhoea was reported at a rate of 227.5 per 1000 patient–years. Only 2.6% of diarrhoea events were reported as serious. Of 39,788 (33.6%) diarrhoea events with a known time to onset, almost 60% occurred within the first 3 months of treatment. The rate of serious liver enzyme and bilirubin elevations (including drug-induced liver injury) was 4.0 per 1000 patient–years. Bleeding was reported at a rate of 24.2 per 1000 patient–years. Most (81.3%) bleeding events were non-serious. The rates of myocardial infarction, ischaemic stroke, and venous thromboembolism were 3.3, 3.3, and 2.0 per 1000 patient–years, respectively. Gastrointestinal perforation was reported at a rate of 0.9 per 1000 patient–years.

Conclusion

Post-marketing safety data on established and potential adverse events associated with nintedanib in patients with fibrosing ILDs, collected over 9 years, demonstrated a safety profile that was similar to that established in clinical trials and provided in the product labels. Education of patients about the adverse events that may be associated with nintedanib, and the effective management of adverse events when they occur, is important to minimise the impact of adverse events and help patients remain on treatment.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12325-024-03023-4.

Plain Language Summary

Nintedanib is a drug that is used to treat interstitial lung diseases (ILDs) that lead to fibrosis (scarring) of the lungs. The results of clinical trials showed that the most frequent adverse events seen in patients treated with nintedanib are gastrointestinal events, particularly diarrhoea. This analysis looked at safety data from patients with ILDs treated with nintedanib. Data were collected from the time that nintedanib was approved for the treatment of ILDs (15 October 2014) to 15 October 2023. The findings showed that the safety profile of nintedanib in the real world was similar to that seen in clinical trials. Diarrhoea was the most frequent adverse event reported, but was only regarded as serious in about 1 in 38 events. It is important that clinicians using nintedanib to treat ILDs educate patients about the adverse events that may occur and how to manage them.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12325-024-03023-4.

Key Summary Points

Why carry out this study?

Nintedanib is a tyrosine kinase inhibitor approved for the treatment of idiopathic pulmonary fibrosis, other forms of progressive pulmonary fibrosis, and systemic sclerosis-associated interstitial lung disease.

We analysed real-world post-marketing safety data for nintedanib in patients with fibrosing interstitial lung diseases.

What was learned from the study?

Post-marketing safety data collected over 9 years showed that nintedanib had a safety profile in the real world that was similar to that seen in clinical trials.

Patients need education about the adverse events that may be associated with nintedanib and the effective management of such events if they occur.

Introduction

Interstitial lung diseases (ILDs) are a heterogeneous group of lung disorders that may result in pulmonary fibrosis [1, 2]. Nintedanib, an intracellular inhibitor of tyrosine kinases [3, 4], has been approved for the treatment of idiopathic pulmonary fibrosis (IPF), progressive fibrosing ILDs other than IPF (generally referred to as progressive pulmonary fibrosis or PPF), and ILD associated with systemic sclerosis (SSc-ILD). The efficacy and safety of nintedanib were investigated in the phase III INPULSIS trials in patients with IPF [5], the SENSCIS trial in patients with SSc-ILD [6], and the INBUILD trial in patients with PPF [7]. In these trials, nintedanib slowed the progression of ILD, with adverse events that were predominantly gastrointestinal and similar across the trials [8–10]. Safety data from the open-label extensions of these trials suggested that the safety profile of nintedanib was consistent over longer-term treatment [11–13].

A global pharmacovigilance programme is collecting data on post-marketing adverse events in patients with IPF, PPF, and SSc-ILD who are treated with nintedanib, regardless of causal relationship. A previous analysis of post-marketing safety data collected between October 2014 and October 2018 suggested that the safety profile of nintedanib in patients with IPF in clinical practice was similar to that observed in clinical trials and described in the product labels [14]. Here we present an update on the post-marketing safety data for nintedanib collected in patients with IPF, SSc-ILD, and PPF.

Methods

Data Source and Collection

Data on adverse events in patients with IPF, SSc-ILD, and PPF who were treated with nintedanib were collected via the spontaneous reporting of adverse events (including reports from authorities, patients, and healthcare professionals), and solicited reporting from various studies (e.g., patient support programmes, compassionate use programmes, and non-interventional studies). These studies varied in their duration and in the frequency with which information on adverse events was solicited. Safety data from clinical trials were not included.

Data are presented as cases or adverse events. The minimum data required to identify a case were an identified patient, use of nintedanib, and an adverse event. Cumulative post-marketing safety data collected in the Boehringer Ingelheim Global Safety Platform (GSP) from 15 October 2014 (the date of the first regulatory approval of nintedanib) to 15 October 2023 were included in the analysis. Sales data were used to estimate cumulative exposure to nintedanib. It was assumed that each patient was treated with two capsules per day and that all capsules were used. This analysis did not require institutional review board approval as it was based on aggregate patient de-identified data from a pharmacovigilance database.

Adverse Events

Adverse events were coded using preferred terms in the Medical Dictionary for Regulatory Activities (MedDRA) version 26.0. Single preferred terms and specific groupings of preferred terms such as standardised MedDRA queries (SMQs) (standard sets of preferred terms related to a medical condition or area of interest) were used. SMQs can comprise broad or narrow terms, with narrow terms being those that are highly likely to represent the condition of interest.

Based on the safety profile of nintedanib established in clinical trials and important potential risks arising from its inhibition of the vascular endothelial growth factor receptor (VEGFR) [3, 4, 15], the following safety topics were assessed: diarrhoea (MedDRA preferred term); drug-induced liver injury (MedDRA preferred term); liver enzyme and bilirubin elevations (SMQs ‘liver related investigations, signs and symptoms’ [broad terms], ‘drug related hepatic disorders—comprehensive search’ [narrow terms], ‘cholestasis and jaundice of hepatic origin’ [narrow terms], ‘hepatitis non-infectious’ [narrow terms]); bleeding (SMQ ‘haemorrhage terms [excluding laboratory terms]’ [narrow terms]); myocardial infarction (MI) (SMQ ‘myocardial infarction’ [narrow terms]); ischaemic stroke (SMQ ‘ischaemic central nervous system vascular conditions’ [narrow terms]); venous thromboembolism (SMQ ‘embolic and thrombotic events, venous [narrow terms]); and gastrointestinal (GI) perforation (SMQ ‘gastrointestinal perforation’ [narrow terms]). Bleeding events were also assessed in subgroups of patients by use of anticoagulation or antiplatelet therapy (yes/no) defined on the basis of the World Health Organisation Anatomical Therapeutic Chemical group “antithrombotic agents”.

Serious adverse events were defined as events that resulted in hospitalisation or prolonged hospitalisation, resulted in persistent or clinically significant disability or incapacity, were life-threatening, resulted in death, or were considered serious for another medical reason. For the purpose of regulatory reporting, all events suggestive of drug-induced liver injury were considered serious events. Adverse events were considered as related or unrelated to nintedanib on the basis of the opinion of the individual reporting the event for inclusion in the GSP. The incidence of each adverse event was calculated as the rate per 1000 patient–years.

Results

Exposure

At the time of this analysis, nintedanib was approved for the treatment of IPF in 86 countries, for SSc-ILD in 84 countries, and for progressive fibrosing ILDs other than IPF (PPF) in 84 countries. The total number of cases in the GSP was 157,775 (149,677 IPF, 2878 SSc-ILD, 5220 PPF). Of these cases, 102,386 (64.9%) were aged ≥ 65 years, 24,960 (15.8%) were aged < 65 years, and 30,429 (19.3%) were of unreported age. Cumulative exposure to nintedanib was 380,557 patient–years (249,251 patient–years for nintedanib 150 mg bid and 131,305 patient–years for nintedanib 100 mg bid). Use of anticoagulation therapy was reported in 12,594 cases (8.0%).

Diarrhoea

The rate of diarrhoea was 227.5 per 1000 patient–years. Overall, 86,594 cases (comprising 118,399 events) were reported. Of the 86,594 cases, 77,896 (90.0%) were derived from studies, 8275 (9.6%) from spontaneous reporting, and 423 (0.5%) from individual patient use.

Of the 118,399 events, the vast majority were non-serious; 3030 events (2.6%) were reported as serious. Following an event, nintedanib was withdrawn in 22.5% of cases; the dose was reduced in 8.4% of cases and the dose was unchanged in 32.9% of cases; the action taken was unknown in 35.8% of cases. No data was collected on the proportion of patients who had nintedanib reintroduced following withdrawal. The characteristics of the cases of diarrhoea, including comorbidities and medication use, are presented in Table S1. The time to onset of diarrhoea was reported for 39,788 events (33.6%). Almost 60% of these events had a time of onset within the first 3 months of treatment.

Serious Liver Enzyme and Bilirubin Elevations and Drug-Induced Liver Injury

The rate of serious liver enzyme and bilirubin elevations (including drug-induced liver injury) was 4.0 per 1000 patient–years. Overall, 1561 cases (comprising 1776 events) were reported. Of the 1561 cases, 1149 (73.6%) were derived from studies, 386 (24.7%) from spontaneous reporting, and 26 (1.7%) from individual patient use. The time to onset was reported for 863 events (48.6%); of these, 550 (63.7%) occurred within the first 3 months of treatment. The most frequently reported events were hepatic enzyme increased (23.3%), liver function test increased (7.7%), and liver disorder (7.7%). Of the 1776 events, 1386 (78.0%) were reported as related to nintedanib. Following an event, nintedanib was withdrawn in 56.9% of cases, the dose reduced in 5.6% of cases, and the dose unchanged in 9.7% of cases; the action taken was unknown in 27.4% of cases. No data was collected on the proportion of patients who had nintedanib reintroduced following withdrawal.

The characteristics of the cases of serious liver enzyme and bilirubin elevations (including drug-induced liver injury) are presented in Table S2. Fifty-one events (2.9%), reported in 49 cases, were fatal. Of these 49 cases, severe comorbidities including underlying or new onset diseases or concomitant hepatotoxic medication were reported in 27 cases (Table 1). Nineteen cases had insufficient information documented and only three cases were suggestive of a causal association with nintedanib.

Table 1.

Characteristics of cases of serious liver enzyme and bilirubin elevations (including drug-induced liver injury) with fatal outcome

Characteristic	N
Serious liver enzyme and bilirubin elevations (including drug-induced liver injury)	1561
Fatal cases of serious liver enzyme and bilirubin elevations (including drug-induced liver injury)	49
Involved severe comorbidities (e.g., acute exacerbation of idiopathic pulmonary fibrosis, respiratory failure, multiorgan failure, sepsis, COVID-19)	12
Involved events of liver cancer and/or metastasis to liver	9
Involved strong confounders (e.g., hepatic fibrosis/cirrhosis, alcohol use, concomitant hepatotoxic medication)	6
Suggestive of causal association with nintedanib	3
Insufficient information documented	19

The rate of drug-induced liver injury (a subgroup of serious liver enzyme and bilirubin elevations) was 0.23 per 1000 patient–years. Overall, 90 cases (comprising 93 events) were reported. Of the 90 cases, 60 (66.7%) were derived from studies, 27 (30.0%) from spontaneous reporting, and 3 (3.3%) from individual patient use. The time to onset of drug-induced liver injury was reported for 56 events (60.2%); of these events, 40 (71.4%) occurred within the first 3 months of treatment. Four events of drug-induced liver injury (4.3%), reported in four cases, were fatal. Following an event, nintedanib was withdrawn in 56.9% of cases; the dose was reduced in 5.6% of cases and unchanged in 9.7% of cases; the action taken was unknown in 27.4% of cases.

Bleeding

The rate of bleeding was 24.2 per 1000 patient–years. Overall, 9313 cases (comprising 12,091 bleeding events) were reported. Of the 9313 cases, 8272 (88.8%) were derived from studies, 985 (10.6%) from spontaneous reporting, and 56 (0.6%) from individual patient use.

Time to onset of bleeding was reported for 4294 events (35.5%), of which 53% occurred within the first 3 months of treatment. The most frequent bleeding adverse events were epistaxis (22.9% of events), contusion (12.5% of events), and haematochezia (12.2% of events) (Table S3). Most bleeding events were non-serious; 2382 events (19.7%) were serious. Overall, 173 events (1.4%) reported in 161 cases were fatal. Fatal events included gastrointestinal haemorrhage (23 events; 13.3%), cerebral haemorrhage (16 events; 9.2%), haemorrhage (15 events; 8.7%), haemoptysis (12 events; 6.9%), and pulmonary alveolar haemorrhage (10 events; 5.8%). Of the 12,091 bleeding events, concomitant use of anticoagulation (2394 events,19.8%) and no use of anticoagulation (2520 events, 20.8%) were reported in a similar proportion of events, but data were missing for 7177 (59.4%) events. Following a bleeding event, nintedanib was withdrawn in 28.8% of cases; the dose was reduced in 5.2% of cases and unchanged in 26.6% of cases; the action taken was unknown in 39.0% of cases.

Myocardial Infarction (MI)

The rate of MI was 3.3 per 1000 patient–years. Overall, 1289 cases (comprising 1390 events) were reported. Of the 1289 cases, 1026 (79.6%) were derived from studies, 239 (18.5%) from spontaneous reporting, and 24 (1.9%) from individual patient use. Of the 1390 events, 429 (30.9%) were reported as related to nintedanib. Following an event, nintedanib was withdrawn in 34.7% of cases; the dose was reduced in 2.7% of cases and unchanged in 18.1% of cases; the action taken was unknown in 44.4% of cases. The characteristics of the cases with MI, including comorbidities and medication use, are presented in Table S4. The most frequent comorbidities reported for patients in the cases of MI were hypertension (12.1%), coronary artery disease (7.9%), and diabetes mellitus (5.9%). Time to onset of MI was available for 657 events (47.3%), 40% of which occurred more than 1 year after initiation of nintedanib.

Ischaemic Stroke

The rate of ischaemic stroke was 3.3 per 1000 patient–years. Overall, 1285 cases (comprising 1448 events) were reported. Of the 1285 cases, 1064 (82.8%) were derived from studies, 214 (16.7%) from spontaneous reporting, and 7 (0.5%) from individual patient use. The most frequent events were cerebrovascular accident (59.9%), transient ischaemic attack (22.9%), and cerebral infarction (3.9%). The characteristics of the cases of ischaemic stroke, including comorbidities and medication use, are presented in Table S5. The most frequent comorbidity in the cases of ischaemic stroke was hypertension (12.2%). Time to onset of ischaemic stroke was available for 725 events (50.1%). There was no discernible pattern of time to onset.

Venous Thromboembolism

The rate of venous thromboembolism was 2.0 per 1000 patient–years. Overall, 781 cases (comprising 866 events) were reported. Of the 781 cases, 617 (79.0%) were derived from studies, 150 (19.2%) from spontaneous reporting, and 14 (1.8%) from individual patient use.

The most frequent events were pulmonary embolism (52.5%), pulmonary thrombosis (26.1%), and deep vein thrombosis (12.7%). The characteristics of the cases of venous thromboembolism, including comorbidities and medication use, are presented in Table S6. The most frequent comorbidities in the cases of venous thromboembolism were hypertension (12.2%) and gastro-oesophageal reflux disease (5.9%). Time to onset of venous thromboembolism was available for 390 events (45.0%), of which 33% occurred over 1 year after initiation of nintedanib.

Gastrointestinal Perforation

The rate of gastrointestinal perforation was 0.9 per 1000 patient–years. Overall, 356 cases (comprising 395 events) were reported. Of the 356 cases, 262 (73.6%) were derived from studies, 87 (24.4%) from spontaneous reporting, and 7 (2.0%) from individual patient use.

The most frequent events were intestinal perforation (19.5%), gastric perforation (11.1%), and large intestine perforation (9.9%) (Table S7). Of the 395 events, 228 (57.7%) were reported as related to nintedanib. The characteristics of the cases of gastrointestinal perforation, including comorbidities and medication use, are presented in Table S8. The most frequent comorbidities in the cases of gastrointestinal perforation were hypertension (10.4%) and gastro-oesophageal reflux disease (5.1%). Time to onset of gastrointestinal perforation was available for 165 events (41.8%). There was no discernible pattern in the time to onset.

Discussion

Post-marketing safety data for nintedanib in patients with fibrosing ILDs, collected over 9 years following its approval by regulatory authorities, demonstrated a safety profile similar to that established in clinical trials for all three indications (IPF, SSc-ILD, progressive fibrosing ILDs other than IPF). The most frequently reported adverse event associated with nintedanib was diarrhoea. Diarrhoea is a known side effect of tyrosine kinase inhibition and may be the result of localised changes in the gut following inhibition of the VEGFR [16]. Other real-world studies have also found diarrhoea to be the most frequent adverse event in patients treated with nintedanib and the adverse event that most frequently leads to dose reduction [17]. It is recommended in the product label to treat diarrhoea in patients taking nintedanib with adequate hydration and antidiarrheal medication (e.g. loperamide), and consider dose reduction or treatment interruption if diarrhoea continues. Among nintedanib-treated patients who experienced diarrhoea adverse events in the INPULSIS, SENSCIS, and INBUILD trials, most patients (88–94%) reported events that were mild or moderate in intensity [8–10]. In the post-marketing safety data, serious diarrhoea events were rare, with only 2.6% of the diarrhoea events reported as serious. Overall, nintedanib was withdrawn in 22.5% of cases of diarrhoea. No data were collected on the reintroduction of nintedanib following resolution of diarrhoea, as recommended in the product label.

Clinical trial data have shown that nintedanib is associated with a risk of liver enzyme elevations and drug-induced liver injury [8–10]. An analysis of pooled data from trials in patients with IPF, SSc-ILD, and PPF showed that liver enzyme elevations are more frequent among female than male patients [18]. It is recommended that liver enzymes be monitored prior to initiation of nintedanib, at regular intervals during the first 3 months of treatment, and then periodically or as clinically indicated. Temporary dose reductions or discontinuations may be required. In the post-marketing safety data, the rate of serious liver enzyme and bilirubin elevations (including drug-induced liver injury) was low (4.0 per 1000 patient–years). Most events occurred in the first 3 months of treatment and were managed by dose reduction or discontinuation. Events of drug-induced liver injury were rare, reported at a rate of 0.23 per 1000 patient–years.

As an inhibitor of the VEGFR, nintedanib may be associated with an increased risk of bleeding [19, 20]. Patients with a known risk of bleeding should only receive nintedanib if the anticipated benefit outweighs the potential risk. Clinical trials of nintedanib excluded patients requiring fibrinolysis, full-dose therapeutic anticoagulation, or high-dose antiplatelet therapy [5–7]. Post-marketing data show that in clinical practice, nintedanib is being used in patients receiving anticoagulation or antiplatelet therapy. The rate of bleeding in the post-marketing data (24.2 per 1000 patient–years) was lower than the rate reported in patients treated with nintedanib in the INPULSIS and INBUILD trials (118 and 116 patient–years, respectively) [10, 14]. Most cases of bleeding in the post-marketing safety data were non-serious and there was no evidence that the use of anticoagulation therapy had a relevant impact on the occurrence or severity of bleeding events. However, patients receiving nintedanib and full-dose anticoagulation should be monitored closely for bleeding and their anticoagulation treatment adjusted if necessary.

Patients with IPF and other forms of ILD represent a population at high risk of cardiovascular disease [21–23]. In clinical trials, patients with a history of myocardial infarction within 6 months, unstable angina within 6 months, or thrombotic events (including stroke and transient ischaemic attacks) within 12 months were excluded [5–7]. In these trials, the overall frequency of cardiovascular adverse events was low and similar between the nintedanib and placebo groups [8–10, 24]. The rate of myocardial infarction in the post-marketing safety data (3.3 per 1000 patient–years) was lower than those reported in the INPULSIS trials (17 per 1000 patient–years) and INBUILD trial (11.4 per 1000 patient–years) and based on claims data from 7298 patients with IPF (13.8 per 1000 patient–years) [10, 14, 22]. The most frequent comorbidities reported for patients in the cases of MI were known risk factors for cardiovascular events, including hypertension, coronary artery disease, and diabetes mellitus. In the post-marketing safety data, the rate of ischaemic stroke (3.3 per 1000 patient–years) was lower than the rates reported in the INBUILD trial (7.0 per 1000 patient–years) and based on claims data from patients with IPF (19.2 per 1000 patient–years) [10, 22]. The rate of venous thromboembolism in the post-marketing safety data (2.0 per 1000 patient–years) was lower than the rates reported in the INBUILD trial (7.0 per 1000 patient–years) and based on claims data from patients with IPF (37.6 per 1000 patient–years) [10, 22]. Although events are rare, patients should be advised about signs of acute myocardial ischaemia and other arterial thromboembolic events and the urgency to seek medical care. The product labels state that physicians should use caution when treating patients at higher cardiovascular risk including known coronary artery disease.

Patients treated with VEGFR inhibitors might have an increased risk of gastrointestinal perforation. Therefore, caution should be used when treating patients with nintedanib who have known contributing factors e.g. recent abdominal surgery, diverticular disease, or concomitant use of corticosteroids or non-steroidal anti-inflammatory drugs. In clinical trials of nintedanib, low rates of gastrointestinal perforation were reported in patients with IPF (0.3%) and progressive fibrosing ILDs (0.3%) [8, 10]. Similarly, a low incidence rate of gastrointestinal perforation (0.9 per 1000 patient–years) was reported in the post-marketing safety data. This is lower than the rate reported in the claims data from patients with IPF (3.5 per 1000 patient–years) [22] and consistent with the rate that would be expected in this patient population. Nevertheless, nintedanib should be permanently discontinued in patients who develop gastrointestinal perforation. Nintedanib can be reintroduced after recovery and careful assessment of the patient’s condition and risk factors.

Analyses of pharmacovigilance data have limitations, including under-reporting of adverse events (particularly non-serious events) and incomplete information capture when events are reported [25, 26]. Many cases in the nintedanib post-marketing safety data lacked information on the time to onset of the event, comorbidities, concomitant medication use, and outcomes. For example, for almost 50% of the fatal bleeding events, it is unknown whether the patient was taking concomitant anticoagulant or antiplatelet medication. Similarly, data on concomitant medications were missing for 54% of the cases of serious liver enzyme and bilirubin elevations (including drug-induced liver injury). Sales data were used to estimate cumulative exposure to nintedanib and thus the incidence rates of adverse events. This may have led to underestimation of incidence rates, as not all capsules that are sold are taken by patients. Rates of serious and fatal outcomes are confounded by concomitant diseases and medications that are not reported in the database and cannot be excluded. This manuscript focused on the important potential risks of nintedanib and did not cover all the adverse events that may be associated with nintedanib. As a result of the small number of patients with diagnoses other than IPF, rates of adverse events per indication were not analysed. Comparisons among pharmacovigilance, clinical trial, and epidemiological data should be undertaken with caution because of differences in patient populations and methodology.

Conclusion

Post-marketing safety data on established and potential adverse events associated with nintedanib in patients with fibrosing ILDs demonstrated a safety profile that was similar to that established in clinical trials and provided in the product labels. Education of patients about the adverse events that may be associated with nintedanib, and the effective management of adverse events when they occur, is important to minimise the impact of adverse events and help patients remain on treatment.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Kamila Sroka-Saidi, Elvira Erhardt, Ivana Ritter contributed to the study conception and design. Elvira Erhardt contributed to the data analysis. Nazia Chaudhuri, Arata Azuma, Kamila Sroka-Saidi, Elvira Erdhardt, Ivana Ritter, and Sergio Harari contributed to the writing or critical review of the manuscript. All authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and have given their approval for this version to be published. The authors did not receive payment for the development of this manuscript.

Funding

The clinical trials and pharmacovigilance programme described in this paper were supported by Boehringer Ingelheim. The Rapid Service Fee and Open Access fee were supported by Boehringer Ingelheim.

Data Availability

To ensure independent interpretation of clinical study results and enable authors to fulfil their roles and obligations under the ICMJE criteria, Boehringer Ingelheim grants all authors access to relevant clinical study data. In adherence with the Boehringer Ingelheim Policy on Transparency and Publication of Clinical Study Data, researchers can request access to clinical study data, typically one year after the approval has been granted by major regulatory authorities or after termination of the development programme. Researchers should use https://vivli.org/ to request access to data and visit https://www.mystudywindow.com/msw/datasharing for further information.

Declarations

Conflicts of Interest

Nazia Chaudhuri reports consulting fees from Boehringer Ingelheim, Redx, Liminal BioSciences, Vicore Pharma AB, Bridge Biotherapeutics, tranScrip and speaker fees from Boehringer Ingelheim. Nazia Chaudhuri is an Editorial Board member of Advances in Therapy. Nazia Chaudhuri was not involved in the selection of peer reviewers for the manuscript nor any of the subsequent editorial decisions Arata Azuma reports research support from Boehringer Ingelheim; consulting fees from Boehringer Ingelheim and Kyorin Pharma Co.; speaker fees from Boehringer Ingelheim and participation on a data safety monitoring board for Taiho Co. and Toray Co. Kamila Sroka-Saidi, Elvira Erhardt and Ivana Ritter are employees of Boehringer Ingelheim. Sergio Harari reports research support from Aerovate, AstraZeneca, and Boehringer Ingelheim and consulting fees from Roche.

Ethical Approval

This analysis did not require institutional review board (IRB) approval as it was based on aggregate patient de-identified data from a pharmacovigilance database.