Nintedanib in Progressive Fibrosing Interstitial Lung Diseases...diffusing capacity of the lung for carbon monoxide ranging from 30 to less than 80% of the predicted value. Randomization

T h e n e w e ngl a nd j o u r na l o f m e dic i n e

n engl j med nejm.org 1

The authors’ full names, academic degrees, and affiliations are listed in the Appendix. Address reprint requests to Dr. Brown at the Department of Medicine, National Jewish Health, 1400 Jackson St., Denver, CO 80206, or at brownk@ njhealth . org.

*A complete list of investigators in the INBUILD trial is provided in the Supple-mentary Appendix, available at NEJM.org.

This article was published on September 29, 2019, at NEJM.org.

DOI: 10.1056/NEJMoa1908681Copyright © 2019 Massachusetts Medical Society.

BACKGROUNDPreclinical data have suggested that nintedanib, an intracellular inhibitor of tyro-sine kinases, inhibits processes involved in the progression of lung fibrosis. Although the efficacy of nintedanib has been shown in idiopathic pulmonary fibrosis, its ef-ficacy across a broad range of fibrosing lung diseases is unknown.

METHODSIn this double-blind, placebo-controlled, phase 3 trial conducted in 15 countries, we randomly assigned patients with fibrosing lung disease affecting more than 10% of lung volume on high-resolution computed tomography (CT) to receive nintedanib at a dose of 150 mg twice daily or placebo. All the patients met criteria for progression of interstitial lung disease in the past 24 months despite treatment and had a forced vital capacity (FVC) of at least 45% of the predicted value and a diffusing capacity of the lung for carbon monoxide ranging from 30 to less than 80% of the predicted value. Randomization was stratified according to the fibrotic pattern (a pattern of usual interstitial pneumonia [UIP] or other fibrotic patterns) on high-resolution CT. The primary end point was the annual rate of decline in the FVC, as assessed over a 52-week period. The two primary populations for analy-sis were the overall population and patients with a UIP-like fibrotic pattern.

RESULTSA total of 663 patients were treated. In the overall population, the adjusted rate of decline in the FVC was −80.8 ml per year with nintedanib and −187.8 ml per year with placebo, for a between-group difference of 107.0 ml per year (95% confidence interval [CI], 65.4 to 148.5; P<0.001). In patients with a UIP-like fibrotic pattern, the adjusted rate of decline in the FVC was −82.9 ml per year with nintedanib and −211.1 ml per year with placebo, for a difference of 128.2 ml (95% CI, 70.8 to 185.6; P<0.001). Diarrhea was the most common adverse event, as reported in 66.9% and 23.9% of patients treated with nintedanib and placebo, respectively. Abnormalities on liver-function testing were more common in the nintedanib group than in the placebo group.

CONCLUSIONSIn patients with progressive fibrosing interstitial lung diseases, the annual rate of decline in the FVC was significantly lower among patients who received nintedanib than among those who received placebo. Diarrhea was a common adverse event. (Funded by Boehringer Ingelheim; INBUILD ClinicalTrials.gov number, NCT02999178.)

A BS TR AC T

Nintedanib in Progressive Fibrosing Interstitial Lung Diseases

K.R. Flaherty, A.U. Wells, V. Cottin, A. Devaraj, S.L.F. Walsh, Y. Inoue, L. Richeldi, M. Kolb, K. Tetzlaff, S. Stowasser, C. Coeck, E. Clerisme-Beaty, B. Rosenstock,

M. Quaresma, T. Haeufel, R.-G. Goeldner, R. Schlenker-Herceg, and K.K. Brown, for the INBUILD Trial Investigators*

Original Article

P19-08802New England Journal of Medicine2019_______________________________________________________________________________________________________________

Page 1 of 49

Please follow your local copyright law



P atients with a spectrum of lung

disorders, including idiopathic pulmonary fibrosis (IPF),1,2 have a progressive fibrosing

clinical phenotype that is characterized by an increasing extent of fibrosis on high-resolution computed tomography (CT), decline in lung func-tion, worsening of symptoms and quality of life, and early death despite current therapy.3-6 On the basis of the clinical and pathophysiological simi-larities among these diseases, it has been postu-lated that such disorders with a progressive phe-notype have a common pathobiologic mechanism regardless of the cause and thus could all have a response to similar treatment.4

Nintedanib is an intracellular inhibitor of ty-rosine kinases.7 Preclinical data have suggested that nintedanib inhibits processes involved in the progression of lung fibrosis.7-11 In patients with IPF and systemic sclerosis-associated interstitial lung disease, treatment with 150 mg of ninted-anib twice daily reduced the rate of decline in the forced vital capacity (FVC).12-14 We conducted the INBUILD trial to investigate the efficacy and safety of nintedanib in patients with fibrosing interstitial lung diseases with a progressive phe-notype.

Me thods

Trial Design and Oversight

The INBUILD trial was a randomized, double-blind, placebo-controlled, parallel-group trial con-ducted at 153 sites in 15 countries.3 The trial was carried out in compliance with the protocol (avail-able with the full text of this article at NEJM.org) and with the principles of the Declaration of Helsinki and the Harmonised Tripartite Guide-line for Good Clinical Practice of the Interna-tional Conference on Harmonisation; the trial was approved by the local authorities. All the patients provided written informed consent before trial entry.

All the authors had access to the data, which were analyzed by the sponsor, Boehringer Ingel-heim. The authors assume responsibility for the accuracy and completeness of the data and for the fidelity of the trial to the protocol. Medical writing assistance, funded by the sponsor, was provided by FleishmanHillard Fishburn.

Patients

Recruitment began in February 2017 and ended in April 2018. Eligible patients were adults (≥18 years

of age) with a physician-diagnosed fibrosing in-terstitial lung disease. Because patients with IPF had already been studied, specific efforts were made to enroll patients with a progressive fibrotic phenotype other than IPF. Enrolled patients had features of fibrosing lung disease affecting more than 10% of lung volume on high-resolution CT, as confirmed by central review (Section B in the Supplementary Appendix, available at NEJM.org).

The patients were required to meet at least one of the following criteria for progression of interstitial lung disease within the 24 months before screening, despite standard treatment with an agent other than nintedanib or pirfeni-done: a relative decline in the FVC of at least 10% of the predicted value, a relative decline in the FVC of 5% to less than 10% of the predicted value and worsening of respiratory symptoms or an increased extent of fibrosis on high-resolu-tion CT, or worsening of respiratory symptoms and an increased extent of fibrosis. At the time of enrollment, patients were required to have an FVC of at least 45% of the predicted value and a diffusing capacity of the lung for carbon monox-ide (corrected for hemoglobin) of 30 to less than 80% of the predicted value.

Patients who were treated with azathioprine, cyclosporine, mycophenolate mofetil, tacrolimus, rituximab, cyclophosphamide, or oral glucocor-ticoids (at a dose of more than 20 mg per day for glucocorticoids) were excluded. At the discretion of the investigator, initiation of these medications was allowed after 6 months of trial treatment in patients with clinically significant deterioration of interstitial lung disease or connective tissue disease. Key exclusion criteria are provided in Sec-tion C in the Supplementary Appendix.

Trial Treatment

Patients were randomly assigned in a 1:1 ratio to receive oral nintedanib (at a dose of 150 mg twice daily) or placebo with the use of interac-tive-response technology. Since some studies have suggested that the progression of fibrosing interstitial lung disease is more rapid in patients with an imaging pattern of usual interstitial pneu-monia (UIP) on high-resolution CT than in those with other fibrotic patterns,15-18 randomization was stratified according to the imaging pattern (UIP-like fibrotic pattern or other fibrotic patterns) on the basis of central review. An enrichment design19 was planned, with stratification of the trial population so that two thirds of the patients


Page 2 of 49



Nintedanib in Fibrosing Interstitial Lung Diseases

had a UIP-like fibrotic pattern (as identified ac-cording to the criteria of the INPULSIS trials13) (Section B in the Supplementary Appendix) and one third had other fibrotic patterns (i.e., a 2:1 ratio). However, stratification caps were not im-plemented, since recruitment led to a ratio close to 2:1 without the need for active management.

For each patient, the trial consisted of two parts: Part A, which was conducted during the first 52 weeks, and Part B, which was a variable treatment period beyond week 52 during which patients continued to receive either nintedanib or placebo until all the patients had completed Part A. (Details are provided in Section D in the Supplementary Appendix.) At the end of the tri-al, all the patients had the option of receiving nintedanib in an open-label extension trial. Pa-tients who had adverse events were permitted to interrupt treatment or to reduce the dose of nintedanib to 100 mg twice daily. Patients who discontinued either nintedanib or placebo were asked to attend all visits as originally planned. The first database lock took place after the last patient had completed the week 52 visit.

End Points

The primary end point was the annual rate of decline in the FVC, as assessed over the 52-week period. Spirometry was performed at baseline and at weeks 2, 4, 6, 12, 24, 36, and 52, in accor-dance with international guidelines20 with the use of sponsor-provided spirometers (Flowscreen CT).

The main secondary end points were the ab-solute change from baseline in the total score on the King’s Brief Interstitial Lung Disease (K-BILD) questionnaire at week 52, the time until the first acute exacerbation of interstitial lung disease or death over the 52-week period, and the time until death over the 52-week period. (The K-BILD ques-tionnaire is a self-administered health-status questionnaire that has been developed in patients with interstitial lung diseases.21 It consists of 15 items in three domains: breathlessness and ac-tivities, psychological factors, and chest symp-toms. Domain and total scores range from 0 to 100, with higher scores representing better health status. The minimal clinically important differ-ence has not been established, but a change of between 4 and 8 points has been suggested to represent a meaningful change.22-24) Further end points included a composite of an acute exacer-bation of interstitial lung disease or death or

death alone as assessed during the period until the first database lock.

In line with the categorization of acute exac-erbations of IPF in the latest international work-ing group report,25 we defined acute exacerba-tions of interstitial lung disease as acute, clinically significant respiratory deteriorations characterized by evidence of new, widespread alveolar abnormality meeting all the following criteria: acute worsening or development of dys-pnea (typically of <1 month duration), CT with new bilateral ground-glass opacity or consolida-tion superimposed on a background pattern con-sistent with fibrosing interstitial lung disease, and deterioration not fully explained by cardiac failure or fluid overload. Infection was not an ex-clusion criterion for an acute exacerbation. Safety was assessed by clinical and laboratory evalua-tion and the recording of adverse events, as coded with the use of the Medical Dictionary for Regulatory Activities, version 22.0.

Statistical Analysis

All analyses were conducted in patients who had received at least one dose of nintedanib or place-bo. The two primary populations were the overall population and patients with a UIP-like fibrotic pattern on high-resolution CT. The primary end point and safety were assessed in the overall popu-lation, in patients with a UIP-like fibrotic pattern, and in patients with other fibrotic patterns. Analy-ses of the secondary and further efficacy end points were limited to the overall population and to patients with a UIP-like fibrotic pattern. The primary end point was analyzed with the use of a random coefficient regression model (with ran-dom slopes and intercepts), including baseline FVC (reported in milliliters) and imaging pattern as covariates. The slope of the FVC decline was calculated for every patient and the average com-pared between the two assigned groups. The analysis was based on all measurements obtained over the 52-week period, including those from patients who had discontinued nintedanib or pla-cebo. The model allowed for missing data on the assumption that data were missing at random.

To maintain a type I error rate of 5%, a Hoch-berg procedure was used.26 For the primary end point, statistical significance was indicated if the analyses in the two primary populations were significant at a two-sided 5% level or the analysis in either primary population was sig-


Page 3 of 49




nificant at a two-sided 2.5% level. Sensitivity analyses for the primary end point were per-formed to analyze the potential effects of miss-ing data and assumptions made in the primary analysis model. (Details are provided in Sections E and F in the Supplementary Appendix.) All other end points were exploratory (Section G in the Supplementary Appendix). Because the sta-tistical analysis plan did not include a provision for correcting for multiple comparisons for sec-ondary or other outcomes, results are reported as point estimates and 95% confidence intervals.

The widths of the confidence intervals have not been adjusted for multiple comparisons, so the intervals should not be used to infer definitive treatment effects.

In the determination of sample size, between-group differences in the adjusted rate of decline in the FVC were assumed to be 75 to 100 ml per year in patients with a UIP-like fibrotic pattern and 60 to 75 ml per year among those with other fibrotic patterns. This calculation was based on assumptions that the rate of decline in the FVC in the trial population would be similar to that

Figure 1. Enrollment and Randomization in the Overall Population.

A list of other key exclusion criteria is provided in Section C in the Supplementary Appendix.

663 Underwent randomizationand were treated

1010 Patients were assessed for eligibility

347 Were excluded304 Did not meet inclusion criteria or met exclusion

criteria88 Did not have fibrosing lung disease of >10%

extent on high-resolution CT74 Did not have diffusing capacity for carbon

monoxide of 30% to <80% of predicted value142 Did not meet ≥1 other inclusion criterion

or met additional exclusion criteria22 Withdrew consent

5 Had adverse event16 Had other or unknown reason

332 Were assigned to receive nintedanib 331 Were assigned to receive placebo

311 Completed 52-wk observation period314 Completed 52-wk observation period

18 Did not complete 52-wkobservation period

20 Did not complete 52-wkobservation period

252 Did not prematurely discontinuenintedanib before wk 52

80 Prematurely discontinuednintedanib

65 Had adverse event11 Withdrew

1 Had protocol deviation3 Had other reason

282 Did not prematurely discontinueplacebo before wk 52

49 Prematurely discontinued placebo34 Had adverse event

9 Withdrew2 Had protocol deviation1 Was lost to follow-up3 Had other reason


Page 4 of 49




in patients with IPF but slightly lower in patients with a fibrotic pattern other than a UIP-like pat-tern, along with the assumption that as in patients with IPF,13 the rate of decline in the FVC would be 50% lower in the nintedanib group than in the placebo group. We determined that a sample of 600 patients, including two thirds (400 patients) with a UIP-like fibrotic pattern, would provide a power of approximately 90% to detect a between-group difference of 100 ml per year in patients with a UIP-like fibrotic pattern (assumed stan-dard deviation, 300 ml) and to detect a between-group difference of 92 ml per year in the overall population (assumed standard deviation, 337 ml).

R esult s

Patients

A total of 663 patients underwent randomization and received at least one dose of nintedanib or placebo (332 in the nintedanib group and 331 in the placebo group); of the 663 patients, 412 (62.1%) had a UIP-like fibrotic pattern (Fig. 1). The base-

line characteristics of the patients were similar in the two primary populations and in the two treatment groups (Table 1, and Sections H and I in the Supplementary Appendix).

In the overall population, the mean (±SD) age was 65.8±9.8 years, the FVC was 69.0±15.6% of the predicted value, and the diffusing capacity of the lung for carbon monoxide was 46.1±13.6% of the predicted value. The most frequent diag-noses of interstitial lung disease were chronic hypersensitivity pneumonitis (in 26.1% of the patients) and autoimmune interstitial lung dis-eases (in 25.6%). Section J in the Supplementary Appendix provides details regarding the use of restricted background therapies at baseline (a use that was balanced in the two trial groups) and regarding restricted therapies that were added over the 52 weeks of treatment.

In the overall population, 252 patients (75.9%) in the nintedanib group and 282 (85.2%) in the placebo group completed 52 weeks of treatment (Fig. 1). The mean duration of exposure during 52 weeks was 10.3±3.8 months in the nintedanib

CharacteristicNintedanib (N = 332)

Placebo (N = 331)

Male sex — no. (%) 179 (53.9) 177 (53.5)

Age — yr 65.2±9.7 66.3±9.8

Former or current smoker — no. (%) 169 (50.9) 169 (51.1)

UIP-like fibrotic pattern on high-resolution CT — no. (%) 206 (62.0) 206 (62.2)

Criteria for disease progression in previous 24 mo — no. (%)

Relative decline in FVC of ≥10% of predicted value 160 (48.2) 172 (52.0)

Relative decline in FVC of 5% to <10% of predicted value plus wors-ening of respiratory symptoms or increased extent of fibrosis on high-resolution CT

110 (33.1) 97 (29.3)

Worsening of respiratory symptoms and increased extent of fibrosis on high-resolution CT

62 (18.7) 61 (18.4)

FVC

Mean value — ml 2340±740 2321±728

Percent of predicted value 68.7±16.0 69.3±15.2

Diffusing capacity for carbon monoxide†

Mean value — mmol/min/kPa 3.5±1.2 3.7±1.3

Percent of predicted value 44.4±11.9 47.9±15.0

Total score on K-BILD questionnaire‡ 52.5±11.0 52.3±9.8

* Plus–minus values are means ±SD. FVC denotes forced vital capacity, and UIP usual interstitial pneumonia.† The values for diffusing capacity for carbon monoxide were corrected for the hemoglobin level.‡ Scores on the King’s Brief Interstitial Lung Disease (K-BILD) questionnaire range from 0 to 100, with higher scores rep-

resenting better health status.

Table 1. Characteristics of the Overall Population at Baseline.*


Page 5 of 49




group and 11.2±2.6 months in the placebo group. The mean duration of exposure before the first database lock was 15.0±6.8 months in the nin-tedanib group and 16.2±5.5 months in the pla-cebo group.

Primary and Other FVC End Points

In the overall population, the adjusted rate of decline in the FVC over the 52-week period (the

primary end point) was −80.8 ml per year in the nintedanib group and −187.8 ml per year in the placebo group (between-group difference, 107.0 ml; 95% confidence interval [CI], 65.4 to 148.5; P<0.001). In patients with a UIP-like fibrotic pat-tern, the adjusted rate of decline in the FVC over the 52-week period was −82.9 ml per year in the nintedanib group and −211.1 ml per year in the placebo group (between-group difference, 128.2

End PointNintedanib (N = 332)

Placebo (N = 331)

Difference (95% CI)

Primary end point

Rate of decline in the FVC at 52 wk — ml/yr†

Overall population −80.8±15.1 −187.8±14.8 107.0 (65.4 to 148.5)‡

Patients with a UIP-like fibrotic pattern −82.9±20.8 −211.1±20.5 128.2 (70.8 to 185.6)‡

Patients with other fibrotic patterns −79.0±21.6 −154.2±21.2 75.3 (15.5 to 135.0)§

Main secondary end points

Absolute change from baseline in total score on K-BILD questionnaire at 52 wk¶

Overall population 0.55±0.60 −0.79±0.59 1.34 (−0.31 to 2.98)§

Patients with a UIP-like fibrotic pattern 0.75±0.80 −0.78±0.79 1.53 (−0.68 to 3.74)§

Acute exacerbation of interstitial lung disease or death at 52 wk — no. with event/total no. (%)

Overall population 26/332 (7.8) 32/331 (9.7) 0.80 (0.48 to 1.34)§∥

Patients with a UIP-like fibrotic pattern 17/206 (8.3) 25/206 (12.1) 0.67 (0.36 to 1.24)§∥

Death at 52 wk — no. with event/total no. (%)



Additional end points assessed during period until first database lock

Acute exacerbation of interstitial lung disease or death — no. with event/total no. (%)



Death — no. with event/total no. (%)



* Changes from baseline are adjusted means ±SE based on the statistical models. The two primary populations for analy-sis were the overall population and patients with a UIP-like fibrotic pattern.

† For the primary end point, the patients with a UIP-like fibrotic pattern included 206 in each treatment group. The pa-tients with other fibrotic patterns included 126 in the nintedanib group and 125 in the placebo group.

‡ P<0.001.§ The widths of the confidence intervals have not been adjusted for multiple comparisons, so the intervals should not be

used to infer definitive treatment effects.¶ For the analysis of the scores on the K-BILD questionnaire, 332 patients were included in the nintedanib group and 330

in the placebo group in the overall population; among the patients with a UIP-like fibrotic pattern, included were 206 patients and 205 patients, respectively.

∥ The difference was assessed as a hazard ratio.

Table 2. Efficacy End Points.*


Page 6 of 49




ml; 95% CI, 70.8 to 185.6; P<0.001) (Table 2, and Section K in the Supplementary Appendix). Box-and-whisker plots of the adjusted annual rate of decline in the FVC are provided in Section L in the Supplementary Appendix.

In patients with other fibrotic patterns, the adjusted rate of decline in FVC was −79.0 ml per year in the nintedanib group and −154.2 ml per year in the placebo group (between-group differ-ence, 75.3 ml; 95% CI, 15.5 to 135.0). The treat-ment effect between subgroups according to the imaging pattern was consistent (Section M in the Supplementary Appendix). All sensitivity analyses related to the handling of missing data were sup-portive of the findings in the primary analysis (Sections E and F in the Supplementary Appen-dix). In addition, the curves of observed change from baseline in the FVC in the two groups sepa-rated early and continued to diverge (Fig. 2).

Main Secondary End Points

At week 52, the adjusted mean absolute change from baseline in the total score on the K-BILD questionnaire (measuring activity level, psycho-logical factors, and chest symptoms) was 0.55 points in the nintedanib group and −0.79 points

in the placebo group (between-group difference, 1.34 points; 95% CI, −0.31 to 2.98) in the overall population and 0.75 points and −0.78 points, re-spectively (between-group difference, 1.53; 95% CI, −0.68 to 3.74) in patients with a UIP-like fi-brotic pattern. The percentage of patients who either died or had an acute exacerbation of inter-stitial lung disease over the 52-week period was 7.8% in the nintedanib group and 9.7% in the placebo group (hazard ratio, 0.80; 95% CI, 0.48 to 1.34) in the overall population and 8.3% and 12.1%, respectively (hazard ratio, 0.67; 95% CI, 0.36 to 1.24) in patients with a UIP-like fibrotic pattern. The percentage of patients who died over the 52-week period was 4.8% in the ninted-anib group and 5.1% in the placebo group (haz-ard ratio, 0.94; 95% CI, 0.47 to 1.86) in the overall population and 5.3% and 7.8% (hazard ratio, 0.68; 95% CI, 0.32 to 1.47) in patients with a UIP-like fibrotic pattern.

Additional Prespecified End Points

During the period until the first database lock, the percentage of patients in the overall popula-tion who either died or had an acute exacerbation of interstitial lung disease was 12.3% in the nin-

Figure 2. Decline from Baseline in Forced Vital Capacity (FVC).

Shown is the observed mean change from baseline in FVC over the 52-week trial period in the overall population and in patients with an imaging pattern of usual interstitial pneumonia (UIP) on high-resolution computed tomography in the nintedanib group and the place-bo group. The I bars indicate the standard error.

Mea

n C

hang

e fr

om B

asel

ine

in F

VC

(ml)

0

−100

−50

−150

−200

−2500 6 12 24 36

Week

No. of PatientsOverall population

NintedanibPlacebo

Patients with UIP-like fibroticpattern

NintedanibPlacebo

332331

206206

326325

203202

320326

200202

322325

199201

314320

193197

298311

180190

285296

171176

265274

160162

4 522

Placebo, overallpopulation

Nintedanib, overallpopulation

Placebo, UIP-likefibrotic pattern

Nintedanib, UIP-likefibrotic pattern


Page 7 of 49




tedanib group and 17.8% in the placebo group (hazard ratio, 0.68; 95% CI, 0.46 to 1.01); the per-centage of patients who died was 8.1% and 11.5%, respectively (hazard ratio, 0.70; 95% CI, 0.43 to 1.15). Similar results were observed in patients with a UIP-like fibrotic pattern (Table 2, and Section N in the Supplementary Appendix).

Adverse Events

In the overall population over the 52-week period, the percentages of patients with any adverse events and serious adverse events were similar in the nintedanib group and the placebo group (Ta-ble 3). A greater percentage of patients in the nintedanib group than in the placebo group had adverse events leading to a permanent dose re-duction (33.1% vs. 4.2%) and to discontinuation of either nintedanib or placebo (19.6% vs. 10.3%). Fatal adverse events were less frequent in the nintedanib group than in the placebo group (3.3% vs. 5.1%). The most frequent adverse event was diarrhea, which was reported in 222 pa-tients (66.9%) in the nintedanib group and in 79 patients (23.9%) in the placebo group. The worst episode of diarrhea (according to the Common Terminology Criteria for Adverse Events, version 4.03) was grade 3 in 23 patients in the nintedanib group and in 5 patients in the placebo group. No patients in either group had grade 4 or 5 diarrhea. Nausea, vomiting, abdominal pain, decreased ap-petite, and weight decrease were more frequent in the nintedanib group than in the placebo group. Adverse events that were evaluated during the pe-riod until the first database lock were consistent with those reported over the 52-week period (Sec-tion O in the Supplementary Appendix).

Hepatic adverse events were more common in patients treated with nintedanib than in those treated with placebo (Table 3). On the basis of laboratory tests, elevations in alanine aminotrans-ferase, aspartate aminotransferase, or both to lev-els that were at least three times the upper limit of the normal range were observed in 43 patients (13.0%) treated with nintedanib and 6 patients (1.8%) treated with placebo over the 52-week pe-riod (Section P in the Supplementary Appendix). One patient in each treatment group had concur-rent elevations in liver enzymes and bilirubin that met the criteria for Hy’s law. Elevations in liver enzymes were reversible, with a normalization of

EventNintedanib (N = 332)

Placebo (N = 331)

no. of patients (%)

Adverse event

Any 317 (95.5) 296 (89.4)

Any except for progression of interstitial lung disease†

317 (95.5) 295 (89.1)

Most frequent adverse events‡

Diarrhea 222 (66.9) 79 (23.9)

Nausea 96 (28.9) 31 (9.4)

Bronchitis 41 (12.3) 47 (14.2)

Nasopharyngitis 44 (13.3) 40 (12.1)

Dyspnea 36 (10.8) 44 (13.3)

Vomiting 61 (18.4) 17 (5.1)

Cough 33 (9.9) 44 (13.3)

Decreased appetite 48 (14.5) 17 (5.1)

Headache 35 (10.5) 23 (6.9)

Alanine aminotransferase increased 43 (13.0) 12 (3.6)

Progression of interstitial lung disease† 16 (4.8) 39 (11.8)

Weight loss 41 (12.3) 11 (3.3)

Aspartate aminotransferase increased 38 (11.4) 12 (3.6)

Abdominal pain 34 (10.2) 8 (2.4)

Severe adverse event§ 60 (18.1) 73 (22.1)

Serious adverse event¶ 107 (32.2) 110 (33.2)

Fatal adverse event

Any 11 (3.3) 17 (5.1)

Any except for progression of interstitial lung disease†

10 (3.0) 14 (4.2)

Adverse event leading to treatment discontinuation

65 (19.6) 34 (10.3)

Adverse event leading to permanent dose reduction

110 (33.1) 14 (4.2)

* Listed are adverse events that were reported over the 52-week trial period or until 28 days after the last dose in patients who discontinued nintedanib or placebo before week 52.

† The phrase “progression of interstitial lung disease” was based on the pre-ferred term “interstitial lung disease” in the Medical Dictionary for Regulatory Activities (MedDRA).

‡ Listed are adverse events that were reported in more than 10% of the patients in either treatment group. These events were coded with the use of preferred terms in MedDRA.

§ A severe adverse event was defined as an event that was incapacitating or that caused an inability to work or to perform usual activities.

¶ A serious adverse event was defined as an event that resulted in death, in hospitalization or prolongation of hospitalization, or in persistent or clinically significant disability or incapacity; or was life-threatening, a congenital anoma-ly or birth defect, or deemed to be serious for any other reason.

Table 3. Adverse Events in the Overall Population.*


Page 8 of 49




values or a trend toward normalization on dose reduction and after treatment interruption or discontinuation, along with some instances of spontaneous normalization. The side-effect pro-file of nintedanib was similar in patients with a UIP-like fibrotic pattern and in those with other fibrotic patterns (Section Q in the Supplemen-tary Appendix).

Discussion

In the INBUILD trial, we enrolled patients who had a broad range of fibrosing interstitial lung diseases, which were identified on the basis of the presence of pulmonary fibrosis on chest imaging and common progressive clinical fea-tures. The patients who received nintedanib had a slower progression of interstitial lung disease than those who received placebo, as shown by a lower annual rate of decline in the FVC over the 52-week period, both in the overall trial popula-tion and in patients with a UIP-like fibrotic pat-tern on high-resolution CT. The absolute treat-ment effects in these patient populations were similar in magnitude to those observed in pooled data from the INPULSIS trials in patients with IPF (a between-group difference of 107.0 ml in the overall population and 128.2 ml in patients with a UIP-like fibrotic pattern in our trial, as compared with a between-group difference of 109.9 ml in the INPULSIS trials). The relatively lower rate of FVC decline with nintedanib than with placebo was similar in patients with a UIP-like fibrotic pattern and in those with other fi-brotic patterns. Changes in health-related quality of life, as measured on the K-BILD questionnaire over the 52-week period, were small in the two treatment groups.

The safety and side-effect profile of nintedanib in patients with progressive fibrosing interstitial lung diseases was similar to that observed in patients with IPF27 and systemic sclerosis-associ-ated interstitial lung disease.14 More patients stopped nintedanib than stopped placebo because

of adverse effects, with diarrhea being the most common adverse event. Liver-enzyme elevations were more frequently observed in patients in the nintedanib group than in the placebo group.

Beyond exploring the effects of nintedanib, the INBUILD trial provided insights into the natu-ral history of progressive fibrosing interstitial lung diseases. The annual rates of decline in the FVC in placebo-treated patients in the overall popula-tion and in patients with a UIP-like fibrotic pat-tern were similar to those seen in pooled data from the INPULSIS trials in patients who met a case definition for IPF (−187.8 ml per year, −211.1 ml per year, and −223.5 ml per year, respectively). In our trial, the rate of decline in the FVC in placebo-treated patients with other (non–UIP-like) fibrotic patterns (−154.2 ml per year) was only slightly lower than that in patients with a UIP-like fibrotic pattern and was still clinically relevant, a finding that was consistent with pre-vious observations.15-18 These data support the hypothesis that progressive fibrosing interstitial lung diseases, regardless of clinical diagnosis, have a similar pathobiologic mechanism.

In conclusion, we found that patients who received nintedanib had a slower rate of progres-sion of interstitial lung disease than those who received placebo, independent of the fibrotic pattern on high-resolution CT. This change in physiological outcomes was not accompanied by meaningful changes in measures of quality of life, although nintedanib was associated with a higher frequency of diarrhea, nausea, and vomiting.

Supported by Boehringer Ingelheim.Disclosure forms provided by the authors are available with

the full text of this article at NEJM.org.A data sharing statement provided by the authors is available

with the full text of this article at NEJM.org.We thank the patients who participated in this trial; the mem-

bers of the data and safety monitoring committee (Fernando J. Martinez [chair], Kevin J. Anstrom, John A. Belperio, Ulrich Cos-tabel, and R. Gisli Jenkins) and the adjudication committee (Aryeh Fischer, Jan Grutters, and Peter R. Kowey); Harold Col-lard for his contribution to the steering committee; and Eliza-beth Ng and Wendy Morris of FleishmanHillard Fishburn for their medical writing assistance with earlier versions of the manuscript.

Appendix

The authors’ full names and academic degrees are as follows: Kevin R. Flaherty, M.D., Athol U. Wells, M.D., Vincent Cottin, M.D., Anand Devaraj, M.D., Simon L.F. Walsh, M.D., Yoshikazu Inoue, M.D., Luca Richeldi, M.D., Martin Kolb, M.D., Kay Tetzlaff, M.D., Susanne Stowasser, M.D., Carl Coeck, M.D., Emmanuelle Clerisme-Beaty, M.D., Bernd Rosenstock, Ph.D., Manuel Quaresma, Lic., Thomas Haeufel, M.D., Rainer-Georg Goeldner, Ph.D., Rozsa Schlenker-Herceg, M.D., and Kevin K. Brown, M.D.

The authors’ affiliations are as follows: the Division of Pulmonary and Critical Care Medicine, University of Michigan, Ann Arbor


Page 9 of 49




(K.R.F.); the National Institute for Health Research Respiratory Biomedical Research Unit, Royal Brompton and Harefield NHS Founda-tion Trust (A.U.W.), the National Heart and Lung Institute, Imperial College (A.U.W., A.D., S.L.F.W.), and the Department of Radiol-ogy, Royal Brompton and Harefield NHS Foundation Trust (A.D.) — all in London; the National Reference Center for Rare Pulmonary Diseases, Louis Pradel Hospital, Hospices Civils de Lyon, Claude Bernard University Lyon 1, Unité Mixte de Recherche 754, Lyon, France (V.C.); the Clinical Research Center, National Hospital Organization Kinki-Chuo Chest Medical Center, Sakai City, Japan (Y.I.); Fondazi-one Policlinico A. Gemelli IRCCS, Università Cattolica del Sacro Cuore, Rome (L.R.); McMaster University and St. Joseph’s Healthcare, Hamilton, ON, Canada (M.K.); Boehringer Ingelheim International, Ingelheim am Rhein (K.T., S.S., E.C.-B., M.Q., T.H.), the Depart-ment of Sports Medicine, University of Tübingen, Tübingen (K.T.), and Boehringer Ingelheim Pharma, Biberach (B.R., R.-G.G.) — all in Germany; Boehringer Ingelheim, Brussels (C.C.); Boehringer Ingelheim Pharmaceuticals, Ridgefield, CT (R.S.-H.); and the Depart-ment of Medicine, National Jewish Health, Denver (K.K.B.).

References

1. Raghu G, Remy-Jardin M, Myers JL, et al. Diagnosis of idiopathic pulmonary fi-brosis: an official ATS/ERS/JRS/ALAT clinical practice guideline. Am J Respir Crit Care Med 2018; 198(5): e44-e68.2. Lederer DJ, Martinez FJ. Idiopathic pulmonary fibrosis. N Engl J Med 2018; 378: 1811-23.3. Flaherty KR, Brown KK, Wells AU, et al. Design of the PF-ILD trial: a double-blind, randomised, placebo-controlled phase III trial of nintedanib in patients with progressive fibrosing interstitial lung disease. BMJ Open Respir Res 2017; 4(1): e000212.4. Wells AU, Brown KK, Flaherty KR, Kolb M, Thannickal VJ. What’s in a name? That which we call IPF, by any other name would act the same. Eur Respir J 2018; 51(5): 1800692.5. Cottin V, Wollin L, Fischer A, Qua-resma M, Stowasser S, Harari S. Fibrosing interstitial lung diseases: knowns and un-knowns. Eur Respir Rev 2019; 28(151): 180100.6. Kolb M, Vašáková M. The natural his-tory of progressive fibrosing interstitial lung diseases. Respir Res 2019; 20: 57. 7. Wollin L, Wex E, Pautsch A, et al. Mode of action of nintedanib in the treat-ment of idiopathic pulmonary fibrosis. Eur Respir J 2015; 45: 1434-45.8. Wollin L, Maillet I, Quesniaux V, Hol-weg A, Ryffel B. Antifibrotic and anti- inflammatory activity of the tyrosine ki-nase inhibitor nintedanib in experimental models of lung fibrosis. J Pharmacol Exp Ther 2014; 349: 209-20.9. Redente EF, Aguilar MA, Black BP, et al. Nintedanib reduces pulmonary fibrosis in a model of rheumatoid arthritis-associated interstitial lung disease. Am J Physiol Lung Cell Mol Physiol 2018; 314: L998-L1009.

10. Huang J, Maier C, Zhang Y, et al. Nint-edanib inhibits macrophage activation and ameliorates vascular and fibrotic manifestations in the Fra2 mouse model of systemic sclerosis. Ann Rheum Dis 2017; 76: 1941-8.11. Wollin L, Distler JHW, Redente EF, et al. Potential of nintedanib in treatment of progressive fibrosing interstitial lung dis-eases. Eur Respir J 2019; 54(3).12. Richeldi L, Costabel U, Selman M, et al. Efficacy of a tyrosine kinase inhibitor in idiopathic pulmonary fibrosis. N Engl J Med 2011; 365: 1079-87.13. Richeldi L, du Bois RM, Raghu G, et al. Efficacy and safety of nintedanib in idiopathic pulmonary fibrosis. N Engl J Med 2014; 370: 2071-82.14. Distler O, Highland KB, Gahlemann M, et al. Nintedanib for systemic sclero-sis–associated interstitial lung disease. N Engl J Med 2019; 380: 2518-28.15. Kim EJ, Elicker BM, Maldonado F, et al. Usual interstitial pneumonia in rheumatoid arthritis-associated interstitial lung dis-ease. Eur Respir J 2010; 35: 1322-8.16. Walsh SLF, Sverzellati N, Devaraj A, Keir GJ, Wells AU, Hansell DM. Connec-tive tissue disease related fibrotic lung disease: high resolution computed tomo-graphic and pulmonary function indices as prognostic determinants. Thorax 2014; 69: 216-22.17. Salisbury ML, Gu T, Murray S, et al. Hypersensitivity pneumonitis: radiologic phenotypes are associated with distinct survival time and pulmonary function trajectory. Chest 2019; 155: 699-711.18. Adegunsoye A, Oldham JM, Bellam SK, et al. Computed tomography honey-combing identifies a progressive fibrotic phenotype with increased mortality across diverse interstitial lung diseases. Ann Am Thorac Soc 2019; 16: 580-8.

19. Guidance for industry: adaptive de-sign clinical trials for drugs and biologics (draft guidance). Silver Spring, MD: Food and Drug Administration, 2013.20. Miller MR, Hankinson J, Brusasco V, et al. Standardisation of spirometry. Eur Respir J 2005; 26: 319-38.21. Patel AS, Siegert RJ, Brignall K, et al. The development and validation of the King’s Brief Interstitial Lung Disease (K-BILD) health status questionnaire. Tho-rax 2012; 67: 804-10.22. Patel AS, Siegert RJ, Keir GJ, et al. The minimal important difference of the King’s Brief Interstitial Lung Disease questionnaire (K-BILD) and forced vital capacity in interstitial lung disease. Respir Med 2013; 107: 1438-43.23. Sinha A, Patel AS, Siegert RJ, et al. The King’s Brief Interstitial Lung Disease (KBILD) questionnaire: an updated mini-mal clinically important difference. BMJ Open Respir Res 2019; 6(1): e000363.24. Nolan CM, Birring SS, Maddocks M, et al. Kings Brief Interstitial Lung Disease questionnaire: responsiveness and mini-mum clinically important difference. Eur Respir J 2019; 54(3).25. Collard HR, Ryerson CJ, Corte TJ, et al. Acute exacerbation of idiopathic pul-monary fibrosis: an international work-ing group report. Am J Respir Crit Care Med 2016; 194: 265-75.26. Hochberg Y. A sharper Bonferroni procedure for multiple tests of signifi-cance. Biometrika 1988; 75: 800-2.27. Lancaster L, Crestani B, Hernandez P, et al. Safety and survival data in patients with idiopathic pulmonary fibrosis treat-ed with nintedanib: pooled data from six clinical trials. BMJ Open Respir Res 2019; 6(1): e000397.Copyright © 2019 Massachusetts Medical Society.


Page 10 of 49


Supplementary Appendix

This appendix has been provided by the authors to give readers additional information about their work.

Supplement to: Flaherty KR, Wells AU, Cottin V, et al. Nintedanib in progressive fibrosing interstitial lung dis-eases. N Engl J Med. DOI: 10.1056/NEJMoa1908681


Page 11 of 49


2

Contents

Section A: List of investigators

Section B: HRCT features

Section C: Key exclusion criteria

Section D: Trial design

Section E: Sensitivity analyses of the primary endpoint related to handling missing data

Section F: Tipping point analysis of the primary endpoint

Section G: Statistical analyses of endpoints other than the primary endpoint

Section H: Clinical ILD diagnoses

Section I: Baseline characteristics in subgroups by fibrotic pattern on HRCT

Section J. Restricted therapies taken at baseline and initiated between first and last trial drug

intake over 52 weeks

Section K: Between-group adjusted difference in the annual rate of decline in FVC (mL/year)

over 52 weeks (primary endpoint)

Section L: Box and whisker plots of the adjusted annual rates of decline in FVC over 52 weeks

Section M. Annual rate of decline in FVC (mL/year) over 52 weeks in subgroups by HRCT

pattern

Section N: Time to first acute exacerbation of ILD or death and time to death using data up to

first database lock

Section O: Adverse events reported up to first database lock

Section P. Liver enzyme and bilirubin elevations

Section Q. Adverse events in subgroups by fibrotic pattern on HRCT


Page 12 of 49


3

Section A: List of investigators

Argentina: S. Quadrelli, Fundación Sanatorio Güemes, Buenos Aires; M. Otaola, Centro de Investigaciones Metabólicas (CINME), Buenos Aires; M. A. Bergna, CEMER-Centro Medico De Enfermedades Respiratorias, Florida; P. Elias, INSARES, Mendoza; G. Arce, Instituto Médico de la Fundación Estudios Clínicos, Rosario; A. Cazaux, Centro Dr. Lazaro Langer S.R.L, Alberdi Sur; Belgium: W. Wuyts, UZ Leuven, Leuven; J. Guiot, CHU Sart Tilman, Angleur; B. Bondue, Erasme University Hospital, Bruxelles; C. Dahlqvist, Cliniques universitaires UCL de Mont-Godinne, Yvoir; Canada: L. Homik, Concordia Hospital Respiratory Research, Winnipeg; S. Shapera, Toronto General Hospital, Toronto; A. Cantin, CHUS-Fleurimont Centre Recherche Clinique, Sherbrooke; M. Kolb, St. Joseph's Healthcare Firestone Institute for Respiratory Health, Hamilton; Chile: M. Salinas Fénero, Instituto Nacional del Tórax Departamento de Neumonología, Providencia, Santiago; R. Maturana Rozas, Hospital Clínico Regional de Concepción Dr. Guillermo Grant Benavente, Concepción; A. Silva Orellana, Centro de Investigación del Maule, Talca; China: Z. Xu, Peking Union Medical College Hospital, Beijing; Q. Luo, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou; J. Kang, The First Hospital of China Medical University, Shenyang; H. Cai, Nanjing Drum Tower Hospital Neurology, Nanjing; France: S. Marchand-Adam, Hôpital Bretonneau - CHRU de Tours, Tours; E. Bergot, Hôpital Côte de Nacre - CHU de Caen, Caen; A. Gamez-Dubuis, Hôpital Arnaud de Villeneuve - CHRU de Montpellier, Montpellier; F. Riviere, Hôpital d'Instruction des Armées Percy, Clamart; R. Kessler, Hôpital Civil - Nouvel Hôpital Civil - CHU de Strasbourg, Strasbourg; H. Nunes, Hôpital Avicenne, Bobigny; C. Marquette, Hôpital Pasteur - CHU de Nice, Nice; L. Wemeau, Hôpital Calmette - CHRU de Lille, Lille; S. Jouneau, Hôpital Pontchaillou - CHU de Rennes, Rennes; F. Lebargy, Hôpital Maison Blanche - CHU de Reims, Reims; B. Crestani, Hôpital Bichat - AP-HP, Paris; V. Cottin, Hôpital Louis Pradel - Groupement Hospitalier Est - CHU de Lyon HCL, Bron; M. Reynaud-Gaubert, Hôpital Nord - AP-HM, Marseille; Germany: S. Blaas, Klinik Donaustauf Zentrum für Pneumologie, Donaustauf; F. Bonella, Ruhrlandklinik Westdeutsches Lungenzentrum, Essen; W. Randerath, Krankenhaus Bethanien GmbH, Solingen; J. Hetzel, Universitaetsklinikum Tuebingen, Tuebingen; D. Koschel, Fachkrankenhaus Coswig GmbH, Coswig; M. Kreuter, Thoraxklinik-Heidelberg gGmbH am Universitätsklinikum Heidelberg, Heidelberg; A. Prasse, Medizinische Hochschule Hannover, Hannover; D. Skowasch, Universitätsklinikum Bonn AöR, Bonn; S. Stieglitz, Petrus Krankenhaus Wuppertal, Wuppertal; Italy: R. Refini, Azienda Ospedaliera Universitaria Senese, Siena; S. Cerri, Clinica Malattie Apparato Respiratorio Policlinico di Modena, Modena; A. Pesci, Azienda Ospedaliera San Gerardo di Monza U.O. Clinica Pneumologica, Monza; S. Tomassetti, AUSL di Forlì U.O., Forli; C. Vancheri, A.O.U. Policlinco Vittorio Emanuele U.O.P.I. interstiziopatie e malattie rare del polmone, Catania; F. Varone, Policlinico Universitario A. Gemelli U.O.C. Pneumologia, Rome; Japan: N. Sakamoto, Nagasaki University Hospital, Nagasaki; S. Abe and H. Hayashi, Nippon Medical School Hospital, Tokyo, Bunkyo-ku; T. Saito, Ibarakihigashi National Hospital, Ibaraki, Naka-gun; T. Suda, Hamamatsu University Hospital, Shizuoka, Hamamatsu; H. Kitamura, Kanagawa Cardiovascular and Respiratory Center, Kanagawa, Yokohama; M. Okamoto, Kurume University Hospital, Fukuoka, Kurume; Y. Kondoh, Tosei General Hospital, Aichi, Seto; S. Makino and T. Takeuchi, Osaka Medical College Hospital, Osaka, Takatsuki; Y. Yamada and C. Kono, JR Tokyo General Hospital, Tokyo, Shibuya-ku; Y. Inoue, National Hospital Organization Kinki-Chuo Chest Medical Center, Osaka, Sakai; H. Sugiura, Tohoku University Hospital, Miyagi, Sendai; K. Kishi and H. Takaya, Toranomon Hospital, Tokyo, Minato-ku; H. Yamauchi, Jichi Medical University Hospital, Tochigi, Shimotsuke; K. Ichikado, Saiseikai Kumamoto Hospital, Kumamoto; K. Tomii, Kobe City Hospital Organization Kobe City Medical Center General


Page 13 of 49


4

Hospital, Hyogo, Kobe; H. Takahashi, Sapporo Medical University Hospital, Hokkaido, Sapporo; S. Izumi, National Center for Global Health and Medicine, Tokyo, Shinjuku-ku; T. Kawamura, National Hospital Organization Himeji Medical Center, Hyogo, Himeji; Y. Nishioka, Tokushima University Hospital, Tokushima; Y. Miyazaki, Tokyo Medical and Dental University, Tokyo, Bunkyo-ku; Republic of Korea: J.W. Song, Asan Medical Center, Seoul; J.S. Park, Seoul National University Bundang Hospital, Seongnam; Y. Kim, The Catholic University of Korea, Bucheon St.Mary's Hospital, Bucheon; Poland: E. Jassem, University Clinical Center, Gdansk; J. Kus, National Institute of Tuberculosis and Lung Diseases Outpatient Clinic, Warsaw; W. Piotrowski, Department of Pneumology and Allergy, Medical University of Lodz, Lodz; A. Barczyk, Leszek Giec Upper-Silesian Medical Centre, Katowice; D. Ziora, Dept & Clinic of Internal Diseases/Pulmon.dept, Zabrze; Russian Federation: E. Bazdyrev, Res.Inst.-Compl.Iss.Cardi.Dis., Kemerovo; S. Moiseev, Sechenov First Moscow State Medical University, Moscow; S. Avdeev, Sechenov First Moscow State Medical University, Moscow; M. Ilkovich, Scientific Research Institute of Pulmonology Pavlov State Medical University Clinical Center f. Inst, St. Petersburg; V. Yakusevich, Clinical Hospital of Emergency Care n.a. N.V. Solovyev, Yaroslavl; Spain: C. Valenzuela, Hospital Universitario de La Princesa, Madrid; O. Acosta, Hospital Universitario de Canarias, San Cristóbal de La Laguna; M. Martínez, Hospital Universitari i Politècnic La Fe, Valencia; L. Gómez Carrera, Hospital Universitario La Paz, Madrid; M. Molina-Molina, Hospital Universitari de Bellvitge, L'Hospitalet de Llobregat; D. M. Castillo Villegas, Hospital de la Santa Creu i Sant Pau, Barcelona; M. Aburto, Hospital de Galdakao, Galdakao; J.A. Rodríguez Portal, Hospital Universitario Vírgen del Rocío, Sevilla; A. Villar, Hospital Universitari Vall d'Hebron, Barcelona; A. León Jiménez, Hospital Puerta del Mar, Cádiz; J. Sauleda, Hospital Son Espases, Palma de Mallorca; M. Arias, Hospital Central de Asturias, Oviedo; United Kingdom: T. M. Maher, Royal Brompton Hospital, London; P. Beirne, St James's University Hospital, Leeds; H. Stone, Royal Stoke University Hospital, Stoke-on-Trent; B. Hope-Gill, University Hospital Llandough, Cardiff; N. Hirani, Royal Infirmary of Edinburgh, Edinburgh; N. Chaudhuri, Wythenshawe Hospital, Manchester; United States of America: C. Andrews, Diagnostics Research Group, San Antonio; A. Gifford, Dartmouth-Hitchcock Medical Center, Lebanon; L. Jones, University of Florida College of Medicine, Jacksonville; L. Morrison, Duke University Medical Center, Durham; D. Antin-Ozerkis, Yale University School of Medicine, New Haven; N. Bhatt, The Ohio State University Wexner Medical Center, Columbus; T. Kulkarni, University of Alabama at Birmingham Lung Health Center, Birmingham; T. Moua, Mayo Clinic, Rochester; N. Ettinger, The Lung Research Center, Chesterfield; L. Pitts, University of Kansas Medical Center, Kansas City; S. Veeraraghavan, The Emory Clinic, Atlanta; M. Padilla, Icahn School of Medicine at Mount Sinai - Mount Sinai Medical Center, New York; E. R. Fernández Pérez, National Jewish Health, Denver; G. Giessel, Pulmonary Associates of Richmond, Inc., Richmond; M. Strek, University of Chicago Medical Center, Chicago; S. Danoff, Johns Hopkins Asthma and Allergy Center, Baltimore; J. Burk, Texas Pulmonary and Critical Care Consultants, PA, Fort Worth; M. Rossman, The Hospital of the University of Pennsylvania, Philadelphia; N. Patel, Columbia University Medical Center-New York Presbyterian Hospital, New York; E. Belloli, Pulmonary and Critical Care Medicine, Ann Arbor; D. Hotchkin, The Oregon Clinic, Portland; S. Weigt, Peter Morton Medical Building, Los Angeles; M.B. Scholand, University of Utah Health, Salt Lake City; R. Kaner, New York-Presbyterian Hospital/ Weill Cornell Medical Center, New York; B. Sigal, Southeastern Research Center, Winston-Salem; Z. Safdar, Houston Methodist Hospital, Houston; L. Tolle, Cleveland Clinic, Cleveland; R. Martinez, Pulmonary and Sleep of Tampa Bay, Brandon; M. Glassberg, University of Miami Pulmonary Research Office, Miami; R. Hallowell, Beth Israel Deaconess Medical Center, Boston; J. Golden, University of California San Francisco, San Francisco; M. Schwartz, Pulmonary and Critical Care Associates of Baltimore,


Page 14 of 49


5

Towson; E. Britt, University of Maryland Medical Center, Baltimore; L. Morrow, Creighton University School of Medicine, Omaha; Y. Mageto, Baylor University Medical Center, Dallas; K. Buch and S. Chaaban, University of Kentucky Medical Center, Lexington; H. Poonyagariyagorn, The Oregon Clinic - Pulmonary, Critical Care and Sleep Medicine – West, Portland; D. Dilling, Loyola University Medical Center, Maywood; O. Shlobin, Inova Fairfax Medical Campus, Falls Church; K. Thavarajah, Henry Ford Health System, Detroit; A. Nambiar, Medical Arts and Research Center (MARC), San Antonio; I. Rosas, Brigham and Women's Hospital, Boston; R. Bascom, Penn State Milton S. Hershey Medical Center, Hershey; J. Oldham, UC Davis Medical Center, Sacramento; S. Schmidt, Spectrum Health, Grand Rapids; J. Dematte D'Amico, Northwestern Memorial Hospital, Chicago; J. Falk, Cedars-Sinai Medical Center, Los Angeles; C. Glazer, UT Southwestern Medical Center, Dallas; G. Criner, Temple University Hospital, Oaks.


Page 15 of 49


6

Section B: HRCT features

Eligible patients had fibrosing lung disease on HRCT, defined as reticular abnormality with

traction bronchiectasis, with or without honeycombing, with a disease extent of >10%.

The following co-existing features were accepted:

ground glass opacity

upper lung or peribronchovascular predominance

mosaic attenuation

air trapping

centrilobular nodules

The following co-existing features were not allowed:

widespread consolidation

progressive massive fibrosis

HRCT patterns were used to classify patients as having a UIP-like fibrotic pattern or other

fibrotic patterns (based on central review). Patients who met criteria A, B and C; A and C; or B

and C below were classified as having a UIP-like fibrotic pattern on HRCT. Patients who did not

meet these criteria were classified as having other fibrotic patterns on HRCT.

A Definite honeycomb lung destruction with basal and peripheral predominance.

B Presence of reticular abnormality and traction bronchiectasis consistent with fibrosis

with basal and peripheral predominance.

C Atypical features are absent, specifically nodules and consolidation.

Ground glass opacity, if present, is less extensive than reticular opacity pattern.


Page 16 of 49


7

Section C: Key exclusion criteria

Patients with alanine transaminase, aspartate transaminase, or total bilirubin >1.5 times the

upper limit of normal; forced expiratory volume in 1 second/forced vital capacity <0.7 (pre-

bronchodilator); creatinine clearance <30 mL/min calculated by Cockcroft–Gault formula;

chronic hepatic disease (Child-Pugh A, B or C); or bleeding risk (defined as genetic

predisposition to bleeding; requirement for fibrinolysis, full-dose therapeutic anticoagulation or

high-dose antiplatelet therapy; history of hemorrhagic central nervous system event within 12

months; hemoptysis or hematuria, active gastrointestinal bleeding or gastrointestinal ulcers, or

major injury or surgery, within 3 months; or international normalized ratio >2, prolongation of

prothrombin time and activated partial thromboplastin time by >1.5 time upper limit of normal) at

screening were excluded. Patients with significant pulmonary arterial hypertension (PAH)

(defined as previous clinical or echocardiographic evidence of significant right heart failure;

history of right heart catheterization showing a cardiac i

parenteral therapy with epoprostenol or treprostinil) were excluded. Patients were excluded if

history of myocardial infarction within 6 months, history of unstable angina within 6 months or

history of thrombotic event (including stroke and transient ischemic attack) within 12 months of

screening. Previous treatment with nintedanib or pirfenidone was not permitted. Treatment with

azathioprine, cyclosporine, mycophenolate mofetil, tacrolimus, oral glucocorticoids >20 mg/day,

or the combination of oral glucocorticoids, azathioprine and N-acetylcysteine within 4 weeks of

randomization was not permitted. Treatment with cyclophosphamide within 8 weeks of

randomization or rituximab within 6 months of randomization was not permitted. Women who

were pregnant, nursing, or who planned to become pregnant while in the trial, and women of

childbearing potential not willing or able to use highly effective methods of birth control for 28

days prior to and 3 months after nintedanib administration were excluded. Patients with a life


Page 17 of 49


8

expectancy <2.5 years for a disease other than ILD (in the opinion of the investigator) were

excluded.


Page 18 of 49


9

Section D: Trial design

Figure S1A: Overall trial design

Figure S1B: Trial design on patient level


Page 19 of 49


10

Section E: Sensitivity analyses of the primary endpoint related to handling missing data

Table S1. Sensitivity analyses for the primary endpoint based on multiple imputation

Analysis

Missing FVC data at week 52 in patients still alive at week 52

Patients who died before week 52

Handling of missing data at week 52

Underlying assumption regarding persistence of efficacy post-withdrawal

Handling of missing data at week 52

Underlying assumption regarding persistence of

efficacy after death

Primary

No imputation Assumes missing at

random

No imputation Assumes missing at

random

Sensitivity 1 Data imputed using the

slope estimates for

nintedanib- and placebo-

treated patients from

pattern 2

Assumes the rate of decline

in patients with missing

data at week 52 is similar to

the rate of decline in the

respective treatment group

in patients from pattern 2

Data imputed using the

same slope estimates for

placebo-treated patients

from pattern 2, but

truncated to force the

slope in patients who

died to be more severe

than in those who

survived

Assumes the rate of

decline in patients who

died before week 52 is

similar to or worse than

the rate of decline in


who prematurely

discontinued Sensitivity 2 Data imputed using the

slope estimates from


from pattern 2

Assumes the rate of decline

in all patients with missing




from pattern 2

Sensitivity 3 Data imputed using the

slope estimates for

Assumes rate of decline in

all patients with missing

Data imputed using the

slope estimates for all

Assumes the rate of

decline in patients who


Page 20 of 49


11


from patterns 1 or 2


the rate of decline in all


placebo-treated patients,

but truncated to force the

slope in patients who

died to be more severe

than in those who

survived

died before week 52 is

similar to or worse than



Pattern 1: Patients with an FVC value at week 52 who received trial medication until week 52. Pattern 2: Patients with an FVC value at week 52 who prematurely discontinued trial medication before week 52. Slope: Annual rate of decline in FVC (mL/year).


Page 21 of 49


12

Figure S2A. Sensitivity analyses of the primary endpoint related to handling missing data in the overall population

Figure S2B. Sensitivity analyses of the primary endpoint related to handling missing data in patients with a UIP-like fibrotic pattern on HRCT


Page 22 of 49


13

Section F: Tipping point analysis of the primary endpoint

A tipping point analysis was conducted to assess how large deviations from the assumption that

missing data were missing at random could be to impact the conclusions of the primary

analysis, i.e., at what point the result would change from being statistically significant to not

statistically significant. As a first step, non-monotone missing data were imputed from the

collected data using Markov Chain Monte Carlo (MCMC) to generate sets of longitudinal FVC

(mL) data with a monotone missingness pattern. Such a pattern is the pre-requisite for applying

sequential imputation and means that once a patient had a missing FVC value at a particular

time point, FVC values at all subsequent time points were also missing. The tipping point

analysis for the longitudinal FVC data was based on the Multiple Delta Adjustment Method.

Data were imputed using sequential regression, and a delta adjustment, dependent on a shift

parameter and the weeks between visits, was added to each imputed value. The selection of

shift values was based on considerations of plausible and implausible cumulative changes of

FVC (mL) values at week 52 in the INBUILD trial, as well as observations made based on the

INPULSIS trials, given the possible range of missing data patterns. The selected range of shift

parameters ranged between -20mL/week and +10mL/week, so a patient with only one missing

FVC value at week 52 would have an adjustment of -320mL when the shift parameter was set to

-20ml/week, up to an adjustment of +160mL when the shift parameter was set to +10mL/week

for the imputed value at week 52. Depending on the number of monotone missing assessments,

this covered a median adjustment in imputed FVC (mL) at week 52 on a patient level of -453 mL

to +227 mL. For patients with more than 1 monotone missing assessment, the effect of the

adjustments was cumulative.


Page 23 of 49


14

Figure S3A. Tipping point analysis of the primary endpoint in the overall population


Page 24 of 49


15

Figure S3B: Tipping point analysis of the primary endpoint in patients with a UIP-like

fibrotic pattern on HRCT


Page 25 of 49


16

Section G: Statistical analyses of endpoints other than the primary endpoint

Analysis of the primary endpoint in subgroups based on HRCT pattern was based on

a random coefficient regression model (with random slopes and intercepts) including baseline

FVC (mL) and treatment-by-subgroup and treatment-by-subgroup-by-time interactions. The P

value for treatment-by-subgroup-by-time interaction was obtained from a test of heterogeneity

across all expression levels of the subgrouping, with no adjustment for multiple testing.

Time to first acute exacerbation of ILD or death over 52 weeks and time to death over 52 weeks

were analyzed using a stratified log-rank test (stratified by HRCT pattern: UIP-like fibrotic

pattern or other fibrotic patterns). A Cox proportional hazards model stratified by the same

factors as the log-rank test was used to derive the hazard ratio and a 95% CI. HRCT pattern

was not included as a covariate in the analysis of these endpoints in patients with a UIP-like

fibrotic pattern on HRCT. Breslow’s method for handling ties was used. The same analyses

were used in the assessment of these endpoints using all data up to first database lock.

Absolute change from baseline in K-BILD questionnaire total score at week 52 in the overall

population was analyzed using a restricted maximum likelihood-based repeated measures

approach with fixed effects for baseline, HRCT pattern (UIP-like fibrotic pattern or other fibrotic

patterns), visit, treatment-by-visit interaction, baseline-by-visit interaction and random effect for

patient as covariates. HRCT pattern was not included as a covariate in the analysis of this

endpoint in patients with a UIP-like fibrotic pattern on HRCT.


Page 26 of 49


17

Section H: Clinical ILD diagnoses

Table S2: Clinical ILD diagnoses (grouped) in the overall population

Nintedanib (n=332)

Placebo (n=331)

Hypersensitivity pneumonitis 84 (25.3) 89 (26.9)

Autoimmune ILDs 82 (24.7) 88 (26.6)

Rheumatoid arthritis-associated ILD 42 (12.7) 47 (14.2)

Systemic sclerosis-associated ILD 23 (6.9) 16 (4.8)

Mixed connective tissue disease-

associated ILD

7 (2.1) 12 (3.6)

Other autoimmune ILDs 10 (3.0) 13 (3.9)

Idiopathic non-specific interstitial pneumonia 64 (19.3) 61 (18.4)

Unclassifiable idiopathic interstitial

pneumonia

64 (19.3) 50 (15.1)

Other ILDs* 38 (11.4) 43 (13.0) Data are no (%) of patients.

*Included sarcoidosis, exposure-related ILDs and selected other terms in “Other fibrosing ILDs”.


Page 27 of 49


18

Section I: Baseline characteristics in subgroups by fibrotic pattern on HRCT

Table S3: Baseline characteristics of patients with a UIP-like fibrotic pattern and other

fibrotic patterns on HRCT*

Patients with a UIP-like fibrotic pattern on HRCT

Patients with other fibrotic patterns on

HRCT Nintedanib

(n=206) Placebo (n=206)

Nintedanib (n=126)

Placebo (n=125)

Male – no. (%) 120 (58.3) 127 (61.7) 59 (46.8) 50 (40.0)

Age – yr 67.5±8.1 68.5±8.7 61.6±10.9 62.6±10.4

Former or current smoker – no. (%) 118 (57.3) 118 (57.3) 51 (40.5) 51 (40.8)

Criteria for ILD progression in 24

months before screening (grouped)

– no. (%)

Relative decline

predicted

100 (48.5) 98 (47.6) 60 (47.6) 74 (59.2)

Relative decline –<10%

predicted combined with

worsening of respiratory

symptoms and/or increased

extent of fibrotic changes on

HRCT

76 (36.9) 68 (33.0) 34 (27.0) 29 (23.2)

Worsened respiratory symptoms

and increased extent of fibrotic

changes on HRCT only

30 (14.6) 39 (18.9) 32 (25.4) 22 (17.6)

Clinical ILD diagnosis (grouped) –

no. (%)

Hypersensitivity pneumonitis 44 (21.4) 46 (22.3) 40 (31.7) 43 (34.4)

Autoimmune ILDs 62 (30.1) 65 (31.6) 20 (15.9) 23 (18.4)

Rheumatoid arthritis-

associated ILD

36 (17.5) 41 (19.9) 6 (4.8) 6 (4.8)


Page 28 of 49


19

Systemic sclerosis-associated

ILD

17 (8.3) 7 (3.4) 6 (4.8) 9 (7.2)

Mixed connective tissue

disease-associated ILD

4 (1.9) 8 (3.9) 3 (2.4) 4 (3.2)

Other autoimmune ILDs 5 (2.4) 9 (4.4) 5 (4.0) 4 (3.2)

Idiopathic non-specific interstitial

pneumonia

34 (16.5) 37 (18.0) 30 (23.8) 24 (19.2)

Unclassifiable idiopathic

interstitial pneumonia

43 (20.9) 34 (16.5) 21 (16.7) 16 (12.8)

Other ILDs† 23 (11.2) 24 (11.7) 15 (11.9) 19 (15.2)

FVC

mL 2363±763 2374±720 2302±703 2235±736

% of predicted value 70.6±17.0 70.6±14.7 65.6±13.8 67.2±15.8

DLco, mmol/min/kpa† 3.5±1.2 3.7±1.3 3.6±1.2 3.7±1.2

DLco, % of predicted value‡ 44.6±12.1 48.5±15.9 44.0±11.6 46.7±13.2

K-BILD questionnaire total score 53.1±10.8 53.1±9.4 51.4±11.3 51.1±10.5

*Plus–minus values are means ± SD. †Included sarcoidosis, exposure-related ILDs and selected other

terms in “Other fibrosing ILDs”. ‡The DLco value was corrected for the hemoglobin level.

DLco denotes diffusion capacity of the lungs for carbon monoxide, FVC forced vital capacity, HRCT high-

resolution computed tomography, K-BILD King’s Brief Interstitial Lung Disease, UIP usual interstitial

pneumonia. Not all patients provided data for all variables.


Page 29 of 49


20

Section J. Restricted therapies taken at baseline and initiated between first and last trial drug intake over 52 weeks.

Table S4. Restricted therapies taken at baseline

Overall population Patients with UIP-like fibrotic pattern on HRCT

Nintedanib (n=332)

Placebo (n=331)

Nintedanib (n=206)

Placebo (n=206)

restricted therapy at baseline

57 (17.2) 59 (17.8) 42 (20.4) 44 (21.4)

Biologic disease-modifying antirheumatic drugs

14 (4.2) 17 (5.1) 14 (6.8) 13 (6.3)

Denosumab 3 (0.9) 8 (2.4) 3 (1.5) 5 (2.4) Abatacept 3 (0.9) 3 (0.9) 3 (1.5) 2 (1.0) Etanercept 3 (0.9) 1 (0.3) 3 (1.5) 1 (0.5) Tocilizumab 2 (0.6) 2 (0.6) 2 (1.0) 2 (1.0) Adalimumab 2 (0.6) 1 (0.3) 2 (1.0) 1 (0.5) Infliximab 0 (0.0) 2 (0.6) 0 (0.0) 2 (1.0) Rituximab 1 (0.3) 0 (0.0) 1 (0.5) 0 (0.0) Ascorbic acid; collagen 0 (0.0) 1 (0.3) - - Other† 0 (0.0) 1 (0.3) 0 (0.0) 1 (0.5)

Glucocorticoids‡ 3 (0.9) 5 (1.5) 1 (0.5) 2 (1.0) Meprednisone 1 (0.3) 2 (0.6) 1 (0.5) 2 (1.0) Prednisone 2 (0.6) 2 (0.6) - - Prednisolone 0 (0.0) 1 (0.3) - -

Immunomodulatory medications for ILD 3 (0.9) 4 (1.2) 3 (1.5) 4 (1.9) Mycophenolate mofetil 2 (0.6) 1 (0.3) 2 (1.0) 1 (0.5) Ciclosporin 0 (0.0) 2 (0.6) 0 (0.0) 2 (1.0) Rituximab 1 (0.3) 0 (0.0) 1 (0.5) 0 (0.0) Tacrolimus 0 (0.0) 1 (0.3) 0 (0.0) 1 (0.5)

Non-biologic disease-modifying antirheumatic drugs

35 (10.5) 42 (12.7) 27 (13.1) 34 (16.5)

Hydroxychloroquine 13 (3.9) 9 (2.7) 10 (4.9) 8 (3.9) Leflunomide 10 (3.0) 8 (2.4) 8 (3.9) 8 (3.9) Methotrexate 5 (1.5) 10 (3.0) 4 (1.9) 6 (2.9)


Page 30 of 49


21

Sulfasalazine 5 (1.5) 5 (1.5) 4 (1.9) 5 (2.4) Hydroxychloroquine sulfate 5 (1.5) 3 (0.9) 4 (1.9) 2 (1.0) Methotrexate sodium 1 (0.3) 3 (0.9) 1 (0.5) 3 (1.5) Mycophenolate mofetil 2 (0.6) 1 (0.3) 2 (1.0) 1 (0.5) Ciclosporin 0 (0.0) 2 (0.6) 0 (0.0) 2 (1.0) Doxycycline 0 (0.0) 2 (0.6) 0 (0.0) 2 (1.0) Chloroquine phosphate 0 (0.0) 2 (0.6) 0 (0.0) 1 (0.5) Penicillamine 0 (0.0) 2 (0.6) 0 (0.0) 1 (0.5) Bucillamine 0 (0.0) 1 (0.3) 0 (0.0) 1 (0.5) Iguratimod 0 (0.0) 1 (0.3) 0 (0.0) 1 (0.5) Tacrolimus 0 (0.0) 1 (0.3) 0 (0.0) 1 (0.5) Minocycline hydrochloride 1 (0.3) 0 (0.0) - -

Data are n (%) of patients. Medications based on customized drug grouping category. *Only included in customized drug grouping if high dose. †Ascorbic acid; boswellia serrata resin; chondroitin sulfate; collagen; glucosamine hydrochloride; hyaluronic acid; manganese sulfate; methylsulfonylmethane; sodium; sodium borate. ‡Only included in customized drug grouping if high dose and route of administration was oral, intravenous, or intramuscular.


Page 31 of 49


22

Table S5. Restricted therapies initiated between first and last trial drug intake over 52 weeks

Overall population Patients with UIP-like fibrotic pattern on HRCT

Nintedanib (n=332)

Placebo (n=331)

Nintedanib (n=206)

Placebo (n=206)

weeks 36 (10.8) 70 (21.1) 21 (10.2) 43 (20.9)

Biologic disease-modifying antirheumatic drugs

2 (0.6) 2 (0.6) 0 (0.0) 1 (0.5)

Rituximab 2 (0.6) 2 (0.6) 0 (0.0) 1 (0.5) Glucocorticoids* 33 (9.9) 57 (17.2) 19 (9.2) 36 (17.5)

Prednisone 20 (6.0) 27 (8.2) 10 (4.9) 20 (9.7) Prednisolone 8 (2.4) 15 (4.5) 4 (1.9) 9 (4.4) Methylprednisolone sodium succinate 5 (1.5) 14 (4.2) 3 (1.5) 10 (4.9) Methylprednisolone 4 (1.2) 9 (2.7) 2 (1.0) 5 (2.4) Hydrocortisone 1 (0.3) 2 (0.6) 1 (0.5) 1 (0.5) Meprednisone 1 (0.3) 1 (0.3) 1 (0.5) 0 (0.0) Steroids 2 (0.6) 0 (0.0) 1 (0.5) 0 (0.0) Betamethasone sodium phosphate 1 (0.3) 0 (0.0) 1 (0.5) 0 (0.0) Deflazacort 1 (0.3) 0 (0.0) - - Dexamethasone 0 (0.0) 1 (0.3) - - Dexamethasone sodium phosphate 0 (0.0) 1 (0.3) - -

Immunomodulatory medications for ILD 9 (2.7) 21 (6.3) 3 (1.5) 13 (6.3) Mycophenolate Mofetil 3 (0.9) 7 (2.1) 0 (0.0) 3 (1.5) Azathioprine 1 (0.3) 5 (1.5) 1 (0.5) 2 (1.0) Tacrolimus 3 (0.9) 3 (0.9) 2 (1.0) 3 (1.5) Ciclosporin 0 (0.0) 4 (1.2) 0 (0.0) 4 (1.9) Rituximab 2 (0.6) 2 (0.6) 0 (0.0) 1 (0.5) Cyclophosphamide 0 (0.0) 2 (0.6) 0 (0.0) 2 (1.0)

Non-biologic disease-modifying antirheumatic drugs

7 (2.1) 19 (5.7) 3 (1.5) 12 (5.8)

Mycophenolate mofetil 3 (0.9) 7 (2.1) 0 (0.0) 3 (1.5) Tacrolimus 3 (0.9) 3 (0.9) 2 (1.0) 3 (1.5) Azathioprine 1 (0.3) 5 (1.5) 1 (0.5) 2 (1.0)


Page 32 of 49


23

Ciclosporin 0 (0.0) 4 (1.2) 0 (0.0) 4 (1.9) Cyclophosphamide 0 (0.0) 2 (0.6) 0 (0.0) 2 (1.0)

Data are n (%) of patients. Medications based on customized drug grouping category. *Only included in customized drug grouping if high dose and route of administration was oral, intravenous, or intramuscular.


Page 33 of 49


24

Section K: Between-group adjusted difference in the annual rate of decline in FVC

(mL/year) over 52 weeks (primary endpoint)

Figure S4A. Between-group adjusted difference in the annual rate of decline in FVC

(mL/year) over 52 weeks in the overall population (primary endpoint). The bars indicate the

standard error.


Page 34 of 49


25

Figure S4B. Between-group adjusted difference in the annual rate of decline in FVC

(mL/year) over 52 weeks in patients with a UIP-like fibrotic pattern on HRCT (primary

endpoint). The bars indicate the standard error.


Page 35 of 49


26

Section L: Box and whisker plots of the adjusted annual rates of decline in FVC over 52

weeks. Based on rates calculated from the primary model for each patient. Q1 denotes the first

quartile, and Q3 the third quartile. The whiskers are drawn from the box to the most extreme

point that is less than or equal to 1.5 times the interquartile range.

Figure S5. Box and whisker plot of the adjusted annual rates of decline in FVC over 52

weeks in the overall population


Page 36 of 49


27

Figure S6. Box and whisker plot of the adjusted annual rates of decline in FVC over 52

weeks in patients with a UIP-like fibrotic pattern on HRCT


Page 37 of 49


28

Section M. Annual rate of decline in FVC (mL/year) over 52 weeks in subgroups by HRCT

pattern.

Figure S7. Between-group adjusted difference in the annual rate of decline in FVC

(mL/year) over 52 weeks in subgroups by HRCT pattern


Page 38 of 49


29

Section N: Time to first acute exacerbation of ILD or death and time to death using data

up to first database lock


Page 39 of 49


30

Figure S8: Kaplan-Meier estimate of time to first acute exacerbation of ILD or death in the

overall population using data up to first database lock


Page 40 of 49


31

Figure S9: Kaplan-Meier estimate of time to first acute exacerbation of ILD or death in patients with a UIP-like fibrotic pattern on HRCT using data up to first database lock


Page 41 of 49


32

Figure S10: Kaplan-Meier estimate of time to death in the overall population using data

up to first database lock


Page 42 of 49


33

Figure S11: Kaplan-Meier estimate of time to death in patients with a UIP-like fibrotic pattern on HRCT using data up to first database lock


Page 43 of 49


34

Section O: Adverse events reported up to first database lock

Table S3. Adverse events in the overall population using data up to first database lock

Event Nintedanib

(n=332) Placebo (n=331)

no. of patients (%)

Any adverse event 325 (97.9) 306 (92.4)

Any adverse event except progression of ILD† 324 (97.6) 305 (92.1)


Diarrhea 232 (69.9) 84 (25.4)

Nausea 100 (30.1) 33 (10.0)

Vomiting 63 (19.0) 17 (5.1)

Nasopharyngitis 53 (16.0) 48 (14.5)

Decreased appetite 53 (16.0) 21 (6.3)

Alanine aminotransferase increased 48 (14.5) 12 (3.6)

Dyspnea 48 (14.5) 54 (16.3)

Bronchitis 46 (13.9) 57 (17.2)

Weight decreased 43 (13.0) 12 (3.6)

Aspartate aminotransferase increased 43 (13.0) 12 (3.6)

Headache 38 (11.4) 26 (7.9)

Fatigue 34 (10.2) 21 (6.3)

Abdominal pain 34 (10.2) 9 (2.7)

Cough 37 (11.1) 50 (15.1)

Progression of ILD† 24 (7.2) 52 (15.7)

Severe adverse event§ 81 (24.4) 97 (29.3)

Serious adverse event¶ 138 (41.6) 154 (46.5)

Fatal adverse event 15 (4.5) 30 (9.1)

Fatal adverse event except progression of ILD† 14 (4.2) 26 (7.9)

Adverse event leading to treatment discontinuation 70 (21.1) 45 (13.6) *Adverse events with onset between first trial drug intake and last drug intake plus 28 days up to first database lock. Mean (±SD) duration of exposure up to first database lock was 15.0±6.8 months in the nintedanib group and 16.2±5.5 months in the placebo such adverse event. †“Progression of ILD” was based on the preferred term “interstitial lung disease” in the Medical Dictionary for Regulatory Activities (MedDRA). ‡Adverse events, coded using preferred terms in the MedDRA, reported in >10% of patients in either treatment group. §Adverse event that was incapacitating or that caused an inability to work or to perform usual activities. ¶Adverse event that


Page 44 of 49


35

resulted in death, was life-threatening, resulted in hospitalization or prolongation of hospitalization, resulted in persistent or clinically significant disability or incapacity, was a congenital anomaly or birth defect, or was deemed to be serious for any other reason. ILD denotes interstitial lung disease.


Page 45 of 49


36

Section P. Liver enzyme and bilirubin elevations

Table S4: Liver enzyme and bilirubin elevations in the overall population

Over 52 weeks Up to first database lock Nintedanib

(n=332) Placebo (n=331)

Nintedanib (n=332)

Placebo (n=331)

Alanine aminotransferase

40 (12.0) 5 (1.5) 44 (13.3) 5 (1.5)

4 (1.2) 1 (0.3) 8 (2.4) 1 (0.3)

1 (0.3) 0 (0.0) 2 (0.6) 0 (0.0)

Aspartate aminotransferase

18 (5.4) 5 (1.5) 22 (6.6) 5 (1.5)

7 (2.1) 1 (0.3) 9 (2.7) 1 (0.3)

1 (0.3) 1 (0.3) 1 (0.3) 1 (0.3)

Alanine aminotransferase and/or aspartate aminotransferase

43 (13.0) 6 (1.8) 47 (14.2) 6 (1.8)

10 (3.0) 1 (0.3) 14 (4.2) 1 (0.3)

2 (0.6) 1 (0.3) 3 (0.9) 1 (0.3)

Total bilirubin

limit of normal 3 (0.9) 5 (1.5) 3 (0.9) 6 (1.8)

1 (0.3) 1 (0.3) 1 (0.3) 1 (0.3)

Alkaline phosphatase

17 (5.1) 5 (1.5) 17 (5.1) 6 (1.8)

8 (2.4) 2 (0.6) 8 (2.4) 3 (0.9)


Page 46 of 49


37

Data are no (%) of patients based on worst values over 52 weeks (or until 28 days after last trial drug intake in patients who discontinued trial drug

before week 52) or between first trial drug intake and last drug intake plus 28 days up to first database lock.


Page 47 of 49


38

Section Q. Adverse events in subgroups by fibrotic pattern on HRCT

Table S5. Adverse events in patients with a UIP-like fibrotic pattern and other fibrotic patterns on HRCT*

Patients with a UIP-like fibrotic pattern on HRCT

Patients with other fibrotic patterns on HRCT

Event Nintedanib (n=206)

Placebo (n=206)

Nintedanib (n=126)

Placebo (n=125)

no. of patients (%)

Any adverse event 198 (96.1) 184 (89.3) 119 (94.4) 112 (89.6)

Any adverse event except progression of ILD† 198 (96.1) 183 (88.8) 119 (94.4) 112 (89.6)


Diarrhea 143 (69.4) 51 (24.8) 79 (62.7) 28 (22.4)

Nausea 57 (27.7) 16 (7.8) 39 (31.0) 15 (12.0)

Bronchitis 27 (13.1) 29 (14.1) 14 (11.1) 18 (14.4)

Nasopharyngitis 31 (15.0) 28 (13.6) 13 (10.3) 12 (9.6)

Dyspnea 25 (12.1) 25 (12.1) 11 (8.7) 19 (15.2)

Vomiting 34 (16.5) 13 (6.3) 27 (21.4) 4 (3.2)

Cough 22 (10.7) 21 (10.2) 11 (8.7) 23 (18.4)

Decreased appetite 31 (15.0) 13 (6.3) 17 (13.5) 4 (3.2)

Headache 19 (9.2) 14 (6.8) 16 (12.7) 9 (7.2)

Progression of ILD† 12 (5.8) 28 (13.6) 4 (3.2) 11 (8.8)

Alanine aminotransferase increased 28 (13.6) 4 (1.9) 15 (11.9) 8 (8.4)

Fatigue 21 (10.2) 11 (5.3) 12 (9.5) 9 (7.2)

Weight decreased 30 (14.6) 6 (2.9) 11 (8.7) 5 (4.0)


Page 48 of 49


39

Aspartate aminotransferase increased 22 (10.7) 5 (2.4) 16 (12.7) 7 (5.6)

Upper respiratory tract infection 10 (4.9) 11 (5.3) 14 (11.1) 8 (6.4)

Abdominal pain 20 (9.7) 3 (1.5) 14 (11.1) 5 (4.0)

Severe adverse event§ 37 (18.0) 50 (24.3) 23 (18.3) 23 (18.4)

Serious adverse event¶ 63 (30.6) 77 (37.4) 44 (34.9) 33 (26.4)

Fatal adverse event 7 (3.4) 16 (7.8) 4 (3.2) 1 (0.8)

Fatal adverse event except progression of ILD† 6 (2.9) 13 (6.3) 4 (3.2) 1 (0.8)

Adverse event leading to treatment discontinuation 45 (21.8) 22 (10.7) 20 (15.9) 12 (9.6) *Adverse events reported over 52 weeks (or until 28 days after last trial drug intake in patients who discontinued trial drug before week 52). Data

†“Progression of ILD” was based on the preferred term “interstitial lung disease” in the Medical Dictionary for Regulatory Activities (MedDRA). ‡Adverse events, coded using preferred terms in the MedDRA, reported in >10% of patients in any of the subgroups shown. §Adverse event that was incapacitating or that caused an inability to work or to perform usual activities. ¶Adverse event that resulted in death, was life-threatening, resulted in hospitalization or prolongation of hospitalization, resulted in persistent or clinically significant disability or incapacity, was a congenital anomaly or birth defect, or was deemed to be serious for any other reason.


Page 49 of 49


Documents

Nintedanib in Progressive Fibrosing Interstitial Lung Diseases...diffusing capacity of the lung for carbon monoxide ranging from 30 to less than 80% of the predicted value. Randomization