Scleroderma, Systemic
Conditions
Brief summary
Systemic Sclerosis (SSc) is a devastating disease of unknown etiology. Patients suffer from multiple organ fibrosis whereas lung fibrosis (interstitial lung disease, ILD) is one of the main driver for mortality. There is preclinical evidence for efficacy of nintedanib in SSc and associated ILD (SSc-ILD) and the anti-fibrotic efficacy of nintedanib was proven in idiopathic pulmonary fibrosis patients, who are presenting a similar pattern regarding lung fibrosis. Hence it is the purpose of the trial to confirm the efficacy and safety of nintedanib 150 mg bid in treating patients with SSc-ILD, compared with placebo. The trial will be conducted as a double blind, randomised, placebo-controlled trial with primary efficacy evaluation at week 52 and placebo-controlled treatment until last patient out (up to a maximum of 100 weeks). Respiratory function is globally accepted for assessment of treatment effects in patients with lung fibrosis. The chosen endpoint (Forced Vital Capacity (FVC) decline) is easy to obtain and is part of the usual examinations done in patients with SSc-ILD.
Interventions
DRUGNintedanib
DRUGPlacebo
Sponsors
Boehringer Ingelheim
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
DOUBLE (Subject, Investigator)
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Age \>= 18 years * 2013 American College of Rheumatology (ACR) / EULAR classification criteria for SSc fulfilled * SSc disease onset (defined by first non-Raynaud symptom) within 7 years * SSc related Interstitial Lung Disease confirmed by High Resolution Computer Tomography (HRCT); Extent of fibrotic disease in the lung \>= 10% * FVC \>= 40% of predicted normal * Carbon Monoxide Diffusion Capacity (DLCO) 30% to 89% of predicted normal
Exclusion criteria
* Aspartate Aminotransferase (AST), Alanine Aminotransferase (ALT) \>1.5 x ULN * Bilirubin \>1.5 x ULN * Creatinine clearance \<30 mL/min * Airway obstruction (pre-bronchodilator Forced Expiratory Volume in 1 second (FEV1)/FVC \<0.7) * Other clinically significant pulmonary abnormalities * Significant Pulmonary Hypertension (PH) * Cardiovascular diseases * More than 3 digital fingertip ulcers or a history of severe digital necrosis requiring hospitalization or severe other ulcers * Bleeding risk (such as predisposition to bleeding, fibrinolysis, full-dose anticoagulation, high dose antiplatelet therapy, history of hemorrhagic central nervous system (CNS) event within last year * international normalised ratio (INR) \>2, prolongation of prothrombin time (PT) and partial thromboplastin time (PTT) by \>1.5 x ULN) * History of thrombotic event within last year * Clinical signs of malabsorption or needing parenteral nutrition * Previous treatment with nintedanib or pirfenidone * Treatment with prednisone \>10 mg/day, azathioprine, hydroxychloroquine, colchicine, D-penicillamine, sulfasalazine, cyclophosphamide, rituximab, tocilizumab, abatacept, leflunomide, tacrolimus, newer anti-arthritic treatments like tofacitinib and cyclosporine A, potassium para-aminobenzoate * Unstable background therapy with either mycophenolate mofetil or methotrexate * Previous or planned hematopoietic stem cell transplantation * Patients with underlying chronic liver disease (Child Pugh A, B, C hepatic impairment)
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Annual Rate of Decline in Forced Vital Capacity (FVC) Over 52 Weeks | up to week (wk) 52 after the start of administration | Forced vital capacity (FVC) is the total amount of air exhaled during the lung function test. For this endpoint reported means represent the adjusted rate. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Absolute Change From Baseline in Saint George's Respiratory Questionnaire (SGRQ) Total Score at Week 52. | Baseline and up to 52 weeks after the start of administration | This is the second key secondary endpoint. The Saint George's Respiratory Questionnaire measures the health status in patients with chronic airflow limitation. It consists of 2 parts that cover 3 domains: symptoms, activities, and impacts. The symptom domain relates to the effect, frequency and severity of respiratory symptoms. The activity domain relates to activities that cause or are limited by breathlessness. The impact domain evaluates a range of aspects concerned with social functioning and psychological disturbances resulting from airways disease. The scores of these domains range from 0 (no impairment) to 100 (worst possible). The calculated total score summarises the impact of the disease on overall health status. A high score corresponds to worse health. Least square mean is actually the adjusted mean. Adjusted mean was based on all analysed patients in the model (not only patients with a baseline and measurement at Week 52). |
| Annual Rate of Decline in FVC in Percentage (%) Predicted Over 52 Weeks | up to 52 weeks after the start of administration | Annual rate of decline in FVC in percentage (%) predicted over 52 weeks. For this endpoint reported means represent the adjusted rate. |
| Absolute Change From Baseline in FVC in mL at Week 52 | Baseline and up to 52 weeks after the start of administration | Absolute change from baseline in FVC in mL at Week 52. Least square mean is actually the adjusted mean. Adjusted mean was based on all analysed patients in the model (not only patients with a baseline and measurement at Week 52). |
| Relative Change From Baseline [%] of mRSS at Week 52 | Baseline and up to 52 weeks after the start of administration | Relative change from baseline \[%\] of mRSS at Week 52. The modified Rodnan Skin Score (mRSS) is an evaluation of the patient's skin thickness rated by clinical palpation using a 0 to 3 scale. The scale differentiates between 0 = normal skin, 1 = mild thickness, 2 = moderate thickness, and 3 = severe thickness with inability to pinch the skin into a fold. The palpation is done for each of the 17 surface anatomic areas of the body: face, anterior chest, abdomen, fingers (right and left separately), forearms, upper arms, thighs, lower legs, dorsum of hands and feet. The sum of these individual values is defined as the total skin score. The mRSS has a range from 0 (no thickening) to 51 (severe thickening in all 17 areas). A high score corresponds to worse skin thickness. Least square mean is actually the adjusted mean. Adjusted mean was based on all analysed patients in the model (not only patients with a baseline and measurement at Week 52). |
| Time to Death | From date of first trial drug intake up to date of death or last contact date (ie., up to 100 weeks) | Time to event analysis of patients with death. The number of observed patients with death are reported. |
| Absolute Change From Baseline in the Modified Rodnan Skin Score (mRSS) at Week 52 | Baseline and up to 52 weeks after the start of administration | This is the first key secondary endpoint. The modified Rodnan Skin Score (mRSS) is an evaluation of the patient's skin thickness rated by clinical palpation using a 0 to 3 scale. The scale differentiates between 0 = normal skin, 1 = mild thickness, 2 = moderate thickness, and 3 = severe thickness with inability to pinch the skin into a fold. The palpation is done for each of the 17 surface anatomic areas of the body: face, anterior chest, abdomen, fingers (right and left separately), forearms, upper arms, thighs, lower legs, dorsum of hands and feet. The sum of these individual values is defined as the total skin score. The mRSS has a range from 0 (no thickening) to 51 (severe thickening in all 17 areas). A high score corresponds to worse skin thickness. Least square mean is actually the adjusted mean. Adjusted mean was based on all analysed patients in the model (not only patients with a baseline and measurement at Week 52). |
| Absolute Change From Baseline in Carbon Monoxide Diffusion Capacity (DLco) in % Predicted at Week 52 | Baseline and up to 52 weeks after the start of administration | Absolute change from baseline in Carbon Monoxide Diffusion Capacity (DLco) in % predicted at Week 52. Least square mean is actually the adjusted mean. Adjusted mean is based on all analysed patients in the model (not only patients with a baseline and measurement at week 52). |
| Absolute Change From Baseline in Digital Ulcer Net Burden at Week 52 | Baseline and up to 52 weeks after the start of administration | Absolute change from baseline in digital ulcer net burden (defined as the number of new digital ulcers (DUs) plus the number of DUs that have been verified at any earlier assessment during the trial) at Week 52. It is calculated at a visit by counting the total number of fingertips with ulcers (i.e. number of fingers with presence of digital ulcer ticked Yes) at the corresponding visit Least square mean is actually the adjusted mean. Adjusted mean is based on all analysed patients in the model (not only patients with a baseline and measurement at week 52). |
| Absolute Change From Baseline in Health Assessment Questionnaire-Disability Index (HAQ-DI) Score at Week 52 | Baseline and up to 52 weeks after the start of administration | Absolute change from baseline in Health Assessment Questionnaire-Disability Index (HAQ-DI) score at Week 52. The HAQ-DI score is calculated as follows: Each question is scored 0-3 (where 0= without difficulty & 3= unable to do). There are 8 categories (Dressing & Grooming, Arising, Eating, Walking, Hygiene, Reach, Grip, Activities), each including 2 or 3 questions. The score for each category corresponds to maximum question score within each category. Finally, HAQ-DI score corresponds to sum of the sub-scores of all 8 categories divided by number of categories completed. Please note that if there are fewer than 6 categories with responses, then a score cannot be calculated. The HAQ-DI score scale has 25 possible values (i.e., 0, 0.125, 0.250, 0.375 … 3). A high score corresponds to worse impairment. Least square mean is actually the adjusted mean. Adjusted mean is based on all analysed patients in the model (not only patients with a baseline and measurement at week 52). |
| Absolute Change From Baseline in Functional Assessment of Chronic Illness Therapy (FACIT) Dyspnoea Score at Week 52 | Baseline and up to 52 weeks after the start of administration | Absolute change from baseline in Functional Assessment of Chronic Illness Therapy (FACIT) dyspnoea score at Week 52. FACIT-Dyspnoea (Dyspnoea) 10 Item Short Form include a 4-point rating scale (no shortness of breath=0; mildly short of breath=1; moderately short of breath = 2; severely short of breath =3; or I did not do this in the past 7 days =4). A raw score is calculated as: Sum individual item scores \* 10 / number of items answered. Raw scores are then converted to scale scores using the table included in the FACIT Dyspnoea Scale Short Form Scoring Guideline. FACIT dyspnea scale score ranges between 0 and 75.9. The FACIT-Dyspnea short forms are scored such that a high score represents high levels of dyspnea. Least square mean is actually the adjusted mean. Adjusted mean is based on all analysed patients in the model (not only patients with a baseline and measurement at week 52). |
| The Percentage (%) of Responder Based on Combined Response Index in Systemic Sclerosis (CRISS) at Week 52 | Week 52 | The percentage (%) of responder based on Combined Response Index in Systemic Sclerosis (CRISS) at Week 52. This is a composite endpoint, based on the mRSS, FVC percent predicted, HAQ-DI, patient's global impression of overall health Visual Analogue Scale (VAS) and physician's global impression of patient's overall health VAS, as well as the absence of significant worsening of interstitial lung disease, a new scleroderma renal crisis, left ventricular failure or pulmonary arterial hypertension. The CRISS index score represents a probability of improvement and ranges between 0 and 1. This is a 2 stage process to predict probability of improvement: Step 1 - absence of major organ progression (SRC etc.) - score 0 Step 2 - predicted probability of improvement - (score 0 - 1) |
Countries
Argentina, Australia, Austria, Belgium, Brazil, Canada, Chile, China, Czechia, Denmark, Finland, France, Germany, Greece, Hungary, India, Ireland, Israel, Italy, Japan, Malaysia, Mexico, Netherlands, Norway, Poland, Portugal, Spain, Sweden, Switzerland, Thailand, United Kingdom, United States
Participant flow
Recruitment details
This was a randomised, placebo-controlled, double-blind, parallel design trial. Abbreviation used: treatment (trt) baseline (bl.) categorical (cat.) continuous (cont.) number (no.) patient (pt) Placebo (pl.) discontinued (disc.) primary analysis (PA) Antitopoisomerase Antibody (ATA)
Pre-assignment details
All subjects were screened for eligibility to participate in trial. Subjects attended specialist sites to ensure that they (the subjects) met all implemented inclusion/exclusion criteria. Subjects were not to be randomised to trial drug if any of the specific entry criteria was violated
Participants by arm
| Arm | Count |
|---|---|
| Placebo Patients were administered orally placebo matching nintedanib 150 milligram (mg) soft gelatine capsules, twice daily with a possibility to interrupt treatment or to reduce to 100mg to manage adverse events. | 288 |
| Nintedanib Patients were administered orally 150 milligram (mg) soft gelatin capsules, twice daily with a possibility to interrupt treatment or to reduce to 100mg to manage adverse events. | 288 |
| Total | 576 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Adverse Event | 20 | 28 |
| Overall Study | Not treated | 2 | 2 |
| Overall Study | Other than stated above | 7 | 14 |
| Overall Study | Protocol Violation | 2 | 2 |
| Overall Study | Withdrawal by Subject | 7 | 5 |
Baseline characteristics
| Characteristic | Placebo | Nintedanib | Total |
|---|---|---|---|
| Age, Continuous | 53.4 Years STANDARD_DEVIATION 12.6 | 54.6 Years STANDARD_DEVIATION 11.8 | 54.0 Years STANDARD_DEVIATION 12.2 |
| Baseline pulmonary efficacy variables - Forced Vital Capacity (FVC) | 2541.0 mL STANDARD_DEVIATION 815.5 | 2458.5 mL STANDARD_DEVIATION 735.9 | 2499.7 mL STANDARD_DEVIATION 777.2 |
| Ethnicity (NIH/OMB) Hispanic or Latino | 18 Participants | 22 Participants | 40 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 270 Participants | 266 Participants | 536 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 3 Participants | 2 Participants | 5 Participants |
| Race (NIH/OMB) Asian | 81 Participants | 62 Participants | 143 Participants |
| Race (NIH/OMB) Black or African American | 16 Participants | 20 Participants | 36 Participants |
| Race (NIH/OMB) More than one race | 2 Participants | 2 Participants | 4 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants | 1 Participants | 1 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) White | 186 Participants | 201 Participants | 387 Participants |
| Sex: Female, Male Female | 212 Participants | 221 Participants | 433 Participants |
| Sex: Female, Male Male | 76 Participants | 67 Participants | 143 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 9 / 288 | 10 / 288 |
| other Total, other adverse events | 239 / 288 | 270 / 288 |
| serious Total, serious adverse events | 79 / 288 | 88 / 288 |
Outcome results
Primary
Annual Rate of Decline in Forced Vital Capacity (FVC) Over 52 Weeks
Forced vital capacity (FVC) is the total amount of air exhaled during the lung function test. For this endpoint reported means represent the adjusted rate.
Time frame: up to week (wk) 52 after the start of administration
Population: Treated Set
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Placebo | Annual Rate of Decline in Forced Vital Capacity (FVC) Over 52 Weeks | -93.3 millilitre (mL)/year (yr) | Standard Error 13.5 |
| Nintedanib | Annual Rate of Decline in Forced Vital Capacity (FVC) Over 52 Weeks | -52.4 millilitre (mL)/year (yr) | Standard Error 13.8 |
Comparison: The primary analysis is a restricted maximum likelihood (REML) based approach using a random slope \& intercept model. The analysis included the fixed, categorical effects of treatment, ATA status \& gender, fixed continuous effects of time \& baseline FVC (mL), age and height as well as the treatment-by time \& baseline-by-time interactions. Random effects was included for patient response for both time \& intercept.Within-patient errors are modelled by an unstructured variance-covariance matrixp-value: 0.03595% CI: [2.88, 79.01]random coefficient regression
Comparison: This is a sensitivity analysis (SA) on primary endpoint including only on-trt measurements of FVC \[mL\]. The random coefficient model was used. The analysis included fixed, categorical effects of trt, ATA status \& gender, fixed continuous effects of time \& bl. FVC (mL), age, height, trt -by time \& bl.-by-time interactions. Random effects included for patient response for both time \& intercept.~Within-patient errors were modelled by an Unstructured variance-covariance matrix.p-value: 0.037895% CI: [2.44, 83.83]random coefficient regression
Comparison: In multiple imputation SA 1, missing FVC values at wk 52 in pts who were alive at wk 52 were imputed assuming similar rate of FVC decline as in pts from corresponding trt group who prematurely disc. trial drug but had wk 52 FVC value. Missing FVC values at wk 52 in pts who died before wk 52 were imputed assuming similar rate of FVC decline as in pl. pts with wk 52 FVC value who prematurely disc. trial drug with most severe declines.The imputation model was similar to statistical model of PA.p-value: 0.104695% CI: [-6.22, 66.22]random coefficient regression
Comparison: In multiple imputation SA 2, missing FVC values at wk 52 in pts who were alive at wk 52 were imputed assuming similar rate of FVC decline as in pts from pl. group who prematurely disc. trial drug but had a wk 52 FVC value. Missing FVC values at wk 52 in pts who died before wk 52 were imputed assuming similar rate of FVC decline as in pl. pts with a wk 52 FVC value who prematurely disc. trial drug with most severe declines. The imputation model was similar to the statistical model of the PAp-value: 0.07495% CI: [-3.19, 69.06]random coefficient regression
Comparison: In multiple imputation SA 3, missing FVC values at wk 52 in pts who were alive at wk 52 were imputed assuming a similar rate of FVC decline as in all pts in the pl. group who were included in the PA. Missing FVC values at wk 52 in pts who died before wk 52 were imputed assuming a similar rate of FVC decline as in all placebo patients included in the primary analysis with the most severe declines. The imputation model was similar to the statistical model of the PA.p-value: 0.064495% CI: [-2.03, 69.75]random coefficient regression
Comparison: This is sensitivity analysis using the model similar to the primary analysis but including a different set of covariates: the fixed, categorical effects of treatment, ATA status, the fixed continuous effects of time, baseline FVC (mL), and the treatment-by-time and baseline-by-time interactions.Random effects was included for patient response for both time and intercept.p-value: 0.035195% CI: [2.88, 79.01]random coefficient regression
Comparison: This is sensitivity analysis using the model similar to the primary analysis but including a different set of covariates: the fixed, categorical effects of treatment, ATA status (Positive / Negative), gender and mycophenolate mofetil /sodium background therapy use (Yes / No), fixed continuous effects of time, age , height and baseline FVC (mL), the treatment-by-time and baseline-by-time interactions. Random effects was included for patient response for both time and interceptp-value: 0.034995% CI: [2.92, 79.04]random coefficient regression
Secondary
Absolute Change From Baseline in Carbon Monoxide Diffusion Capacity (DLco) in % Predicted at Week 52
Absolute change from baseline in Carbon Monoxide Diffusion Capacity (DLco) in % predicted at Week 52. Least square mean is actually the adjusted mean. Adjusted mean is based on all analysed patients in the model (not only patients with a baseline and measurement at week 52).
Time frame: Baseline and up to 52 weeks after the start of administration
Population: Treated set
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| Placebo | Absolute Change From Baseline in Carbon Monoxide Diffusion Capacity (DLco) in % Predicted at Week 52 | -2.77 % predicted DLco | Standard Error 0.54 |
| Nintedanib | Absolute Change From Baseline in Carbon Monoxide Diffusion Capacity (DLco) in % Predicted at Week 52 | -3.21 % predicted DLco | Standard Error 0.54 |
Comparison: Based on mixed model repeated measures (MMRM) approach was used, with fixed categorical effects of ATA status, visit, treatment-by-visit interaction and baseline-by-visit interaction. Visit was the repeated measure. Within-patient errors were modelled by unstructured variance-covariance structure.p-value: 0.566895% CI: [-1.94, 1.06]MMRM
Secondary
Absolute Change From Baseline in Digital Ulcer Net Burden at Week 52
Absolute change from baseline in digital ulcer net burden (defined as the number of new digital ulcers (DUs) plus the number of DUs that have been verified at any earlier assessment during the trial) at Week 52. It is calculated at a visit by counting the total number of fingertips with ulcers (i.e. number of fingers with presence of digital ulcer ticked Yes) at the corresponding visit Least square mean is actually the adjusted mean. Adjusted mean is based on all analysed patients in the model (not only patients with a baseline and measurement at week 52).
Time frame: Baseline and up to 52 weeks after the start of administration
Population: Treated set
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| Placebo | Absolute Change From Baseline in Digital Ulcer Net Burden at Week 52 | 0.06 fingers | Standard Error 0.04 |
| Nintedanib | Absolute Change From Baseline in Digital Ulcer Net Burden at Week 52 | 0.03 fingers | Standard Error 0.05 |
Comparison: Based on mixed model repeated measures (MMRM) approach was used, with fixed categorical effects of ATA status, visit, treatment-by-visit interaction and baseline-by-visit interaction. Visit was the repeated measure. Within-patient errors were modelled by unstructured variance-covariance structure.p-value: 0.591495% CI: [-0.16, 0.09]MMRM
Secondary
Absolute Change From Baseline in Functional Assessment of Chronic Illness Therapy (FACIT) Dyspnoea Score at Week 52
Absolute change from baseline in Functional Assessment of Chronic Illness Therapy (FACIT) dyspnoea score at Week 52. FACIT-Dyspnoea (Dyspnoea) 10 Item Short Form include a 4-point rating scale (no shortness of breath=0; mildly short of breath=1; moderately short of breath = 2; severely short of breath =3; or I did not do this in the past 7 days =4). A raw score is calculated as: Sum individual item scores \* 10 / number of items answered. Raw scores are then converted to scale scores using the table included in the FACIT Dyspnoea Scale Short Form Scoring Guideline. FACIT dyspnea scale score ranges between 0 and 75.9. The FACIT-Dyspnea short forms are scored such that a high score represents high levels of dyspnea. Least square mean is actually the adjusted mean. Adjusted mean is based on all analysed patients in the model (not only patients with a baseline and measurement at week 52).
Time frame: Baseline and up to 52 weeks after the start of administration
Population: Treated set
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| Placebo | Absolute Change From Baseline in Functional Assessment of Chronic Illness Therapy (FACIT) Dyspnoea Score at Week 52 | 0.34 Unit on a scale | Standard Error 0.41 |
| Nintedanib | Absolute Change From Baseline in Functional Assessment of Chronic Illness Therapy (FACIT) Dyspnoea Score at Week 52 | 0.99 Unit on a scale | Standard Error 0.42 |
Comparison: Based on mixed model repeated measures (MMRM) approach was used, with fixed categorical effects of ATA status, visit, treatment-by-visit interaction and baseline-by-visit interaction. Visit was the repeated measure. Within-patient errors were modelled by unstructured variance-covariance structure.p-value: 0.272795% CI: [-0.51, 1.79]MMRM
Secondary
Absolute Change From Baseline in FVC in mL at Week 52
Absolute change from baseline in FVC in mL at Week 52. Least square mean is actually the adjusted mean. Adjusted mean was based on all analysed patients in the model (not only patients with a baseline and measurement at Week 52).
Time frame: Baseline and up to 52 weeks after the start of administration
Population: Treated Set
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| Placebo | Absolute Change From Baseline in FVC in mL at Week 52 | -101.03 mL | Standard Error 13.62 |
| Nintedanib | Absolute Change From Baseline in FVC in mL at Week 52 | -54.63 mL | Standard Error 13.94 |
Comparison: Based on mixed model repeated measures (MMRM) approach was used, with fixed categorical effects of ATA status, visit, treatment-by-visit interaction and baseline-by-visit interaction. Visit was the repeated measure. Within-patient errors were modelled by unstructured variance-covariance structure.p-value: 0.017795% CI: [8.09, 84.73]MMRM
Secondary
Absolute Change From Baseline in Health Assessment Questionnaire-Disability Index (HAQ-DI) Score at Week 52
Absolute change from baseline in Health Assessment Questionnaire-Disability Index (HAQ-DI) score at Week 52. The HAQ-DI score is calculated as follows: Each question is scored 0-3 (where 0= without difficulty & 3= unable to do). There are 8 categories (Dressing & Grooming, Arising, Eating, Walking, Hygiene, Reach, Grip, Activities), each including 2 or 3 questions. The score for each category corresponds to maximum question score within each category. Finally, HAQ-DI score corresponds to sum of the sub-scores of all 8 categories divided by number of categories completed. Please note that if there are fewer than 6 categories with responses, then a score cannot be calculated. The HAQ-DI score scale has 25 possible values (i.e., 0, 0.125, 0.250, 0.375 … 3). A high score corresponds to worse impairment. Least square mean is actually the adjusted mean. Adjusted mean is based on all analysed patients in the model (not only patients with a baseline and measurement at week 52).
Time frame: Baseline and up to 52 weeks after the start of administration
Population: Treated set
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| Placebo | Absolute Change From Baseline in Health Assessment Questionnaire-Disability Index (HAQ-DI) Score at Week 52 | 0.022 unit on a scale | Standard Error 0.024 |
| Nintedanib | Absolute Change From Baseline in Health Assessment Questionnaire-Disability Index (HAQ-DI) Score at Week 52 | 0.054 unit on a scale | Standard Error 0.024 |
Comparison: Based on mixed model repeated measures (MMRM) approach was used, with fixed categorical effects of ATA status, visit, treatment-by-visit interaction and baseline-by-visit interaction. Visit was the repeated measure. Within-patient errors were modelled by unstructured variance-covariance structure.p-value: 0.344795% CI: [-0.035, 0.099]MMRM
Secondary
Absolute Change From Baseline in Saint George's Respiratory Questionnaire (SGRQ) Total Score at Week 52.
This is the second key secondary endpoint. The Saint George's Respiratory Questionnaire measures the health status in patients with chronic airflow limitation. It consists of 2 parts that cover 3 domains: symptoms, activities, and impacts. The symptom domain relates to the effect, frequency and severity of respiratory symptoms. The activity domain relates to activities that cause or are limited by breathlessness. The impact domain evaluates a range of aspects concerned with social functioning and psychological disturbances resulting from airways disease. The scores of these domains range from 0 (no impairment) to 100 (worst possible). The calculated total score summarises the impact of the disease on overall health status. A high score corresponds to worse health. Least square mean is actually the adjusted mean. Adjusted mean was based on all analysed patients in the model (not only patients with a baseline and measurement at Week 52).
Time frame: Baseline and up to 52 weeks after the start of administration
Population: Treated set
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| Placebo | Absolute Change From Baseline in Saint George's Respiratory Questionnaire (SGRQ) Total Score at Week 52. | -0.88 unit on scale | Standard Error 0.87 |
| Nintedanib | Absolute Change From Baseline in Saint George's Respiratory Questionnaire (SGRQ) Total Score at Week 52. | 0.81 unit on scale | Standard Error 0.88 |
Comparison: Based on mixed model repeated measures (MMRM) approach was used, with fixed categorical effects of ATA status, visit, treatment-by visit interaction and baseline-by-visit interaction. Visit was the repeated measure. Within-patient errors were modelled by unstructured variance-covariance structure.p-value: 0.171195% CI: [-0.73, 4.12]MMRM
Secondary
Absolute Change From Baseline in the Modified Rodnan Skin Score (mRSS) at Week 52
This is the first key secondary endpoint. The modified Rodnan Skin Score (mRSS) is an evaluation of the patient's skin thickness rated by clinical palpation using a 0 to 3 scale. The scale differentiates between 0 = normal skin, 1 = mild thickness, 2 = moderate thickness, and 3 = severe thickness with inability to pinch the skin into a fold. The palpation is done for each of the 17 surface anatomic areas of the body: face, anterior chest, abdomen, fingers (right and left separately), forearms, upper arms, thighs, lower legs, dorsum of hands and feet. The sum of these individual values is defined as the total skin score. The mRSS has a range from 0 (no thickening) to 51 (severe thickening in all 17 areas). A high score corresponds to worse skin thickness. Least square mean is actually the adjusted mean. Adjusted mean was based on all analysed patients in the model (not only patients with a baseline and measurement at Week 52).
Time frame: Baseline and up to 52 weeks after the start of administration
Population: Treated Set
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| Placebo | Absolute Change From Baseline in the Modified Rodnan Skin Score (mRSS) at Week 52 | -1.96 unit on scale | Standard Error 0.26 |
| Nintedanib | Absolute Change From Baseline in the Modified Rodnan Skin Score (mRSS) at Week 52 | -2.17 unit on scale | Standard Error 0.27 |
Comparison: The mixed model repeated measures (MMRM) approach was used. The model assumed that data were missing at random \& that patients who dropped out would have behaved similarly to those who remained in trial.p-value: 0.578595% CI: [-0.94, 0.53]MMRM
Secondary
Annual Rate of Decline in FVC in Percentage (%) Predicted Over 52 Weeks
Annual rate of decline in FVC in percentage (%) predicted over 52 weeks. For this endpoint reported means represent the adjusted rate.
Time frame: up to 52 weeks after the start of administration
Population: Treated set
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Placebo | Annual Rate of Decline in FVC in Percentage (%) Predicted Over 52 Weeks | -2.6 % predicted/yr | Standard Error 0.4 |
| Nintedanib | Annual Rate of Decline in FVC in Percentage (%) Predicted Over 52 Weeks | -1.4 % predicted/yr | Standard Error 0.4 |
Comparison: Based on a random coefficient regression with fixed categorical effects of treatment, ATA status, fixed continuous effects of time, baseline FVC \[% pred\], \& including treatment-by-time and baseline-by-time interactions. Random effect was included for patient specific intercept \& time.~Within-patient errors are modelled by an Unstructured variance-covariance matrix.~Inter-individual variability is modelled by a Variance-Components variance-covariance matrix.p-value: 0.03311.15% CI: [0.09, 2.21]MMRM
Secondary
Relative Change From Baseline [%] of mRSS at Week 52
Relative change from baseline \[%\] of mRSS at Week 52. The modified Rodnan Skin Score (mRSS) is an evaluation of the patient's skin thickness rated by clinical palpation using a 0 to 3 scale. The scale differentiates between 0 = normal skin, 1 = mild thickness, 2 = moderate thickness, and 3 = severe thickness with inability to pinch the skin into a fold. The palpation is done for each of the 17 surface anatomic areas of the body: face, anterior chest, abdomen, fingers (right and left separately), forearms, upper arms, thighs, lower legs, dorsum of hands and feet. The sum of these individual values is defined as the total skin score. The mRSS has a range from 0 (no thickening) to 51 (severe thickening in all 17 areas). A high score corresponds to worse skin thickness. Least square mean is actually the adjusted mean. Adjusted mean was based on all analysed patients in the model (not only patients with a baseline and measurement at Week 52).
Time frame: Baseline and up to 52 weeks after the start of administration
Population: Treated set
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| Placebo | Relative Change From Baseline [%] of mRSS at Week 52 | -3.92 percent change | Standard Error 5.89 |
| Nintedanib | Relative Change From Baseline [%] of mRSS at Week 52 | -10.20 percent change | Standard Error 5.98 |
Comparison: Based on mixed model repeated measures (MMRM) approach was used, with fixed categorical effects of ATA status, visit, treatment-by-visit interaction and baseline-by-visit interaction. Visit was the repeated measure. Within-patient errors were modelled by unstructured variance-covariance structure.p-value: 0.454795% CI: [-22.77, 10.21]MMRM
Secondary
The Percentage (%) of Responder Based on Combined Response Index in Systemic Sclerosis (CRISS) at Week 52
The percentage (%) of responder based on Combined Response Index in Systemic Sclerosis (CRISS) at Week 52. This is a composite endpoint, based on the mRSS, FVC percent predicted, HAQ-DI, patient's global impression of overall health Visual Analogue Scale (VAS) and physician's global impression of patient's overall health VAS, as well as the absence of significant worsening of interstitial lung disease, a new scleroderma renal crisis, left ventricular failure or pulmonary arterial hypertension. The CRISS index score represents a probability of improvement and ranges between 0 and 1. This is a 2 stage process to predict probability of improvement: Step 1 - absence of major organ progression (SRC etc.) - score 0 Step 2 - predicted probability of improvement - (score 0 - 1)
Time frame: Week 52
Population: Treated set
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Placebo | The Percentage (%) of Responder Based on Combined Response Index in Systemic Sclerosis (CRISS) at Week 52 | 11.8 (%) of responder based on CRISS |
| Nintedanib | The Percentage (%) of Responder Based on Combined Response Index in Systemic Sclerosis (CRISS) at Week 52 | 12.2 (%) of responder based on CRISS |
Comparison: The comparison between both treatment groups was performed using a Cochran-Mantel-Haenszel test. CRISS score at Week 52 was transformed into 100 binary responder endpoints using multiple imputation. These were analyzed using a Cochran-Mantel-Haenszel test, stratified by ATA status OR and the 95% confidence interval (CI) as obtained from all 100 imputations were combined using Rubin´s rule.p-value: 0.911595% CI: [0.57, 1.88]Cochran-Mantel-Haenszel
Secondary
Time to Death
Time to event analysis of patients with death. The number of observed patients with death are reported.
Time frame: From date of first trial drug intake up to date of death or last contact date (ie., up to 100 weeks)
Population: Treated set
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Placebo | Time to Death | 9 Participants |
| Nintedanib | Time to Death | 10 Participants |
p-value: 0.753595% CI: [0.47, 2.84]Regression, Cox