Articles
Safety, tolerability, pharmacokinetics, and pharmacodynamics of GLPG1690, a novel autotaxin inhibitor, to treat idiopathic pulmonary fibrosis (FLORA): a phase 2a randomised placebo-controlled trial Toby M Maher, Ellen M van der Aar, Olivier Van de Steen, Lisa Allamassey, Julie Desrivot, Sonia Dupont, Liesbeth Fagard, Paul Ford, Ann Fieuw, Wim Wuyts
Summary
Background Idiopathic pulmonary fibrosis (IPF) causes irreversible loss of lung function. People with IPF have increased concentrations of autotaxin in lung tissue and lysophosphatidic acid (LPA) in bronchoalveolar lavage fluid and exhaled condensate. GLPG1690 (Galapagos, Mechelen, Belgium) is a novel, potent, selective autotaxin inhibitor with good oral exposure. We explored the effects of GLPG1690 in patients with IPF.
Lancet Respir Med 2018
Methods This was a randomised, double-blind, placebo-controlled phase 2a study done in 17 centres in Italy, Ukraine and the UK. Eligible patients were aged 40 years or older, non-smokers, not taking pirfenidone or nintedanib, and had a centrally confirmed diagnosis of IPF. We used a computer-generated randomisation schedule to assign patients 1:3 to receive placebo or 600 mg oral GLPG1690 once daily for 12 weeks. The primary outcomes were safety (adverse events), tolerability, pharmacokinetics, and pharmacodynamics. Spirometry was assessed as a secondary outcome. This trial is registered with ClinicalTrials.gov, number NCT02738801.
NIHR Respiratory Clinical Research Facility, Royal Brompton Hospital, London, UK (Prof T M Maher MD); Fibrosis Research Group, National Heart and Lung Institute, Imperial College, London, UK (Prof T M Maher); Galapagos, Mechelen, Belgium (E M van der Aar PhD, O Van de Steen MD, L Fagard MSc, P Ford MD, A Fieuw MD); Alten, Brussels, Belgium (L Allamassey MSc); Galapagos, Romainville, France (J Desrivot PhD, S Dupont PhD); and Unit for Interstitial Lung Diseases, Department of Pulmonary Medicine, University Hospitals Leuven, Leuven, Belgium (Prof W Wuyts MD)
Findings Between March 24, 2016, and May 2, 2017, 72 patients were screened., of whom 49 were ineligible and 23 were enrolled in eight centres (six in Ukraine and two in the UK). Six patients were assigned to receive placebo and 17 to receive GLPG1690. 20 patients completed the study after one in each group discontinued because of adverse events and one in the GLPG1690 group withdrew consent. Four (67%) patients in the placebo group and 11 (65%) in the GLPG1690 group had treatment-emergent adverse events, most of which were mild to moderate. The most frequent events in the GLPG1690 group were infections and infestations (ten events) and respiratory, thoracic, and mediastinal disorders (eight events) with no apparent differences from the placebo group. Two (12%) patients in the GLPG1690 group had events that were judged to be related to treatment. Serious adverse events were seen in two patients in the placebo group (one had a urinary tract infection, acute kidney injury, and lower respiratory tract infection and the other had atrioventricular block, second degree) and one in the GLPG1690 group (cholangiocarcinoma that resulted in discontinuation of treatment). No patients died. The pharmacokinetic and pharmacodynamic profiles of GLPG1690 were similar to those previously shown in healthy controls. LPA C18:2 concentrations in plasma were consistently decreased. Mean change from baseline in forced vital capacity at week 12 was 25 mL (95% CI –75 to 124) for GLPG1690 and –70 mL (–208 to 68 mL) for placebo.
Published Online May 20, 2018 http://dx.doi.org/10.1016/ S2213-2600(18)30181-4
Correspondence to: Prof Toby M Maher, NIHR Respiratory Clinical Research Facility, Royal Brompton Hospital, London SW3 6NP, UK
[email protected]
Interpretation Our findings support further development of GLPG1690 as a novel treatment for IPF. Funding Galapagos. Copyright © 2018 Elsevier Ltd. All rights reserved.
Introduction Autotaxin is the primary enzyme responsible for the production of lysophosphatidic acid (LPA),1 which is essential for a diverse range of cellular processes. LPA and autotaxin have been implicated as key factors in several disorders and pathologies.1 Increased autotaxin concentrations have been found in lung tissue in people with idiopathic pulmonary fibrosis (IPF), and increased concentrations of LPA have been found in bronchoalveolar lavage fluid (BALF) and exhaled breath condensate,2,3 suggesting that the autotaxin-LPA pathway has a pathogenic role in this disorder. Autotaxin inhibition
might be a novel therapeutic target in the treatment of IPF. IPF is characterised by progressive accumulation of collagen scar tissue in the lungs that leads to irreversible loss of lung function4 and in most patients death due to respiratory failure.5 Prognosis is poor,5,6 with 5 year survival of 20–30%.5 The cause of IPF remains poorly understood. Based on pivotal phase 3 trial results,7–9 the antifibrotic treatments pirfenidone and nintedanib received approval worldwide for the treatment of IPF. Both agents slow disease progression,10,11 but neither stabilises or improves lung function, and both therapies have tolerability issues
www.thelancet.com/respiratory Published online May 20, 2018 http://dx.doi.org/10.1016/S2213-2600(18)30181-4
1
Articles
Research in context Evidence before this study We searched PubMed with the term “idiopathic pulmonary fibrosis” for articles that were published from Jan 1, 2000, to Jan 29, 2018, and contained the term in the title or abstract. Of the 5443 articles retrieved, 180 were clinical trials. After excluding those that listed nintedanib or pirfenidone in the title or abstract, 145 articles remained. Among these, multiple potential pharmacotherapies for idiopathic pulmonary fibrosis (IPF) were discussed, including interferon, immunoglobulin, N-acetylcysteine, recombinant human pentraxin-2, thrombomodulin, ciclosporin, antibodies against CD20 or tumour necrosis factor, octreotide, bosentan, colchicine, everolimus, imatinib, lecithinised superoxide dismutase, collagen, endothelin-A-receptor antagonists, polymyxinB-immobilised fibre, warfarin, heparin, C-C motif chemokine-2 inhibitors, tyrosine-kinase inhibitors, and Feiwei granules. The breadth of targets under investigation for these drugs highlights the pressing need for novel therapies for the treatment of IPF and the interest in this field.
and substantial discontinuation rates.7–9,12–14 As such, an unmet need exists for more effective and better tolerated novel therapies. GLPG1690 (Galapagos, Mechelen, Belgium) is a potent and selective inhibitor of autotaxin that in rats has been associated with reduced concentrations of LPA C18:2 species in plasma after oral administration.15 Compared with pirfenidone, GLPG1690 was significantly superior in reducing the Ashcroft fibrotic score at prophylactic15 and therapeutic16 doses in mice with bleomycin-induced pulmonary fibrosis. An additive inhibitory effect on profibrotic mediators was seen in ex-vivo assessment of fibroblasts isolated from IPF lung tissue after use of combined GLPG1690 and nintedanib.17 A phase 1 first-inhuman study showed that GLPG1690 had good oral exposure and was generally well tolerated.18 Plasma concentrations of LPA C18:2 decreased with increasing concentrations of GLPG1690.18 The evidence implicating autotaxin and LPA in IPF and the efficacy and tolerability issues associated with current IPF treatments suggest that GLPG1690 could provide a novel treatment option for IPF. In this study we aimed to assess the safety, tolerability, pharmacokinetics, and pharmacodynamics of GLPG1690 in people with IPF. We additionally explored efficacy, biomarkers, functional respiratory imaging, and health-related quality of life. See Online for appendix
Methods
Study design and participants We did a multicentre, randomised, double-blind, parallelgroup, placebo-controlled, exploratory phase 2a study. We recruited patients from 17 clinical study centres in Italy, Ukraine, and the UK after a screening period of up to 4 weeks. Eligible patients were men or women aged 2
Added value of this study We believe this study to be unique among IPF clinical trials because it reports phase 2 results, including innovative endpoints, for a treatment with a novel mechanism of action in IPF. This small proof-of-concept study was intended to bridge the gap between the early pharmacokinetic and pharmacodynamic findings for GLPG1690 (Galapagos, Mechelen, Belgium) and assess its characteristics in people with IPF before moving to larger trials. Implications of all the available evidence Our results and the previous preclinical and phase 1 data support the further development of GLPG1690 for the treatment of patients with IPF. Longer-term data will provide further insights into the potential of GLPG1690 to address the unmet need in the treatment of IPF, including therapies with improved tolerability that are able to halt disease progression.
40 years or older with a diagnosis of IPF based on the American Thoracic Society/European Respiratory Society/Japanese Respiratory Society/Latin American Thoracic guidelines.19 Diagnoses were confirmed by central review of high-resolution CT chest scans obtained within 12 months before screening and, if available, surgical lung biopsy samples. The central review was done by an experienced radiologist or pathologist with expertise in interstitial lung diseases. Eligible individuals had forced vital capacity (FVC) at least 50% predicted, diffusing capacity for the lungs for carbon monoxide of at least 30% of predicted normal corrected for haemoglobin, and FEV1:FVC ratio 0·70 or greater, based on prebroncho dilator spirometry. Patients also had to have stable disease and a minimum life expectancy of 12 months (judged by an investigator). Exclusion criteria included interstitial lung disease associated with known primary causes; an acute IPF exacerbation within 6 weeks before or during the screening period; a lower respiratory tract infection requiring antibiotics within 4 weeks before or during the screening period; smoking within 3 months before screening; a history of lung volume reduction surgery or lung transplant; and treatment with pirfenidone, nintedanib, or prednisone at a steady dose greater than 15 mg per day or an experimental IPF therapy within 4 weeks before screening (all were prohibited throughout the course of the study). Full inclusion and exclusion criteria are available in the appendix. Independent ethics committees at each participating site approved the study and it was done in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. All study participants provided written informed consent. The study protocol is available in the appendix.
www.thelancet.com/respiratory Published online May 20, 2018 http://dx.doi.org/10.1016/S2213-2600(18)30181-4
Articles
Randomisation and masking We used a computer-generated randomisation schedule created by the contract research organisation (Chiltern International Ltd, Slough, UK), using permuted blocks, to assign patients in a 1:3 ratio to receive placebo or 600 mg oral GLPG1690 once daily for 12 weeks. Group allocations were obtained via a computerised web-based interactive voice response system (Medidata Balance, New York, NY, US). Patients and all study personnel were masked to treatment assignment. Each GLPG1690 dose was given as three capsules (size 00) containing 200 mg drug and the placebo was provided as matching capsules. At each visit, patients received medication kits containing enough doses until the next visit and labelled with the allocated randomisation number.
Procedures Screening was done at the study centres in the 4 weeks before random treatment assignment. Enrolled patients attended visits on the day before week 1 (baseline), in weeks 1 (days 5–9), 2 (day 12–16), 4 (days 25–31), 8 (days 53–59), and 12 (day 81–87) and 2 weeks after the end of treatment (days 95–101). Thus, patients were in the study for up to 18 weeks. They took placebo or 600 mg GLPG1690 once daily in the morning at home, unless attending a study visit when the dose was taken at the study centre. Safety was assessed by monitoring of adverse events (including acute IPF exacerbations) and concomitant medications throughout the study. At each visit, a physical examination and 12-lead electrocardiogram ([ECG] all visits except baseline) were done, vital signs were monitored, and blood and urine samples were taken for clinical laboratory tests. Additional blood samples were taken for biomarker analysis at baseline, before the dose at the week 1, 4, and 12 visits, and at the final visit after the end of treatment; for pharmacokinetic and pharmacodynamic testing at baseline, before the dose at the week 1, 2, and 8 visits, and at the final visit after the end of treatment; for pharmacokinetic testing at 1·5, 4·0, and 6·0 h after the dose at the week 4 visit; and for pharmacodynamic testing at 1·5 and 6·0 h after the dose at the week 4 visit. BALF samples were taken via bronchoscopy at baseline and after the dose at the week 12 visit. Spirometry was done in the study centres at screening, baseline, and after the dose during all visits from baseline, and at home at baseline and every day after the dose. All spirometry tests had to be done at roughly the same time (within 1 h) in the morning. Spirometry testing at the study sites was done according to the acceptability and repeatability criteria of American Thoracic Society/European Respiratory Society guide lines20 with the NDD Easy On-PC (TrueFlow) PC-based spirometer (New Diagnostic Design Medizintechnik, Zurich, Switzerland). The spirometry tests at home were done with hand-held devices (EasyOne Air
spirometer, New Diagnostic Design). All spirometry testing was done more than 6·0 h after administration of short-acting β agonists or anticholinergics, at least 12 h after administration of a long-acting bronchodilator, and at least 24 h after administration of longer-acting agents. High-resolution CT scans were done at baseline and after the dose at the week 12 visit. These were used to generate functional respiratory imaging measurements to assess specific airway volume and specific airway resistance at total lung capacity. Health-related quality of life was assessed with the St George’s Respiratory Questionnaire (SGRQ),21 which was completed at baseline and before the dose at the week 4, 12, and 14 visits. Patients read and self-completed the SGRQ alone. Study staff reviewed the questionnaires afterwards to ensure all questions were completed and asked patients to complete any missing items.
Outcomes The primary outcomes were the safety and tolerability, pharmacokinetics, and pharmacodynamics of GLPG1690. We measured concentrations of the study drug in plasma with a liquid chromatography tandem mass spectrometry method validated by Galapagos (Romainville, France) and did a non-compartmental analysis with Phoenix WinNonlin version 6.4 to assess pharmacokinetics. The pharmacokinetic parameters were maximum observed concentration (Cmax) in plasma, the time at which Cmax occurred, trough concentration in plasma, and the area under the plasma drug concentration–time curve for the 24 h dosing interval. LPA species for target engagement
72 patients screened
49 patients excluded* 44 did not meet eligibility criteria 3 withdrew consent 2 eligibility assessments incomplete 2 other reasons
23 patients randomly assigned treatment
6 assigned to receive placebo once daily
17 assigned to receive 600 mg GLPG1690 once daily
1 discontinued treatment (adverse event)
5 completed the study
2 discontinued treatment 1 adverse event 1 withdrew consent
15 completed the study
Figure 1: Trial profile *Some patients were excluded for more than one reason.
www.thelancet.com/respiratory Published online May 20, 2018 http://dx.doi.org/10.1016/S2213-2600(18)30181-4
3
Articles
Placebo group (n=6)
GLPG1690 group (n=17)
Median (IQR) age (years)
64·0 (54·0–69·0)
67·0 (61·0–73·0)
Median (IQR) BMI (kg/m²)
32·2 (26·6–36·8)
28·0 (24·8–32·5)
White ethnic origin
6 (100%)
17 (100%)
Men
5 (83%)
10 (59%)
Women
1 (17%)
7 (41%)
Sex
Smoking status Former
3 (50%)
6 (35%)
Never
3 (50%)
11 (65%)
Duration of IPF (years)* Mean (95% CI)
1·0 (0·5–1·6)
1·9 (0·7–3·1)
Median (IQR)
1·0 (0·8–1·3)
0·9 (0·4–2·5)
Screening DLCO (% predicted of normal)† Mean (95% CI)
40·6 (29·3–51·9)
37·8 (34·3–41·3)
Median (IQR)
35·5 (33·0–53·2)
36·9 (31·6–42·5)
Baseline FVC (L) Mean (95% CI)
2·7 (1·7–3·7)
2·8 (2·3–3·2)
Median (IQR)
2·3 (2·1–3·3)
3·0 (1·9–3·5)
Baseline FVC (%) Mean (95% CI)
69·7 (46·4–92·9)
75·3 (67·9–82·7)
Median (IQR)
58·5 (57·0–83·0)
74·0 (66·0–84·0)
BMI=body mass index. IPF=idiopathic pulmonary fibrosis. DLCO=diffusing capacity for the lungs for carbon monoxide. FVC=forced vital capacity. *(Date informed consent signed – date of initial IPF diagnosis) / 365·25. †Corrected for haemoglobin.
Table 1: Baseline characteristics in the safety population
Placebo group (n=6)
GLPG1690 group (n=17)
≥1 event
4 (67%)
11 (65%)
≥1 serious event
2 (33%)
1 (6%)
≥1 event resulting in death
0
0
≥1 event by worst severity Mild
0
4 (24%)
Moderate
3 (50%)
6 (35%)
1 (17%)
1 (6%)
≥1 event related to treatment
Severe
0
2 (12%)
≥1 event leading to temporary discontinuation of study drug
0
2 (12%)
≥1 event leading to permanent discontinuation of study drug
1 (17%)
1 (6%)
We classified adverse events with the Medical Dictionary for Regulatory Activities version 18.1. Patients could be included in more than one category.
Table 2: Treatment-emergent adverse events in the safety population
were assessed in plasma (LPA C18:2) and BALF (seven species; appendix) with an internally qualified liquid chroma tography tandem mass spectrometry method, based on the peak area ratio of the analyte over the internal standard LPA C17:0. Secondary outcomes included pulmonary function (measured at the study centre or at home), biomarkers of 4
disease activity (including autotaxin, Krebs von den Lungen 6/mucin type 1, surfactant protein A and D, chemokine motif ligand 18, matrix metalloproteinases 1 and 7, and markers of extracellular matrix turnover; appendix), functional respir atory imaging, and healthrelated quality of life. Pulmonary function was assessed with spirometry, FEV1, FVC, FEV1:FVC ratio, and forced expiratory flow between 25% and 75% of the exhaled volume. We measured biomarkers of disease activity before and after treatment. For the functional respiratory imaging assessments, we took increased airway volume and decreased airway resistance to be signs of disease progression. In the assessment of health-related quality of life, we took a mean change of 5–8 points on the SGRQ to be an important change, as has previously been estimated as the minimum important difference in IPF.22
Statistical analysis We aimed to randomly assign around 24 patients to receive GLPG1690 to explore its effect on IPF. We did not use strict statistical criteria or a formal sample size calculation. Rather, we based the number of patients on that judged to allow reasonable estimations of safety, pharmacokinetics, and pharmacodynamics. We calcul ated a retrospective power calculation estimating how much of the effect of GLPG1690 on FVC the study was powered to detect (appendix). The safety population included all patients who received at least one dose of the study drug or placebo. The pharmacokinetic population included all patients who received at least one dose of GLPG1690 and for whom evaluable pharmacokinetic data were available. The pharmacodynamic population included all patients who received at least one dose of study drug or placebo and had data from at least one pharmacodynamic data assessment after baseline. We did secondary efficacy analyses in the intention-to-treat population, which included all patients who received at least one dose of study drug or placebo and had data from one or more visits after baseline. We used SAS version 9.4 for all statistical analyses. Safety endpoints, including treatment-emergent adverse events, were analysed descriptively. We classified adverse events with the Medical Dictionary for Regulatory Activities version 18.1 and graded severity as mild, moderate, or severe based on the judgment of the investigator. Descriptive statistics were calculated with use of derived pharmacokinetic parameters. For pharmacodynamic data, plasma LPA C18:2 peak area ratios were summarised descriptively. Although we measured Cmax, we did no formal analysis of GLPG1690 trough (ie, predose) concentrations to assess achievement of steady state. Visual inspection of the GLPG1690 trough concentrations in plasma at the week 1 visit, however, indicated steady state by the time of the first pharmacokinetic assessment. This finding is supported by the first-in-human study, in which trough
www.thelancet.com/respiratory Published online May 20, 2018 http://dx.doi.org/10.1016/S2213-2600(18)30181-4
Articles
concentrations were established 4 days after the first dose.18 We measured the area under the effective-time curve over 6 h and the maximum percentage reduction from baseline at the week 4 visit, which were determined from individual effect–time profiles with the trapezoidal summation rule. Absolute values, percentage red uctions from baseline, area under the effective-time curve, and the maximum percentage reduction from baseline were compared between treatment groups with an ANCOVA model in which we included disease severity (baseline FVC and screening diffusing capacity for the lungs for carbon monoxide), age, sex, treatment group, country, and baseline values as covariates. We used a paired t test to compare absolute values in samples taken before the dose and at 1·5 and 6·0 h after the dose at the week 4 visit and before the dose at the week 12 visit, as well as percentage reductions from baseline. BALF LPA peak area ratio reductions from baseline were also calculated. For secondary outcomes, we analysed continuous parameters descriptively and compared changes from baseline between treatment groups with the same ANCOVA model. Because the endpoint analysis was prespecified, we made no correction for multiple analyses. Comparisons between placebo and GLPG1690 were exploratory. Within-group comparisons were done with a paired t test. Missing data were imputed, primarily via last observation carried forward (LOCF), including for patients who discontinued the study early, and we did an observed-case sensitivity analysis. For biomarker data, the LOCF and observed-case analyses were prespecified, and we present the data for the latter. All 95% CIs are pointwise. To maximise the safety and integrity of the study, an independent medical safety review was done by an independent pulmonologist experienced in IPF (WW). This involved regular review of unblinded safety data to monitor risks and benefits and to assess any potential safety issues arising during the study. This study is registered with ClinicalTrials.gov, number NCT02738801.
Role of the funding source The funder of the study supervised the study design, data collection, statistical analyses, data interpretation, and the writing of the report. The corresponding author had full access to all data in the study and had final responsibility for the decision to submit for publication.
Results Between March 24, 2016, and May 2, 2017, we screened 72 patients in 17 study centres, 49 of whom were ineligible and 23 of whom were enrolled to the study (at six study sites in Ukraine and two in the UK). Only two patients were screened in Italy and neither met the criteria for inclusion. Six patients were assigned to the placebo group and 17 to the GLPG1690 group, of whom five and 15, respectively, completed the study (figure 1).
Infections and infestations
Placebo group (n=6)
GLPG1690 group (n=17)
Number of patients (%)
Number of events
Number of patients (%)
Number of events 10
3 (50%)
8
7 (41%)
Lower respiratory tract infection 2 (33%)
3
2 (12%)
3
Nasopharyngitis
1 (17%)
1
2 (12%)
2 0
Orchitis
1 (17%)
1
0
Urinary tract infection
1 (17%)
3
0
0
Respiratory, thoracic, and mediastinal disorders
2 (33%)
4
4 (24%)
8
Cough
1 (17%)
1
2 (12%)
2
Dyspepsia
1 (17%)
1
2 (12%)
2
Productive cough
0
0
2 (12%)
2
Haemothorax
1 (17%)
1
0
0
Pneumothorax spontaneous
1 (17%)
1
0
0
2 (33%)
2
2 (12%)
2
2 (33%)
2
1 (6%)
1
1 (17%)
1
2 (12%)
2
Gastrointestinal disorders Diarrhoea General disorders and administration site conditions Peripheral swelling Investigations PR prolongation on electrocardiogram Musculoskeletal and connective tissue disorders
1 (17%)
1
1 (6%)
1
1 (17%)
1
2 (12%)
2
1 (17%)
1
0
0
2 (33%)
6
1 (6%)
1
Arthralgia
1 (17%)*
3
0
0
Joint swelling
1 (17%)*
1
0
0
Osteoarthritis
1 (17%)
1
0
0
Pain in extremity
1 (17%)*
1
0
0
Muscular weakness
0
0
1 (6%)
0
1 (17%)
2
0
0
Atrioventricular block second degree
1 (17%)
1
0
0
Bradycardia
1 (17%)
1
0
0
1 (17%)
3
0
0
Acute kidney injury
1 (17%)
2
0
0
Dysuria
1 (17%)
1
0
0
1 (17%)
1
0
0
1 (17%)
1
0
0
Cardiac disorders
Renal and urinary disorders
Vascular disorders Orthostatic hypotension
We show events reported in ≥10% of patients in at least one group. We classified adverse events with the Medical Dictionary for Regulatory Activities 18.1. Patients could be included in more than one category. *Occurred in one man (joint swelling in the knee) with no history of connective tissue disease and with a usual interstitial pneumonia pattern on high-resolution CT at screening while receiving placebo, and were mild and judged to be unrelated to treatment.
Table 3: Most frequent treatment-emergent adverse events in the safety population
All 23 enrolled patients were included in the safety, pharmacodynamic, and intention-to-treat populations. 16 patients were included in the pharmacokinetic population after one patient withdrew from the study due to an adverse event before week 1. Demographics and baseline clinical characteristics were similar in the two treatment groups (table 1). The median age of the overall study population was 66·0 years (IQR 60·0–72·0). All patients were white, and most were men. The median body-mass index of the overall study population was 28·6 kg/m² (IQR 24·8–35·8). No
www.thelancet.com/respiratory Published online May 20, 2018 http://dx.doi.org/10.1016/S2213-2600(18)30181-4
5
Articles
A
Change in plasma LPA C18:2 concentration from baseline (%)
100
Placebo 600 mg GLPG1690
p=0·0030 p=0·0001
50
0
−50
−100
n=5 n=16
n=5 n=15
n=5 n=15
0
1·5 Time after dose at week 4 (h)
6·0
B Change in plasma LPA C18:2 concentration from baseline (%)
200 150 100 p=0·0014 50 0 −50 −100
n=6 n=17 0
n=5 n=16 4
n=6 n=15 12
Time from baseline (weeks)
n=5 n=15 14 (follow-up)
Figure 2: Percentage changes in LPA C18:2 concentrations in plasma in the pharmacodynamic population (A) Change in LPA C18:2 concentrations in plasma from baseline at different times after the dose given at the week 4 visit for placebo or GLPG1690 in the observed-case analysis. (B) Change in LPA C18:2 concentrations in plasma over time from baseline in the placebo group and the GLPG1690 group in the observed-case analysis; changes from baseline were non-significant at week 4 (p=0·29) and week 14 (p=0·60). Data are medians with IQRs (boxes), 1·5 times IQR values (whiskers), mean values (x symbols), and outliers (closed circles). LPA=lysophosphatidic acid.
patient had received nintedanib or pirfenidone before entering the study. Treatment-emergent adverse events were reported in similar proportions of patients in the placebo and GLPG1690 groups (four [67%] and 11 [65%], respectively). Most treatment-emergent adverse events were mild to moderate. Two (12%) patients in the GLPG1690 group had events that were judged to be related to treatment (table 2). The most frequently occurring treatment-emergent adverse events are shown in table 3. Two patients in the placebo group and one in the GLPG1690 group had at least one serious treatment-emergent adverse event: one patient in the placebo group had a urinary tract infection, acute kidney injury, and lower respiratory tract infection; the other patient in the placebo group had atrioventricular block (second degree, leading to discontinuation of study 6
treatment), haemothorax, and pneumothorax; and the patient in the GLPG1690 group had cholangiocarcinoma that resulted in discontinuation of study treatment (symptoms were noted 1 day after the first intake of study drug, but had occurred during screening). No patients died or had acute IPF exacerbations. No clinically relevant findings were seen in the laboratory tests, ECG assessments, or measurements of vital signs over time in either treatment group. The median time to Cmax for GLPG1690 was 4·0 h (IQR 1·5–4·0). Geometric mean steady-state Cmax was 4·4 μg/mL (geometric coefficient of variation 104%) and the geometric mean area under the drug concentration– time curve for the 24 h dosing interval was 40 μg/mL per h (100%). Pharmacodynamic analyses showed that concentrations of LPA C18:2 in plasma decreased after administration of GLPG1690 at the week 4 (figure 2) and week 12 visits and that they returned to baseline concentrations at the follow-up visit (figure 2). Reductions of LPA C18:2 concentrations were seen with GLPG1690 in observed cases at week 4 (maximum percentage reduction from baseline 35·5% for placebo and 89·4% for GLPG1690, p=0·0008) and with area under the effective-time curve over 6 h (104·1% for placebo and 496·0% for GLPG1690, p=0·0007). In BALF, LPA C18:2 and LPA C20:4 concen trations were below the level of quantification for more than 25% of baseline samples obtained from patients in the GLPG1690 treatment group, which prevented interpretation. For other LPA species, concentrations were low in the two treatment groups with no significant difference between them at week 1 (appendix). Mean FVC measured in the study centres decreased over the 12 week treatment period in the placebo group but remained similar to or greater than baseline values in the GLPG1690 treatment group (mean –70 mL [95% CI –208 to 68] in the placebo group and 25 mL [–75 to 124] in the GLPG1690 group with LOCF, figure 3). Further spirometry data are provided in the appendix. Results from observed-case analysis were similar to those with LOCF (mean FVC –87·5 mL [95% CI –345 to 170] for placebo and 8 mL [–101 to 116] for GLPG1690, appendix). The overall patterns of change in FVC assessed by spirometry at home were similar to those seen with study centre spirometry testing (figure 3). Among blood biomarkers (appendix) only surfactant protein A concentrations showed significant changes, with greater increases being seen in the GLPG1690 group than in the placebo group. In BALF samples, autotaxin was undetectable and changes in total cell count were similar in the two treatment groups (appendix). Mean change from baseline in specific airway volume in the observed-case analysis was 3·038 mL/L (95% CI 0·799 to 5·310) in three patients in the placebo group and 0·079 mL/L (–3·722 to 2·631) in 15 in the GLPG1690 group (p=0·0137). Values for specific airway resistance were –0·035 kPa/s (–0·047 to –0·015) in three patients in
www.thelancet.com/respiratory Published online May 20, 2018 http://dx.doi.org/10.1016/S2213-2600(18)30181-4
Articles
A 0·2 p=0·009
Change in FVC (L)
the placebo group and 0·004 kPa/s (–0·045 to 0·056) in 14 in the GLPG1690 group (p=0·0255) in the observedcase analysis. The mean change from baseline in total SGRQ score at week 12 was similar in the two treatment groups (figure 4). For domain scores, mean changes from baseline after 12 weeks of treatment were 2·90 units in the placebo group versus –5·45 in the GLPG1690 group for the symptom domain, 4·14 versus –2·32 units in the activity domain, and –3·90 versus 3·22 units in the impact domain.
0
−0·2
Discussion
Placebo 600 mg GLPG1690 0
4
8 Time from baseline (weeks)
12
14 (follow-up)
B Weekly average home spirometry change in FVC (L)
GLPG1690 is the first autotaxin inhibitor to be assessed as a treatment for IPF and represents a potential novel therapeutic class for this disease. This phase 2a study was designed to explore the effects of GLPG1690 in patients with IPF, with the aim of subsequently doing a larger study of longer duration. Similar proportions of patients in the placebo and GLPG1690 groups had treatmentemergent adverse and serious adverse events and laboratory parameter abnormalities, and, overall, GLPG1690 was well tolerated. No patients had acute IPF exacerbations during the study. The pharmacokinetics and pharmacodynamics of GLPG1690 were similar to those previously reported in healthy individuals.18 Although the study was not designed to investigate efficacy, FVC and functional respiratory imaging results provided promising preliminary efficacy signals. The spirometry results were similar in the two groups irrespective of whether they were measured in the study centres or at home. Among biomarkers of disease activity, we only saw a significant change in concentrations of surfactant protein A (in blood) with GLPG1690. Finally, health-related quality of life overall did not differ between the two treatment groups, although findings for specific domains might indicate a beneficial effect with GLPG1690. The most frequent treatment-emergent adverse events in the GLPG1690 group were infections and infestations (lower respiratory tract infections and nasopharyngitis) and respiratory, thoracic, and mediastinal disorders (cough and dyspepsia). The most frequent events with other IPF pharmacotherapies are nausea in 36% of patients and rash in 28–32% receiving pirfenidone7,8 and diarrhoea in 62–63% and nausea in 23–26% receiving nintedanib,11 and all are probably related to the underlying disease. Of note, treatment exposures have been longer in the studies of pirfenidone and nintedanib than in this study (>12 months vs 12 weeks), although a pooled analysis of nintedanib studies showed that 66% and 82% of participants reported diarrhoea or nausea, respectively, in the first 92 days of treatment.10 Treatment discontinuation data from clinical trials are variable and subject to complicating factors, such as disease severity and comorbidities, but real-world studies seem to support the findings. A US retrospective chart review, however, revealed that 21% of patients discontinued pirfenidone and 26% discontinued
0·2
0
−0·2
−0·4 0
1
2
3
4
5 6 7 Time from baseline (weeks)
8
9
10
11
12
Figure 3: Mean (95% CI) changes in FVC from baseline in the placebo and GLPG1690 groups in the intention-to-treat population (A) Spirometry results from study centre visits; placebo group n=6 and GLPG1690 group n=17; changes were non-significant at weeks 4 (p=0·13), 12 (p=0·3), and 14 (p=0·06). (B) Spirometry results from measurements at home; placebo group n=6 and GLPG1690 group n=16. FVC=forced vital capacity.
nintedanib, mainly because of gastro-intestinal events.14 The small number of patients in our study precluded comparison of the adverse event profile for GLPG1690 with that for pirfenidone and nintedanib and, therefore, firm conclusions cannot be drawn. In exploring the mechanism of action of GLPG1690, we found sustained reductions in plasma concentrations of LPA C18:2 during treatment that reverted to baseline concentrations by the week 14 follow-up visit. Once daily dosing with 600 mg GLPG1690 was sufficient to achieve this sustained effect, which might translate to a potential clinical advantage over current IPF therapies, as pirfenidone needs to be taken three times daily after a 14-day titration period and nintedanib needs to be taken twice daily.23,24 Preclinical data for IPF in mice has shown target engagement of GLPG1690 with various LPA species in BALF.16 In this study, despite robust findings in the plasma analysis, the BALF LPA analysis was hindered by two species being below the limit of detection and practical challenges associated with BALF collection, especially dilution caused by the installation of fluid. Additionally,
www.thelancet.com/respiratory Published online May 20, 2018 http://dx.doi.org/10.1016/S2213-2600(18)30181-4
7
Articles
B
Total
140 120 100 80 60 40 20 0
Symptoms
140 120 100 80 60 40 20 0
C
D
140 120 100 80 60 40 20 0
Impacts
Placebo 600 mg GLPG1690
Activity
A
140 120 100 80 60 40 20 0
0
4
12
14 (follow-up)
Time from baseline (weeks)
0
4
12
14 (follow-up)
Time from baseline (weeks)
Figure 4: Health-related quality of life measured with the St George’s Respiratory Questionnaire in the intention-to-treat population (A) Total score. (B) Symptom score. (C) Activity score. (D) Impact score. Data were available for six patients in the placebo group at all timepoints and for 17 patients for symptom scores and 16 for all other scores in the GLPG1690 group. Data are median values with IQRs (boxes) and 1·5 times IQR values (whiskers), mean values (x symbols), and outliers (closed circles). Differences between groups were non-significant at all timepoints.
LPA analysis has several known challenges, including variability between assay methods and laboratories. Variability is suggested to be due at least partly to LPA generation from other lysophospholipids during ex-vivo handling and processing.25 The method we used aimed to minimise interference with LPA measurement, including LPA production during sample processing. The validity of the exploratory FVC findings in the GLPG1690 group is supported by the results in the placebo group being similar and in line with those seen in other clinical trials involving IPF patients. For example, the mean decline in FVC from baseline in our placebo group at week 12 (–70 mL with LOCF) is similar to that at 3 months in pooled analyses of pirfenidone (approximately −70 mL)26 and nintedanib (approximately −80 mL).27 However, as for the other secondary outcomes, this small study was not powered to detect significant differences between treatment groups and the data should be interpreted with caution. Small proof-of-concept studies have strengths and limitations but are important for disorders, such as IPF, in which the need for more effective therapies remains unmet, because they can quickly support or refute the potential of new treatment options without unnecessarily exposing high-risk patients to investigational therapies with limited benefit. This small study was intended to bridge the gap between the early pharmacokinetic and pharmacodynamic findings for GLPG1690 and to assess the characteristics of the drug in people with IPF before moving to larger trials. Despite the small number of patients (particularly in the placebo group) and the short treatment duration, our findings provided reassurance 8
about the safety of GLPG1690 in patients with IPF. Additionally, useful data were generated for target engagement, pharmacokinetics, and pharmacodynamics, and we generated some preliminary efficacy findings. As is common with small safety studies where the expected number of events is unknown, we did not do power calculations. Although there is no best method for imputation of missing data,28 we used the LOCF method for some endpoints, which might have affected efficacy findings due to the small number of patients. The functional respiratory imaging data should be interpreted with some caution due to the small numbers of patients who had available serial scans, particularly in the placebo group (n=3). This study combined multiple proof-of-concept endpoints with preliminary efficacy assessments. Overall, GLPG1690 was well tolerated by patients with IPF over 12 weeks, showing a similar safety profile to placebo. Pharmacokinetic and pharmacodynamic findings were consistent with those seen in healthy individuals in a phase 1 study,18 and the reductions in LPA C18:2 concentrations in plasma confirmed target engagement. Although efficacy measurements were secondary endpoints and the study was not powered to detect differences between groups for these or other endpoints, our data, particularly for FVC, are encouraging and support further clinical assessment of GLPG1690 as a treatment for IPF. Contributors TMM, EMvdA, OVdS, LA, JD, SD, LF, and WW designed the study. Data were collected by TMM, OVdS, LA, JD (pharmacokinetic data), SD (pharmacodynamic data), LF, and AF. OVdS, LA, JD, SD, PF, and AF analysed the data. All authors were involved in data interpretation, reviewed and revised drafts of the paper, and approved the final draft. Declaration of interests TMM has, via his institution, received industry-academic funding from GlaxoSmithKline and UCB, and has received consultancy or speaker’s fees from Apellis, Astra Zeneca, Bayer, Biogen Idec, Boehringer Ingelheim, Cipla, GlaxoSmithKline, ProMetic, Roche, Samumed, and UCB. EMvdA is an employee of and has received warrants from Galapagos. OVdS works as a contractor for and has received personal fees from Galapagos. LA was an employee of Alten Belgium at the time of the study and worked as a contractor for Galapagos, and has received personal fees from Galapagos. JD was an employee of Galapagos at the time of the study. SD is an employee of and has received warrants from Galapagos. LF, PF, and AF are employees of Galapagos. WW reports travel costs from Galapagos and grants from Boehringer Ingelheim and Roche paid to his institution. Acknowledgments The study was funded by Galapagos, Mechelen, Belgium. We thank all patients, families, and investigators involved in the study. Medical writing support, including development of a draft outline and subsequent drafts in consultation with the authors, assembling tables and figures, collating author comments, copy editing, fact checking and referencing, was provided by Hannah Mace, Aspire Scientific, Bollington, UK, and funded by Galapagos. Simon Hatch, an external medical consultant to Galapagos, provided support with data interpretation and reviewed the drafts of the paper for accuracy. Nadia Verbruggen, Galapagos, Mechelen, Belgium, provided support with statistical analysis and reviewed drafts of the paper for accuracy. TMM is supported by a National Institute for Health Research Clinician Scientist Fellowship (CS-2013-13-017) and a British Lung Foundation Chair in Respiratory Research (C17-3). WW is supported by a Senior Clinical Investigator Fellowship Research Foundation – Flanders.
www.thelancet.com/respiratory Published online May 20, 2018 http://dx.doi.org/10.1016/S2213-2600(18)30181-4
Articles
References 1 Nakanaga K, Hama K, Aoki J. Autotaxin—an LPA producing enzyme with diverse functions. J Biochem 2010; 148: 13–24. 2 Oikonomou N, Mouratis MA, Tzouvelekis A, et al. Pulmonary autotaxin expression contributes to the pathogenesis of pulmonary fibrosis. Am J Respir Cell Mol Biol 2012; 47: 566–74. 3 Montesi SB, Mathai SK, Brenner LN, et al. Docosatetraenoyl LPA is elevated in exhaled breath condensate in idiopathic pulmonary fibrosis. BMC Pulm Med 2014; 14: 5. 4 King TE Jr, Pardo A, Selman M. Idiopathic pulmonary fibrosis. Lancet 2011; 378: 1949–61. 5 Ley B, Collard HR, King TE Jr. Clinical course and prediction of survival in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 2011; 183: 431–40. 6 Raghu G, Weycker D, Edelsberg J, Bradford WZ, Oster G. Incidence and prevalence of idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 2006; 174: 810–16. 7 King TE Jr, Bradford WZ, Castro-Bernardini S, et al. A phase 3 trial of pirfenidone in patients with idiopathic pulmonary fibrosis. N Engl J Med 2014; 370: 2083–92. 8 Noble PW, Albera C, Bradford WZ, et al. Pirfenidone in patients with idiopathic pulmonary fibrosis (CAPACITY): two randomised trials. Lancet 2011; 377: 1760–69. 9 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. 10 Center for Drug Evaluation and Research. Application number: 205832Orig1s000. Medical review(s). Sept 22, 2014. https://www.access data.fda.gov/drugsatfda_docs/nda/2014/205832Orig1s000MedR.pdf (accessed Feb 12, 2018). 11 Nathan SD, Albera C, Bradford WZ, et al. Effect of pirfenidone on mortality: pooled analyses and meta-analyses of clinical trials in idiopathic pulmonary fibrosis. Lancet Respir Med 2017; 5: 33–41. 12 Wuyts WA, Kolb M, Stowasser S, Stansen W, Huggins JT, Raghu G. First data on efficacy and safety of nintedanib in patients with idiopathic pulmonary fibrosis and forced vital capacity of ≤50% of predicted value. Lung 2016; 194: 739–43. 13 Bando M, Yamauchi H, Ogura T, et al. Clinical experience of the long-term use of pirfenidone for idiopathic pulmonary fibrosis. Intern Med 2016; 55: 443–48. 14 Galli JA, Pandya A, Vega-Olivo M, Dass C, Zhao H, Criner GJ. Pirfenidone and nintedanib for pulmonary fibrosis in clinical practice: Tolerability and adverse drug reactions. Respirology 2017; 22: 1171–78. 15 Desroy N, Housseman C, Bock X, et al. Discovery of 2-[[2-Ethyl-6-[4-[2-(3-hydroxyazetidin-1-yl)-2-oxoethyl]piperazin-1-yl]8-methylimidazo[1,2-a]pyridin-3-yl]methylamino]-4-(4-fluorophenyl) thiazole-5-carbonitrile (GLPG1690), a first-in-class autotaxin inhibitor undergoing clinical evaluation for the treatment of idiopathic pulmonary fibrosis. J Med Chem 2017; 60: 3580–90.
16 van der Aar EM, Heckmann B, Blanque R, et al. Pharmacological profile and efficacy of GLPG1690, a novel autotaxin inhibitor for the treatment of idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 2016; 193: A4532. 17 Coornaert B, Duys I, Van Der Schueren J, van der Aar E, Heckmann B. Autotaxin inhibitor GLPG1690 affects TGFβ-induced production of the pro-fibrotic mediators CTGF, IL-6 and ET-1 in fibroblasts. Am J Respir Crit Care Med 2017; 195: A2404. 18 van der Aar EM, Fagard L, Desrivot J, et al. Favorable human safety, pharmacokinetics and pharmacodynamics of the autotaxin inhibitor GLPG1690, a potential new treatment in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 2016; 193: A2701. 19 Raghu G, Collard HR, Egan JJ, et al. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med 2011; 183: 788–824. 20 Miller MR, Hankinson J, Brusasco V, et al. Standardisation of spirometry. Eur Respir J 2005; 26: 319–38. 21 Jones PW, Quirk FH, Baveystock CM, Littlejohns P. A self-complete measure of health status for chronic airflow limitation. The St. George’s Respiratory Questionnaire. Am Rev Respir Dis 1992; 145: 1321–27. 22 Swigris JJ, Brown KK, Behr J, et al. The SF-36 and SGRQ: validity and first look at minimum important differences in IPF. Respir Med 2010; 104: 296–304. 23 Boehringer Ingelheim. Ofev highlights of prescribing information. January, 2018. http://docs.boehringer-ingelheim.com/Prescribing% 20Information/PIs/Ofev/ofev.pdf (accessed Feb 12, 2018). 24 Genentech. Esbriet highlights of prescribing information. October, 2017. https://www.gene.com/download/pdf/esbriet_ prescribing.pdf (accessed Feb 12, 2018). 25 Ononato JM, Shipkova P, Minnich A, Aubrey AF, Easter J, Tymiak A. Challenges in accurate quantitation of lysophosphatidic acids in human biofluids. J Lipids Res 2014; 55: 1784–96. 26 Noble PW, Albera C, Bradford WZ, et al. Pirfenidone for idiopathic pulmonary fibrosis: analysis of pooled data from three multinational phase 3 trials. Eur Respir J 2016; 47: 243–53. 27 Richeldi L, Cottin V, du Bois RM, et al. Nintedanib in patients with idiopathic pulmonary fibrosis: Combined evidence from the TOMORROW and INPULSIS® trials. Respir Med 2016; 113: 74–79. 28 Thabut G, Crestani B, Porcher R, Richeldi L. Missing data in IPF trials: do not let methodological issues undermine a major therapeutic breakthrough. Eur Respir J 2015; 46: 607–14.
www.thelancet.com/respiratory Published online May 20, 2018 http://dx.doi.org/10.1016/S2213-2600(18)30181-4
9
