Disease progression in COPD: What is it? How should it be measured? Can it be modified? Professor Paul Jones MD, PhD, FERS Emeritus Professor of Respiratory Medicine; St George s, University of London Global Medical Expert; GSK* * Financial conflict of interest
FEV 1 (% of value at age 25) Disease progression as measured by FEV 1 100 Never smoked or not susceptible to smoke 75 50 Disability Smoked regularly and susceptible to its effects Stopped at 45 25 Death Stopped at 65 0 25 Age (years) 50 75 Fletcher et al, 1976.
Rate of decline of FEV 1 : the ISOLDE trial FEV 1 (litre) 1.50 1.45 1.40 1.35 1.30 76 ml Fluticasone propionate* Placebo 50 ml/year p=0.16 1.25 1.20 59 ml/year 0 3 1 2 2 4 3 6 Months * Not licensed for COPD Burge et al BMJ 2000; 320:1297
Rate of decline in FEV 1 (ml/yr) TORCH: Exacerbation rate and FEV 1 decline Exacerbations per year n=1735 n=1862 n=1306 P<0.001 37% faster 65% faster Adjusted for smoking status, gender, baseline FEV 1, region, BMI, prior exacerbations, treatment, time, time by treatment and covariate by time Celli et al AJRCCM 2008; 178: 332
Exacerbations and worsening in health status over 3 years 3.0 ANOVA p<0.0003 SGRQ slope (units/yr) 2.5 2.0 1.5 1.0 Getting worse faster 0.5 0 None in 3 years Infrequent <1.65/yr Frequent >1.65/yr Exacerbation Category Spencer et al. Eur Respir J 2004;23:1-5.
Change in steps per day over 3 years SE Daily activity is lower after 3 years with all degrees of severity Washki et al Am J Resp Crit Care Med 2015; 192: 295-30
FEV 1 and SGRQ changes FEV 1 (% predicted) SGRQ total score 52 51 50 49 50 48 46 44 48 42 47 40 46 38 45 36 44 34 43 32 42 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Time (years) 30 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Time (years) SGRQ, St George s Respiratory Questionnaire Five-year study of male outpatients with stable COPD (n=137) Oga et al; Respir Med; 2007; 101; 146-153
ECLIPSE: Post-bronchodilator FEV1 rate of decline over 3-years 40 38 35 31 Normal or no decline 30 % patients 25 20 23 15 10 8 5 0 Fast decline >40ml Slow decline 20-40ml No change +/- 20ml Improvement >20ml Vestbo et al; NEJM 2011; 365:1184-92.
Half of people with COPD (as determined by FEV 1 ) do not appear to have abnormal rates of FEV 1 decline FEV 1 decline in early COPD cohort 1 TR3 = 53 ml/yr (47% of COPD population) TR4 = 27 ml/yr (53% of COPD population) At time of diagnosis no knowledge of how the patient reached this point Reference: 1.Adapted from Lange P et.al. NEJM 2015 Vol 373 : 111-122 2. supplementary appendix
Long-term trial issues
TORCH: Early study withdrawal on placebo: FEV1 rate of decline Healthy survivor issue Placebo patients with 3-years of data annual decline = 54 ml/year 2 Early withdrawal patients annual decline = 76 ml/year 2 Annual decline in last 6-months leading to withdrawal >100ml/year 1. Adapted from Vestbo J et al Clin. Respir J (2011) 5. 44-49 2. Celli B et al Am J Respir Crit Care Med (2008) 178. 332 338
Clinically Important Deterioration (CID) - Rationale
A composite approach to assess shortterm worsening in COPD Measurable deterioration Loss of lung function Occurrence of first exacerbation Deterioration in health status Naya I et al. Thorax. 2015; 70(3): A34 (S57)
Clinically Important Deteriorations (CID) in COPD Decrease of 100 ml from baseline in trough FEV 1 and/or Deterioration in SGRQ 4 units and/or Moderate/severe COPD exacerbation
Long term risk based on composite CID status (+) or (-) at 6-months (TORCH post hoc analysis): Placebo n=1524 Salmeterol n=1521 At 6 months CID+ (n=2870 [54%]) Fluticasone n=1534 FP/ SAL combination n=1533 Reason for CID+ status Exacerbation = 33% FEV 1 = 23% SGRQ = 17% 2 or more causes = 27% Naya I et al. Thorax. 2015; 70(3): A34 (S57) Naya et al; European Respiratory Society Congress 2016 (PA304)
FEV1 Mean change (95% CI) SGRQ mean change (95% CI) TORCH: 3-year outcome on FEV 1 & SGRQ based on composite CID status at 6-months FEV 1 (ml) deterioration over time SGRQ total score deterioration over time 150 6 100 Patients CID - 4 Patients CID + 50 0-40 10 60 110 160-50 -100 Patients CID + - 117 ml p<0.001 2 0-40 10 60 110 160-2 -4-6 -8 Patients CID - + 6.4 units p<0.001-150 Time after day 182 (weeks) -10 Time after day 182 (weeks) Naya I et al. Thorax. 2015; 70(3): A34 (S57), DOF: RF/CPD/0041/16
Subjects with 1 exacerbation (%) All cause death (% subjects) TORCH: 3-year risk assessments based on composite CID status at 6-months Future risk of exacerbations on treatment Future risk of all cause mortality on treatment Patients CID+ [73%] Patients CID+ [8.3%] Patients CID- [60%] CID+ patients had a 61% increased risk of an exacerbation (p<0.001) Median time to event 520 days [CID-] vs. 265 days [CID+] Patients CID- [6.6%] CID+ patients had a 41% increased risk of all-cause death (p<0.001) Time after day 182 (weeks) Time after day 182 (weeks) All treatment groups combined in TORCH. At 6-Mo. 2870 [54%] patients were CID (CID+) and 2422 [46%] were (CID-). Naya et al; European Respiratory Society Congress 2016 (PA304)
CID in short-term trials
Time to first clinically important deterioration: Relative Risk (RR) reduction RR: 43% (95% CI 31, 53; p<0.001) UMEC/VI 55/22mcg TIO 18mcg Singh et al. Int J COPD; 2016; 11; 1413 Post-hoc analysis of ZEP117115: Maleki-Yazdi M et al. Respir Med 2014; 108:1752 1760
All components of the composite endpoint were significantly better with UMEC/VI vs. tiotropium Proportion of patients (%) RR: 43% (p<0.001) Proportion of patients with a first clinically important deterioration event RR: 26% p=0.025 RR: 53% (p<0.001) RR: 47% (p=0.044) Composite Exacerbations SGRQ 4 units FEV 100 ml decline UMEC/VI 55/22mcg TIO 18mcg *RR = Relative Risk reduction Singh et al. Am J Respir Crit Care Med 2015:191; A5760 (poster presentation) Post-hoc analysis of ZEP117115: Maleki-Yazdi M et al. Respir Med 2014; 108:1752 1760
Preventing short-term worsening: comparison between dual and mono bronchodilator therapy Post hoc analysis of time to a first composite CID (A) Dual bronchodilator therapy (UMEC/VI) vs. tiotropium (B) Dual bronchodilator therapy (UMEC/VI) vs. placebo (C) Open triple therapy (ICS/LABA + UMEC) vs.ics/laba + placebo Tiotopium n=869 UMEC/VI n=878 Placebo n=280 UMEC/VI n=413 75% ICS/LABA + PBO n=818 ICS/LABA + UMEC n=819 56% 68% 41% 44% 46% HR 0.62 (0.54, 0.71) P<0.001 Singh et al ERS abstract PA1487 HR 0.37 (0.30, 0.45) P<0.001 0 50 100 150 Time to event (days) Malaki-Yazdi et al ERS abstract PA1001 HR 0.52 (0.45, 0.59) P<0.001 Singh et al ERS abstract PA1487 21
Summary All components contribute to CID worsening CID distinguishes between treatments in the short-term Short term worsening measured over 6 months may predict long-term outcome Prevention of short-term worsening looks a promising therapeutic target As a surrogate for long term trials To identify potential disease modifying treatments more quickly