Update on COPD & Asthma Disclosures · 2016. 6. 2. · 1 Update on COPD & Asthma Michael C. Peters,...
Transcript of Update on COPD & Asthma Disclosures · 2016. 6. 2. · 1 Update on COPD & Asthma Michael C. Peters,...
1
Update on COPD & Asthma
Michael C. Peters, M.D. MASDivision of Pulmonary & Critical Care Medicine
Cardiovascular Research InstituteUniversity of California San Francisco
UCSF Primary Care MedicineSan Francisco, CA
May 27, 2016
Disclosures
• No Pharma Disclosures
• NHLBI - Asthma Clinical Research Network
• NHLBI – Severe Asthma Research Program
• Parker B. Francis Foundation
Update on the Management of COPD
2
To review COPD
• COPD is a leading cause of death worldwide, and mortality is increasing
• Exacerbations are the major complication of COPD•Associated with increased loss of lung function•And Mortality
• There are effective strategies for decreasing exacerbations
• COPD = Inflammatory Disease
• O2 therapy
• Pharmacologic Therapy: (“it’s not just for symptoms anymore”)
- Decreasing exacerbations- Change natural history?
•Pulmonary Rehab: reduces symptoms, depression, health care utilization; improves Q of L, exercise
COPD
• Smoking Cessation modifies natural history(lung function, mortality)
3
Question #1: Which of the following is NOT true?
1. COPD mortality has plateaued
2. Hospitalization for exacerbation predicts mortality
3. Most exacerbations are caused by infection
4. There are effective strategies for decreasing exacerbations
Question #1: Which of the following is NOT true?
1. COPD mortality has plateaued
2. Hospitalization for exacerbation predicts mortality
3. Most exacerbations are caused by infection
4. There are effective strategies for decreasing exacerbations
Percent Change in Age-Adjusted Death Rates (US, 1965–1998)
Proportion of 1965 Rate
0.0
0.5
1.0
1.5
2.0
2.5
3.0
1965 – 1998 1965 – 1998 1965 – 1998 1965 – 1998 1965 – 1998–59% –64% –35% +163% –7%
CHD Stroke Other CVD COPD All othercauses
Hey Doc, Do I Have COPD????
Simel and RennieEvidence-based Clinical DiagnosisMcGraw Hill, 2008
•CHRONIC Obstructive Pulmonary Disease• NEED SPIROMETRY: FEV1/FVC < 0.70
• Physical Exam:>90% SpecificityPoor Sensitivity
• > 55 Pack Years• Wheezing on Auscultation• Self-reported wheezing
Likelihood Ratio: 156
High Probability For COPD
4
Hey Doc, Do I Have COPD????
Simel and RennieEvidence-based Clinical DiagnosisMcGraw Hill, 2008
•CHRONIC Obstructive Pulmonary Disease• NEED SPIROMETRY: FEV1/FVC < 0.70
• Physical Exam:>90% SpecificityPoor Sensitivity
• > 55 Pack Years• Wheezing on Auscultation• Self-reported wheezing
Likelihood Ratio: 156
High Probability For COPD
Respiratory Symptoms Smokers with Normal Pulmonary Function
Woodruff PG et a l. N Engl J Med 2016;374:1811-1821
Symptom Scores
Prevalence of Symptoms and Risk of Respiratory Exacerbations
Woodruff PG et a l. N Engl J Med 2016;374:1811-1821
20%
Anthonisen et alJAMA 272:1497-505, 1994
• No benefit of screening adults with no symptoms
• No evidence that treating asymptomatic individuals prevents future symptoms, or reduces the subsequent decline in lung function.
Qaseen, Ann Int Med 155:179-91, 2011
5
• Other: – Proteases/inflammation– Repetitive bacterial/viral infections – Genetics, especially a1-antitrypsin
deficiencyNHLBI/WHO Global Initiative for Chronic Obstructive Lung Disease. April 2001; (Updated 2003).American Thoracic Society Statement Statement. Am J Respir Crit Care Med. 1995;152(suppl 5):S77-S120.
Risk Factors for COPD Give it to me Straight. Is it BAD?
GOLD Guidelines 2007
GOLD 1: (Mild COPD) FEV1 > 80% predicted
FEV1/FVC < 0.70
GOLD 2: (Moderate COPD) FEV1 50-80% predicted
GOLD 3: (Severe COPD) FEV1 30-50% predicted
GOLD 4: (Very Severe COPD) FEV1 <30% predicted
GOLD 2007
N = 2164 stable COPDN = 337 “Healthy Smokers”N = 245 Never Smokers
Characterized Extensively at:Baseline3, 6, 12, 18, 24, 30, 36 months
Evaluation of COPD Longitudinally to Identify Predictive Surrogate End-Points
(ECLIPSE)Eur Respir J 2008; 31:869-73
2007 Gold Guidelines Not Good Enough
Respir Res 2010; 11:122
Agusti Respir Res 2010; 11:122
Symptom Scores
6
Respir Res 2010; 11:122
Agusti Respir Res 2010; 11:122
2007 Gold Guidelines Not Good Enough COPD Assessment: A New Model
Risk
GO
LD
Clas
sifi
cati
on
of A
irfl
ow
Lim
itat
ion
(C) (D)
(B)(A)
4
3
2
1
≥2 or
1
0
Risk
Exac
erba
tion
Hist
ory
mMRC 0-1CAT < 10
mMRC ≥ 2CAT ≥10Symptoms
(mMRC or CAT score)
When assessing risk, choose the highest risk according to GOLD grade or exacerbation history
Patient Category
Characteristics Spirometric Classification
Exacerbations per year
mMRC CAT
A Low Risk, Less Symptoms GOLD 1-2 ≤1 0-1 <10
B Low Risk, More Symptoms GOLD 1-2 ≤1 ≥2 ≥10
C High R isk, Less Symptoms GOLD 3-4 ≥2 0-1 <10
D High R isk, More Symptoms GOLD 3-4 ≥2 ≥2 ≥10
GOLD Guidelines 2015
≥1 leading to hospital admission
(no hospital admission)
Risk
GO
LD
Clas
sifi
cati
on
of A
irfl
ow
Lim
itat
ion
(C) (D)
(B)(A)
4
3
2
1
≥2 or
1
0
Risk
Exac
erba
tion
Hist
ory
mMRC 0-1CAT < 10
mMRC ≥ 2CAT ≥10Symptoms
(mMRC or CAT score)
When assessing risk, choose the highest risk according to GOLD grade or exacerbation history
GOLD Guidelines 2015
≥1 leading to hospital admission
(no hospital admission)
Patient Category
Characteristics Spirometric Classification
Exacerbations per year
mMRC CAT
A Low Risk, Less Symptoms GOLD 1-2 ≤1 0-1 <10
B Low Risk, More Symptoms GOLD 1-2 ≤1 ≥2 ≥10
C High R isk, Less Symptoms GOLD 3-4 ≥2 0-1 <10
D High R isk, More Symptoms GOLD 3-4 ≥2 ≥2 ≥10
GOLD Guidelines 2015
7
Hospitalized Severe AECOPD and Mortality:Severity of AECOPD
1- no AECOPD 2- AECOPD ED
N = 305 men with COPDx 5 years
Soler-Cataluna Thorax 2005
3- AECOPD Hosp4- AECOPD Readmit
Question #2:Which of the Following Is the Best
Predictor of a Future Acute Exacerbations of COPD?
1. Spirometry2. Symptoms3. Smoking Status4. Socio-Economic Status5. Prior Exacerbation History
Predictors of Acute Exacerbations of COPD
Number of Exacerbations≥2 vs. 0 1 vs. 0
Odds Ratio (95% CI) Odds Ratio (95% CI)
Exacerbation in Prior Year 5.7 (4.5-7.3) 2.2 (1.8-2.8)FEV1 per 100ml decrease 1.1 (1.08-1.1) 1.1 (1.0-1.1)
SGRC (symptom score) per 4 points
1.1 (1.0-1.1) 1.1 (1.0 – 1.1)
GERD 2.1 (1.6-2.7) 1.6 (1.2-2.1)
WBC Count 1.1 (1-1.1) 1.1 (1.0-1.1)
Hurst NEJM 2010
Acute Exacerbations of COPD• Some patients seldom exacerbate
• Some patients exacerbate frequently
• Best predictor of ≥2 AECOPD/year (“Frequent Exacerbator”) = previous frequent exacerbations
• Spirometry does not correlate well with clinical features of disease
• “Frequent Exacerbator” is a stable phenotype
8
COPD Exacerbations
• “Exacerbations are to COPD what myocardial infarctions are to coronary artery disease”
• “They are the acute, often trajectory-changing, and sometimes deadly manifestations of a chronic disease”
- Gerard J Criner, MDTemple University School of Medicine
Philadelphia, PA, USA
The Battle Plan.
• Improve Symptoms
• Prevent Progressive Loss of Lung Function
• Prevent Acute Exacerbations
COPD Exacerbations (AECOPD): The Major Complication of COPD
• Characterized by episodic increases in dyspnea, sputum production and cough
• 16 million office visits/year
• 500,000 hospitalizations/year
• 110,000 deaths/year• $18 billion in direct health care costs
Mannino et al. MMWR Surveill Summ 2002; 51:1-16NHLBI: http://www.nhlbi.gov/resources/docs/02_chtbk.pdf
9
Question #3:Which of the Following DOES NOT
Reduce Acute Exacerbations of COPD?
1. Inhaled Corticosteroids2. Long Acting Beta Agonist3. Long Acting Muscarinic Agonists4. Azithromycin5. EMR training
Question #3:Which of the Following DOES NOT
Reduce Acute Exacerbations of COPD?
1. Inhaled Corticosteroids2. Long Acting Beta Agonist3. Long Acting Muscarinic Agonists4. Azithromycin5. EMR training
Prevention of AECOPDAmerican College of Chest Physicians & Canadian
Thoracic Society Guideline• PICO (population, intervention, comparator,
outcome)
• Literature Search
• Quality Assessment (AGREE II, DART)
• Grading Evidence (GRADEpro)• Recommendations (CHEST)
Criner et al. CHEST 147:894-942, 2015
Prevention of AECOPDRecommendations
• Influenza Vaccine (Grade 1B)
• Pulmonary Rehab (Grade 1C)
• Smoking Cessation (Grade 2C)
• Pneumococcal Vaccine (Grade 2C)Mod-severe-very severe; recent AECOPD<4 weeks
Criner et al. CHEST 147:894-942, 2015
Non-Pharmacologic Treatments/Vaccinations:
10
• LAMA vs PBO (Grade 1A)
• LABA vs PBO (Grade 1B)
• LAMA vs LABA (Grade 1C)
• COMBO Therapy vs MonoTherapy (Grade
1B,C)
Criner et al. CHEST 147:894-942, 2015
Maintenance Inhaled Therapy:
Prevention of AECOPDRecommendations
• Macrolide (Grade 2A)(Frequent AECOPD despite Tx)
• Systemic Corticosteroids (Grade 2B)(For AECOPD – prevent next 30 days)
• Roflumilast (Grade 2A)(Chr Bronchitis, ≥1 AECOPD in year)
• Do not use statins for AECOPD (Grade 1B)
Criner et al. CHEST 147:894-942, 2015
Oral Therapy:
Prevention of AECOPDRecommendations
NEJM 365:689-98, 2011
• NHLBI – COPD Clinical Research Network
• N = 1130
• Moderately-severe COPDFEV1/FVC < 70%; FEV1 <80%
• “Exacerbation Prone”
• Primary Outcome: Time to first AECOPD
The MACRO Study(Azithromycin 250mg/day x 1 year)
NEJM 365:689-98, 2011
11
Rates of Acute Exacerbations of Chronic Obstructive Pulmonary Disease per Person-Year, According to Study Group.
Albert RK et al. NEJM 2011
Macrolides Decrease AECOPD
Ray WA et a l. N Engl J Med 2012;366:1881-1890
Ray WA et al. NEJM 2012
Macrolides May Increase risk of Cardiovascular Death
• Macrolides can prolong QT and QTc leading to arrhythmias, including torsades de pointes
• Most arrhythmias with macrolides occur in patients with underlying risk factors
• Incidence of arrhythmias in absence of additional risk factors is very low, perhaps 1 in 100,000.
Mosholder, NEJM 2013
Am J Respir Crit Care Med2014; 189:1173-1180
“Macrolide-associated arrhythmias can be reduced by not prescribing to patients with comorbidities of concern…the majority of which can be discovered by:
• History• ECG before initiating therapy• ECG a short time after initiating therapy”
Am J Respir Crit Care Med2014; 189:1173-1180
12
Ray WA et a l. N Engl J Med 2012;366:1881-1890
Roflumilast
• Oral Tablet• 500 ug Once Daily• Phosphodiesterase-4
Inhibitor
Articles
www.thelancet.com Vol 385 March 7, 2015 861
The scientifi c oversight of the study was provided by a steering committee responsible for providing scientifi c advice about the study design, execution, interpretation, and publication of results. A major adverse cardiovascular event adjudication committee, comprising independent cardiologists, adjudicated all cardiovascular events in a masked manner (appendix p 6). As sponsor of this study, Takeda (Takeda Development Centre Europe Ltd, London, UK) was responsible for study oversight and overall project management. PPD Global Ltd (Cambridge, UK) managed the administration, coordination, and monitoring of the study, including data management, statistical analysis, and the Interactive Voice Response System–Interactive Web Response System. SAS version 9.1.3 was used for all statistical analyses.
This study is registered with ClinicalTrials.gov, number NCT01329029.
Role of the funding sourceThe study was funded by Takeda. The steering committee, consisting of four academic investigators (PMAC, KFR, LMF, and FJM) and two employees of Takeda (U-MG and MB), developed the design and concept of the studies, approved the statistical plans, had full access to and interpreted the data, wrote the report, and had responsibility for the decision to publish the report. Data collection was coordinated by the two employees of Takeda (U-MG and MB). An academic author (FJM) wrote a draft of the report and an employee of Takeda (MB) did the statistical analysis. All authors vouch for the accuracy and completeness of the data and the analyses. Takeda did not place any restrictions on the academic authors regarding statements made in the fi nal report. The corresponding author had full access to all the data in the study and fi nal responsibility to submit for publication.
ResultsPatient recruitment began on April 3, 2011, and the study ended on May 27, 2014. Of 2708 patients recruited, 1945 were randomly assigned and 1935 actually received treatment (969 in the rofl umilast group and 966 in the placebo group; fi gure 1). Table 1 shows the demographic and baseline characteristics of the randomly assigned patients who received at least one dose of study medication. The mean pre-bronchodilator FEV1 was 1·0 L (SD 0·32) and the mean post-bronchodilator FEV1 was 1·1 L (SD 0·33). As anticipated in view of the inclusion criteria, 1900 (98%) of 1935 patients were using a combination of inhaled corticosteroid–longacting β2 agonist according to the protocol. 1346 (70%) of 1935 patients were also using a longacting muscarinic antagonist during the course of the study, with similar numbers in each study group. Despite the use of these inhaled therapies, study participants had a history of frequent exacerbations and impaired health status.
Figure 1 shows patient disposition throughout the study. The patient withdrawal rate was similar in both treatment
groups (269 [28%] of 969 in the rofl umilast group vs 192 [20%] of 966 in the placebo group). However, more patients withdrew in the fi rst 12 weeks post-randomisation in the rofl umilast group than in the placebo group (appendix p 10). Adherence to treatment was very high (99%) in both groups. 312 (16%) of 1935 participants had protocol violations (appendix p 15), which were mainly failures to meet the spirometric entry criteria.
Figure 2 illustrates and table 2 enumerates the eff ect of rofl umilast versus placebo on the rate of moderate-to-severe exacerbations analysed with Poisson regression and negative binomial regression in the intention-to-treat and per-protocol populations. The numerical reductions were similar in both analyses. In the intention-to-treat population, the frequency of moderate-to-severe exacer-bations was 13·2% lower in the rofl umilast group than in the placebo group in the Poisson regression analysis (rate ratio [RR] 0·868 [95% CI 0·753–1·002], p=0·0529), and was 14·2% lower (0·858 [0·740–0·995], p=0·0424) in the negative binomial regression analysis. The reduction in the moderate-to-severe exacerbation rate was greater in the per-protocol population (analysed with Poisson regression) than in the intention-to-treat population (also analysed with Poisson regression; fi gure 2). Importantly, the eff ect of rofl umilast was similar irrespective of concomitant treatment with a longacting muscarinic antagonist (appendix p 16). Table 2 also shows the eff ects of rofl umilast treatment according to predefi ned, alternative defi nitions for exacerbations.
In view of the small number of anticipated events, we analysed the eff ect of rofl umilast on severe exacerbations and on those necessitating hospital admission using
Figure 2: Mean rate of moderate or severe chronic obstructive pulmonary disease exacerbations per patient per yearRate ratios, 95% CIs, and p values are based on a Poisson regression analysis.
0·921
0·742
Number at riskPatients with at leastone exacerbation (n)
Rate ratio (95% CI)Two-sided p value
Intention to treat
Placebo 432; roflumilast 380
0·868 (0·753–1·002)0·0529
Placebo 369; roflumilast 310
0·806 (0·688–0·943)0·0070
Per protocol0
0·1
0·2
0·3
0·4
0·6
0·8
0·9
1·0
0·5
0·7M
ean
rate
of c
hron
ic ob
stru
ctiv
e pul
mon
ary d
iseas
eex
acer
batio
ns p
er p
atie
nt p
er ye
ar
Placebo group Roflumilast group
0·927
0·805
Martinez et al. Lancet 2015
Side Effects, GIDiarrheaWeight LossNausea
N Engl J Med. 2014 Jun 5;370(23):2201-10
Effect of Corticosteroids on Treatment Failure Rates after AE COPD
Niewoehner et al., NEJM 340:1941, 1999
2 week = Solumedrol 125mg q6hr x 3d, Prednisone 60mg qd x 4d, 40mg qd x 4d,20mg qd x 4d
8 week = additional 10mg qd x 5 week, then 5 mg qd x 1 week
Rate
of T
reat
men
t Fai
lure
(%)
Month0 1 2 3 4 5 6
60
50
40
20
0
10
308 week2 weekPlacebo
Leuppi et alJAMA 2013; 309:2223-2231
• Randomized, noninferiority multicenter trial• N = 314, ED with AECOPD• Prednisone, 40 mg/day x 5 days
vs• Prednisone, 40 mg/day x 14 days
13
Time to Reexacerbation of COPD
(Intention-to-treat) (Per-Protocol)
Leuppi et al.JAMA 2013;309(21):2223-2231
Summary
• Azithromycin prevents COPD Exacerbations– Potential Risk of Cardiac Arrhythmias
• Roflumilast offers some benefit in bronchitis patients
• 5 days of corticosteroids is the appropriate time frame
• No indication for statins in preventing AECOPD
Goals of TreatmentFor Primary Care Physicians
• Improve Symptoms
• Prevent Progressive Loss of Lung Function
• Prevention of Acute Exacerbations
Decline in FEV1 in COPD
Fletcher and PetoBMJ, 1977;1:1645-1648
14
Smoking Cessation: the Lung Health StudyAnthonisen et al. JAMA 272:1497 (1994)
Research Question:Does smoking intervention, ± ipratropiumchange the course of “mild” COPD
n = 5887 smokers; ages 35-60 (mean 48); FEV1 = 63%
Effect of Smoking Cessation on FEV1
JAMA 272:1497,1994.
Sustained Quitters2.9
2.8
2.7
2.6
2.5
2.4
Continuing Smokers
Follow-up in years
1 2 3 4 5Screen 2
.
Post
Bro
nchd
ilato
r FEV
1(li
ters
)
Effects of a Smoking Cessation Intervention on 14.5-year Mortality
Anthonisen et alAnn Intern Med 2005; 142:233-239
P=0.03
Smoking Cessation
Usual Care
Celli et alAm J Respir Crit Care Med 178:332-38, 2008
Therapy Reduces Lung Decline
(TORCH)
Placebo
Salmeterol + Fluticasone
15
Tashkin et alNEJM 359:1543-54, 2008
Tiotropium Reduces Lung Decline
Goals of TreatmentFor Primary Care Physicians
• Improve Symptoms
• Prevent Progressive Loss of Lung Function
• Prevention of Acute Exacerbations
Goals of TreatmentFor Primary Care Physicians
• Improve Symptoms– LABA– LAMA– ICS
• Prevent Progressive Loss of Lung Function
• Prevention of Acute Exacerbations
Downward Spiral In FunctionAssociated With COPD
Disease
Dyspnea
Inactivity
Deconditioning
16
Pulmonary Rehabilitation
• Benefits all levels of disease severity • Reduces respiratory symptoms • Reduces anxiety and depression • Reduces medical and hospital usage • Improves exercise performance • Improves quality of life• Is typically provided as outpatient• Can be initiated as an inpatient until functional
ability has improved
Update on the Management of Asthma
Definition of Asthma• Obstruction that is reversible either spontaneously or with treatment; [NAEPP-EPR, 1991]
• Chronic inflammatory disorder (MCs, Eos, Tcells, Macs, PMNs, Epi); variable obstruction; [NAEPP-EPR2, 1997]
• Variable symptoms, obstruction, BHR; inflammation; interaction [NAEPP-EPR3, 2007]
17
Definition of Asthma• Chronic inflammatory disorder; many different cells; BHR; variable/reversible symptoms and obstruction; phenotypes? [GINA, 2011]
• Heterogeneous; Chronic airway inflammation; variable/reversible symptoms and obstruction;•Different phenotypes or clusters [GINA, 2014]
45M
anaging Asthma Long Term
FIGURE 16. STEPWISE APPROACH FOR MANAGING ASTHMA IN YOUTHS ≥12 YEARS OF AGE AND ADULTS
Key: Alphabetical order is used when more than onetreatment option is listed within either preferred oralternative therapy. ICS, inhaled corticosteroid; LABA, long-acting inhaled beta2-agonist; LTRA, leukotriene receptorantagonist; SABA, inhaled short-acting beta2-agonist
Notes:
• The stepwise approach is meant to assist, not replace, theclinical decisionmaking required to meet individual patientneeds.
• If alternative treatment is used and response is inadequate,discontinue it and use the preferred treatment before stepping up.
• Zileuton is a less desirable alternative due to limited studies as adjunctive therapy and the need to monitor liver function. Theophylline requires monitoring of serumconcentration levels.
• In step 6, before oral corticosteroids are introduced, a trialof high-dose ICS + LABA + either LTRA, theophylline, orzileuton may be considered, although this approach hasnot been studied in clinical trials.
• Step 1, 2, and 3 preferred therapies are based on EvidenceA; step 3 alternative therapy is based on Evidence A forLTRA, Evidence B for theophylline, and Evidence D forzileuton. Step 4 preferred therapy is based on Evidence B,and alternative therapy is based on Evidence B for LTRAand theophylline and Evidence D zileuton. Step 5 preferred therapy is based on Evidence B. Step 6 preferredtherapy is based on (EPR—2 1997) and Evidence B for omalizumab.
• Immunotherapy for steps 2–4 is based on Evidence B forhouse-dust mites, animal danders, and pollens; evidence isweak or lacking for molds and cockroaches. Evidence isstrongest for immunotherapy with single allergens. The roleof allergy in asthma is greater in children than in adults.
• Clinicians who administer immunotherapy or omalizumabshould be prepared and equipped to identify and treat anaphylaxis that may occur.
EPR-3, NHLBI, 2011
described an early-onset, symptom-predominant group withminimal eosinophilic disease. Cluster 4 described an eosino-philic inflammation–predominant group with few symptoms,late-onset disease, and a greater proportion of males.
Discriminant function modeling identified the majority ofinput parameters used in the cluster analysis of both popula-tions to be significant determinants of cluster membership(Table E1 of the online supplement). The discriminant functionmodel of primary-care and refractory asthma clusters requiredseven of eight input parameters (excluding atopic status) andfive of seven parameters (excluding atopic status and sex),respectively. The accuracy of the discriminant function modelsfor predicting cluster membership was 94.6% (primary care)and 96.8% (refractory asthma).
Cluster analysis was performed from baseline data in 68patients of the prospective study dataset. Three clusters wereidentified (Table E2); all were comparable with clusters ob-served in the larger refractory asthma population. The originalstudy demonstrated a significant reduction in severe exacerba-tion frequency in the sputum arm, with no significant differencein corticosteroid usage between the groups. The present cluster-specific analysis revealed that all of the benefit for preventingexacerbations occurred in the inflammation-predominant co-hort (3.53 [SD, 1.18] vs. 0.38 [SD, 0.13] exacerbation/patient/yr,P 5 0.002) (Table 4). In addition, sputum-guided therapyallowed successful downtitration of corticosteroid therapy inearly symptom-predominant asthma (Table 4; mean difference,1,829 mg beclomethasone equivalent/d [95% confidence interval,307–349 mg]; P 5 0.02), without compromising asthma control.
A univariate ANOVA with the cluster model as a covariateidentified both treatment grouping and the cluster model assignificant determinants for observed differences in exacerba-tion frequency (P 5 0.002, study groups; P 5 0.03, clustermodel), but only the cluster model was a significant determinantfor differences in inhaled corticosteroid dose (P 5 0.07 fortreatment groups and P 5 0.005 for cluster model).
DISCUSSION
The need for classifying asthma heterogeneity has gainedurgency with the parallel development of better tools for mea-suring disease characteristics that highlight disparity in clinical,physiologic, and pathologic markers, together with novel andspecific molecular therapies that are only likely to be efficaciousin particular subgroups of asthma. This study is the first to applyprinciples of cluster analysis for the identification of clinicalasthma phenotypes. We have further shown that phenotypesconstructed in this way exhibit clinically relevant differences inoutcome, with management strategies that use a measure ofeosinophilic inflammation for titrating corticosteroid therapy.
Asthma classification is complicated by the multidimensionalnature of the disease. This prompted our consideration ofcluster analysis techniques for this purpose. We selected the k-means clustering algorithm as it maximizes separation betweenclusters, thereby offering the greatest scope for identifyingdistinct groups within the population. Both familiar and pre-viously uncharacterized asthma subgroups were identified thatare more representative of multidimensionality. The identifica-
Figure 1. Clinical phenotypes of asthma. A summary of phenotypes identified using cluster analysis in primary- and secondary-care asthmapopulations. The clusters are plotted according to their relative expression of symptoms and inflammation because these are the two clinically pertinentand modifiable dimensions of the disease. The plot highlights greater discordance to be a feature of secondary-care asthma. Although reasons forthis dissociation are unclear, the use of measures of airway inflammation in these subgroups is clinically informative. BMI 5 body mass index.
Haldar, Pavord, Shaw, et al.: Clinical Asthma Phenotypes 221
Haldar AJRCCM 2008
Asthma Phenotypes
Fahy, NRI, 2015
18
Not all asthma is the same!!
(Heterogeneity)
(Phenotypes)
Question #4 - Asthma
Inhaled Corticosteroids are effective (at some dose) in all asthmatics.
1. True2. False
True or False? Question #4 - Asthma
Inhaled Corticosteroids are effective (at some dose) in all asthmatics.
1. True2. False
True or False?
19
Patients(%)
FEV1 Percent Change From Baseline
30
25
20
15
10
5
0<-30 -30 to
<-20-20 to
<-10-10 to
<00 to<10
10 to<20
20 to<30
40 to<50
5030 to<40
Beclomethasone (n=246)
Montelukast (n=375)
Patients (≥15 Years) Not Controlled on PRN Beta-Agonists FEV1: Distribution of Individual Patient Responses
Malmstrom et al.Ann Intern Med. 130:487-495, 1999
Eosinophils
Charcot-Leyden Crystals
A Large Subgroup of Mild-to-Moderate AsthmaIs Persistently Noneosinophilic
• Asthma is a heterogeneous disease
• Prior ACRN data (n=995; 2.7 SI; ≥2% eos):
• ~50% of asthmatics – poor response to steroids• Eosinophilic airway inflammation not ubiquitous
McGrath et al (ACRN)Am J Respir Crit Care Med 185:612–619, 2012
Sputum Eosinophil Percentage (No ICS)
TH2 Genes Overexpressed in Asthma
Woodruff et alAm J Respir CCM 180:388, 2009
Th2High
Th2High
Th2High
Th2Low
Th2Low
Th2Low
20
Th2 Status Predicts Corticosteroid Response
the inflammatory and pathological processes and thus representproximal targets for directed therapy. Second, phenotyping bygene expression identifies the genes that may contribute to thiscomplex disease based on DNA sequence variation and willhelp reduce phenotypic heterogeneity in genetic studies. Third,these gene expression array data facilitate the identification ofsecreted proteins that mark the phenotype and may be de-veloped into diagnostic tests.
One strength of our approach to molecular phenotyping is theuse of multiple sample types to provide a broad view of geneexpression in the airway. Our preparations of epithelial cells andalveolar macrophages are valuable for gene expression profilingbecause they represent relatively homogeneous populations ofcells. On the other hand, bronchial biopsies are valuable becausethe mix of cell types they contain (including inflammatory cells)permits direct measurement of Th1 and Th2 cytokines. However,because of the cell type heterogeneity inherent to bronchialbiopsies, the cellular source of these cytokines remains uncertain.
Finally, these data reveal that a significant percentage ofpatients with asthma have a Th2-low phenotype that manifestsclinical features of asthma, airway obstruction, airway hyper-responsiveness, and bronchodilator reversibility despite a paucityof Th2-driven inflammation. The causes of Th2-low asthma
remain obscure, but possibilities include neutrophilic inflamma-tion, IL-17–driven inflammation, intrinsic defects in barrierfunction and chronic subclinical infection by viruses, and atypicalintracellular bacteria. Whatever the mechanism, our data suggestthat it does not relate to uncontrolled allergic inflammation butrather to alternative mechanisms that are not steroid responsive.Future work should be directed at identifying noninvasive bio-markers that correlate with these molecular phenotypes in theairway and at developing a deeper understanding of the patho-physiologic mechanisms underlying asthma that is not driven byTh2 inflammation.
There are limitations inherent to our study design. First, oursubjects were not treated with inhaled or oral steroids for at least4 weeks before enrollment, and the use of leukotriene antagonistswas excluded. Although these steps facilitated identification ofthis Th2 signature in the absence of the confounding effects ofantiinflammatory medications, these inclusion/exclusion criterialimit the generalizability of the study to patients with more severeasthma. Whether these markers can be used to detect Th2-driveninflammation in patients who are poorly controlled despitecorticosteroid treatment remains to be determined. Second,repeat bronchoscopy was performed 1 week after initiation oftreatment, which was continued for a total of 8 weeks. Thus, thetiming of the second bronchoscopy may underestimate the extentto which inhaled steroids influence inflammation and our molec-ular signature. Third, we excluded healthy control subjects whohad a history of allergic rhinitis. Thus, it is possible that our Th2signature may also be detectable in the upper or lower airway ofatopic nonasthmatic subjects. Finally, our study design does notdefinitively rule out the possibility that the Th2-low phenotype inpart represents patients with well-controlled asthma. However,the majority of our subjects had not been using inhaled steroidsfor a long period before enrollment (i.e., we did not stop the use ofinhaled steroids in recruitment of our subjects but rather soughtpatients who were not using inhaled steroids chronically of theirown accord), minimizing the effect of prior steroid therapy in ouranalyses. Furthermore, patients with Th2-high and Th2-lowasthma did not significantly differ with regard to airway obstruc-tion or reversibility at baseline in this study, which lends furthercredence to the notion that the severity of their disease wascomparable and not due to differences in asthma control.
Conflict of Interest Statement: P.G.W. received more than $100,001 researchgrant to study biomarkers in asthma from Genentech and is the co-inventor ona patent for the development of biomarkers in asthma. B.M. received more than$100,001 in salary from employment by Genentech and has $10,001 to $50,000in stock owernship or options from Genentech. D.F.C. is an employee ofGenentech. G.J. is a full-time employee of Roche/Genentech Inc. A.R.A. doesnot have a financial relationship with a commercial entity that has an interest in
Figure 4. Alveolar macrophage gene expression in Th2-high and Th2-low asthma. Mean (1SEM) expression levels of 15-lipoxygenase(ALOX15) and TNF-a as determined by real-time PCR (*P , 0.03 forsubjects with Th2-high asthma vs. healthy control subjects). n 5 15healthy control subjects (gray bars); n 5 5 patients with Th2-lowasthma (blue bars); and n 5 9 patients with Th2-high asthma (red bars).
Figure 5. Responsiveness ofTh2-high asthma to inhaledsteroids and reproducibilityof phenotypic markers afterplacebo in a randomized con-trolled trial. (A) FEV1 measuredat baseline (week 0), after 4and 8 weeks on daily flutica-sone (500 mg BID), and 1 weekafter the cessation of flutica-sone (week 9). *See Table 2for number of subjects and Pvalues. There was no signifi-cant change in FEV1 in re-
sponse to placebo at any time point in either group. (B) Heatmap depicting unsupervised hierarchical clustering of POSTN, CLCA1, andSERPINB2 as in Figure 1C in bronchial epithelium of patients with asthma 1 week after the initiation of fluticasone or placebo treatment (n 5 19receiving fluticasone; 13 receiving placebo). Cluster at baseline for individual subjects in Figure 1C and treatment are indicated below the heatmap.
394 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 180 2009
Woodruff et alAm J Respir CCM 180:388, 2009
• N=135, prednisone x ≥6 months, eosinophils >300
21
A Large Subgroup of Mild-to-Moderate AsthmaIs Persistently Noneosinophilic
• Asthma is a heterogeneous disease
• Prior ACRN data (n=995; 2.7 SI; ≥2% eos):
• ~50% of asthmatics – poor response to steroids• Eosinophilic airway inflammation not ubiquitous
McGrath et al (ACRN)Am J Respir Crit Care Med 185:612–619, 2012
Sputum Eosinophil Percentage (No ICS)
Steroids in Eosinophil Negative Asthma (SIENA)
1. Does the response to ICS differ between subjects who are persistently EOS– and
those who are EOS+?
Co-Primary Research Questions:
2. Does the response to LMA differ between subjects who are persistently EOS– and
those who are EOS+?
22
In terimHxLimited P ESp iro w/IP B MRSIP erio stineNOEo sinoph i l sGenetics B loodDiaryReviewReview El ig& Comp l ian ceQuestionnaires
EOS-
EOS+
Run-in
Wk 0 3
SI SI
SIENA: Schematic
3-month trea tment;1st mo c ensored
3-month trea tment;1st mo c ensored
3-month trea tment;1st mo c ensored
Con sen tH&PSp iro w/Alb MR(Mch )CBC , IgEImmunoCAPSIP erio stineNOEo sinoph i l sP regnan cy testQuestionnaires
LMA + Int ICS LMA + Int ICS LMA + Int ICS
PBO + Int ICS PBO + Int ICS PBO + Int ICS
PBO + Int ICS PBO + Int ICS PBO + Int ICS
LMA + Int ICS LMA + Int ICS LMA + Int ICS
Phone Visit
18 30 4224 36
Phone Visit
Phone Visit
Phone Visit
Phone Visit
In terimHxLimited P ESp iroDiaryReviewQuestionnaires
In terimHxLimited P ESp iroDiaryReviewQuestionnaires
In terimHxLimited P ESp iroDiaryReviewQuestionnaires
6
RandomizeIn terimHxLimited P ESp iro(SI)(P erio stin )(eNO)DiaryReviewQuestionnaires
12
Phone Visit
Phone Visit
In terimHxLimited P ESp iroDiaryReviewQuestionnaires
In terimHxLimited P ESp iroDiaryReviewQuestionnaires
In terimHxP ESp iroDiary ReviewQuestionnairesP regnan cy test
Alb MR =Alb u tero l MaximumReversib i l i ty SI=Sp u tum In d u ctio nIP B MR = Ip ra tro p iumMa ximumReversib i l i ty IC S= In ha led Co rtico stero idMch =Meth a ch o l in eP C2 0 LMA=Lo n g -a ctin g Mu sca rin icAn ta g o n ist
ICS + Int ICS ICS + Int ICS ICS + Int ICS
ICS + Int ICS ICS + Int ICS ICS + Int ICS
Single-blind Placebo
(SI)
Phone VisitV 1 2 3 4 5 6 7 8 9
(See AppendixAforlistof Questionnaires)
N = 384Alternative Treatment?
Tiotropium Step-Up for Uncontrolled Asthma
Peters et al.N Engl J Med 363:18, 2010
Eur Respir J; 43:343-73, 2014
23
Eur Respir J; 43:343-73, 2014
Recommendations:• In adults with severe asthma – use sputum eos in
experienced centers
• In severe allergic asthma – therapeutic trial of omalizumab
• Do not use methotrexate for asthma
• Do not use azithromycin for asthma
Eur Respir J; 43:343-73, 2014
Recommendations:• Use anti-fungals for ABPA
• Do not use anti-fungals without ABPA
• Consider bronchial thermoplasty only as part of a study
NAEPP GUIDELINES
“If there is a clear and positive response for at least 3 months, a careful step down in therapy should be attempted to identify the lowest dose required to maintain control. (Evidence D)”
Evidence D = Panel Consensus Judgment
Expert Panel Report 3 (EPR-3): Guidelines for the Diagnosis and Management of Asthma-Summary Report 2007.National Asthma Education and Prevention Program.J Allergy Clin Immunol. 2007 Nov;120(5 Suppl):S94-138.
GINA GUIDELINES
“Controller treatment may be stopped if the patient’s asthma remains controlled on the lowest dose of controller and no recurrence of symptoms occurs for 1 year (Evidence D)”
Global strategy for asthma management and prevention: GINA executive summary.Eur Respir J. 2008 Jan;31(1):143-78.
Evidence D = Panel Consensus Judgment
24
Is There Really A Difference
Between Asthma And COPD?
Pathophysiology in COPD versus Asthma
Asthma• Inflammation
• Bronchialhyperresponsiveness
• Varying airwayobstruction
COPD• Loss of elastic
recoil• Changes in small
airways• “Inflammation”• Fixed airway
obstruction
Inflammation inCOPD versus Asthma
Calverley, Barnes. AJRCCM 2000; 161:341-344
COPD AsthmaPredominant Cells
Macrophages EosinophilsNeutrophils Activated Mast Cells
CD-8 T-Lymphocytes CD-4 T Lymphocytes
Predominant CytokinesInterleukin 8 Interleukin 4
Leukotriene B4 Interleukin 5Tumor Necrosis Factor alpha Interleukin 13
COPD Asthma OverlapIN COPD
Postma DS, Rabe KF .N Engl J Med 2015; 373: 1241-1249
n engl j med 373;13 nejm.org September 24, 20151244
T h e n e w e ngl a nd j o u r na l o f m e dic i n e
factor for the development of COPD19 (Fig. 2). The prevalence of bronchial hyperresponsiveness among patients with COPD has been reported to be 60%,20 and it may occur even in patients with mild disease, in whom the baseline level of FEV1should minimally influence the measurement of bronchial hyperresponsiveness.20 One study showed bronchial hyperresponsiveness in 90% of patients with COPD who did not have a his-tory of asthma.21 A recent study showed that more severe bronchial hyperresponsiveness is associ-ated with higher residual volume (a measure of air trapping that is related to small-airway dys-function) in COPD.22 In addition, bronchial hyper-responsiveness is associated with airway inflam-mation — that is, increased levels of neutrophils, macrophages, and lymphocytes in sputum and bronchial-biopsy specimens22 and increased levels of CD8 lymphocytes and eosinophils in periph-
eral lung tissue23 — in patients with COPD. The association of increased eosinophil levels with bronchial hyperresponsiveness was once thought to be limited to patients with asthma.11
What are the clinical implications of bronchial hyperresponsiveness in COPD? Studies have shown that the decline in FEV1 is accelerated in patients with COPD who have bronchial hyper-responsiveness,24,25 and that the decline is even more prominent in smokers.25 Thus, the associa-tion of bronchial hyperresponsiveness with the course of lung-function change and with response to inhaled glucocorticoids differs between COPD and asthma. Bronchial hyperresponsiveness is a risk factor for death from COPD in the general population.26,27 Thus, bronchial hyperresponsive-ness is a marker for more severe disease in both asthma and COPD, but there are not adequate data to indicate whether there is a long-term
Figure 2. Risk Factors for Asthma and COPD and the Influence of Environment and Aging.
There are a large number of established risk factors for asthma and COPD. Although some of these factors are clearly related to one disease and not the other, there is already considerable overlap early in life and early in the develop-ment of either disease. With time and increasing age of patients, cumulative environmental hazards, and the result-ing chronicity of the diseases, the clinical presentations appear more and more similar, making it sometimes impos-sible to clearly differentiate asthma from COPD in a given person. Although a number of characteristics overlap, some features of one or the other underlying disease may still be more prominent. Careful examination of these fac-tors may be helpful to identify the predominant disease phenotype when predicted lung-function values alone are not sufficient. Spirometry (graph) shows limited reversibility of airway obstruction after bronchodilator (BD) use by patients with ACOS. BHR denotes bronchial hyperresponsiveness, BPD bronchopulmonary dysplasia, GERD gastro-esophageal reflux disease, and Th2 type 2 helper T cells.
Asthma
ACOS
COPD
Risks Outcomes
INFLUENCE OF ENVIRONMENT AND AGING ON SEVERITY AND CHRONICITY OF DISEASE
Air
flow
(lite
rs/m
in)
Volume of expired air (liters)
0
4
6
0 1 3 4
The New England Journal of Medicine Downloaded from nejm.org at SAN FRANCISCO (UCSF) on October 7, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
25
Asthma Summary• Asthma not a single disease but a
heterogeneous group of diseases• Patients respond differently to medications
based upon underlying “endotype/phenotype”• “Th2-High” or Allergic Asthma responds to
corticosteroids• Treatments for “Th2-Low” or Non-Allergic
Asthma remain unclear