ASCEND (IPF)

Published

King TE, et al. "A phase 3 trial of pirfenidone in patients with idiopathic pulmonary fibrosis". The New England Journal of Medicine. 2014. 370(22):2083-2092.
PubMed • Full text • PDF

Clinical Question

Among patients with idiopathic pulmonary fibrosis, does pirfenidone reduce the composite rate of FVC decline or death when compared to placebo?

Bottom Line

Among patients with idiopathic pulmonary fibrosis, the oral TGFβ/TNFα modulator pirfenidone modestly reduces the rate of FVC decline compared to placebo. Although ASCEND demonstrated an improvement in progression-free survival (PFS), there was no difference in overall survival.

Major Points

Idiopathic pulmonary fibrosis (IPF) is a progressive disease with a median survival of 2-5 years, for which few approved therapies exist. Pirfenidone is an oral agent that reduces fibroblast proliferation and collagen deposition via regulation of TGFβ and TNFα, and has been shown to be active in preclinical IPF models. Pirfenidone was investigated as a potential therapy for the disease in three phase 3 studies including SP3 (2010) and the two CAPACITY trials (2010), which suggested that pirfenidone may reduce the FVC decline associated with IPF and prolong progression-free survival (PFS). Some of the data was conflicting from these earlier studies, and thus the FDA required an additional phase 3 study.

The 2014 Assessment of Pirfenidone to Confirm Efficacy and Safety in Idiopathic Pulmonary Fibrosis (ASCEND) trial randomized 522 patients with IPF to pirfenidone or placebo for 52 weeks. A diagnosis of IPF was based on CT imaging and clinical findings, and an FVC 50-90% was required. Pirfenidone dosage was uptitrated over the first 2 weeks of therapy to a target dose of 2,403 mg daily divided into three equal doses and taken with meals. The primary endpoint was FVC decline at 52 weeks in the intention-to-treat population. The main secondary endpoints were change in 6-minute walk test (6MWT) at 52 weeks and PFS (defined as an FVC decline of ≥10%, 6MWT decline of ≥50 m, or death from any cause). At 52 weeks, pirfenidone reduced the rate of the study's primary endpoint, decline in FVC (mean FVC decline 235 vs. 428 ml; P<0.001). For the secondary endpoint of 52-week PFS, patients in the pirfenidone group fared better as well (HR 0.57, 95% CI 0.43-0.77; P<0.001). Pirfenidone was generally well tolerated though was associated with a higher rate of GI upset, dermatologic side effects, and reversible LFT abnormalities.

The survival data reported in ASCEND deserve special mention. First, ASCEND used PFS as the main survival endpoint. In IPF trials, PFS is variably defined based upon changes in 6MWT distance, DLCO, and/or FVC; reported PFS may therefore vary between studies. Despite these caveats, a 2010 Cochrane meta-analysis of PFS from the CAPACITY and SP3 studies suggests that pirfenidone reduces the risk of progression by approximately 30% compared to placebo (HR 0.70; 95% CI 0.56-0.88; P=0.002). This HR is similar to that seen in ASCEND (HR 0.57, 95% CI 0.43-0.77; P<0.001). All-cause mortality, which is a component of PFS, was separately reported as a secondary outcome in the ASCEND and CAPACITY studies. In ASCEND, all-cause mortality at 52 weeks was 4.0% with pirfenidone and 7.2% with placebo (HR 0.55; 95% CI 0.26-1.15; P=0.10). A pooled analysis of all-cause mortality in ASCEND and CAPACITY studies demonstrated improved all-cause mortality at 52 weeks (HR 0.52; 95% CI 0.31-0.8; P=0.01) but this was not preserved at 72 weeks.

Taken together, the efficacy of pirfenidone and the considerable unmet need in patients with IPF led to its FDA breakthrough designation in October 2014. Concurrently published with ASCEND was INPULSIS, which detailed a reduction in FVC decline with the tyrosine kinase inhibitor nintedanib in patients with IPF. Nintedanib was also FDA approved for IPF in October 2014.

Guidelines

NICE IPF (UK) (2013, adapted)

• Pirfenidone is recommended as an option for treating idiopathic pulmonary fibrosis only if:
  • the person has a forced vital capacity (FVC) between 50% and 80% predicted and
  • the manufacturer provides pirfenidone with the discount agreed in the patient access scheme
• Treatment with pirfenidone should be discontinued if there is evidence of disease progression (a decline in per cent predicted FVC of 10% or more within any 12 month period).
• People currently receiving pirfenidone that is not recommended according to the above guidelines should have the option to continue treatment until they and their clinician consider it appropriate to stop.

ATS/ERS/JRS/ALAT IPF (2015, adapted)

• Suggest that clinicians use pirfenidone in patients with IPF (conditional recomendation, moderate confidence in effect estimates)

Design

• Multicenter, randomized, placebo-controlled trial
• N=522 (94.1% of all enrolled completed)
  • Pirfenidone (n=261)
  • Placebo (n=261)
• Setting: 127 sites in 9 countries
• Enrollment: 2011-2013
• Follow-up: 52 weeks
• Analysis: Intention-to-treat
• Primary outcome: ≥10% decline in FVC or all-cause mortality

Population

Inclusion Criteria

• Age 40-80 years
• IPF confirmed by high-resolution CT with definite or probable usual interstitial pneumonia (UIP)
  • If probable UIP then a confirmatory surgical lung biopsy was performed
  • Diagnosis was confirmed centrally by the researchers
• Clinical symptoms for ≥1 year, diagnosis 6-48 months prior, no improvement in disease severity in the prior year
• Fibrotic changes including honeycombing and reticular changes of a greater extent than emphastematous changes on CT scan
• FVC 50-90% of predicted and change with bronchodilators <10% relative difference
• DLCO 30-90% of predicted
• FEV₁:FVC ≥0.80
• 6MWT ≥150m
• Able to consent and likely to adhere to study requirements

Exclusion Criteria

• FEV₁/FVC <0.80 after bronchodilator or increase ≥12% and increase of 200 mL in FEV₁ or FVC after bronchodilator
• Smoking in prior 3 months or unwilling to avoid smoking during the study
• Exposure to evironmental factors known to cause pulmonary fibrosis
• Known cause of ILD or connective sittue disease
• Asthma or COPD
• Active infection
• Likely lung tranplant in ≤1 year (US participants) or on waiting list (all other participants)
• Significant malignancy or other health condition likely to affect health status in following 2 years
• Severe liver disease or creatinine clearance <30 mL/min by Cockcroft-Gault
• Unstable or progressive cardiac or pulmonary disease in prior year
• Significant worsening of clinical status between screening and day 1
• Unlikely to comply to the study requirements
• Pregnancy or lactation
• Alcohol or other substance abuse in prior 2 years
• Personal or family history of long QT syndrome
• Significant liver-specific tests
• Prior treatment with the study medication
• Known hypersensitivity to study drug
• Use of any investigational, cytotoxic, immunosuppressive, cytokine-modulating, or receptor antagonist medications
• Use of fluvoxamine or sildenafil (except for ED use)

Baseline Characteristics

From the pirfenidone group. Groups were similar.

• Demographics: Age 68 years, male 80 years, enrolled in US 76%
• PMH: Former smoker 66%
• Health data: FVC 68% of predicted, FEV₁:FVC 0.84, DLCO 44% of predicted
• Dyspnea score: 34.0 (from UCSD SOB questionnaire, out of 120 with higher numbers indicating worse dyspnea)
• 6MWT: 415 m
• Time since diagnosis: 2 years
• CT findings:
  • Definite UIP pattern: 96%
  • Possible UIP pattern with diagnosis on lung biopsy: 4%
• Lung biopsy: 31%

Interventions

• Randomized to a group:
  • Pirfenidone - Pirfenidone 2,403 mg po qday, divided in three doses
  • Placebo
• The therapy was continue for 52 weeks
• Use of other experimental agents was not allowed

Outcomes

Presented as pirfenidone vs. placebo. Outcomes are at 52 weeks.

Primary Outcome

≥10% decline in FVC or all-cause mortality

16.5% vs. 31.8% (RRR 47.9%; P<0.001; NNT 7)

Secondary Outcomes

6MWT distance

27m higher with pirfenidone (Relative difference 44.2%; P=0.04)

PFS

Lower with pirfenidone (HR 0.57; 95% CI 0.43-0.77; P<0.001)

Dyspnea

From the UCSD SOB questionnaire

No difference

All-cause mortality

4.0% vs. 7.2% (HR 0.55; 95% CI 0.26-1.15; P=0.10)

Pooled with 692 participants in CAPACITY trials: 3.5% vs. 6.7% (HR 0.52; 95% CI 0.31-0.87; P=0.01; NNT 31)

IPF mortality

1.1% vs. 2.5% (HR 0.44; 95% CI 0.11-1.72; P=0.23)

Pooled with CAPACITY trials: 1.1% vs. 3.5% (HR 0.32; 95% CI 0.14-0.76; P=0.006; NNT 42)

Additional Analyses

Decline in FVC

Mean: -235 vs. -428 mL (relative difference 45.1%; P<0.001)

Slope of decline at week 52: -122 vs. -262 mL (AD 140 mL; Relative reduction 53.5%; P<0.001)

No change in FVC: 22.7% vs. 9.7% (RRI 132.5%; P<0.001)

6MWT decrease ≥50 m or all-cause mortality

25.9% vs. 35.7% (RR 27.5%)

Disease progression or all-cause mortality

Pirfenidone with 43% relative risk reduction (HR 0.57; 95% CI 0.43-0.77; P<0.001)

Adverse Events

Cough: 25.2% vs. 29.6%

Nausea: 36.0% vs. 13.4%

Headache: 25.9% vs. 23.1%

Diarrhea: 22.3% vs. 21.7%

URI: 21.9% vs. 20.2%

Fatigue: 20.9% vs. 17.3%

Rash: 28.1% vs. 8.7%

Dyspnea: 14.7% vs. 17.7%

Dizziness: 17.6% vs. 13.0%

Worsening of IPF: 9.4% vs. 18.1%

Dyspepsia: 17.6% vs. 6.1%

Anorexia: 15.8% vs. 6.5%

Vomiting: 12.9% vs. 8.7%

Weight loss: 12.6% vs. 7.9%

GERD: 11.9% vs. 6.5%

Insomnia: 11.2% vs. 6.5%

Discontinuation of therapy

Related to GI side effects: 2.2% vs. 1.1%

Related to dermatologic side effects: 2.9% vs. 0.4%

Serious adverse events or deaths

Lower in pirfenidone group

Criticisms

• Findings not necessarily generalizable to patients with advanced disease
• Unclear if results will persist for >1 year
• No efficacy data for exacerbations of IPF

Funding

InterMune, the manufacturer of Esbriet and Pirfenex (the brand names of pirfenidone)