Effect of interferon gamma-1b on survival in patients with idiopathic pulmonary fibrosis (INSPIRE): a multicentre, randomised, placebo-controlled trial.

Vincent Cottin

Vincent Cottin

Vincent Cottin

Lize Kiakouama

Lize Kiakouama

Lize Kiakouama

+ 0 Evaluators

This article changes clinical practice

This study demonstrates that interferon gamma-1b cannot be recommended to treat patients with idiopathic pulmonary fibrosis (IPF). Off-label use of interferon gamma-1b should be strongly discouraged.

Previous studies have suggested that interferon gamma-1b might be effective in treating patients diagnosed with IPF. This large multicenter, randomized, placebo-controlled phase III trial definitely refutes the findings that interferon gamma-1b improves overall survival or any secondary endpoint in IPF. Recombinant interferon gamma-1b is approved for patients with chronic granulomatous disease and patients with severe malignant osteopetrosis. Based on its ability to decrease fibroblast activation and collagen secretion and to modulate the Th1/Th2 balance toward a Th1 phenotype, it has been considered as a potential treatment for IPF. Three previous phase II clinical trials have suggested that patients with IPF might benefit from treatment with interferon gamma-1b; an exploratory analysis of one previous study further suggested that treatment with interferon gamma-1b might be associated with improved survival in a subgroup of patients with mild to moderate functional impairment {1}. To determine whether treatment with interferon gamma-1b improves survival in patients with IPF, the authors here conducted a large, multicentre, placebo-controlled, randomized, phase III trial comparing 200µg subcutaneous interferon three times a week versus placebo (2:1). IPF was defined by ATS/ERS diagnostic criteria, including lung biopsy in 55% of the patients. Eligible patients had a forced vital capacity between 55% and 90% of predicted value and a diffusion capacity for carbon monoxide between 35% and 90% of predicted value, corresponding to a mild to moderate functional impairment. The study was designed to have 90% power to detect a treatment effect equivalent to a 50% reduction in 3-year mortality (with alpha value of 0.025 using one-sided Log-rank test). A total of 826 patients were enrolled. Although the primary endpoint of overall survival was intended to be measured 90-96 weeks after enrollment of the 600th patient, the study was terminated at the second interim analysis due to absence of minimum benefit of treatment as compared to placebo and as pre-specified per protocol. After a median duration of treatment of 64 weeks, there was no benefit of mortality with interferon gamma-1b treatment compared to placebo (hazard ratio 1.15, 95 % CI 0.77-1.71, p=0.497). Treatment with interferon gamma-1b did not significantly influence any of the secondary endpoints, i.e. survival time without transplantation, survival without respiratory-related hospital admission, change in dyspnea, change in 6-minute walk distance, change in forced vital capacity, transfer capacity, quality of life score, rate of disease progression, acute respiratory decompensation or acute exacerbation of IPF. Although the prevalence of severe adverse events was comparable in both groups, patients receiving interferon were more likely to report constitutional symptoms (influenza-like illness, fatigue, fever and chills). The strengths of the study include its methodology, with overall mortality as a primary endpoint, a very large sample size (it is the largest trial of IPF to date), a neat design, the good overall treatment compliance (91% on interferon and 94% on placebo) and assessable data (99% in both groups), despite significant adverse events of treatments, and the long duration of the study. The main limitations of the study are inherent to conducting a large placebo-controlled trial in such a severe disease. Eligibility was limited to a subset of patients with moderate functional impairment. The number of deaths or acute exacerbations of IPF during the study was lower than expected, suggesting that inclusion bias may have occurred with patients who had rapidly progressive disease being less likely to be included, although disease progression prior to enrollment was a mandatory inclusion criterion. Nevertheless, this study refutes previously reported, less robust trials and definitely demonstrates that interferon gamma-1b is ineffective to treat IPF and cannot be recommended. Participation of patients in other ongoing clinical trials in IPF should be strongly encouraged to find novel treatments for IPF.

Stephen Frankel

Stephen Frankel

Stephen Frankel

+ -1 Evaluators

This randomized, placebo-controlled trial of interferon gamma for the treatment of idiopathic pulmonary fibrosis (IPF) is an important negative study that concludes the debate over the use of interferon gamma in patients with IPF. Perhaps even more importantly, this study demonstrates the proof of principle that large, multi-center, randomized, placebo-controlled, investigational treatment trials can be efficiently and effectively conducted in patients with IPF. IPF is a chronic, debilitating interstitial lung disease (ILD) with a mean survival of 2-5 years from the time of diagnosis. To date, there have been no proven effective therapies for IPF. In 1999, Ziesche et al. published a small treatment trial comparing interferon gamma-1b plus low-dose prednisolone with prednisolone alone in 18 patients with IPF {1}. At 12 months, the group that had received interferon gamma had an improvement in their resting PaO2 compared with the prednisolone group. Based upon this and other small trials, the GIPF-001 study was performed {2}. GIPF-001 was a 330-patient, double-blind, randomized, multi-center, placebo-controlled trial of interferon gamma for the treatment of IPF with a primary endpoint of progression-free survival. Published in 2004, this trial found that interferon gamma did not affect progression-free survival and, hence, was a negative study. However, the majority of the endpoints were progression as measured by resting PaO2, which retrospectively was found to be an invalid marker for disease progression. An intention to treat (ITT) analysis of the secondary endpoint survival suggested a trend towards (non-statistically significant [p=0.08]) improved survival in the interferon gamma group. The authors concluded that, “owing to the size and duration of the trial, a clinically significant survival benefit could not be ruled out.” In the INSPIRE trial, 826 patients with mildly-to-moderately severe IPF were enrolled from 81 centers in North America and Europe over 2.5 years. The patients were randomly assigned (2:1) to either 200 micrograms of interferon gamma-1b subcutaneously injected three times per week or to placebo injections. The primary endpoint was survival time measured from the time of enrollment. At the second interim analysis, the trial was stopped for non-efficacy as the treatment group failed to achieve the minimum benefit threshold compared with placebo. At a median duration of therapy of 64 weeks, the mortality rate in the treatment group was 15% (hazard ratio for mortality: 1.15, CI 95% of 0.77-1.71, p=0.497) whereas the mortality rate in the placebo group was 13%. Secondary endpoint analyses demonstrated similar results (non-efficacy). Trial retention and treatment adherence were excellent as 99% of patients completed the study. While a negative study, this trial is important and worthy of attention for a number of reasons. Firstly, because a number of smaller trials had suggested benefit, significant numbers of patients were started on open-label, compassionate use interferon gamma for their IPF. This study definitely demonstrated the lack of efficacy of interferon gamma. Secondly, the interferon gamma studies have collectively demonstrated the ability of investigators in ILD to effectively conduct large, prospective, randomized, placebo-controlled, multi-centered clinical research trials of investigational agents. While taken for granted in 2009, it must be appreciated that just 10 years ago the ILD and IPF literature was predominantly based upon case series and small, uncontrolled trials. The identification of valid endpoints, the classification of disease types and severity, and the ability to effectively enroll patients at a rate that makes these studies feasible are all recent developments, and the INSPIRE trial remains the largest completed study and among the best designed trials in IPF research to date. Trial registration: NCT00075998