Although forced vital capacity has long been used to assess the status and progression of idiopathic pulmonary fibrosis (IPF), it’s been largely on a “gestalt” basis, because there hasn’t been sufficient data to properly determine FVC’s test performance characteristics. FVC has a natural 2.5% intra-patient variation in healthy people on sequential measurements; this intra-patient variability may increase with worsening illness. Therefore, there’s significant potential “noise” to consider when using longitudinal spirometry to ascertain progression in IPF or any other disease.
Roland du Bois, Derek Weycker, Talmadge King et al try to help us with this, by continuing to mine their large data set of 1,156 patients in past (negative) trials of interferon gamma for IPF for new insights. Previously, they’ve used it to create a scoring system for survival prediction in IPF, and to propose the minimal clinically important change in 6-minute walk distance in IPF to be 24-45 m. In the December Blue Journal, they propose that the minimally clinically important difference (MCID) in percent-predicted FVC in people with IPF is a change of 2-6%.
They conclude this after a statistical trick I’m not qualified to critique: multiplying the estimated standard deviation of FVC at baseline for all randomized subjects by the square root of one minus the “reliability coefficient” (a measure of intrapatient reproducibility of FVC). This produced the MCID. The reliability coefficient was determined from a sample of 91 patients who had 2 PFTs close together in time.
They also used an “anchoring” method in which they quantified the average change in FVC associated with a subjective change in symptoms. Among all patients, a mean change of 2.2% up or down in %-predicted FVC correlated with feeling “somewhat better” or “somewhat worse,” respectively.
Clinical Takeaway: Large declines (5-10%) in percent-predicted FVC correlate with poor clinical outcomes. A 2-6% change in percent-predicted FVC might be meaningful in following the clinical course of IPF, as authors propose here. Importantly, these values were either statistical means (symptoms-based method) or dependent on extrapolation from a small sample (statistical/distribution method), probably reducing their applicability to individual patients. In addition, the 2.5% intra-patient variability on sequential measurements of FVC (which variability seems to increase in sicker patients), and spirometers’ variability up to 2.5% in measuring FVC longitudinally, introduce further uncertainty of clinical utility of small changes in FVC. A validation study would help clarify this.
The definition of minimally clinically important differences in forced vital capacity, six minute walk distance, diffusion capacity, and other parameters may be helpful in clinical practice, but the arguably larger and more proximate effect of studies such as this one might be to provide achievable and defensible measures of success in clinical trials of pharmacologic agents for idiopathic pulmonary fibrosis. For example, defining a 2% change in forced vital capacity as clinically significant in IPF sets pirfenidone‘s performance in the recent CAPACITY trials in a slightly better light.
Although approved for treatment of IPF in Europe and Japan, pirfenidone’s observed effects in CAPACITY disappointed the FDA, who insisted in mid-2011 that another clinical trial be performed before they would approve pirfenidone for U.S. use. After watching its stock price fall from $48 to $11, InterMune, makers of pirfenidone, launched the new Phase 3 ASCEND-3 trial in July 2011. ASCEND-3′s primary endpoint will be change in percent predicted forced vital capacity at week 52. (This current study by Du Bois et al was also funded by InterMune, according to authors’ disclosures.)
Du Bois RM et al. Forced Vital Capacity in Patients with Idiopathic Pulmonary Fibrosis. Test Properties and Minimal Clinically Important Difference. Am J Respir Crit Care Med 2011;184:1382-1389.