乌鸦传媒

PARIS -- For idiopathic pulmonary fibrosis (IPF) patients with severely impaired diffusion capacity for carbon monoxide (DL_CO), adding sildenafil (Revatio, Viagra) to standard nintedanib (Ofev) failed to improve quality-of-life (QoL) or dyspnea, the INSTAGE trial found.

At 12 weeks, the primary endpoint of change in the 100-point St. George's Respiratory Questionnaire (SGRQ) score was -1.28 points in the nintedanib-sildenafil group of IPF patients with DL_CO ≤35% compared with -0.77 points in the nintedanib-placebo group (-0.52 difference, 95% CI -3.33 to 2.30, P=0.72), reported Fernando Martinez, MD, of Weill Cornell Medical College and New York-Presbyterian Hospital in New York City.

"My own personal view is that there was a suggestion of improving effect over the 24 weeks," Martinez told 乌鸦传媒. "It is likely that the study was too short and underpowered to demonstrate statistical significance."

Change in total SGRQ score at 24 weeks were increases of 0.23 in the sildenafil arm compared with 2.42 in the placebo arm (-2.19 difference, 95% CI -5.40 to 1.02). But Martinez cautioned that with the primary study endpoint being negative, any other findings should be viewed within that context.

"It is likely that nintedanib has an effect on QoL in these more physiologically affected patients and that the effect of adding sildenafil was 'tempered' in this setting," Martinez said.

The results of the phase III trial were presented here at the European Respiratory Society (ERS) meeting and published in the .

In his presentation at ERS, Martinez highlighted that the study was designed and powered based on preliminary data from STEP-IPF (Sildenafil Trial of Exercise Performance in Idiopathic Pulmonary Fibrosis), but the control arm in that study used placebo. "STEP-IPF had no anti-fibrotic therapy at all," he said. "It was done when there wasn't anti-fibrotic therapy available."

The current study did suggest benefit with sildenafil in certain measures of lung function -- notably changes in forced vital capacity (FVC) and brain natriuretic peptide (BNP) levels.

For FVC, the adjusted mean change from baseline favored the nintedanib-sildenafil arm at both 12 weeks (7.0 mL vs -25.5 mL with nintedanib alone, respectively) and 24 weeks (-20.8 mL vs -58.2 mL).

And the prespecified endpoint of a ≥5% decline in predicted FVC value or death occurred in 31.4% of those on sildenafil-nintedanib compared with 50.7% of patients those on nintedanib-placebo (HR 0.56, 95% CI 0.38-0.82).

"The hint of improvement in FVC was unexpected but supported by evolving mechanistic data," Martinez said.

For BNP levels, the change at 24 weeks from baseline was -11.6 ng/L in the sildenafil arm compared with 39.7 ng/L in the placebo arm (-51.3 ng/L difference, 95% CI -85.1 to -17.6).

For 24 weeks, the INSTAGE study randomized 274 IPF patients with a DL_CO ≤35% to standard dosing of 150 mg nintedanib twice daily plus 20 mg sildenafil three times daily (n=138) or nintedanib plus placebo (n=136). Other inclusion criteria were age ≥40 years, IPF diagnosis within the past 6 years, and a chest high-resolution CT (within 18 months) confirming the IPF diagnosis.

At 12 weeks, the secondary endpoint of dyspnea showed no meaningful difference between the two arms based on change from baseline on the 120-point University of California San Diego Shortness of Breath Questionnaire, where higher scores equate to more breathlessness (1.46 points with sildenafil vs 4.40 points with placebo). And at 24 weeks (4.44 vs 6.85 points, respectively).

Baseline characteristics were similar between the nintedanib-sildenafil and nintedanib-placebo arms: mean age (70.3 vs 70.0, respectively), with most of the patients being men (80.3% vs 77.9%) and white (75.2% vs 69.9%). Of note, roughly one-third of patients had emphysema at baseline (37.2% vs 33.1%) and more than half of patients were nintedanib-naive (55.5% vs 64.0%).

The investigators found no new safety signals with nintedanib monotherapy for these IPF patients -- previous trials studied the drug in those with more mild disease. The number of deaths from any cause (14 with sildenafil vs 15) were similar between the two arms, as were those that occurred during treatment (4 vs 5) and those deemed respiratory related (11 vs 12).

Comment