Anti-acid therapy in idiopathic pulmonary fibrosis: insights from the INPULSIS® trials

Idiopathic pulmonary fibrosis (IPF) is a specific form of progressive fibrosing interstitial lung disease characterized by worsening lung function and dyspnea [1]. IPF is believed to develop as a result of an aberrant wound healing response to epithelial injury, in which activated alveolar epithelial cells release fibrogenic growth factors, promoting fibroblast proliferation and transformation to myofibroblasts and excess deposition of extracellular matrix [2]. IPF typically presents during the sixth or seventh decade and is more common in men than in women [1]. The natural history of IPF is variable but its prognosis is poor [3]. Acute deteriorations in respiratory function, known as acute exacerbations, are a common feature of the clinical course of IPF and a major cause of morbidity and mortality [4].

Gastroesophageal reflux disease (GERD) is known to be a common comorbidity in patients with IPF, but its reported prevalence varies widely according to the definition used [5]. In the INSIGHTS-IPF registry of patients with IPF in Germany, 29% of patients reported GERD on enrollment [6], while in a prospective study of 65 patients with IPF, acid gastroesophageal reflux occurred in 87% of patients subjected to 24-h pH monitoring [7]. It has been hypothesized that microaspiration caused by GERD plays a role in the pathogenesis of IPF by causing injury to the lung epithelium and the initiation of inflammatory cascades [8‐11] but a causal relationship between GERD and IPF has not been established.

Proton pump inhibitors used to treat GERD have pleiotropic antioxidant and anti-inflammatory effects [12]. Proton pump inhibitors target the H+/K+ ATPase, which has recently been shown to be expressed in the hyperplastic alveolar epithelium in patients with IPF [13]. Use of anti-acid medication has been associated with a decreased rate of decline in FVC and improved survival in some observational studies in patients with IPF [14, 15]. Surgical intervention to treat GERD (i.e. Nissen fundoplication) in patients with IPF has also been associated with improved survival [14, 16]. The latest international clinical practice guidelines for IPF included a conditional recommendation for the use of anti-acid therapy in patients with asymptomatic GERD, based on very low quality evidence [17]. However, there are no data from randomized controlled trials showing an improvement in outcomes in patients with IPF treated with anti-acid medication and the benefits and risks of anti-acid medication in patients with IPF remain a topic of debate [18‐20].

Nintedanib is an intracellular inhibitor of tyrosine kinases that has been approved for the treatment of IPF in many countries and regions, including the US, Europe and Asia. In the latest international clinical practice guidelines for IPF [17], nintedanib received a conditional recommendation for use based on data from three randomized controlled trials [21, 22]. In the two replicate Phase III INPULSIS® trials, nintedanib slowed disease progression in patients with IPF by reducing the rate of decline in forced vital capacity (FVC) by about 50% [22]. The most frequently reported adverse events in patients treated with nintedanib were gastrointestinal, particularly diarrhea.

The design of the INPULSIS® trials has been described [22]. Briefly, participants had to have been diagnosed with IPF within the previous 5 years and to have an FVC of ≥50% predicted and a diffusing capacity of the lungs for carbon monoxide (DLco) of 30–79% predicted. In the absence of a surgical lung biopsy, patients were required to have honeycombing and/or a combination of traction bronchiectasis and reticulation in the absence of atypical features of usual interstitial pneumonia on high-resolution computed tomography (HRCT). Patients were randomized 3:2 to receive nintedanib 150 mg bid or placebo for 52 weeks, with a follow-up visit 4 weeks later. The primary endpoint was the annual rate of decline in FVC (mL/year), analyzed using a random coefficient regression model including sex, age and height as covariates. Key secondary endpoints were time to first investigator-reported acute exacerbation (defined in [22]) and change from baseline in St George’s Respiratory Questionnaire (SGRQ) total score, both over 52 weeks. Time to first acute exacerbation was analyzed using the log rank test, with hazard ratios and confidence intervals obtained using the Cox’s proportional hazards model adjusted for sex, age and height. Change from baseline in SGRQ total score was analyzed using a mixed model for repeated measures including treatment and visit as fixed effects, baseline score as a covariate, and treatment-by-visit and baseline score-by-visit as interaction terms.

To evaluate whether anti-acid medication use at baseline was associated with differences in the natural course of IPF, analyses of the annual rate of decline in FVC (mL/year) and the time to absolute decline in FVC ≥10% predicted or death over 52 weeks were conducted in patients treated with placebo who were versus were not receiving anti-acid medication at baseline. A Cox regression model with terms for trial, subgroup, sex, age and height was used to assess time to absolute decline in FVC ≥10% predicted or death over 52 weeks. The original statistical approach for the annual rate of decline in FVC was repeated, but with an interaction term time-by-baseline anti-acid medication use included as a fixed effect in the model.

Safety was assessed via clinical and laboratory evaluation and the recording of adverse events with onset after the first dose and up to 28 days after the last dose of study drug. Adverse events were coded according to the Medical Dictionary for Regulatory Activities version 16.1. Safety analyses were descriptive.

The results of this post-hoc subgroup analysis of data from the INPULSIS® trials showed that use of anti-acid medication at baseline generally did not influence the treatment effect of nintedanib in patients with IPF. In the placebo group, the annual rate of decline in FVC was numerically higher in patients who were receiving anti-acid medication at baseline than in those who were not (− 252.9 vs − 205.4 mL/year; difference of − 47.5 mL/year [95% CI: –105.1, 10.1]; p = 0.1057) and a greater proportion of patients receiving anti-acid medication at baseline experienced decline in FVC ≥10% predicted or died over 52 weeks (47.5% vs 37.5%). These findings should be interpreted with caution given that these analyses were conducted post-hoc; that patients were not randomized according to reported use of anti-acid medication; that information on the type, dose, and duration of anti-acid medication used throughout (or prior to) the study are lacking; that the efficacy of anti-acid medication in suppressing acid reflux is not known, and that it is unknown whether the patients were taking anti-acid medication to treat symptomatic GERD (acid or non-acid), assumed silent GERD, or even IPF. Similar caveats apply to the interpretation of data from a post-hoc analysis of pooled data from three randomized trials of pirfenidone in patients with IPF, in which use of anti-acid therapy at baseline was not associated with a difference in the occurrence of disease progression in either the placebo group [18, 23] or the pirfenidone group [24].

Elevated pepsin levels in bronchoalveolar lavage fluid have been documented in patients manifesting acute exacerbations of IPF [8]. In a prospective observational study of data from placebo-treated patients with IPF in the IPFnet trials, there were no adjudicated acute exacerbations in the 124 patients who used anti-acid medication at baseline (of whom 117 continued to take it for the entire study period) versus 9 acute exacerbations in the 118 patients who did not use anti-acid medication at baseline [15]. In contrast, in our post-hoc analysis of data from the INPULSIS® trials, acute exacerbations reported by the site investigators occurred in a greater proportion of placebo-treated patients who were receiving anti-acid medication at baseline than in those who were not (11.7% [19 patients] versus 5.0% [13 patients]). An increased risk of investigator-reported acute exacerbations (and adjudicated confirmed or suspected acute exacerbations) was also observed in a risk factor analysis based on pooled data from these trials [25]. There are multiple hypotheses that may explain these findings. Patients who were receiving anti-acid medication at baseline may have had a greater extent of IPF than patients who were not. The patients receiving anti-acid medication at baseline had worse health-related quality of life at baseline according to SGRQ total score and a greater frequency of fatal adverse events, but did not have greater lung function impairment based on FVC or DLco. Secondly, it is possible that the increased frequency of acute exacerbations in patients taking anti-acid medication at baseline may be due to insufficiently controlled microaspiration in patients who had co-existing GERD. Abnormal acid gastro-oesphageal reflux has been documented in patients with IPF taking proton pump inhibitors [26], and the use of antiacid medication may not have fully suppressed acid reflux in patients in these trials who had concomitant GERD. Thirdly, it may be hypothesized that patients receiving anti-acid medication at baseline were at greater risk of infectious respiratory events (e.g. pneumonia) due to altered host defense against bacteria as a result of increased gastric pH. In our analysis, pneumonia was reported in a numerically higher proportion of placebo-treated patients receiving anti-acid medication at baseline than placebo-treated patients who were not. Epidemiological data suggest a potential adverse relationship between proton pump inhibitor use and community acquired pneumonia [27‐29], and a post-hoc analysis of data from placebo-treated patients in trials of pirfenidone suggested that in patients with FVC < 70% predicted at baseline, unadjudicated pulmonary infections were more common in patients who were taking anti-acid medication at baseline [18, 19, 23].

The proportion of patients experiencing diarrhea adverse events was similar in the subgroups by anti-acid medication use at baseline, but in the nintedanib group, the frequency of permanent treatment discontinuations due to diarrhea, although low, was numerically higher in patients receiving anti-acid medication at baseline. Overall, the adverse event profile of nintedanib in patients in both subgroups by anti-acid medication use was as expected based on the safety and tolerability profile reported in the overall patient population [22].