Pulmonary hypertension (PH) commonly complicates the course of patients with idiopathic
pulmonary fibrosis (IPF). It is associated with impaired functional ability and worse
survival (1). The prevalence of PH has a variably reported range between 15% in those
with mild to moderate restriction and 84% in those with more advanced disease (2,
3). The high end of this range underscores that most patients are likely to develop
PH as their disease progresses. The increasing armamentarium of drugs to treat pulmonary
arterial hypertension has raised the notion of therapy for PH complicating IPF. It
remains uncertain whether the presence of PH is the driver of worse outcomes or whether
it is a surrogate for disease severity. If it is indeed an adaptive phenomenon, then
ameliorating this may not result in benefit and, worse yet, might result in harm.
In contrast, if PH in this setting is a maladaptive response, then targeting it may
result in beneficial outcomes.
The INSTAGE (Efficacy and Safety of Nintedanib Co-administered with Sildenafil in
Idiopathic Pulmonary Fibrosis Patients with Advanced Lung Function Impairment) study
was a prospective, double-blind, randomized clinical trial comparing the benefits
of nintedanib with those of nintedanib plus add-on sildenafil in patients with IPF
with single-breath Dl
CO <35% predicted (4). This major inclusionary criterion replicated that of the STEP-IPF
(Sildenafil Trial of Exercise Performance in Idiopathic Pulmonary Fibrosis) study,
which examined the effects of sildenafil versus placebo in patients with IPF (5).
Among patients with IPF with Dl
CO <35%, the prevalence of PH is about 50%, and in this regard, this cutpoint represents
an enrichment strategy for underlying resting PH (6). The INSTAGE study failed to
meet its primary endpoint of a change in the St. George’s Respiratory Questionnaire
at 12 weeks, but there was a favorable trend in a number of secondary endpoints, including
the University of California, San Diego, Shortness of Breath questionnaire, as well
as a salutary effect on FVC change. Therefore, although it was a negative study based
on the chosen primary endpoint, the INSTAGE study was suggestive of a possible benefit,
perhaps best demonstrated in a further enriched population.
In this issue of the Journal, Behr and colleagues (pp. 1505–1512) present a post hoc
analysis of the INSTAGE cohort, categorized by the presence or absence of echocardiographic
evidence of right heart dysfunction (RHD) (7). This parallels a subgroup analysis
of the STEP-IPF study, which demonstrated a significant improvement in 6-minute-walk
distance of 99 m in the sildenafil patients who had evidence of RHD on echocardiography
(8). Thus, the current subgroup analysis was eagerly anticipated, with the hope that
this too might demonstrate a similar difference. Moreover, a positive result would
have further established echocardiography as a complimentary enrichment strategy that,
coupled with the Dl
CO, could identify patients with IPF who were likely to benefit from sildenafil. Unfortunately,
however, the presence of RHD did not appear to predict benefit from nintedanib plus
sildenafil over nintedanib alone. Notably, there was no difference in the effects
of combined therapy in those with RHD with regard to the primary endpoint (St. George’s
Respiratory Questionnaire score), or secondary endpoints including the University
of California, San Diego, Shortness of Breath questionnaire or change in FVC. Indeed,
the only difference between the two groups was a change in the B-type natriuretic
peptide, with greater stabilization of this biomarker noted in the RHD group receiving
combined therapy.
The disappointing message from this subgroup analysis is that those with RHD did not
have a magnified response when treated with sildenafil. Why did evidence of RHD on
echocardiography not discern an enriched target group for sildenafil therapy? One
possible explanation could be insufficient standardization or reader variability between
participating centers, as there were no central reads of the echocardiograms. We have
certainly learned the value of central adjudication of high-resolution computed tomography
scans for inclusion in IPF studies (9). If RHD is to be the target of a subgroup analysis
in future clinical trials, then perhaps the same due diligence for echocardiography
studies should be applied. This may be especially important in patients with IPF and
other forms of advanced lung disease, given the disappointing performance characteristics
of echocardiography in determining PH in these patients (10, 11).
On the flip side of this negative subgroup analysis, a potentially heartening message
is that if sildenafil does indeed benefit patients with Dl
CO <35%, then everyone may be a candidate, not just those with RHD. Indeed, although
there may not be evidence of PH at rest, many of these patients with Dl
CO <35% likely do have exercise-induced PH (12). Another possible reason for the failure
of echocardiography to discern a target group is that perhaps the quality-of-life
benefit suggested by the INSTAGE study was not a result of the vasodilatory properties
of sildenafil but, rather, the synergistic antifibrotic properties of sildenafil coupled
with nintedanib, suggested by the lack of deterioration in the FVC in the dual-therapy
group (4).
Whether the suggested benefits of sildenafil together with nintedanib will be seen
in a broader population of patients with IPF (e.g., Dl
CO >35%) remains uncertain. Similarly, whether similar or more robust results will
be seen with the combination of pirfenidone and sildenafil remains unknown but is
eagerly anticipated (13).
The encouraging news from the INSTAGE study, reinforced by this current subgroup analysis,
is that patients with IPF with severe disease are a worthy study population. These
patients have typically been excluded from prior pharmaceutical clinical trials and
have limited options. They are therefore readily available, recruitable, and retainable
with a clinical course that is inevitable. In addition, patients with more severe
disease do not appear to be at unduly heightened risk for treatment-related adverse
events. The inclusion of patients with more severe disease may be a valuable enrichment
strategy when evaluating future pharmacologic therapies for IPF. This is especially
important in the current era, in which antifibrotic therapy is the established standard
of care resulting in a more attenuated disease trajectory. In the phase 3 INPULSIS
studies, 30% of patients with mild to moderate disease when receiving nintedanib had
a 10% FVC decline at 52 weeks, whereas in the INSTAGE study of patients with more
advanced IPF, 36.8% of patients receiving nintedanib met the endpoint of a 10% relative
FVC decline or death at 24 weeks (7, 9). Patients with IPF with more advanced disease
therefore have a higher event rate, which may enable shorter studies with fewer patients
required to power a difference (14). There is often a silver lining and much to be
gained from negative studies; in this regard, it is hoped that the INSTAGE study and
the current subgroup analysis will help set the stage for future studies focusing
on the later part of the IPF journey.