For most rare disorders, therapeutic regimens are based on experience or are delineated from pathophysiologic considerations. Cochrane reviews rely on double-blind placebo controlled trials, which are extremely scarce in uncommon diseases. Consequently, a definitive therapeutic approach is not disclosed by those analyses, as can be seen in the case of sarcoidosis (1). However, in idiopathic pulmonary fibrosis (IPF), a Cochrane review confirmed the usefulness of pirfenidone based on a limited database of phase III studies. The recent evidence-based guidelines of the American Thoracic Society, European Respiratory Society, Japanese Respiratory Society, and Latin American Thoracic Association for the diagnosis and management of idiopathic pulmonary fibrosis, however, provide a weak recommendation against the use of pirfenidone (2), although Japanese and European authorities subsequently approved pirfenidone as the first drug licensed for the treatment of IPF. According to Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) methodology, a weak negative recommendation means that the drug should only be used in selected cases and that the task force expects that future research is likely to result in changes in the recommendation (3).

Similarly, the task force made a weak negative recommendation against the use of anticoagulant therapy in IPF. Once again, this GRADE recommendation can be used to justify anticoagulation in selected IPF patients, whether in a handful of patients or in a large minority. The recommendation was based on a single open-label study by H. Kubo and colleagues from Japan that reported a significant improvement in survival with warfarin therapy, compared with corticosteroid therapy (4). The survival advantage could be ascribed to a reduction in mortality from acute exacerbations: the excess number of deaths in the corticosteroid arm was almost identical to the excess number of fatal acute exacerbations. Acute exacerbations in the anticoagulant and corticosteroid arms were treated with heparin and high-dose corticosteroid therapy, respectively. Based on these observations and the lack of treatment options, many clinicians have resorted to the empirical use of heparin in acute exacerbations of IPF, although it was unclear whether the putative survival benefit required prior treatment with warfarin.

The findings of Kubo and coworkers must be interpreted with great caution due to the selective withdrawal of a large minority of patients from the anticoagulant arm, leading to major imbalances in patient numbers in what was already a small patient cohort. Whether it was appropriate for pirfenidone and anticoagulant therapy to receive the same level of recommendation can be debated. However, this reservation does not detract from the fact that there is a sound basis for the view that anticoagulation might be beneficial in IPF. Ex vivo and in vivo studies have demonstrated dysregulation of the coagulation cascade, which was associated with some evidence of disease progression. In an epidemiologic study from Denmark, past or current thromboembolism was associated with a diagnosis of IPF (5), in keeping with earlier reports of a procoagulant milieu in the lower respiratory tract of IPF patients, as identified in bronchoalveolar lavage (6). Coagulation factors X and VII are locally overexpressed in human and murine fibrotic lung tissue and induce myofibroblast differentiation and fibroblast proliferation by binding to proteinase-activated receptor-1 and -2, respectively (7, 8). Moreover, there is evidence that heparin might block pro-proliferative effects of growth factors on alveolar epithelial type II cells (9). Based on these observations and the controversial findings of Kubo and colleagues, many clinicians considered that anticoagulation should be discussed as a treatment option in appropriate patients, despite the widely discussed flaws in the study by Kubo and coworkers (10).

In this issue of the Journal (pp. 88–95), a double-blind placebo-controlled clinical trial of warfarin therapy, performed by the American Idiopathic Pulmonary Fibrosis Clinical Research Network, is reported (11). The consortium invested a tremendous effort in accomplishing a placebo-controlled trial using anticoagulation, despite the obvious difficulties, and should be strongly commended for this achievement. Overall, the study reaches the high standards of recent therapeutic trials in IPF, and the interim analysis by the Data and Safety Monitoring Board, resulting in the termination of the study, can be regarded as definitive. There were no significant or marginal benefits on disease progression with warfarin therapy, and treatment was associated with excess mortality and a high prevalence of side effects. However, the cause of excess mortality remained obscure and was not due to hemorrhagic complications. It appears that warfarin may worsen the spontaneous course of IPF by causing exacerbations and progression. Potentially, it blocks beneficial reactions of the organism on inflammation and remodeling.

Based on their findings, the authors suggest that the observed activation of the coagulation cascade in IPF may be a protective reaction rather than a pathomechanism and should not, in future, be the target of a therapeutic intervention. Thus, the increased prevalence of thromboembolism in IPF (5) could be a “side effect” of a response that serves to constrain fibrotic pathomechanisms. A new insight delivered by the study of Noth and colleagues could thus be that in the pathogenesis of IPF, beneficial effects of vitamin K–dependent mechanisms must be postulated and must not be interfered with by therapy.

It is remarkable that this study was funded by public money. It is highly unlikely that scrutinizing established therapies or therapies delineated from pathomechanistic considerations will be funded by the pharmaceutical industry. For this reason, a formal trial of the efficacy of antireflux therapy in preventing or slowing disease progression in IPF will be difficult to achieve, despite increasing indirect evidence that this treatment approach might be fruitful. Therapeutic guidelines of industrialized nations should not be wholly dependent on studies conducted by pharmaceutical companies to approve novel drugs. The exploration of current agents that have a reasonable safety record in other contexts is desperately needed. Although the study of Noth and coworkers is definitively negative, the invested public money was well spent. The publication of the trial will save lives by preventing the fatal effects of a therapeutic approach that was endorsed for use in a minority of patients in the recent IPF guidelines (2). Other therapeutic approaches with existing agents, extrapolated from pathomechanistic studies, might demonstrate benefits but might also disclose unpleasant surprises, as in the study of Noth and colleagues. Therefore, initiatives scrutinizing established therapies beyond pharmaceutical phase III studies of patent-protected drugs will be of tremendous benefit in uncommon disorders. We welcome the fact that funding bodies increasingly recognize this need, with researchers now able to receive funding to address the inescapable limitations of evidence-based recommendations in rare diseases.

With regard to weak negative guideline recommendations in the recent IPF guidelines (2), it would appear that a year is a long time in interstitial lung disease. With the emergence of further data, including the recent independent analysis of a Japanese database, the weak negative recommendation made for the use of pirfenidone is now open to question, with many clinicians arguing for a weak positive recommendation, at least in mild to moderate disease. The validity of the weak negative recommendation against triple therapy is likely to be profoundly influenced by the recent publication of the discontinued arm of a study testing prednisone, azathioprine, and N-acetylcysteine for pulmonary fibrosis (PANTHER) (12). Based on the current study, it now appears that a strong negative recommendation against warfarin therapy in IPF is warranted. Thus, only a year after publication of the American Thoracic Society/European Respiratory Society/Japanese Respiratory Society/Latin American Thoracic Association guidelines, three of the four weak negative pharmaceutical recommendations made in that document need to be reconsidered. The current guideline recommendations should not be left in place for a further decade: urgent reappraisal is essential. With the exploration of novel therapies gathering pace, more frequent review and update of guidelines is surely required.

References

Section:

ChooseTop of pageReferences <<CITING ARTICLES