A First Glimpse at the Early Origins of Idiopathic Pulmonary Fibrosis

Familial pulmonary fibrosis (FPF) is most often diagnosed by the presence of clinically evident idiopathic pulmonary fibrosis (IPF), or any idiopathic interstitial pneumonia, in at least two members of the same family (1). For almost 30 years, we have known that asymptomatic family members recruited on the basis of FPF have detectable abnormalities shared by their symptomatic relatives (2). Further studies have demonstrated that significant percentages of asymptomatic, or self-identified unaffected, family members in these FPF kindreds have physiologic, radiologic, and histopathologic abnormalities (3, 4). This suggests that a lack of respiratory symptoms, or loss of function, should not be a benchmark on which to exclude disease affection status. Although this may not be the most comforting message, in this issue of the Journal, Kropski and colleagues (pp. 417–426) provide further evidence that this statement is almost certainly true (5). We must consider what obligation we might have to inform our patients with FPF that some of their asymptomatic family members may be at risk for, or in some cases may already have, the same disease.

Kropski and colleagues present the collected data and cross-sectional analysis of 75 asymptomatic, or minimally symptomatic, first-degree relatives recruited on the basis of FPF who completed a battery of testing, including respiratory questionnaires, blood sample collection, high-resolution chest CT, and in most cases, bronchoscopy with transbronchial biopsy (5). Although there is much to be learned from a review of this data, and from subsequent analysis of this population, this study provides evidence that radiologic, histopathologic, genetic, and other potentially pathogenic abnormalities noted in patients with pulmonary fibrosis can be detected in asymptomatic relatives recruited on the basis of a family history of PF. A number of peripheral blood biomarkers previously found elevated in patients with FPF and IPF (e.g., matrix metalloproteinase-7) (6) were also noted to be elevated in these asymptomatic relatives and were, in some cases, associated with the imaging findings. There was no evidence that cellular inflammation played a strong role in early stages of FPF. Some of these potentially pathogenic processes deserve further comment.

Comparable to previous findings by the Vanderbilt group in patients with IPF (7), this manuscript demonstrates that 30% of the asymptomatic relatives of FPF kindreds have detectable herpesvirus DNA in their lungs. Those with detectable viruses were more likely to have increased measures of endoplasmic reticulum stress. Similarly, herpesviruses have been demonstrated to lead to endoplasmic reticulum stress and pulmonary fibrosis in aged mice (8). However, it should be also noted that endoplasmic reticulum stress was also commonly found in those without detectable herpesviruses in this study (5), and a similarly elevated prevalence of herpesviruses in the lungs of patients with IPF has not always been found in other studies (9, 10).

In addition, this study confirms that reduced telomere length likely occurs early in the course of pulmonary fibrosis, as suggested by the fact that genetic variants in, or adjacent to, numerous genes controlling telomere length have now been associated with both FPF and IPF (DKC1, NHP2, NOP10, OBFC1, TERT, TERC, and TINH2) (11–17). Although the precise mechanisms by which reduced telomere length leads to pulmonary fibrosis remain uncertain, cellular senescence and/or apoptosis and genomic instability are thought to play some role (18).

Finally, this manuscript provides further evidence that the MUC5B promoter polymorphism (rs35705940) is associated with both FPF and an increased expression of MUC5B protein in the lung (19). Although it is intriguing that both the MUC5B promoter polymorphism and the resultant increase in MUC5B protein expression appear to be present in early (5, 20) as well as late (5, 19) stages of pulmonary fibrosis, it remains unclear how increases in MUC5B expression contribute to pulmonary fibrosis. Further complicating this story, MUC5B deficiency has recently been identified to result in critical impairments in macrophage host defense and bacterial burden in the lungs of an animal model (21), and the MUC5B promoter polymorphism appears to have an inverse relationship with the bacterial burden in the lower airways of patients with IPF (22).

Although the authors should be congratulated for this study, there are also limitations worth noting. Accumulating evidence suggests that early stages of sporadic IPF that are comparable to these findings in FPF (5) may be also detectable (23). However, we should be cautious in extrapolating the findings of early disease detection in FPF to sporadic IPF. FPF tends to present earlier than sporadic IPF (24) and can have imaging (25) and histopathologic (5) findings discordant from those expected to be present in sporadic IPF. Although the authors should be applauded for their collection and phenotypic characterization of this valuable cohort, the controls for the bronchoalveolar lavage analyses were on average 13 years younger than the asymptomatic family members. Given the prominent role that age alone may have in this disorder (26), these age differences potentially limit the strength of some of the conclusions that can be drawn. In addition, we do not know whether the development of pulmonary fibrosis can result from each of these potentially pathogenic processes independently or whether the cumulative and/or cooperative effects of multiple processes will be required for the transition to a progressive, clinically evident disease. Finally, as the authors duly note, the greatest value of this cohort will come from the longitudinal follow-up evaluations that hopefully will allow us to determine what factors predict progression from detectable abnormalities to pulmonary fibrosis. If we can ultimately agree that a lack of respiratory symptoms should not alone exclude the detection of FPF, perhaps we should consider what abnormalities are sufficient to define disease in FPF.