Incorporating genetics into the identification and treatment of Idiopathic Pulmonary Fibrosis

Identification of the constellation of findings (liver abnormalities, cytopenias, premature graying of the hair) consistent with rare genetic mutations in TERT or TERC mutations is critical, as these patients are at risk for bone marrow failure and cryptogenic liver cirrhosis [40]. Evidence suggests that in autosomal dominant forms of FIP caused by coding mutations in TERT, a unique form of genetic anticipation causes a shift from a pulmonary fibrosis predominant phenotype to one characterized by bone-marrow failure over successive generations [21]. Patients carrying TERT mutations have a poor prognosis with reduced life expectancy [41].

One of the few therapeutic options for patients with pulmonary fibrosis is lung transplantation. In the case of patients with TERT mutations, a small observational study suggests that complications of lung transplantation, such as renal failure, may be more common in IPF patients with telomerase mutations and/or shortened telomere syndrome [42], suggesting that genotyping could be important in determining transplant eligibility. This is particularly relevant in post-transplant patients who require significant immunosuppression because patients with these telomerase mutations experience increased rates of bone marrow suppression and medication-related complications [42], which may reflect their underlying diminished their underlying diminished bone marrow reserves. The authors caution that this observation has yet to be confirmed in larger cohorts of patients but suggest that careful consideration of the patients’ hematologic and hepatic status is warranted pre-transplantation [42].

Three observational studies have illustrated that common genetic variants associated with disease are also associated with differences in survival. In 2013, Noth and colleagues reported that several variants in TOLLIP were associated with IPF; however, carriers of the minor allele (G) at rs5743890 had decreased decreased risk of IPF but those with IPF who had this allele experienced increased mortality [28]. Another study published the same year described a survival advantage for individuals with the minor allele at rs35705950, the MUC5B promoter polymorphism strongly associated with disease [33]. Another functional SNP found in Toll-like receptor 3 (TLR3) has been associated with increased mortality and with accelerated disease progression in patients with IPF [43]. The mechanism for these observed differences in mortality remains unknown, but could be related to underlying differences in disease pathogenesis or in the clinical response to commonly prescribed therapies.

As numerous investigators have shown, various genetic variants, both rare and common, in telomere-related genes are associated with disease status [20, 23, 24]. Telomere length itself is also associated with transplant-free survival time for patients with IPF independent of age, sex, forced vital capacity, or diffusing capacity of carbon monoxide [44]. Additional studies are necessary to establish what the clinically relevant telomere length thresholds might be and how this measurement could function as an IPF biomarker or affect choice of therapy.

We suggest that further studies will continue to elucidate phenotypic differences between patients with IPF who have different disease-associated genetic variants. The clinical significance of these specific genetic variants remains unknown. Though strongly statistically associated with disease, the effect size of most common variants is small, whereas the effect size of rare variants is large (Fig. 1). The relationship between different common variants and their potential interaction with rare variants in disease pathogenesis will be an area of future investigation.

Because the clinical significance of these more common genotypes remains unknown, routine genotyping of individuals with IPF is not recommended. Furthermore, no evidence exists to suggest that genetic data should determine the selection of approved therapies for IPF, such as pirfenidone [45], nintedanib [46], or lung transplantation [47], for any individual patient. At this time, the specific treatment options for any given patient should be made on the basis of the published known risks and benefits of the medications, all of which have been studied independently of genotype [45, 46].

However, future investigations and clinical trials will need to take into account potential genotypic variation between different genotypes, especially as they likely affect the primary outcomes for clinical trials [48, 49]. Failure to control for genotypes, such as the MUC5B promoter polymorphism, would be tantamount to failing to control for other factors such as age, sex, and baseline lung function that are known to influence clinical outcomes. Post-hoc analysis of existing clinical trial data stratifying groups by the presence of common risk alleles might also generate intriguing hypotheses to be validated in prospective studies.

Given the unpredictable clinical course of IPF, its poor prognosis, and lack of available mortality-modifying medical treatment, it is important to identify individuals with early disease. Given the low prevalence of IPF, it is not a disease for which physicians routinely screen asymptomatic patients. However, the growing evidence for inherited disease risk may prompt the pulmonary community to reconsider the need to seek out patients for early diagnosis among at-risk populations.

In the case of FIP, it is known that first-degree relatives of patients with pulmonary fibrosis are at high risk of developing lung abnormalities, but the clinical significance of these abnormalities are unclear [8]. In 1986, Bitterman and colleagues studied family members of patients with autosomal-dominant FIP. They found that first-degree family members without clinically apparent disease had bronchoalveolar lavage fluid with increased inflammatory cells, but whether these individuals developed pulmonary fibrosis was not studied [8]. Twenty-seven years later, follow-up evaluation of two of these patients revealed interim development of radiographic evidence of pulmonary fibrosis, as well as symptomatic and measurable respiratory impairment [50]. Though this study was limited by its small sample size, it illustrates that alveolar inflammation in first-degree relatives of FIP patients can progress to overt pulmonary fibrosis and that these patients can experience a long duration of preclinical disease. More recently, extensive phenotyping of first-degree relatives of patients with FIP revealed evidence of dysfunction in pathways associated with the development of pulmonary fibrosis, including telomere shortening, endoplasmic reticulum stress, and elevated MUC5B levels [51]. These findings were observed in relatives with and without evidence of disease by high-resolution computed tomography or transbronchial lung biopsy, suggesting that these at-risk individuals have molecular abnormalities that precede symptoms or clinical detection. More than one-third of the at-risk subjects had histologically abnormal lung tissue, and 14.7 % had evidence of early interstitial lung disease [51]. Further observation will be required to determine the significance of these findings and of the results of the Framingham Heart study [32] with respect to identifying which asymptomatic subjects will progress to pulmonary fibrosis and whether early intervention prevents clinical worsening.

Although numerous studies have now shown that asymptomatic individuals who are at risk based on pedigree or who carry known risk alleles like the rs35705950 variant have higher rates of interstitial lung abnormalities [32, 51], there is no data to suggest that intervention is indicated. In part, this is due to the lack of data concerning the natural history of asymptomatic interstitial lung abnormalities. However, family members of patients with FIP should remain vigilant for the development for respiratory symptoms and should refrain from exposure to known environmental pulmonary toxins such as tobacco smoke [52].