Progress in Understanding and Treating Idiopathic Pulmonary Fibrosis

Abstract

This is a time of substantial progress in the evaluation and care of patients with idiopathic pulmonary fibrosis (IPF). In addition to the approval and widespread availability of the first IPF-specific therapies, there have been improvements in imaging interpretation and lung biopsy methods to enable more expeditious and more accurate diagnosis. Recent advances in identifying genetic factors that underlie susceptibility to IPF and affect prognosis have raised the possibility of personalized therapeutic approaches in the future. Further, evolving work is elucidating novel mechanisms influencing epithelial, mesenchymal, and inflammatory cell responses during the injury-repair process, thus advancing understanding of disease pathogenesis. As analytic approaches mature, the field is now poised to harness the power of rapidly advancing “omics” technologies to further accelerate progress.

INTRODUCTION

Interstitial lung diseases (ILDs) represent a heterogeneous group of pulmonary parenchymal disorders resulting from a variety of environmental insults, systemic diseases, and idiopathic conditions. Of the ILDs, idiopathic pulmonary fibrosis (IPF) is the most common and most severe. Current estimates are that >100,000 people in the United States and Europe are living with IPF (1). These patients have an average life expectancy of 3–5 years after diagnosis in the absence of lung transplantation. While morbidity and mortality associated with this disease remain unacceptably high, there has been rapid progress in a number of areas over the last several years, including improved approaches for diagnosis, development of two therapies that have been approved by the US Food and Drug Administration (FDA), and identification of underlying genetic predisposition. The term IPF has been in common use since the 1970s and is linked to the pathological entity of usual interstitial pneumonia (UIP), which is defined by predominant subpleural fibrosis with fibroblastic foci, temporal heterogeneity, and microscopic honeycombing. Understanding of the pathobiology of IPF continues to evolve, with strong evidence supporting the concept that repetitive injury to susceptible alveolar epithelial cells (AECs) drives pathological interactions with fibroblasts, leading to excessive matrix deposition that destroys gas-exchanging units. Progress in understanding and treating IPF has led some investigators to call for renaming this disease to better align with current understanding of disease pathogenesis, as well as its relationship with other ILDs (2). This review focuses on areas of recent progress and highlights areas where more work is needed.

CURRENT APPROACH TO DIAGNOSIS

Patients with IPF present with chronic exertional dyspnea, cough, bibasilar inspiratory crackles, and sometimes digital clubbing. IPF patients are more commonly male than female, frequently have a tobacco smoking history, and are typically greater than 60 years old. Pulmonary function testing reveals evidence of restriction and reduced diffusing capacity for carbon monoxide. A diagnosis of IPF requires exclusion of pneumoconiosis, hypersensitivity pneumonitis, drug-induced interstitial lung disease, and rheumatologic disease. High-resolution computerized tomography (HRCT) scanning is now the primary diagnostic modality for evaluating patients with suspected IPF.

Although heterogeneity in radiologic appearance, pathology, and patient characteristics can make the diagnosis of IPF challenging, the approach to diagnosis of IPF is becoming more standardized. In 2011, the American Thoracic Society (ATS), the European Respiratory Society (ERS), the Japanese Respiratory Society (JRS), and the Latin American Thoracic Association (ALAT) jointly issued clinical practice guidelines (3). The ATS/ERS/JRS/ALAT guidelines categorized HRCT appearance as (a) UIP pattern, (b) possible UIP pattern, or (c) inconsistent with UIP pattern based on defined CT scan criteria. The UIP pattern category was defined as subpleural basal predominance of reticular abnormalities and honeycombing with or without traction bronchiectasis, and the absence of features inconsistent with the UIP diagnosis. The possible UIP pattern category was assigned if honeycombing was absent. Based on high-quality evidence that HRCT appearance can be highly specific for IPF, these guidelines indicated that the presence of a UIP pattern in combination with appropriate history and physical exam was sufficient for IPF diagnosis. Surgical lung biopsy was suggested for cases where IPF was suspected but the HRCT appearance did not show the UIP pattern. Using this approach, only about half of individuals with IPF/UIP have HRCT scans that are read as having a UIP pattern (4), so a large number of patients require additional diagnostic studies.

A recent Fleischner Society white paper (4) proposed a modification of these diagnostic criteria taking into account the clinical probability of IPF, which in increased in those who are older than 60, are current or former smokers, and have no history of other potential causes of fibrosis. Under these recommendations, the HRCT pattern is categorized as definite UIP, probable UIP, pattern indeterminate for UIP, or features most consistent with non-IPF diagnosis. These guidelines suggest that a probable UIP pattern in which honeycombing is absent could still be diagnostic for UIP if the clinical probability of IPF is high. This classification also suggests that some patients who would be identified as inconsistent with UIP pattern under the 2011 guidelines could have a reasonable likelihood of UIP on biopsy; therefore, some patients in the category of pattern indeterminate for UIP could still benefit from biopsy if there is high clinical suspicion.

An increasingly frequent occurrence is the detection of asymptomatic interstitial changes on abdominal or chest CT imaging obtained for other purposes, as it is now recognized that up to 7% of former tobacco smokers over 50 years of age have identifiable interstitial abnormalities (5, 6). Although these interstitial changes can progress, development of clinical symptoms over five-year intervals appears infrequent in unselected populations (7). These patients present a challenge for clinicians, as there is potential for early diagnosis and treatment but also risk of overdiagnosis of ILD/IPF. Several ongoing cohort studies have been designed to provide more clarity as to which patients with early asymptomatic interstitial changes on CT warrant further evaluation and/or early treatment (NCT03437486, NCT03478553).

The recommended modality for lung biopsy is also evolving. While thoracoscopic biopsy is still the standard approach, in-hospital mortality is in the range of 1.7% for elective procedures and as high as 16% for nonelective procedures (8, 9). Although traditional transbronchial biopsies have limited utility, transbronchial cryobiopsy is emerging as an alternative to surgical biopsy. This procedure results in substantially larger biopsies than transbronchial forceps, and a successful diagnosis can be made in 70–80% of cases (10). This procedure, however, is not well standardized and is currently best used at experienced centers. In inexperienced hands, the diagnostic yield may not be sufficient, and the risk of bleeding and pneumothorax can be substantial.

Although the most recent clinical practice guidelines do not recommend serologic evaluation for patients with suspected IPF, emerging evidence suggests there may be a distinct subgroup of patients who meet current IPF criteria but have more prominent features of autoimmunity and a better prognosis (11). Further work will be needed to refine classification of these patients and understand whether distinct mechanisms are responsible for their disease.

Because of the uncertainty inherent in the clinical, radiological, and pathological evaluation of patients with suspected IPF, multidisciplinary conferences with interactions among clinicians, radiologists, and pathologists are increasingly seen as necessary for consistent evaluation of patients with ILD. Several studies have shown that multidisciplinary discussion increases diagnostic confidence and may change the final consensus diagnosis in up to 20% of cases (12). However, outside of academic referral centers, access to multidisciplinary conferences is limited. One promising approach calls for adopting the “virtual multidisciplinary team” model (13) from cancer care centers in order to rapidly expand access and improve diagnostic standardization.

Since an accurate diagnosis of IPF has implications for prognosis and treatment, development of specific biomarkers would be beneficial if they could limit the need for surgical biopsy and improve specificity using noninvasive or minimally invasive techniques. To date, individual protein biomarkers in blood or bronchoalveolar lavage have not proven to be sensitive or specific enough to improve the accuracy of diagnosis of IPF. Several investigators have studied transcriptional signatures in peripheral blood cells and shown that these differ between IPF patients and controls (14–16), but the potential utility of the peripheral blood transcriptome in IPF diagnosis is uncertain. A more promising approach may be to utilize transcriptional profiling of samples obtained by transbronchial biopsy (17). Several studies have now used a machine learning approach to develop algorithms with high specificity for UIP/IPF that could prove to be clinically useful for improving diagnostic accuracy and minimizing the need for surgical biopsy (18, 19).

CLINICAL ADVANCES AND ONGOING CHALLENGES

The approach to treatment of ILD has changed dramatically over the last several years. The PANTHER-IPF study published in 2012 showed not only that immune-suppressive therapy is ineffective in IPF but that the combination of prednisone, azathioprine, and N-acetyl cysteine (NAC) is harmful (20). This study was followed by landmark studies in 2014 indicating that two antifibrotic therapies, pirfenidone (21) and nintedinab (22), were effective in reducing the decline in forced vital capacity (FVC) in patients with moderately advanced IPF. These studies led to FDA approval of both drugs. Nintedinab is an intracellular tyrosine kinase inhibitor. Pirfenidone is an antifibrotic molecule whose target(s) remains uncertain; however, this drug has been shown to reduce fibroblast proliferation and differentiation (23). Although each of these treatments has substantial gastrointestinal side effects, it is becoming clear that most patients can tolerate these therapies for an extended period and that continued therapy remains effective at reducing FVC decline (24, 25). In addition, new studies have suggested that patients with milder disease (FVC > 90% predicted) have similar reduction in FVC decline compared to patients with more advanced disease, suggesting that early treatment is warranted (26–28). While some investigators have questioned the relevance of reduced FVC decline as an important surrogate endpoint for clinical efficacy in IPF (28a), pooled analyses of clinical trials with pirfenidone (which has been studied in more patients than nintedanib) suggest that long-term treatment modestly reduces all-cause and IPF-related mortality (29). This finding suggests that FVC decline is a relevant surrogate endpoint and that these drugs may have long-term disease-modifying effects. However, neither drug has been shown to reduce symptoms or improve quality of life for patients with IPF. At present, there is no compelling reason to choose one drug over the other in most patients, and patient preference related to administration and potential side effects may sway the choice of treatment. Available data do not provide clear guidance as to the approach to treating a patient who is progressing on pirfenidone or nintedanib. While in practice many experts recommend switching to the other agent, criteria for defining treatment failure have not yet been established. Further, an obvious question is whether combination therapy might have added benefit. Although this remains to be answered, the INJOURNEY trial has recently shown that combination therapy is feasible and tolerable (30).

A 2015 update to the clinical practice guidelines for IPF by ATS/ERS/JRS/ILAT (31) gave conditional recommendations for use of both pirfenidone and nintedinab and recommended against use of other therapies tested to date, with the exception of antacid therapy, which has been suggested as an interdiction to reduce the injurious effects of gastric acid aspiration in this disease. Despite some contradictory data (32), use of proton pump inhibitors or histamine-2 receptor antagonists was recommended on the basis of retrospective studies. For patients with end-stage or progressive disease, lung transplantation remains the treatment of choice. Available data suggest that IPF patients who undergo bilateral lung transplantation may have improved survival compared to those who undergo single lung transplantation (33).

Several important questions related to the care of IPF patients remain unanswered. Acute exacerbations remain a major cause of morbidity and mortality among IPF patients, and there has been limited progress towards understanding the causes, mechanisms, and optimal treatment of these life-threatening disease complications (33a). Pulmonary hypertension is common in severe IPF and leads to worse outcomes (34), yet it is uncertain whether treating secondary pulmonary hypertension could be beneficial, and studies to date have been negative. In addition, obstructive sleep apnea and sleep-disordered breathing are vastly overrepresented in the IPF population (35), often in the absence of typical symptoms (36), and some investigators have suggested there may be reason to perform sleep studies on all IPF patients (36a). However, it remains to be seen whether obstructive sleep apnea treatment impacts disease outcome. Exercise training can improve six-minute-walk distance in patients with IPF (37), but whether this intervention has durable effects requires additional study. Another important question is whether pirfenidone and nintedinab, approved for IPF, have efficacy in other forms of ILD. Numerous ongoing trials are addressing this issue.

Although there is a clear need for additional therapeutic options, the results of most recent studies involving new agents have been disappointing. A large randomized placebo-controlled study targeting lysyl oxidase-like 2 (LOXL2), which catalyzes collagen cross-linking, was terminated early for futility (38). A randomized study of CC-chemokine ligand 2 (39), which regulates recruitment of monocyte-derived macrophages, was also stopped early, and a randomized study of an anti-IL-13 monoclonal antibody did not achieve efficacy endpoints (40). However, despite these results, the field of IPF clinical research has been buoyed by the introduction of pirfenidone and nintedinab, and more studies with novel therapeutics are planned or ongoing. Some studies have shown promising early-phase data (41).

An important issue in the field of IPF clinical studies relates to the design of future phase III clinical trials. Should potential therapies be evaluated in direct comparison to current therapies or as add-on therapies to one of the approved agents? This issue is especially relevant since many investigators believe that rational combination therapies targeting both upstream factors in the injury-repair process and fibroblast activity/matrix deposition are the best strategy for achieving maximal efficacy. In addition, given the rapid advancement in understanding of genetic susceptibility to IPF (discussed below), it will likely be important to stratify patients in future studies based on genetic variables. The best example of the potential importance of this approach to date is a re-evaluation of the PANTHER-IPF study in which patients were genotyped for common single-nucleotide polymorphisms (SNPs) associated with risk for IPF. These investigators found a significant interaction between response to NAC therapy and a SNP in TOLLIP that correlated with efficacy in this subgroup (42), suggesting that personalized pharmacogenetics approaches could be a productive strategy for testing new therapies.

PROGRESS IN UNDERSTANDING GENETIC PREDISPOSITION TO IDIOPATHIC PULMONARY FIBROSIS

It is now recognized that IPF is a gene-by-environment disease with a heterogeneous set of susceptibility genes, along with an ill-defined group of environmental risk factors that includes tobacco smoking. Both common SNPs and rare genetic mutations have been linked to development of IPF (Table 1) (43). To assess the role of common genetic variation in IPF, several genome-wide association studies (GWAS) have now been performed (44–46), resulting in identification of SNPs at 17 different loci that associate with development of IPF, most notably in the promoter region of the Mucin 5B gene, MUC5B (47). This SNP (rs35705950), which has now been confirmed in multiple studies, is located adjacent to a FOXA2 binding site in a region of the MUC5B promoter that is differentially methylated in IPF (48). The minor (T) allele is present in ~18% of the Caucasian population, compared to 60–70% of IPF patients of European ancestry and is associated with increased MUC5B mRNA expression in normal (although not IPF) lungs (47). Although minor allele carriers of rs35705950 have increased risk of developing disease, IPF patients who carry the risk allele appear to have slower disease progression than noncarriers (49). rs35705950 is much rarer among IPF patients of Asian ancestry (49a), underscoring a need for further study of genetic risk for IPF in ethnically diverse populations. Animal studies have suggested that MUC5B regulates airway host defense (50); however, the mechanisms by which altered MUC5B expression influences fibrotic remodeling remain uncertain.

Table 1.

Genetic variants linked to IPF by GWAS and Next-Generation Sequencing Studies

			Common Variants				Rare variants
Locus	Gene		SNP	OR	Progression/Survival		FIP	IPF		Reference
3q26	TERC		rs6793295	1.3			1–2%	0.80%		44, 51,52, 71
4q22	FAM13A		rs2609255	1.29	Protective					44
4q32	NAF1						~1%			60a
5p15	TERT		rs2736100	0.73			7–15%	4–5%		44, 51, 52, 70, 71
6q24	DSP		rs2076295	1.43						44
7q22	Intergenic		rs47274443	1.3						44
8p21	SFTPC						1–2%			61–63
10q22	SFTPA2						<1%			64
10q24	OBFC1		rs11191865	0.8						44
11p15	MUC5B		rs35705950	4.51	Protective					44–49
	MUC2		rs7934606	1.61						44
	TOLLIP		rs111521887	1.48						46
	TOLLIP		rs5743894	1.49						46
	TOLLIP		rs2743890	0.61	Protective					46
13q34	ATP11A		rs1278769	0.79						44
14q12	TINF2						<1%			59
14q21	MDGA2		rs7144383	1.44						46
15q14–15	Intergenic		rs2034650	0.77						44
15q25	AKAP13		rs62025270	1.27						45
16p13	ABCA3						<1%			65, 66
16p13	PARN						2–5%	1.2% – 2.7%		55, 60, 70, 71
17q21	MAPT		rs1981997	0.71						44
17q21	SPPL2C		rs17690703	0.7						46
19q13	DPP9		rs12610495	1.29						44
20q13	RTEL1						3–7%	2.3–3%		53, 54, 55, 70, 71
Xq28	DKC1						<1%			57

OR = odds ratio; FIP = familial interstitial pneumonia

Information regarding rare genetic variants has been generated primarily from studies in individuals with the familial form of IPF, known as familial interstitial pneumonia (FIP). Currently identified disease-associated genes fit into either the telomerase pathway [telomerase reverse transcriptase (TERT) (51, 52), regulator of telomere elongation helicase (RTEL1) (53–55), telomerase RNA component (TERC) (51, 52), dyskerin (DKC1) (57), telomere interacting factor 2 (TINF2) (59), poly(A)-specific ribonuclease (PARN) (55, 60)], nuclear assembly factor 1 (NAF1) (60a) or the surfactant protein pathway [surfactant protein C (SFTPC) (61–63), surfactant protein A2 (SFTPA2) (64), and ATP-binding cassette member A3 (ABCA3) (65, 66)]. Our current estimate is that 15–20% of FIP families share a loss-of-function mutation in one of the telomerase pathway genes, with TERT and RTEL1 being the most common (43). Rare genetic variants in the surfactant protein pathway are much less common in FIP, accounting for no more than 1–2% of cases. Patients with telomerase pathway rare variants have very short telomeres as measured in white blood cells, more rapid disease progression, and often other manifestations of the short-telomere syndrome, including liver and bone marrow disease (51, 52, 67, 68).

The degree of similarity in the genetic underpinnings of familial and sporadic IPF has been an unresolved question in the field. The prevalence of the MUC5B SNP minor allele appears to be similar in patients with familial and sporadic IPF (47), suggesting that common genetic variants are shared in both forms of the disease. For rare genetic variants, prior studies have shown that mutations in the surfactant protein pathway are uncommon in sporadic IPF (69). In contrast, recent data indicate that rare variants in the telomerase pathway occur at a relatively high frequency in patients with sporadic IPF. A recent study using whole-exome sequencing data from 262 subjects with sporadic IPF and unaffected controls found that rare variants in TERT, RTEL1, and PARN were overrepresented in sporadic IPF cases (70). We recently reported data from whole-genome sequencing of 1,510 patients with sporadic IPF and demonstrated that rare variants in TERT, RTEL1, TERC, and PARN were present in ~8.5% of IPF patients, significantly higher than the percentage of control populations (71). In addition, this study identified an interaction between rare variants in TERT and the MUC5B promoter SNP. These findings showed that the MUC5B risk allele was substantially less common in IPF patients who harbored a TERT rare variant than in IPF patients without a telomerase mutation, thus suggesting that the MUC5B polymorphism and TERT rare variants may be separable, independent risk pathways for development of IPF. The finding that rare genetic variants in telomerase pathway genes occur frequently in sporadic IPF points to a potential role for genetic testing. We recently published recommendations for genetic testing in familial IPF (72), and ongoing discussions regarding the role for genetic testing in sporadic IPF are warranted.

Although increasing knowledge regarding the genetics of ILD has not yet translated to improved treatment approaches, identification of disease-associated genes has enhanced understanding of the pathobiology of IPF. The identification of a mutant form of surfactant protein C that segregated with disease in a large FIP family in 2002 (61) led to the identification of endoplasmic reticulum (ER) stress as a common abnormality in IPF epithelium that likely contributes to disease pathogenesis through regulation of epithelial cell survival and repair after injury (73–75). Likewise, the description of telomerase pathway mutations in FIP in 2007 (51, 52) led to the identification of short telomeres as a common phenotype in both familial and sporadic IPF. Overall, peripheral blood cell telomere length is much shorter in IPF than in other chronic degenerative and inflammatory diseases. Short telomeres in peripheral white blood cells (<10th percentile adjusted for age) are identified in at least a third of patients with familial and sporadic IPF (67, 68). Short peripheral blood telomeres are an important biomarker that is independently associated with a worse prognosis (76). Telomere length measured in type II AECs, in contrast to peripheral blood cells, is uniformly reduced in IPF (58, 67) and may not correlate well with peripheral blood telomere length (77). A recent study shows that telomere shortening in the lungs is limited to AECs, and telomeres are shorter in cells from fibrotic areas compared to nonfibrotic areas in the IPF lung (79). Although the interpretation of the studies is not entirely clear, one explanation for these findings could be that peripheral blood telomere length primarily reflects genetic risk, whereas telomere shortening in the alveolar epithelium is a common factor in disease progression. Severe reduction in telomere length is known to result in senescence and cell cycle arrest; however, the exact mechanisms relating telomerase mutations and telomere shortening to fibrotic remodeling in the lung are still being explored. Although prior animal model studies were disappointing in explicating telomere-related disease mechanisms, two recent papers have shown that epithelial specific deletion of the shelterin protein telomere repeat binding factor 1 (TRF1) results in spontaneous fibrotic remodeling with evidence of DNA damage and accumulation of senescent lung epithelial cells (80, 81).

Despite rapid progress in understanding the genetics of pulmonary fibrosis, much remains unknown. The underlying genetic predisposition is unknown in 80% of FIP patients, and current evidence suggests that a variety of genes may be involved, making identification of additional disease-associated genes challenging. In addition, the relationship between rare and common genetic variants is incompletely understood; however, an important clue might be that several common polymorphisms in telomere-associated genes, including TERT, TERC, and OBFC1, are overrepresented in IPF (44), suggesting that combinations of common and rare genetic variants could work in concert to regulate disease-associated phenotypes like telomere shortening. Further progress in the genetics of IPF will likely require large, well-phenotyped cohorts with expanded genetic search space (i.e., whole-genome sequencing) to investigate promoter, intronic, and other regulatory regions, coupled with assessment of gene expression data and in vitro functional evaluation of individual genetic variants.

EVOLVING UNDERSTANDING OF IDIOPATHIC PULMONARY FIBROSIS PATHOGENESIS

In UIP, collagen deposition in the distal lung parenchyma is thought to occur in association with accumulation of activated (myo)fibroblasts in areas subjacent to the epithelial surface, which is composed of hyperplastic type II AECs or epithelium with a bronchiolar appearance. Collapse of remodeled alveoli creates focal areas of fibrosis and stretches the adjacent lung parenchyma, resulting in microhoneycombing and traction bronchiolectasis, which are characteristic of the disorder. In addition to AECs and fibroblasts, inflammatory cells are thought to contribute to the pathogenesis of this disorder by modulating epithelial–fibroblast interactions. The predominant hypothesis explaining disease pathogenesis is that repetitive, environment-derived (micro)injuries to susceptible AECs result in increased cell death, impaired re-epithelialization, and pathological interactions with fibroblasts that lead to persistent activation with excessive collagen and matrix production. AEC susceptibility is thought to be related to genetic predisposition, aging, or senescence. This paradigm has been bolstered in recent years by genetic data and preclinical models showing that epithelial injury can drive subsequent fibrosis.

The importance of stress responses modulating injury-repair responses in AECs has been an area of active investigation in IPF, and altered mitochondrial function has emerged as a potential factor in AEC dysfunction. In AECs, expression of the mitochondrial protective factor PTEN-induced putative kinase 1 (PINK1) is reduced by aging and ER stress (82), resulting in accumulation of damaged mitochondria and decreased cell viability, both of which are characteristic of AECs in IPF. Recently, it was shown that activating transcription factor 3 (ATF3) can mediate downregulation of the PINK1 promoter (83), thus identifying a mechanism linking these processes. Impairment of mitochondrial bioenergetics potentially leads to a feedback loop with persistent ER stress that facilitates fibrotic remodeling. In this regard, it was recently shown that iodothyronine deiodinase 2 (DIO2), an enzyme that activates thyroid hormone, is increased in lungs of patients with IPF (84). This study also showed that DIO2-deficient mice had more severe fibrosis following bleomycin treatment and that supplementation with active thyroid hormone reduced fibrosis and improved mitochondrial bioenergetics through interactions with peroxisome proliferator-activated receptor γ coactivator α (PGC-1α) and PINK1 (84). Together, these studies (82–84) suggest that interventions to improve mitochondrial function in AECs could reduce vulnerability of these cells to injury, thus reversing fibrotic susceptibility.

Recovery from injury of the distal lung requires proliferation and differentiation of AECs to reestablish barrier and gas exchange functions. Identification of subpopulations of AECs that are responsible for regeneration and repair of the injured alveolus is an area of intense interest. Several recent high-profile papers have markedly advanced the field by identifying a progenitor cell niche for AECs that appears to be regulated by Wnt signaling (85, 86), along with other pathways (87), and may be critical for AEC self-renewal and restoration of homeostasis after injury.

Pathological persistence and activation of fibroblasts are responsible for the excess collagen and other matrix products in IPF. Although the origin of pathogenic fibroblasts remains a topic of debate, it was recently shown that the transcription factor T-box gene 4 (TBX4) is an important lineage marker for myofibroblasts that participate in lung fibrosis (88). In addition, it is increasingly recognized that extracellular matrix is functionally abnormal in IPF and may itself drive fibroblast activation (89). Increased matrix stiffness has been shown to activate fibroblasts via mechanosensing pathways (90), resulting in a feed-forward mechanism of progressive fibrotic remodeling once architectural distortion reaches a tipping point.

In addition to epithelial cells and fibroblasts, recent progress has focused on elucidating the role of macrophages in mediating lung fibrosis. Recent papers have indicated that production of transforming growth factor–β (TGF-β) by alternatively activated macrophages plays an important role in fibroblast activation (91, 92). Also, an atypical monocyte population, identified as segregated-nucleus-containing atypical monocytes (SatM) (93), was shown to activate fibroblasts through non-TGF-β-dependent mechanisms during lung fibrosis, thus indicating that several subpopulations of monocytes/macrophages could be involved in regulation of the fibrotic process.

FUTURE DIRECTIONS

We have highlighted recent progress in improving IPF diagnosis and treatment, defining the underlying genetic susceptibility, and advancing understanding of disease pathogenesis. This is a time of rapid progress in IPF, and there are a number of new techniques and approaches whose promise is yet to be fully realized. Expression profiling of lung and peripheral blood cells has been helpful in identifying pathways that are dysregulated in IPF; however, new techniques such as single-cell RNA sequencing show promise to elucidate key factors that contribute to profibrotic interactions between epithelial (94, 95), mesenchymal (96), and immune/inflammatory cells. This approach can be layered onto next-generation sequencing data, epigenetic data, and other “omics” information to elucidate a more complete picture of the pathobiology of IPF. Advances in primary cell and organoid culture have the potential to provide important information about cellular interactions that regulate differentiation and reparative capacity. Together, these approaches hold promise for identifying disease-relevant pathways and cellular phenotypes that can be targeted for therapeutic benefit.