The Diagnosis and Treatment of Pulmonary Fibrosis

Abstract

Background

The different types of pulmonary fibrosis are a subgroup of the interstitial lung diseases (ILDs). They are associated with a chronic and often progressive course.

Methods

This review is based on pertinent publications retrieved by a selective search in the EMBASE and PubMed databases, with an emphasis on articles published from 2000 to 2020.

Results

The most common type of pulmonary fibrosis is idiopathic pulmonary fibrosis (IPF). Among other relevant types, the most important ones are fibrosing hypersensitivity pneumonitis (fHP) and ILDs associated with systemic diseases, all of which are rare and generally carry a poor prognosis. The essential prerequisite to accurate diagnosis is aninterdisciplinary approach, taking account of the clinical, histological, and radiological aspects. The main complications of pulmonary fibrosis are acute exacerbations and pulmonary hypertension; comorbidities are also of prognostic relevance. Treatment of pulmonary fibrosis depends on the subtype and clinical behavior. For IPF, antifibrotic therapy is indicated; fHP, on the other hand, is mainly treated by antigen avoidance and immune modulation. The predominant mode of treatment for systemic disease-associated pulmonary fibrosis is immune suppression. Antifibrotic agents can also be useful in the treatment of other types of progressivepulmonary fibrosis besides IPF.

Conclusion

The differential diagnosis of pulmonary fibrosis, though complex, is clinically essential, as different types of pulmonary fibrosis are treated differently.

The different types of pulmonary fibrosis are a subgroup of the interstitial lung diseases (ILD), a heterogeneous group of diseases affecting the interstitium and/or the alveoli and often the bronchi and bronchioli as well.

Definition.

The different types of pulmonary fibrosis are a subgroup of the interstitial lung diseases (ILD), a heterogeneous group of diseases affecting the interstitium and/or the alveoli and often the bronchi and bronchioli as well.

The ILDs are subdivided into four categories (figure 1):

Figure 1.

The classification of all types of interstitial lung disease (ILD)isin four categories:

I) known causes, II) idiopathic types, III) granulomatous types, IV) further types.

Within these categories, there are types that always take a fibrosing course (red): IPF and idiopathic NSIP.

There are other entities that can take a fibrosing course (blue): drug-associated ILD, collagenosis-associated ILD, pneumoconioses, unclassifiable ILD, fibrosing exogenous allergic alveolitis, fibrosing sarcoidosis.

HP, hypersensitivity pneumonitis; ILD, interstitial lung disease; LAM, lymphangioleiomyomatosis; PAP, pulmonary alveolar proteinosis;

PLHC, pulmonary Langerhans-cell histiocytosis

• ILDs associated with known causes, e.g., connective tissue diseases, pneumoconioses, or drugs

• idiopathic types, such as idiopathic pulmonary fibrosis (IPF) and idiopathic nonspecific interstitial pneumonitis

• granulomatous ILDs, such as sarcoidosis and hypersensitivity pneumonitis (HP)

• other, mostly very rare types.

Aside from the reversible, generally “inflammatory” types of ILD, there are also the so-called fibrosing types, which involve pulmonary fibrosis. Figure 1 contains an overview of the categories of ILD and their subtypes involving pulmonary fibrosis. These diseases take a chronic course; the damage they cause is usually irreversible and often progressive. In this review, we present new information on the precise diagnosis and treatment of the various types of pulmonary fibrosis and their progressive phenotypes (PF-ILD). For reasons of space, we only briefly discuss idiopathic nonspecific interstitial pneumonitis (e1) and fibrosing sarcoidosis (e2). For a further discussion of these, the reader is referred to the corresponding literature.

Clinical course.

Pulmonary fibroses take a chronic course; the damage they cause is usually irreversible and often progressive.

Learning objectives

This review should enable the reader to:

• know the differential diagnosis of pulmonary fibrosis

• be familiar with the multidisciplinary approach to diagnostic assessment and therapeutic decision-making for patients with pulmonary fibrosis

• understand the different therapeutic approaches for different types of pulmonary fibrosis.

Methods

This review is based on pertinent publications retrieved by a selective search in the EMBASE, PubMed and Cochrane Library databases, with an emphasis on articles published from 2000 to 2020. The search employed the searching terms “pulmonary fibrosis,” “biomarker,” “therapy,” “progressive phenotype,” and “patient related outcomes.”

Incidence.

The incidence of idiopathic pulmonary fibrosis lies between 2.8 and 19 cases per 100 000 persons per year.

Prognosis.

According to the findings of the Global Burden of Disease Study, the number of deaths from ILD can be expected to double in the next 20 years.

The classification#epidemiology#and prognosis of the pulmonary fibroses

The most common type of pulmonary fibrosis is idiopathic pulmonary fibrosis (IPF) (1). Other common types of pulmonary fibrosis arise in the setting of systemic diseases, mainly systemic sclerosis and rheumatoid arthritis, or are forms of HP that take a fibrosing course. The precise incidence and prevalence of these conditions in Germany are unknown; it is assumed that more than 100 000 persons in the country suffer from pulmonary fibrosis (www.dzl.de). The incidence of IPF lies between 2.8 and 19 cases per 100 000 persons per year (e3). Approximately one in three persons with rheumatoid arthritis has an ILD, although only 10–20% receive the diagnosis on clinical grounds (2, e4, e5). Up to 50% of patients with systemic sclerosis have pulmonary fibrosis, depending on the subtype (3, e6– e9).

History-taking.

The history is often complex; a questionnaire that can be downloaded at no cost is recommended as an aid to history-taking.

Definitive diagnosis.

The gold standard for definitive diagnosis is interdisciplinary discussion in an ILD board.

The prognosis of the pulmonary fibroses differs from one entity to another. According to the findings of the Global Burden of Disease Study, the number of deaths from ILD can be expected to double in the next 20 years (4), mainly owing to pulmonary fibrosis. The median survival time of patients with untreated IPF is three to five years (1). Cause-of-death analyses in systemic sclerosis have revealed the importance of pulmonary fibrosis accompanying systemic disease, as ILD has now replaced renal crises as the main cause of death in scleroderma patients (5). The situation is similar for rheumatoid arthritis-associated ILD. Analyses have led to the result that the prognosis of patients with ILD is significantly worse than that of patients without ILD (6) and can resemble the prognosis of IPF (e10). Systemic sclerosis–associated ILD takes a more favorable course than IPF, with slower loss of pulmonary function; in HP, mortality is markedly higher in the fibrotic type (e11).

Pathophysiology

The precise causes of most of the idiopathic fibrosing interstitial lung diseases are unknown. Recent studies have linked their pathogenesis to predisposing genetic factors and further environmental influences. Up to 20% of cases of fibrosing lung disease have been found to be familial or due to a familial predisposition, depending on type. The genetic abnormalities predominantly involve soluble mediators and structural proteins. Occupational exposures and environmental factors such as cigarette smoking – an important risk factor for IPF (7), fine particulate matter, organic substances from molds or bird-derived proteins, comorbidities such as gastro-esophageal reflux, and other factors have also been linked to the development of fibrosing lung diseases. ILD are likely triggered by predisposing genetic aberrations that elevate the risk of the disease along with multiple additional risk factors from the environment.

Pathophysiology.

Recent studies have linked the pathogenesis of the idiopathic fibrosing interstitial lung diseases to predisposing genetic factors and further environmental influences. Up to 20% of cases of fibrosing lung disease have been found to be familial or due to a familial predisposition, depending on type.

The diagnostic evaluation and differential diagnosis of the pulmonary fibroses

The various types of pulmonary fibrosis are diagnosed on the basis of the clinical, radiological, and, where necessary, histological findings. In view of the complexity of these diseases and their wide differential diagnosis (eTable), the diagnosis should be established in an interdisciplinary collaboration (figure 2). Specialized ILD centers can help optimize the initial diagnostic evaluation. The symptoms most commonly leading patients to consult a doctor are non-specific, namely, exertional dyspnea and cough. Moreover, any clinical evidence of systemic disease must be noted. In taking the history, the physician should ask about the patient’s past medical history, medications, cigarette smoking, use of illicit substances, possibly toxic environmental exposures, family history, and travel history, preferably with the aid of the relevant patient questionnaire developed by the German Respiratory Society (8). The website www.pneumotox.com is very helpful for answering the question whether any particular drug might be the cause of pulmonary fibrosis. Commonly used drugs that can cause it include amiodarone, nitrofurantoin, chemotherapeutic agents, and immune suppressants (e.g., bleomycin and checkpoint inhibitors).

eTable 1. The differential diagnosis of pulmonary fibroses and PF-ILDs.

Pulmonary fibrosis entities	Typical radiological pattern	Typical histological pattern	Typical BAL findings
Idiopathic pulmonary fibrosis (IPF) / idiopathic NSIP	UIP pattern / NSIP pattern	UIP: heterogeneous interstitial fibrosis; fibroblast foci; myogenic metaplasia; marked destruction of architecture NSIP: uniform, chronic interstitial pneumonitis; little interstitial fibrosis	IPF: neutrophilia & mild eosinophilia NSIP: a) cellular type: lymphocytic b) fibrosing type: lymphocytic or mixed-cell
Connective tissue disease– or rheumatoid arthritis (RA)–associated interstitial lung disease (ILD)	collagenoses: NSIP, OP, NSIP-/OP overlap, LIP, rarely UIP RA-ILD: UIP > NSIP	NSIP: uniform, chronic interstitial pneumonitis; little interstitial fibrosis OP: intra-alveolar granulation tissue proliferation LIP: chronic interstitial pneumonitis; prominent follicular hyperplasia UIP: heterogeneous interstitial fibrosis; fibroblast foci; myogenic metaplasia; marked destruction of architecture	RA-ILD: often neutrophilic collagenoses: LIP: lymphocytic other collagenoses: depending on subtype, occasionally lymphocytic
Asbestosis	UIP/NSIP and pleural plaques	UIP: heterogeneous interstitial fibrosis; fibroblast foci; myogenic metaplasia; marked destruction of architecture NSIP: uniform, chronic interstitial pneumonitis; little interstitial fibrosis	mixed-cell, with neutrophils predominating
Drug-or radiation-induced ILD	nonspecific, all patterns are possible (e.g., NSIP, UIP, consolidations, alveolitis)	cf. individual types of pulmonary involvement	depending on pattern of damage, often lymphocytic or eosinophilic
Fibrosing HP	various fibrosis patterns, mainly reticulations, traction bronchiectases, less commonly UIP or NSIP pattern, usually not more prominent at the bases; possible demonstration of a mosaic pattern, centrilobular nodules	typical HP: interstitial pneumonitis; granulomas; organizing pneumonitis; otherwise depending on particular manifestation: UIP: heterogeneous interstitial fibrosis; fibroblast foci; myogenic metaplasia; marked destruction of architecture NSIP: uniform, chronic interstitial pneumonitis; little interstitial fibrosis	non-fibrosing HP: lymphocytic fibrosing, chronic HP: partly lymphocytic (or mixed-cell)
Fibrosing sarcoidosis	reticulations most prominent in the upper fields, bronchiectases, lymphadenopathy	sarcoidosis: granulomatous inflammation; sparse pneumonitis; heterogenous fibrosis	often lymphocytic, occasionally neutrophilic
Unclassifiable ILD	nonspecific, all patterns possible, often UIP-like and NSIP patterns	cf. individual types of pulmonary involvement	nonspecific pattern
Rare causes (Dyskeratosis congenita; Hermansky-Pudlak syndrome)	UIP/NSIP	UIP: heterogeneous interstitial fibrosis; fibroblast foci; myogenic metaplasia; marked destruction of architecture NSIP: uniform, chronic interstitial pneumonitis; little interstitial fibrosis	neutrophilic

HP, hypersensitivity pneumonitis; ILD, interstitial lung disease; LIP, lymphocytic interstitial pneumonitis; NSIP, nonspecific interstitial pneumonitis; OP, organizing pneumonitis; UIP, usual interstitial pneumonitis

Figure 2.

Diagnostic algorithm for suspected pulmonary fibrosis. (German consensus recommendations for the evaluation of interstitial lung diseases are in preparation under the aegis of the German Respiratory Society. This algorithm is based on Travis WD, Costabel U, Hansell DM, et al.: An official American Thoracic Society/European Respiratory Society statement: Update of the international multidisciplinary classification of the idiopathic interstital pneumonias. Am J Respir Crit Care Med 2013; 188: 733–48)

ANA, antinuclear antibodies; ANCA, antineutrophilic cytoplasmic antibodies; BAL, bronchoalveolar lavage; HP, hypersensitivity pneumonitis; ENA, extractable nuclear antibodies; CCP, cyclic citrullinated peptide; CT, computed tomography; HRCT, high-resolution computed tomography; ILD, interstitial lung disease; IPF, idiopathic pulmonary fibrosis; RA, rheumatoid arthritis; RF, rheumatoid factor; TBB, transbronchial biopsy; UIP, usual interstitial pneumonitis

Differential diagnoses.

The main elements of the differential diagnosis of pulmonary fibrosis are IPF, idiopathic nonspecific pneumonitis, pulmonary fibrosis associated with a systemic disease, and fibrosing HP.

Auscultation often reveals typical inspiratory crackles, called sclerosophonia. The remainder of the physical examination should concentrate on any potential clinical evidence of rheumatological systemic disease. Pulmonary function testing is used mainly to assess the severity of disease; it generally demonstrates restrictive deficits of both ventilation (as measured by the forced vital capacity [FVC]) and diffusion (low diffusion capacity for carbon monoxide [DLCO]). Nonetheless, normal values on pulmonary function tests do not rule out pulmonary fibrosis.

Laboratory testing is based on serologic screening for systemic diseases with antinuclear antibodies, rheumatoid factor, and collagenosis- and vasculitis-associated antibodies (1, e12), and, where indicated, a myositis panel (9).

Bronchoalveolar lavage.

The main questions that can be answered by bronchoalveolar lavage involve cellular morphology, differential cytology, and microbiology

Lymphocytosis.

Lymphocytosis (> 30%) can be found in HP, sarcoidosis, nonspecific interstitial pneumonitis, toxic drug reactions, and collagenoses.

Imaging is an essential component of the diagnostic evaluation of pulmonary fibrosis, and histology often is as well. The various radiological and histological patterns can be characteristic for certain ILD entities but are never pathognomonic. Some patterns, such as the UIP pattern, can appear in multiple types of pulmonary fibrosis (eTable). In individual cases, only a consideration of the overall picture in an interdisciplinary setting enables the proper classification of the disease pattern and the disease entity (7, e13).

Imaging should be performed with high-resolution, non-contrast, thin-section computed tomography (CT), performed in volume technique rather than with sequential acquisition (which leads to gaps between slices). The important CT patterns in pulmonary fibrosis (Figure 3, eFigures 2– 7) are the usual interstitial pneumonitis (UIP) pattern and the pattern of nonspecific interstitial pneumonitis (NSIP); attention should also be paid to the characteristic findings of non-idiopathic ILDs, e.g., the mosaic pattern or ground-glass-density micronodules in HP (e14). Depending on the clinical and radiological findings, the indication for an endoscopic diagnostic study should be established by an interdisciplinary team. Such studies involve either bronchoalveolar lavage or transbronchial biopsy (figure 2). The main questions that can be answered by bronchoalveolar lavage involve cellular morphology, differential cytology, and microbiology (eTable). Lymphocytosis (> 30%) can be found in HP, sarcoidosis, nonspecific interstitial pneumonitis, toxic drug reactions, and collagenoses. The findings of IPF include neutrophilia and eosinophilia, but the cytologic picture is nonspecific. Forceps biopsies should be avoided in pulmonary fibrosis in favor of a transbronchial cryobiopsy performed in an experienced center; the diagnostic yield of a cryobiopsy carried out by experienced hands is the same as that of a surgical lung biopsy (10). If IPF is suspected and certain patterns (mainly the UIP pattern) are found in the proper clinical context, no further studies are needed to establish the diagnosis (Figure 2, Figure 3a). If fibrosing HP is suspected and the pattern is typical (e14), bronchoalveolar lavage suffices (Figure 3b). If a systemic disease or pneumoconiosis is thought to be the cause of pulmonary fibrosis, a bronchoalveolar lavage can be considered, but a biopsy is generally not indicated. In all other cases, cryobiopsy is indicated as a rule. If cryobiopsy fails to provide a definitive answer, the interdisciplinary team must consider the indication for surgical lung biopsy. The histopathological findings, like the radiological findings, are classified by pattern (Figure 4, eFigure 1). Finally, all findings should be discussed in an interdisciplinary ILD board. If no diagnosis can be made with confidence, e.g., because the findings are contradictory or because a biopsy could not be performed for medical reasons, the patient should be diagnosed with unclassifiable ILD.

Figure 3.

Figure 3: Imaging of pulmonary fibroses

1. This 64-year-old former smoker has been suffering from a dry cough of progressive severity for several months. The native thin-section computer-tomographic images with axial reconstruction show a reticular pattern, most pronounced in basal and peripheral areas, with traction bronchiectases (arrowheads) and multilayered honeycombing (arrows). This is a typical pattern of UIP.

2. This 65-year-old woman owns parakeets. Native thin-section CT shows extensive ground-glass infiltrates (*) with a mosaic pattern (arrowheads) and a few traction bronchiectases (arrows). The pattern of radiological findings is not consistent with UIP. Interdisciplinary diagnostic evaluation revealed that the findings were due to an exogenous allergic alveolitis.

eFigure 2 a, b.

This 64-year-old man with a history of smoking (60 pack-years, quit 18 years ago) has been suffering from a dry cough of progressive severity for several months. The native thin-section computer-tomographic images with axial and sagittal reconstruction (eFigures 2a, 2b) show a reticular pattern, most pronounced in basal and peripheral areas, with traction bronchiectases (arrowheads) and multilayered honeycombing (arrows). There is also some degree of centrilobular emphysema (*). This is a typical pattern of UIP. Idiopathic pulmonary fibrosis (IPF) was diagnosed by the ILD Board and antifibrotic treatment was initiated. One year later, the patients symptoms and pulmonary function tests were stable.

eFigure 7:

The patient is a 68-year-old male former smoker with rheumatoid arthritis (RA) under treatment with methotrexate and leflunomide who has suffered from increasingly severe dyspnea for three years. The non-contrast thin-section CT shows a mainly basal and peripheral reticular pattern (*) with traction bronchiectases (arrows) and multilayered honeycombing (arrowheads). This is a typical UIP pattern. Interdisciplinary evaluation led to the diagnosis of RA-associated pulmonary fibrosis.

eFigure 3 a, b, c.

The patient is a 63-year-old male former smoker with progressive dyspnea. The non-contrast thin-section CT reveals, in transverse reconstruction (eFigure 3a), parasagittal reformatting (eFigure 3b), and a miminal intensity projection (eFigure 3c), a reticular pattern, most pronounced in basal and peripheral areas, with traction bronchiectases (arrows). There is also some degree of centrilobular emphysema (*). In view of the lack of honeycombing, this is probably a UIP pattern.

Figure 4.

Figure 4: Lung biopsy findings a) with a UIP pattern in IPF, b) with fibrosis in HP

1. Prominent fibrotic reorganization and destruction of architecture in the (alveolar) parenchyma with extensive involvement of the visceral pleura. Areas of complete and incomplete scarring and apparently uninvolved parenchyma are mixed together in a heterogeneous distribution. Particularly where these two types of tissue abut one another, there are often accumulations of myofibroblasts of variable thickness, embedded in an immature extracellular matrix of myxoid appearance; the covering epithelium here is mostly hyperplastic. These are the so-called fibroblastic foci (see insert). Other specific patterns of damage, e.g., dominant pneumonitis, and epithelioid-cell reaction, granulomatous inflammatory infiltrates, or intra-alveolar reorganizing processes, such as in an organizing pneumonitis, are typically not found. Hematoxylin-Eosin stain, enlargement 20 ×, insert 40 ×; scale bar 500 µm, insert 100 µm.

2. Typical appearance of fibrosing exogenous allergic alveolitis with interstitial lymphocytic infiltrates of varying thickness and accompanying granulomatous infiltration, with the multifocal formation of multinucleated giant cells, negative images of cholesterol crystals, and accompanying deposition of extracellular matrix and fibrosis. Focally, particularly in areas of highly active pneumonitis, there are frequent areas of intra-alveolar granulation-tissue-like mesenchymal proliferation, indicating organizing pneumonitis (see insert). Hematoxylin-Eosin stain, enlargement 100 ×, insert 400 ×, scale bar 100 µm, insert 100 µm.

HP, hypersensitivity pneumonitis; IFP, idiopathic pulmonary fibrosis;

UIP, usual interstitial pneumonia

eFigure 1.

Lung biopsies a–c) with a UIP pattern in IPF; d–f) fibrosis in HP

a–c) Prominent fibrotic reorganization and destruction of architecture in the (alveolar) parenchyma with extensive involvement of the visceral pleura.

Areas of complete and incomplete scarring and apparently uninvolved parenchyma are mixed together in a heterogeneous distribution. Particularly where these two types of tissue abut one another, there are often accumulations of myofibroblasts of variable thickness, embedded in an immature extracellular matrix of myxoid appearance; the covering epithelium here is mostly hyperplastic. These are the so-called fibroblastic foci. There a few inflammatory cells dispersed in the interstitium, mainly macrophages and lymphocytes, but these are not characteristic of fibrotic reorganizing processes. Other specific patterns of damage, e.g., dominant pneumonitis, and epithelioid-cell reaction, granulomatous inflammatory infiltrates, or intra-alveolar reorganizing processes, such as in an organizing pneumonitis, are typically not found. a) 20×, b) 40×, c) 60× enlargement.

d–e) Typical appearance of subacute, fibrosing exogenous allergic alveolitis with interitial lympocytic infiltrates of varying thickness indicating interstitial pneumonitis, and accompanying granulomatous infiltration, with the multifocal formation of multinucleated giant cells, negative images of cholesterol crystals, and accompanying deposition of extracellular matrix and fibrosis. Focally, particularly in areas of highly active pneumonitis, there are frequent areas of intra-alveolar granulation-tissue-like mesenchymal proliferation, indicating organizing pneumonitis. Classic fibroblastic foci, such as are typical of UIP in particular, cannot be seen in the classic picture of exogenous allergic alveolitis. d) 100×, e) 200×, f) 400× enlargement.

Specific features of different pulmonary fibroses

Idiopathic pulmonary fibrosis

IPF is a type of fibrosis limited to the lung that has no recognizable cause and is associated with the UIP pattern. It mainly arises in old age among male smokers and ex-smokers. Its course is variable, but always chronically progressive and irreversible, with a loss of 200 mL of vital capacity per year in untreated patients (11), regardless of baseline pulmonary function.

Idiopathic pulmonary fibrosis.

IPF is a type of fibrosis limited to the lung that has no recognizable cause and is associated with the UIP pattern. It mainly arises in old age among male smokers and ex-smokers.

Pulmonary fibrosis as a manifestation of systemic disease

The pulmonary manifestations of rheumatological diseases take many different forms, and nearly any systemic disease can be associated with an ILD. The differential diagnosis should always include drug toxicity, as certain immune suppressants are pneumotoxic. It must be pointed out in this regard, however, that the pulmonary toxicity of methotrexate is overrated (12). Methotrexate can indeed rarely induce subacute or acute pneumonitis, but it does not cause chronic pulmonary fibrosis (which is, rather, a manifestation of the disease under treatment). Recent studies suggest that methotrexate can actually slow the development of pulmonary fibrosis (e15, e16).

Pulmonary fibrosis does not always become clinically manifest later than the systemic disease with which it is associated; in fact, in up to 20% of cases, it shows the way to the initial diagnosis of a rheumatic disease (13). In some 10% of cases, pulmonary fibrosis precedes the overt systemic disease and is initially misdiagnosed as idiopathic (14). Idiopathic ILD with clinical and radiological evidence consistent with a collagenosis, but unaccompanied by the full clinical picture of such a disease, is called interstitial pneumonia with autoimmune features (IPAF); this newly introduced term is currently only used for scientific purposes (15).

Fibrosing exogenous allergic alveolitis

Exogenous allergic alveolitis is an immune-mediated allergic disease that is induced by the inhalation of fine particles (usually, organic particles) by persons predisposed to this type of reaction. The inducing substances include proteins derived from birds and components of micro-organisms and molds. The fibrosing type of exogenous allergic alveolitis is often difficult to distinguish from other types of pulmonary fibrosis, and from IPF in particular (16), especially when the culprit antigen cannot be identified (17).

Fibrosing exogenous allergic alveolitis.

Exogenous allergic alveolitis is an immune-mediated allergic disease that is induced by the inhalation of fine particles (usually, organic particles) by persons predisposed to this type of reaction.

The phenotype of progressive pulmonary fibrosis

While IPF nearly always takes a progressive course, the other types of pulmonary fibrosis can potentially stabilize or improve under treatment. Depending on the entity, however, 20–35% of cases of pulmonary fibrosis take on a progressive phenotype (PF-ILD) despite the current best treatment (18). Although there is still no uniform definition of PF-ILD, the term is used when functional and/or radiological worsening is seen in conjunction with more severe symptoms. The responsible pathophysiological mechanisms are not yet entirely clear but seem to overlap with those underlying IPF. In any case of PF-ILD, it should be investigated whether external factors are present, such as a persistent exposure in HP, and the further management should be discussed by an interdisciplinary team.

The complications and comorbidities of the pulmonary fibroses

Worsened respiratory function is the main cause of death in IPF, and probably also in other types of pulmonary fibrosis. The most significant mode of worsening is the acute exacerbation, which carries an in-hospital mortality of more than 50% and a mean survival time of less than 5 months. Acute exacerbations can be idiopathic or they can be triggered, for example, by an infection or in the postoperative setting. They are clinically characterized by rapid respiratory worsening, which always requires further diagnostic evaluation. High-resolution non-contrast thin-section computed tomography must be performed and generally shows ground-glass-type changes, consolidations, or both. The treatment consists of high-dose steroids in combination with antibiotics, although there is very little evidence supporting this particular form of treatment and treatments differ internationally (19).

Pulmonary hypertension usually develops as a complication of fibrosis but can also arise independently as a complication of the underlying systemic disease. It is characterized by markedly increased shortness of breath and carries a poor prognosis, particularly when associated with IPF. There is not yet any promising approach to the treatment of pulmonary hypertension in ILD beyond diuretic administration; most studies of drugs to treat pulmonary arterial hypertension have not shown any useful effect. It is, therefore, up to the specialist to direct the treatment in each individual case (20).

Aside from acute respiratory exacerbations, comorbidities play a central role, as they markedly influence the affected patients’ quality of life, as well as their prognosis (21, e17, e18). Lung cancer, in particular, determines the outcome in many cases (figure 5). It is presumed that the comprehensive treatment of comorbidities can improve the prognosis of patients with pulmonary fibrosis, although this has not been unequivocally demonstrated (22). A first indication that this is the case comes from the WRAP-IPF phase II trial, which dealt with the hypothesis that surgical treatment of gastroesophageal reflux might be relevant to the prognosis. The conceptual background of this trial was the presumption that micro-aspirations due to reflux might induce pulmonary fibrosis or worsen it if already present. Even though the findings of this trial were not statistically significant, it did reveal clinically relevant effects on the course of pulmonary function parameters as well as on the rates of exacerbation, hospitalization, and mortality (23).

Figure 5.

The prognostic significance of comorbidities in patients with idiopathic pulmonary fibrosis (after Ref. 21). The prevalence of the more common comorbidities of IPF and their association with mortality are shown (the “comorbidome”). In combination with IPF, lung cancer carries the highest mortality, but cardiovascular diseases, COPD, and diabetes also have a major influence on the prognosis of patients with IPF.

CHD, coronary heart disease; COPD, chronic obstructive pulmonary disease; IPF, idiopathic pulmonary fibrosis

* Risk: the comorbidities with the strongest association with mortality are indicated. The closer to the center, the higher the risk of death; size indicates prevalence.

Pulmonary hypertension as a complication of fibrosis.

Pulmonary hypertension usually develops as a complication of fibrosis but can also arise independently as a complication of the underlying systemic disease. It is characterized by markedly increased shortness of breath and carries a poor prognosis, particularly when associated with IPF

The treatment of pulmonary fibrosis

Non-pharmacological treatment options

Aside from the prevention of infection by vaccination against pneumococci and influenza (www.rki.de) and general measures, the goal of non-pharmacological treatment is to exert a beneficial effect on the symptoms of the disease and on the patients’ physical condition. Long-term oxygen therapy is one form of non-pharmacological treatment. There are guideline criteria for long-term oxygen therapy at rest (e19, e20); yet a recent randomized, open crossover trial showed that its use during exertion, in particular, is associated with a statistically significant improvement of the quality of life (albeit with a relatively weak effect) (24). Immediately after specialized pulmonary rehabilitation, significant effects are seen on physical performance, the quality of life, and subjective shortness of breath. The effect is often lost, however, mainly when the patient does not keep up with the recommended lung exercises (25). In pulmonary fibrosis, just as in advanced cancer, palliative medical measures should be initiated early in order to sustain the quality of life and keep symptoms under control (26). This is particularly so for patients who are not candidates for lung transplantation, which is the only way that pulmonary fibrosis can be treated with curative intent (7).

Non-pharmacological treatment options.

Aside from the prevention of infection by vaccination against pneumococci and influenza (www.rki.de) and general

measures, the goal of non-pharmacological treatment is to exert a beneficial effect on the symptoms of the disease and on the patients’ physical condition.

Pharmacological treatment options

Treatment with drugs is directed against the particular type of pulmonary fibrosis from which the patient is suffering. The treatment of IPF has undergone a major change in recent years. Immune-modulating and anti-oxidative drugs used to be given (27), but the findings of the placebo-controlled PANTHER trial imply that they should no longer be used, because excess mortality was seen when immune modulation was the mainstay of treatment (28). The treatment of choice for IPF is now antifibrotic therapy with nintedanib or pirfenidone, which should be recommended to symptomatic patients as soon as the diagnosis is made, according to the current German IPF guideline (29). Nintedanib, a receptor tyrosine kinase inhibitor, has been studied in patients with IPF in three randomized and controlled trials, the phase II TOMORROW trial and the twin phase III trials INPULSIS-1 and -2 (e21; 30). In the INPULSIS trials, 1066 patients were treated with either nintedanib or placebo for 52 weeks. The trials revealed a statistically significant reduction of the annual decline in FVC (primary endpoint), which was approximately halved (absolute difference between nintedanib and placebo at 52 weeks: 109.9 ml). In the pooled analysis (but not in INPULSIS-1 alone), a statistically significant effect was also seen in one of the main secondary endpoints, the time until the first adjudicated acute exacerbation. Nonetheless, no effect on the quality of life was seen in the overall patient cohort, though a beneficial effect has been described in patients with more advanced disease (e22). Pirfenidone has pleiotropic effects, some of which are still not clear; it has been studied in four randomized and controlled phase III trials (e23, 31, 32). Pooled analysis of the CAPACITY- and ASCEND trials showed that pirfenidone slows the decline of FVC to a statistically significant extent (absolute difference between perfenidone and placebo at 52 weeks: 148 mL). Moreover, the treatment yielded statistically significant improvements in progression-free survival, walking distance, and subjective shortness of breath (e24). The current German guideline recommends both drugs with recommendation grade A and evidence level 1a (29). In addition, an analysis of the German INSIGHTS-IPF registry revealed that antifibrotic therapy can significantly lower mortality (hazard ratio [HR]: 0.63, 95% confidence interval [0.45; 0.87]; p = 0.005) (e25). As disease progression continues to be seen in many patients despite the demonstrated effects of antifibrotic drugs, further treatment strategies are now under discussion. Combining the approved therapies is not recommended at present (29), because only toxicity studies and no efficacy studies on this question are available to date. Several new drugs, some of them promising, are now being studied in phase I, II, and III trials (33).

The treatment of fibrosing HP is based on a much weaker evidence base than that of IPF. It is important for the patient to avoid further exposure to the culprit antigen; this is associated with a better prognosis even when fibrosis is already present. Thus, a diligent search for persistent antigen exposure should be carried out repeatedly throughout the patient’s course (e26, 34). Immune suppressants are often used, but there are only sparse data on their effects. In a retrospective study involving 60 patients, one year of treatment with either mycophenolate or azathioprine resulted in an increased DLCO and a constant FVC compared to the pretreatment lung function parameters, while also enabling a marked reduction of the prednisone dose (35). In another retrospective study, a combination of immune modulators was found to have a similar therapeutic effect to that of (high-dose) steroid therapy, with a much lower rate of side effects (e27). In the absence of a current treatment guideline, fibrosing HP is generally treated with immune modulators; data are increasingly becoming available on the use of antifibrotic drugs as well. An open feasibility study suggests that a combination of antifibrotic drugs and immune modulators can have positive effects (e28), but large-scale randomized trials of this treatment strategy for fibrosing HP are currently lacking.

Fibrosing HP.

The treatment of fibrosing HP is based on much weaker evidence than that of IPF. The patient must avoid further antigen exposure; this improves the prognosis even when fibrosis is already present. There should be repeated searches for persistent antigen exposure throughout the patient’s course.

The risk of progression in pulmonary fibrosis associated with a systemic inflammatory disease.

In pulmonary fibrosis associated with systemic inflammatory diseases, the risk of progression of ILD is probably lower if the activity level of the underlying disease is low. The inflammatory systemic disease should be brought into remission if possible.

In pulmonary fibrosis associated with systemic inflammatory diseases, the risk of progression of ILD is probably lower if the activity level of the underlying disease is low. It follows that the inflammatory systemic disease should be brought into remission if possible. Thus, immune modulation is currently considered the treatment of choice for this type of pulmonary fibrosis as well (36). There is, however, sparse evidence for any specific drug.

The therapeutic role of methotrexate was discussed above; in ILD associated with rheumatoid arthritis, the use of abatacept (e29) and rituximab (e30, e31) is under discussion. The evidence concerning the use of these drugs is derived mainly from retrospective studies or registry data and consists mainly of clinical experience.

It must also be borne in mind in the choice of treatment that certain biological agents, mainly the TNF-alpha inhibitors, can worsen a preexisting rheumatoid arthritis-associated ILD (e32) or increase the risk that one will develop (e33).

For the treatment of systemic sclerosis–associated pulmonary fibrosis, the main drugs now used are the immune mediators mycophenolate and cyclophosphamide, usually in combination with low-dose prednisone; the efficacy of these drugs was investigated in the Scleroderma Lung Study (SLS) I and II (37, 38). Cyclophosphamide is recommended in guidelines even though its effect is relatively weak, as documented in a Cochrane analysis dealing mainly with its use against systemic sclerosis–associated ILD (e34). Mycophenolate has comparable therapeutic effects to cyclophosphamide (38), but with lower discontinuation rates and fewer side effects.

Antifibrotic therapy with nintedanib to treat systemic sclerosis–associated ILD was studied in the recent SENSCIS trial (39). This randomized and controlled phase III trial revealed a beneficial effect on the course of pulmonary function, and the side effects of nintedanib were well tolerated. The relative effect size (ca. 45%) was similar to that seen in IPF studies, although the absolute values differed in that the decline was markedly less in systemic sclerosis. Nintedanib was taken in combination with preexisting mycophenolate therapy by approximately half of the patients, but mycophenolate status was not considered in the randomization assignment to nintedanib or placebo; the relative differences in the subgroups were similar, although the decline of FVC in the mycophenolate groups was markedly less (39). Nintedanib was recently approved for this application, but the issue of its use with other drugs in combination or in sequence has not yet been clarified (e35).

There are also data concerning tocilizumab, an IL-6 antagonist, in systemic sclerosis. In a phase III trial, no statistically significant effect on the primary endpoint (skin involvement) was found, but a clinically relevant improvement compared to placebo was seen in the main secondary endpoint, namely, the course of FVC. The findings of this trial must be considered negative overall in the light of the failure to reach the primary endpoint (e36).

The treatment of progressive phenotypes of non-IPF pulmonary fibrosis

As PF-ILDs resemble IPF both in their clinical features and in their prognosis, trials of antifibrotic drugs have been carried out in this group of diseases as well. The INBUILD trial (40), a randomized and controlled phase III trial carried out on 663 patients with a PF-ILD (iNSIP, fHP, unclassifiable ILD, ILD in collagenosis), investigated the therapeutic effect of nintedanib. Statistically significant effects of nintedanib on the course of pulmonary function were described that closely resembled those seen in the INPULSIS trials, not only for the UIP radiological pattern but for all of the patterns studied, with consistency across all types of pulmonary fibrosis. Similar findings were obtained in a randomized, controlled trial (RELIEF) conducted by the German Center for Lung Research (Deutsches Zentrum für Lungenforschung) (e37). Even if the precise diagnostic algorithm, the definition of PF-ILD, and the question of preceding and accompanying treatments remain to be fully clarified, these trials are the first ones showing the way to new therapeutic opportunities for these groups of patients whose prognosis has been poor until now.

The treatment of progressive phenotypes of non-IPF pulmonary fibrosis.

Very recent studies have yielded new options for antifibrotic treatment in patients with these diseases, whose prognosis has been poor until now.

Further information on CME.

• Participation in the CME certification program is possible only over the Internet at cme.aerzteblatt.de. This unit can be accessed until 4 March 2022. Submissions by letter, e-mail or fax cannot be considered.

• All new CME units can be accessed for 12 months. The results can be downloaded four weeks after the start of the CME unit. Please note the relevant access deadlines on cme.aerzteblatt.de.

• This article has been certified by the North Rhine Academy for Continuing Medical Education. Participants in the CME program can manage their CME points with their 15-digit “uniform CME number” (einheitliche Fortbildungsnummer, EFN), which is found on the CME card (8027XXXXXXXXXXX). The EFN must be stated during registration on www.aerzteblatt.de (“Mein DÄ”) or else entered in “Meine Daten,” and the participant must agree to communication of the results.

eFigure 4 a, b.

The patient is an 81-year-old male former smoker with progressive dyspnea. The non-contrast thin-section CT reveals, in transverse reconstruction (eFigure 4a) and sagittal reformatting (eFigure 4b), a mainly peripheral mixed reticular (arrows) and consolidating pattern (i.e., organizing pneumonitis) (arrowheads) without traction bronchiectases, honeycombing, nodules, cysts, or any clear apicobasal gradient. This is, therefore, an indeterminate pattern for UIP.

eFigure 5 a, b.

The patient is a 64-year old woman, with scl 70-positive scleroderma. The non-contrast thin-section CT reveals, in transverse reconstruction (eFigure 5a) and sagittal reformatting (eFigure 5b), a marked, mainly basal destruction of lung parenchyma with traction bronchiectases (arrows). There is also notable dilatation of the esophagus (*) with chyme retention and a fluid level. There is no evidence of honeycombing, nodules, cysts, consolidations, or any relevant ground-glass infiltrates. This is thus a probable UIP pattern. The ILD Board considered these findings to be compatible with scleroderma-associated pulmonary fibrosis, also because of the nearly horizontal pattern of pulmonary parenchymal destruction.

eFigure 6 a, b:

The patient is a 76-year-old man with progressive dyspnea. The non-contrast thin-section CT reveals a perilymphatic reticular pattern (arrowheads) with ground-glass infiltrates, some of them dense (*), and traction bronchiectases (arrows) (eFigure 6a). There is no evidence of honeycombing, nodules, or cysts. This pattern of radiological findings is incompatible with UIP. Further diagnostic evaluation yielded the diagnosis of an antisynthetase syndrome. Over the ensuing year, the ground-glass infiltrates largely resolved under treatment. The remaining findings included mainly basal traction bronchiectases (arrows) and a mosaic pattern as in NSIP (eFigure 6b).