**1. Introduction**

Lung involvement is common in patients with rheumatic diseases (RDs), causing substantial morbidity and mortality in these patients. Individual RDs tend to be associated with a characteristic lung disease pattern where the key structure of the injury, as well as the critical cells and cytokines involved, are different [1].

Among the diverse forms of lung involvement, interstitial lung disease (ILD) and pulmonary arterial hypertension (PAH) are the two most important manifestations in patients with RD, leading to grave prognoses [2]. Of note, rheumatoid arthritis (RA), myositis, and systemic sclerosis (SSc) are the major systemic RDs, in which a significant proportion of the patients develop and die from ILD and/or PAH. However, recent advances in pharmacologic interventions have been shown to delay the disease progression of these lung conditions and also improve patient survival. In this review, we introduced updated knowledge on the treatment options for ILD and PAH in RDs, focusing on targeted therapies.

#### **2. Interstitial Lung Disease (ILD)**

Lung involvement in RDs most commonly takes the form of ILD. In particular, more than two thirds of the patients exhibit ILD in SSc and myositis [3,4]. Clinically symptomatic ILD is less frequent in RA than SSc and myositis, found only in about 10% of the patients [5]. However, the associated morbidity and mortality are never less. RD-associated ILD (RD-ILD) can be classified as usual interstitial pneumonia (UIP), nonspecific interstitial pneumonia (NSIP), organizing pneumonia (OP), diffuse alveolar damage (DAD), lymphoid interstitial pneumonia (LIP) and others, according to the radiologic and/or pathologic– morphologic patterns presented by the revised 2013 American Thoracic Society/European

99

**Citation:** Kang, E.H.; Song, Y.W. Pharmacological Interventions for Pulmonary Involvement in Rheumatic Diseases. *Pharmaceuticals* **2021**, *14*, 251. https://doi.org/ 10.3390/ph14030251

Academic Editor: Barbara Ruaro

Received: 29 January 2021 Accepted: 3 March 2021 Published: 10 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Respiratory Society classification of idiopathic interstitial pneumonias [6]. In RDs, the two most predominant types of ILDs are NISP and UIP [1]. While the prognosis is similar between RD-NISP and idiopathic NSIP, the prognosis of the RD-UIP other than RA-UIP is better than that of idiopathic UIP or idiopathic pulmonary fibrosis (IPF) [7].

#### *2.1. Rheumatoid Arthritis Associated ILD (RA-ILD)*

Potential targets of the lung injury in RA include almost all components of the lung structures. Thus, lung injury associated with RA encompasses a wide spectrum of disorders such as parenchymal (ILD), airway (bronchiectasis or bronchiolitis), pleural (pleurisy), and vascular diseases. Among them, ILD is most common.

#### 2.1.1. Clinical Features of RA-ILD

Approximately 10% of RA patients suffer clinically significant RA-ILD [5], with 8–9 times the lifetime risk of ILD development among RA patients compared to that of the general population (7.7% vs. 0.9%, respectively) [8]. Thirty-four percent of RA-ILD was found to occur within one year of RA diagnosis [9] and the risk of RA-ILD increases with RA duration and autoantibody (rheumatoid factor and/or anticitrullinated protein antibody) titers [10,11].

Unlike other RD-ILDs in which NSIP is the most prevalent histopathologic pattern, up to half of the RA-ILD cases show the UIP pattern, with NSIP as the second most common pattern [1]. With a heterogeneous progression rate across individuals, symptoms of RA-ILD usually progress over time once clinically present. The UIP patterns tend to demonstrate extensive disease at baseline and more rapid pulmonary function decline during followup, thus are associated with a worse prognosis [1,12]. The mortality of the patients with RA-ILD was as high as three times that of the patients with RA alone, with more than one third of RA-ILD patients being dead at five-years after ILD diagnosis [8,9].

Although it is considered that the natural progression of RA-ILD is heterogeneous, as in IPF, and a subgroup of those whose lung function progressively declines would show a grave prognosis, the exact natural history of RA-ILD is not fully known, particularly regarding acute exacerbation. In addition to the chronic lung function loss, acute exacerbation is another cause of the mortality associated with RA-ILD. However, a substantial proportion of patients with RA-ILD would suffer mixed patterns.

Acute exacerbation is a fatal condition characterized by a rapidly progressive respiratory failure. It has been well recognized not only in IPF and NSIP [6], but also in RD-ILD, particularly RA-ILD [13]. The radiographic characteristics of the acute exacerbation are defined as ground-glass opacities (GGOs) or consolidations newly overlaid on the background reticular abnormalities. Two thirds of RA-ILD patients with acute exacerbation died during the initial episode, causing a 2.5-fold increased mortality among RA-ILD patients [14]. The risk factors for the acute exacerbation of RA-ILD include UIP histology, old age, and methotrexate (MTX) use [14]. In IPF, more advanced lung fibrosis is also known as a reliable risk factor of the acute exacerbation [15].

#### 2.1.2. Pharmacologic Treatment of RA-ILD

Unfortunately, there have been no randomized controlled trials (RCTs) for RA-ILD and the EULAR and ACR recommendations do not specify how to treat RA-ILD yet [16,17]. In general, physicians either adopt the treatment strategies against the corresponding pattern of idiopathic interstitial pneumonia or practice empirical therapies with immunosuppressives. However, due to the heterogeneous clinical behaviors of RA-ILD, the treatment goal is hard to define: treatment to achieve recovery versus treatment to stabilize or slow progression. In the case of acute exacerbation, most therapies are immunosuppressants with potent anti-inflammatory effects, assuming reversibility of the lesions, while in case of chronic progression, the antifibrotic approach will take the priority [2]. Therefore, physicians need to understand the dominant pathogenic mechanism underlying the clinical

behavior of individual cases of RA-ILD, which can vary at different time points even in the same patient.

#### Treatment with Conventional Agents

The data on the effect of individual immunosuppressive agents on RA-ILD are limited. Mycophenolate mofetil (MMF) was associated with a modest improvement in forced vital capacity (FVC) and the diffusing capacity of carbon monoxide (DLCO), reducing prednisone dosage in an observational study on 125 patients with RD-ILD including 18 RA-ILD cases [18]. In the case of the acute exacerbation of RA-ILD, high dose steroid therapy is delivered in combination with other immunosuppressive agents including MMF, azathioprine, or cyclophosphamide (CYC) [13,14].

#### Treatment with Targeted Agents

Although hope has been placed on biologics for RA-ILD treatment based on their excellent efficacy against articular inflammation, the evidence suggests that the safety profile of biologics including TNF inhibitors and others is uncertain for ILD treatment. Since biologics could, albeit rarely, exacerbate or cause the de novo development of ILD [19], experts recommend carefully assessing patients before and after treatment. While anti-TNF therapy has been suspected to aggravate ILD or induce its development [19], clear causal evidence is lacking [20,21]. Rituximab and abatacept have been more favorably suggested for patients with RA-ILD over anti-TNF therapy [22,23]. However, even these drugs have also been associated with lung toxicity and their data are scarce due to the less common use compared to TNF inhibitors [19].

Besides biologics, there are targeted agents with antifibrotic effects, recently introduced to treat IPF and RD-ILD (Table 1). These agents include nintedanib and pirfenidone.

Nintedanib is a competitive inhibitor of the nonreceptor and receptor tyrosine kinases that shows an antifibrotic effect [24]. The nonreceptor targets of nintedanib include Lck, Lyn, and Src and the receptor targets include platelet-derived growth factor receptors (PDGFR) α/β, fibroblast growth factor receptors (FGFR) 1/2/3, vascular endothelial growth factor receptors (VEGFR) 1/2/3, and fms-related tyrosine kinase 3 (FLT3). The targets of both categories play important roles in fibrosis [24]. In the phase II proof-of-concept TOMORROW study on patients with IPF diagnosed based on biopsy and/or high resolution computed tomography (HRCT) whose % predicted FVC of ≥50% and DLCO of 30–79% [25], 52-week nintedanib treatment (150 mg twice/day) was associated with less FVC decline, fewer acute exacerbations, and the preservation of health-related quality of life compared to placebo in patients with IPF. The antifibrotic effect of nintedanib on IPF was also replicated in phase 3 INPULSIS 1 and 2 trials, which showed significant reductions in the annual FVC decline in the treated group compared to the placebo (between-group difference of 125.3 and 93.7 mL/year in INPULSIS 1 and 2 trials, respectively, *p* < 0.001 in both results) [26]. Unlike treatment consistency on lung function preservation, however, the benefits on acute exacerbation were only observed in the INPULSIS 2. Nintedanib reduced the FVC decline among diverse subgroups defined by sex, age (<65 or ≥65 years), race (White or Asian), the baseline FVC% predicted (≤70% or >70%) and DLCO% (≤40% or >40%), composite physiologic index at baseline (≤45 or >45), the total baseline score on St George's Respiratory Questionnaire (≤40 or >40), smoking status (never-smoker or current/ex-smoker), corticosteroids for systemic use at baseline (yes or no), and bronchodilator use at baseline (yes or no) [27,28].

The recent INBUILD study, an RCT that assessed the efficacy and safety of nintedanib in patients with non-IPF ILD (entry criteria: % predicted FVC of ≥45% and DLCO of 30–79%), included a subgroup of patients with RD-ILDs, mostly RA-ILD [29]. In the RD subgroup, nintedanib reduced the rate of FVC decline compared to placebo at 52 weeks with a between-group difference of 104 mL/year (*p* < 0.001) [29]. Based on the INBUILD trial, nintedanib was approved by the U.S. Food and Drug Administration (FDA) for progressive-ILD, including RA-ILD in March 2020.



FVC = forced vital capacity, ILD = interstitial lung disease, IPF = idiopathic pulmonary fibrosis, RD = rheumatic disease, RA = rheumatoid arthritis, RCT = randomized controlled trial, SSc = systemic sclerosis.

> Pirfenidone, another antifibrotic drug approved for IPF, might show similar benefits for patients with the RA-UIP as the drug did for patients with IPF. Unlike nintedanib, the mechanism of the action of pirfenidone is not well known. Among the three phase 3 clinical trials in patients with IPF [30,31], the Japanese trial showed that pirfenidone reduced the decline in vital capacity at week 52 (between-group difference, 0.07 L/year) and improved progression-free survival compared to placebo [30]. In the remaining two CAPACITY trials (004 and 006) on patients with multinational backgrounds (entry criteria: % predicted FVC of 50–90% and DLCO of 35–90%), the primary endpoint (% predicted FVC change from baseline to week 72) was met in trial 004 (−8.0% vs. −12.4%, *p* < 0.05) but not in 006 (−9.0% vs. −9.6%, *p* = 0.501) [31]. In the ASCEND trial [32], there was a relative reduction of 47.9% in the proportion of IPF patients who had an absolute decline of ≥10% in the FVC or who died (16.5% vs. 31.8%, *p* < 0.001). There was also a relative

increase of 132.5% in the proportion of patients with no decline in FVC (22.7% vs. 9.7%, *p* < 0.001). In another multinational clinical trial on non-IPF ILD, the use of pirfenidone was associated with a lower risk of FVC decline greater than 10% (odds ratio = 0.44, 95% CI: 0.23–0.84) with a between-group difference in FVC of 95.3 mL/year at 24 weeks [33]. TRAIL1 is a 52-week multicenter randomized, double-blind, placebo-controlled, phase 2 study (ClinicalTrials.gov number, NCT02808871) of the safety, tolerability, and efficacy of pirfenidone in patients with RA-ILD, aiming to enroll as many as 270 subjects with RA-ILD and the results are awaited [34].

There is an overlap of adverse event profiles between nintedanib and pirfenidone [35]. In patients treated with nintedanib, gastrointestinal adverse events (e.g., diarrhea) were most common with mild-to-moderate intensity, accounting for the majority of the drug discontinuation [25,26,29]. Similarly, the most common adverse event of pirfenidone was gastrointestinal including nausea and vomiting, experienced by 40% and 18% of treated patients, respectively, out of 2059 person-years of exposure [36]. However, the two drugs have different pharmacokinetic profiles. Nintedanib is metabolized predominantly by ester cleavage and then glucuronidated to be excreted via the biliary system [35]. The use of nintedanib is associated with liver function test abnormalities in less than 5% of patients, and is not recommended for those with moderate-to-severe hepatic dysfunction. Regular liver function monitoring is required. Nintedanib has a low potential for drug–drug interactions, especially with drugs metabolized by cytochrome P450 enzymes. Pirfenidone is metabolized by various cytochrome P450 enzymes in the liver and predominantly excreted via the urine [35]. Similar rates of liver function test abnormalities were observed with pirfenidone as with nintedanib [25,26,31,32]. Pharmacokinetically, no drug–drug interaction was observed between nintedanib and pirfenidone. There are RCTs that examined the effect of nintedanib added on pirfenidone treatment [37,38]. More reports of nausea and vomiting were observed with nintedanib added on pirfenidone than used alone. However, most of them were mild-to-moderate as with the single drug treatment and the combination did not provide a new safety signal.

#### General Treatment Strategy for RA-ILD

Due to the heterogeneous progression patterns across individuals, it is hard to define an optimal treatment strategy in RA-ILD. Although the baseline extent of lung injury has been acknowledged as the most reliable risk factor of both progression and acute exacerbation, the cut-off extent to initiate treatment is not well defined. The international guidelines on RA management have yet to specify when to initiate treatment and what to be the first line agent [16,17]. Moreover, the treatment of acute exacerbation should be different from that of chronic progression [2]. However, it seems reasonable to use nintedanib when FVC loss is progressive enough to deteriorate symptoms (e.g., dyspnea or exercise capacity) or of when the current status is severe enough to qualify for the previous RCTs including the INBUILD trial (FVC of ≥45%, DLCO of 30–79%) [29].

#### *2.2. Systemic Sclerosis Associated Interstitial Lung Disease (SSc-ILD)*

The pulmonary manifestations of SSc can be both direct and indirect. The former category includes parenchymal (ILD) and vascular diseases (PAH) and the latter includes aspiration due to gastroesophageal reflux associated with esophageal sphincter fibrosis.

#### 2.2.1. Clinical Features of SSc-ILD

The key pathophysiology of SSc encompasses a triad of immune activation, vasculopathy, and fibrosis. Due to fibrosis and vasculopathy, the lung involvement of SSc most often manifests as ILD and/or PAH. ILD develops more frequently in the diffuse than limited cutaneous subset, particularly in the presence of antitopoisomerase I antibodies. SSc-specific anti-U3 RNP and anti-Th/To are also associated with SSc-ILD [39].

As many as 90% of SSc patients show ILD [3]. Among 3656 SSc patients from the European Scleroderma Trials and Research group, ILD was observed by plain chest radiography in approximately half of the patients with the diffuse cutaneous SSc and in one third of the patients with the limited cutaneous SSc [40]. Regarding severity, moderate (FVC of 50–75%) to severe (FVC < 50%) restrictive lung disease was found in 40% of SSc patients [41]. The pulmonary function of SSc-ILD patients mostly declines during the first several years after the onset of non-Raynaud's symptoms, but the progression rate of decline is best predicted by the baseline severity of the lung function or fibrosis [1]. A greater impairment at baseline predicts more progression in the future. Further, there are a significant proportion of patients not treated without progression [42], suggesting that the clinical course of SSc-ILD is variable [43].

The predominant histologic pattern of SSc-ILD is NSIP based on biopsy and/or HRCT [1,3,44]. However, the histologic pattern does not affect the clinical outcome of SSc-ILD [1,7,44]. Because fibrosis is the main histologic finding other than inflammation, even GGO represents fine reticulation and is rarely reversed but replaced later by overt fibrotic findings such as reticulation or honeycombing [45,46].

The mortality of SSc patients showed an overall three-fold increased standardized mortality ratio compared to the general population [47]: the survival of SSc patients was 74.9% at 5 years and 62.5% at 10 years from the diagnosis, with the mortality risk in the presence of ILD being 2.9-fold compared to in the absence of ILD [47].

#### 2.2.2. Pharmacologic Treatment of SSc-ILD

The lung function of SSc patients with FVC of ≥80% at baseline rarely declines [43]. Thus, symptomatic patients are the primary target of treatments, particularly those whose ILD is moderate-to-severe or shows progression. The recent European consensus statements presented an agreement on screening of ILD for all SSc patients at baseline particularly in the presence of risk factors (diffuse cutaneous subset, antitopoisomerase antibody, low DLCO) preferably by HRCT but also with pulmonary function test (PFT) and clinical assessment as supporting tools [48]. The statements recommend treatment for all severe cases defined by PFT or HRCT, and for those who progress based on clinical assessment, HRCT, and/or PFT. However, the statements did not specify the threshold to define severe ILD or when to initiate or escalate treatment. Those who have symptoms that are progressively deteriorating (newly developed functional class II and more), whose FVC and/or DLCO decline is large enough (≥10% and ≥15%, respectively) [49,50], or whose FVC or DLCO is severe enough to justify treatment-related adverse events (refer to the inclusion criteria for Scleroderma Lung Study or SENSCIS trial below) [51–53] would be reasonable candidates to initiate or escalate treatment.

#### Treatment with Conventional Agents

Until recently, the treatment SSc-ILD has relied on immunosuppressants such as CYC and MMF based on the RCT data [51,52,54]. Since GGO of SSc-ILD often indicates fine fibrosis rather than inflammation [45,46], these immunosuppressive treatments stabilize rather than improve such lesions.

In the Scleroderma Lung Study (SLS) I on 158 SSc patients with symptomatic ILD (active alveolitis on imaging and FVC ranging between 45–85%) [51], the mean absolute difference of FVC% predicted at 12 months (2.53%, *p* < 0.03) was significantly in favor of oral CYC over placebo. When the patients were followed for another one year after stopping the medication, the effect was maintained or even greater at six months after treatment termination but disappeared by one year [52]. According to the subgroup analysis, a greater response to CYC was linked to more severe lung disease at baseline [52]. In SLS II, the efficacy of the two-year treatment with MMF was comparable to that of the one-year treatment with oral CYC at 24 months (mean FVC improvement from baseline of 2.17% in the MMF group versus 2.86% in the CYC group) [54]. The adverse events were less with the former drug. The 2017 updated EULAR recommendations still suggest CYC preferentially over MMF [55], but MMF is a viable first line treatment for patients who are susceptible to the toxicity of CYC.

Treatment with Targeted Agents

#### **Antifibrotics**

The SENSCIS trial was performed on 576 patients with SSc-ILD (mean age 54.6 years, 74–76% female) of at least 10% extent of the lung [53]. The entry criteria of ILD required % predicted FVC ≥ 40% and % predicted DLCO ranging in 30–90%. The % predicted FVC of included patients was 72% and DLCO 53% at baseline. Of the patients included in the study, 48.4% were taking background MMF. After 52 weeks of treatment, nintedanib (150mg twice daily) was associated with a lower annual rate of FVC decline compared to the placebo treatment (−52.4 mL/year versus −93.3 mL/year). The curves for the FVC change from the baseline separated by 12 weeks and continued to diverge until the end of study. The effect of nintedanib was additive when combined with MMF: the annual rates of change in FVC among the patients receiving mycophenolate were −40.2 mL/year in the nintedanib group and −66.5 mL/year in the placebo group, and the corresponding rates among the patients who were not receiving mycophenolate were −63.9 mL/year and −119.3 mL/year. Of note, the SENSCIS trial did not show any treatment effect on the skin fibrosis. According to the subgroup analysis, the effect of nintedanib was consistent regardless of the use of MMF, suggesting that there is no synergy but additive effect between nintedanib and MMF [56]. The gastrointestinal adverse (e.g., diarrhea) events were more common in the SENSCIS trial compared to the INPULSIS trials probably due to the gastrointestinal involvement by SSc itself [26,53]. The post hoc analysis on the SENSCIS trial showed that the effect of nintedanib was significant to prevent % predicted FVC decline of >10% but not of >5% [57]. An open label extension study is ongoing to provide long term data on nintedanib use in patients with SSc-ILD (ClinicalTrials.gov number, NCT03313180). Based on the SENSCIS trial [53], nintedanib was approved to treat SSc-ILD by the FDA in 2019 and by the European Medicines Agency (EMA) in 2020.

The acceptable safety and tolerability of the antifibrotic agent pirfenidone were reported among patients with SSc-ILD [58], but its efficacy needs to be assessed in further clinical trials. There are two ongoing RCTs on SSc-ILD. The efficacy and safety of pirfenidone is now being assessed in a phase 3 study on 144 SSc-ILD patients (ClinicalTrials.gov number, NCT03856853). In addition, the SLS III study (phase 2) will randomize 150 patients with SSc-ILD to pirfenidone versus placebo, with background MMF, over 18 months of follow-up (ClinicalTrials.gov number, NCT03221257), to assess the effect of combination treatment of pirfenidone and MMF versus MMF alone.

#### **Biologics**

Promising benefits in SSc-ILD from biologics have recently been suggested. In the phase II faSScinate trial that assessed the skin fibrosis as the primary outcome and lung function as the secondary outcome, 48-week treatment with tocilizumab (TCZ) provided an encouraging numerical improvement in the modified Rodnan skin scores and evidence of less decline in lung function [59]. During another 48-week open-label extension period of TCZ treatment, skin score improvement and FVC stabilization were observed in the placebo-treated patients who transitioned to tocilizumab [60]. In the faSScinate study, the patient population showing benefits from TCZ had a shorter duration of disease (mean disease duration, 1.6 years), increased serum acute phase reactants, progressive skin disease, and low normal mean FVC levels (mean predicted FVC%, 81%) at the study baseline. These findings may suggest that more inflammatory process is involved during the early phase of the disease, where immunosuppressive or anti-inflammatory treatment can provide more benefits. In the subsequent focuSSced study of the multinational background on 210 patients with early diffuse cutaneous SSc (duration ≤ 5 years) [61], TCZ and placebo were compared regarding skin fibrosis (primary outcome) and lung function (secondary outcome). After 48 weeks, the skin fibrosis endpoint was not met but the FVC loss was more in favor of the TCZ group than the placebo (between-group difference, 241 mL, *p* = 0.002). The results of the focuSSced study on FVC should be interpreted with caution because the SSc-ILD was not the primary outcome and only 65% of enrolled patients had SSc-ILD at baseline. Controlled studies are awaited to assess whether TCZ has an

effect in patients with severe established SSc lung disease. In addition, the paradoxical response of aggravating ILD has been also reported from TCZ treatment [19], indicating that meticulous monitoring of ILD is mandatory before and after treatment despite the overall benefits of TCZ in patients with ILD.

Rituximab has also been used to treat SSc-ILD. In an open-label RCT on 60 patients with SSc-ILD, the RTX treatment showed a significantly better patient response than the CYC treatment in terms of improvement in FVC and skin fibrosis [62]. However, a propensity score matched observational study showed that the proportion of patients with FVC decline of ≥10% was similar regardless of RTX treatment. Nevertheless, patients treated with RTX were more likely to stop or decrease steroid use [63]. According to the experts, rituximab was considered as one of the first induction agents for SSc-ILD in addition to MMF or CYC [64]. An ongoing trial (RECITAL) is comparing rituximab and CYC for RD-ILD including SSc-ILD [65].

#### General Treatment Strategy for SSc-ILD

For those who are asymptomatic with minimal extent of ILD (e.g., FVC > 80%), watchful monitoring without treatment is justified [42,43]. For those who need treatment, the EULAR guidelines endorse CYC over MMF as the first line treatment for SSc-ILD [55] while the European consensus statements consider CYC and MMF, and nintedanib as equivalent options for treatment initiation or escalation [48]. The subgroup analysis of SENSCIS showed a comparable FVC change between placebo with baseline MMF group (−66.5 mL/year) versus nintedanib without baseline MMF group (−63.5 mL/year) [53]. Thus, whether to use antifibrotics as a first line treatment needs further data (e.g., head-tohead comparison between immunosuppressants versus antifibrotics) and discussion. A reasonable approach would be to switch to or add nintedanib when refractory to CYC or MMF. Although some evidence emerged suggestive of the benefits with TCZ and RTX on lung function, such evidence is only exploratory and needs to be confirmed in an RCT that examines lung function as the primary end point.

#### *2.3. Myositis Associated Interstitial Lung Disease (Myositis-ILD)*

As in SSc, the pulmonary manifestations of myositis can be categorized as direct and indirect. The direct involvements are mostly ILD. The indirect involvements consist of aspiration pneumonia due to pharyngeal muscle weakness and hypoventilation due to respiratory muscle weakness [66].

#### 2.3.1. Clinical Course of Myositis-ILD

A majority of myositis-associated lung involvement manifests as ILD. Unlike in SSc, isolated PAH is rare in myositis [67]. ILD manifests early in the course of myositis, often found at the time of diagnosis [4,68,69]. Being prevalent in up to two thirds of myositis patients [4,69], more than 90% of myositis patients develop ILD in the presence of antiaminoacyl tRNA synthetase (ARS) antibody [70]. The clinical course of myositis-ILD is diverse but distinguished from SSc-ILD in that as many as 20% of myositis-ILD cases manifest as a rapidly progressive form that challenges successful treatment [69]. This contrasts with SSc-ILD, where fibrosis is the hallmark of the disease, causing more chronic progression. Such difference indicates that inflammatory process is a dominant process in myositis-ILD, generally in proportion to symptom deterioration rate, and culminates in the rapidly progressive form in which respiratory failure can occur even within days [71,72].

While NSIP is most common for myositis-ILD in general [1], DAD is more common in case of the rapidly progressive type [70]. The rapidly progressive ILD tends to cluster in patients with either classic DM or amyopathic dermatomyositis than polymyositis, particularly in the presence of antimelanoma differentiation-associated gene 5 (anti-MDA5) [73]. Myositis-ILD of the rapidly progressive type is refractory to treatment leading to a high mortality [72].

The treatment response of myositis-ILD follows the underlying histological pattern [69,74,75]. OP is well treated by steroids alone, whereas DAD and UIP respond poorly even to combination therapy [69,74]. The treatment response of NSIP depends upon the degree of inflammation compared to fibrosis [75]. Biomarker studies showed that high serum levels of ferritin, IL-18, or soluble CD206 (macrophage mannose receptor) were associated with a poor treatment response among patients with anti-MDA5-positive ILD [76,77].

#### 2.3.2. Pharmacologic Treatment of Myositis-ILD

Due to the rarity of the disease and the heterogeneous clinical subsets within the disease, the treatment of myositis-ILD most often relies upon empirical therapy consisting of steroids combined with conventional immunosuppressives such as CYC, cyclosporine, or MMF without sufficient data supported by RCTs [1]. Even more limited are the data for targeted treatments. A recent Japanese study showed that the initial combination treatment with high dose steroid, tacrolimus, and CYC with or without plasmapheresis was associated with a better response in anti-MDA5-positive ILD patients compared to step-up therapy from initial high-dose steroids [78]. This finding also put emphasis on the early aggressive treatment for anti-MDA5-positive, rapidly progressive ILD. Moreover, a Spanish expert group agreed that anti-MDA5-positive ILD should be similarly treated [79]. Of note, despite the disease-specific anti-MDA5 autoantibody, successful treatment with rituximab has only been anecdotally reported [80].

#### *2.4. Future Treatment Strategies for Rheumatic Disease Associated Interstitial Lung Disease*

Nintedanib and pirfenidone slow but do not halt the progression of IPF. Thus, interest in the combination of the two drugs is growing, but more data are needed on the safety and efficacy of combination therapy. There are a few studies that showed relatively acceptable tolerability and safety of the two drugs combined for a short-term period (12~24 weeks) [37,38]: the combination was completed in two thirds of patients despite a higher adverse event rate compared to the single drug treatment. However, there have yet to be large randomized controlled trials assessing whether combination therapy has increased efficacy compared to treatment with a single antifibrotic drug.

Several clinical trials evaluating the efficacy of the new drugs on pulmonary outcomes are underway. The ISABELA 1 and 2 trials are now ongoing to examine the effect of autotaxin inhibitor GB-0998 among patients with IPF (ClinicalTrials.gov number, NCT03711162; NCT03733444) [81]. A phase 2 RCT of bortezomib plus MMF versus MMF alone on patients with SSc-ILD is currently ongoing (ClinicalTrials.gov number, NCT02370693). Romilkimab, a monoclonal antibody targeting IL-4 and IL-13, showed efficacy after 24 weeks on skin fibrosis associated with diffuse SSc under background immunosuppressive agents in a phase 2 pilot study [82]. However, it failed to show benefits in the of IPF after 52 weeks [83]. Guselkumab (monoclonal antibody that blocks IL-23) is under investigation in SSc patients with skin fibrosis as the primary outcome of and lung function as one of the secondary outcomes (ClinicalTrials.gov number, NCT04683029).

#### **3. Pulmonary Arterial Hypertension (PAH)**

Abnormal proliferation, vasoconstriction, and thrombosis of the pulmonary vasculature are the main pathogenic mechanisms of PAH. Right heart catheterization (RHC) is the gold standard to diagnose PAH using the criteria of a mean pulmonary arterial pressure of ≥25 mmHg and a pulmonary capillary wedge pressure of ≤15 mmHg [84].

#### *3.1. Systemic Sclerosis Associated Pulmonary Arterial Hypertension (SSc-PAH)*

Not only isolated PAH but also PH secondary to either left heart dysfunction or ILD progression can occur in SSc, either separately or in combination. The SSc-PAH shows a unique phenotype of PAH with a worst prognosis compared to idiopathic or non-SSc RD-PAH [85,86].

#### 3.1.1. Clinical Features of SSc-PAH

The mortality risk of SSc patients increases by 3-fold in the presence of PAH, showing PAH is a deadly complication of SSc patients [87]. Therefore, patients should be screened for PAH based on symptoms, echocardiography, and DLCO patterns and undergo RHC if identified as high-risk [88,89]. With RHC, SSc-PAH was found in approximately 10% of these high-risk candidates. The limited rather than the diffuse cutaneous subset shows a higher prevalence of SSc-PAH, which was observed in up to half of the patients with CREST syndrome [90]. It was reported that 50% of SSc-PAH occurs within five years from the first non-Raynaud phenomenon symptom with the mean interval from SSc diagnosis to PAH occurrence of 6.3 years [91].

Like those with idiopathic PAH, patients with SSc-PAH are either clinically silent or show only nonspecific symptoms until their disease is advanced. Therefore, active surveillance is the key to early intervention, which indeed led to better survival compared to passive identification [92]. In SSc patients, male sex, old age, overt vasculopathy such as telangiectasia and digital ulcers, anticentromere or anti-U3 RNP antibodies, and the limited cutaneous subset (e.g., CREST syndrome) were identified as risk factors for PAH [93]. However, both the sensitivity and the specificity of these risk factors are limited. Other findings suggestive of PAH include elevated levels of the N-terminal probrain natriuretic peptide (NT-proBNP) or disproportionately low DLCO [94,95]. In particular, echocardiographic measures are one of the important screening tools to identify candidates for RHC. For example, a tricuspid regurgitation (TR) velocity of ≥2.5 m/s is considered one of the findings highly suggestive of PAH [96]. However, the sensitivity at this TR velocity threshold is limited, missing 20% of the mild PAH patients [96,97]. DETECT is a multidimensional algorithm to identify SSc patients at a high risk of PAH, used as the active surveillance strategy to avoid the delayed diagnosis of PAH [1,96]. Among 57 DETECT enrollees with SSc-PAH, 44% progressed during a median follow-up of 12.6 months [98]. The factors associated with progression were male gender, disproportionately low DLCO, and poor functional capacity. A more simplified and practical algorithm proposed in 2013 to initially screen SSc patients uses PFT, echocardiography, and NT-proBNP for referral for RHC [99] (Table 2).

In a recent prospective observational study, 93 SSc-ILD patients were screened and underwent RHC following the above 2013 algorithm. Of these, 31.2% were found to have RHC-proven PAH, and the survival rate was 91% at three years [100]. This improved survival is striking compared to the exceptionally grave prognosis (survival of 39% at three years) previously reported for SSc patients with coexisting ILD and PAH [101]. Based on this finding, the poor prognosis of SSc-PAH patients compared to patients with idiopathic or PAH of other rheumatic diseases could be due to delayed diagnosis and treatment [85,86], and early treatment is paramount for better survival.

#### 3.1.2. Pharmacological Treatment of SSc-PAH Treatment with Conventional Agents

Immunosuppression against the hyperimmune state of SSc is never sufficient to stop SSc-PAH progression due to fibrotic and proliferative pathways driven by nonimmune cells. An acute vasoreactive response is observed in 10% of the patients with RD-PAH, but the response to calcium channel blockers beyond 3–4 months is preserved in less than 1% of these patients [102].


**Table 2.** Screening algorithm for SSc-PAH proposed in 2013 a.

<sup>a</sup> Cited and modified from "Recommendations for screening and detection of connective tissue disease-associated pulmonary arterial hypertension" by Khanna D, et al. Arthritis Rheum 2013; 65: 3194–3201 [99]. <sup>b</sup> Symptoms: dyspnea upon rest or exercise, fatigue, presyncope/syncope, chest pain, palpitations, dizziness, lightheadedness. Signs: loud pulmonic sound, peripheral edema. <sup>c</sup> Without overt systolic dysfunction, greater than grade I diastolic dysfunction, greater than mild mitral or aortic valve disease, or evidence of PAH in echocardiography. DLCO = diffusion capacity of carbon monoxide; FVC = forced vital capacity; NT-proBNP = N-terminal probrain natriuretic peptide; PAH = pulmonary arterial hypertension; PFT = pulmonary function test; RA = right atrium; RV = right ventricle; SSc = systemic sclerosis; TTE = transthoracic echocardiography.

#### Treatment with Targeted Agents and Risk Stratification

Although the autoimmune process may be the fundamental mechanism orchestrating the initiation and progression of RD-PAH, its downstream effects involve nonimmune components as well, leading to pulmonary vascular remodeling. In particular, the treatment effect of anti-inflammatory agents is limited in lung diseases associated with SSc, indicating that noninflammatory pathways play a dominant role. In PAH, endothelial dysfunction due to abnormally regulated vasoactive and/or proliferative mediators induces vascular constriction and remodeling. Currently, three pathways important in endothelial function are targeted by PAH treatments, endothelin-1, nitric oxide (NO), and prostacyclin pathways [103]. Endothelin receptor antagonist (ERA), phosphodiesterase 5 inhibitor (PDE5i), and soluble guanylate cyclase (sGC) stimulator, which potentiate the effect of nitric oxide (NO), as well as prostacyclin analogs (PCA), are commercially available (Table 3).


**Table 3.** Efficacy of drug monotherapy for group 1 PAH according to WHO functional class a.

<sup>a</sup> Cited and modified from "2015 European Society of Cardiology (ESC)/European Respiratory Society (ERS) guidelines for the diagnosis and treatment of pulmonary hypertension" by Galiè N et al. Eur Heart J. 2016; 37:67–119 [97], <sup>b</sup> Class of recommendation, <sup>c</sup> level of evidence, ERA = endothelin receptor antagonist; PDE5i = phosphodiesterase 5 inhibitor; sGC = soluble guanylate cyclase; PAH = pulmonary arterial hypertension; PCA = prostacyclin analog or receptor antagonist; WHO = World Health Organization; FC = functional class.

> According to the 2017 EULAR recommendations [55], ERA (ambrisentan, bosentan, and macitentan), PDE5i (sildenafil and tadalafil), and an sGC stimulator (riociguats) are considered the first-line options for treating SSc-PAH based on high-quality RCTs that showed improvement in exercise capacity, the time to clinical worsening (defined as a composite of death, hospitalization, and disease progression), and/or PAH-related hemodynamics in heterogeneous patients with PAH including RD-PAH. These studies were not powered to show the independent efficacy of the SSc-PAH subgroup, but the EULAR recommendations are based on the extrapolation of the high-quality RCT results. In the subgroup analyses, the direction of the results was not different between patients with idiopathic PAH and RD-PAH. In the idiopathic PAH, significant improvements in exercise capacity and reductions in mortality (pooled effect, 44% reduction; *p* < 0.041) by these treatments were demonstrated [104].

> A high-quality RCT showed that continuous intravenous epoprostenol (a prostacyclin analog) administration improved exercise capacity, functional class, and hemodynamics in patients with SSc-PAH [105]. The administration of continuous intravenous epoprostenol requires an indwelling central venous catheter and abrupt cessation of the drug may cause PAH rebound that can be fatal. Thus, the EULAR recommendations suggest that this treatment should be considered for severe SSc-PAH of class III and IV [55]. In addition to epoprostenol, other PCAs (iloprost and treprostinil) showed similar results in high-quality RCTs on heterogeneous patients with PAH including RD-PAH and are also approved for the treatment of PAH including RD-PAH [55]. Although not commented in EULAR recommendations, selexipag, an oral selective prostacyclin IP receptor agonist, has shown efficacy in the phase 3 GRIPHON trial of reducing the risk of a primary composite endpoint of morbidity and mortality by 40% and 41% in PAH and RD-PAH (majority, SSc-PAH),

respectively [106,107]. Selexipag is the only oral prostacyclin receptor agonist that showed a reduction of morbidity and mortality in a RCT.

In addition to the application of individual drugs that target each corresponding pathway, the advancements in recent treatment strategies include the risk assessment of PAH patients and linking the baseline severity of PAH to the subsequent treatment intensity or escalation [97,103]. According to the 2015 European Society of Cardiology (ESC)/European Respiratory Society (ERS) pulmonary hypertension guidelines [97], a multiparametric approach should be considered to stratify patients into low-, intermediateor high-risk groups for 1-year mortality using clinical (clinical signs of right heart failure, the progression of symptoms, and syncope), functional class (WHO or NYHA class), exercise (6-min walking distance and cardiopulmonary exercise testing), biochemical (NT-proBNP), and echocardiographic/hemodynamic parameters (Table 4).

**Table 4.** 2015 European Society of Cardiology (ESC)/European Respiratory Society (ERS) guidelines for the risk assessment of patients with PAH a.


<sup>a</sup> Cited and modified from "2015 European Society of Cardiology (ESC)/European Respiratory Society (ERS) guidelines for the diagnosis and treatment of pulmonary hypertension" Galiè N, et al. Eur Heart J. 2016;37:67–119 [97]. <sup>b</sup> Occasional syncope during brisk or heavy exercise, or occasional orthostatic syncope in an otherwise stable patient. <sup>c</sup> Repeated episodes of syncope even with little or regular physical activity. 6MWD = 6-min walking distance; BNP = brain natriuretic peptide; CI = cardiac index; CMR = cardiac magnetic resonance; NT-proBNP = N-terminal probrain natriuretic peptide; PAH = pulmonary arterial hypertension; RA = right atrium; RAP = right atrial pressure; SvO2 = mixed venous oxygen saturation; VE/VCO2 = ventilatory equivalents for carbon dioxide; VO2 = oxygen consumption; WHO = World Health Organization.

> Another important feature of the 2015 ESC/ERS guidelines for PAH is the endorsement of pre-emptive combination therapy targeting different endothelial pathways, particularly for high-risk patients [97]. The most compelling evidence for such strategy came from the recent AMBITION trial where a 50% reduction in the composite endpoint of clinical failure events at 24 weeks was seen from treatment with the upfront combination of ambrisentan and tadalafil compared to either drug alone in patients with PAH and RD-PAH [108]. Similarly, the add-on effect of a study drug (either macitentan, selexipag, or riociguat) was observed in other RCTs when combined with a background PAH treatment using a different class [106,109,110], also endorsing sequential combination therapy. Similar evidence for upfront or initial combination or sequential combination therapy was also observed in patients with RD-PAH including SSc-PAH [107,111,112]. The 2015 ESC/ERS

guidelines also depict a treatment algorithm that recommends treatment escalation (double to triple, maximal medical therapy including intravenous PCA) when the reassessment in 3–6 months shows an insufficient response such as residual intermediate- or high-risk status [97]. However, the evidence for the benefits of combination therapy remains to be replicated, specifically for SSc-PAH in future studies.

#### **4. Pulmonary Manifestations of Other Rheumatic Diseases**

Other than RA, SSc, and myositis, Sjogren's syndrome (SjS) and mixed connective tissue disease (MCTD) also show a substantial prevalence of pulmonary manifestations including ILD and PAH. Although the data on the epidemiology are growing in SjS and MCTD, RCT data regarding the treatments are very limited. However, we may be able to apply similar treatment strategies in SjS and MCTD as in other RDs described in this review, based on the dominant underlying mechanism of fibrosis versus inflammation or chronic progression versus acute exacerbation.

#### *4.1. Sjogren's Syndrome Associated Pulmonary Involvements*

SjS is a systemic autoimmune disease primarily affecting exocrine glands, which ultimately leads to the destruction of the given tissue [113]. The autoimmunity of SjS activates both immune cells and glandular epithelial cells showing a histological lesion of focal lymphocytic infiltrates, enriched in CD4+ T-cells, around the salivary and lachrymal ducts and even in lung tissues [114]. Of note, the focus score of salivary glands has been shown to correlate with the increased prevalence of airway disease and ILD in SjS [115]. These findings may suggest that the glandular and extraglandular lesions of SjS share similar pathogenic pathways involving autoimmunity to epithelial cells.

The prevalence of pulmonary involvement in patients with SjS has been estimated to range from 10% to 20% [116]. Airway disease and ILD are the predominant forms of lung involvement among the pulmonary manifestations of SjS, but lymphoproliferative disorders as well as cystic lesions are also observed [116]. The most common histologic type of ILD associated with SjS was found to be NSIP followed by UIP and LIP [117,118], with NSIP present in up to 45% of biopsied cases [117] and mostly of the fibrotic type [118]. Unlike RA-UIP, SjS-UIP tends to have a better response to immunotherapy as compared to IPF [119]. LIP constitutes 15% of SjS-ILD, of which clinical course is variable from complete resolution without treatment to progression and possible death or transformation to lymphoma [120].

#### *4.2. Mixed Connective Tissue Disease Associated Pulmonary Involvements*

MCTD is characterized by mixed features of two or more RDs including but not limited to SSc, myositis or systemic lupus erythematosus, with disease specific high titer anti-U1 RNP antibodies [121]. The common pulmonary manifestations of MCTD are ILD and PAH. A Norwegian nationwide cross-sectional study showed that 35% of MCTD patients had lung fibrosis in HRCT after a mean disease duration 9 years [122]. A higher rate of ILD was found (up to two thirds of the patients) in the hospital-based study [123]. In the Norwegian study, 19% of MCTD associated ILD (MCTD-ILD) was severe in extent involving >50% of lung parenchyma [122]. However, the retrospective cohort study on unselected 53 patients with a mean disease duration of 9 years in Brazil showed that 51% of the patients had ILD at baseline with a mean FVC that remained stable at around 77% over 10 years of follow-up and a mean DLCO that declined from 84% to 71% [124]. Such discrepant findings leave a debate on the natural course of MCTD-ILD, and further cohort studies at a larger scale are needed. Together with ILD, PAH is a major prognostic factor of MCTD [125,126]. The Norwegian population-based prevalence of pulmonary hypertension was reported to be less than 5% over 5.6 years [125], but PAH has been recognized as the leading cause of death in patients with MCTD, explaining 41% of all deaths [126].

#### **5. Conclusions**

ILD and PAH are the two most important prognostic factors in patients with RD and are associated with significant morbidity and mortality. Early diagnosis and early treatment are the key steps to the successful management of these two conditions, improving survival. In addition to conventional immunosuppressants, new targeted treatments are under investigation and some have already been incorporated into the international recommendations for treating RD-related ILD or PAH. However, more data are needed to ensure that the efficacy and safety of drugs used to treat RD as a whole are demonstrated for a specific RD.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**

CYC = cyclophosphamide, DAD = diffuse alveolar damage, DLCO = diffusion capacity of carbon monoxide, FVC = forced vital capacity, GGO = ground glass opacity, HRCT = high resolution computed tomography, ILD = interstitial lung disease, IPF = idiopathic pulmonary fibrosis, LIP = lymphoid interstitial pneumonia, MCTD = mixed connective tissue disease, MMF = mycophenolate, NSIP = nonspecific interstitial pneumonia, OP = organizing pneumonia, PAH = pulmonary arterial hypertension, RA = rheumatoid arthritis, RCT = randomized controlled trial, RD = rheumatic disease, SjS = Sjogren's syndrome, SSc = systemic sclerosis, UIP = usual interstitial pneumonia

#### **References**


Society (ERS): Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC), International Society for Heart and Lung Transplantation (ISHLT). *Eur. Heart J.* **2016**, *37*, 67–119.


### *Review* **The Relationship between Pulmonary Damage and Peripheral Vascular Manifestations in Systemic Sclerosis Patients**

**Barbara Ruaro 1,\*, Marco Confalonieri 1, Francesco Salton 1, Barbara Wade 2, Elisa Baratella 3, Pietro Geri 1, Paola Confalonieri 1, Metka Kodric 1, Marco Biolo <sup>1</sup> and Cosimo Bruni <sup>4</sup>**


**Abstract:** Systemic sclerosis (SSc) is an autoimmune disease, characterized by the presence of generalized vasculopathy and tissue fibrosis. Collagen vascular disorder in SSc is due to fibroblast and endothelial cell dysfunctions. This leads to collagen overproduction, vascular impairment and immune system abnormalities and, in the last stage, multi-organ damage. Thus, to avoid organ damage, which has a poor prognosis, all patients should be carefully evaluated and followed. This is particularly important in the initial disease phase, so as to facilitate early identification of any organ involvement and to allow for appropriate therapy. Pulmonary disease in SSc mainly involves interstitial lung disease (ILD) and pulmonary arterial hypertension (PAH). High-resolution computed tomography (HRCT) and pulmonary function tests (PFT) have been proposed to monitor parenchymal damage. Although transthoracic echocardiography is the most commonly used screening tool for PAH in SSc patients, definitive diagnosis necessitates confirmation by right heart catheterization (RHC). Moreover, some studies have demonstrated that nailfold videocapillaroscopy (NVC) provides an accurate evaluation of the microvascular damage in SSc and is able to predict internal organ involvement, such as lung impairment. This review provides an overview of the correlation between lung damage and microvascular involvement in SSc patients.

**Keywords:** systemic sclerosis; pulmonary involvement; microvascular involvement; pulmonary arterial hypertension; interstitial lung disease; nailfold capillaroscopy

#### **1. Introduction**

Systemic sclerosis (SSc), a heterogeneous disease, is characterized by immune dysfunction, often leading to organ damage due to inflammation, endothelial dysfunction and fibrosis [1–4]. SSc involves microcirculation structural and functional alterations [5–9]. The main cause of death in SSc patients is not only collagen overproduction but also the effects collagen overproduction has on the pulmonary system. This includes fibrosis or pulmonary artery hypertension (PAH) [10–15]. Recent guidelines have recommend screening with high resolution computed tomography (HRCT) to diagnose interstitial lung diseases (ILD) in SSc patients at the baseline visit and once the diagnosis of ILD has been established [16,17], whilst a combination of HRCT and pulmonary function tests is recommended to quantify the extent and severity of ILD [16–19]. Screening for PAH in SSc is transthoracic echocardiography, which has a sensitivity of 90%, even if definitive diagnosis is to be confirmed by right heart catheterization (RHC) [20–25]. Although nailfold videocapillaroscopy (NVC) is

**Citation:** Ruaro, B.; Confalonieri, M.; Salton, F.; Wade, B.; Baratella, E.; Geri, P.; Confalonieri, P.; Kodric, M.; Biolo, M.; Bruni, C. The Relationship between Pulmonary Damage and Peripheral Vascular Manifestations in Systemic Sclerosis Patients. *Pharmaceuticals* **2021**, *14*, 403. https:// doi.org/10.3390/ph14050403

Academic Editor: Antoni Camins Espuny

Received: 21 March 2021 Accepted: 19 April 2021 Published: 23 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

the validated method for assessing peripheral vascular damage [26–29], several studies have demonstrated that NVC is capable of predicting internal organ involvement [27–35].

This review aims at providing updated information on the link between pulmonary damage, i.e., ILD and PAH, and peripheral vascular manifestations, evaluated by NVC, in SSc patients.

#### **2. Pulmonary Manifestations**

Pulmonary disease in SSc includes interstitial lung disease (ILD) and pulmonary arterial hypertension (PAH) [36–38]. All SSc patients should be screened to detect any ILD and PAH development, at diagnosis and periodically thereafter. Indeed, although there has been no statistically significant change in the SSc mortality rate over the past 40 years, the proportion of deaths due to ILD and PAH has increased [36–41]. ILD and PAH are the two main causes of death in SSc patients and account for 33% and 28% of deaths, respectively [36–41]. Although ILD is reported to be more common in diffuse cutaneous SSc (dcSSc) whilst PAH is reportedly more common in limited cutaneous SSc (lcSSc), both pulmonary manifestations have been described in each of the disease subsets. Patients with rarer phenotypes associated with antiTh/To and anti U3RNP antibodies may have PAH and ILD concomitantly. Pulmonary disease may even occur in SSc with no skin involvement, i.e., scleroderma sine scleroderma [36–41].

Although the clinical course varies from mild and asymptomatic to severely debilitating, most patients have some degree of pulmonary fibrosis [39,40]. The most common early symptoms related to SSc pulmonary manifestations are exertional dyspnea and dry cough and, in most cases, are non-specific findings [42–46]. Should this be the case, then a differential diagnosis should be made to investigate/exclude SSc-ILD, PAH, deconditioning, chronic anemia and/or left heart involvement with a reduced or preserved ejection fraction [42–46]. Considering the frequency of lung involvement in SSc and its impact on prognosis, early recognition of lung involvement and prompt appropriate treatment is a must [36–41]. Although there are inherent challenges in the management of both PAH and ILD, with the early diagnosis, treatment may have a higher chance of efficacy for each of these lung complications [2,10].

The etiologic or enhancing factors of pulmonary involvement in SSc patients are still a question of debate. A previous study implicated genetic factors, i.e., HLA class II (1–3). Others have implicated immunologic factors for which certain autoantibodies such as anti–topoisomerase I (anti–topo I) may be markers [1–3]. The few studies that have addressed the impact of race or ethnicity on lung involvement in early SSc suggest a worse prognosis for nonwhite groups (e.g., African Americans, the Japanese population and Choctaw Indians). However, ethnicity is not only defined by racial or genetic factors but also by sociodemographic and cultural factors [1–4].

#### **3. Interstitial Lung Disease**

ILD complicates diffuse cutaneous SSc (dcSSc) in 53% of cases but may also be associated in 35% of cases with limited cutaneous SSc (lcSSc), as reported by the European Scleroderma Trials and Research group (EUSTAR) [39,42]. Furthermore, several autopsy studies reported that parenchymal involvement, in the form of ILD, was present in up to 90% of SSc patients. Risk factors for ILD development include African American ethnicity, skin score, serum creatinine and creatine phosphokinase levels, hypothyroidism and cardiac involvement [1,2,42–44]. Genetic factors, specific serological findings and anti-topoisomerase and anti-endothelial cell antibodies can predict the presence of lung involvement [1,2,42–44]. It is also reported that the patients with dcSSc have a higher incidence of interstitial disease [1–3]. Predictors of severe restrictive lung disease (defined by a forced vital capacity (FVC) of 50% predicted) include African American ethnicity, male gender, the degree of physiological abnormalities at diagnosis (FVC and diffusing lung capacity for carbon monoxide (DLCO)) and a younger age [1,2,46–48].

Unfortunately, there are limited treatment options for this manifestation. This is due to the paucity of high-quality, randomized, controlled trials that specifically target SSc-ILD. Moreover, historically, studies have favored cyclophosphamide (CYC) for SSc-ILD treatment, as also suggested in the most recent European League against Rheumatism (EULAR) recommendations [2,49]. Supportive data have shown that nintedanib, a multi-tyrosine kinase inhibitor, and tocilizumab (TCZ) significantly inhibit the progressive functional decline [2,49]. Current innovative proposals have also recently been made on the basis of clinical and preclinical evidence for rituximab (RTX) and pirfenidone (PIRF), as well as hematopoietic stem cell and lung transplantation [2,49]. However, the safety and efficacy of emerging experimental therapies for SSc-ILD do require further investigation.

Other findings were that high-resolution computed tomography (HRCT) evidenced interstitial abnormalities in as many as 90% of patients, and 40–75% had changes in pulmonary function tests (PFT) [46–52].

#### **4. Imaging**

SSc-ILD is diagnosed by HRCT, which is a simple non-invasive, sensitive investigation, able to detect parenchymal lung disease [53–60]. However, despite its high sensitivity, HRCT may be normal in some patients with pulmonary function test abnormalities or abnormal chest auscultation (i.e., crackles) [46–48,58]. The absence of lung involvement in HRCT at the time of disease presentation may lower the long-term risk of developing SSc-ILD, as 85% of patients have a normal HRCT at an average 5 year follow-up [46–55]. These factors stress the importance of making an SSc-ILD diagnosis by combining clinical findings, pulmonary function tests and HRCT abnormalities.

A common HRCT pattern of SSc-ILD is characterized by a greater proportion of ground-glass opacities with a lower degree of reticulation, suggestive of nonspecific interstitial pneumonia (NSIP). The predominant observations in the basal areas of the lungs are low lung volumes and interstitial reticular thickening. In the late lung involvement stages, pulmonary fibrosis manifests as traction bronchiectasis and honeycomb cysts, a marker for usual interstitial pneumonia (UIP) [55–64] (Figure 1). These two alterations have been observed in up to 33% of SSc-ILD patients, suggesting that these patients may have a mixture (or overlap) of UIP and NSIP patterns [55–64].

**Figure 1.** A 43-year-old female with a diagnosis of systemic sclerosis. Axial high-resolution CT scan obtained in the supine position shows subtle ground-glass opacities in the subpleural regions of the lung bases, suspicious for NSIP (non-specific interstitial pneumonia) (**a**). When these findings are not prominent, an additional scan can be acquired in the prone position to differentiate ground-glass opacities due to gravitational phenomena from interstitial lung disease. In this patient the ground-glass opacities persist in the prone position, confirming the interstitial lung involvement (**b**). Additional systemic sclerosis-related findings should be searched for, such as a dilated esophagus on images reconstructed using an appropriate mediastinal window setting (**c**).

> Even when treated, ground-glass opacities progress to fibrosis and lead to honeycombing/traction bronchiectasis and/or bronchiectasis formation over time in up to 60% of patients [55–64]. There is a correlation between ground-glass opacities/consolidation and active inflammation, whilst reticular opacities/honeycombing correlate with fibrotic

lesions. There is a better treatment response in patients with HRCT features of ground-glass opacities, as they are markers of inflammation and reversible lung injury [50–52,64–67].

Although the HRCT pattern correlates well with histology, nowadays lung biopsies are rarely performed, except for the exclusion of other parenchymal processes [52–55]. However, when performed, histology analysis shows interstitial fibrosis with temporal homogeneity and a modest inflammatory cell infiltrate (i.e., fibrotic NSIP) [42–44,52–55].

It has been observed that HRCT is more sensitive than chest radiography (CR) in diagnosing and characterizing SSc-related lung diseases, as there may be a normal CR in early lung involvement and even in some patients with pulmonary symptoms [58,64,67]. Moreover, ILD HRCT findings correlate more closely with pulmonary function test abnormalities, demonstrating that SSc-related lung injury is a restrictive disorder, associated with low lung volumes, and a diffusion disorder, which impairs carbon monoxide diffusion capacity [55,58].

Chest HRCT findings also have prognostic implications in SSc and SSc-ILD. The absence of lung involvement in HRCT at the time of disease presentation is a good longterm prognostic indicator of SSc-ILD [42,64]. Conversely, the presence of SSc-ILD and its extent, quantified by both visual semi-quantitative and software-based quantitative methods, are able to predict disease-related mortality [64,65]. Along with parenchymal features, lung vessels have also been recently investigated on HRCT. It was observed that the extent of lung volume occupied by vessels has a statistically significant correlation with the extent of SSc-ILD, ILD-related restrictive functional changes and decline in the diffusion capacity of carbon monoxide (DLCO) among SSc patients with or without ILD [51,65–67].

Recently, various radiation-free modalities have been tested, and it seems that lung MRI may be a promising tool for SSc-ILD detection and prognostication [68]. This may be due to the fact that MRI is capable of differentiating inflammation-predominant versus fibrosis-predominant lesions, offering information as to the choice for more antiinflammatory or more anti-fibrotic targeting medications [68]. Moreover, ultrasound lung investigations are becoming widespread in the SSc-ILD field due to their potential for ILD screening [69–71]. Ultrasound correlates well with ILD extent and lung impairment and has a significant prognostic value in the evaluation of lung involvement, even if further studies are required to support the use of the technique [69–71].

#### **5. Pulmonary Function Tests**

Pulmonary function tests (PFT) are essential, readily available non-invasive tests able to detect SSc-related pulmonary changes. PFT in SSc-ILD is characterized by a restrictive ventilatory defect with a decrease in functional vital capacity (FVC) and/or total lung capacity (TLC), a preserved forced expiratory volume in 1 s (FEV1), a normal or increased FEV1/FVC ratio and a DLCO reduction [72–77]. SSc survival has been inversely correlated with the degree of restrictive ventilatory defect on pulmonary function tests. Several studies have reported an 87% 10-year survival rate in patients with minimal to absent restriction and a 75% and 58% 10-year survival rate in patients with moderate or severe restriction [72–77]. Both FVC and DLCO have been identified as adverse prognostic markers in SSc-related lung injury. Indeed, almost all patients have a reduced DLCO, along with other pulmonary function test abnormalities. However, a reduced DLCO is the single most significant marker of poor outcome and correlates with the extent of lung disease on HRCT [72–77]. An important factor, not to be overlooked, is the fact that although patients with early SSc-ILD may have signs of lung disease at HRCT and a DLCO decrease, they may also have preserved lung volumes [72–77].

Recent studies have demonstrated that more than 60% of SSc-ILD patients had normal PFT at HRCT [72–77]. Therefore, although PFT is an important diagnostic tool for SSc-ILD, it is not sensitive enough make an early detection [72–77]. Regular annual PFT after SSc diagnosis may be useful to evidence any changes in lung function that are indicative of ILD [72–77].

The reduced DLCO levels observed in SSc-ILD are due to a variable combination of a reduction in alveolar volume and/or thickening of the alveolar–capillary membrane. Impaired DLCO in SSc-induced lung injury is usually secondary to two main pathological

conditions, i.e., ILD and PAH, even if it may be observed without these complications. Indeed, an isolated DLCO impairment, with reduced FVC/TLC and clinical and/or radiological signs of parenchymal lung involvement, has been attributed to lung vasculopathy and could be considered a good prognostic sign, even it may rarely be associated with the future development of PAH or SSc-ILD [72–78].

Up until 2010, the most common outcome test used in clinical lung disease studies was the DLCO evaluation, which was later surpassed by FVC. Indeed, the FVC percentage predicted the primary endpoint in 70.4% of studies, whilst only 11.3% of DLCO evaluations were predictive. To the best of our knowledge, only five studies specifically aimed to validate PFT: two concluded that the extent of SSc-ILD was best measured by DLCO whilst the other three did not favor any PFT parameter. These studies also showed validity measures for total lung capacity (TLC). Despite the current preference for FVC, available evidence suggests that DLCO and TLC should not yet be discounted as potential surrogate markers for SSc-ILD progression [55,72–78].

#### **6. Pulmonary Arterial Hypertension**

The highest prevalence of PAH amongst the various connective tissue diseases is observed in SSc patients, and it may occur in all forms [20–25]. The main pathophysiological alteration in SSc-PAH is small vessel vasculopathy [20–25,43,51]. This is usually diagnosed 10 to 15 years after SSc onset and is associated with early mortality [20–25,51]. PAH was previously defined as an average pulmonary artery pressure (mPAP) of ≥ 25 mmHg, assessed by right heart catheterization (RHC), with an mPAP between 21 mmHg and 24 mmHg, which was considered "borderline pulmonary hypertension" (Bo-PAH) [19,78]. At the 6th World Symposium of Pulmonary Hypertension, PAH was finally defined as an mPAP of ≥ 21 mmHg with a peripheral vascular resistance (PRV) of ≥ 3 Woods Units (WU) [19,78].

The presence of PAH in SSc may be the result of vaso-occlusive pulmonary artery hypertension (SSc-PAH), left ventricular heart dysfunction or pulmonary hypoxic disease, classified as group 1, 2 and 3 PAH, respectively [19,78]. Group 1 includes patients with isolated PAH without ILD, whilst PAH patients with ILD are classified into group 3, in the PH classification [19,78].

#### **7. Screening**

Our understating of this condition has been changed by the development of systematic algorithms for early diagnosis over the last decade and the data from the follow-up cohorts of incidental SSc-PAH [24]. Indeed, echocardiograph assessment is the most frequently used screening tool to identify candidates for RHC, and a tricuspid regurgitation (TR) velocity of ≥ 2.5 m/s is considered to be highly suggestive of PAH [78]. However, the sensitivity at this TR velocity threshold is limited and misses 20% of mild PAH patients [19,78].

Other studies documented that 55–86% of patients with an echocardiography finding suggestive of pulmonary hypertension (e.g., a right ventricular systolic pressure (RVSP) of 30 to 40 mmHg or higher, with or without symptoms) will have pulmonary hypertension on RHC. When the measurement of RVSP is combined with an increase in right atrial or right ventricular size, reduced pulmonary artery acceleration and decreased right ventricular function, the specificity of echocardiography for pulmonary hypertension diagnosis will be higher [78–84].

The multi-dimensional DETECT algorithm, the forced vital capacity (FVC)/diffusion capacity for carbon monoxide (DLCO) ratio or N-terminal-pro-brain natriuretic peptide (NT-pro-BNP) are all proven screening tools that support the early diagnosis of SSc-PAH [24,75,79]. Reduced DLCO levels in PAH are due to vascular remodeling, which leads to vessel wall tightening and arterial stiffness. The presence of a baseline isolated marked reduction in DLCO (<55% of predicted) in SSc patients might characterize a peculiar SSc subset that may precede the development of PAH, and the progression of pulmonary vascular disease can be linked to decreasing DLCO trends [66,72–74].

#### **8. Right Heart Catheterization**

Right heart catheterization (RHC) is the gold standard investigation for making a definitive diagnosis of pulmonary arterial hypertension (PAH) [80–84]. The RHC provides useful information on the degree of hemodynamic impairment, determines response to PAH therapy and establishes prognosis, providing information for clinical decision-making in PAH management [80–84] (Figure 2). Despite widespread acceptance, there are no internationally accepted clinical guidelines presenting the best practice for performing RHC. Therefore, to ensure the correct evaluation of directly measured hemodynamic or calculated parameters from RHC, procedures such as the position of the pressure transducer and catheter balloon inflation volume should be standardized with care [80–84]. The assessment of pulmonary arterial wedge pressure is particularly vulnerable to over- or under-wedging, which may lead to false readings. Moreover, errors in RHC measurement and data interpretation can complicate the differentiation of PAH from other disorders and lead to a misdiagnosis. Apart from diagnosis, the role of RHC in conjunction with non-invasive tests is on continuous expansion, encompassing the monitoring of treatment response and establishing the prognosis of patients diagnosed with PAH. However, it has been proposed that further standardization of RHC is warranted if we are to ensure its optimal use in routine clinical practice [80–84].

**Figure 2.** From below: pulmonary arterial pressure, respiratory and EKG waveforms during arterial catheterization. The first part of the pressure trace reflects the pressure in a pulmonary artery (large swings, dicrotic notch), then the balloon is inflated and the tip of the Swan Ganz catheter floats until it wedges in a small artery (small swings synchronous with respiratory rate), allowing a pulmonary arterial wedge pressure (PAWP) to be obtained, which is an indirect measure of left ventricle pressure.

#### **9. Peripheral Vascular Manifestations**

Raynaud's phenomenon (RP), secondary to SSc, is the most frequent vascular manifestation in SSc patients. Secondary RP, the most common presenting feature of the disease, is observed in 95% of scleroderma patients and may precede diagnosis by many years [85–88]. During RP, the skin usually turns white (ischemia), blue (deoxygenation) and then red (reperfusion) [85–88].

Secondary Raynaud's phenomenon (SRP) occurs in response to cold temperature or emotional stress, in the setting of underlying vascular disturbance, and is often associated with digital pain and ischemic ulcers [89–92]. However, it may occasionally lead to gangrene with tissue loss or the need for digital amputation [89,90].

As there is an obliterative vasculopathy of the peripheral arteries and microcirculation in SRP, it often leads to critical ischemia in scleroderma. There is often a luminal narrowing of >75% of digital arteries due to underlying intimal fibrosis and luminal occlusion caused by thrombi [85–88]. Endothelial cell injury and activation lead to vascular dysfunction and vasospasm that may quickly obstruct the already limited blood flow of the vasculopathic digital arteries [41].

Conversely, primary RP (PRP) is an isolated finding without underlying pathology (idiopathic). The suggested criteria for PRP include symmetric attacks, the absence of tissue necrosis, ulceration or gangrene, the absence of a secondary cause, negative tests for antinuclear antibodies and a normal erythrocyte sedimentation rate [85,86].

As a diagnosis of PRP is made at a time when no underlying disease has yet been identified, predicting whether or when it may turn into SRP is a difficult task [93–96]. As NVC detects morphological microcirculation abnormalities, it is able to distinguish SRP from both PRP and healthy subjects [97–100]. Therefore, primary RP patients should be carefully followed-up by NVC so as to allow for an early detection of the first signs of any transition to the secondary form of RP in the most reliable manner [101–104].

#### **10. Nailfold Videocapillaroscopy**

Morphological signs that represent the microvascular damage can be observed in nailfold videocapillaroscopy (NVC) images in SRP patients; these alterations include giant capillaries, microhemorrhages, capillary loss, the presence of avascular areas and angiogenesis [105–107]. These sequential capillaroscopic changes are typical of the microvascular involvement observed in more than 95% of SSc patients and are described as an "SSc pattern" [93–95,105]. The nailfold capillaries in PRP patients usually have a normal shape without any specific alterations. Whilst the presence of abnormal capillaroscopic findings, i.e., giant capillaries and microhemorrhages, are diagnostic of the early NVC pattern of scleroderma microangiography [105], the NVC technique is able to identify three morphological patterns specific to various SSc stages (early, active and late patterns) [105]. As reported hereafter, the early NVC pattern is characterized by a few enlarged/giant capillaries and capillary microhemorrhages, no evident capillary loss and a relatively well-preserved capillary distribution [105]. The most frequent alterations in the active NVC pattern are giant capillaries and capillary microhemorrhages, with a moderate capillary loss and a mild disorganization of the capillary architecture. There is severe capillary loss with evident avascular areas and disorganization of the normal capillary array in the late NVC pattern [105]. NVC provides a quantitative assessment of the microvascular damage, i.e., a quantification of certain characteristics and a semi-quantitative scoring. The characteristic capillaroscopic diagnostic parameters, i.e., irregularly enlarged capillaries, giant capillaries, microhemorrhages and progression parameters, which include fewer capillaries, capillary ramifications and capillary architectural disorganization, can be scored from 0 to 3 according to increasing severity and have been combined to create a semi-quantitative scale [105].

In healthy and primary RP subjects, NVC evaluation is characterized by morphological/structural homogeneity, evidencing 10–12 capillaries per linear millimeter, morphology of the capillary to "U" or "hairpin shape" and diameters of capillary branches of <20 μm [93–95]. Although it is quite common to observe normal nailfold capillaries

in primary RP, capillaries with efferent branch enlargement or tortuosity may also be present [93–95].

The European League Against Rheumatism (EULAR) Study Group on Microcirculation in Rheumatic Diseases (EULAR SG MC/RD) has recently reported a simple consensus definition to name a single capillary as "(ab)normal". The authors tried to standardize and clarify the differences between scleroderma and non-scleroderma patterns, avoiding confusion caused by the various different definitions used to describe non-scleroderma abnormal capillary morphology (e.g., "ramifications", "neoangiogenesis" or "meandering") [28].

In conclusion NVC, which combines a microscope (with system that ranges from 50 × up to 500 × magnification) and a digital video camera, represents a method for an early diagnosis and follow-up of nailfold microangiopathy—one of the earliest signs of morphological damage and change in SSc—and is a non-invasive, user-friendly, wellaccepted, accessible and portable tool [28].

That is why abnormal nailfold capillaroscopic images, i.e., "scleroderma patterns", were included in the 2013 European League Against Rheumatism and American College of Rheumatology's classification criteria for SSc [103]. Several studies have also demonstrated that NVC is a promising tool for the prediction of clinical complication markers of severity and progression of SSc organ involvement [49–54].

#### **11. The Correlation Between Peripheral Vascular and Pulmonary Involvement**

Various studies have demonstrated that NVC alterations are associated with different SSc clinical complications and organ involvement [106–112]. Moreover, other authors reported on the correlation between NVC alterations and SSc-ILD diagnosed by HRCT [110,111]. Caetano et al., made a cross-sectional analysis of 48 SSc-ILD patients with HRCT and the presence of ground-glass opacities and/or fibrosis [111]. The same authors investigated the association between NVC findings, the presence and extent of ILD, as well as functional impairment. Capillary loss and avascular areas were significantly associated with the presence of ILD. The receiver operating characteristic (ROC) curve analysis confirmed the association between capillary loss and ILD (the area under the ROC curve, 90.1%; 95% CI, 81.8–91.4). Avascular areas and capillary loss were associated with a worse pulmonary function. No additional statistically significant difference was observed between ILD and other NVC findings (i.e., capillary dimension (*p*-value = 0.328), abnormal capillary morphology (*p*-value = 0.790) or the presence of hemorrhages (*p*-value = 0.187)) [111]. In another cross-sectional study, Guillen-del-Castillo et al. evaluated 134 SSc patients (58 with ILD on HRCT) with at least eight NVC (200 × magnification) images through both quantitative and qualitative examinations [110]. The SSc ILD patients had a lower median capillary density (4.86/mm vs 5.88/mm, *p*-value = 0.005) and higher median neoangiogenesis (0.56/mm vs 0.31/mm, *p*-value = 0.005). Moreover, more neoangiogenesis capillaries were observed in PAH patients (0.70/mm vs 0.33/mm, *p*-value = 0.008). A multivariate linear regression analysis emphasized a correlation between neoangiogenesis and decreased FVC (*p*-value < 0.001) and between the number of giant capillaries and reduced DLCO (*p*-value = 0.016) [110]. Guillen-del-Castillo et al., demonstrated that the late pattern was associated with lower FVC (*p*-value = 0.018) [110]. Jehangir et al.,'s case-control study made use of a dermascope to study the NVC pattern in 65 subjects: 10 patients with primary Raynaud's phenomenon (RP), 40 with SSc and 15 age- and gender-matched controls. When testing HRCT and NVC patterns, only one patient with the early pattern had ILD, whereas those with an active or late pattern had a higher percentage of 55% and 100%, respectively [108] (Figure 3).

**Figure 3.** Vasculopathy in systemic sclerosis patients and the link between pulmonary damage and peripheral vascular manifestations.

Markusse et al. performed NVC in 287 SSc patients aimed at assessing whether it could improve the detection of patients at high risk of cardiopulmonary involvement (82). The study population included 51% ILD patients, 59% with a DLCO decrease and 16% with a systolic pulmonary artery pressure (sPAP) of >35 mmHg. The NVC pattern showed a stable association with the presence of ILD or sPAP. The odds ratio (OR) for ILD was 1.3–1.4 (*p*-value < 0.05 for analyses with anti-RNAPIII, anti-RNP). The OR for DLCO was 1.5 (*p*-value < 0.05 for analyses with ACA, anti-Scl70, anti-RNAPIII, anti-RNP). The OR for sPAP was 2.2–2.4 (*p*-value < 0.05 for analyses with anti-RNAPIII, anti-RNP) [82].

It has been demonstrated that SSc-associated PAH is correlated with capillaroscopic changes identified by NVC [112–127]. NVC data are also markers of SSc severity and progression, such as reduced capillary density, which is associated with a high risk of developing PAH [112–123]. Some authors have supported the possibility of an early identification of a subset of patients with severe disease [112–127]. Hofstee et al. studied capillary density and dimensions and their association with pulmonary hemodynamic characteristics in 21 healthy controls, 20 idiopathic PAH patients and 40 SSc patients (21/40 had SSc-PAH, as determined by RHC) [123]. This study reported a significantly lower capillary density in SSc-PAH patients than in those without SSc-PAH (*p*-value = 0.001), although no statistically significant difference was observed for capillary dimensions (*p*-value > 0.05) [123].

Similarly, Corrado et al. evaluated 25 healthy subjects, 21 idiopathic PAH patients and 39 SSc patients (19/39 affected by SSc-PAH, determined by RHC, as mPAP ≥ 25 mmHg, PCWP ≤ 15 mmHg and pulmonary vascular resistance > 3 Wood units) [118]. The authors observed that the presence of PAH had a significantly inverse correlation with capillary density (*p*-value < 0.05) and correlated with both capillary dimension and giant capillaries (*p*-value < 0.05), which correlated with abnormal capillary morphology (*p*-value < 0.01) in SSc patients [118].

Riccieri et al.'s cross-sectional study evaluated NVC alterations in 12 consecutive SSc-PAH patients, confirmed by RHC. They demonstrated that NVC damage is correlated with the grade of PAH.

Indeed, the NVC score (combining a semiquantitative score for density, dimension, presence of hemorrhage and morphology) and avascular area grading had a statistically significant correlation with PAH (*p*-value = 0.03 and *p*-value = 0.003, respectively) [121]. Furthermore, they observed that the active/late pattern was more common in SSc-PAH patients than in those without SSc-PAH (73% vs 50%, *p*-value < 0.05) [121]. This led to two recent meta-analyses confirming that microvascular changes detected by NVC are significantly associated with SSc-PAH, in particular lower capillary density and higher capillary width [112,117].

Kim et al. investigated the relationship between clinical manifestations and quantitative analysis of computerized NVC. They observed a strong correlation between capillary dimension and capillary loss with SSc-PAH (*p*-value < 0.05) and digital ulceration (*p*-value < 0.01) [114]. Moreover, a cross-sectional pilot study also assessed sublingual microvasculature by videocapillaroscope. The authors demonstrated that PAH patients had a lower sublingual microvasculature flow index and a higher vascular tortuosity than healthy age- and gender-matched control subjects [128].

In addition to the cross-sectional evidence, Smith et al.'s prospective longitudinal research studied 66 consecutive SSc patients. They defined the NVC pattern according to Cutolo's classification and clinical evaluation was performed according to Medsger's disease severity scale (DSS), with an 18–24 month follow-up [120]. They observed a statistically significant association between the NVC patterns and the development of future severe peripheral vascular or lung involvement. Indeed, the OR for future severe lung involvement, based on simple/multiple regression was 2.54/2.33 (*p*-value < 0.05) for early patterns, 6.43/5.44 for active patterns and 16.30/12.68 for late patterns.

#### **12. Conclusions**

The characteristic features of SSc include extensive fibrosis, fibroproliferative vasculopathy and systemic autoimmunity (autoantibodies and T cell autoantigen reactivity). However, the vascular pathology in SSc is not necessarily an inflammatory process and may be better characterized as a vasculopathy. Some autopsy studies have shown that vasculopathy is a systemic process [128]. One such study is that of D'Angelo et al., who reported that SSc patients had widespread intimal proliferation in the pulmonary arteries [128]. Moreover, complex interactions between endothelial cells, vascular smooth muscle cells, extracellular matrix and circulating mediators contribute to vascular remodeling, vasospasm and vessel occlusion [128].

There is evidence to support the hypothesis that microangiopathy may be an important component of internal organ involvement and that NVC is a candidate biomarker for the assessment of pulmonary damage.

There are numerous reports finding a good correlation between distinctive quantitative and qualitative NVC features and the presence of ILD on HRCT, as well as lung functional parameters such as FVC and DLCO. Therefore, it might well be that microangiopathy is a pivotal process in the establishment and progression of SSc-ILD. Similarly, as NVC is capable of detecting the early microvascular changes associated with the presence of PAH, it may well play a significant role in early prediction of SSc-PAH.

The gold standard with which to diagnose SSc-ILD is still the chest HRCT, whilst RHC is the validated test with which to diagnose PAH. Due to the poor prognosis for SSc patients with organ damage, in particular those with pulmonary manifestations, all these patients should be carefully evaluated from the early disease phase and followed-up. This should hopefully facilitate early identification and the choice of an early appropriate therapeutic regimen.

This review reports on the different clinical manifestations and tests, i.e., clinical, radiological and pulmonary function tests that may be used for the early prediction of lung involvement in SSc patients. It also includes evidence from the literature on how NVC may also be a promising tool for early identification and/or prediction of pulmonary complications. The data herein reported support the possibility that NVC might be incorporated together with other parameters in high-performance algorithms (e.g., DETECT algorithm for SSc-PAH, ILD screening procedures) in the early detection of lung involvement in SSc.

**Author Contributions:** B.R., C.B. and M.C. contributed to study design. F.S., B.W., E.B. and P.C. contributed to manuscript drafting. P.G., M.B. and M.K. contributed to literature revision. All the authors critically revised the manuscript and approved the final version. All authors have read and agreed to the published version of the manuscript.

**Funding:** The authors received no financial support for the research, authorship, and/or publication of this article.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare that there are no conflict of interest.

#### **References**


### *Review* **The Treatment of Lung Involvement in Systemic Sclerosis**

**Barbara Ruaro 1,\*, Marco Confalonieri 1, Marco Matucci-Cerinic 2, Francesco Salton 1, Paola Confalonieri 1, Mario Santagiuliana 1, Gloria Maria Citton 1, Elisa Baratella <sup>3</sup> and Cosimo Bruni <sup>2</sup>**


**Abstract:** Systemic sclerosis (SSc) patients are often affected by interstitial lung disease (ILD) and, although there have been recent treatment advances, it remains the leading cause of death among SSc, with a 10-year mortality up to 40%. African Americans and subjects with diffuse cutaneous SSc or anti-topoisomerase 1 antibodies are most commonly affected. Currently, early ILD diagnosis can be made, and it is pivotal to improve the prognosis. The diagnostic mainstay test for SSc-ILD is high-resolution computed tomography for the morphology and pulmonary function tests for the functional aspects. Treatment planning and intensity are guided by the disease severity and risk of progression. Traditionally, therapy has depended on combinations of immunosuppressants, particularly cyclophosphamide and mycophenolate mofetil, which can be supplemented by targeted biological and antifibrotic therapies. Benefits have been observed in trials on hematopoietic autologous stem cell transplantation for patients with progressive SSc, whilst lung transplantation is reserved for refractory SSc-ILD cases. Herein, recent advances in SSc-ILD treatment will be explored.

**Keywords:** systemic sclerosis; scleroderma; interstitial lung disease; pulmonary function tests; high-resolution computed tomography

#### **1. Introduction**

Scleroderma or Systemic Sclerosis (SSc), a disease characterized by fibrosis, vasculopathy, and inflammation, may affect different organ and systems, with severe prognostic implications [1,2]. When SSc pathogenetic processes manifest at lung level [3], pulmonary disease may manifest both as interstitial lung disease (ILD) and/or pulmonary arterial hypertension (PAH) [4,5]. The European Scleroderma Trials and Research (EUSTAR) group reported that 53% of cases with diffuse cutaneous SSc (dcSSc) have ILD, as do 35% of cases with limited cutaneous SSc [6]. Moreover, high-resolution computed tomography (HRCT) evidences interstitial abnormalities in 90% of SSc patients [7], and pulmonary function tests (PFT) showed alterations in 40–75% [8]. There has been no significant change in SSc mortality rate over the past 40 years [9,10], although an increase in mortality due to ILD and PAH [11,12] is significant, a decrease in deaths due to renal crisis has been recorded [13]. Nowadays, ILD and PAH are the two leading causes of death in SSc, accounting for 33% and 28% of deaths, respectively [10–12]. The survival of systemic sclerosis-related interstitial lung disease (SSc-ILD) patients is reported to be 29–69% at 10 years [9,12]. Early autopsy studies demonstrated that up to 100% of patients had parenchymal involvement [14]. Considering the frequency and the prognosis of SSc-ILD patients, it is essential to attempt to identify pulmonary disease early, at a potentially reversible stage [15].

Unfortunately, there are limited treatment options for this manifestation, given that the paucity of high-quality, randomized, controlled trials specifically targeting SSc-ILD are

**Citation:** Ruaro, B.; Confalonieri, M.; Matucci-Cerinic, M.; Salton, F.; Confalonieri, P.; Santagiuliana, M.; Citton, G.M.; Baratella, E.; Bruni, C. The Treatment of Lung Involvement in Systemic Sclerosis. *Pharmaceuticals* **2021**, *14*, 154. https://doi.org/ 10.3390/ph14020154

Academic Editor: Félix Carvalho Received: 19 January 2021 Accepted: 10 February 2021 Published: 13 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

scanty, and, historically, studies have favored cyclophosphamide (CYC) for the treatment of SSc-ILD, as also suggested in the most recent European League against Rheumatism (EULAR) recommendations [16–19]. The most recent and supportive data showed the positive effect of nintedanib, a multi-tyrosine kinase inhibitor, as a significant inhibitor of progressive functional decline [20]. Innovative proposals have also recently been made on the basis of clinical and preclinical evidence for rituximab (RTX), tocilizumab (TCZ), and pirfenidone (PIRF), as well as hematopoietic stem cell transplantation and lung transplantation [21]. However, the safety and efficacy of emerging experimental therapies for SSc-ILD require further investigation. The aim of this review is to summarize the state-of-the art in SSc-ILD treatments.

#### **2. Management Principles**

As SSc-ILD is a very heterogeneous disease, management tends to differ according to the profile of the patient. Furthermore, with the advent of the new aforementioned treatment options, it is pivotal to detect ILD [22–24] as early as possible and also to assign the right treatment as early as possible [7,19]. Toward this aim, precise and objective ILD classification tools that allow for patient stratification at ILD detection and diagnosis play a major role [25]. Indeed, patients must be classified by a severity assessment of ILD at diagnosis, performed by HRCT and PFT, and then by the evaluation of the risk of ILD progression [25,26].

The HRCT variables predictive of mortality and ILD progression in SSc–ILD were studied and reported in a recent meta-analysis of 27 studies, which concluded that the extent of disease on HRCT was an independent predictor of both mortality and ILD progression [27].

It is a must to detect the subset of clinical ILD patients with progressive disease, defined as a decline in Forced Vital Capacity (FVC) levels of >10% from baseline or a ≥ 5% to < 10% relative decline in FVC and a ≥ 15% relative decline in Diffusion Lung Capacity of Carbon monoxide (DLCO) over 12 months [28]. Despite this cut-off being proposed for clinical trials and applied also in clinical practice, smaller changes may also be of clinical importance, in particular worsening symptoms attributable to ILD [25,26,29]. DLCO alone was also one of the most consistent predictors of mortality, a finding that may well help in the identification of patients with a poor prognosis, even if these preliminary findings should be confirmed and expanded by further rigorous studies [25–27]. The likelihood of progression, comorbidities, and toxicity risks and current data on efficacy are often the basis for decisions taken to initiate or advance treatment [30]. The goal of treating clinical SSc-ILD is the stabilization or prevention of progressive disease.

#### **3. Treatment Options**

The 2017 EULAR recommendations for the treatment of SSc state that the physician's assessment of symptoms, disease severity, and/or disease progression form the basis for decisions to initiate SSc-ILD treatment, with a tailored risk–benefit evaluation especially in progressive SSc-ILD patients [18]. There was also a recommendation for the use of CYC and hematopoietic stem-cell transplantation for SSc-ILD patients. After the release of this recommendations, evidences for a positive effect of mycophenolate mofetil (MMF) and nintedanib have also become available [17,20]. Several studies have also been reported that tocilizumab and rituximab might be able to slow down ILD progression [21]. A summary of the treatment options further discussed is presented in Table 1.


**Table 1.** Summary of current treatment options for systemic sclerosis related interstitial lung disease.

DLCO = diffusion lung capacity of carbone monoxyde; FVC = forced vital capacity; HRCT = high resolution computed tomography; ILD = interstitial lung disease; OS = observational study; PROs = patient reported outcomes; RCT = randomized clinical trial; CR/CS = case report/case series.

#### **4. Conventional and Biologic Immunosuppressants**

SSc is a connective tissue disease where inflammation and immune abnormalities play a central role [64–67]. The immune system, especially B and T lymphocytes, is involved in fibroblast activation and fibrogenesis as they secrete proinflammatory and profibrotic cytokines and growth factors [64–67]. That is why traditional immunosuppressant, e.g., CYC, MMF and azathioprine (AZA) have been so far considered the milestones of SSc-ILD treatment.

#### *4.1. Cyclophosphamide*

CYC is the most commonly used immunosuppressant, and it has been tested in numerous open-label studies, as well as in a few randomized control trials (RCT) [68]. CYC is recommended as first-line therapy in SSc–ILD patients in the EULAR guidelines [18].

In the Scleroderma Lung Study (SLS) I, 1-year course of oral CYC up to 2 mg/kg/day showed a statistically significant but small improvement in FVC (2.5% improvement) vs. placebo and little sustained benefit after discontinuation [16]. Similar results were not confirmed in the Fibrosing Alveolitis in Scleroderma Trial (FAST), which reported no statistically significant difference between the placebo and CYC group [31]. The clinical significance of this is modest, yet real improvement in FVC is still under debate and it seems that there will be decades of pros and cons. Noteworthy is the fact that the SLS I patients most likely had a stable SSc-ILD, as only 15% of them needed to restart an immunosuppressive treatment after the end of the study [16].

The SLS II (head-to-head comparison of oral CYC up to 2 mg/kg/day for 1 year plus 1 year of placebo versus MMF at up to 1.5 g twice daily for 2 years) showed that the benefits of MMF on FVC and on improvement in dyspnea were similar to those obtained with oral CYC at 2 years (MMF 2.2%, CYC 2.9%), with a safety profile favoring MMF [17].

In conclusion, it seems that CYC can either stabilize worsening SSc-ILD or modestly improve stable SSc-ILD; these data were also confirmed in a recent comparison between intravenous and oral CYC administration analyzing patients derived from the SLS1, SLS2, and EUSTAR cohorts. These results showed non-different effect on FVC change and ILD progression for the two routes of administration, despite a significantly lower CYC dosages in the intravenous group and a significantly different safety profile [32].

#### *4.2. Mycophenolate Mofetil*

MMF inhibits lymphocyte proliferation and is a safer, less toxic alternative to CYC for the treatment of SSc–ILD. Indeed, the safety and efficacy of MMF in SSc–ILD patients has been reported in several case series, uncontrolled studies, and, more recently, 2 metaanalyses [17,33,40,41]. Recently, the SLS II study, which reported on SSc–ILD patients treated with MMF for 2 years or CYC for 1 year followed by one year of placebo, showed that both treatment regimens led to a significant improvement in the pre-specified measurements of lung function over the 2-year study period. However, even if MMF was better tolerated and had lower toxicity levels, the hypothesis that it would be more efficacious at 24 months than CYC was not confirmed [17]. Although these data support the potential clinical efficacy of both CYC and MMF for progressive SSc–ILD, there is a possible preference for MMF due to its better tolerability and toxicity profile [17]. Lastly, Owen et al. demonstrated that MMF therapy was associated with a clinical stability for up to 36 months and lower frequency of early adverse events compared to AZA for SSc–ILD patients with a decline in pulmonary function [42].

#### *4.3. Azathioprine*

Although some small case series and retrospective studies suggested that AZA could be used as maintenance immunosuppressive treatment for SSc–ILD [34,35], a randomized unblinded clinical trial comparing the use of CYC and AZA (a purine analog) as first-line treatment did not evidence the efficacy of AZA in the treatment of SSc–ILD [36]. In addition, the very recent study by Owen et al. showed the better efficacy and tolerability of MMF versus AZA in the management of SSc–ILD [42].

#### *4.4. Rituximab*

A few case reports and open-label uncontrolled studies reported an improvement in SSc–ILD with RTX. Indeed, RTX therapy in SSc has gained favor after reports on its promising effects on both ILD and skin thickening [52,53,69]. The largest observational study available so far was published by the EUSTAR group and included 254 SSc patients treated with RTX, showing a good safety profile, steroid sparing agent potential and good efficacy profile on the skin but not on the lung. At pulmonary level, the combination of RTX + MMF determined a significant reduction in FVC decline over time, compared to monotherapy, therefore hypothesizing a higher promising potential for the combination treatment [54]. A similar safety profile and potential beneficial effect was also confirmed in an observational cohort receiving biosimilar RTX [55].

A recent open-label, randomized, controlled trial of head-to-head RTX vs. monthly pulse CYC reported on a population of 60 early, treatment naïve, anti-SCL-70+, dcSSc with ILD patients receiving either CYC or RTX. At the end of 6 months, the authors observed that FVC improved from 61.3% to 67.5% in the RTX group, whilst it did not in the CYC group (59.3% to 58.1%) [37]. The currently ongoing Rituximab versus Cyclophosphamide in Connective Tissue Disease-ILD (RECITAL) study (NCT01862926) is investigating the same topic in a larger cohort of connective tissue diseases related ILD, with a longer follow-up (48 weeks) [70].

#### *4.5. Tocilizumab*

The first two studies on TCZ, an anti-IL-6 soluble receptor monoclonal antibody, reported inconclusive results [48,49]. TCZ was administered in a phase 2 study (FaSScinate), and the data suggested this drug played a role in the IL-6 pathway in SSc-ILD and treatment of early SSc with elevated C-Reactive protein (CRP) and that it led to the stabilization of the FVC% in the tocilizumab group vs. a clinically meaningful decline in the placebo group over 48 weeks [48]. In this view, the phase 3 double-blind randomized placebo-controlled study (FocuSSced) of TCZ enrolled 210 early dcSSc patients. Similarly, a reduced FVC decline was seen in the TCZ group (difference between groups 4.2 (95% CI 2.0–6.4) favoring TCZ; *p* = 0.0002), with a trend for a lower rate of patients requiring rescue immunosuppressive therapy for ILD indication (*p* = 0.08) [49].

Although the primary (skin) endpoint was not met, both trials showed some efficacy and a good safety profile for using TCZ in SSc and evidenced a potential benefit of treating subclinical ILD patients with high risk features (early dcSSc, and elevated CRP) [50].

#### *4.6. Abatacept*

Abatacept is a recombinant fusion protein that inhibits T cell activation. An observational study was carried out on 20 patients with SSc-associated polyarthritis and myopathy to evaluate the safety and efficacy of TCZ. However, despite having a good safety profile, there was no change in lung fibrosis in patients treated with abatacept [51]. A similar, more recent, observational experiment from the EUSTAR group, which included 27 SSc patients (15 with ILD), confirmed the good safety profile of the drug as well as a beneficial effect on joint and muscle disease. In addition, a possible positive effect was also seen for skin fibrosis (despite the lack of a control group), while no significant change in lung function was detected [56]. Finally, a recent phase II multicentre double-blind placebo-controlled trial of abatacept in early dcSSc showed a trend for a significant lower decline of FVC% predicted (mean difference 2.79, 95% CI −0,69–6,27, *p* = 0,11 favouring Abatacept), although the change in skin fibrosis as a primary endpoint was, again, not met [57]. Despite this, there is a promising potential for Abatacept in SSc, which could be investigated in a phase III study.

#### **5. Other Treatment Options**

#### *5.1. Immunoglobulins*

A randomized control trial assessing the change in skin fibrosis as primary endpoint failed to demonstrate a significant beneficial reduction of modified Rodnan skin score in Intravenous immunoglobulin (IVIg) administration versus placebo [71]. Different authors have shown the beneficial effect of the use of IVIg in SSc patients with arthritis [72] and inflammatory myopathy [73], and they have demonstrated some potential benefit on earlystage ILD, with the authors reporting a regression in ground glass opacity, septal thickening, and a full recovery of lung function [74]. A randomized phase II trial (NCT04137224) is currently testing a possible similar effect and the safety of subcutaneous immunoglobulins and IVIg [75].

#### *5.2. Autologous Hematopoietic Stem Cell Transplant*

Hematopoietic stem cell transplant (HSCT) is an emerging treatment option, aimed at regenerating the patient's immune system [43,76]. It is based on the use of high-intensity immunosuppression (conditioning regimen) aimed at a strong reduction/eradication of the "auto-reactive" immune system, followed by a re-population with antigen-naïve T cells previously isolated from the same individual [76]. It has been proposed for patients with dcSSc (with or without SSc-ILD) that is severe and refractory to standard therapy, who will probably benefit from the procedure but are more unlikely to develop post-transplant complications [44,77]. Indeed, improved survival compared to CYC has been reported by three trials, i.e., Autologous Stem Cell Systemic Sclerosis Immune Suppression Trial (ASSIST), Autologous Stem Cell Transplantation International Scleroderma trial for (ASTIS), and Scleroderma: Cyclophosphamide Or Transplantation trial (SCOT). Moreover, there was an improvement in skin thickening and FVC, as well as quality of life [38,39,45]. In addition to these results, an observational analysis of ILD extent in SSc patients receiving HSCT versus CYC was recently published, showing significant reduction in total ILD extent and, in particular, in the extent of ground glass opacifications, which were not seen in the CYC group [46]. With this promising background, a phase III randomized clinical trial (NCT044644) is currently testing upfront HSCT versus intravenous CYC induction followed by maintenance with MMF in early dcSSc, including pulmonary endpoints [78].

#### *5.3. Lung Transplant*

Lung Transplant is a life-saving option and remains a therapy for appropriately selected candidates with treatment-refractory lung disease [47,79]. An early referral should be made for advancing disease so as to provide these patients with a multi-disciplinary evaluation before transplant is considered an option. A few recent studies have demonstrated an increase in survival after lung transplantation [47,79].

#### *5.4. Non-Pharmacologic Therapy*

SSc-ILD should be managed by a multidisciplinary team [80]. Among non-pharmacologic options, pulmonary rehabilitation is aimed at improving lung function [81]; in particular, when an SSc-ILD patient is being considered for a transplant, pulmonary rehabilitation is a necessary step in their evaluation [81]. Furthermore, supplemental oxygen should be given whenever deemed necessary.

#### **6. Anti-Fibrotic Therapies**

Nintedanib is an intracellular tyrosine kinase inhibitor approved for the first time for the treatment of idiopathic pulmonary fibrosis (IPF) [82,83]. Its pharmacological effect covers numerous pathophysiological pathways, such as fibroblast activation, myofibroblast accumulation, and fibrogenic cytokine and growth factor expression. The increasing number of national and international authorities giving approval for treatment with the antifibrotic agent nintedanib to slow down the rate of decline of pulmonary function in SSc-ILD patients is opening up a new era [84]. The results of the recently published Safety and Efficacy of Nintedanib in Systemic Sclerosis (SENSCIS) trial supported the decision [20]. The SENSCIS trial, a double blind, randomized, placebo-controlled trial, evaluated the efficacy and safety of oral nintenadib (150 mg bid) treatment in patients with SSc-ILD for at least 52 weeks [20]. It reported that almost 50% of the subjects had dcSSc; a similar percentage was on a stable dose of MMF and HRCT evidenced fibrosis in at least 10% of the lungs (the latter as per study inclusion criteria). In this trial, patients with SSc-ILD treated with nintedanib showed a significantly lower rate of annual FVC decline than those receiving placebo, despite no significant improvement or benefit in any of the other organ manifestations. Although the change in FVC was small (absolute mean decline mean −52.4 mL per year in the nintedanib versus −93.3 mL per year in the placebo group), the mean decline reached a previously shown value of minimal clinically important difference (MCID) in the placebo group, but not in the nintedanib treated population [85]. This beneficial effect on FVC preservation was seen both in MMF and non-MMF cotreated patients, with a numerically lower decline in patients receiving the combination treatment [58]. In the SENSCIS study, Nintedanib showed a safety profile similar to the side effects seen in IPF, particularly affecting the gastrointestinal tract (75.7% of treated patients manifested diarrhea) and requiring dose-adjustment/temporary interruption in almost half of the treated patients [59]. Interestingly, the safety profile was similar in patients receiving or not receiving co-treatment with MMF, which itself carries a gastro-intestinal burden in terms of adverse events [58].

With a similar multi-target pathogenetic activity, pirfenidone (a pyridone showing both anti-inflammatory and anti-fibrotic effects) is another antifibrotic agents approved for the management of IPF patients [86]. The initial compassionate use in selected patients with SSc–ILD showed that the drug was well tolerated and, although it did not improve survival, it did stabilize the effects of progressive pulmonary fibrosis [60,61]. Recently, the Safety and Tolerability of Pirfenidone in Participants with Systemic Sclerosis-related Interstitial Lung Disease (LOTUSS) study, a phase II, open-label, randomized, 16-week study, assessed the safety and tolerability of pirfenidone in SSc–ILD patients. The drug was reported to have an acceptable tolerability profile that was not affected by concomitant treatment with MMF, although data as to its efficacy is not yet available [62]. Indeed, pirfenidone can be associated with adverse events of the gastrointestinal system and the skin in patients with IPF, two organs very frequently involved in SSc-ILD. Sometimes these

adverse events can lead to drug discontinuation [63]. Given the promising effect in the stabilization of SSc-ILD, the drug is now tested versus placebo in SSc-ILD patients receiving MMF as a background immunosuppressive therapy in a placebo-controlled multi-center double blind randomized SLS study III [87].

#### **7. Conclusions**

Although ILD is a common finding in SSc, currently there is a paucity of detailed data to help in predicting which subsets of patients will or will not develop organ and potentially life-threatening disease. Despite this, the potential risk of morbidity and mortality supports the need for a thorough and early monitoring of the signs and symptoms of the development and progression of ILD.

At time of writing, the standard of care includes the use of CYC and MMF (which have only provided modest improvements in FVC) and Nintedanib (which is not available worldwide). Preliminary data on newer therapies, like biologics, stem cell transplant, and other anti-fibrotics suggest improved efficacy and safety profiles compared to those obtained with conventional immunosuppressive therapy.

Following the SLS II trial, a Delphi consensus treatment algorithm advocated MMF as first-line treatment of SSc-ILD and suggested that second-line treatment should include CYC or rituximab as an induction therapy, followed by MMF as a maintenance therapy [88].

A more recently published European consensus on SSc-ILD identification and management stressed the importance of different factors guiding treatment initiation, including speed of disease progression, survival rate, response rate after previous treatment, prolongation of time to progression, speed of improvement of patients' symptoms, safety and tolerability, scientific evidence of efficacy, and impact on quality of life [15].

In addition, disease severity and speed of progression could be the main drivers of treatment escalation [15]. Although consensus and recommendations are nowadays available, these do not fully cover the different clinical scenarios, in particular regarding time to initiation and a possibly more effective treatment protocol. In this context, SSc experts still relay on their clinical experience and take into account the different abovementioned factors to guide their decision in a patient-tailored, customized treatment regimen [89], possibly informed also by molecular biomarkers [90].

Although a substantial amount of evidences in SSc-ILD management resemble IPF, SSc-ILD is not IPF, as patients with SSc-ILD have a systemic disease. It has been hypothesized that future therapeutic options may be provided by targeting the self-perpetuating fibrosis [91], although whether an early (immunosuppressive) aggressive treatment will lead to a modification of disease progression and prevention of irreversible lung damage remains a question of debate. In this context, limiting fibrogenesis by the use of antifibrotic therapy and controlling inflammation/immunological abnormalities through immunosuppressants could well become the new paradigm of treatment in SSc-ILD. If available and well-tolerated, a combination regimen with immunosuppression and anti-fibrotic may allow a multi-target treatment and, potentially, a multiple organ/system benefit. Specifically, immunosuppression could also be personalized according to non-ILD organ complications such as cutaneous involvement, arthritis, myositis, and cardiomyopathy.

Clearly, there is a need for guidance in the new treatment regimens, in particular regarding the use of upfront or add-on combination treatment with immunosuppressants/antifibrotic, which could be the possible second/third level option in case of treatment failure. Hopefully, our understanding of the pathogenesis of SSc-ILD will evolve, along with the development of specific therapies for the organ systems affected by this disease, thus improving patients' survival, function, and quality of life [92,93].

**Author Contributions:** B.R. and M.M.-C. contributed to the study design; B.R., C.B., and M.C. contributed to manuscript drafting; F.S., M.S., G.M.C., E.B. and P.C. contributed to literature revision. All authors have critically revised the manuscript and approved the final version. All authors have read and agreed to the published version of the manuscript.

**Funding:** The authors received no financial support for the research, authorship, and/or publication of this article.

**Acknowledgments:** The authors would like to express their appreciation to Barbara Wade for her language assistance.

**Conflicts of Interest:** The authors declare that there is no conflict of interest.

#### **References**

