Interstitial Lung Disease in Systemic Lupus Erythematosus and Systemic Sclerosis: How Can We Manage the Challenge?

Interstitial lung disease (ILD) is a severe and frequent manifestation of connective tissue diseases (CTD). Due to its debilitating potential, it requires serious evaluation and treatment. The prevalence of ILD in systemic lupus erythematosus (SLE) is still controversial. Therefore, in order to establish the diagnosis of ILD, an overlap syndrome must be excluded. Increasing the identification of SLE-associated ILD cases should become a target. To treat this complication, various therapies are now being proposed. To date, no placebo-controlled studies were conducted. Regarding another CTD, systemic sclerosis (SSc), SSc-associated ILD is considered one of the leading causes of mortality. The incidence of ILD varies among disease subtypes, being influenced by diagnostic method, but also by disease duration. Due to the high prevalence of this complication, all SSc patients should be investigated for ILD at the time of SSc diagnosis and during the course of the disease. Fortunately, progress was made in terms of treatment. Nintedanib, a tyrosine kinases inhibitor, showed promising results. It appeared to decrease the rate of progression of ILD compared to placebo. This review aimed to provide up-to-date findings related to SLE-associated ILD and SSc-associated ILD, in order to raise awareness of their diagnosis and management.


Introduction
Data from the literature on connective tissue disease-associated interstitial lung disease (CTD-ILD) focused on systemic sclerosis (SSc)-ILD and rarely on systemic lupus erythematosus (SLE)-ILD. SLE is a chronic, autoimmune condition that can present with a wide range of clinical and immunological manifestations. Pulmonary involvement in SLE is various. SLE-associated ILD, while rare, is a predictor of poor prognosis [1].
SSc is a heterogeneous autoimmune disorder characterized by multiorgan vascular and fibrotic abnormalities. On the other hand, pulmonary involvement in SSc is frequent, ILD being considered one of the most common cause of death. Despite being one of the most significant SSc complications, there is no established treatment for it yet. A new treatment for ILD is, therefore, definitely needed. Recently, nintedanib was successfully approved for SSc-ILD in Japan, Europe, and the United States.

Systemic Lupus Erythematosus
In addition to other systemic manifestations, pulmonary involvement has an impact on patients' quality of life and disease prognosis [1]. ILD, a pulmonary manifestation of several connective tissue disorders (CTDs), frequently causes important morbidity and mortality [2]. Scleroderma is one of the CTD's most frequently reported ILD; however, et al enrolled 505 SLE patients and additionally observed the presence of tachypnea in 55 ILD patients [28]. Low levels of albumin and chest tightness/shortness of breath should indicate the presence of ILD, regarding a recent study that found statistical significant differences between the SLE-ILD and SLE-non ILD groups [29].
ILD can be divided into several manifestations: follicular bronchitis, organizing pneumonia (usual interstitial or lymphocytic interstitial pneumonia) and non-specific interstitial pneumonia (NSIP) [9,[30][31][32][33]. The onset of the symptoms is usually insidious [34]. Regarding symptoms at presentation, patients may be asymptomatic or present cough and dyspnea [9,35]. Considering the asymptomatic presentation in most patients, this could draw attention to a subclinical course of the disease [5].
The characteristics of ILD include aberrant remodeling of lung tissue driven by an excessive synthesis and deposition of extracellular matrix as well as a defective repair of alveolar epithelial cell following chronic injury [15,36]. Vivero and Padera investigated different histological characteristics associated with ILD secondary to CTD. It seems that each CTD has a particular histological pattern. Using lung biopsy, a certain disease can be confirmed. In SLE-related NSIP, mononuclear or lymphoplasmacytic interstitial and peribronchiolar infiltrates were described [37].
Laboratory changes such as the presence of acute phase reactants in high titers, elevated double-stranded DNA (dsDNA) antibody levels, and a low level of complement should indicate a possible pulmonary involvement related to SLE [9]. Other antibodies such as anti-La, anti-Scl-70 and anti-U1RNP predispose to the development of ILD [9,37]. Vasser et al. confirmed the increased prevalence of auto-antibodies; for example, anti-La, anti-Sm, RNP, and lupus anticoagulant in their cohort [7]. Interestingly, Chen et al. found remarkably lower levels of anti-dsDNA antibodies and elevated levels of serum complement C3 in a cohort of patients with SLE-associated ILD [28].
In order to establish the diagnosis of SLE-associated ILD, an overlap syndrome should be firstly excluded. This possible cause can lead to a false diagnosis. Then, a high resolutioncomputed tomography (HR-CT) should be performed [9,38]. SLE-ILD is often clinically diagnosed, using frequently imaging methods such as HRCT and excluding other possible causes [39]. It is very important to make a correct differential diagnosis with: infections, tuberculosis, pulmonary oedema, cancers or even sarcoidosis [34,40].
Additional functional tests may reveal asymptomatic abnormalities, usually a restrictive dysfunction and a decrease in diffusing capacity for carbon monoxide (DLCO) [9,41].
At the onset of the disease, the radiological image may be normal. Bibasilar irregular linear opacities may also be found. Subsequently, it can be highlighted a diffuse or bibasilar infiltrates, pleural disease, or a honeycombing aspect [34]. Ground glass opacities were reported having a great frequency [19], followed by pulmonary consolidation, honeycombing aspect, or even traction bronchiectasis [42,43]. The identification on thoracic CT of two novel distinct fibrosis patterns (the island-like fibrosis and heterogeneous lung destruction signs) may be considered a starting point in differentiating patients with idiopathic pulmonary fibrosis from those with CTD-related ILD [44].
HR-CT is used to identify lung involvement and the specific pattern of disease [34]. Some studies reported NSIP as the most frequent pattern in SLE patients with ILD [45,46], although data are still controversial.

Management of ILD in SLE
Information on treatment in ILD-SLE is limited [2]. Some data from the literature are presented in Table 1. All patients with SLE should use hydroxychloroquine as their first-line treatment, according to the European League Against Rheumatism (EULAR). Glucocorticoids may also be added, if necessary. Immunosuppressive agents are indicated in refractory cases or even as the initial therapy in life-threatening cases [47].
Current treatments only reduce the progression of the disease [48]. The choice of a certain immunosuppressive treatment is based on the experience in scleroderma-associated ILD [49]. No placebo-controlled trials were performed in order to assess the efficacy and safety profile of corticosteroids and immunosuppressant drugs for the treatment of ILD secondary to SLE [34].
The therapeutic strategy is based on expert opinion [50]. It was recently agreed that corticosteroids should be used as first-line therapy associated with cyclophosphamide (CYC) or mycophenolate mofetil (MMF), followed by either rituximab or intravenous immunoglobulins (IVIG) [2,51]. More precisely, Koo et al. indicated, as first-line treatment for mild and moderate forms, the use of corticosteroid and azathioprine/MMF, drugs used also as a maintenance therapy; for refractory forms, high-dose steroids and a steroidsparing agent (for example, CYC) can be used [4,[50][51][52]. In most patients with SLE-ILD, systemic corticosteroids may normalize inspiratory vital capacity and DLCO, despite there being minimal evidence regarding their efficacy [53].
Currently, the impact of biological therapies on patients having various forms of ILD is being investigated [48]. Belimumab, a recombinant monoclonal antibody that targets the B-cell activating factor (BLys), is the first biological agent approved in SLE treatment [54]. Rituximab may be useful as second-line therapy [50,55]. Notably, patients having a severe and progressive disease may respond to rituximab, which may also prevent lung transplantation [56]. Intravenous immunoglobulin (IVIg) and plasmapheresis are used to treat refractory cases [39]. Antifibrotic treatments and lung transplantation should be properly considered for patients with fibrotic CTD-associated ILD and a progressive and severe disease course [2]. In SLE, it is mentioned that the clinical evolution of ILD is gradual and frequently stabilizes over time [9,53]. In terms of prognosis, it is not yet known whether ILD can be considered a negative factor in SLE patients. One study claimed that ILD is an independent risk factor for mortality. In another cohort, no significant differences in survival were found in patients with concomitant ILD [2,8,20].

ILD in SSc
Pulmonary involvement in SSc is frequent and ILD is considered one of the most common causes of death having a 10-year mortality up to 40% [65]. The incidence of ILD varies significantly between disease subtypes, but it is also influenced by the diagnostic method used or by disease duration [66]. SSc-ILD is most frequently defined by NSIP in up to 78% of the cases [67].
There are certain clinical characteristics that were associated with ILD presence. Patients with lcSSc (limited systemic sclerosis) and positive anticentromere antibodies (ACA) are less likely to present ILD compared to those having a dcSSc (diffuse systemic sclerosis), anti-Scl-70, and anti-topoisomerase I antibodies (ATA) [66]. Additionally, male gender, higher levels of mRSS (modified Rodnan Skin Score), the presence of gastroesophageal reflux disease, digital ulcers, and pulmonary hypertension were associated with the presence of ILD [68]. Additionally, the degree of skin involvement assessed by mRSS is considered to be a prognostic factor for progressive ILD, characterized by a FVC-predicted decline of more than 10% over a 12-month period [69]. ATA is also a prognostic factor for a decreased pulmonary function and severe ILD in patients with short disease duration [66,70].
ILD linked to SSc has a complicated and poorly known pathogenesis which includes fibrosis and various degrees of inflammation. As a consequence of innate and adaptive immune system inaccurate activation, fibroblasts are activated and myofibroblasts produce excessive extracellular matrix [65,66]. There are three main stages that lead to the development of ILD in SSc. In the early phase, there is a triggering event that occurs in a susceptible subject determined by genetic and epigenetic factors. Early inflammation leads to a pronounced pro-fibrotic phenotype fibroblasts and TGF-β activation. In the established phase the ongoing inflammation and failure of resolution processes determine an excessive matrix deposition. From this point forward, some participants will advance to stability and even regression and some will progress to a more severe form of ILD [67].
All SSc patients need to be screened for ILD at the time of SSc diagnosis and over the disease course due to the high prevalence of this complication. Part of the patients may describe typical symptoms such as dyspnea, non-productive cough, or fatigue. However, in some cases, there are no specific symptoms and, in these situations, screening for ILD is extremely important for an early diagnosis [71]. Screening and diagnosis are made with the following elements: clinical examination, pulmonary function testing (including forced vital capacity [FVC] and DLCO), HRCT, and chest X-rays [71,72]. However, the gold standard for diagnosis is HRCT, as it is more sensitive than X-ray [73].

Management of ILD in SSc
Although ILD is one of the most important complications in SSc, a definite treatment is not yet standardized. It is based mostly on an immunosuppressive strategy and a "wait and watch" approach [74,75]. Methotrexate, cyclophosphamide (CYC), or mycophenolate mofetil (MMF) can be used, although they do not have a very high efficacy and are associated with important adverse effects. Thus, a new treatment for ILD is highly needed; since 2019, various studies on rituximab, tocilizumab, and nintedanib were published. Following study findings, tocilizumab was authorized for SSc-ILD in March 2021 in United States and rituximab was approved in Japan in September 2021. Recently, nintedanib was also successfully approved for SSc-ILD in the United States, Europe, and Japan [74,76].
Nintedanib is a tyrosine kinases inhibitor previously approved for idiopathic pulmonary fibrosis following INPULSIS studies [77]. Its effect on pulmonary fibrosis was observed in both in vitro and in vivo studies, reducing bleomycin or silica pulmonary fibrosis, having a significant antifibrotic and anti-inflammatory action [78]. Moreover, the in vitro effect persisted in circulating fibrocytes from SSc patients, preventing their transition into myofibroblasts [79].
After nintedanib proved antifibrotic effects in patients with idiopathic pulmonary fibrosis, it was further investigated to see if the beneficial effects can be demonstrated in a fibrosis. Thus, the safety and efficacy of nintedanib was investigated in SENSCIS trial, conducted between 2015 and 2017. It was a randomized, double-blind, placebo-controlled trial that involved 576 patients with SSc; the patients were over 18 years old, with a history of Raynaud's phenomenon of at least 7 years prior to screening, from 32 countries. Patients were randomized 1:1 into a placebo group and a treated group receiving 150 mg of nintedanib twice daily. The extent of pulmonary fibrosis was diagnosed using HRCT with a mean extent of fibrosis of 36 ± 21.3%. In almost half of the participants, MMF was administrated at baseline. After 52 weeks, 232 patients from the nintedanib group and 257 from the placebo group completed the study. In the nintedanib group, the adjusted yearly change in FVC was 52.4 mL/year, while in the placebo group, it was 93.3 mL/year, proving a significantly suppressed reduction in FVC decline. Furthermore, in patients receiving MMF, the mean adjusted annual decrease in FVC was −40.2 mL in the nintedanib group compared to −66.5 mL in the placebo group. Either way, nintedanib proved its beneficial effects compared to placebo, and a more significant effect when combined with MMF [80]. Further on, many scientists performed extensive studies using the same cohort of patients from SENSCIS trial and demonstrated various results presented in Table 2.  Secondary: changes from baseline in mRSS and SGRQ scores after 52 weeks.
The decline in FVC from baseline was smaller for patients in the nintedanib group compared to placebo at 52 weeks and the effect persisted up to 100 weeks, similar between Japanese and non-Japanese participants.
The mRSS and SGRQ scores and the rate of adverse effects were similar between Japanese and non-Japanese patients.

Seibold et al. [83] 2020
The study included the same patients (n = 576) from the SENSCIS trial.
Primary: data about possible digestive (diarrhea, liver enzyme, and bilirubin elevation) and cardiovascular adverse effects.
The most common adverse effect was diarrhea in both groups (75.7% in the nintedanib group and 31.6% in the placebo group).
Almost half of these patients had the first episode in the first 30 days of treatment.
70.2% of them had one or two diarrhea episodes 94% of the cases being described as mild or moderate.
Secondary: the predisposition to develop intestinal side effect.
Antidiarrheal medication was prescribed in 48% of nintedanib patients and in 9% of placebo group.
Dose reduction was considered in order to maintain the continuity of the treatment in 40.6% of the nintedanib group participants.
Other digestive adverse events (nausea, vomiting, abdominal pain) or weight loss were more common in patients treated with nintedanib.
Primary: the annual decline in FVC rate (mL/year), the baseline changes in mRSS and SGRQ scores after 52 weeks Fewer Asian participants received MMF and methotrexate compared to non-Asian patients.
62 were from nintedanib group.
Secondary: adverse events developed over 52 weeks of treatment.
The benefit of slowing the FVC decline and the adverse events described were similar between Asian and non-Asian patients throughout the entire study.
Highland et al. [85] 2021 The study included the patients from the SENSCIS trial (n = 576).

Primary: the annual decline in FVC (mL/year) after 52 weeks of treatment
More than 90% of cases continued MMF treatment until week 52.
They were divided based on the treatment with MMF at baseline (median dose of 2000 mg): 139 from nintedanib group and 140 from placebo group.
Secondary: changes from baseline in mRSS and SGRQ scores after 52 weeks. The annual percentage rate of FVC decline, the absolute change in mL of FVC baseline the absolute and relative decrease in FVC of more than 5% predicted and more than 10% predicted were also evaluated.
Nintedanib usage was associated with a reduced rate of decline in FVC regardless of the association with MMF MMF association did not improve skin fibrosis or quality of life in both nintedanib and placebo groups and had a similar adverse event profile.

Maher et al. [86] 2021
The study included the same patients (n = 576) from the SENSCIS trial.
Primary: the absolute decrease/increase in FVC A decline in FVC% predicted was described in 55.7% of participants in nintedanib group and in 66.3% of participants in placebo group.
After 52 weeks, 13.6% of nintedanib group had an absolute decline in FVC between 5-10% predicted, while only 3.5% had an absolute decline in FVC between 10 and 15% predicted. In placebo group, the absolute decline in FVC and death were present in a higher proportion. Almost half of the patients from both groups were taking MMF.
The frequency of adverse events did not differ in SENSCIS-ON as compared to SENSCIS trial.
Fewer patients discontinued the treatment in SENSCIS-ON group.
The progression of FVC over 52 weeks was similar.
Primary: the annual decline in FVC (mL/year) after 100 weeks In both intent-to-treat and on-treatment analysis, the adjusted mean annual rate of decline in FVC after 100 weeks was slower in patients with nintedanib compared to placebo.
Secondary: the effect of nintedanib in patients who followed the treatment until the end of the study Denton et al. [89] 2022 The study included the same patients (n = 576) from SENSCIS trial.
Primary: the relationship between the degree of fibrotic ILD and the absolute rate of FVC decline in 52 weeks.
The degree of fibrotic ILD on HRCT at baseline and the decline in FVC predicted at 52 weeks were not related with the overall population or with subgroups of participants.
the extent of fibrotic ILD at baseline was not associated with the rate of decline in FVC over 52 weeks.
Secondary: the extent of fibrotic ILD and predicted FVC at baseline the higher the FVC predicted values at baseline the greater the decline in FVC during the trial, suggesting that a greater respiratory reserve leads to a higher decline in FVC.
nintedanib is beneficial irrespective of the fibrotic ILD extent on HRCT. Kreuter et al. [90] 2022 The study included the same patients (n = 576) from the SENSCIS trial.
Primary: annual decline in FVC (mL/year) over 52 weeks.
Baseline PRO measures scores were worse in dcSSc and correlated with a lower FVC, greater extent of fibrosis on HRCT, higher mRSS, presence of GERD, and other upper gastrointestinal symptoms in patients receiving MMF.
Additionally, it was linked with hospitalization during the study and with supplemental oxygen usage.
Secondary: changes from baseline in PRO measures questionnaires: SGRQ, FACIT-dyspnea, and HAQ-DI after 52 weeks.
After 52 weeks, in patients with FVC ≥ 70% predicted, PRO measures scores had better results compared with FVC < 70% predicted.
Kreuter et al. [91] 2022 The study included the same patients (n = 576) from the SENSCIS trial.
FVC decline rate in patients who were hospitalized for an all-cause event or SSc-related causes.
After 52 weeks, 13.7% of the participants experienced an all-cause hospitalization event or died, of which 7.4% were related to SSc.
FVC decline and time to first all-cause hospitalization or death were linked with statistically significant results after 52 weeks.
78 were hospitalized for all-cause events, 42 for SSc causes and 75 were admitted in ER or hospital followed by admission to ICU.
The risk of first hospitalization events related to FVC decline rate.
FVC decline was also correlated with time to first SSc-related hospitalization or death, yet no relationship was found between FVC decline and risk of hospitalization or death of any cause.
The reduction in FVC decline resulting after nintedanib treatment may reduce the risk of hospitalization.
Kuwana et al. [92] 2022 The study included the same patients (n = 576) from the SENSCIS trial divided after ATA status, baseline MRSS and SSc subtype. Primary: the annual decline in FVC over 52 weeks in the 3 groups.
Almost 60% of the patients were ATA-positive in both groups. Patients in the placebo group who were ATA-positive or negative had the same adjusted yearly rate of reduction in FVC.
In the nintedanib group there were 173 ATA positive patients versus 177 in the placebo group.
Compared to ATA-positive patients, nintedanib's effect on lowering the annual rate of decline in FVC was more pronounced in the ATA-negative patients.
There were 218 patients with mRSS < 18 in the nintedanib group and 226 in the placebo group.
Secondary: the correlations between FVC and MRSS at baseline versus week 52 and the decline rate of FVC considering MRSS at baseline a continuous variable.
The reduction in mRSS was similar between nintedanib and placebo groups regardless of the ATA status.
In the placebo group, the rate of decline in FVC over 52 weeks was similar in ATA positive and negative patients, higher in patients with a mRSS ≥ 18 at baseline compared to those with a mRSS < 18 and higher in patients with dcSSc compared to lcSSc. Nintedanib had no effect on reducing the mRSS in any of the subgroups. Maher et al. [93] 2022 The study included the same patients (n = 576) from the SENSCIS trial.
Primary: the decline in FVC at baseline and over 52 weeks in the two groups Participants from SENSCIS trial had a substantial impairment in FVC at baseline, with an FVC (mL) over 25% (around 950 mL) lower than it would be anticipated in those without lung disease.
As the average duration of SSc in the SENSCIS study was about 3.4 years, the predicted loss of lung function was about 280 mL/year since the onset of SSc.
They were compared to a matched population for age, sex, ethnicity and height.
Secondary: to estimate the "effective lung age" and compare the results to the real age After 52 weeks, the decline in FVC was four-fold greater in the placebo group, while in the nintedanib group it was only two-fold greater compared to hypothetical healthy references.
Volkmann et al. [94] 2022 The study included 574 patients from the SENSCIS trial with information on cough and dyspnea. 353 had both cough and dyspnea, 156 had only cough or dyspnea, 65 no complaints.
Primary: the annual decline of FVC over 52 weeks; the proportion of patients with relative/absolute decline in FVC > 5% and >10% at week 52; time to absolute decline in FVC >10% or death over 52 weeks in all the subgroups based on cough and/or dyspnea.
Most of cases reported coughing (76.6%) several days a week and some (23.4%) reported coughing a few days a month.
Cough was described in 229 patients from the nintedanib group and in 232 of the placebo group.
Secondary: changes in SGRQ scores at week 52.
In the dyspnea group, 70% reported it several days a week and 29.6% described dyspnea a few days a month.
Dyspnea was present in 209 of the nintedanib participants and 193 of the placebo participants.
Both cough and dyspnea were associated with a greater extent of fibrotic ILD, a lower FVC % predicted and a lower DLCO% predicted at baseline.
The rate of FVC decline after 52 weeks was similar in placebo patients regardless of the presence of cough or dyspnea.
In the nintedanib group, the effect of reducing FVC decline was more pronounced in those without cough or dyspnea, although no heterogeneity was observed between the subgroups.

Comparison between ILD SLE and SSc
As shown in Table 3, the prevalence of ILD in SLE differs from SSc. ILD is a rare manifestation in SLE, whereas in SSc, it occurs frequently up to 75%. The HLA genetic background was described in ILD-associated SSc, whereas in SLE, it is poorly understood. Several antibodies were described in relation to CTD, with some antibodies occurring in both diseases. NSIP is the most common manifestation of ILD in both SSC and SLE. It appears that pulmonary manifestation occurs a few years after CTD diagnosis. The clinical manifestation is also similar in these two diseases. It can range from asymptomatic to common pulmonary manifestations such as dyspnea or cough. For diagnosis and staging, both diseases require HT-CT. Treatment primarily involves the treatment of the underlying disease. In SLE, various therapies were tried, starting from corticosteroids to biological treatment or antifibrotic agents [9,65,95].

Pathology
An aberrant inflammatory response due to cytokine release; The impaired apoptosis and abnormal fibroblast proliferation leading to alveolar injury was described.
An injury to the alveolar epithelium, the vasculature, or both have been proposed as the initial event, followed by an aberrant immune response with fibroblast recruitment and activation. Extracellular matrix overproduction and an important scarring process replace the standard pulmonary architecture. The fibrotic variant of NSIP is more frequent than the cellular variant. Survival is not different between the two types of NSIP variants.

Conclusions
Many organs and systems can be affected in SLE and SSc. Although the major focus is on manifestations that lead to high mortality, lung damage is an important predictor of disease progression and prognosis. This article presented a comprehensive review of the current literature data on the management of ILD in SSc and SLE and information on the use of nintedanib in SSc. It gained significant attention in recent years for its potential therapeutic benefits.
Even if pulmonary fibrosis is a rare manifestation in SLE, prompt diagnosis of this complication is truly important. There is still ongoing controversy about the true prevalence of ILD in SLE, given the overlapping syndromes. Overall, the prognosis for patients with SLE-ILD is encouraging.
On the other hand, in SSc, ILD is a frequent manifestation associated with high morbidity and mortality. The prevalence is highly variable given the different diagnostic method used. There are limited data on the management and treatment of this complication, but new therapies seem promising, and nintedanib emerged as a treatment option for patients with ILD in SSc, improving lung function and slowing the progression of ILD. Despite the fact that it has a favorable safety profile, the optimal dosing and treatment duration remain to be evaluated through further studies.
In conclusion, nintedanib represents a significant advancement in the field of fibrotic diseases, and its integration into clinical practice has the potential to improve patient outcomes and quality of life in SS. On the other hand, in SLE cases, immunosuppressive treatment remains a valid option.