SLE is a chronic autoimmune disease with multiorgan and multisystem involvement, in which lung damage is an important predictor of disease progression and prognosis. Lung involvement in SLE is far more common than in other CTDs, and up to 50–70% of SLE patients may have lung damage [20]. The clinical manifestations of lung damage vary widely in patients with SLE, ranging from a small amount of asymptomatic pleural effusion to life-threatening alveolar hemorrhage [20]. Therefore, early diagnosis and treatment of SLE lung damage directly affect the prognosis of patients.
Narvaez et al. calculated the prevalence of primary respiratory diseases in a retrospective cohort of patients with SLE, including 3,215 patients, the most common being pleural diseases (21%), followed by lupus pneumonia (3.6%), pulmonary thromboembolism (2.9%), primary pulmonary hypertension (2.4%), and ILD (2%) [8]. Kamen et al. showed that in SLE patients with lung symptoms, some common proinflammatory cytokines were 2–3 times higher than those in SLE patients without lung symptoms, including interferon-γ (IFN-γ), tumor necrosis factor (TNF)-α, and interleukin (IL)-6. In addition, the IL-8/IL-10 ratio was three times higher than that in patients with pulmonary symptoms [21–22]
ILD is characterized by abnormal remodeling of lung tissue due to excessive synthesis and deposition of the extracellular matrix as well as defects in alveolar epithelial cell repair after chronic injury [23]. ILD mainly involves the alveolar space and interstitial tissue, eventually leading to the loss of alveolar-capillary functional units [24]. The main pathological features are the accumulation of inflammatory cells, proliferation of fibroblasts, and excessive production and abnormal deposition of extracellular matrix in the interstitial tissue, leading to dysfunction of lung gas exchange, dyspnea, respiratory failure, and even death [25].
Some studies have reported that non-specific interstitial pneumonia (NSIP) is the most common manifestation in patients with SLE [26, 27]. The clinical features of SLE-ILD have been reported to include older age, longer illness duration, increased ratios of Raynaud's phenomenon, moist rales, and tachypnea [28]. In addition, studies have reported laboratory changes suggesting ILD may be associated with SLE. In a group of patients with SLE-ILD, anti-double-stranded deoxyribonucleic acid (dsDNA) antibody levels were significantly reduced and serum complement 3 (C3) levels were elevated [28], whereas other antibodies such as anti-cyclic citrullinated peptide (CCP) and anti-ribonuclear protein (RNP) were predisposed to ILD [29]. Table 3 summarizes current data from relevant studies of SLE-ILD, including recommended treatments [30].
Table 3
Current research reports on SLE-ILD.
Prevalence | Pathogenesis [20,21] | Clinical features [28] | Laboratory features [28,29] | Treatment [30] |
3–9% [5] | IFN-γ | elderly age | anti-dsDNA↓ | corticosteroids with CYC/MMF |
29% [9] | TNF-α | longer illness duration | serum C3↓ | rituximab |
| IL-6 | increased ratios of Raynaud's phenomenon | anti-CCP/anti-Scl-70/anti-RNP (+) | IVIG |
| IL-8/IL-10 | moist rales | | |
| | tachypnea | | |
*SLE: systemic lupus erythematosus; ILD: interstitial lung disease; IFN: interferon; TNF: tumor necrosis factor; IL: interleukin; dsDNA: double-stranded deoxyribonucleic acid; C3: complement 3; CCP: cyclic citrullinated peptide; Scl-70: topoisomerase I; RNP: ribonucleoprotein; CYC: cyclophosphamide; MMF: mycophenolate mofetil; IVIG: intravenous immunoglobulins.
GWAS uses SNPs in the human genome as molecular genetic markers to conduct genome-wide control or association analyses and find genetic variants affecting complex traits by comparison. Mendelian randomization based on gamete formation to follow the "parental alleles are assigned randomly to progeny" Mendel's random distribution law theory, using GWAS data analysis between different exposure and the results of cause and effect, so the relationship between genes and disease is not subject to the interference of the common confounding factors such as environment, make the effect estimate is closer to the real situation [31–32]. MR Analysis effectively overcomes the limitations of traditional randomized controlled studies and has unique advantages in inferring causal associations between phenotypes and diseases.
In this study, we used GWAS databases for SLE and ILD, and MR analysis was used to explore the potential causal relationships. Large-scale GWAS data were used to provide powerful and reliable instrumental variables, and sensitivity analysis was used to confirm causality inference. There was no heterogeneity or horizontal pleiotropy among the instrumental variables, and no single SNP had a large effect on the overall results. This study has certain limitations: first, GAWS data in this study were limited to East Asian populations, and further exploration is needed in other ethnic populations. Second, the number of SNPs strongly associated with ILD was small. At the same time, more clinical and basic research are still needed to clarify the pathogenesis and treatment of the two diseases.