Advancements in mesenchymal stromal cells for treating connective tissue disease-associated interstitial lung disease: from basic research to clinical application

Abstract

Interstitial lung disease (ILD) is a common complication associated with connective tissue disease (CTD). It is characterized by a progressive decline in lung function, significantly affecting prognosis and potentially leading to respiratory failure or even death. The primary pathological mechanisms behind ILD include abnormal immune system activation and pulmonary fibrosis. Current treatment options primarily include immunosuppressive agents, antifibrotic drugs, and oxygen therapy to alleviate symptoms. However, these treatments have notable limitations, including individual variability and uncertain long-term effectiveness, which pose significant challenges in managing CTD-associated ILD (CTD-ILD).In recent years, Mesenchymal Stromal Cells (MSCs) have gained considerable attention as a promising therapeutic option for CTD-ILD. This is due to their immunomodulatory, antifibrotic, and tissue repair properties. Both clinical and preclinical studies have shown the therapeutic potential of MSCs in this context. However, this article also discusses the challenges and limitations of MSC-based therapies, highlighting the need for further clinical studies to validate their effectiveness.

Introduction

Connective tissue diseases (CTDs) are autoimmune disorders primarily involving connective tissues. These diseases include systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), systemic sclerosis (SSc), Sjögren’s syndrome (SS), and idiopathic inflammatory myopathies (IIMs). CTDs are progressive and can affect multiple organ systems, including the vasculature, bones, joints, skin, muscles, and internal organs. Pulmonary involvement, particularly interstitial lung disease (ILD), is a near-universal and prognostically critical manifestation across CTDs, strongly correlating with accelerated disease progression and elevated mortality. Among these pulmonary complications, ILD accounts for 25–30%, primarily affecting the lung interstitium. ILD is characterized by varying degrees of inflammation or fibrosis, leading to impaired gas exchange, progressive decline in lung function, and respiratory symptoms such as dyspnea. Severe cases can culminate in respiratory failure and death, significantly compromising patients’ quality of life [1]. CTD-associated ILD (CTD-ILD) is commonly seen in patients with IIMs, SSc, SS, RA, and SLE, with prevalence rates of 40%, 30–40%, 40%, 10%, and 12% [2], respectively. Recent cohort studies indicate that the mortality rates of dermatomyositis/polymyositis-associated ILD (DM/PM-ILD) and SSc-associated ILD (SSc-ILD) are 40–70% and 12.5%, respectively, significantly higher than those of patients without ILD, who exhibit 5-year survival rates exceeding 80% [3]. Currently, treating CTD-ILD involves addressing both the underlying CTD and the ILD itself. For CTDs, glucocorticoids and disease-modifying antirheumatic drugs (DMARDs) remain the cornerstone of therapy. However, long-term use of glucocorticoids and immunosuppressants increases the risks of infections, osteoporosis, and malignancies (approximately 1–3-fold), and the therapeutic response is often suboptimal. While some patients with CTD-ILD have stable disease, many others experience a heterogeneous disease course resembling idiopathic pulmonary fibrosis (IPF), ranging from slow to rapid progression. ILD can be divided into two phases: inflammatory and fibrotic. Unlike IPF, anti-inflammatory treatments can significantly improve ILD in its early inflammatory phase. However, once the fibrotic phase begins, the progression of fibrosis becomes difficult to reverse. In this context, novel antifibrotic agents such as pirfenidone and nintedanib have shown promise in slowing disease progression. Research indicates that pirfenidone inhibits the transformation of fibroblasts into myofibroblasts in RA-ILD lung specimens by reducing IL-6 and TNF-α levels, thereby attenuating pulmonary fibrosis [4, 5]. However, variability in efficacy across patients and the need for further investigation into its long-term effects remain challenges. Consequently, there is an urgent need to explore novel therapeutic approaches to improve the prognosis and quality of life for patients with CTD-ILD.

Mesenchymal stromal cells (MSCs), as defined by the International Society for Cell & Gene Therapy (ISCT), are mesoderm-derived stromal cells. Contemporary consensus discourages the term “Mesenchymal Stromal Cells” due to insufficient evidence for inherent stemness properties such as sustained self-renewal and multipotent differentiation. Instead, MSCs function primarily as paracrine signaling entities that coordinate immunomodulation and trophic support through microenvironmental modulation.MSCs can be isolated from diverse tissues, including bone marrow, umbilical cord, cord blood, placenta, adipose tissue, muscles, dermal tissue, amniotic fluid, menstrual blood, and urine. Their mechanisms of action include: (1) immune regulation through secretion of soluble mediators (transforming growth factor-beta(TGF-β), prostaglandin E2(PGE2), indoleamine 2,3-dioxygenase (IDO)) and direct cellular interactions that suppress pathological immune activation while promoting regulatory networks, (2) tissue homeostasis restoration via homing to injury sites, macrophage polarization toward pro-reparative phenotypes, releasing extracellular vesicles (EVs) carrying antifibrotic miRNAs (e.g., miR-29a-3p), and activating endogenous repair pathways (e.g., Wnt/β-catenin signaling) [6]and (3) metabolic reprogramming of local niches through mitochondrial transfer and nutrient sensing. Clinically applied since the mid-2000s for immune-dysregulated conditions, such as graft-versus-host disease and idiopathic pulmonary fibrosis, MSCs demonstrate efficacy rooted in cytokine-mediated immunomodulation and microenvironmental reshaping. In recent years, MSCs have emerged as a promising therapeutic candidate for CTD-ILD due to their ability to suppress aberrant immune responses and mitigate fibrosis, as well as the growing body of evidence supporting their efficacy.

Clinical characteristics and pathophysiological mechanisms of CTD-ILD

Shared clinical features and pathophysiological mechanisms of CTD-ILD

The most common clinical manifestation of CTD-ILD is dyspnea, which often presents insidiously and progresses gradually. Approximately 80–90% of patients experience varying degrees of dry cough, sometimes accompanied by small amounts of white, viscous sputum, while hemoptysis is observed in a minority of cases. In up to 80% of patients, lung auscultation reveals inspiratory crackles (commonly referred to as Velcro rales). Hypoxemia-related signs, such as tachycardia, cyanosis, and clubbing of the fingers, are also observed in certain patients [7]. Systemic manifestations are closely linked to the underlying CTD, such as skin sclerosis in SSc, joint pain in RA, and muscle weakness in IIMs. In advanced disease stages, continued progression of pulmonary fibrosis may result in severe restrictive ventilatory impairment, manifesting as reduced lung capacity and impaired gas exchange. This diversity in clinical manifestations contributes to the complexity and heterogeneity of CTD-ILD.

The chest’s High-resolution computed tomography (HRCT) is a critical imaging tool for diagnosing, classifying, assessing disease severity, and predicting prognosis in CTD-ILD. HRCT enables the detailed visualization of the extent, distribution, and specific features of interstitial lung abnormalities, facilitating the differentiation of various subtypes of ILD. Characteristic HRCT findings in CTD-ILD include interstitial changes, such as thickening of alveolar walls, reticular opacities, ground-glass opacities (GGO), traction bronchiectasis due to peribronchial fibrosis, honeycombing, and fibrotic remodeling. These features correlate with the extent of alveolar-interstitial injury and fibrosis progression, while their specificity varies among distinct CTD subtypes. The most common subtype of ILD in patients with IIMs, SSc, and SS is nonspecific interstitial pneumonia (NSIP) [8]. HRCT findings in NSIP include diffuse reticular opacities, primarily distributed in the basal regions of the lungs, patchy ground-glass opacities, and mild honeycombing. Compared to usual interstitial pneumonia (UIP), the honeycombing observed in NSIP is less pronounced, suggesting a lower degree of fibrosis and generally correlating with a relatively favorable prognosis. Conversely, most cases of RA-ILD exhibit a UIP pattern characterized by reticular opacities, traction bronchiectasis, and significant honeycombing predominantly involving the basal and peripheral regions of the lungs [9]. This UIP pattern in RA-ILD closely resembles the imaging features of IPF and is typically associated with advanced disease stages and poorer outcomes. Studies have demonstrated that median survival in RA-ILD patients with a UIP pattern is significantly shorter compared to those with non-UIP patterns (3.2 years vs. 6.6 years, respectively) [10]. Additionally, other less common subtypes of CTD-ILD include organizing pneumonia (OP), NSIP-OP overlap, and lymphocytic interstitial pneumonia (LIP) [11] (Table 1).

Table 1.

Comparative summary of key CTD-ILD subtypes and potential benefits of MSC therapy

Feature	IIMs-ILD	SSc-ILD	RA-ILD	SS-ILD
Clinical Features	Muscle weakness, rapidly progressive ILD (RP-ILD); poor prognosis in anti-MDA5 antibody-positive patients	Skin sclerosis, Raynaud’s phenomenon; higher risk in diffuse cutaneous SSc (dcSSc) and anti-Scl-70 antibody-positive patients	Joint destruction, high titers of RF/anti-CCP antibodies, iBALT-driven fibrosis	Dry mouth/eyes; lymphoid follicle formation and lymphocytic infiltration
Radiologic Patterns	Predominantly NSIP (basal reticulation, ground-glass opacities)	NSIP or UIP	Predominantly UIP (honeycombing), resembling IPF	Predominantly NSIP (diffuse reticulation, ground-glass opacities), rarely UIP or OP
Biomarkers	KL-6, SP-D, sCD40L, CXCL9/10, anti-MDA5 antibodies	IL-34, CXCL4, KL-6, LOX, SP-D	RF, anti-CCP antibodies, MMP-7, KL-6	KL-6, anti-SSA antibodies, lymphoid follicle-associated factors
Prognosis	Mortality 40–70% (RP-ILD); better survival in anti-ARS antibody-positive patients	5-year survival ~ 70%; rapid fibrosis progression with high FAP expression	5-year mortality 39%; median survival 3.2 years in UIP pattern	Lower mortality but high disability; poor prognosis
Potential Benefits of MSC Therapy	Mortality 40–70% (RP-ILD); better survival in anti-ARS antibody-positive patients	Inhibit fibroblast activation, reduce collagen deposition, and modulate M1/M2 macrophage polarization.	Suppress iBALT formation, downregulate TGF-β signaling, and enhance alveolar repair.	Inhibit lymphocytic infiltration, regulate B-cell function, and reduce lymphoid follicles.

NSIP: Nonspecific Interstitial Pneumonia, UIP: Usual Interstitial Pneumonia, OP: Organizing Pneumonia

A complex interaction of immune-mediated inflammation and fibrosis drives the pathogenesis of CTD-ILD. Chronic immune dysregulation in CTD patients is characterized by the overactivation of T cells and B cells, as well as the production of autoantibodies, which triggers an inflammatory response in the lung interstitium and damages alveolar epithelial cells and capillary endothelial cells. Persistent inflammation leads to the release of pro-inflammatory cytokines such as TGF-β, interleukin-6 (IL-6), and interleukin-13 (IL-13), which activate fibroblasts and promote their proliferation. These fibroblasts further secrete collagen and other extracellular matrix (ECM) components, resulting in alveolar wall thickening and irreversible fibrosis, which reduces alveolar elasticity and compliance, impairing gas exchange and ultimately leading to a decline in pulmonary function. Additionally, airway remodeling and microvascular abnormalities may develop in some patients. Airway remodeling is characterized by bronchial wall thickening and airway obstruction, while microvascular abnormalities, commonly seen in SSc-ILD, include pulmonary vasoconstriction, capillary loss, and aberrant angiogenesis. These changes exacerbate hypoxemia and contribute to disease progression. Furthermore, aberrant repair mechanisms such as epithelial-mesenchymal transition (EMT), which involves the transformation of alveolar epithelial cells into mesenchymal cells during the repair process, play a crucial role in the pathogenesis of CTD-ILD, further exacerbating fibrosis.

In the early inflammatory phase, CTD-ILD is primarily driven by lymphocyte activation, differentiation, and autoimmune responses that generate a pro-fibrotic microenvironment. Autoimmune processes in CTD activate and recruit immune cells, which release inflammatory mediators, including tumor necrosis factor-alpha (TNF-α), interleukin-1 (IL-1), and IL-6. These mediators initiate both inflammatory and fibrotic processes in lung tissue. Preclinical studies have shown that T cells are critical in promoting fibrosis [12]. Elevated B cell levels in the lung tissue of patients with IPF, RA-ILD, and SSc further suggest their involvement in the fibrotic progression of these diseases.

In the later fibrotic phase, activated fibroblasts differentiate into myofibroblasts, which produce ECM components and perpetuate the fibrotic process [13]. This leads to the establishment of a self-sustaining, pro-fibrotic feedback loop that drives progressive structural changes in the lung tissue. Myofibroblasts, key mediators of fibrosis, secrete collagen and other ECM proteins and originate from monocytes recruited to the lungs during fibrosis. These monocytes differentiate into fibroblasts or myofibroblasts, a process observed in both IPF and SSc. Neutrophils also play a multifaceted role in the fibrotic environment by secreting elastases and matrix metalloproteinases (MMPs), which degrade ECM and activate TGF-β, further driving ECM accumulation. Several soluble mediators contribute to fibrosis: IL-13, TGF-β, platelet-derived growth factor (PDGF), and C-C motif chemokine ligand 2 (CCL2). IL-13 promotes fibroblast-to-myofibroblast differentiation via the c-Jun N-terminal kinase (JNK) signaling pathway. TGF-β induces EMT and regulates the differentiation of fibroblasts into myofibroblasts, driving ECM deposition through signaling pathways such as MAPK, ERK, and PI3K/AKT. PDGF enhances ECM-related gene expression in fibroblasts, while CCL2 promotes fibroblast recruitment and differentiation, further contributing to excessive ECM deposition and fibrosis progression [14].

In addition to these cellular mechanisms, autoantibodies play a crucial role in the pathogenesis of CTD-ILD. For example, anti-topoisomerase I (Scl-70) antibodies are associated with both the presence and severity of ILD in SSc, while anti-SSA antibodies are linked to an increased risk of ILD in SS. These factors collectively contribute to the development and progression of fibrosis in CTD-ILD.

Heterogeneity in different types of CTD-ILD

IMs-ILD

IIMs are a heterogeneous group of autoimmune diseases that primarily affect skeletal muscles and are often associated with multisystem involvement. The most common subtypes include dermatomyositis(DM), anti-synthetase syndrome (ASA), and immune-mediated necrotizing myopathy (IMNM). ILD is the most frequent pulmonary complication of IIMs and significantly impacts patient prognosis. The prevalence of ILD in IIM patients is relatively high and varies across geographic regions. It is estimated that more than 40% of DM patients develop ILD, with prevalence rates of approximately 50% in Asia, 23% in North America, and 26% in Europe—interstitial lung disease [15]. With advances in diagnostic technology and increased awareness of the disease, the number of reported cases of IIMs-ILD has been rising. Compared to other CTD-ILD, IIMs-ILD demonstrates greater heterogeneity in clinical presentation, treatment response, and prognosis, which are primarily influenced by the antibody profile and disease subtype.

Studies have shown that myositis-specific antibodies (MSAs) and myositis-associated antibodies (MAAs) enhance the identification of ILD in IIM patients and provide guidance for clinical decision-making based on antibody profiling [16]. Among MSAs, anti-MDA5, anti-Jo-1, and anti-Ro-52 antibodies are all associated with an increased risk of ILD in IIM patients. The presence of anti-MDA5 antibodies is strongly linked to severe pulmonary manifestations, poor therapeutic response, and unfavorable prognosis, particularly in rapidly progressive ILD (RP-ILD). Conversely, patients with positive anti-aminoacyl-tRNA synthetase (anti-ARS) antibodies generally have better survival rates and respond more favorably to treatment [17, 18].

Although the pathogenesis of IIMs-ILD remains incompletely understood, multiple studies have identified potential mechanisms, including immune-mediated inflammation, fibrosis, and the dysregulation of specific biomarkers. In experimental autoimmune myositis (EAM) mouse models, the excessive formation of neutrophil extracellular traps (NETs) has been identified as a key pathogenic mechanism in IIMs-ILD, potentially contributing to inflammation and fibrosis [19]. A retrospective analysis of muscle biopsies from South Korea revealed that elevated HLA-DR expression on muscle fibers is closely associated with ILD in IIM patients [20]. Additionally, a cross-sectional study in China indicated that a decreased proportion of peripheral Th1 cells and an increased proportion of Th2 cells may play significant roles in developing and progressing IIMs-ILD [21]. Research has also demonstrated that elevated levels of interferon-gamma (IFN-γ)-induced chemokines, CXCL9 and CXCL10, can recruit proliferative M2 macrophages into the lungs, which subsequently promote pulmonary fibrosis through the secretion of TGF-β [22] (Table 1).

A study from Nanjing identified key mechanisms underlying the rapid progression of IIMs-ILD, including cytokine storms, coagulation abnormalities, and immune responses associated with anti-MDA5 antibody positivity. Biomarkers such as soluble CD40 ligand (sCD40L) and D-dimer were found to be correlated with RP-ILD [23]. Numerous studies have established that biomarkers play critical roles in the pathophysiology of IIMs-ILD and are associated with poor prognosis. Verified biomarkers include alveolar surfactant markers (Krebs von den Lungen-6 [KL-6]and surfactant protein D [SP-D]), inflammatory markers (chitinase-like protein 40 [YKL-40]), macrophage activation markers (soluble CD163 [sCD163]), and matrix metalloproteinase-7 (MMP-7) [22].

SSc-ILD

SSc is a rare but severe connective tissue disease characterized by excessive collagen deposition, vascular damage, and multi-organ fibrosis driven by autoimmune processes. Its clinical manifestations include skin sclerosis, Raynaud’s phenomenon, and visceral organ involvement. Among SSc patients, ILD is a common and potentially fatal complication. Approximately 80% of SSc patients develop ILD, with a higher prevalence observed in males, patients with diffuse cutaneous SSc (dcSSc), and those who test positive for anti-Scl-70 antibodies [24] (Table 1).

The pathogenesis of SSc-ILD is complex, involving a multifactorial interplay of fibrosis, autoimmunity, inflammation, and vascular injury. The fibrotic process begins with damage to alveolar epithelial and endothelial cells, triggering immune responses, the release of inflammatory mediators, and the abnormal activation of fibroblasts. Both genetic and environmental factors influence this process. Studies have demonstrated that genetic polymorphisms in the HLA region and abnormalities in genes associated with innate immunity, B-cell, and T-cell activation contribute to the increased susceptibility to SSc-ILD. Dysregulated gene functions may lead to immune hyperactivation, further driving fibrosis development and progression.

Significant progress has been made in identifying molecular markers for predicting and diagnosing SSc-ILD. These markers reflect the multifaceted characteristics of the disease, including inflammation, fibrosis, and cellular injury. For example, cytokines such as IL-34 and IL-6, chemokines such as CXCL4 and CCL18, and inflammatory proteins such as serum amyloid A (SAA) are significantly elevated in SSc-ILD patients. They may serve as predictive markers during the early stages of the disease. Other markers, including enzymes such as chitinase-1 (YKL-1) and lysyl oxidase (LOX), extracellular matrix proteins, and specific lung injury markers such as SP-D, carbohydrate antigen 153, and KL-6, have been identified as critical biomarkers [25].

In advanced stages, SSc-ILD is characterized by prominent pulmonary fibrosis, with persistent activation of fibroblasts and excessive accumulation of ECM. Fibroblast activation protein (FAP) is highly expressed in the lung tissues of SSc-ILD patients. It is a key driver of fibrosis by promoting ECM deposition, exacerbating tissue stiffness, and impairing pulmonary function. The aberrant fibroblast phenotype is closely associated with establishing a fibrotic microenvironment, which includes inflammatory cell recruitment, local hypoxia, and amplification of pro-fibrotic signaling [26].

RA-ILD

RA is a chronic autoimmune disease primarily characterized by progressive joint destruction. However, extra-articular manifestations, including ILD, significantly contribute to the disease burden. RA-ILD is one of the most common and challenging pulmonary complications, posing significant management challenges due to its complex pathogenesis and poor prognosis.RA-ILD patients often present with high titers of autoantibodies, including rheumatoid factor (RF) and anti-cyclic citrullinated peptide (anti-CCP) antibodies, suggesting that autoimmunity plays a critical role in disease initiation and progression. A hallmark of RA-ILD pathogenesis is the formation of inducible bronchus-associated lymphoid tissue (iBALT), driven by chronic immune activation and inflammation [10]. iBALT not only regulates immune responses but also secretes pro-fibrotic factors that activate lung fibroblasts, thereby promoting pulmonary fibrosis. This fibrotic process leads to progressive lung function decline and significantly impacts patients’ quality of life and survival (Table 1).

Genetic factors also play a critical role in RA-ILD development. Mutations in the RTEL1 and TERT genes, which regulate telomere length, have been associated with earlier onset of RA-ILD [14]. Telomere shortening exacerbates cellular senescence and fibrosis in lung tissues, accelerating disease progression. Furthermore, poorly controlled RA disease activity and higher arthritis severity are strongly associated with the onset of new ILD, highlighting the close link between systemic inflammation and RA-ILD.

RA-ILD is typically associated with poor clinical outcomes. Recent cohort studies have shown significantly higher mortality rates in RA-ILD patients compared to those without ILD, with 5-year and 10-year mortality rates reaching 39.0% and 60.1%, respectively, compared to 18.2% and 34.5% in non-ILD RA patients32. UIP is the most common radiological pattern in RA-ILD and is associated with shorter median survival compared to non-UIP patterns (3.2 years vs. 6.6 years) [27]. UIP in RA-ILD shares clinical and pathological features with IPF, including prominent fibrosis and honeycombing on imaging [28]. However, UIP in RA-ILD is characterized by more significant lymphocyte infiltration, which exacerbates inflammation and fibrosis. These patients have survival rates comparable to those with IPF but face limited and less effective treatment options, underscoring the need for targeted therapeutic strategies (Table 1).

SS-ILD

SS is a systemic autoimmune disease characterized by exocrine gland dysfunction and multi-organ involvement. ILD is one of the most common and severe pulmonary complications of SS, with a prevalence of approximately 20%. Among the pathological subtypes of SS-ILD, NSIP is the most frequent, accounting for 41–45% of cases, followed by UIP (10%) and OP (4%) [29]. Different pathological subtypes exhibit varying prognostic and disease progression characteristics, but overall, ILD significantly increases disease burden and mortality in SS patients.HRCT findings, such as extensive reticulation, and histopathological features, such as lymphoid follicle formation in lung biopsies, are strong predictors of disease progression and poor prognosis. Serum biomarkers, such as elevated KL-6 levels, are strongly associated with disease activity and mortality risk in SS-ILD patients, making them valuable tools for early disease detection and monitoring [30] (Table 1).

The pathogenesis of SS-ILD is driven by chronic inflammation and immune-mediated tissue damage. Persistent pulmonary inflammation leads to the secretion of aberrant cytokines and chemokines, such as TGF-β and IL-6, which activate fibroblasts and promote collagen deposition and interstitial fibrosis. Lymphocytic infiltration and lymphoid follicle formation are hallmark features of pulmonary pathology in SS-ILD, closely linked to systemic immune dysregulation and exocrine gland dysfunction. This chronic inflammatory and fibrotic process not only impairs lung function but also significantly increases disability and mortality risks.

Basic research on MSCs in CTD-ILD

The translational potential of MSCs in CTD-ILD hinges on their multifaceted modulation of the fibroinflammatory microenvironment. Immunomodulation—mediated through hierarchical interactions with immune cells via soluble factors, EVs, and epigenetic signals—constitutes the primary mechanism, resolving chronic inflammation while preserving host defense. Concurrent antifibrotic actions disrupt pathological TGF-β/Smad3 signaling and the transdifferentiation of myofibroblasts. Functional restoration of damaged lung tissue occurs via paracrine activation of endogenous repair pathways (e.g., HGF/KGF-mediated stimulation of alveolar progenitors and EV-delivered miRNA reprogramming), which collectively mitigate epithelial injury and optimize the reparative niche.

Potential mechanisms of MSCs in treating CTD-ILD

Immunomodulatory effects of MSCs in CTD

The immunomodulatory capacity of MSCs is critically relevant to CTD-ILD through their ability to rebalance the dysregulated immune microenvironment and concurrently target both inflammation and fibrosis [31]. For instance, MSC-derived PGE2 and tumor necrosis factor-stimulated gene-6 (TSG-6) polarize pro-inflammatory M1 macrophages toward an anti-inflammatory M2 phenotype, attenuating NLRP3 inflammasome activation and IL-1β-driven alveolar injury. Simultaneously, programmed death ligand 1 (PD-L1) expression on MSCs inhibits Th1/Th17 differentiation through PD-1 receptor engagement on T cells, while IDO-mediated tryptophan depletion expands regulatory T cells (Tregs), restoring immune tolerance [32]. These effects are further amplified by MSC-secreted IL-1 receptor antagonist (IL-1RA) and soluble TNF receptor-1 (sTNFR1), neutralizing key cytokine driving fibrosis [6]. Notably, the inflammatory CTD-ILD microenvironment itself enhances MSC immunoregulatory function through “licensing,” where cytokines like IFN-γ and TNF-α boost glycolytic flux and stabilize HIF-1α, increasing production of IDO and PGE2(Fig. 1).

Fig. 1.

Immunomodulatory Role and Antifibrotic Effects of MSCs in CTD-ILD: This figure illustrates the progression of CTD-ILD from the early inflammatory phase to the late fibrotic phase, highlighting the critical role of MSCs in remodeling the immune microenvironment. CTD triggers autoimmune responses. In the inflammatory phase, these responses activate immune cells that release pro-inflammatory cytokines (e.g., TNF-α, IL-1, IL-6), leading to alveolar injury and chronic inflammation.MSCs help restore immune balance by converting M1 to M2 macrophages via PGE2 and TSG-6, suppressing inflammasome activity and IL-1β-mediated damage. They also inhibit Th1/Th17 responses through PD-L1 and promote Treg expansion via IDO. Anti-inflammatory molecules, such as IL-1RA and sTNFR1, further reduce fibrosis-driving signals. During the fibrotic phase, MSCs counteract fibroblast activation and collagen deposition by releasing anti-fibrotic factors, such as HGF and VEGF, thereby regulating EMT and promoting tissue repair. This figure demonstrates the dual therapeutic potential of MSCs in targeting both inflammation and fibrosis in CTD-ILD

Furthermore, MSC-derived extracellular vesicles (MSC-EVs) emerge as central effectors of this paracrine activity, inheriting the immunomodulatory properties of parental MSCs while mitigating the risks associated with cell transplantation [33]. These heterogeneous lipid-bound nanoparticles (50–500 nm), encompassing exosomes, microvesicles, and apoptotic bodies, serve as natural carriers of bioactive cargo, including microRNAs (e.g., miR-29b, miR-21), anti-inflammatory cytokines (e.g., IL-10, TGF-β), and matrix-remodeling enzymes (e.g., MMPs, TIMPs). In CTD-ILD, MSC-EVs directly target lung fibroblasts and alveolar macrophages through surface adhesion molecules (e.g., integrins, tetraspanins), delivering miR-29b to suppress TGF-β/Smad3-driven collagen synthesis and TSG-6 to inhibit NLRP3 inflammasome activation, thereby concurrently attenuating fibrosis and inflammation. Studies demonstrate that small EVs derived from human umbilical cord MSCs (hUC-MSCs) can improve inflammatory conditions in collagen-induced arthritis (CIA) mouse models by modulating T-cell function, offering new insights into potential therapies for RA-ILD [34]. Sun et al. verified that transplantation of MenSC-derived small EVs significantly ameliorated bleomycin-induced pulmonary fibrosis by repairing alveolar epithelial cell injury in a mouse model in vivo and in vitro. Thus, MSCs and their secretome, particularly EVs, represent a multifaceted therapeutic strategy for CTD-ILD by simultaneously suppressing autoimmune inflammation, neutralizing profibrotic cytokines, inhibiting fibroblast activation, and promoting tissue repair.

Antifibrotic effects of MSCs in CTD-ILD

The antifibrotic effects of MSCs in CTD-ILD directly target fibroblast activation and EMT—key drivers of pulmonary fibrosis. Central to this process is the direct inhibition of TGF-β/Smad3 signaling—a master regulator of fibroblast activation and ECM deposition. MSCs secrete hepatocyte growth factor (HGF) and TSG-6, which competitively bind to TGF-β receptors on lung fibroblasts, blocking downstream phosphorylation of Smad2/3 and suppressing α-SMA-mediated myofibroblast differentiation [35]. Concurrently, MSC-derived PGE2 upregulates intracellular cAMP in fibroblasts, inhibiting collagen I/III synthesis via CREB-dependent transcriptional repression [36]. In SSc, a rare chronic autoimmune disease characterized by progressive fibrosis, MSC treatment exhibits anti-fibrotic effects, resulting in reduced skin and lung fibrosis in various SSc animal models.

MSCs secret IL-10 and CCL2 to recruit Tregs and M2 macrophages that actively degrade existing ECM via MMP-9/12 secretion while suppressing IL-13-producing Th2 cells—a key driver of fibroblast-to-myofibroblast differentiation [38]. Notably, MSC-educated macrophages exhibit reduced lysyl oxidase (LOX) expression, an enzyme critical for collagen cross-linking and mechanical stress-induced myofibroblast persistence [39]. At the same time, MSCs migrate to damaged lung tissues and exert antifibrotic effects by secreting bioactive factors such as IDO, PGE2, VEGF, IGF-1, HGF, TGF-β1, and IL-6. These factors suppress fibroblast proliferation, reduce collagen overproduction, and inhibit fibrosis progression [37] (Fig. 1).

MSCs-Evs amplify these antifibrotic effects through the precision delivery of non-coding RNAs. Exosomal miR-29b-3p directly silences TGF-βR1 and COL1A1 mRNA in lung fibroblasts, while miR-21a-5p disrupts glycolysis in activated myofibroblasts by targeting phosphofructokinase (PFKM), effectively “starving” fibrotic cells of energy substrates, thereby inhibits the transformation of fibroblasts into myofibroblasts.

Tissue repair capacity of MSCs in CTD-ILD

The tissue repair capacity of MSCs in CTD-ILD is orchestrated through synergistic paracrine signaling and microenvironmental reprogramming. Following intravenous administration, MSCs exhibit preferential homing to pulmonary capillary beds, where they locally release a repertoire of reparative molecules, including VEGF, IGF-1, and HGF. These factors collectively enhance alveolar epithelial regeneration—VEGF restores microvascular integrity, while HGF activates β-catenin signaling in type II pneumocytes, enhancing surfactant production and epithelial restitution [38] (Fig. 1).

Within fibrotic lungs, MSCs establish a pro-reparative niche by suppressing pro-inflammatory cytokines (e.g., IL-1β, TNF-α) and elevating anti-inflammatory mediators (e.g., IL-10). This immune reprogramming attenuates pathological remodeling and improves gas exchange, demonstrating therapeutic potential for CTD-ILD [39].

Animal studies on MSCs in treating CTD-ILD

Animal studies robustly demonstrate the therapeutic potential of MSC therapy in CTD-ILD, highlighting its anti-inflammatory, antifibrotic, and tissue-reparative properties, along with a favorable safety profile. Widely utilized models—including silica-induced silicosis, bleomycin-induced fibrosis, transgenic IPF, radiation-induced injury, and SSc-ILD—collectively recapitulate core hallmarks of human disease: sustained epithelial injury, dysregulated TGF-β/Wnt signaling, macrophage-driven inflammation, and progressive ECM deposition. In silica models where NLRP3 inflammasome activation dominates, adipose-derived MSCs (AD-MSCs) suppress epithelial apoptosis through Caspase-3 reduction and elevated Bcl-2/Bax ratios while mitigating fibrosis via inflammasome modulation [40]. Bleomycin models characterized by TGF-β-dominated epithelial-mesenchymal crosstalk reveal that hUC-MSCs disrupt fibrogenic feedback loops through the decrease of IL-6/IL-10/TGF-β reduction and attenuated M2 macrophage accumulation [41]. For transgenic IPF models featuring TGF-α-driven aberrant re-epithelialization, WJ-MSCs rebalance ECM homeostasis via AKT/MMP-2 suppression coupled with MMP-9/TLR-4 enhancement [42].

Complementing these findings, radiation-induced fibrosis models demonstrate olfactory mucosa-derived MSCs (OM-MSCs) reprogramming macrophage polarization from pro-inflammatory M1 to reparative M2 phenotypes, highlighting conserved macrophage plasticity across species [43]. Similarly, in autoimmune-mediated SSc-ILD models, hUC-MSCs reduce collagen deposition and alveolar wall thickening through dual immunomodulatory and fibrinolytic actions targeting vasculopathy [44].

Mechanistically, MSC interventions demonstrate their maximal efficacy during the early inflammatory phases, attenuating cytokine storms, inhibiting fibroblast activation, and ameliorating alveolar epithelial damage rather than reversing established fibrosis. These preclinical outcomes consistently converge on conserved therapeutic targets—particularly TGF-β signaling dysregulation, macrophage polarization dynamics, and EMT/ECM imbalance—establishing mechanistic coherence that strongly supports clinical translation of MSC-based disease-modifying strategies for CTD-ILD.

The therapeutic role of MSCs in CTD-ILD patients

Safety and efficacy of MSC therapy

Clinical evidence increasingly supports the safety and therapeutic potential of MSC-based therapies for interstitial lung diseases, though distinctions exist between IPF and CTD-ILD. For IPF, early-phase trials demonstrate consistent safety: a phase Ib trial of autologous adipose-derived stromal vascular fraction (ADSCs-SVF) reported no significant adverse events during acute or long-term follow-up, while intravenous BM-MSCs(20–200 × 10⁶ cells) showed no serious adverse events over 60 weeks. Beyond safety, efficacy signals include attenuated lung function decline (mean FVC decline: 3.0%; DLCO decline: 5.4%) and slowed disease progression. Bronchoscopic BM-MSCs reduced FVC decline rates (8% vs. higher controls), and placenta-derived MSCs halted fibrosis advancement [45, 46]. Similarly, a phase I trial of nebulized human umbilical cord MSC-derived extracellular vesicles (hUCMSC-EVs) in pulmonary fibrosis patients (N = 24) significantly improved lung function (FVC, MVV), respiratory quality of life (SGRQ, LCQ), and even showed fibrosis regression on CT in two advanced cases [47].

In CTD-ILD, evidence is more limited but promising. For systemic sclerosis (SSc-ILD), combined MSC-cyclophosphamide therapy improved skin scores (mRSS), stabilized lung function, and reduced fibrotic biomarkers (TGF-β/VEGF). Umbilical cord-derived MSCs achieved 92.7% 5-year survival with ILD stabilization/improvement in 72% of patients and no treatment-related adverse events [48, 49] (Table 2). A retrospective cohort study further confirmed that mesenchymal stem cell transplantation (MSCT) significantly improved survival in SSc patients, particularly those with pulmonary hypertension (PAH) and ILD. These outcomes suggest MSC therapy may simultaneously target autoimmune dysregulation and fibrotic progression in CTD-ILD [50].

Table 2.

Clinical trials of MSCs treatment in CTD-ILD patients

Institution	ClinicalTrials.gov ID	Title	Intervention	Phase	Conclusion
Mayo Clinic, USA	NCT03929120	Allogeneic Bone Marrow MSCs (BMD-MSCs) Treatment of CTD-ILD	IV, BWD-MSCs (0.5-1 × 10^6 cells/kg)	I	None
Xijing Hospital, China	NCT03798028	MSC Treatment for Moderate to Severe RA: A Multicenter RCT	IV, UC-MSCs (1 × 10^6 cells/kg)	NA	None
Sabana University, Colombia	NCT04432545	Allogeneic WJ-MSC Infusion for Refractory Diffuse Skin SSc with Lung Involvement	IV, WJ-MSCs (2 × 10^6 cells/kg)	NA	None
Federal Research Clinical Center, Russia	NCT02594839	Safety and Efficacy of Allogeneic BM-MSCs in Rapidly Progressive ILD Patients	IV, BM-MSCs (2 × 10^9 cells/kg)	I-II	Safety
The Affiliated Drum Tower Hospital of Nanjing University Medical School, China	NCT00962923	Allogeneic Mesenchymal Stromal Cells Transplantation for Systemic Sclerosis (SSc)	IV, cyclophosphamide and (1 × 106 cells/kg )	I-II	Safety, improvement, or stability in HRCT images was observed in 72.0% of ILD patients. The overall 5-year survival rate was 92.7% (38 out of 41 patients).

BM-MSCs bone marrow-derived mesenchymal stromal cells, UC-MSCs umbilical cord blood-derived mesenchymal stromal cells, WJ-MSCs Wharton’s Jelly-derived mesenchymal stromal cells, MSCT mesenchymal stromal cells transplantation

Tissue Source-Dependent therapeutic profiles of MSCs

The therapeutic impact of MSCs in CTD-ILD varies significantly by tissue source, reflecting distinct functional profiles.BM-MSCs excel in anti-inflammatory cytokine secretion (e.g., IL-10) but are limited by low yield, favoring early-stage inflammatory disease. hUC-MSCs exhibit rapid proliferation and robust antifibrotic activity via HGF/TSG-6, which is ideal for advanced fibrosis. Conversely, AD-MSCs promote angiogenesis through VEGF, which may worsen late-stage fibrosis but benefit vascular-deficient subtypes like SSc-ILD. Clinical selection should align source strengths with the disease phase: BM-MSCs for inflammatory conditions, hUC-MSCs for fibrotic conditions, and AD-MSCs for vascular repair. Standardizing MSC potency remains critical to address donor- and batch-dependent variability [47].

Innovations in clinical trial design

To enhance MSC therapy for CTD-ILD, future large-scale randomized trials must implement stratified designs that delineate patients by specific CTD-ILD subtypes and disease activity. Primary endpoints should incorporate objective physiological and radiological assessments (e.g., FVC, DLCO, HRCT fibrosis scores). In contrast, secondary endpoints track disease-specific biomarkers (e.g., KL-6, SP-D, MMP-7) to monitor lung injury and fibrosis. Furthermore, rigorous and long-term safety monitoring, particularly for rare adverse events such as tumorigenicity, remains paramount. This biomarker-driven patient stratification is crucial for optimizing treatment efficacy; for instance, early-stage patients presenting with elevated inflammatory markers, such as IL-6 and C-reactive protein (CRP), are hypothesized to benefit most from the potent immunomodulatory capabilities of MSCs. Conversely, individuals with advanced fibrosis, characterized by extensive tissue remodeling, may require more targeted interventions, such as MSC-derived extracellular vesicles (MSC-EVs) specifically engineered and loaded with antifibrotic agents like miRNA-29, aimed at directly mitigating profibrotic pathways.

Conclusions and future perspectives

Growing attention has been drawn to the applications of MSCs in fundamental and clinical research, which has led to numerous studies focusing on new areas of MSC application and MSC-sourced bioactive factors to achieve optimal outcomes. However, their clinical translation faces significant challenges, primarily due to the scarcity and limitations of existing clinical evidence. Dedicated CTD-ILD trials remain notably sparse compared to IPF studies, and those conducted often suffer from small sample sizes, inconsistent endpoints, and lack of standardization—factors that compromise statistical power and obscure definitive efficacy signals. This challenge is compounded by profound disease heterogeneity: CTD-ILD encompasses diverse subtypes (e.g., SSc, RA, IIMs) with distinct immune-fibrotic drivers, yet most trials inadequately stratify populations or focus narrowly on SSc-ILD. MSC product variability further complicates outcomes, as differences in tissue source (bone marrow, umbilical cord, adipose), isolation methods, and culture conditions lead to unpredictable therapeutic potency and reproducibility.

Delivery hurdles also persist—systemic administration risks off-target effects and poor lung retention, while MSC-EVs exhibit inefficient pulmonary targeting after intravenous infusion. Host-microenvironment interactions introduce additional unpredictability, as the immunosuppressed state of CTD-ILD patients and dynamic immune cell responses (e.g., Treg/Th17 balance, macrophage polarization) may compromise the efficacy of MSCs isolation methods and culture conditions yield unpredictable potency. Unresolved long-term safety concerns (e.g., tumorigenicity, pro-fibrotic risks) in small-scale trials necessitate robust phase II/III studies featuring standardized protocols, subtype stratification, and rigorous longitudinal monitoring to translate mechanistic promise into clinical benefit. Notably, MSCs exhibit dual fibrosis roles: resident MSCs (LR-MSCs) adopt pro-fibrotic phenotypes (e.g., α-SMA + myofibroblast differentiation exacerbating ECM deposition), while exogenous MSCs suppress fibrosis. This paradox underscores the need for microenvironment-specific conditioning (e.g., genetic engineering, cytokine priming) to direct regenerative activity [51, 52].

Future advancements in MSC-based therapies for CTD-ILD will likely focus on three transformative approaches: induced pluripotent stem cell-derived MSCs (iMSCs), engineered MSCs, and MSC-EVs. iPSC technology offers a scalable solution to donor-dependent heterogeneity by generating standardized, clinically grade iMSCs with consistent anti-fibrotic and immunomodulatory profiles, circumventing limitations of primary MSC sources. Engineered MSCs offer a robust solution by enabling the modification of MSCs to express targeted anti-fibrotic factors (e.g., TGF-β antagonists, MMPs) or immunomodulatory cytokines (e.g., IL-10), thereby amplifying their intrinsic capacity to suppress pro-inflammatory cytokines (TNF-α, IL-1β, IL-6) and reduce extracellular matrix accumulation. Concurrently, biomaterial innovations, such as hydrogel microcapsules, optimize MSC retention and paracrine activity, while RNA interference strategies achieve precision gene silencing in fibrotic niches. Crucially, engineered extracellular vesicles (EVs) derived from MSCs—particularly optimized for pulmonary delivery via size control and surface modification—emerge as superior cell-free alternatives, combining low immunogenicity, enhanced stability, and efficient cargo delivery (e.g., miR-29b, miR-186) to concurrently modulate multiple fibrotic pathways (TGF-β/Smad, Wnt/β-catenin) and promote myofibroblast dedifferentiation [53]. Further refinements include EV functionalization with hyaluronic acid to boost anti-inflammatory uptake or miRNA loading (e.g., miR-486-5p) to reverse fibrotic gene signatures [33]. Complementing this, lipid nanoparticles (LNPs) offer tunable platforms for targeted drug co-delivery, exemplified by inhaled nintedanib-LNPs achieving 8000-fold higher lung concentrations than oral administration, or dexamethasone-loaded liposomes selectively reprogramming M2 macrophages. The convergence of these approaches—such as MSC-EVs loaded with nintedanib (MSC-Exo-Nin)—demonstrates synergistic attenuation of fibrosis by simultaneously suppressing TGF-β signaling and oxidative injury, outperforming monotherapies. Thus, integrating engineered MSC products (genetically modified cells, optimized EVs, or hybrid LNPs) with established antifibrotic agents offers a promising paradigm: leveraging MSC plasticity to enhance drug bioavailability and multi-pathway engagement while mitigating systemic toxicity, ideally deployed early to halt inflammation or as adjuncts to standard care in advanced disease [54].

Ultimately, large-scale, randomized controlled trials are indispensable for verifying the long-term efficacy, safety, and cost-effectiveness of treatments. Concurrently, the field must converge on harmonized protocols for cell manufacture, quality control, and biomaterial-assisted delivery. Close collaboration among stem-cell biologists, materials scientists, and clinicians will be essential for resolving mechanistic questions, refining delivery technologies, and keeping patient-centric design at the forefront. When these interconnected challenges are met, MSC-based therapies are poised to evolve from experimental interventions into accessible, affordable, and reproducible options for individuals with CTD-ILD and other fibrotic lung diseases.

Abbreviations

Abbreviation

Full Form

MSC

Mesenchymal Stromal Cells

CTD

Connective Tissue Disease

ILD

Interstitial Lung Disease

RA

Rheumatoid Arthritis

SSc

Systemic Sclerosis

SLE

Systemic Lupus Erythematosus

NSIP

Nonspecific Interstitial Pneumonia

UIP

Usual Interstitial Pneumonia

OP

Organizing Pneumonia

MMP

Matrix Metalloproteinase

TGF-

Transforming Growth Factor Beta

PGE2

Prostaglandin E2

VEGF

Vascular Endothelial Growth Factor

IGF-1

Insulin-Like Growth Factor 1

hUC-MSC

Human Umbilical Cord-derived Mesenchymal Stromal Cells

BM-MSC

Bone Marrow-derived Mesenchymal Stromal Cells

hMSC

Human Mesenchymal Stromal Cells

P-MSC

Placenta-derived Mesenchymal Stromal Cells

TGF-1

Transforming Growth Factor Beta 1

sCD40L

Soluble CD40 Ligand

KL-6

Krebs Von den Lungen-6 (a biomarker)

SP-D

Surfactant Protein D

CD163

Cluster of Differentiation 163

FVC

Forced Vital Capacity

DLCO

Diffusing Capacity of the Lung for Carbon Monoxide

FDA

Food and Drug Administration

IFN-

Interferon Gamma

NK

Natural Killer (cells)

Tregs

Regulatory T cells

M2 Macrophages

M2 Phenotype Macrophages

PD-L1

Programmed Death Ligand 1

EVs

Extracellular Vesicles

TNF-

Tumor Necrosis Factor Alpha

PDGF

Platelet-Derived Growth Factor

MAPK

Mitogen-Activated Protein Kinase

NF-B

Nuclear Factor Kappa-light-chain-enhancer of activated B cells

TLR-4

Toll-Like Receptor 4

EMT

Epithelial-Mesenchymal Transition

AKT

Protein Kinase B

Authors’ contributions

Jin-fang Gao and Yan-li Yang conceptualized the manuscript. Jiao Yuan wrote it, and Jiao Yuan, Ya-Lin Zhao, Yu-wei Xu, and Lei Wang edited it.

Funding

This work was supported by grants from the Shanxi Province Clinical Research Center for Dermatologic and Immunologic Diseases (Rheumatic diseases) Scientific Research Project (LYZX-202302), the Shanxi Provincial Administration of Traditional Chinese Medicine Scientific Research Topics program mission statement (2024ZYYA014), the Key Research and Development Program of Shanxi Clinical Medical Research Center for Dermatology and Immunological Diseases(2023ZYYA2007), the Shanxi Bethune Hospital Institutional Fund Project(2022YJ02), and Shanxi Provincial Department of Science and Technology Project (202403021211172).

Data availability

Data sharing is not applicable to this article, as no datasets were generated or analyzed during the current study.

Competing interests: The authors declare no conflicts of interest.

Declerations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.