Bronchopulmonary dysplasia and prematurity-associated lung disease: emerging therapies to improve lifelong respiratory outcomes

Abstract

Survival of extremely preterm infants has improved owing to advances in perinatal and neonatal care. However, bronchopulmonary dysplasia (BPD) has remained unabated, and research into preventive measures for the broader spectrum of prematurity-associated lung disease (PLD) has met with limited success. We summarize emerging therapeutic strategies with the potential to favorably influence the lifelong trajectory of lung health of preterm infants, especially the most immature. This narrative review examines recent preclinical and clinical studies on therapies for BPD prevention, focusing on mechanism-based interventions, biologic strategies, and innovative translational tools. Preclinical evidence converges on actionable targets in alveolarization, inflammatory pathways, redox and mitochondrial homeostasis, vascular and matrix remodeling. Novel molecules and drug-repurposing strategies are being tested in animal models, while patient-derived platforms may enable personalized therapeutic approaches. Mechanism-based approaches are progressing through Phase 1–2 clinical trials; pending efficacy analyses, insulin-like growth factor-1 replacement may become the first pharmacological therapy specifically approved to reduce or attenuate BPD. In parallel, cell-based therapies and extracellular vesicle strategies offer a complementary paradigm, acting through pleiotropic biological reprogramming of the injured lung. Ethical-regulatory complexity and the challenges of standardizing and scaling cell-based and cell-free therapies may limit clinical translation.

Conclusion: A growing repertoire of therapies for BPD holds promise for modifying early respiratory illness and the long-term burden of PLD. Realizing this potential will benefit from both mechanism-informed combinatorial approaches and pleiotropic strategies, supported by patient-derived translational tools and adequately powered efficacy evaluations. Equally essential are tailored neonatal-first regulatory pathways linking the continuum of BPD to the later respiratory morbidity of PLD, a life-course condition with significant public health relevance beyond the neonatal period.

What is Known: • Bronchopulmonary dysplasia (BPD) is the current clinical and regulatory endpoint for neonatal pulmonary therapies. • BPD definitions capture the most severe manifestations of prematurity-associated lung disease (PLD), a broader continuum of respiratory morbidity reaching into adulthood; attenuating BPD may improve the whole spectrum of PLD.

What is New: • Therapies targeting alveolar, inflammatory, oxidative, vascular, and remodeling pathways are in clinical development; pending late-stage trials, IGF-1 replacement may become the first disease-modifying therapy to reduce BPD. • Novel targets, experimental platforms, and repurposed drugs hold promise for precision therapies in preterm infants.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1007/s00431-026-07366-8.

Introduction

Preterm birth remains a major global health challenge, affecting approximately 10% of live births worldwide and representing a leading cause of neonatal morbidity and mortality [1]. Advances in perinatal and neonatal care, including antenatal corticosteroids, surfactant replacement therapy, and refined respiratory support strategies, have substantially improved survival, particularly among extremely low gestational age neonates (ELGANs, born below 28 weeks of gestation). However, a survival improvement at the more severe end of the spectrum of prematurity has been accompanied by an increased burden of chronic respiratory morbidity, both in childhood and adulthood [2–5]. So far, the need for oxygen and respiratory support during the neonatal intensive care unit admission has been variably integrated to diagnose and grade bronchopulmonary dysplasia (BPD) [6, 7]. This disease entity has received the greatest epidemiological and research attention, and has long been considered the main determinant of long-term pulmonary outcome in preterm infants, thereby becoming the current clinical and regulatory endpoint for pulmonary therapies; yet it has itself gone through profound changes from its initial description to contemporary times [8, 9]. While the “classic” BPD was characterized by severe airway injury, fibrosis, and emphysematous changes, the “new” BPD reflects an arrest of lung development and is typical of extreme prematurity, where impaired alveolarization and dysmorphic pulmonary vasculature predominate over other abnormalities [10]. However, in recent years, there is growing recognition that even mild degrees of prematurity can alter lifelong respiratory health [11]. These observations underscore how respiratory disease in preterm infants is itself a multifaceted entity, shaped by the interplay of distinct injurious exposures. This argues for moving beyond a purely neonatal definition of BPD toward endotypes and clinical phenotypes linked, across the lifespan, to healthcare needs, prolonged oxygen therapy, recurrent hospitalization, wheezing disorders, pulmonary hypertension, and the adult development of early-onset chronic obstructive pulmonary disease (COPD)-like phenotypes [12]. For these reasons, the framework of prematurity-associated lung disease (PLD) has been conceptualized to collectively describe the heterogeneous respiratory consequences of prematurity from birth to adulthood and to unify efforts towards healthier respiratory outcomes for all preterm infants [13].

Importantly, PLD can present with a range of multisystem comorbidities; these conditions reflect overlapping perinatal insults and shared pathogenic pathways, with adverse consequences on overall health that may translate into a substantial individual and societal burden of morbidity as these patients reach adulthood and older age [14, 15]. Interventions targeting lung protection, inflammation, oxidative stress, vascular development, and tissue regeneration may have lasting effects on lung structure and function, and potentially extend their benefit to other organ systems involved through shared mechanisms. However, while many therapies show promise in reducing early respiratory morbidity, robust evidence demonstrating sustained improvements in long-term outcomes remains limited. Notably, targeted single-axis interventions have not translated into clear benefit [16]. Taken together, these observations call for strategies addressing the full mechanistic complexity of PLD, with the potential to reshape the respiratory developmental course of preterm-born individuals.

This review aims to summarize emerging treatments in preterm infants that may influence long-term pulmonary outcomes. By integrating pathophysiological insights with clinical evidence, we seek to identify which interventions have the potential not only to improve survival and short-term respiratory stability but also to favorably modify the lifelong trajectory of lung health in this vulnerable population.

Pathophysiological background

Lung development proceeds through highly regulated stages-embryonic, pseudoglandular, canalicular, saccular, and alveolar-culminating in the coordinated development of alveolar structures and the pulmonary capillary network required for efficient gas exchange. Extremely preterm birth typically occurs during the late canalicular or early saccular stage of lung development, when alveolarization and microvascular maturation remain incomplete [10, 17]. Prenatal factors, such as intrauterine growth restriction, may already disrupt critical developmental processes during fetal life and are frequently associated with preterm birth itself. Acting on this immature lung, many possible extrauterine noxae can trigger additional pathogenetic pathways [13].

While respiratory support is often lifesaving in preterm infants, ventilation and supplemental oxygen can amplify lung injury. Ventilator-induced lung injury (volu-baro-atelecto- and bio-trauma) can activate inflammatory cascades that interfere with alveolar and vascular maturation [18]. Similarly, hyperoxia promotes oxidative stress due to immature antioxidant defenses, leading to epithelial and endothelial damage and disruption of growth factor signaling [19]. Inflammation represents a unifying mechanism linking mechanical and oxidative injury. Proinflammatory cytokines that impair septation and angiogenesis, lack of surfactant, inefficient antioxidant mechanisms, lower compliance, and inadequate fluid clearance may perpetuate a vicious cycle of persistent early inflammatory activation that contributes to airway remodeling and long-term airflow limitation [20]. Beyond these pathways, three interconnected axes of extrapulmonary influences target the developing lung. The first is the gut-lung axis: early antibiotic exposure and gut dysbiosis appear to shape inflammation and immune training, with consequences for BPD risk in observational data [21–24]. The second is neuro-pulmonary signaling. Mechanical ventilation, repeated airway management, and hyperoxia appear to sensitize vagal afferents, and the resulting cholinergic drive modulates pulmonary inflammation and airway innervation. This frontier remains hypothesis-generating and awaits preclinical confirmation [25]. The third is metabolic reprogramming: perinatal insults skew nutrient sensing through mTOR and AMPK and shift substrate utilization; metabolic signatures include mitochondrial dysfunction, oxidative stress, and altered purine and amino acid metabolism, and remain detectable into adulthood [14, 26, 27]. Nutrient availability intersects this last axis directly. Nitric oxide promotes vasodilation, angiogenesis, and alveolarization; with preterm infants frequently suffering arginine deficiency, its endogenous synthesis is limited in this population [28]. Together, these factors highlight how early exposures can permanently alter lung development and influence lifelong respiratory trajectories.

Emerging treatments

Insulin-like growth factor-1 (IGF-1)

Insulin-like growth factor-1 (IGF-1) is the principal fetal growth factor supporting coordinated development of lung, brain, and eye during the last trimester of pregnancy. It is supplied predominantly through the placenta until approximately 30 weeks of postmenstrual age (PMA), when the fetal liver progressively acquires the capacity to produce it [29, 30]. In the developing lung, IGF-1 drives alveolar epithelial proliferation and differentiation and induces vascular endothelial growth factor (VEGF) transcription, thereby coupling alveologenesis to microvascular growth, while attenuating transforming growth factor-β (TGF-β) signaling and the fibroproliferative response to neonatal injury [29, 31]. Preclinical evidence confirms the essential nature of the IGF-1 axis as a driver of lung development: IGF-1R-null mice suffer pulmonary hypoplasia and die of respiratory failure [32]. Despite strong evidence for inflammatory involvement in PLD, the fundamental deficit in preterm lung disease may lie in interrupted development: alveolarization, microvascular growth, and maturational trajectories are truncated at birth, often further disrupted by antenatal insults [33]. Accordingly, anti-inflammatory strategies per se—including budesonide/surfactant and azithromycin trials—have yielded disappointing results, underscoring that effective therapies may need to actively stimulate lung growth rather than simply modulate its inflammatory sequelae [16, 34].

Extreme prematurity disrupts placental endocrine function at a particularly vulnerable window. Circulating within days, IGF-1 falls to approximately one-fifth of fetal levels and remains below the physiological intrauterine range for weeks [35]. The deficit is amplified in ELGANs by the immature liver’s inability to assemble the ternary IGF-1/IGF binding protein-3/acid-labile-subunit complex, which normally prolongs IGF-1 bioavailability; the half-life of free IGF-1 drops to less than 1 h, so that physiological replacement requires continuous intravenous infusion [29]. Low postnatal IGF-1 has been consistently associated with subsequent BPD, retinopathy of prematurity, intraventricular hemorrhage and impaired somatic and brain growth [30].

In preclinical models of pre-eclampsia, chorioamnionitis and hyperoxia, recombinant human IGF-1 combined with its binding protein (IGFBP3) produced dose-dependent improvements in alveolar count, microvascular density and lung compliance, and prevented right ventricular hypertrophy [29, 32]. In mechanically ventilated preterm lambs delivered at approximately 128 days' gestation (equivalent to ≈28 weeks in humans), a 7-day continuous infusion improved gas exchange, enhanced alveolar and capillary development, upregulated VEGF-receptor, and reduced terminal bronchiolar smooth-muscle accumulation, without hepatic, renal, or hemodynamic toxicity [35].

Clinically, the Phase 2a trial of OHB-607 (IGF-1/IGFBP3 complex) missed its primary ophthalmological endpoint but a secondary analysis reported a 53% relative reduction in severe BPD (89% in the evaluable set that reached the target physiological exposure of 28–109 µg/L, a subgroup defined by achieved exposure rather than by randomization) [30, 36]. These findings shifted the clinical development of OHB-607 toward BPD prevention and provided the rationale for the multicenter Phase 2b trial (NCT03253263), which completed randomization of 338 infants born at 23⁺⁰–27⁺⁶ weeks to continuous infusion of OHB-607 until 29⁺⁶ weeks PMA versus standard care [30]. By acting during the canalicular-to-saccular window when distal airspaces and the alveolar capillary bed are simultaneously formed, IGF-1 replacement may preserve the structural ceiling against which lifelong lung function develops. Recruitment is complete. Should the efficacy analyses prove favorable, IGF-1 replacement could become the first pharmacological therapy specifically approved to reduce the incidence or severity of BPD. Whether this translates into measurable improvements in spirometric outcomes and PLD-related morbidities during adolescence and adulthood will require the long-term follow-up of this trial and future longitudinal cohorts.

Mesenchymal stromal cells (MSCs)

Mesenchymal stromal cells (MSCs) have emerged as a promising therapeutic strategy for BPD because of their pleiotropic effects. Rather than engrafting and differentiating into lung tissue, MSCs appear to exert most of their therapeutic impact through paracrine mechanisms, releasing a wide range of bioactive molecules (cytokines, growth factors, and extracellular vesicles (EVs)) that modulate macrophage-mediated inflammation, promote tissue repair, reduce fibrotic reactions and oxidative stress, protect surfactant, while supporting lung alveolar and vascular development [37–39]. Beyond their respiratory effects, MSCs may also exert broader systemic benefits in preterm infants, with emerging preclinical and early clinical evidence suggesting potential effects on neurodevelopmental injury, intestinal inflammation, and sepsis-related complications [40–43]. Among the possible sources, a substantial body of preclinical evidence supports the human umbilical cord (UC) or UC-blood derivation for BPD. MSC populations are functionally heterogeneous, with implications for their therapeutic application in BPD. Single-cell RNA sequencing revealed distinct transcriptional clusters with divergent biological properties, including progenitor-like and fibroblast-like subpopulations, and the transcriptomic profile of MSCs has been shown to correlate with their regenerative efficacy [44, 45].

Phase I trials have demonstrated that administering allogeneic UC-derived MSCs in ELGANs at high risk of BPD is feasible and appears safe, with no major treatment-related adverse events reported [39, 46–48]. To date, only one phase II study has further explored the clinical application of MSC therapy (NCT01828957), though not powered for efficacy [41]. At 5-year follow-up (NCT01897987), mortality, growth, and neurodevelopmental outcomes did not differ between groups, and a non-significant trend towards lower respiratory morbidity was observed in the treated group [49]. Notably, this remains the only clinical program in this field to have reported respiratory outcomes well beyond infancy.

Several challenges remain before MSC-based therapies can be widely implemented. Open questions include the standardization of cell manufacturing, pre-injection conditioning, functional and quality-control testing, donor and tissue selection, dosing strategies, timing and route of administration, patient selection, and outcome measures. The functional heterogeneity noted above probably underlies much of the variability across preclinical studies, making the selection or enrichment of subpopulations with superior reparative capacity a central translational goal [39, 50].

Extracellular vesicles

EVs are nanosized lipid-bilayer particles of 30–5000 nm, carrying a heterogeneous cargo of proteins, lipids, mRNAs, regulatory RNAs, and even mitochondria, and mediating intercellular communication both locally and at a distance [51]. EVs released by MSCs (MSC-EVs) are now considered the effectors of their paracrine therapeutic activity and recapitulate most of their regenerative properties (combining anti-inflammatory, pro-angiogenic and anti-fibrotic effects), while overcoming practical barriers to implementation, like storage and large-scale manufacturing, as well as theoretical safety limitations of live-cell therapy such as engraftment, immunogenicity, and phenotypic instability under inflammatory stress [52, 53].

In the preterm lung, EVs are detected from 22 weeks of gestation in tracheal aspirates and amniotic fluid, and their size and surface markers change across the canalicular and saccular stages, suggesting a possible involvement in developmental processes [54]. EV profiles diverge in those who go on to develop BPD. EVs recovered from their tracheal aspirates are smaller and display an altered immune/epithelial surface-marker signature, with depletion of protective microRNAs. By contrast, circulating EVs from hyperoxia-exposed rat pups carry alveolar epithelial proteins and pyroptosis-related mediators, contributing to concomitant lung and brain damage [54]. Intravenous, intraperitoneal or intratracheal MSC-EVs in neonatal rodents exposed to hyperoxia consistently improve alveolarization, reduce septal fibrosis, restore vascular density, attenuate pulmonary hypertension and shift pulmonary macrophages towards an M2 anti-inflammatory phenotype, with the cardiopulmonary benefit of a single neonatal dose persisting into young adulthood [55–59]. Antenatal MSC-EVs preserve lung development in chorioamnionitis models [60]; multifactorial injury combining endotoxin, mechanical ventilation, and hyperoxia is mitigated by intratracheal EVs with concurrent neuroprotection [61]; and—closer to the clinical context—intravenous bone-marrow MSC-EVs improve gas exchange, alveolar–capillary growth, and feeding tolerance in mechanically ventilated preterm lambs [62].

The EVENEW Phase I/II trial (NCT06279741) of intratracheal allogeneic UC MSC-derived EVs (EXOB-001) in ELGANs is currently recruiting in Europe and represents the first regulatory milestone for a cell-free therapy in BPD. Standardization of isolation, characterization, dosing and pharmacokinetics, together with rigorous adherence to Minimal Information for Studies of Extracellular Vesicles (MISEV) reporting guidelines, remains the principal hurdle before EVs can meaningfully reshape the life-course trajectory of lung function in extreme prematurity.

Surfactant protein-D (SP-D)

Surfactant Protein-D (SP-D), together with SP-A, belongs to the collectin family of innate immunity proteins and is synthesized by alveolar type II cells and club cells. Unlike the hydrophobic surfactant proteins, SP-D has minimal biophysical activity but is a central modulator of pulmonary inflammation and innate host defense [63]. SP-D opsonizes bacteria, viruses, and fungi, attenuates toll-like receptor 4 (TLR4)/nuclear factor kappa B (NF-κB) and NOD-like receptor family pyrin domain-containing 3 (NLRP3)-inflammasome signaling, downregulates pro-inflammatory cytokine release, limits matrix metalloproteinase-2/9 (MMP-2/9) production by alveolar macrophages, and inhibits neutrophil extracellular trap formation [63]. SP-D also contributes to surfactant lipid homeostasis and to the regulation of postnatal surfactant pool sizes.

SP-D’s immunomodulatory activity depends on its assembly into higher-order oligomeric structures, with smaller fragments retaining only partial function [63]. In preterm infants who develop BPD, both the amount and the assembly of bronchoalveolar SP-D appear impaired: levels measured in the first hours of life rise with gestational age, are significantly lower in those who later develop the disease [64], and are skewed towards poorly active low-oligomeric forms [65]. SP-D is induced by chorioamnionitis [64], a condition that itself raises BPD risk and is associated with high secretory phospholipase A2 (sPLA2) activity [66, 67]. Consistent with the role described above, SP-D acts as a counter-regulatory response that protects only when adequate: higher early levels accordingly correlate with shorter invasive ventilation and hospital stay [68], whereas sPLA2 activity rises with longer ventilation [64, 69]. Certain SP-D polymorphisms further link the protein to respiratory distress at birth, oxygen supplementation and respiratory support, indicating a constitutive contribution [70]. Serum SP-D in the first week of life has not consistently mirrored these bronchoalveolar findings [71], confirming that the alveolar compartment is the relevant biological space.

Preclinical evidence supports a therapeutic role for SP-D replacement. In mechanically ventilated premature lambs, intratracheal recombinant human SP-D added to commercial surfactant reduced total inflammatory cells, neutrophil elastase activity, and interleukin-8 in the lung compared with surfactant alone; in neonatal rodent models of hyperoxia and endotoxemia, exogenous SP-D attenuated pro-inflammatory cytokine release, oxidative injury, and alveolar damage [63].

These observations have now reached early clinical testing. The first-in-neonates Phase 1b randomized trial of zelpultide alfa (recombinant human SP-D) enrolled 37 extremely preterm infants and showed that daily intratracheal administration for up to 7 days was safe and well-tolerated, with no dose-limiting toxicities; preliminary efficacy indicators included a lower incidence of BPD and a shorter duration of mechanical ventilation compared with air-sham, alongside a higher mortality in the treated arm; the study was not powered for efficacy [64]. These findings have prompted the ZELA trial (NCT06897839), a randomized, double-blind, placebo-controlled Phase 2b/3 study currently recruiting neonates born at 22⁺⁰–27⁺⁶. It is too early to assess whether early SP-D replacement will durably modify the diseased trajectory underlying PLD, but the convergence of mechanistic biology, biomarker data, preclinical signals, and first-in-human evidence makes it a credible candidate.

Anakinra

Interleukin-1 (IL-1α and IL-1β) sits upstream of the inflammatory cascade sustaining BPD. IL-1β activates NF-κB and the NLRP3-inflammasome, amplifies tumor necrosis factor (TNF)-α, IL-6, and IL-8, recruits and primes monocytes, and links lung injury to the parallel inflammatory damage of brain and gut that defines extreme prematurity [72]. In preterm infants, tracheal-aspirate IL-1β is elevated days before the clinical onset of BPD, and correlates with the duration of mechanical ventilation and supplemental oxygen, particularly when Ureaplasma colonization is present [73].

The pharmacological case for anakinra is unusual in the BPD landscape because the drug already exists, making it a candidate for repurposing rather than de novo development. Recombinant human IL-1 receptor antagonist (IL-1Ra, anakinra) has been used in pediatric and adult inflammatory diseases for more than two decades—including in neonates with CINCA/NOMID syndrome—with a well-characterized safety, pharmacokinetic, and efficacy profile [72]. In murine models combining perinatal LPS and postnatal hyperoxia, daily IL-1Ra prevents the alveolar simplification, septal disorganization and gas-exchange impairment of experimental BPD [74], attenuates airway remodeling [75], and preserves pulmonary vascular density while limiting the rise in pulmonary vascular resistance characteristic of BPD-associated pulmonary hypertension [76].

Clinical translation is led by the investigator-initiated Anakinra Pilot trial (NCT05280340), an open-label Phase I/IIa dose-escalation study in 24 infants born at 24⁺⁰–27⁺⁶ weeks, administering intravenous anakinra over the first 21 days of life [72]. The trial has completed enrolment and is currently in the follow-up phase, with results awaited; no prospective clinical efficacy data are therefore yet available in this population. Meanwhile, De Rose et al. reported promising respiratory outcomes in a retrospective neonatal cohort, a hypothesis generating observation. However, only 4 patients were ELGANs with BPD (the population most at risk and targeted by the ongoing trial), underscoring the continued need for trial data in this specific population [77].

From the bench: emerging preclinical therapies

Preclinical studies from the last 5 years in experimental models of BPD/PLD are uncovering a range of mechanistically distinct but potentially complementary therapeutic targets (Table 1, Fig. 1). Beyond descriptive pathology, these approaches focus on actionable biological processes. These could be broadly grouped into inflammasome-driven and NF-κB-mediated inflammation, oxidative and mitochondrial stress, and the vascular/angiogenic axis. Additional themes include endoplasmic reticulum stress, matrix and airway remodeling, gut–lung microbiota modulation, and epigenetic/RNA-based strategies, illustrating an increasingly multi-compartmental view of BPD pathogenesis (Fig. 1, Table 1).

Table 1.

Preclinical studies from the last 5 years of novel interventions in in vivo translational models of bronchopulmonary dysplasia/prematurity-associated lung disease, organized by key biological mechanisms.

Studies are grouped by predominant mechanistic targets

Author, year	Model	Drug/intervention	Studied signaling pathway(s)	Key outcomes
Section 1: inflammasome, pyroptosis, and NF-κB signaling
León Silva et al., 2025 [82]	Neonatal rat, hyperoxia	VX-765 (caspase-1 inhibitor)	Caspase-1/IL-1β/IL-18/GSDMD; pyroptosis	Improved alveolar structure and pulmonary vascular density; ↓ vascular remodeling, right ventricular hypertrophy, aortic stiffness and cardiac fibrosis
Feng et al., 2026 [83]	Neonatal mouse hyperoxia + AECII hyperoxia	Dexmedetomidine	PINK1/Parkin-mediated mitophagy → NLRP3 inflammasome/pyroptosis	↓ lung injury, NLRP3 activation, cytokines and pyroptosis; preserved mitochondrial integrity
Zhang et al., 2026 [84]	Neonatal mouse, hyperoxia	Betaine	FOXO1–NLRP3 axis; macrophage pyroptosis	↓ macrophage pyroptosis and inflammation; ↑ alveolar development
Yang et al., 2025(a) [85]	Neonatal rat + AECII	Nesfatin-1	HMGB1/TLR4/NF-κB/NLRP3	↓ inflammation, fibrosis, apoptosis, neutrophil influx in BALF; improved lung injury
Zhang et al., 2024(a) [86]	Rat (prenatal LPS) + placenta	Inhaled hydrogen (H₂)	TLR4–NF-κB–IL-6/NLRP3; placenta–lung axis	↓ placental inflammation (IL-6, IL-1β, IL-18); improved neonatal lung development, ↑ survival
Hao et al., 2025 [87]	Neonatal rat, prolonged hyperoxia	High-dose (50 mg/kg/die dose) Nintedanib	NF-κB; IL-1β/CXCL-1/MCP-1; apoptosis	↓ histological injury (↑ RAC/↓ MLI), ↓ apoptosis, ↓ inflammation; ↓ NF-κB. No unplanned mortality reported (see Ding et al., 2025)
Section 2: oxidative stress and redox signaling
Aslan et al., 2023 [88]	Neonatal rat, hyperoxia	Prophilactic molsidomine	NO donor; oxidative stress; TNF-α/IL-1β; apoptosis	↓ histological injury, macrophage infiltration, oxidative/inflammatory stress, ↓ apoptosis
Deng et al., 2022 [89]	Neonatal rat + AECII hyperoxia	CGRP (calcitonin gene-related peptide)	Notch1/Hes1/HERP; oxidative stress (MDA/SOD)	↑ AECII viability; ↓ AECII➝ AECI transdifferentiation➝ preserved AECII progenitor pool; improved epithelial regeneration
Guzmán-Navarro et al., 2021 [90]	Rat, hyperoxia–hypoxia	Prenatal indole-3-carbinol	AhR➝ antioxidant genes (Cyp1a1, Nqo1, Gsta1); ↑ NF-κB (immunomodulatory)	↑ RAC (partial); ↓ fibrosis (early timepoint); ↓ lung injury/inflammatory infiltrates
Montgomery et al., 2026 [91]	Neonatal C57BL/6N mouse, hyperoxia	Intranasal FAD (flavin adenine dinucleotide)	FAD-glutathione reductase-GSH/GSSG redox axis	Improved redox potential (GSH/GSSG ratio); ↑ RAC ↓ MLI and ↓BALF neutrophils, ↑ BALF macrophages
Pini et al., 2026 [92]	Neonatal rat, strong hyperoxia	β3-adrenoceptor agonist at three doses (BRL37344)	β3-adrenoceptor signaling; TGF-β; VEGF/VEGFR2; fibroblast/Circulating Endothelial Progenitor Cells	↑ survival, alveolarization, lung volume; ↓ TGF-β ↓ fibrosis and oxidative stress; ↑ vascular density, VEGF and Circulating Endothelial Progenitor Cells. Protective at 3 mg/kg; 6 mg/kg ↑ mortality under hyperoxia
Reçica et al., 2024 [93]	Rat pups, > 95% O₂	Prophilactic resveratrol	NO; SOD/GPx; TNF-α/IL-1β	↓ airway hyperreactivity; restored relaxation; ↑ antioxidant defenses; ↓ inflammation; improved lung histology
Zhao et al., 2023 [94]	Neonatal mouse (males only), hyperoxia	Irisin	Nrf2/HO-1 pathway; oxidative stress	↓ MDA ↑GSH; ↑ angiogenesis and alveolarization (↑ RAC ↓ MLI)
Chu et al., 2023 [95]	Premature rat, hyperoxia	Erythromycin	GSH; TNF-α/IL-1β	↑ GSH; ↓ TNF-α/IL-1β; attenuated histologic lung injury (qualitative)
Ozdemir et al., 2022 [96]	Neonatal rat, hyperoxia	Apocynin	SOD/GSH/GPx/oxidative stress; TNF-α/IL-1β	↓ histopathological injury, oxidative stress and ↓TNF-α and IL-1β, ↓ alveolar macrophage scores; ↓ TUNEL + apoptotic cells
Yang et al., 2023 [97]	Neonatal rat + AECII, hyperoxia	Nesfatin-1	SIRT1/PGC-1α pathway; oxidative stress	↑ cell viability; ↓ apoptosis (↓ Bax, ↑ Bcl-2); ↓ ROS, ↓ pulmonary MDA, ↑ pulmonary SOD; improved alveolarization in vivo: (↑ RAC, ↓ MLI)
Section 3: angiogenesis and vascular development
Chang et al., 2022 [98]	Fetal pulmonary endothelial cells, hyperoxia + neonatal mouse, hyperoxia + neonatal mouse Cpt1a-KO, hyperoxia	L-carnitine + baicalin	CPT1a/fatty acid oxidation; endothelial angiogenesis/vascular development	↓ endothelial apoptosis (↓ caspase-3); ↑ endothelial migration/tube formation; ↓ alveolar and vascular simplification (↑ RAC, ↑ vWF)
Ding et al., 2025 [99]	Neonatal rat, hyperoxia	Dose-ranging study with intraperitoneal nintedanib	Src Tyrosine kinase; IL1, TNF-α, Caspase-3	Narrow therapeutic window (low dose; lethal at a 50 mg/kg/die dose); Src phosphorylation unchanged (Src-independent mechanism by nintedanib), Restored RAC/MLI; ↑ pulmonary vascular density; ↓ right ventricular hypertrophy; ↓ vessel wall thickness; ↓ IL-1α, ↓ TNF-α, ↓ Caspase-3
Huang et al., 2021 [100]	Neonatal mouse, hyperoxia	Roxadustat	HIF-1α stabilization; VEGF; eNOS; mTOR/HIF-1α	↑ pulmonary angiogenesis (↑ VEGF, eNOS, vWF, vessel density); ↓ MLI; ↑ survival/body weight; ↑RAC
Xiang et al., 2022 [101]	Neonatal mouse, hyperoxia + RAW264.7 macrophages, hyperoxia	Metformin, purmorphamine (Shh agonist)	Shh/Gli1; M1 → M2 macrophage polarization; VEGF, TNF-α, CD31	↑ M2 macrophages; ↓ TNF-α/iNOS; ↑ angiogenesis (CD31/VEGF); improved vascular development. No effect on secondary septa. Purmorphamine reversed metformin effects on M2 polarization and VEGF
Daniel et al., 2025 [102]	Neonatal rat, hyperoxia, normoxia & intermittent hypoxia (H-IH) + acute hypoxic challenge (12 wk)	Sodium nitrite vs. iNO 10 ppm	NO/sGC/cGMP signaling; S-nitrosylation by proteomics, xantine oxidase activity; irreversibile nitration	NaNO₂: prevented all H-IH-induced abnormalities (alveolar surface area, alveolar number, MLI, septal thickness, RVH, medial wall area); restored NO and cGMP; prevented nitration; benefits persisted into adulthood (12 wk) with improved SpO₂ in acute hypoxic challenge. iNO 10 ppm: prevented PH but had minimal effects on alveolar morphology and did not prevent nitration. iNO in normoxic controls caused alveolar hypoplasia, decreased NOx and cGMP, and increased nitration (evidence of iNO toxicity at a clinical dose on the developing lung)
Section 4: epigenetics, miRNA, and lncRNA
Sugar et al., 2021 [103]	Double-hit murine model (prenatal intraamniotic LPS + neonatal hyperoxia)	miR-29b delivered via lipid nanoparticles	Epigenetics: histone methylation (e.g. H3K4me3, H3K27me3, H4K20me3, PRMT1/PRMT5)	↓ septal thickness; ↓ αSMA/MMP-9; ↑ PDGFα/PDGFRα; partial/full restoration of methylation patterns; partial restoration of PRMT1; no significant improvement in terminal airspaces or alveolar area
Ji et al., 2021 [104]	Neonatal mouse, short-exposure hyperoxia + BEAS-2B cells, hyperoxia	Adenoviral vector overexpressing lncRNA CASC2; rescue with miR-194-5p agomiR or sh-CAV1	CASC2–miR-194-5p–CAV1 axis; TGF-β1, CD31	↓ TGF-β (cytoplasmic CASC2 competitively binds miR-194-5p, derepressing CAV1, leading to TGF-β1 pathway inactivation; causality confirmed by full reversal of protection with miR-194-5p agomiR or sh-CAV1) ↓ apoptosis (TUNEL); ↑ angiogenesis (CD31), ↑ proliferation (ki67); attenuated lung injury
Heyob et al., 2023 [105]	Double-hit murine model (prenatal intraamniotic LPS + neonatal hyperoxia)	DNMT inhibitors: decitabine, RG108	DNA methylation; DNMT; TGF-β1/p-SMAD2/3 signaling; surfactant protein C	Decitabine: modest improvement in alveolarization; ↓ p-SMAD2/3; ↑ surfactant protein C. RG108: no evident benefit
Section 5: cell-based therapies and biologic agents
Song et al., 2026 [106]	Neonatal rat, prolonged hyperoxia	Rat bone marrow-derived MSCs overexpressing PRDX6, generated by lentiviral PRDX6-GFP transduction	PRDX6–MANF secretion (paracrine/autocrine); VEGF/CD31/vWF	↑ BMSC lung recruitment; ↑ MANF in lung/BALF/serum ↓ apoptosis (caspase-3); ↓ TNF-α/IL-6/IL-1β; ↑ angiogenesis (VEGF/CD31); ↓ vascular remodeling
Chaubey et al., 2021 [107]	Neonatal mouse, hyperoxia; human RDS/BPD lung tissues for SSEA-1 expression comparison	Recombinant human FUT9	SSEA-1/FUT9; lung stem/progenitor cells	Hyperoxia: ↓ SSEA-1, ↓ BALF total cells/neutrophils FUT9 injection: ↓ BALF protein leak, ↓ TUNEL cell death, ↑ Ki67 + proliferation; no significant reduction in lung parenchymal IL-6/IL-1β and no reversal of PH/RVH
Toth et al., 2022 [108]	Rhesus macaque, chorioamnionitis (intra-amniotic LPS)	Combined anti-IL-1β (anakinra) + anti-TNFα blockade (adalimumab) administered before LPS	IL-1/TNF–NF-κB; alveolar signaling (VEGF, PDGF, WNT); CCL/CXCL inflammatory chemokine signaling	Normalized lung injury score/septal numbers, preserved AT1–alveolar capillary interactions, restored epithelial/endothelial cellular quorum, blunted inflammatory activation
Section 6: mixed/multi-pathway interventions
Xu et al., 2026 [109]	Neonatal mouse, hyperoxia from postnatal day 8	Apelin-12	Nrf2/↑HO-1; NF-κB; NLRP1 inflammasome	↓ IL-1β, IL-18; ↓ oxidative stress; improved lung function and alveolar architecture. Paradoxical suppression of NLRP1 by hyperoxia; cytokine release persisted despite NLRP1 suppression, implicating non-canonical inflammasome-independent pathways
Graumuller et al., 2026 [110]	Neonatal mouse, hyperoxia	Dimethyl fumarate	Nrf-2, paradoxical ↓ HO-1, ↓ TGF-β–Smad2 and NF-κB	Improved lung histology, ↑ cell proliferation, ↑ angiogenesis (vessel count)
Tayman et al., 2021 [111]	Neonatal rat, intra-amniotic LPS + postnatal hyperoxia	Apocynin	NADPH oxidase; Nrf2, ROS and antioxidant axis; caspase1/3, NLRP3 inflammasome, inflammatory cytokines	↑ survival; improved alveolarization (↑ RAC, ↓ MLI) and ↑ surfactant protein B/C + pneumocytes; ↓ histopathological injury score and fibrosis; coordinated ↓ of oxidative stress and ↑ of antioxidant capacity; ↑ Nrf2; ↓ NLRP3, caspase-1, caspase-3 and pro-inflammatory cytokines; ↓ neutrophil infiltration (MPO)
Das et al., 2021 [112]	Neonatal mouse, BPD + experimental PH	AVR-48 (chitin derivative)	TLR4; M2 macrophage polarization; NF-κB/TNF-α/IL-1β/TGFβ inflammatory signaling; SP-C, VEGF, Ang2, eNOS, VEGF-D	↓ NF-κB/TNF-α/IL-1β/TGFβ inflammatory signaling ↓ BALF total cells/protein; reduced apoptosis (↓ TUNEL + ↓ caspase-3); ↑ IL-10; normalized neutrophil/macrophage balance; ↑ SP-C + ATII cells’ proliferation (↑ Ki67 +, PCNA); ↑ RAC; ↑ VEGF, VEGF-D, Ang2, eNOS ↓ septal thickness,; ↑ vWF, ↓ Ang2, ↓ Fulton’s Index/RV hypertrophy
Soni et al., 2023 [113]	Neonatal mouse, BPD from 75% O₂	AICAR (AMPK direct activator)	AMPK-ULK1–autophagy axis; autophagy-NLRP3 inflammasome link; alveolar macrophage M1/M2 polarization	↑ alveolarization (↑ RAC, ↓ MLI) and ↑ ACTA2⁺ septal-tip myofibroblasts (secondary septation); ↑ CD31 (vascular density); ↓ neutrophil and macrophage accumulation; ↑ M2 polarization in BAL. Anti-inflammatory effect is autophagy-dependent
Sudhadevi et al., 2024 [114]	Neonatal mouse 95% hyperoxia → adult follow-up	Intranasal Fingolimod (S1PR1 modulator) vs intraperitoneal fingolimod	SPHK1/S1P/S1PR1; Lysyl oxidase (LOX)/collagen cross-linking	Improved alveolarization (↓ MLI), ↓ BALF protein/total cells/neutrophils, fingolimod decreased ↓ S1PR1 and ↓ LOX in both airway and alveolar regions → ↓ collagen cross-linking; no lymphopenia with intranasal route. Adults (PN56): persistent ↓ MLI, ↓ peribronchial collagen and ↓ smooth muscle hypertrophy (myosin) ↓ airway hyperreactivity to methacholine. Notably, the ↑ S1PR1 and ↑ LOX persisted into adulthood of hyperoxia untreated mice, selectively in airways (not in alveoli), implicating airway-specific pathology in BPD sequelae. This persistent effect was reduced by the use of fingolimod
Zhong et al., 2022 [115]	Neonatal mouse, hyperoxia + lung fibroblasts	LOX inhibition (BAPN)	Lysyl oxidase (LOX)–TGF-β–Smad2/3; ECM remodeling; extracellular aldehydes	LOX inhibition →↓ aldehydes, improved ECM organization and alveolar development
Yang et al., 2025(b) [116]	Neonatal rat, hyperoxia	Prostaglandin E1	ER stress (GRP78, CHOP); apoptosis (caspase-3, Bcl-2/Bax)	↓ IL-1β/IL-6/TNF-α; ↓ caspase-3/Bax-mediated apoptosis; ↓ pulmonary edema; improved alveolar histology
Zhang et al., 2023 [117]	Neonatal rat, > 90% hyperoxia	Glutamine	ER stress (IRE1α/JNK; GRP78, CHOP); apoptosis	↓ inflammation, oxidative stress and apoptosis; ↑ RAC, ↓ MLI, RAC improved lung function (inspiratory capacity, compliance, airway resistance, tissue elasticity)
Liang et al., 2021 [118]	Neonatal rat, 90% O₂	Adenoassociated virus encoding SEMA3A	ERK/JNK; NF-κB; apoptosis and inflammation	Endogenous SEMA3A (mRNA and protein) significantly reduced in BPD model lung; AAV-SEMA3A restored body weight, lung weight, and lung-to-body weight ratio. Improved alveolar architecture: ↑ RAC, ↓ MLI, reduced inflammatory infiltrate. ↓ TUNEL⁺ apoptotic cells; ↑ Bcl-2, ↓ Bax, ↓ cleaved Caspase-3. ↓ lung IL-1β, MCP-1, TNF-α; ↓ p-NF-κB; ↓ p-ERK1/2 and ↓ p-JNK (with unchanged total ERK/JNK)
Lee et al., 2023 [119]	Neonatal rat, hyperoxia	Recombinant Hsp70/heat stimulation at birth	Hsp70-TLR4-NF-κB axis (Anti-apoptotic; anti-inflammatory)	↑ survival; ↓ early alveolar apoptosis and macrophage infiltration; ↓ histological injury
Xie et al., 2025 [120]	Neonatal mouse, hyperoxia	Sodium propionate	IL-17 pathway; epithelial apoptosis (Bax, Bcl, caspase-3)	↑ alveolarization (RAC); ↓ septal thickening; ↓ IL-17/IL-6/TNF-α; ↓ AECII apoptosis
Kryeziu et al., 2023 [121]	Neonatal rat, > 95% O₂	Quercetin	NO/cGMP signaling; TNF-α/IL-1β	↓ airway smooth muscle hyperreactivity; restored tracheal relaxation; ↓ TNF-α/IL-1β; no structural outcomes
Sopi et al., 2026 [122]	Neonatal rat, > 95% O₂	Rho-kinase inhibitors (Y-27632, fasudil)	Rho/Rho-kinase; NO–cGMP signaling (no direct measure eNOS, cGMP)	Restored tracheal smooth muscle relaxation
Chen et al., 2023 [123]	Neonatal mouse, hyperoxia	Intranasal Lactobacillus johnsonii	Gut–lung microbiota axis; angiogenesis; cytokines (IL-1β, IL-6, TNF-α on lung tissue); ZO-1/occludin (intestinal tight junctions)	↑ body weight; ↑ alveolarization (↓ MLI); ↑ pulmonary angiogenesis (↑ VEGF, ↑ vWF); ↓ pulmonary IL-1β and IL-6; ↑ intestinal ZO-1 and occludin; normalized gut microbiota (↓ Staphylococcus and Enterobacter, ↑ Lactobacillus)
Zhang et al., 2024(b) [124]	Neonatal mouse, hyperoxia + MLE 12 alveolar epithelial cells, hyperoxia at	Lactobacillus reuteri, 3-IAld, Recombinant IL-22; Anti-IL-22 neutralizing antibody	IL-22/STAT3	↑ IL-22 → STAT3 phosphorylation (in MLE-12 in vitro); ↑ alveolarization (↓ MLI, ↑ RAC); ↑ body weight; ↓ pulmonary IL-1β/IL-6/TNF-α; ↑ alveolar epithelial markers SPC and AQP5; ↑ pulmonary vascular markers FLK-1/VEGFR2 and VEGF. Mechanistic causality: (1) 3-IAld alone phenocopies L. reuteri benefit; (2) recombinant IL-22 alone is sufficient to protect; (3) anti-IL-22 abolishes L. reuteri rescue, confirming IL-22 as necessary mediator; (4) IL-22 induces STAT3 phosphorylation in MLE-12, blocked by anti-IL-22
Guillier et al., 2021 [125]	Rat, antenatal low protein diet	Nebulized curcumin	PPARγ–FABP4 axis; downregulation of profibrotic pathways: TGFβ, epithelial-mesenchymal transition, Wnt/β-catenin, Notch, Sonic Hedgehog; Anti-inflammatory; oxidative stress; NF-κB	↓ neonatal lung injury following antenatal insult; ↓ inflammation and structural damage ↓ LPD-induced abnormal alveolarization ↓ FABP4 mRNA at P11/P21 and ↓ FABP4⁺/CD206⁺ alveolar macrophage density at P21; ↑ pulmonary vessel; transcriptomic reversal of profibrotic pathways activated; well tolerated, no mortality or detectable adverse event

Search strategy and abbreviations are reported in Supplemental material

Fig. 1.

Preclinical targets of translational research in animal models of bronchopulmonary dysplasia from the last 5 years. Created in BioRender. Zanetto, L. (2026) https://BioRender.com/7lqdafp

Several of these strategies converge with interventions already discussed in this review. Cytokine blockade with anti–IL-1β (anakinra) and anti-TNF-α (adalimumab), tested preclinically in chorioamnionitis-driven injury, supports the broader rationale for targeted immunomodulation. MSC and MSC-EV therapies are evolving from raw cell products toward engineered paracrine platforms, for instance MSCs overexpressing antioxidant or neurotrophic factors. The nutrient and microbiota axes are likewise represented at the bench: L-citrulline, a nitric oxide precursor, preserves alveolar and vascular growth and attenuates pulmonary hypertension in rodent hyperoxia models [28], and early-phase studies in preterm infants have characterized its pharmacokinetics and tolerability, though a dosing regimen has not been established [78, 79]. Among microbiota-targeted approaches, probiotic trials have mostly targeted other complications of prematurity, with no effect on BPD in meta-analysis [22]. A single trial (ChiCTR2400093781) with BPD as a primary endpoint awaits confirmation in independent populations [21].

A notable gap is the still-limited translational use of patient-derived platforms: nearly all signals still come from neonatal rodent hyperoxia. Hyperoxia-only models capture neither the interacting mechanisms outlined above, nor the genetic susceptibility that modulates human BPD/PLD. Patient-derived airway organoids generated from bronchoalveolar lavage of preterm infants with grade 3 BPD have very recently been reported [80], recapitulating epithelial heterogeneity, hyperoxia-induced injury responses, and dexamethasone-driven anti-inflammatory remodeling, with sex-specific differences validated against human BPD lung specimens. In addition, induced pluripotent stem cell-derived lung organoids may serve as viable substitutes when access to primary fetal tissue is limited, and represent a valuable approach to reproduce key stages of lung development. Such platforms may offer a scalable human-relevant alternative to animal models, and could potentially enable patient-stratified preclinical screening, bridging a long-standing translational gap between rodent models and the developing preterm lung [81].

Finally, gene therapy work in animal models currently seems most useful as a way to identify functional targets for downstream pharmacologic modulation, rather than as a near-term clinical option. Given the multifactorial, multi-compartment biology of PLD, lasting benefit will more plausibly emerge from combined approaches across vascular, immune and matrix programs than from correction of any single gene or pathway.

Discussion

The development of novel therapies for BPD and PLD is shaped by significant regulatory, ethical, and methodological challenges inherent to neonatal research [126]. Neonates and infants are classified as a vulnerable population and require stringent safeguards to ensure ethical conduct, including enhanced oversight, risk minimization, and rigorous justification of potential benefits [127]. These requirements, while essential, contribute to increased complexity, cost, and duration of clinical development programs. In addition, market-related factors play a non-negligible role, as the relatively small target population limits the expected return on investment, potentially reducing industry-driven innovation in this field.

Historically, most therapeutic strategies have followed an “adult-first” development paradigm, with subsequent extrapolation to pediatric populations [126]. This approach often leads to substantial delays in the availability of new treatments for neonates, particularly in conditions such as BPD where pathophysiology is unique and not fully recapitulated in adult disease. There is growing recognition of the need to promote “pediatric-first” or “neonate-first” study designs, supported by tailored regulatory pathways that encourage early inclusion of these populations in drug development [126, 128]. Equally important is a reframing of the perceived commercial scope of these therapies. PLD should not be regarded as a condition confined to the extremely preterm infant or to early childhood: it represents a continuum spanning the full gradient of prematurity and continues to shape respiratory trajectories throughout adolescence, adulthood, and older age, with documented links to accelerated lung-function decline and COPD in later life [13, 129]. Effective interventions could therefore yield benefits well beyond the immediate target population, with implications for adult respiratory health at a societal scale—a perspective that may shift the cost-opportunity calculus for industry and support a more sustained commitment to drug development in this field [130]. Realizing this potential, however, requires a dynamic regulatory landscape, evolving to value these longer-term benefits rather than BPD alone. In addition, despite progressive efforts by regulatory agencies worldwide to mandate and incentivize pediatric drug development, harmonization across regulatory systems remains incomplete, and stronger global coordination will be essential to accelerate the safe and timely translation of innovative therapies to neonatal populations. Compounding these challenges, a definitional conundrum still affects BPD-centered outcomes. The operational nature of BPD and the coexistence of its definitions mean that both incidence and treatment-effect estimates shift depending on the definition applied in each trial [6, 7, 10]. This does not make BPD dispensable. It remains the outcome against which neonatal pulmonary trials have long been measured, and thus the endpoint that allows the therapies reviewed here to be compared with earlier ones. Disaggregated into its components of respiratory support, oxygen requirement, and pulmonary vascular disease, and combined with extended follow-up, it retains validity. Within the framework of PLD, however, therapeutic strategies can no longer be evaluated solely for their short-term efficacy, such as reductions in BPD incidence at 36 weeks PMA, but increasingly for their potential to promote physiological lung development and ultimately to modify long-term outcomes. Read as the most severe manifestation of the PLD continuum, BPD is best complemented rather than replaced by endpoints such as respiratory support at discharge, home oxygen use, pulmonary hypertension, post-discharge respiratory health care utilization, and later lung-function measures. With the exception of the 5-year follow-up of one trial, none of the therapies reviewed here has yet reported such outcomes, and the respiratory follow-up currently prespecified is not designed to capture PLD (Table 2). Preventing or attenuating BPD is expected to shift the whole spectrum of PLD, but whether that shift persists beyond infancy is something current endpoints cannot demonstrate. Later-phase and future trials should therefore consider prespecifying respiratory assessment that extends sufficiently to establish an effect on PLD, and not on BPD alone.

Table 2.

Status and respiratory follow-up of clinical trials discussed in this review

Agent	Trial and trial phase	Population	Primary endpoint	Status	Prespecified respiratory follow-up
OHB-607 (rhIGF-1/rhIGFBP-3)	NCT03253263, Phase 2b	295, 23⁺⁰–27⁺⁶ weeks	Severe BPD or death at 36 weeks PMA	Enrolment complete	Chronic respiratory morbidity outcomes at 24 months corrected age
UC-blood derived MSCs	NCT01828957, Phase 2	66, 23–28 weeks	BPD severity	Completed	See next line
UC-blood derived MSCs	NCT01897987, follow-up of NCT01828957	59, 23–28 weeks	Composite morbidity of respiratory symptom-associated emergency department visits, hospital readmissions and oxygen use	Completed	Five-year follow-up
EXOB-001 (UC-MSC-derived EVs)	NCT06279741, Phase 1/2	36 (phase 1) and 203 (phase 2), 23⁺⁰–27⁺⁶ weeks	Safety; prevention of severe BPD	Recruiting	Lung ultrasound score and respiratory morbidity to 2 years corrected age
Zelpultide alfa (rhSP-D)	NCT04662151, Phase 1b	37, 23⁺⁰–28⁺⁶	Safety and tolerability	Completed	No prespecified long-term respiratory endpoint
Zelpultide alfa (rhSP-D)	NCT06897839, Phase 2b/3	366, 22⁺⁰–27⁺⁶	Incidence of grade 2 or grade 3 BPD or death	Recruiting	Chronic respiratory morbidity through 12 months corrected age
Anakinra	NCT05280340, Phase 1/2a	24 infants, 24⁺⁰–27⁺⁶ weeks	Safety, feasibility, pharmacokinetics	Enrollment complete, in follow-up	No prespecified long-term respiratory endpoint

Conclusions

PLD sits at the intersection of developmental biology, lung injury and lifelong respiratory trajectory, and its therapeutic landscape is rapidly expanding. The most clinically credible interventions discussed in this review include growth-factor strategies, mesenchymal stromal cell and extracellular vesicle therapies, and cytokine blockade. Altogether, these efforts point toward a future in which PLD is managed not by targeted single-axis agents but through combinatorial, mechanism-informed approaches, tailored to the individual patient and to the specific phase of disease. Realizing this promise will require a broader recognition of PLD as a life-course condition with public health relevance well beyond the neonatal intensive care unit.

Supplementary information

Below is the link to the electronic supplementary material.

Abbreviations

AECII

Alveolar epithelial type II cells

BPD

Bronchopulmonary dysplasia

CASC2

Cancer susceptibility candidate 2

cGMP

Cyclic guanosine monophosphate

COPD

Chronic obstructive pulmonary disease

ELGAN

Extremely low gestational age neonate

EV

Extracellular vesicle

FUT9

Fucosyltransferase 9

GSDMD

Gasdermin D

HIF-1α

Hypoxia-inducible factor 1α

HO-1

Heme oxygenase 1

IGF-1

Insulin-like growth factor 1

IGFBP-3

Insulin-like growth factor-binding protein 3

IL-1β/-22

Interleukin-1β/-22

IRE1α

Inositol-requiring enzyme 1α

lncRNA

Long non-coding RNA

LNP

Lipid nanoparticle

LOX

Lysyl oxidase

miR

MicroRNA

MISEV

Minimal Information for Studies of Extracellular Vesicles

MMP

Matrix metalloproteinase

MSC

Mesenchymal stromal cell

NADPH

Nicotinamide adenine dinucleotide phosphate

NF-κB

Nuclear factor κB

NLRP3

NLR family pyrin domain–containing 3

NO

Nitric oxide

Nrf2

Nuclear factor erythroid 2–related factor 2

PERK

Protein kinase R–like endoplasmic reticulum kinase

PGE1

Prostaglandin E1

PLD

Prematurity-associated lung disease

PMA

Postmenstrual age

rhFUT9

Recombinant human FUT9

rhIGF-1

Recombinant human insulin-like growth factor 1

rhSP-D

Recombinant human surfactant protein D

SP-D

Surfactant protein D

sPLA2

Secretory phospholipase A2

STAT3

Signal transducer and activator of transcription 3

TGF-β

Transforming growth factor β

TLR4

Toll-like receptor 4

TNF-α

Tumor necrosis factor α

UC

Umbilical cord

VEGF

Vascular endothelial growth factor

Authors’ Contributions

EB and DDL conceptualized the review. LZ performed the literature search and drafted the manuscript, table and figure. LB contributed to the literature search and critically revised the manuscript. DDL and EB provided supervision, critically revised the manuscript for important intellectual content, and approved the final version. SS, EV, EP, DN revised the figure and table and the manuscript for intellectual content. VN revised the manuscript for important intellectual content. All authors read and approved the final manuscript.

Funding

Open access funding provided by Università degli Studi di Padova within the CRUI-CARE Agreement.

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

NA.

Competing interests

EB and LB declare that they acted as external experts for ExoBiologics and Oak Hill Bio. DDL received speaker or consulting fee or research assistance from Airway Therapeutics, Oak Hill Bio, Chiesi Farmaceutici, Medtronic, Getinge, Zoll, Radiometer, Mediprema. VN is an employee and shareholder in Oak Hill Bio Holdings (previously known as Oak Hill Bio Ltd). LZ, SS, EV, DN, EP, declare no conflicts of interest.