Developing a mechanistic understanding of pediatric acute respiratory distress syndrome one cell at a time

Soon after the advent of invasive mechanical ventilation, a previously unreported syndrome was identified that consisted of severe hypoxemia and bilateral lung infiltrates on chest x-ray.¹ At that time, the condition was fatal. Ashbaugh et al. published the first description of these patients in 1967 and coined the term acute respiratory distress in adults which would become known as ARDS.² In this report, Ashbaugh noted the presence of hyaline membranes lining their alveoli on autopsy. The definition of ARDS has been refined several times but the hallmarks of severe hypoxemia and chest x-ray infiltrates remain core features.³ Pediatric acute respiratory distress syndrome (PARDS) was first defined as a unique syndrome in 2015.⁴ PARDS, like its adult counterpart, is a heterogenous multifactorial lung disease that is defined by hypoxemia, non-cardiogenic pulmonary edema, and occurs within 7 days of a precipitating insult.⁵

ARDS and PARDS have traditionally been categorised into three phases.⁶ The exudative phase occurs within the first⁷ days of an insult and is characterised by activation of the innate immunity, recruitment of additional leukocytes to the alveolus, and damage to the alveolar capillary barrier.⁶ The proliferative phase occurs next and is characterised by the proliferation and differentiation of alveolar type II (ATII) cells into alveolar type I (ATI) cells.⁶ The proliferative phase can begin days after the initial insult and can last for days or even weeks.⁶ Patients unable to successfully repair the damage enter the third phase, known as the fibroproliferative phase, which is characterised by collagen deposition by fibroblasts leading to lasting lung damage and dysfunction.⁶ This schematic is likely an oversimplification, but is a useful framework from which to discuss the current research in ARDS & PARDS.

PARDS & ARDS research has largely focused on the exudative phase and translating promising therapies from preclinical models or single centre studies. Most large, randomised clinical trials in ARDS and PARDS have shown no benefit of the intervention tested. Relatively fewer studies have sought to understand the specific cells involved in human lung repair and regeneration after acute injury. The reason for this is that lung tissue is rarely available from these patients. The little that we know in humans are targeted confirmatory studies of tracheal aspirate and bronchoalvleolar lavage fluids that test whether the same development, repair, and regenerative pathways important in mice are operative in humans. For example, both transforming growth factor-β (TGF-β) and Wnt/β-catenin signalling are activated.^{7, 8} While many other examples can be given, each is rooted in established mouse in vivo and human in vitro biology with confirmation in human specimens. In this issue of JEM, Song et al. took the opposite approach and asked what are the important cell and cell signalling pathways that are altered in pediatric ARDS and how are they different than normal human lung.

The authors examine the case of a 10-year-old female placed on extracorporeal membrane oxygenation (ECMO) support for severe PARDS complicated by air leak. The patient ultimately survived but required surgical intervention for her air leak. The authors performed single-cell RNAseq on lung tissue obtained during her surgical intervention and peripheral blood mononuclear cells (PBMCs) from blood samples taken at several different time points while the patient was on ECMO. The comparison lung tissue samples were taken from two healthy boys aged 1 and 3 years. The authors found that monocyte derived macrophages and lung stromal cells showed reparative transcriptomic signatures in the PARDS subject. The transcriptomic signatures from the PBMCs showed a gradual decrease in inflammatory and interferon-related gene expression over time. The biggest strength of this is study is the direct evaluation of lung tissue from a survivor of PARDS. Most studies assessing lung tissues directly are postmortem, as lung biopsy is often not necessary or advisable in cases of PARDS. This limits our knowledge regarding the mechanisms critical to successful lung repair and regeneration in patients able to successfully recover from their injury. The study is hampered by several weaknesses, however. One is the sample size; however, lung tissue from PARDS patients is a rarity. The second is the control population. Patients who are much younger are likely to have transcriptomic differences as they are continuing to grow and develop. Third, it is unclear to what extent PBMC transcriptomic profile mirrors ongoing processes in pulmonary cells. For these and other reasons, any conclusions drawn with regards to important cells or processes in lung repair and regeneration should be made with caution.

These findings are complemented by other recent—omics studies in pediatric ARDS. One recent study using bulk mRNAseq to evaluate the nasal epithelial transcriptome of PARDS patients found that one subgroup of patients had loss of epithelial stem cells genes and those patients had a longer course of PARDS.⁹ Another analysed tracheal aspirates of intubated PARDS patients using single-cell RNAseq and found a subset of monocytes that expressed repair related transcripts.¹⁰ These monocytes were more common in patients with a lower level of illness severity and were hypothesised to be critical for the clearance of pro-inflammatory cells thus paving the way for repair and regeneration.¹⁰ Additionally, there is evidence that cross talk between multiple different reparative macrophage phenotypes may help promote repair and regeneration in pediatric patients with severe lung injury.¹¹ Taken together, these studies have begun to delve into the cell populations, pathways, and cell–cell crosstalk involved in repair and regeneration in pediatric patients recovering from PARDS. These studies also demonstrate the promise of unbiased approaches to rapidly provide mechanistic insight and provide potential therapeutic targets.

Significant gaps in knowledge remain about the unique aspects of pediatric lung injury and recovery. For example, we do not know the unique aspects that ongoing development has on how a small child responds to and heals from PARDS. That process is likely different from how a young adult responds and heals from acute lung injury. For example, the signalling pathways above are critical for the normal lung development. How a disease process and recovery interact with ongoing development is unknown. Does this difference account for the better pulmonary outcomes of PARDS compared to ARDS? Future research in PARDS will need to focus on identifying mechanisms that explain the clinical heterogeneity intrinsic to ARDS & PARDS and how these mechanisms change over time with development. It will also be crucial to investigate the mechanisms critical to healthy regeneration and repair and how these mechanisms are influenced by a child's position on the developmental continuum. Identification of critical mechanisms for healthy regeneration and repair would present attractive targets for intervention to prevent fibrosis and promote recovery.