Covid-19: some patients will develop a fibrotic lung disease

Luglio 23, 2022

268

A new study by researchers from University of California-San Diego has found that the lung issues that many Post COVID individuals develop are similar to Idiopathic Pulmonary Fibrosis (IPF) with potential serious and fatal outcomes in months or years that follow.

The study findings were published in the peer reviewed journal: eBio Medicine.
https://www.thelancet.com/journals/ebiom/article/PIIS2352-3964(22)00366-8/fulltext

The major discovery we report here is that lung disease in severe COVID-19 resembles IPF, the most common form of ILD, at a fundamental level—showing similar gene expression patterns (ViP and IPF signatures), shared prognostic signatures (in circulating PBMCs), dysfunctional cell states (AT2 and monocytes), and dysfunctional AT2 processes (ER stress, telomere instability, and senescence)].

It is also noteworthy that this semblance between COVID-19 and IPF was identified through a comprehensive and unbiased computational approach (see Figure 1) which compared gene expression signatures in COVID-19 against transcriptomic datasets representing a plethora of neoplastic, granulomatous, allergic/infectious, and vasculopathic pathologic conditions of the lung.

Figure 1Study design: Artificial Intelligence-guided navigation of COVID-19 lung disease. (From top to bottom) Step 0: Over 45,000 human, mouse, and rat gene expression databases were mined using machine learning tools called Boolean Equivalent Correlated Clusters (BECC133) to identify invariant host response to viral pandemics (ViP). In the absence of a sufficiently large number of COVID-19 datasets at the onset of the COVID-19 pandemic, these ViP signatures were trained on only two datasets from the pandemics of the past (Influenza and avian flu; GSE47963, n = 438; GSE113211, n = 118) and used without further training to prospectively analyze the samples from the current pandemic (i.e., COVID-19; n = 727 samples from diverse datasets). A subset of 20-genes classified disease severity called severe-ViP (sViP) signature. The ViP signatures appeared to capture the ‘invariant’ host response, i.e., the shared fundamental nature of the host immune response induced by all viral pandemics, including COVID-19. Step1: The set of ViP/sViP signatures and a CoV-lung specific13 gene signature was analyzed on diverse transcriptomic datasets representing a plethora of lung diseases; these efforts identified COVID-19 lung disease to be the closest to Idiopathic pulmonary fibrosis (IPF); both conditions induced a common array of gene signatures. Step 2: Clinically useful whole-blood and PBMC-derived prognostic signatures previously validated in IPF27 showed crossover efficacy in COVID-19, and vice versa. Step 3: Gene signatures of alveolar type II (AT2) cytopathic changes that are known to fuel IPF were analyzed in COVID-19 lung, and predicted shared features were validated in human and hamster lungs and lung-organoid derived models. Step 4: Protein- protein interaction (PPI) network built using sViP and AT2 cytopathy-related signatures was analyzed to pinpoint ER stress as a major shared feature in COVID-19 lung disease and IPF, which was subsequently validated in human and hamster lungs.

Although two other diseases (sarcoidosis and tuberculosis) that are characterized by focal fibrosis or lower incidence of fibrosis were found to induce some of the COVID-19-associated lung signatures, IPF induced them all and matched COVID-19 closely due to the diffuse nature of fibrosis.

This finding suggests that the nature of the host immune response to injury and the proximal AT2 cytopathies we report here are unlikely to be mere sequelae of cellular and molecular events during any fibrotic remodeling of the lung and may instead be specific to ILDs such as IPF and COVID-19.

Our findings support the following model (see Figure 8d): Upon injury (viral infection/inflammatory cytokines), diffuse alveolar damage is associated with extensive loss of AT1 cells, which are duly replaced by AT2 cells that serve as progenitors that are capable of self-renewal and differentiation into AT1 cells.90, 91, 92

Figure 8Induction of ER stress in AT2 cells is observed in COVID-19 and is sufficient to mimic the host immune response in COVID-19 and IPF. a-b. FFPE uninfected (control) and SARS-CoV-2-infected human (A) and hamster (B) lungs were analyzed for GRP78 expression by IHC. Representative fields are shown on the left. Scale bar = 200 µm. Images were quantified by IHC profiler (ImageJ) and displayed as violin plots on the right. c. Bubble plots show the ROC-AUC values (radii of circles are based on the ROC-AUC) demonstrating the direction of gene regulation (Up, red; Down, blue) for the classification of WT vs GRP78-KO murine lung (bottom row) and uninfected vs. infected hamster lung (top row) samples based on the signatures (below). d. Schematic summarizes the findings in this work and the proposed working model for the contributions of various alveolar cells in fueling the fibrotic progression in both IPF and COVID-19. The significance of the AUC was determined by Welch’s t-test; * = p < 0.05, ** = p < 0.01, and *** = p < 0.001.

But such physiologic regeneration and healing process is impaired when AT2 progenitors are also injured and begin to elicit ER stress responses, which induces telomerase activity.93,94 Because severe COVID-19 has been linked to short telomere lengths94 [not as a cause, i.e, SARS-CoV-2 infection does not shorten telomeres, but as a predisposition95], it is possible that patients who have short and/or dysfunctional telomeres fail to adequately respond to ER stress, and instead, become senescent (through p53 activation), accumulate DNA damage and enter a dysfunctional KRT8+ transitional stem cell state.96

The latter is a phenomenon that has been independently reported by three groups, under different names: PATS, pre-alveolar type-1 transitional cell state69; ADI, alveolar differentiation intermediates68; DATP, damage-associated transient progenitors.67 All three groups reported their presence in fibrotic regions of IPF lungs; an increase of AT2 cells in the transitional state was invariably accompanied by an increase in myofibroblasts.

Our PPI network analyses suggested that AT2 senescence and progenitor arrested state in the setting of a ViP/sViP-immune response (cytokine storm contributed by PBMCs) may support a SASP phenotype, which is a pathological feature of aging and IPF lung.97,98

It is possible that fibroblasts/myofibroblasts also contribute to this vicious cycle of inflammation and AT2 dysfunction. These findings show that these two distinct clinical syndromes, IPF, which predates the current pandemic by many decades, and the novel COVID-19, share a similar profile of host immune response, both in the lung microenvironment (in AT2 cells, to be specific) as well as in the circulating blood/PBMCs.

We not only formally define the nature of that host response and provide computational tools to measure the extent of such response, but also chart the cascade of cytopathic changes in the alveoli that are critical for the profibrogenic state.

It is noteworthy that the same host immune response is seen also in MIS-C, a new disease that co-emerged with COVID-19, and in KD (which shares overlapping features with MIS-C in clinical presentation35).

There are three major impacts and/or implications of the findings reported in this study.

First, our finding that COVID-19 and IPF share fundamental host immune response and alveolar cytopathic features is in keeping with the fact that both diseases share epidemiologic similarities—they primarily impact older adults, males more than females, and are characterized by progressive worsening of dyspnea and lung function.99,100

The induction of ViP/sViP signatures in IPF is consistent with gathering consensus in the past decade that IPF may be a multi-trigger infection-driven chronic inflammatory condition.36,101, 102, 103 Finally, patients with existing ILDs have greater odds of death due to COVID-19 compared with adults without ILD, even after controlling for age, sex, and comorbidities.104, 105, 106

On day 30 of COVID-19, 35% of patients with fibrotic idiopathic ILD had died compared with 19% of those with other ILDs.107 Whether the shared molecular or pathophysiologic features we report here synergize to increase fatality remains unknown; however, these shared features support the rationalization of clinical trials in COVID-19 using FDA-approved drugs for IPF, nintedanib (Ofev®), and pirfenidone (Esbriet®) (NCT04856111 and NCT04653831), and anecdotal case reports and case series have already chronicled the benefits of their use in COVID-19 lung disease.108, 109, 110

Second, most severe COVID-19 patients develop pneumonia and hyperinflammation likely related to a macrophage activation syndrome111 commonly named “cytokine storm”. Although this storm has been implicated, and it is comprised of the generic mix of all the typical cytokines,10 which exact component is the pro-fibrogenic driver remains unclear.

By showing that an IL15/IL15RA-centric cytokine storm is the key shared phenomenon between CoV and IPF lungs, this study provides a link between hyperinflammation and the sequelae of fibrosis. Findings are consistent with prior reports of circulating IL15 as a biomarker for prognostication in IPF and other ILDs.112, 113, 114 As for what may be contributing to such a cytokine response, myeloid cells are likely to be major culprits, but dysfunctional AT2 cells cannot be ruled out.

Prior studies using bleomycin-challenged transgenic mice that lack the Telomeric Repeat Binding Factor 2 in the AT2 cells (Trf2Fl/Fl;Sftpc-CreER) have shown that AT2 cells with telomere dysfunction upregulate an IL15-centric pathways.115 Mutations in the two main components of the telomerase holoenzyme complex that is responsible for creating new telomeric DNA, Telomerase reverse transcriptase (TERT) or telomerase RNA (TERC), are major monogenic causes of pulmonary fibrosis.116

Patients with IPF have shortened telomeres; short telomeres and TERT/TERC-disease associated variants were associated with specific clinical and biological features and reduced transplant-free survival.116,117 Similarly, short telomeres increase the risk of severe COVID-19,95,118,119 pulmonary fibrosis, and poorer outcomes.119,120

It is possible that injury, DNA damage and IL15 signaling in AT2 cells are one component in the profibrogenic cascade and suggest that IL15-targeted therapeutics may be beneficial in the most severe cases of COVID-19 to prevent fibrotic sequelae.
Third, our finding that shared alveolar cytopathic changes (e.g., DNA damage, progenitor state arrest, SASP, and ER stress) fuel fibrogenic programs in both COVID-19 and IPF are insightful because AT2 cells are known to contain an elegant quality control feedback loop to respond to intrinsic or extrinsic stress; a failure of such quality control results in diverse cellular phenotypes (reviewed in88): ER stress,121 defective autophagy, mitochondrial dysfunction, apoptosis, inflammatory cell recruitment, profibrotic signaling, and altered progenitor function96,122,123 that ultimately converge to drive downstream fibrotic remodeling in the lung.

Prior work,70 which led to the discovery of the AT2-senescence signature (which we used here), had demonstrated that senescence of AT2 cells (not loss of AT2) is sufficient to drive progressive pulmonary fibrosis.

Others agree, and this is now an established pathophysiologic trigger in lung fibrosis.64,124, 125, 126 Our findings are in keeping with published work127 that suggests that AT2 senescence may be a targetable disease driver of lung injury in COVID-19.

Although AT2 senescence is a shared phenomenon, our PPI network analyses –which integrated AT2 processes with the immune responses (ViP signatures) – provided valuable clues into how platelet activation and thromboinflammation may be uniquely seen in the setting of COVID-19.

These findings are in keeping with recent publications128, 129, 130 which suggest that SARS-CoV-2 could induce epithelial senescence; senescent AT2 cells would then assume a SASP phenotype, which in turn led to neutrophil and platelet activation, and activation of the clotting cascade.128

As for the limitations of this study, a direct comparison of lung samples from patients who survived COVID-19 but went on to develop (or not) restrictive (fibrotic) lung disease was not possible due to the lack of such datasets at this stage of the pandemic. It also remains unclear how to model post-COVID-19 progressive lung fibrosis in vitro and hence, no attempt was made to do so. Our results suggest that AT2 cells alone may be insufficient for such modeling because a specific host immune response that is carried in the PBMCs is a clear determinant as to who progresses and who does not.

Although AT2-specific modulation of ER-stress pathway and SARS-CoV-challenged (treated vs untreated) hamsters were used to go beyond association and establish causation, our study did not attempt to inhibit/reverse fibrosis in COVID-19 by acting on any profibrogenic cellular pathway/process. Development of novel chemical matter/biologicals and validation of the therapeutic efficacy of such agents will take time, but if successful, our findings show that their benefits will likely transcend beyond PCLD into IPF and other fibrotic lung conditions such as IPF.

In conclusion, this transdisciplinary work provides insights into the pathogenesis of PCLD, formally defines the fibrogenic processes in the lung, and rigorously validates high value gene signatures or even targets (i.e., IL15, senescence pathways, etc.) to track and manipulate the same. The insights, tools, computationally vetted disease models, and biomarkers (prognostic gene signatures) identified here are likely to spur the development of therapies for patients with fibrotic interstitial lung disease of diverse causes, including IPF, all of whom have limited or no good treatment options.