1 Introduction

In any disease, the development of specific and effective medical treatments is strictly associated to the understanding of its pathogenic mechanism. Respiratory diseases such as bronchial asthma, Chronic Obstructive Pulmonary Disease (COPD), Idiopathic Pulmonary Fibrosis (IPF), lung cancer, and Acute Respiratory Distress Syndrome (ARDS) are no exception to this rule. In general, these clinical conditions are characterized by the activation, differentiation, proliferation, and persistence into the bronchi and/or lung tissue of inflammatory, immune and resident cells that triggered by specific risk factors (e.g., allergens, smoking, environmental exposures) may lead to different degrees of inflammation, fibrotic damage, and in some cases to the burgeoning of cancer. In these circumstances, several cellular signaling cascades are abnormally activated and many chemical and molecular mediators are released. Recent evidence has demonstrated the role of intracellular kinases such as the lipid kinase phosphatidylinositol-3-kinase (PI3K) in the pathogenesis of a large spectrum of disorders including respiratory diseases. Based on their structure, distinct phospholipid substrates and different regulatory activity, PI3K is divided into classes I, II, and III. Class I is further divided into two subclasses: subclass IA, which includes the isoforms p110α, p110β, and p110δ, and subclass IB which is only composed of the isoform p110γ (Cantley 2002; Bi et al. 2002). Increased activity or mutations of PI3K results in an enhancement of phosphorylated protein kinases such as AKT that is strictly associated with the pathogenesis of different cancers including lung cancer (Samuels and Ericson 2006). PI3Kβ has instead been involved in the pathogenesis of colon cancer and its main activity is related to the regulation of the formation of blood clots. PI3Kδ is expressed by circulating leukocytes (mast cells, neutrophils, eosinophils, T lymphocytes) and endothelial cells. In a similar manner, PI3Kγ expression is largely restricted to immune cells, although some expression has been also detected in structural cell populations such as fibroblasts, and smooth muscle cells (Chan et al. 2021). Therefore, PI3Kδ and PI3Kγ are particularly involved in respiratory diseases such as asthma or COPD, characterized in their pathogenesis, by the participation of immune and inflammatory cells.

The role played by the PI3K family in the biological modulation of crucial cellular functions entails that any dysfunction of this regulatory activity could be associated with a state of disease, suggesting that the inhibition of PI3K could represent an ideal therapeutic target for the treatment of different diseases (Manning and Cantley 2007). Recent studies have highlighted that the PI3K pathway is also dysregulated in different lung pathologies, spanning from those characterized by inflammation and/or fibrosis to cancer (Vivanco and Sawyers 2002; Janku et al. 2018).

In this chapter, the role of the PI3K signaling pathway in the pathogenesis of different lung diseases will be reviewed with special attention to the more recent evidence of pharmacological approaches based on the inhibition of this pathway.

2 PI3K and Bronchial Asthma

2.1 Background

It has been extensively demonstrated that an imbalance between Th1 and Th2 lymphocytes in favor of Th2 cytokines such as IL-4, IL-5, and IL-13 is crucial for allergen-induced B cell production of IgE and for the recruitment of eosinophils, both crucial steps in the pathogenesis of allergic bronchial asthma (Galli and Tsai 2008). Airway inflammation is initially triggered by the interaction between allergen-specific IgE and mast cells. The subsequent activation of mast cells causes the expression and release of mediators and cytokines that promote airway bronchoconstriction, mucus overproduction, bronchial hyperresponsiveness, and amplify the synthesis of immunoglobulin E (IgE). Furthermore, the activation of eosinophils, through the release of other inflammatory mediators, is responsible for chronic airway inflammation and bronchial remodeling, quintessential features of asthma (Hakim-Rad et al. 2009; Boyce et al. 2009).

2.2 Are the Expression and Functional Profile of PI3K Relevant to the Pathogenesis of Bronchial Asthma?

Several in vitro and in vivo studies have suggested that the expression and functional profile of PI3K is relevant for the pathogenesis of allergic bronchial asthma (Ito et al. 2007). Most of the evidence produced has used both non-selective and selective inhibitors of PI3K that have indirectly demonstrated the fundamental role of PI3K in many pathogenic aspects of bronchial asthma. Kämpe et al. (2012) investigated the role of PI3K in regulating eosinophil and neutrophil degranulation in allergic rhinitis and asthma. These cells were incubated with wortmannin, a non-selective PI3K inhibitor, and then stimulated with C3b-coated sephadex particles. Wortmannin reduced in a dose-dependent way the release of eosinophil cationic protein (ECP), eosinophil peroxidase (EPO), and myeloperoxidase. However, this effect was lower in cells derived from allergic asthmatic patients during pollen season or after bronchial challenge, suggesting that in these conditions, in addition to PI3K, other signaling pathways might be involved. Other PI3K class I pan-inhibitors including CL27c, LY294002, and the same wortmannin have been shown to reduce eosinophil, neutrophil, and lymphocyte activity and recruitment in various experimental murine models of asthma or allergic inflammation. CL27c showed anti-inflammatory effects in a murine acute model of allergic asthma. This prodrug, which can be administered by inhalation, decreases leukocyte infiltration into the lungs, reduces the number of inflammatory cells in broncho-alveolar lavage (BAL) as well as the production of inflammatory mediators, and most importantly, it exerts a significant improvement of lung function. This study is particularly important for the route of administration that has been used to treat animals. Local treatment may have the advantage of allowing the optimization of the therapeutic efficacy, but at the same time, it could reduce the risk of unwanted adverse events that a systemic treatment could cause (Campa et al. 2018). Other authors have demonstrated that intratracheal administration of LY294002 in OVA-sensitized mice significantly reduces inflammatory cell counts and IL-5, IL-13, and CCL11 (eotaxin) levels in BAL. In addition, a marked reduction of eosinophil infiltration and mucus production was described into the peribronchial and perivascular connective tissue. This was functionally accompanied by a reduction of OVA-induced airway hyperresponsiveness to inhaled methacholine. Furthermore, western blot analysis of lung tissue derived from OVA-challenged animals demonstrated a marked reduction of the phosphorylation of AKT, downstream mediator of the PI3K pathway. (Duan et al. 2005; Marwick et al. 2010).

In the same OVA-induced asthma animal model, Kwak and coworkers have shown a significant increase in PI3K activity after allergen challenge and a parallel dramatic reduction of PTEN, known to counteract PI3K activity. They describe that the immunoreactivity for PTEN, localized in the bronchiolar epithelium in control mice, virtually disappeared in OVA-challenged lungs. Once again, the intratracheal administration of either wortmannin or adenoviruses carrying PTEN were capable of reducing IL-4, IL-5, and ECP levels in BAL as well as airway inflammation and bronchial hyperreactivity (Kwak et al. 2003).

2.3 Are PI3K Inhibitors Safe and Clinically Relevant in Allergic Bronchial Asthma?

Taken together, these studies provide further support to the hypothesis that the PI3K pathway plays a major role in the pathogenic mechanisms underlying asthma, suggesting that the inhibition of PI3K could represent an ideal approach for the treatment of asthma (Ito et al. 2007; Duan et al. 2005; Yoo et al. 2017). This observation, although very interesting and apparently promising, has raised many concerns on the use of non-selective inhibitors for therapeutic purposes. PI3K pan-inhibitors, blocking all PI3K isoforms, may cause systemic adverse events precluding their clinical use. New isoform-selective PI3K inhibitors, expected to be less toxic, have been already developed with the aim of representing more specific and effective inhibitors and, at the same time, safer molecules. This is largely based on a growing number of reports showing a specific involvement of isoforms γ and δ in airway inflammation and asthma. Indeed, several works have shown that PI3K isoforms γ and δ are upregulated during lung inflammation and airway remodeling, suggesting a major role for these two isoforms in the pathogenesis of asthma (Fruman and Bismuth 2009; Kok et al. 2009; Rommel et al. 2007).

In 2003, the first PI3Kγ-selective inhibitor, a 5-phenylthiazole derivative, was eventually developed, opening a new, possible therapeutic approach for respiratory disorders marked by airway inflammation. From then on, a variety of single or combined selective-PI3Kγ-PI3Kδ inhibitors have been tested to assess their efficacy and safety in in vitro experimental settings first, then in animal models and more recently in human clinical trials (Bruce et al. 2003).

PI3Kγ is involved in the modulation of mast cell degranulation, ROS release, eosinophil recruitment, neutrophil activation, and Th2 cytokine production (Ghigo et al. 2010). Laffargue et al. (2002) have demonstrated that mast cells derived from PI3Kγ-deficient mice have a weaker degranulation and a reduced ability to respond to IgE receptor cross-linking. In addition, animals passively immunized and antigen-challenged were protected from anaphylaxis. Based on their results these authors hypothesize that PI3Kγ, through its activity on mast cells, has a crucial role in the modulation of allergic inflammation. Moreover, in the same animal model, Wymann et al. (2003) observed reduced eosinophil recruitment at 48 hours post-challenge compared with control group, suggesting that PI3Kγ plays also a role in the maintenance of eosinophilic inflammation in vivo. It is worth reminding here that the permanence of eosinophils within the bronchial tissue is crucial for the establishment of two typical features of severe asthma, chronic airway inflammation and bronchial remodeling.

PI3Kδ is also involved in the early phases of activation and allergen-IgE-induced mast cell degranulation, maturation, and differentiation of Th1, Th2, Th17, Treg cells, B lymphocyte activation, and antibody production. Indeed, T cells isolated from transgenic mice characterized by the absence of PI3Kδ catalytic activity (p110δ D910A/D910A T cells) showed reduced Th1 and Th2 differentiation after stimulation in vitro with anti-CD3 and anti-CD28 (Hawkins and Stephens 2015; Okkenhaug et al. 2002).

Mutation of PI3Kδ also induces alterations in mast cells and neutrophils, in allergen-IgE-induced degranulation and cytokine release. Indeed, inactivation of PI3Kδ restores/protects anaphylactic allergic responses in vivo (Ali et al. 2008; Puri et al. 2004). In this respect, it has been shown that IC87114, a selective inhibitor of the δ isoform, reduces lung tissue eosinophils, leukotriene C4, mucin overproduction, release of IL-4, IL-5, IL-13, vascular cell adhesion molecule-1 (VCAM-1), Chemokine (C–C motif) ligand 5 (CCL5), and CCL11. The same compound is also able to reduce the production of IgE in an OVA animal model of asthma (Lee et al. 2006a).

One of the prevailing features of airway inflammation and bronchial asthma is mucosal edema, mainly due to an increase in vascular permeability. Lee et al. (2006b) have clearly shown that in the murine OVA model of asthma, IC87114, a selective PI3Kδ inhibitor, reduces OVA-induced airway infiltration of inflammatory cells, production of Th2 cytokines, airway hyperresponsiveness, and vascular permeability. Furthermore, the selective inhibition of PI3Kδ reduces the increase of vascular endothelial growth factor (VEGF) levels induced by OVA challenge. Based on these results the authors hypothesized that PI3Kδ, through the inhibition of VEGF, prevents vascular leakage and the formation of mucosal edema, thus diminishing airway inflammation.

In this model, PI3Kδ inhibitors such as IC87114 (El-Hashim et al. 2017) and INK654 (Kim et al. 2020), as well as a double-selective compound blocking both PI3Kγ and PI3Kδ named TG100-115, effectively attenuate allergic bronchial inflammation and airway hyperresponsiveness reducing eosinophils, IL-4, IL-5, and IL-13 levels, total serum IgE (Lee et al. 2006a), airway mucus hypersecretion and, most importantly, improved lung function (Doukas et al. 2009). In addition, IC87114, by blocking VEGF upregulation, decreases vascular permeability and extravascular remodeling, responsible for asthmatic submucosal edema in bronchial tissue (Lee et al. 2006b). Recently, Jing Bi et al. have hypothesized that glucocorticoid insensitivity, a main feature of severe asthma and COPD, might be also related to the regulatory activity of PI3K. In this in vitro study, they observed that peripheral blood mononuclear cells (PBMCs) derived from patients with severe asthma had an evident dose and time-dependent insensitivity to glucocorticoids compared to mild asthma or normal individuals, which was associated with a diminished activity of histone deacetylase 2 (HDAC2) and elevated expression of pro-inflammatory genes such as subunits of the transcription factors NFκB and AP1. Interestingly, BEZ235 (Dactolisib) and LY294002, a dual γ/δ and a PI3K pan-inhibitor, respectively, improved glucocorticoid responsiveness in PBMCs of patients with severe asthma. The results presented by Bi et al. (2020). suggest that PI3K, and more specifically the γ/δ isoforms, could be directly involved in the lack of sensitivity of inflammatory cells to the activity of steroids which is one of the major clinical problems of severe asthma, but certainly also relevant for COPD. A reduction of HDAC activity, at least in COPD, has already been shown to be one of the main reasons for the steroid insensitivity that characterizes this disease. However, this intriguing study has the limitation of being an in vitro study and certainly needs to be confirmed by observations produced in vivo. During the last few years some interesting clinical studies have been performed, that investigated the efficacy of selective PI3K inhibitors in allergic rhinitis and allergic bronchial asthma. Idelalisib is a PI3Kδ inhibitor already approved for the treatment of some hematologic diseases with a recognized inhibitory activity toward allergen-induced activation of mast cells and basophil degranulation. In a phase 1 randomized, double-blind, placebo-controlled study, after 7 days of oral treatment with idelalisib nasal symptom score was registered in allergic rhinitis patients and compared to placebo group. After allergen challenge, idelalisib induced a significant improvement of symptoms, nasal airflow and nasal secretion. In this study, clinical data are consistent with the anti-inflammatory profile of idelalisib. Even so, authors of this study were very cautious on suggesting a regular use for idelalisib in allergic rhinitis, mainly because the study period was very short and did not consider the effect of a longer or chronic treatment as required in allergic rhinitis (Horak et al. 2016).

In a randomized, double-blind, placebo-controlled study, the efficacy and safety of inhaled nemiralisib (GSK2269557), another PI3Kδ inhibitor, were investigated in patients with uncontrolled asthma. The safety profile was good, with exception of some post-inhalation cough that was the most common side effect. After 14 days of treatment, nemiralisib reduced the levels of several pro-inflammatory cytokines such as IL-5, IL-6, IL-8, and IL-13 in sputum. Disappointingly, no significant improvement of lung function was observed. As stated by the authors of the study it is evident that a number of variables may have affected the result of this interesting trial including patient selection, sample size of the study, or duration of treatment (Khindri et al. 2018).

2.4 Closing Remarks

During the last few years, the role of PI3K in allergic bronchial asthma has been thoroughly investigated. Virtually all pathogenic steps of allergic bronchial asthma, from airway inflammation to bronchial remodeling are, at least in part, regulated by the PI3K signaling pathway. This was largely demonstrated by in vitro and preclinical experimental models using different non-selective and selective PI3K inhibitors. In particular, selective δ and γ inhibitors have shown to be particularly effective, and when administered locally by inhalation, reasonably safe. Despite all these positive results, very few functional, clinical advantages have been described. In the face of a great number of in vitro and in vivo studies, very few clinical trials have been performed so far. Different reasons may explain, at least in part, this disappointing “scenario”. Current traditional therapies for allergic bronchial asthma are in most of the cases very effective, determining a substantial lack for an urgent need of new alternative treatments. Thanks to the use of more selective PI3K inhibitors and the possibility to administer them by inhalation some safety issues may be solved. PI3K is certainly one of the main actors in the pathogenesis of asthma, but very likely not the only one, as often happens in biology, other redundant pathways may mimic or intervene when the PI3K pathway is blocked. Hopefully, this will be better investigated in the near future and new clinical trials examining the effect of PI3K inhibitors will be designed.

3 PI3K and Chronic Obstructive Pulmonary Disease (COPD)

3.1 Background

Chronic Obstructive Pulmonary Disease (COPD) is a condition characterized by chronic inflammation, emphysema, and fibrosis of small airways, leading to a slowly progressive, but irreversible lung damage with obstructive respiratory failure (Barnes et al. 2015). COPD represents one of the leading causes of death worldwide. It is associated with smoking or environmental and professional exposures; additional risk factors include age, low or high body mass index, low education level, and poverty (Osman et al. 2017). A genetic predisposition was supposed for the association of emphysema with some Mendelian syndromes as α1-antitripsin deficit, but more likely the genetic background should be assessed in response of the subject to smoke: no-smoking first-degree relatives of COPD patients have a similar risk of developing COPD as the general population, whereas smoking first-degree relatives had a three-fold increased risk (Barnes et al. 2015). Smoking and other “irritants” activate pattern recognition receptors, enabling an innate immune response characterized by an increased number of macrophages and neutrophils in the lungs. These cells show an impairment in phagocytosis with an incomplete clearance of bacteria and apoptotic cells, favoring bacterial colonization, chronic inflammation, and activation of airway epithelial cells and mucus secretion. The action of smoking on innate immune cells persists also after smoking cessation. In successive phases, adaptive immunity also plays a role, with an increased number of mast cells, B and Th1 lymphocytes, that amplify the process (Brusselle et al. 2011). The main pathogenic mechanism is related to oxidative stress, induced by the “irritating” action of smoking on bronchial and alveolar cells. Reactive Oxygen Species (ROS) activate inflammatory genes and proteases through nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB) and mitogen-activated protein kinase (MAPK) and inhibit endogenous antiproteases such as α1-antitrypsin. The increased proteolysis, together with the activation of B cells, generate autoantibodies able to increase inflammation and its persistence. ROS activates transforming growth factor beta (TGF-β) response with increasing fibrosis, causes DNA damage, also interfering with the mechanism of repairing, and decreases the activity of sirtuin-1, favoring an accelerated cellular aging (Barnes et al. 2015).

3.2 Are the Expression and Functional Profile of PI3K Relevant to the Pathogenesis of COPD?

Different signaling pathways such as the Janus kinase-signal transducer and activator of transcription (JAK/STAT) pathway (Montero et al. 2021), TGF-β signaling pathway (TGF-β/Smad), MAPK pathway, NFκB, and PI3K are activated in COPD (Manley et al. 2019; Edwards et al. 2009; Moradi et al. 2021). Several studies have shown that PI3K class I isoforms and specifically PI3Kγ and δ, play a relevant role in the physiopathology of COPD. Although all class I PI3Ks are expressed in leukocytes, PI3Kγ and PI3Kδ are the main and specific mediators in innate and adaptive immune responses. Therefore, it is plausible that PI3K signaling may affect intercellular cross-talks between alveolar immune cells and airway epithelial cells in COPD (Wymann and Marone 2005; So and Fruman 2012). It has been described that PI3K and its downstream mediators are upregulated in primary human bronchial epithelial cells facilitating chronic airway inflammation, mucus secretion, and ultimately alveolar wall damage (Ito et al. 2007; Yoo et al. 2017). Moreover, lungs of COPD patients show exaggerated immune responses with high levels of monocyte infiltration, alveolar macrophage activation, and neutrophil migration due to an increased activation of PI3Kδ (Sriskantharajah et al. 2013). Expression of PI3Kδ and AKT phosphorylation is enhanced in COPD alveolar macrophages. It is oxidative stress that directly induces PI3Kδ-dependent AKT activation, leading to the loss of steroid sensitivity, which is one of the main clinical features of COPD (Barnes 2016). Both γ and δ isoforms, controlling the interactions between leukocytes and epithelial/endothelial cells also regulate chemotaxis and migration of infiltrating cells, such as neutrophils (Okkenhaug et al. 2002; Hirsch et al. 2000). Their activity has been demonstrated in various types of inflammatory cells as described by several experiments with knock-out animals or with pan-PI3K or isoform-specific inhibitors (Fruman and Bismuth 2009; Kok et al. 2009; Rommel et al. 2007; Wymann and Schneiter 2008). Therefore, PI3K isoforms are considered promising therapeutic targets in the development of future treatments for inflammatory lung diseases such as COPD (Ghigo et al. 2010).

The PI3K class I pan-inhibitors, LY294002 and wortmannin, effectively reduce total lymphocyte counts and chemotaxis of neutrophils in vitro and in various murine experimental models of COPD. In vitro studies show that LY294002 reduces neutrophil migratory speed, whereas wortmannin induces alveolar stem cell differentiation and alveolar repair in elastase-induced COPD or in cigarette smoke-exposed mice (Sapey et al. 2011). In this study, the alveolar repairing effect of wortmannin was evaluated in murine models of COPD using CT scans and histology. The radiologic score evaluated in terms of lung density (Hounsfield Units values) showed a complete recovery of animals treated with wortmannin. This was confirmed by histology and by the improvement in lung function suggesting a potential role for PI3K inhibition in COPD.

Different PI3K inhibitors, such as ZSTK474 and GSK045, have been studied in vitro by Gupta et al. (2016) for their inhibitory effects on the production of matrix metalloproteinase-9 (MMP-9) and extracellular reactive oxygen species (ROS) from cultured blood and sputum neutrophils isolated from COPD patients. Blood and sputum neutrophils derived from COPD patients were incubated with dexamethasone, a pan PI3K inhibitor (ZSTK474), or a selective PI3Kδ inhibitor (GSK045). After a period of incubation with the different drugs, neutrophil production of MMP-9 and ROS was measured. GSK045 significantly inhibited MMP-9 and ROS release from neutrophils derived from patients with either a stable disease or during a disease exacerbation. It is important to note that in the same model the effect of a MAPK inhibitor and that of dexamethasone were lower. Authors concluded that the selective inhibition of PI3Kδ could be “a potential strategy for targeting glucocorticoid insensitive MMP-9 and ROS secretion from COPD neutrophils”.

Moreover, IC87114 and theophylline, selective PI3Kδ inhibitors, upregulate HDAC2 expression in vitro and restore steroid response in PBMC from COPD patients (To et al. 2010). Furthermore, Ford et al. (2010) have shown that theophylline increases HDAC2 expression and corticosteroid sensitivity both in vitro and in animal models of COPD, through a specific inhibition of PI3Kδ. PI3Kδ is known to be activated in COPD lungs by oxidative stress, for those selective inhibitors of PI3Kδ mimic the effects of theophylline in restoring corticosteroid sensitivity.

3.3 Are PI3K Inhibitors Safe and Clinically Relevant in COPD?

Some novel PI3Kδ inhibitors including LAS191954 and IHMT-PI3Kδ-372 have been recently evaluated for their activity in airway inflammation and COPD. LAS191954 was tested in a rat model of ConA-induced IL-2 production showing a significant reduction of T cell cytokine production. Feng Li et al. (2020). instead, concentrated their attention on IHMT-PI3Kδ-372, a newly discovered and potent PI3Kδ inhibitor. Inhaled IHMT-PI3Kδ-372 was administered to animals exposed to cigarette smoke; this animal model mimics the inflammatory changes observed in COPD. They described a dose-dependent improvement of lung function as well as an increase in arterial oxygen saturation (PaO2) and a concomitant decrease of arterial carbon dioxide (PaCO2). Inflammatory cytokines such as IL-1, IL-6, IL-8, and TNF as well as the number of alveolar macrophages in BAL were also reduced in a dose-dependent manner. The efficacy, at least in animal models, and the possibility to use IHMT-PI3Kδ-372 by inhalation, likely reducing the possibility of systemic side effects, makes this molecule a good candidate for further studies addressed to the treatment of COPD.

More recently, based on the multitude of experimental and preclinical data supporting the role of PI3K in COPD, the first clinical trials have been performed mainly aimed to evaluate primarily safety and then efficacy. Cahn et al. designed a randomized, double-blind, placebo-controlled study where GSK2265557 was given by inhalation once a day for 14 days, and during this period, possible side effects were recorded as well as cytokine levels in sputum and lung function. The most common adverse events were cough and headache, no other serious side effects were reported. Cytokine levels were reduced in sputum, but even so, airway resistance (Raw) and airway conductance were disappointingly not affected by the treatment. The authors conclude that their results encourage further studies aimed to explore the therapeutic role of PI3K in COPD. Another recent clinical trial evaluated the safety and efficacy of nemiralisib, a PI3Kδ inhibitor, in COPD patients during an acute exacerbation. It was a double-blind, placebo-controlled study. COPD patients having an exacerbation of the disease inhaled nemiralisib once daily for 12 weeks. During this period, safety profile, lung function, rate of new exacerbations, and rescue medication used, were recorded. The most common side effect was cough, particularly in patients receiving higher doses of the drug. Lung function, rate of new exacerbations, use of rescue medication were not statistically different in placebo- versus nemiralisib-treated groups (Fahy et al. 2021).

3.4 Closing Remarks

The possibility to exploit the PI3K pathway for the treatment of COPD has a solid scientific rationale. Based on a variety of evidence, although almost exclusively obtained by in vitro and animal model studies, the inhibition of this pathway can restore the sensitivity to corticosteroids, induce a reduction of airway inflammation and increase the alveolar repairing activity. Unfortunately, in none of the studies where respiratory functional parameters were part of the required outcomes, neither in the animal models nor in human studies, a significant improvement was described. Differently from bronchial asthma where medical treatments are very effective, in COPD there is an urgent need for new and radical approaches capable of changing the course of this devastating disease. Nevertheless, the number of clinical studies investigating the effect of PI3K inhibitors in the clinical context of COPD is surprisingly low. In the few studies performed so far, thanks also to the possibility of administering the PI3K inhibitors by inhalation, the safety profile has been considered acceptable and limited in most of the cases to some post-inhalation cough. Hopefully, in the near future more clinical studies will explore in detail the opportunity to use PI3K inhibitors for the treatment of COPD.

4 PI3K and Idiopathic Pulmonary Fibrosis

4.1 Background

Idiopathic pulmonary fibrosis (IPF) is a chronic, interstitial lung disease of unknown etiology with a median survival of 3–5 years from diagnosis. IPF is characterized by bronchiolar-alveolar epithelial-mesenchymal transition, fibroblast activation, excessive deposition of collagen and extracellular matrix, lung tissue fibrosis, and eventually by the decline of pulmonary function and respiratory failure (Raghu et al. 2011). In the past, IPF was commonly considered an inflammatory-driven disease, nowadays it is rather identified as a primarily fibrotic process initially triggered by an epithelial chronic injury that causes an altered and exaggerated activation of fibroblasts. Although the pathogenesis of IPF is still unclear, genetic and epigenetic factors along with recognized risk factors such as smoking, are considered crucially important for the onset of the fibrotic process (Puglisi et al. 2016). These factors, together with repeated damage of the alveolar epithelium and an altered repairing process, are responsible for an abnormal activation, differentiation, and proliferation of mesenchymal cells. The origin of mesenchymal cells is variable, deriving from alveolar epithelium, resident fibroblasts, and circulating fibrocytes. Many cytokines such as TGF-β and PDGF actively contribute to the proliferation of fibroblasts and the differentiation of these cells into myofibroblasts (Sgalla et al. 2018; Spagnolo et al. 2015). Considering the large variety of pathogenic and clinical similarities, IPF has been, at least in some aspects, compared to cancer (Vancheri 2013).

So far only two drugs, pirfenidone, and nintedanib, have been approved for the treatment of IPF. Both drugs slow disease progression, increase survival but are not able to stop or reverse the relentless course of the disease (Fernández Fabrellas et al. 2018). For this reason, research is strongly focused on the identification of new targets amenable to therapeutic intervention.

4.2 Are the Expression and Functional Profile of PI3K Relevant to the Pathogenesis of IPF?

In the last few years, several studies have revealed an involvement of the PI3K pathway in the pathogenesis of IPF. In vitro studies have demonstrated, for the first time, the central role of this signaling pathway in regulating the proliferation and differentiation of human lung fibroblasts into myofibroblasts induced by TGFβ (Conte et al. 2011). It was also described that PTEN negatively regulates myofibroblast differentiation both in vitro and in vivo (White et al. 2006; Xia et al. 2008). This was confirmed by Le Cras et al. who showed that treatment with the pan-PI3K inhibitor PX866, prevents the progression of lung fibrosis induced by TGFα in vivo (Cras et al. 2010). Specifically, the involvement of the PI3Kγ isoform in lung fibrogenesis was evaluated demonstrating that mice lacking PI3Kγ are protected against bleomycin-induced pulmonary injury and displayed higher survival rates than wild-type mice. AS605240, a specific inhibitor of PI3Kγ, also decreases the production of collagen in response to CXCL1 and CCL2 in lung fibroblasts of mice instilled with bleomycin (Russo et al. 2011). Furthermore, overexpression of the PI3Kγ isoform has been observed in IPF lung homogenates and in fibrotic human lung fibroblasts. Moreover, gene silencing and pharmacological inhibition of PI3Kγ with AS252424 significantly inhibited cell proliferation and α-SMA expression in IPF fibroblasts (Conte et al. 2013). Recently, omipalisib/GSK2126458, a selective inhibitor of class I PI3Ks and mTOR, has been repositioned from oncology to IPF for its anti-fibrotic effects. Biochemical activity studies revealed the specificity of this compound for the PI3Kα isoform showing as well some degree of activity against isoforms β, γ, and δ (Knight et al. 2010). Indeed, omipalisib decreased fibroblast proliferation and TGFβ-induced collagen synthesis in primary human lung fibroblast cultures. Moreover, omipalisib reduced AKT phosphorylation in IPF BAL cells and ex vivo IPF lung tissue (Mercer et al. 2016). Omipalisib has been also investigated in a phase 1 clinical trial (NCT01725139), with 17 IPF patients enrolled, where it displayed a good safety profile (Lukey et al. 2019).

4.3 Are PI3K Inhibitors Safe and Clinically Relevant in IPF?

A critical aspect of PI3K inhibitors is the induction of serious side effects, mainly hyperglycemia and gastrointestinal toxicity, which often result in treatment discontinuation. A very interesting approach with PI3K inhibitors for the treatment of IPF was made by Hettiarachchi et al. (2020). To increase efficacy, possibly diminishing toxicity associated to PI3K inhibitors, they thought to directly target myofibroblasts. Myofibroblasts, mainly present in fibrotic tissues, at the invasive front of cancer and during wound healing, express on their membrane a protein named fibroblast activation protein (FAP) that is virtually a specific marker for these cells (Acharya et al. 2006). Any molecule or drug associated to a ligand for FAP can be the ideal way to selectively target myofibroblasts. These authors, with the aim of inhibiting the activity of myofibroblasts in a selective manner, investigated the effect of omipalisid bound to a FAP ligand (FAPL) in the mouse model of bleomycin-induced fibrosis. With all the limitations of this model that resemble only in part to human IPF, the PI3K inhibitor plus the FAP ligand complex significantly reduced collagen production and the development of lung fibrosis (Hettiarachchi et al. 2020).

Campa et al. (2018) proposed an alternative strategy to reduce toxicity of PI3K inhibitors and maximize their therapeutic efficacy using local and not systemic administration. CL27c, a prodrug pan-PI3K inhibitor, was chemically designed for local therapy and administered by inhalation. Its efficacy was investigated in a murine model of pulmonary fibrosis and asthma. CL27c inhaled, as an aerosol, once activated inside the cytoplasm of lung cells in its active form CL27e, inhibits PI3K activity only locally, minimizing the toxic effects of systemic exposure and possibly increasing efficacy. Indeed, CL27c was able to reduce lung damage and prevent animal mortality, even when administered in a therapeutic setting. Recently, 2-amino-4-methyquinazoline derivates (5d, 5e, and 5 g), a new class of PI3K inhibitors, have been synthesized and tested for their anti-proliferative activities in mouse MLg2908 lung fibroblasts, through the inhibition of the main PI3K downstream signaling pathway such as AKT, p70S6K, and S6RP. In an in vivo fibrotic model, oral administration of 5d suppressed the α-SMA and hydroxyproline increase in lung homogenates and improved the deterioration of lung function induced by bleomycin (Lin et al. 2019).

The PI3K/AKT/mTOR axis is also involved in the mechanism of autophagy. Autophagy is a physiological process that allows the elimination of misfolded proteins or damaged cell components by lysosomal vesicles that plays an important role in tissue development, defense against pathogens, and in counteracting cellular aging. An autophagy deficit has been linked to the development of several diseases, including IPF (Todde et al. 2009). In the course of IPF there are several factors that contribute to increased autophagy, including oxidative stress (Kliment and Oury 2010). TGFβ is one of the inducers of oxidative stress in IPF by increasing reactive oxygen species production and reducing the production of antioxidant enzymes, causing an imbalance in favor of ROS formation (Liu and Desai 2015). TGFβ inhibits autophagy in fibroblasts via the activation of the PI3K/AKT/mTOR axis, for that rapamycin administration results in decreased autophagy and increased fibrosis in bleomycin-induced models (Patel et al. 2012). Reduced autophagy results in decreased collagen degradation and proliferation of fibroblasts together with differentiation into myofibroblasts, cells with a pivotal role in the genesis of IPF (Zhao et al. 2020).

Based on these premises, the PI3K/AKT/mTOR axis could represent a promising potential therapeutic target for IPF. Lukey et al. (2019) investigated the role of omipalisib, in a randomized, placebo-controlled, double-blind study in patients with IPF. The study was exclusively addressed to test safety and pharmacokinetics. Omipalisib, given twice daily to the 17 patients enrolled in the study, exerted a dose-dependent inhibition of the PI3K/mTOR pathway both in blood and BAL cells. The safety profile was good, limited to mild diarrhea in four patients. Although there are no specific data on efficacy, it is interesting to note that 18F-2-fluoro-2-deoxy-D-glucose(FDG)-positron emission tomography/computed tomography scans showed an exposure-dependent reduction in 18F-FDG uptake in fibrotic areas of the lungs. The results of this study, although limited to the evaluation of safety and pharmacokinetics of omipalisib should encourage the launching of other clinical studies in this field. Unfortunately, to date, only one randomized controlled trial is ongoing in IPF. HEC68498, a potent, high selective inhibitor of class I PI3K and mTOR is now under study in a phase I study. It is a double-blind, placebo-controlled trial, but no data are currently available (NCT03502902) (Clinical Trials gov 2021a).

4.4 Closing Remarks

Similar to bronchial asthma or COPD, the PI3K pathway is fully involved in all known pathogenic mechanisms underlying IPF. Many in vitro studies and in vivo animal models of lung fibrosis, clearly demonstrate that the inhibition of PI3K is related to the reduction of fibroblast activation, proliferation, and differentiation. In addition, the release of pro-fibrotic cytokines such as TGFβ or the production of collagen is downregulated. Despite this experimental evidence and a reasonably good safety profile, very few clinical trials have been designed for the evaluation of the clinical efficacy of PI3K inhibitors in IPF. Unfortunately, current therapies may prolong survival, but IPF remains a lethal disease. It is highly desirable that in near future more trials will be specifically designed to investigate the efficacy of PI3K inhibitors in IPF.

5 PI3K and Lung Cancer

5.1 Background

Lung cancer is the second more common cancer after breast cancer and represents the first cause of cancer mortality (Sung et al. 2021). Its classification was recently modified according to new knowledge regarding its molecular profile. It is divided into Small Cell Lung Carcinoma (SCLC), accounting for about 15% of all lung cancers, and Non-Small Cell Lung Carcinoma (NSCLC), including Adenocarcinoma, Squamous Cell Carcinoma, and Large Cell Carcinoma (Inamura 2017). Despite the efforts for an early diagnosis and the new therapeutic interventions, lung cancer remains frequently diagnosed in advanced stages, with a 5-year survival rate ranging to 20–25% (Leiro-Fernández et al. 2014). There is an urgent need for more effective treatments that could improve the current disappointing survival rates. In this view, the PI3K pathway is of great interest, due to its role in cell activation and replication.

5.2 Are PI3K Inhibitors Safe and Clinically Relevant in Lung Cancer?

Chemotherapy remains one of the main treatment strategies for lung cancer. In the last few years, the pharmacological treatment is switching from conventional chemotherapy to a personalized approach, mainly based on markers of cell activity displayed by cancer cells or against molecules belonging to pivotal cancer pathways. In this view, the PI3K/AKT/mTOR pathway is one of the most studied. A great number of mutations or imbalances have been found in molecules involved in this pathway in lung cancer cells, due to either an inhibition of down-regulatory systems or to a chronic activation of the pathway. The use of compounds capable of inhibiting the PI3K pathway, alone or in combination with other treatments, can also provide a way to overcome chemotherapy resistance (Brown and Toker 2015).

Sonolisib (PX-866) is a pan-inhibitor of PI3K, analog of wortmannin, that displays significant antitumor activity in xenograft models and a good safety profile in advanced tumors. The phase I/II study in association with docetaxel in NSCLC is completed but results are still pending (Hong et al. 2012; Clinical Trials gov 2021b). Another pan-inhibitor of PI3K is buparlisib (BKM-120), which also shows some activity on the inhibition of tubulin (Zhao et al. 2017; Wu et al. 2016; Xing et al. 2021). The dose administered was well tolerated at 100 mg/day and tested in few clinical trials in NSCLC (Wu et al. 2016; Xing et al. 2021; Trials and gov 2021; Clinical Trials gov 2021c, a, d, e, f; McGowan et al. 2019), alone or in combination with other molecules. Finally, in this family of PI3K pan-inhibitors, pilaralisib (SAR245408, XL147) resulted safe but not sufficiently effective (Wheler et al. 2017). ASN003 is instead a dual, PI3Kα and δ isoform inhibitor with a potent inhibitory activity on BRAF, a serine/threonine kinase mutated in about 4% of NSCLC (Bustamante and Otterson 2019). A phase I study was completed, but no other clinical trials are currently available on this compound (Bustamante and Otterson 2019; Clinical Trials gov 2022b). Similar action, but limited to the inhibition of PI3Kα and δ is played by pictilisib, which showed a good safety profile in an IB study in association with the standard of care for NSCLC. Progressing to phase II, this study is currently ongoing (Soria et al. 2017; Clinical Trials gov 2021g). Another PI3K inhibitor is taselisib (GDC-0032), specific for the α, γ, and δ isoforms. The molecule was tested in several solid cancers, but when used alone in NSCLC did not sufficiently improve the survival rate (Langer et al. 2019). Eganelisib (IPI-549) is a selective inhibitor of PI3Kγ, an isoform mostly expressed in immune cells. This isoform plays a critical role in cancer microenvironment, increasing the activity of double-negative T cells, myeloid-derived suppressor cells, and tumor-associated macrophages with potent immunosuppressant effects, aimed to escape the anticancer surveillance of the immune system (Hillhouse et al. 2013; Zhang et al. 2019). This indirect mechanism was exploited in the treatment of melanoma and head/neck tumors with encouraging results (Cohen, et al. 2020; Postow, et al. 2020). A phase I clinical trial on Eganelisib plus nivolumab, enrolling NSCLC patients is currently ongoing (Clinical Trials gov 2021h). Two other compounds are currently studied with a selective inhibition for PI3Kα, i.e., alpelisib (BYL719) and serabelisib (TAK-117). Alpelisib was mainly tested to evaluate tolerability and preliminary efficacy in patients with advanced solid cancers including some patients with lung cancer. Hyperglycemia, cutaneous rash, loss of appetite and diarrhea were the most common, although manageable adverse event. Tumor shrinkage was described in some patients, but the overall number of participants and specifically of lung cancer patients was very low. The results of a phase II study that included only two patients are still pending (Ando et al. 2019; Clinical Trials gov 2021i). Serabelisib (TAK-117) is being tested as part of an open phase II clinical trial in combination with canaglifozin, not yet recruiting to date (Juric et al. 2017; Clinical Trials gov 2021j).

BGB-10188 is a PI3Kδ-specific inhibitor (Yang et al. 2020). The phase I/II study, aimed to determine the maximum tolerated dose, included NSCLC patients, evaluating the compound alone or in combination with tislelizumab (Clinical Trials gov 2021k). Another PI3Kδ inhibitor with an excellent in vivo safety profile is Parsaclisib (INCB50465), for which a phase I clinical trial in combination with pembrolizumab was designed. The trial that involves lung cancer patients is still not currently recruiting (Yue et al. 2019; Clinical Trials gov 2021l). Finally, idelalisib, already approved for the treatment of lymphoma, is going to be tested in a phase I/II study in combination with pembrolizumab (Raedler 2015; Clinical Trials gov 2021m).

AZD8186 is considered a PI3Kβ inhibitor, although a weaker inhibition of PI3Kδ has been also described (Barlaam et al. 2015). In vitro studies on PTEN null tumors, proved efficacy of AZD8186 in the downregulation of cholesterol biosynthesis genes and upregulation of markers associated with metabolic stress (Lynch et al. 2017), but its action, when used alone in vivo, resulted limited (Owusu-Brackett et al. 2020). In this view, a phase I clinical trial involving NSCLC patients was carried out (Clinical Trials gov 2021n). In this trial, AZD8186 was used alone and in combination. The study result was completed, but the results are not available yet.

Some molecules have also shown the ability to inhibit the PI3K/AKT/mTOR pathway at multiple levels. Perifosine inhibits both PI3K and AKT, showing an interesting in vitro activity against lung cancer, colorectal cancer, and multiple myeloma (Richardson et al. 2012). It was tested in a phase II clinical trial completed in 2018, but the results of the study are not yet available (Clinical Trials gov 2021o). Gedatolisib (PF-05212384) is a potent pan-class I PI3K and mTOR inhibitor, with a manageable safety profile and antitumor activity in advanced stage NSCLC patients (Mallon et al. 2011). The study encourages two phase I clinical trials including lung cancer patients, alone or associating the compound with docetaxel, cisplatin, or dacomitinib, showing good safety profile and preliminary antitumor activity (Shapiro et al. 2015; Wainberg et al. 2016). However, a recent phase II clinical trial in SCLC proved no benefit for patients treated with this molecule (Udagawa et al. 2020). A similar compound named Voxtalisib (SARD245409/XL765) was tested in phase I clinical trials on advanced solid cancer patients including lung cancer, alone (Papadopoulos et al. 2014) or in combination with pimasertib (Schram et al. 2018). In the first study, the safety was acceptable, however, in the second study, the patients proved poor tolerability and limited antitumor activity.

5.3 Closing Remarks

The PI3K/AKT/mTOR pathway plays a pivotal role in the modulation of cellular proliferation, and therefore in cancer development. Several cancers, including lung cancer, are characterized by an abnormal regulation of this pathway. This can be due to a constitutive reduced efficacy of those mechanisms of control involved in the downregulation of the pathway (e.g., PTEN null mutation) or to a constitutive activation of PI3K/AKT. Exploiting PI3K/AKT/mTOR in lung cancer treatment seems a promising opportunity. However, to date none of the molecules tested progressed toward a phase III clinical trial. It must be said that some clinical trials raised safety concerns and above all registered limited results regarding clinical efficacy. Even so, there is still great interest for this therapeutic approach in lung cancer, proved by the elevated number of clinical trials currently ongoing. Hopefully, the results of these studies will add new and sufficient information for the development of new effective drugs for lung cancer. There is already a signal going into this direction. Some evidence in the treatment of cancer patients suggests the possibility that PI3K inhibitors may enhance the response to conventional treatment (Wright et al. 2021), however, the definitive utility of these molecules in combination to standard treatment needs to be further addressed by additional studies.

6 PI3K and ARDS

6.1 Background

Acute respiratory distress syndrome (ARDS) is an inflammatory disease caused by an uncontrolled systemic and/or local pulmonary inflammatory response due to the effects of risk factors that may involve the lung either directly (toxic gas inhalation, aspiration of gastric content, severe lung infection, etc.) or indirectly (systemic infection, severe trauma, drug overdose, multiple blood transfusion, etc.). These pathological conditions may induce a systemic and local inflammatory response causing inflammatory alveolar damage and pulmonary, non-cardiogenic, edema. From the clinical point of view, ARDS is characterized by acute and progressively increasing dyspnea with severe hypoxemia leading to death in about 50% of cases (Wheeler and Bernard 2007; Thompson et al. 2017). The pathogenic basis of ARDS includes a variety of complex mechanisms that can be summarized in an excessive activation of immune cells, dysregulated inflammation, and increased lung endothelial and epithelial permeability that cause the disruption of the lung microvascular barrier (Huppert et al. 2019; Sapru et al. 2015).

6.2 Are the Expression and Functional Profile of PI3K Relevant to the Pathogenesis of ARDS?

Studies on the role of PI3K/AKT/mTOR pathway on ARDS were mainly focused on the role of the regulatory protein PTEN. Murine, epithelium-specific PTEN-deleted models showed severe hypoxia, increased alveolar permeability, alveolar flooding, and lung fibrosis. These models also showed a reduction of the production of claudin-4, a key factor for the integrity of tight junctions, and increased levels of activated AKT and MMP2. The result is an impairment of the integrity of the alveolus-capillary membrane barrier (Yanagi et al. 2015). These results have also suggested that PI3K could be a potential target for ventilator-induced ARDS. Over-ventilated, isolated lungs from PI3Kγ−/− mice, showed decreased phosphorylation of AKT as well as histological changes typical of lung injury (Lionetti et al. 2006). The involvement of the PI3Kγ isoform was also emphasized in an endotoxin-induced model of acute lung injury. Specifically, the severity of endotoxemia-induced ARDS and lung neutrophil accumulation were significantly reduced in mice lacking the PI3K p110γ catalytic subunit compared with wild-type mice (Yum et al. 2001). The close association of PI3Kδ and injury-induced lung damage has been also extensively studied, although its role has been only recently recognized in ARDS. In this regard, the “Activated PI3Kδ syndrome” (APDS) has been described, a dominant gain-of-function mutation of the PIK3CD gene. The mutated gene was identified in 17 patients, with recurrent respiratory infections, progressive airway damage, lymphopenia, impaired vaccine responses, increased circulating transitional B cells, increased serum levels of IgM and decreased levels of IgG2. Interestingly, these patients also showed high levels of phosphatidylinositol-3-phosphate (PIP3) and phosphorylated AKT. IC87114 and GS-1101, two different selective PI3Kδ blockers, inhibited AKT activity in CD4+ and CD8+ T lymphocytes both on wild type and mutated PI3Kδ in vitro (Angulo et al. 2013). The involvement of PI3Kδ in the modulation of airway neutrophil recruitment and bronchial epithelial functional alterations has been also investigated in a murine model of acute lung injury (ALI). In the LPS-induced model of lung injury, the pharmacological inhibition of PI3Kδ with either IC87114 or AMG319 mitigated LPS-induced edema, lung injury, and neutrophilic inflammation, and increased TNF and IL-6 levels in BAL were also diminished (Yao et al. 2021).

6.3 Are PI3K Inhibitors Safe and Clinically Relevant in ARDS?

Evidence in the treatment of Acute Lung Injury (ALI)/ARDS through the inhibition of the PI3K pathway is still limited and despite some pathophysiological evidence, to the best of our knowledge, no clinical trials aimed to evaluate the inhibition of this pathway in ALI/ARDS are ongoing.

However, more recently PI3Kδ was suggested as a possible target for COVID-19-related ARDS (Zhang et al. 2020). Idelalisib, currently used in hematologic malignancies and studied in allergic diseases, was suggested as a potential pharmacological therapy for COVID-19, alone or in combination with ebastine for their capacity to decrease inflammatory processes (Palma et al. 2020). Macrolides are also widely used in the treatment of COVID-19 patients, being in some instances efficacious mainly by preventing disease progression (Batiha et al. 2021). This action could be mediated by the effect of macrolides at inhibiting PI3K. Their use could be reasonable due to the well-known safety profile and its activity in preventing bacterial super-infection (Bacharier et al. 2015).

Currently, several ongoing clinical trials are testing the inhibition of PI3K/AKT/mTOR in the treatment of COVID-19. Drugs currently studied are duvelisib, a dual inhibitor of PI3Kγ/δ (Clinical Trials gov 2021p), and sirolimus, an mTOR inhibitor (Clinical Trials gov 2021q, r, s, t).

6.4 Closing Remarks

A considerable amount of data suggests that PI3K signaling and its downstream mediators are strongly activated in lung injury and ARDS. However, very few clinical data are currently supporting the use of PI3K inhibitors in a clinical setting. Certainly, the disappointing current survival of patients with ARDS will spur the design of new trials investigating different therapeutic approaches and among these, the inhibition of the PI3K axis seems promising.

7 General Conclusions

Many studies have clearly demonstrated that PI3K is fully involved in most of the steps underlying the pathogenesis of respiratory diseases such as bronchial asthma, COPD, IPF, lung cancer, and ARDS. This is not surprising considering the important regulatory role exerted by PI3K in a wide array of cellular activities such as growth, proliferation, differentiation, cytoskeletal organization, migration/adhesion, survival/apoptosis, and angiogenesis. The use of non-selective and even more isoform-selective PI3K inhibitors has allowed a better understanding of the specific roles played by each isoform in various lung disorders. Several in vitro and in vivo animal models of asthma, COPD, IPF, lung cancer, and ARDS have offered enough evidence to design clinical trials to evaluate the safety profile and efficacy of this therapeutic approach. Nevertheless, few clinical trials have been performed so far with the exception of lung cancer, where a discrete number of trials are currently ongoing. All studies had as primary outcome drug tolerability and only in a few cases, efficacy was assessed. In general, the safety profile was good and limited to manageable adverse events. This was particularly evident when inhibitors were administered locally, by inhalation. In contrast, efficacy data, registered so far, are not brilliant and need further larger and well-powered studies. It is evident that new clinical trials designed for the evaluation of efficacy are not easy to perform. Many variables need to be considered, including the possibility that the positive effect of blocking PI3K in the clinical setting could not be enough to achieve a clinical result. As often happens in biology, redundant signaling pathways could intervene mitigating a potential positive effect. Most of the respiratory diseases described in this chapter are marked by low survival rates and all need a substantial improvement of the current medical treatments. The history of the treatment of respiratory disease with PI3K inhibitors is just at its beginning and certainly needs additional studies and further efforts from researchers and clinicians involved in this fascinating field.