Lymphocyte Depleting and Modulating Therapies for Chronic Lung Allograft Dysfunction

Abstract

Chronic lung rejection, also called chronic lung allograft dysfunction (CLAD), remains the major hurdle limiting long-term survival after lung transplantation, and limited therapeutic options are available to slow the progressive decline in lung function. Most interventions are only temporarily effective in stabilizing the loss of or modestly improving lung function, with disease progression resuming over time in the majority of patients. Therefore, identification of effective treatments that prevent the onset or halt progression of CLAD is urgently needed. As a key effector cell in its pathophysiology, lymphocytes have been considered a therapeutic target in CLAD. The aim of this review is to evaluate the use and efficacy of lymphocyte depleting and immunomodulating therapies in progressive CLAD beyond usual maintenance immunosuppressive strategies. Modalities used include anti-thymocyte globulin, alemtuzumab, methotrexate, cyclophosphamide, total lymphoid irradiation, and extracorporeal photopheresis, and to explore possible future strategies. When considering both efficacy and risk of side effects, extracorporeal photopheresis, anti-thymocyte globulin and total lymphoid irradiation appear to offer the best treatment options currently available for progressive CLAD patients.

Significance Statement

Effective treatments to prevent the onset and progression of chronic lung rejection after lung transplantation are still a major shortcoming. Based on existing data to date, considering both efficacy and risk of side effects, extracorporeal photopheresis, anti-thymocyte globulin, and total lymphoid irradiation are currently the most viable second-line treatment options. However, it is important to note that interpretation of most results is hampered by the lack of randomized controlled trials.

I. Introduction

Lung transplantation is a life-saving therapeutic option in well-selected patients with end-stage chronic lung diseases. Advancements in surgical techniques and early post-transplant care, such as maintenance immunosuppressive therapy and management of infections, have improved post-transplant outcomes in the past decades (Bos et al., 2020). Nevertheless, survival after lung transplantation still lags behind that of recipients of other solid organ transplants, with a median post-transplant survival of only 6.7 years (Chambers et al., 2019). To a larger extent, this poor long-term survival is related to the high incidence of and difficulty managing chronic lung rejection, so-called chronic lung allograft dysfunction (CLAD), a progressive life-threatening condition affecting 50% of patients within five years post-transplant, leading to lung allograft failure, respiratory insufficiency and death (Chambers et al., 2019).

CLAD encompasses two main phenotypes, bronchiolitis obliterans syndrome (BOS) and restrictive allograft syndrome (RAS), along with a mixed phenotype with features of both. BOS is the commonest phenotype in approximately 70% of CLAD patients and is characterized by progressive airway obliteration leading to airflow obstruction. RAS occurs in up to 20%–30% of CLAD patients and is characterized by parenchymal and/or pleural fibrosis with a restrictive pulmonary function decline. RAS has a very poor prognosis, with a median survival of only 1-2 years after diagnosis compared with 3–5 years for BOS. The diagnosis of CLAD is made based on a decline in forced expiratory volume in one second (FEV₁) of ≥20% from post-transplant baseline, defined as the mean of the two best post-operative FEV₁ measurements taken >3 weeks apart, in combination with a concurrent decline in forced vital capacity of ≥20% and persistent opacities on chest imaging for the RAS phenotype (Verleden et al., 2019). CLAD leads to a progressive decline in FEV₁; this decline is often stepwise, in which after an initial decrease a plateau phase is reached. However, some patients have a steep and rapidly progressive decline, while others have a slower decline over years (Belperio et al., 2009; Sato et al., 2013). CLAD severity is graded from 1–4 based on the severity of FEV₁ decline (stage 1: 66–80%, stage 2: 51–65%, stage 3: 36–50%, stage 4: ≤35% of baseline) (Verleden et al., 2019).

It is postulated that CLAD occurs as a result of the host’s adaptive and innate immune responses directed to the lung allograft, in which a complex array of immune cells and mechanisms is involved (Bos et al., 2022b,c). Next to medical non-compliance with immunosuppressive treatment, various risk factors for CLAD have been identified, both alloimmune and non-alloimmune factors, including ischemia-reperfusion injury, acute cellular rejection, antibody-mediated rejection, respiratory infections, gastroesophageal reflux, and air pollution (Verleden et al., 2019).

The type of standard immunosuppressive maintenance treatment after lung transplantation varies between centers, but usually consists of triple therapy with a calcineurin inhibitor (tacrolimus/cyclosporine), a cell cycle inhibitor (mycophenolate mofetil/azathioprine) and corticosteroids (Nelson et al., 2022). Currently, therapeutic options to slow the progressive decline in lung function in CLAD are very limited. These include intensification and optimization of maintenance immunosuppression, such as augmentation of corticosteroids and switching to more potent maintenance immunosuppressive drugs, such as from cyclosporine to tacrolimus and azathioprine to mycophenolate mofetil (Nelson et al., 2022). This, often in combination with the addition of azithromycin (if not already initiated as preventive treatment post-transplant), is usually instituted as an early measure to aim to halt CLAD progression (Verleden et al., 2019). The immunomodulatory properties of azithromycin in CLAD are summarized in a review by Vos et al. (Vos et al., 2012).

Beyond this first line of treatments, several lymphocyte depleting and/or modulating therapies have been studied in patients with progressive CLAD, including methotrexate, cyclophosphamide, alemtuzumab, anti-thymocyte globulin (ATG), total lymphoid irradiation (TLI), and extracorporeal photopheresis (ECP). Most of these therapies have only been evaluated in small retrospective single-center studies, and the effect reported is often temporary with further disease progression over time in the majority of patients. Therefore, there is a compelling need for more effective treatments to prevent the onset and progression of CLAD (Verleden et al., 2019).

This review summarizes the data available to date on the efficacy of lymphocyte depleting and modulating therapies in CLAD beyond optimized maintenance immunosuppressive strategies and explores possible future directions in this area. For this, the electronic databases of PubMed and EMBASE were searched in July 2022 and publications related to our predefined topic were included. There is little data available on use of these modalities for the treatment of RAS, as such, most of the data presented in this review focuses on experience from the treatment of BOS.

II. Immunodepleting Therapies

A. Alemtuzumab

Alemtuzumab is a recombinant humanized IgG1 monoclonal antibody directed against CD52, which is expressed on the cell surface of mainly T and B lymphocytes, and to a lesser extent on natural killer cells, macrophages, and monocytes, and is believed to play a role in cell signaling and homeostasis (Bhowmick et al., 2016; Syed, 2021). Alemtuzumab induces a rapid, profound and prolonged (i.e., several months) lymphocyte depletion through antibody-dependent cell-mediated cytolysis, complement-dependent cytolysis, and induction of apoptosis, but also leads to an expansion of regulatory T and B cells during repopulation (Bhowmick et al., 2016) (Fig. 1). Because of prolonged lymphodepletion, the potential for sustained bone marrow suppression is of concern, especially given the susceptibility of lung transplant recipients to infections and malignancies (Trindade et al., 2020).

Fig. 1.

Overview of main mechanisms of several lymphocyte depleting and/or modulating therapies for CLAD. APC, antigen-presenting cell (e.g., dendritic cell, macrophage, B cell); AZA, azathioprine; Bregs, regulatory B cells; CNI, calcineurin inhibitor; DC, dendritic cell; JAK-I, janus kinase inhibitor; MEK-I, mitogen-activated protein kinase kinase inhibitor; MMF, mycophenolate mofetil; MTX, methotrexate; TNFα-I, tumor necrosis factor alpha inhibitor; Tregs, regulatory T cells. Created with BioRender.com.

Alemtuzumab has been used primarily for the treatment of chronic lymphocytic leukemia (Hallek, 2017) and relapsing-remitting multiple sclerosis (Syed, 2021), but also off-label for induction immunosuppression in solid organ transplantation (Small et al., 2022).

1. Evidence in CLAD

In an effort to more effectively deplete T cells and other immune cells that may contribute to CLAD, Reams et al. investigated the effect of alemtuzumab (30 mg i.v.) in ten BOS patients after failure of prior therapy with methylprednisolone and ATG (Reams et al., 2007). They found a stabilization or improvement of BOS stage in 70% of patients. Alemtuzumab caused a long-lasting decrease in CD4 count and only 27% of patients remained free of infectious complications in the entire cohort, which also included patients with acute cellular rejection (Reams et al., 2007). Another study involving 17 BOS patients mainly demonstrated efficacy of alemtuzumab (30 mg i.v.) in early BOS. BOS-free progression was seen in 53% of patients at 6 months with freedom from FEV₁ decline >10% in 70% of early BOS (stage 1) versus only 14% in advanced BOS (stage 2–3). Also, in this study, the infection rate was high (77%) (Ensor et al., 2017).

Moniodis et al. compared the efficacy of alemtuzumab (30 mg i.v. or s.c.) (n = 13) to ECP (n = 17) for the treatment of CLAD (Moniodis et al., 2018). The rate of FEV₁ decline improved significantly at 3 and 6 months in both groups, compared with pre-treatment, with a benefit also at 1 month in the alemtuzumab group. Subgroup analyses for alemtuzumab in RAS only showed a slowing in slope at 3 months, while the BOS subgroup resembled the overall CLAD cohort. Interestingly, alemtuzumab reduced the number of rapid decliners (>25% drop FEV₁) more markedly than ECP at 1, 3, and 6 months following treatment. There were no differences between alemtuzumab and ECP with regard to infections, with 29% of alemtuzumab-treated patients having a clinically significant infection in the year after treatment. There was no difference in survival at 6 months and 1 year between the alemtuzumab, ECP, and untreated (i.e., slowly progressive CLAD) group (Moniodis et al., 2018).

Trindade et al. examined the safety of alemtuzumab in a specific group of lung transplant recipients with short telomeres who are at increased risk of clinically significant leukopenia (Trindade et al., 2020). In this small study (14 CLAD patients of whom three with short telomeres), alemtuzumab treatment appeared safe, with no significant difference in infections necessitating hospitalization, although it was associated with an increased incidence of neutropenia, thrombocytopenia, and anemia in the short telomere group (Trindade et al., 2020).

Lastly, a conference abstract, looking at 1-year overall survival after alemtuzumab administration in 14 patients with severe CLAD, reported that 64% were alive with a stable FEV₁ in 67% of survivors (Thachuthara-George et al., 2015). Another conference abstract documented that the rate of lung function decline during the 3 months post-treatment (30 mg s.c.) was significantly lower than the 3 months prior to treatment in eight BOS patients with rapid loss of lung function (75% stage 3–4). Clinically symptomatic infections occurred in 50% of patients (Girgis et al., 2020).

Treatment with alemtuzumab appears to attenuate lung function decline, especially in BOS patients. It is, however, difficult to determine whether this change simply represents the natural course of BOS or is a direct treatment effect, although some studies (Moniodis et al., 2018; Thachuthara-George et al., 2015) have documented sustained results. Ensor et al. mainly observed efficacy in BOS stage 1 versus higher stages. Reduced efficacy in more advanced CLAD may be due to a significant delay in therapy to a point beyond where allograft function can be stabilized, because of too severe structural injury to the allograft (Ensor et al., 2017). On the other hand, beneficial results were seen by Girgis et al. where 75% of patients were in CLAD stage 3–4 (Girgis et al., 2020). While alemtuzumab may have a potential benefit in BOS, it carries a high risk of infectious complications. Randomized controlled trials are required to better establish efficacy and safety.

B. Anti-Thymocyte Globulin

ATG is a polyclonal antibody preparation, derived from rabbits or horses immunized with thymocytes or T-cell lines (Mohty, 2007). The polyclonal nature of ATG is reflected in its diverse immune effects, including prolonged (i.e., several weeks) depletion of cytotoxic T cells by complement-mediated lysis, antibody-dependent cell-mediated cytotoxicity, apoptosis (activation-associated and Fas-dependent apoptosis), and opsonization. Alongside T-cell-depleting properties, other potential mechanisms of action involve B-cell apoptosis, depletion of natural killer cells, interference with dendritic cells, modulation of cell surface adhesion proteins and chemokine receptors, and induction of regulatory T cells (Mohty, 2007). One should keep in mind that, despite sharing some common traits, equine and rabbit ATG are strictly different drugs (Mohty, 2007). Rabbit ATG is thought to have a better efficacy and side effect profile than equine ATG, and is more easily accessible than alemtuzumab in some countries.

ATG has been used in conditioning regimens for hematopoietic stem cell transplantation and as induction immunosuppression in solid organ transplants, including lung transplant recipients (Mohty et al., 2014; Small et al., 2022).

Common adverse events related to ATG include transfusion-related reactions, cytokine release syndrome, leukopenia, thrombocytopenia, and infections (Mohty et al., 2014).

1. Evidence in CLAD

In addition to some older studies(Date et al., 1998; Kesten et al., 1996; Snell et al., 1996) published in the early era of lung transplantation that showed some efficacy, there are several larger, recent, retrospective studies that have examined the potency of ATG in slowing CLAD progression. In a study of 25 CLAD patients, 32% had stabilization of FEV₁ for at least 6 months after ATG (1.5 mg/kg/d for 7 days i.v.), with an improved survival rate (Izhakian et al., 2016). However, these patients appeared to have a slower decline in FEV₁ pre-treatment, suggesting an already slower disease progression (Izhakian et al., 2016). January et al. found an increase in FEV₁ (defined by a shift from a negative to a positive slope) in the 6 months after ATG (5–7.5 mg/kg over 3‐6 days) compared with before in 40% of a total of 108 patients (93% BOS) (January et al., 2019). Additionally, 44% of the non-responders had a less negative FEV₁ slope. It is worth noting that this study included 20% BOS stage 0p (10%–20% FEV₁ decline and/or ≥25% decline in FEF_25–75%) patients, and that no predictors of response were identified, neither disease severity at time of treatment, nor steepness of FEV₁ decline or RAS phenotype (January et al., 2019). Kotecha et al. reported 71 patients receiving mostly equine ATG (500 mg on day 1, subsequent dosing days 2–5 based on CD2/CD3 lymphocyte counts) for CLAD (83% BOS). Twenty-three percent were complete responders who had stabilization or improvement in FEV₁, while 40% were partial responders with a ≥20% improved rate of FEV₁ decline (Kotecha et al., 2021). Risk of death or retransplantation was significantly lower in these groups, with a 70% and 65% reduction, respectively. CLAD stage 2–3 and younger age were predictors of partial, but not complete, response. CLAD phenotype did not correlate with response. Interestingly, as many centers only try ATG treatment once, 30% of patients had received ATG twice with a median interval of 3 months (Kotecha et al., 2021). Finally, another small study of 13 CLAD patients (77% BOS; ATG 1.5 mg/kg/d, total target dose 10–20 mg/kg) reported stabilization or improvement (>5%) of FEV₁ in half of the patients (Margallo Iribarnegaray et al., 2021). Most patients who responded were in CLAD stage 1–2 (71%). Worse survival was observed in rapid decliners (monthly FEV₁ drop >100 ml) (Margallo Iribarnegaray et al., 2021).

Most important side effects reported in these studies were mild infusion-related reactions (January et al., 2019; Margallo Iribarnegaray et al., 2021), infections (up to 19%) (January et al., 2019), severe leukopenia (4%) (Izhakian et al., 2016), and neutropenia (14%) (Margallo Iribarnegaray et al., 2021).

ATG appears to be effective in stabilizing or attenuating lung function decline in a subgroup of CLAD patients, including RAS, and may lead to prolonged survival. Although certain predictors of response have been identified, such as early disease stages (Kotecha et al., 2021; Margallo Iribarnegaray et al., 2021), these were not consistent across all studies (January et al., 2019). Multicenter, randomized controlled trials are needed to better determine predictors of response to ATG in CLAD.

C. Total Lymphoid Irradiation

Radiation therapy is undoubtedly best known for its role in cancer treatment, but its use extends beyond this (McKay et al., 2014). TLI targets the main structures of the lymphatic system as most lymphocytes are highly radiation sensitive (Schaue and McBride, 2012). TLI therefore has a strong immunosuppressive nature; it produces a selective and long-lasting (i.e., several weeks) reduction of certain subsets of T-cell and B-cell populations. In general, there is a spectrum of radiosensitivity from B cells through naïve T-helper cells, natural killer cells, toward more radioresistant T-memory cells and natural killer T cells. As a result, irradiation shifts the balance of the immune system. Regulatory T cells and natural killer T cells are relatively radioresistant and their proportion within the lymphoid tissues increases rapidly following irradiation. Further induction and activation of regulatory T cells can occur via TGF-β, which is induced by TLI (Schaue and McBride, 2012). TLI is often administered in ten fractions of 0.8 Gy twice weekly, via mantle, paraaortic and inverted-Y fields (McKay et al., 2014).

1. Evidence in CLAD

A first study in 1998 described poor efficacy of TLI in 11 BOS patients; most patients died within eight weeks of cessation due to further disease progression or infection, and only 36% had sustained stabilization of FEV₁ with a mean follow up of 24–72 weeks (Diamond et al., 1998). Later, Verleden et al. documented a significant attenuation in the rate of FEV₁ decline in a small group (n = 6) compared with historical controls (n = 5), although half of them failed within the first year after TLI (Verleden et al., 2009). The Newcastle Group also reported that TLI significantly decreased the rate of FEV₁ decline in 12 BOS patients (Chacon et al., 2000) and in a further, larger study of 37 BOS patients (Fisher et al., 2005), the majority of whom had BOS stage 2–3. Interestingly, the latter study found that the most pronounced effect appeared to occur in patients with the fastest progression prior to TLI (Fisher et al., 2005). Lastly, in a recent study, the Leuven Group reported the outcome of 20 BOS patients (65% BOS 3) treated with TLI, including the six previously reported (Verleden et al., 2009) patients (Lebeer et al., 2020). Four patients (20%) died during or shortly after TLI due to progressive respiratory insufficiency, while the decline in FEV₁ slowed significantly in 94% of the remaining patients, again especially in those with a rapid decline pre-TLI (≥ 100 ml/mo) (Lebeer et al., 2020). An absolute increase in FEV₁ was seen in 13% 6 months post-treatment, even though these patients were already in BOS 3. Freedom from graft loss was 27% 2 years after TLI (Lebeer et al., 2020). Lastly, a recently published study (Geng-Cahuayme et al., 2022) included 23% RAS patients and showed significant attenuation of FEV₁ slope in both BOS and RAS phenotypes and both rapid and slow decliners. They found that a Karnofsky Performance Status of >70 was a prognostic marker for survival (Geng-Cahuayme et al., 2022).

In addition to these studies, several conference abstract reports were available. Most of these had similar findings with a decrease in FEV₁ decline post-treatment compared with before (Afolabi et al., 1996; Arbeláez et al., 2014; Hunt et al., 2019; Low et al., 2017; Miller et al., 2016; Soresi et al., 2015; Sáez et al., 2014). Hunt et al. reported a mean survival of 4.2 (range 0.75–7.5) years post-TLI, and Soresi et al. a 2-year overall survival of 59% after initiation of treatment (Hunt et al., 2019; Soresi et al., 2015). Schmack et al. attempted to correlate specific lymphocyte phenotypes with response to TLI in a prospective study of 26 patients with progressive BOS (Schmack et al., 2017). They found an inverse correlation between the total number of peripheral B cells, naive B cells, memory B cells, plasmablasts, and naive CD8+ T cells pre-treatment and patient survival (Schmack et al., 2017).

Frequently reported side effects were neutropenia, thrombocytopenia, infections, gastrointestinal symptoms, and fatigue (McKay et al., 2014). The first three often led to treatment being delayed or terminated prematurely (Fisher et al., 2005; Geng-Cahuayme et al., 2022; Lebeer et al., 2020; O’Hare et al., 2011).

Although data on TLI in CLAD remain relatively scarce, the findings are consistent across most studies in which TLI appeared to attenuate the decline in lung function in BOS and RAS. Importantly, it also seemed to be effective in CLAD patients with a rapid decline in lung function at the time of treatment initiation. Early initiation after CLAD onset may be warranted, although good results have also been documented in patients with advanced BOS (Fisher et al., 2005; Lebeer et al., 2020). Reported complications, such as neutropenia, thrombocytopenia, and risk of infection, suggest that TLI should be used with caution, although the incidence of serious side effects was low.

III. Immunomodulating Therapies

A. Methotrexate

Methotrexate is a folic acid analog and acts via several suggested mechanisms, including inhibition of purine and pyrimidine synthesis, suppression of transmethylation reactions with accumulation of polyamines, prolonged (i.e., several weeks) reduction of antigen-dependent T-cell proliferation, apoptosis of T cells through the generation of reactive oxygen species, as well as selective downregulation of B cells, interference with cytokines and matrix metalloproteinases, and promotion of extracellular release of adenosine (Alqarni and Zeidler, 2020; Amrouche and Jamin, 2017; Bedoui et al., 2019). Adenosine is a potent anti-inflammatory mediator that acts through interactions with a variety of immune cell subtypes, such as neutrophils, macrophages, and T cells (Bedoui et al., 2019). In addition, recent insights suggest that methotrexate may also exert its anti-inflammatory effects via inhibition of nuclear factor-κB and the JAK/STAT pathway (Alqarni and Zeidler, 2020; Bedoui et al., 2019).

As a drawback, methotrexate has a high toxicity profile and can cause considerable side effects, including cytopenia, stomatitis, subcutaneous nodulosis, hepatic and renal toxicity, fatigue and lethargy (Bedoui et al., 2019). Although most of these mainly occur when higher doses (usually >30 mg/m²) are used as part of a chemotherapy regimen. Importantly, methotrexate can also cause pulmonary toxicity, such as drug-induced pneumonitis (Pivovarov and Zipursky, 2019).

Methotrexate has been used extensively in the treatment of neoplasms as a chemotherapeutic agent, autoimmune and connective tissue diseases such as rheumatic arthritis (Alqarni and Zeidler, 2020), interstitial lung diseases, including sarcoidosis (van den Bosch et al., 2022), and is commonly used after hematopoietic stem cell transplantation to prevent graft-versus-host disease (GvHD) (Martinez-Cibrian et al., 2021).

1. Evidence in CLAD

Evidence for methotrexate in CLAD is sparse and limited to BOS. A small study of ten patients showed that methotrexate could reduce the rate of lung function decline in BOS (Dusmet et al., 1996). Boettcher et al. also reported some benefit of methotrexate (single dose of 5 mg/kg or 7.5 mg/week) in three BOS patients (Boettcher et al., 2002). The same was found in a larger, retrospective study of 30 BOS patients, the majority of whom had BOS stage 3 at the time of treatment initiation (5–10 mg/week) (Sithamparanathan et al., 2016). A decrease in the rate of lung function decline was seen in 95% of patients treated for at least 6 months (70% of the cohort), with a significant median increase in FEV₁ at 3 and 6 months. The reduced rate of lung function decline remained significant in those treated for at least 12 months. However, methotrexate had to be discontinued in 30% of patients due to nausea, fatigue or leukopenia. This number was higher than that seen in, for example auto-immune diseases, but may be explained by the combination of other immunosuppressants and transplant-related drugs (Sithamparanathan et al., 2016).

Although based on a limited number of small uncontrolled retrospective studies, methotrexate might slow the rate of lung function decline in BOS patients, even in patients with severe BOS. With the recent insights on the involvement of methotrexate in the JAK/STAT pathway, one could reconsider further prospective studies as it is a less expensive alternative to more specific Janus kinase (JAK) inhibitors (discussed later in this review) (Alqarni and Zeidler, 2020). However, toxicity is still a concern and lack of tolerability and side effects were the main cause of drug withdrawal (Sithamparanathan et al., 2016).

B. Cyclophosphamide

Cyclophosphamide is an alkylating agent belonging to the group of oxazaphosporines (Ahlmann and Hempel, 2016). It is an inactive prodrug, requiring bioactivation by P450 enzymes to exhibit cytotoxic activity. Since cyclophosphamide has been used for over 40 years, there is plenty of experience in its use for the treatment of cancer and as a highly potent immunosuppressant for the treatment of autoimmune and immune-mediated diseases including vasculitis, systemic sclerosis, connective tissue disease-related interstitial lung disease (Ahlmann and Hempel, 2016; Barnes et al., 2018; Emadi et al., 2009; van den Bosch et al., 2022).

Cyclophosphamide halts cell division by cross-linking DNA strands (Ahlmann and Hempel, 2016). Therefore, it is a non-specific cell-cycle inhibitor affecting most cell lines, although it has some selectivity toward T and B lymphocytes, causing prolonged (i.e., several weeks) immunosuppressive effects. It is therefore now widely adopted in tumor vaccination protocols and to control alloreactivity after hematopoietic stem cell transplantation (Ahlmann and Hempel, 2016; Nunes and Kanakry, 2019). Interestingly, cyclophosphamide can also increase the number of myeloid-derived suppressor cells (Ahlmann and Hempel, 2016).

Important side effects are cytopenia, nausea, hemorrhagic cystitis, cardio-, liver- and nephrotoxicity, and carcinogenicity with an increased risk of hematological and solid organ malignancies (e.g., secondary acute leukemia, bladder cancer, skin cancer) (Ahlmann and Hempel, 2016; Barnes et al., 2018).

1. Evidence in CLAD

In 1999, Verleden et al. reported the outcome of oral cyclophosphamide (0.5–1 mg/kg daily) in seven BOS patients. In 86% of patients, FEV₁ stabilized or increased 3 and 6 months after initiation and remained stable for at least 24 ± 7 months in five patients who were able to continue treatment (Verleden et al., 1999). Cyclophosphamide was well tolerated and had to be discontinued in only one patient because of persistent leukopenia (Verleden et al., 1999). Other than this study, however, no further data in CLAD are available. As such, the role of cyclophosphamide in the treatment of CLAD remains unclear.

C. mTOR Inhibitors

Mammalian target of rapamycin (mTOR) inhibitors have been used after lung transplantation for several indications, such as a cell cycle inhibitor alternative, as part of a calcineurin inhibitor-sparing regimen or adjunctive immunosuppressive agent in the setting of rejection, cytomegalovirus infection or in patients with malignancies (Fine and Kushwaha, 2016). mTOR inhibitors block mammalian target of rapamycin, a serine/threonine kinase, and thereby inhibit growth factor-stimulated proliferation of lymphocytes and mesenchymal cells. In addition, mTOR inhibitors also interfere with B and dendritic cell maturation and function (Thomson et al., 2009) and possibly NK cell-mediated endotheliitis (Koenig et al., 2019). Common adverse events are gastrointestinal intolerance, leukopenia, edema, thromboembolic events, and drug-induced pneumonitis.

1. Evidence in CLAD

Most recent studies on mTOR inhibitors have focused on their use in maintenance immunosuppression as part of a calcineurin inhibitor-sparing regimen with the aim of preserving kidney function. The combination of low-dose everolimus and low-dose tacrolimus appeared safe, with no difference in incidence of acute rejection or CLAD compared with high-dose calcineurin inhibitor therapy (Gottlieb et al., 2019; Ivulich et al., 2023; Kneidinger et al., 2022). Few studies looked at the use of everolimus or sirolimus as a treatment of CLAD. Cahill et al. found that in patients with rapidly declining pulmonary function, sirolimus resulted in stabilization or improvement of FEV₁ slope (Cahill et al., 2003). Everolimus also improved the FEV₁ slope 3 and 6 months after versus before treatment in a study by Fernandez et al. (David Iturbe et al., 2019). Patrucco et al. also found stabilization in FEV₁ in CLAD patients, however, subgroup analysis showed progressive functional loss in RAS patients (Patrucco et al., 2021). In another small study, three CLAD patients (60%) remained stable after introduction of everolimus, whereas 40% progressed (Turkkan et al., 2022). Nonetheless, side effects often necessitated discontinuation of mTOR inhibitors (Bos et al., 2021; Cahill et al., 2003; Kneidinger et al., 2022).

D. Belatacept and Basiliximab

Little is known about the use of these two agents for CLAD. Belatacept is a selective CD80/86-CD28 T-cell costimulation blocker widely used in kidney transplantation for induction and maintenance immunosuppression (Masson et al., 2014). The role of belatacept in the setting of lung transplantation remains uncertain with only a few small studies reporting its use in maintenance immunosuppression as part of a calcineurin inhibitor-sparing regimen (Huang et al., 2022; Iasella et al., 2018; Timofte et al., 2016), and a conference abstract on antibody-mediated rejection (Zaffiri et al., 2022), while data in CLAD is lacking. Importantly, one randomized controlled trial with 27 lung transplant patients had to be discontinued prematurely due to increased rates of death in the belatacept arm (Huang et al., 2022).

Basiliximab, a chimeric monoclonal antibody that selectively binds to the α–subunit (CD25) of interleukin-2 receptors, is used for induction therapy in lung transplantation (Small et al., 2022). In addition, there are some case series describing its use in maintenance immunosuppression to avoid calcineurin inhibitor-related nephrotoxicity (Högerle et al., 2016; Kim et al., 2021; Ross et al., 2020). Again, there is no data on any potential benefit in CLAD.

E. TNF-Alpha Inhibitors

Tumor necrosis factor alpha (TNFα) is a cytokine that acts as a major regulator of inflammatory reactions via the initiation of signal transduction pathways leading to cytotoxicity and upregulation of various cytokines, chemokines and growth factors (Jang et al., 2021). TNFα is also a key factor in the pathogenesis of CLAD (Bos et al., 2022b).

Several TNFα inhibitors are used for the treatment of inflammatory and autoimmune conditions, such as the monoclonal antibodies infliximab, adalimumab, certolizumab, and golimumab, and the recombinant fusion protein etanercept (Jang et al., 2021). Some of these have also been tested in solid organ transplants to mediate inflammatory responses in ischemia-reperfusion injury and rejection (Pascher and Klupp, 2005).

Anti-TNF agents are generally well tolerated, with common adverse effects being minor. General side effects include infusion-related reactions, injection site reactions, anemia, transaminitis, and mild infections; although there is a risk of severe infections and possibly an increased risk of malignancies, especially lymphomas and non-melanoma skin cancers (Jang et al., 2021).

1. Evidence in CLAD

Next to a few preclinical animal studies (Alho et al., 2003; Aris et al., 2002; Smith et al., 2001), there is one proof-of-concept study that reported the use of infliximab (3 mg/kg i.v. at 0-2-6 weeks minimally) in five patients with progressive BOS (Borthwick et al., 2013). FEV₁ and 6-minute walk distance improved in four patients and stabilized in a fifth patient with rapid lung function decline. All patients remained stable for at least 18 months. Infliximab was generally well tolerated; one patient developed a fungal infection (Borthwick et al., 2013).

F. Extracorporeal Photopheresis

ECP is a leukapheresis-based immunomodulatory procedure, currently approved for the management of cutaneous T-cell lymphoma, GvHD and rejection after solid organ transplantation (Hage et al., 2021). ECP is a procedure in which whole blood is collected from the patient and circulating leukocytes are removed by density centrifugation. The collected buffy coat is then treated with a photosensitizing agent (i.e., 8-methoxypsoralen) and exposed to UV A light before reinfusion into the patient (Cho et al., 2018). The exact mechanisms of therapeutic action are elusive, but ECP is thought to induce apoptosis of lymphoid cells, largely natural killer cells and T cells, and differentiation of activated monocytes into immature dendritic cells which in turn stimulate phagocytosis of lymphoid cells, and maturation and presentation of antigenic peptides (so-called transimmunization). Furthermore, ECP might modify the cytokine profile with induction of anti-inflammatory cytokines (IL-4, IL-10, transforming growth factor beta) and reduction of pro-inflammatory cytokines (e.g., tumor necrosis factor alpha), and stimulate upregulation of regulatory T cells (Cho et al., 2018). Different schedules are being used, often with a more intensive induction phase, followed by a maintenance schedule. However, the treatment effects after ECP initiation take time to come into effect. Next to this, there is no consensus on how long this therapy should be continued and there is uncertainty as to whether a sustained response can be observed and for how long after cessation. Furthermore, ECP is not reimbursed by health systems or insurance providers in many countries.

1. Evidence in CLAD

There are numerous publications describing the effects of ECP in CLAD, including several studies and various conference abstracts. Two prospective studies are available (Table 1). Firstly, a prospective multicenter study with 31 BOS patients (58% stage 2–3) from ten lung transplant centers (Hage et al., 2021). Rate of FEV₁ decline was reduced by 93% at 6 months, with a reduction ≥50% in 95% of patients. Multivariate analysis identified that pre-enrolment FEV₁ rate of decline was associated with both 6- and 12-month mortality. Notably, study enrollment was terminated prematurely due to a higher-than-expected mortality rate within the first year after enrollment of 32% and 41% at 6 and 12 months, respectively. There was no difference in mortality between the ECP group and an observational cohort; worth noting, the slope of FEV₁ decline pre-enrollment was much steeper in the former group (Hage et al., 2021). Another prospective single-center study by Jaksch et al. included 51 BOS patients and reported FEV₁ stabilization (variation <5%) in 61% of patients with an improvement in survival in these patients compared with both non-responders and non-treated BOS patients (Jaksch et al., 2012). Factors associated with inferior treatment response were cystic fibrosis as underlying lung disease and a longer time between transplant and BOS onset (Jaksch et al., 2012).

TABLE 1.

Prospective studies of ECP in BOS

Reference	Study design and period	Number of patients	CLAD stages	Duration ECP	Median slope FEV1 pre-ECP (ml/month)	Median slope FEV1 post-ECP (ml/month)	Response rate to ECP	Mortality within study	Predictors of response
Hage et al., 2021	Prospective, multicenter, 04/2015–07/2016	31 BOS ECP 13 BOS controls (7 crossover)	BOS 1 42% BOS 2 29% BOS 3 29%	6 months	Mean -136 ± 117	Mean -10 ± 58†	- Reduction ≥50% in 95% of patients (data 16/30, 63%) - Higher rate FEV1 decline in non-survivors at 6 and 12 months post-ECP start	39% and 48% at 6 and 12 months (87% CLAD, 13% infection)	FEV1 rate of decline pre-ECP correlated with 6- and 12-month mortality.
Jaksch et al., 2012	Prospective, single-center, 01/2000-06/2010	51 BOS ECP 143 BOS controls At least 3 months treatment	BOS 1 12% BOS 2 20% BOS 3 68%	3 months (51 patients) 12 months (25 patients)	−123†	−14* and -18‡	- FEV1 improvement in 30% (12% 3–6 months, 18% >12 months) - Stabilization in 31%		Negative impact: BOS onset >3 years post-transplant, rapid FEV1 decline pre-ECP, BOS stage, cystic fibrosis as primary lung disease. Overall survival: response to ECP.

Period of 3*, 6^† or 12^‡ months pre-/post-ECP initiation.

Furthermore, several recent retrospective single-center studies included both BOS and RAS patients, varying from 12 to 65 CLAD patients per study, of whom the majority had CLAD stage 2–3 (Table 2). Del Fante, Greer, and Vazirani all reported a significant reduction in rate of lung function decline (Del Fante et al., 2015; Greer et al., 2013; Vazirani et al., 2021) with a stabilization or improvement (≥10%) in lung function around 54%–60% (Del Fante et al., 2015; Greer et al., 2013). Notably, patients who did not complete the initial 3-month induction treatment or at least eight procedures were excluded in Greer’s (Greer et al., 2013) and Del Fante’s (Del Fante et al., 2015) studies, respectively. Robinson et al. looked at the lung function trajectory after forced cessation of ECP due to loss of reimbursement in 12 CLAD patients who had undergone long-term ECP treatment (median 1001 days) (Robinson et al., 2017). FEV₁ significantly and rapidly declined within 6 months of cessation, while lung function was stable in all patients before. Moreover, 58% died within 12 months mostly due to CLAD progression (Robinson et al., 2017).

TABLE 2.

Retrospective, single-center studies of ECP in CLAD

Reference	Study design and period	Number of patients	CLAD stages	Duration ECP	Median slope FEV1 pre-ECP (ml/month)	Median slope FEV1 post-ECP (ml/month)	Response rate to ECP	Mortality within study	Predictors of response
Baskaran et al., 2014	Retrospective, single-center, 01/2000–06/2011	88 BOS		6 months	−127†	−47†	−63% reduction in rate of FEV1 decline, 23% stabilized or improved (data 69/88, 78%)	19% at 2 years	No impact: % reduction of DSA or lung-associated self-antigens, or level of cytokines (IL-1β, IL-2, IL-4, IL-10, IL-17, IFN-γ, IP-10, MCP-1).
Benden et al., 2008	Retrospective, single-center, 1997–2007	12 BOS	BOS 1 42% BOS 2 17% BOS 3 42%	12 cycles	112 From baseline until start ECP	12 After 12 cycles of ECP until last value		33% (100% CLAD) Median OS 4.9 years after ECP start (BOS + ACR cohort)
Del Fante et al., 2015	Retrospective, single-center, 02/2003–12/2013	34 BOS 14 RAS 58 controls At least 8 procedures	CLAD 1 58% CLAD 2 21% CLAD 3 21%	Median 26 (IQR 17-42) procedures	−48 (95% CI -61; -36)†	−19 (95% CI -35; -3)† −4 (95% CI -15; +7) 12-24 months post ECP start	−60% stable graft function at 6 months	42% (85% CLAD, 5% cancer, 10% other), no difference ECP group and controls	Negative impact: rapid decliners (>100 ml FEV1/month), time from transplant to CLAD onset. No impact: CLAD phenotype, CLAD stage, time from CLAD onset to ECP. Overall survival: RAS (trend: P = 0.06), no influence CLAD stage, rapid decline, time CLAD diagnosis and start ECP.
Greer et al., 2013	Retrospective, single-center, 11/2007–09/2011	65 CLAD At least 3 months treatment	CLAD 0p 5% CLAD 1 9% CLAD 2 32% CLAD 3 54%	Median 15 (IQR 12–18) cycles			−12% improvement (≥10%) −42% stabilization in FEV1	2-year OS 97% in responders	Negative impact: rapid decline (>100 ml FEV1/month), RAS phenotype, ≤15% BAL neutrophils. No impact: CLAD stage, time to CLAD diagnosis and initiation of ECP. Overall survival: response to ECP.
Isenring et al., 2017	Retrospective, single-center, 01/2008–12/2012 Update from 2008 study(Benden et al., 2008)	9 of initial 12 BOS, 2 continued ECP after re-transplant	BOS 1 44% BOS 2 11% BOS 3 44%	Range 48–119 months			- Progression in 17% still alive at end of follow up	33% (67% cancer, 33% CLAD)	Negative impact: CLAD stage 2-3.
Karnes et al., 2019	Retrospective, single-center, 01/2000–12/2007 Update from 2010 study(Morrell et al., 2010)	60 BOS	See(Morrell et al., 2010)	6 months	See(Morrell et al., 2010)	See(Morrell et al., 2010)	See(Morrell et al., 2010)	17% <6 months, 50% <16 months of ECP initiation	12-fold higher chance of response if FEV1 decline >40 ml/months pre-ECP. FEV1 at start ECP correlated linearly with time to mortality and mortality at 16 months after start ECP.
Leroux et al., 2022	Retrospective, single-center, 01/2012–07/2019	12 BOS ECP 13 BOS controls At least 6 months treatment	BOS 1 33% BOS 2 25% BOS 3 42%	Median 32 (IQR 12-58) months	−44 (IQR -112; -8)†	+11 (IQR -0.8; +41)†	−75% FEV1 stabilization (±5%) within 12 months −63% improvement (>5%), 25% stabilization within 24 months of initiation - Lower risk of >20% drop in FEV1 in ECP-treated group versus control decliners.	33% (100% sudden death)	No impact: rate of FEV1 decline pre-ECP, time BOS diagnosis and ECP start.
Meloni et al., 2007	Retrospective, single-center	5 BOS At least 4 months treatment	BOS 2 60% BOS 3 40%	Range around 4–32 months			−60% FEV1 stabilization	40% (100% infection)	Tregs stabilized or increased in patients who stabilized and declined in non-responders.
Moniodis et al., 2018	Retrospective, single-center, 01/2005–12/2014	13 BOS ECP 4 RAS ECP 9 BOS alemtuzumab 5 RAS alemtuzumab 78 controls	CLAD 1 88% CLAD 2 12%	6 months	−122 (IQR -164; -77)*	−27 (IQR -82; -36)* and -12 (-56–22)†	- Significant reduction in rate of FEV1 decline at 3 and 6 months	OS at 6 months 0.82 (95% CI 0.55–0.94)	Negative impact: RAS phenotype.
Morrell et al., 2010	Retrospective, single-center, 01/2000–12/2007	60 BOS	BOS 1 8% BOS 2 33% BOS 3 58%	6 months	−116†	−29† and -21‡	- Reduction of FEV1 rate of decline in 79% - FEV1 improved in 25% at 6 and 12 months (data 56/60, 93%)	Median OS 2.6 years after ECP start	No impact: BOS stage, time of BOS onset, rate of FEV1 decline pre-ECP.
Pecoraro et al., 2017	Retrospective, single-center, 11/2013–06/2016	15 BOS 39 controls	BOS 1 7% BOS 2 27% BOS 3 67%	13 cycles			−80% FEV1 stabilization - FEV1 significantly higher 12 months after start ECP compared with controls	13% (50% CLAD, 50% cancer), better OS in ECP versus controls	No impact: CLAD stage, time CLAD onset, time BOS diagnosis and ECP.
Robinson et al., 2017	Retrospective, single-center, Patients who had to stop ECP end 2014 due to stop reimbursement	10 BOS 2 RAS	BOS 2 30% BOS 3 70% RAS unknown	Median 44 (range 8–142) procedures	−13 (range -8–110)	−17 (range -6–163)	- FEV1 rapidly declined within 6 months after ECP cessation	58% within 12 months of treatment cessation (43% CLAD, 43% infection + CLAD, 14% cancer)
Salerno et al., 1999	Retrospective, single-center, 1992–1998	8 BOS 20 controls	BOS 3 88%	Median 6 (range 3–13) months	Range -366; +8 From baseline until start ECP	Range -30; +44	−71% FEV1 stabilization or improvement −63% clinical stabilization	50% alive without retransplant after median 36 months
Vazirani et al., 2021	Retrospective, single-center, 01/2013–06/2018	5 BOS 2 RAS 5 mixed	CLAD 2 17% CLAD 3 83%		Mean 9 (95% CI 5; 12) ml/day in responders Mean 7 (4; 10) m/day in non-responders	Mean 1.4 (95% CI 0; 4) ml/day In responders Mean 5 (3; 7) ml/day in non-responders	−67% (<20% decrease in FEV1 within 6 weeks of ECP start)	Graft-failure in all non-responders (33%) within 6 months of ECP start	Negative impact: female sex, low baseline neutrophil count (< 1.9 × 109/L), prior exposure to ATG. No impact: CLAD phenotype.

OS, overall survival; RAS, restrictive allograft syndrome. Period of 3*, 6^† or 12^‡ months pre-/post-ECP initiation.

Survival seemed to correlate with response to ECP (Greer et al., 2013; Vazirani et al., 2021) though predictors of response varied across the studies. Some studies documented that female sex (Vazirani et al., 2021), a rapid decline in FEV₁ pre-ECP (Del Fante et al., 2015; Greer et al., 2013), RAS phenotype (Greer et al., 2013), a low baseline neutrophil count in blood (< 1.9 × 10⁹/L) (Vazirani et al., 2021) or bronchoalveolar lavage (≤15%) (Greer et al., 2013), prior exposure to ATG (Vazirani et al., 2021), and time from transplant to CLAD onset (Del Fante et al., 2015) adversely affected response to ECP. Although others could not find an impact of sex (Del Fante et al., 2015), CLAD phenotype (Del Fante et al., 2015; Vazirani et al., 2021), timing of CLAD onset (Greer et al., 2013), CLAD stage (Del Fante et al., 2015; Greer et al., 2013), or time from CLAD diagnosis to ECP initiation (Del Fante et al., 2015; Greer et al., 2013).

These findings corroborate the results from multiple previous studies where ECP was administered for BOS, as summarized in Table 2. Again, in these studies, that included between 5 and 88 patients, there was a significant reduction in FEV₁ decline with in most studies a stabilization in lung function in 60%–80% of patients (Baskaran et al., 2014; Benden et al., 2008; Isenring et al., 2017; Karnes et al., 2019; Leroux et al., 2022; Meloni et al., 2007; Moniodis et al., 2018; Morrell et al., 2010; Pecoraro et al., 2017; Salerno et al., 1999). Lastly, there are numerous conference abstracts reporting similar outcomes.

In all published data to date, ECP has generally been found to be a safe treatment without significant adverse effects.

In summary, clinical evidence suggests that ECP is associated with improvement or stabilization in lung function and decreases the rate of lung function decline in BOS, without an increased risk of infections or significant adverse events, with some studies also showing improved survival. Given that this response appeared to be independent of CLAD duration as well as stage at treatment initiation in most studies, ECP should be considered a viable second-line treatment option.

Large prospective clinical trials are needed to help predict response to therapy, and ultimately guide the placement of ECP in the treatment algorithm for CLAD. The results of a multicentre randomized controlled trial comparing ECP plus standard of care versus standard of care alone in patients with progressive CLAD in the UK (NIHR130612) are therefore eagerly awaited.

IV. B-cell-Directed Treatment

The effects of immunomodulatory and lymphodepleting treatments primarily targeting B cells and anti-human leukocyte and donor-specific antibodies, such as rituximab (anti-CD20), bortezomib and carfilzomib (both proteasome inhibitors), are mainly described in the context of antibody-mediated rejection (Neuhaus et al., 2022; Pham et al., 2021; Razia et al., 2022; Roux et al., 2016; Vacha et al., 2017; Yamanashi et al., 2020). However, evidence for their relevance as part of CLAD treatment is lacking. We can speculate that these agents might have a beneficial effect when given in combination with other therapies, as antibodies and various subsets of B cells are involved in CLAD pathogenesis (Bos et al., 2022c). However, combination therapy may increase the complexity of treatment and risk of side effects.

V. Future Directions

Interestingly, there are many similarities between CLAD and pulmonary chronic GvHD after hematopoietic stem cell transplantation, as described elsewhere (Bos et al., 2022a). This could imply that therapies developed for (pulmonary) GvHD may also be effective in CLAD, and vice versa, which deserves further attention. Indeed, efforts are needed from both academia and industry for devoted development of novel (more efficacious and safer) immunosuppressive agents, or drug repurposing, along with innovative trial designs with relevant clinical endpoints focusing on these devastating conditions, which are an unmet need.

A. Tyrosine Kinase Inhibitors

Imatinib and ibrutinib are two tyrosine kinase inhibitors commonly used in chronic GvHD with some evidence for their use in pulmonary GvHD. Imatinib (100–400 mg daily) seemed to stabilize FEV₁ in some BOS patients after allogeneic hematopoietic stem cell transplantation, and in subgroup analyses of some patients treated with imatinib for chronic GvHD (Magro et al., 2009; Olivieri et al., 2013, 2009; Parra Salinas et al., 2021; Stadler et al., 2009; Sánchez-Ortega et al., 2016; Watanabe et al., 2015). There is minimal data from preclinical animal studies regarding the use of imatinib in CLAD, showing that imatinib improved luminal airway obstruction in experimental bronchiolitis obliterans (Pandolfi et al., 2020; von Suesskind-Schwendi et al., 2013; Watanabe et al., 2017), possibly through reduction of migration and differentiation of fibrocytes in the allograft (Watanabe et al., 2017).

Currently, no data are available on ibrutinib (140–420 mg daily) in CLAD nor from pulmonary GvHD-specific studies, although some stabilization of lung function was observed in subgroup analyses of chronic GvHD studies (Doki et al., 2021; Kaloyannidis et al., 2021). More data on the use of tyrosine kinase inhibitors in pulmonary GvHD and CLAD are needed to decide whether there is sufficient efficacy in stabilizing lung function or not.

B. Janus Kinase Inhibitors

Ruxolitinib (5–10 mg b.d.) is a relatively new JAK-1/2 inhibitor used with good results in chronic GvHD and some promising results in pulmonary GvHD as well (Streiler et al., 2020; Zhao et al., 2021). There is currently no data in CLAD yet. Promising results of another JAK-1 inhibitor, itacitinib (400–600 mg daily), in a phase 1 study with 23 BOS patients were recently presented, demonstrating that treatment with itacitinib resulted in stabilization of FEV₁ in all participants who continued treatment with an absolute increase of ≥10% in 22% of patients (Diamond et al., 2022). Further results from phase 2 as well as results from a phase 2 trial in steroid-refractory chronic GvHD (NCT04200365) and phase 1 trial in pulmonary GvHD (NCT04239989) are awaited.

C. Rho Kinase Inhibitors

Belumosudil (200–400 mg daily) is a rho kinase inhibitor recently approved for the treatment of chronic GvHD after failure of at least two prior lines of systemic therapy in the USA, and will soon be available in the UK as well. Its efficacy merits further investigation in both pulmonary GvHD and CLAD, as in two recent phase 2 chronic GvHD studies, it resulted in a ≥ 10% increase in FEV₁ in 55% of 47 (Cutler et al., 2021) and 71% of 17 (Jagasia et al., 2021) subjects with pulmonary GvHD.

D. MEK Inhibitors

MAPK/ERK kinase (MEK) inhibitors inhibit the mitogen-activated protein kinase kinase enzymes MEK1 and/or MEK2. Trametinib, a MEK-1/2 inhibitor, ameliorated the onset of GvHD (Itamura et al., 2021, 2016) and chronic rejection after lung transplantation (Takahagi et al., 2019) in some animal studies, highlighting the need for further translational research.

E. IL-6 Inhibitors

The humanized IL-6 receptor antibody tocilizumab prevents binding of IL-6 to its receptor and signal transducer glycoprotein 130 complex, inhibiting downstream JAK/STAT signaling, and has been used in a limited number of studies after allogeneic hematopoietic stem cell transplantation for both acute and chronic GvHD prevention and treatment (Drobyski et al., 2011; Ganetsky et al., 2019; Kattner et al., 2020; Kennedy et al., 2021; Melgarejo-Ortuño et al., 2021; Roddy et al., 2016; Yucebay et al., 2019). In a study of chronic GvHD patients (8 mg/kg q4w), the response rate for pulmonary GvHD ranged between 17 and 33% within the first year of treatment initiation (Kattner et al., 2020). One case report is available in the setting of CLAD in a patient transplanted for COPA syndrome, a genetic disorder leading to upregulation of pro-inflammatory cytokines (primarily IL-1β and IL-6) and development of interstitial lung disease (Riddell et al., 2021). Involvement of IL-6 in the pathogenesis of CLAD has also been documented (Bos et al., 2022b), and tocilizumab (4 mg/kg monthly for 3 doses) effectively suppressed IL-6 upregulation though without clinical improvement in this patient (Riddell et al., 2021). Moreover, one conference abstract demonstrated stabilization of lung function in nine CLAD patients who received tocilizumab (4–8 mg/kg monthly) for at least 3 months, but in combination with other therapies such as ATG, rituximab and immunoglobulins (Ross et al., 2019). Another conference abstract reported reduced onset of rejection when tocilizumab was added in a preclinical animal model, possibly via transient expansion of regulatory T cells (Aoyama et al., 2016).

Further evaluation of a potential role for tocilizumab in the treatment of pulmonary GvHD and CLAD in larger trials is warranted.

F. Inhaled Liposomal Cyclosporine A

Local intrapulmonary lymphocyte suppression and immunomodulation via nebulized immunosuppressive drugs, such as liposomal cyclosporine, may be an elegant way to prevent systemic side effects and ensure high local efficacy in CLAD. Following prior studies demonstrating a possible beneficial effect for CLAD prevention (Groves et al., 2010; Iacono et al., 2019; Neurohr et al., 2022), currently two studies in CLAD are ongoing (BOSTON-1 and BOSTON-2) which results are eagerly awaited.

VI. Conclusion

CLAD is the leading cause of death beyond the first year after lung transplantation (Chambers et al., 2019). Some patients experience an accelerated loss of lung function, whereas others have a slower progression with intermittent loss of function (Belperio et al., 2009; Sato et al., 2013). Several therapeutic options have been used in attempts to prevent, reverse, or slow CLAD progression; however, there are only limited effective therapeutic options and there is currently no consensus on the most effective option (Verleden et al., 2019). Interpretation of these results is overshadowed by the fact that randomized controlled trials are almost universally lacking; thus, it is unclear whether the attenuated rate of FEV₁ decline represents true treatment response or merely the natural course of the disease. In advanced CLAD stages, a less pronounced decline in lung function may also be due to limited residual lung function (Kotecha et al., 2021). However, some studies showed sustained lung function stabilization (Jaksch et al., 2012; Kotecha et al., 2021; Moniodis et al., 2018; Robinson et al., 2017) or improvement even in advanced CLAD (Del Fante et al., 2015; Girgis et al., 2020; January et al., 2019; Lebeer et al., 2020; Thachuthara-George et al., 2015; Vazirani et al., 2021). Secondly, comparing studies is complicated because of different treatment dosages and regimens used, also with respect to other transplant-related drugs and center-specific policies. Furthermore, the comparison of results is hampered by the use of different definitions of CLAD prior to an international consensus and of treatment response, highlighting the need for standardization and harmonization.

Knowledge of the mechanisms-of-action of existing drugs is an essential prerequisite that allows us to understand how a treatment works, but also the expected side effects, and may allow identification of other treatment options, targeting similar immune cells or pathways. Taking into account the efficacy and risk of side effects, we believe that ECP, ATG, and TLI currently have the most promising data to suggest they could be considered second-line lymphocyte-targeted treatment options for CLAD patients (Fig. 2). As intercurrent infections may drive CLAD onset and progression, however, the need for safer lymphocyte-directed therapies has become clear.

Fig. 2.

Lymphocyte depleting and/or modulating therapies in CLAD. (A) Suggested treatment algorithm for CLAD based on existing data taking into account the efficacy and risk of side effects as well as some potential safer future options that require more investigation. (B) Overview of features associated with ATG, TLI, and ECP treatment. Which therapeutic option is chosen mainly depends on local resources and patient profile (e.g., risk of infection, CLAD phenotype, rapid versus slow lung function decline) and preferences. AZA, azathioprine; CsA, cyclosporine A; IS, immunosuppressive; L-CsA, liposomal cyclosporine A; MMF, mycophenolate mofetil; TAC, tacrolimus.

To improve future treatments in lung transplantation, standardization of care, trial protocols, and relevant study endpoints, which include lung function, overall survival and preferably also quality of life and exercise capacity between different transplant centers, are key. Larger randomized controlled multi-center trials, preferably also including RAS patients, with longer follow up as well as platform trials moving rapidly between investigational agents and further investigation of novel treatment options are urgently needed to define the most appropriate treatment algorithm for CLAD. A list of currently ongoing clinical trials is provided in Table 3.

TABLE 3.

Ongoing clinical trials in CLAD using lymphocyte depleting or modulating drugs (registered at clinicaltrials.gov or NIHR)

Study identifier, country	Title	Study design
NCT02181257, USA	Extracorporeal Photopheresis for the Management of Progressive Bronchiolitis Obliterans Syndrome in Medicare-Eligible Recipients of Lung Allografts	Randomized controlled open-label multicenter trial
NIHR130612, UK	Extracorporeal Photophoresis in the treatment of Chronic Lung Allograft Dysfunction: a randomized controlled trial (E-CLAD UK)	Randomized controlled open-label multicenter trial
NCT04792294, Austria	Multicenter Analysis of Efficacy and Outcomes of Extracorporeal Photopheresis as Treatment of Chronic Lung Allograft Dysfunction	Retrospective multicenter trial
NCT03978637, USA, Canada, Belgium	An Open-Label, Single-Arm, Phase 1/2 Study Evaluating the Safety and Efficacy of Itacitinib in Participants With Bronchiolitis Obliterans Syndrome Following Lung Transplantation	Phase 1–2 open-label multicenter trial
NCT04640025, USA, Canada, Europe	A Phase 2, Open-Label, Multicenter, Rollover Study to Provide Continued Treatment of Participants Previously Enrolled in Studies of Itacitinib (INCB039110)	Phase 2 open-label multicenter trial
NCT03657342, USA and Europe	A Phase III Clinical Trial to Demonstrate Efficacy / Safety of Liposomal Cyclosporine A + Standard of Care (SoC) versus SoC Alone in Treating Chronic Lung Allograft Dysfunction / Bronchiolitis Obliterans in Patients Post Single Lung Transplant (BOSTON-1)	Phase 3 randomized controlled multicenter trial
NCT03656926, USA and Europe	A Phase III Clinical Trial to Demonstrate Efficacy / Safety of Liposomal Cyclosporine A + Standard of Care (SoC) versus SoC Alone in Treating Chronic Lung Allograft Dysfunction / Bronchiolitis Obliterans in Patients Post Double Lung Transplant (BOSTON-2)	Phase 3 randomized controlled multicenter trial
NCT04039347, USA and Europe	A Phase III, Extension Clinical Trial to Demonstrate Efficacy and Safety of Liposomal Cyclosporine A Via the PARI Investigational eFlow Device and SoC in Treating Bronchiolitis Obliterans in Patients Post Single or Double Lung Transplant	Phase 3 open-label multicenter trial

Abbreviations

ATG

anti-thymocyte globulin

BOS

bronchiolitis obliterans syndrome

CLAD

chronic lung allograft dysfunction

ECP

extracorporeal photopheresis

FEV₁

forced expiratory volume in one second

GvHD

graft-versus-host disease

JAK

Janus kinase

MEK

mitogen-activated protein kinase kinase

mTOR

mammalian target of rapamycin

RAS

restrictive allograft syndrome

TLI

total lymphoid irradiation

TNFα

tumor necrosis factor alpha

Authorship Contributions

Participated in research design: Bos, Vos.

Performed data analysis: Bos.

Wrote or contributed to the writing of the manuscript: Bos, Pradère, Beeckmans, Zajacova, Vanaudenaerde, Fisher, Vos.