Combined radiation- and immune checkpoint-inhibitor-induced pneumonitis – The challenge to predict and detect overlapping immune-related adverse effects from evolving laboratory biomarkers and clinical imaging

Abstract

The risk of overlapping pulmonary toxicity induced by thoracic radio(chemo)therapy and immune checkpoint inhibitor therapy in the treatment of patients suffering from non-small cell lung cancer (NSCLC) is one important challenge in successful radioimmunotherapy. In the present opinion we highlight factors that we find important to be considered before treatment initiation, during the treatment sequence, and after treatment completion combined or sequential application of radio(chemo)therapy and immune checkpoint inhibitor therapy. A major aim is to optimize the therapeutic index and to avoid immune related adverse effects. The goals in the future will be focused not only on identifying patients already in the pretreatment phase who could benefit from this complex treatment, but also in identifying patients, who are most likely to have higher grade toxicity. In this respect, proper assessment of clinical performance status, monitoring for the presence of certain comorbidities, evaluation of laboratory parameters such as TGF-α and IL-6 levels, human leukocyte antigens (HLA), and consideration of other potential biomarkers which will evolve in near future are essential. Likewise, the critical parameters must be monitored during the treatment phase and follow-up care to detect potential side effects in time. With the help of high-end imaging which is already used on a daily basis in image guided radiotherapy (IGRT) for intensity modulated radiotherapy (IMRT), its advanced form volumetric modulated arc therapy (VMAT), and adaptive radiation therapy (ART), clinically relevant changes in lung tissue can be detected at an early stage of disease. Concurrent radiotherapy and immunotherapy requires a special focus on adverse events, particularly of the lung, but, when properly approached and applied, it may offer new perspectives for patients with locally advanced NSCLC to be seriously considered as a curative option.

Patients with locally advanced non-small cell lung cancer (NSCLC) are typically treated with fractionated radiotherapy to the thoracic region concurrent with platinum-based chemotherapy [1]. Since the landmark PACIFIC trial, immune consolidation therapy with immune checkpoint inhibitors (ICI) has become a further essential component of multimodality treatment strategies for patients with locally advanced, unresectable stage III NSCLC undergoing definitive radiochemotherapy [2]. Overall and progression-free survival of NSCLC patients receiving durvalumab consolidation after definitive radiochemotherapy was shown to be significantly longer with durvalumab consolidation than without [2], [3]. Even more, secondary endpoints were also in favor of durvalumab at a comparable safety [2], [3]. Hence, consolidation therapy with ICI after definitive radiochemotherapy has become the standard of care for locally advanced NSCLC with programmed death ligand 1 (PD-L1) expression of more than 1%. Similarly, programmed cell death protein 1 (PD-1) antibody treatment with pembrolizumab has become a preferred treatment strategy and first-line therapy for stage IV NSCLC patients with high ≥50% PD-L1 expression and negative actionable driver mutations [4]. Since then, various combinations of PD-1/PD-L1 antibodies with chemotherapy have been approved for first line therapy of metastatic non-small cell lung cancer (NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®), NCCN Guidelines® Version 3.2022 https://www.nccn.org/professionals/physician_gls/pdf/nscl.pdf).

In addition, a prospective trial of concurrent PD-L1 antibody therapy and thoracic radiotherapy has demonstrated the safety of combined treatment [5]. There is raising experimental evidence that concurrent PD-L1 antibody therapy and radiotherapy is more effective than sequential therapy [6]. Therefore, complex multimodality treatment protocols involving induction chemoimmunotherapy followed by radiochemoimmunotherapy offer new perspectives for NSCLC patients. Clinical trials are currently conducted to evaluate the efficacy of concurrent ICI therapy in the setting of induction chemotherapy and concurrent radiochemotherapy in patients with resectable stage III NSCLC with positive PD-L1 expression [e.g. randomized investigator-initiated trial ESPADURVA at University Hospital Essen, (PIs, W. Eberhardt, M. Stuschke: EUDRACT Number: 2019-000058-77; start 12/2019)].

Although there is growing evidence supporting efficacy and significant synergy of combined or sequential radioimmunotherapy, several factors need to be considered before treatment initiation, during the treatment sequence, and finally after treatment completion.

Only a fraction of patients is sensitive to treatment with concurrent radiochemotherapy, ICI, or combined treatment (responders). Some patients fail to ever respond (intrinsic resistance). Other patients develop therapy resistance already after an initial response phase (acquired resistance) or develop adverse effects that impact survival or decrease the quality of life [7], [8], [9]. It is thus highly important to define biomarkers predicting a high probability for response and to identify patients who are at high risk to develop acquired resistance or adverse effects upon combined or sequential radioimmunotherapy.

Cancer immunotherapies aim at exploiting immune mechanisms that facilitate efficient cancer cell killing, e.g., by the (re)-activation of tumor-specific cytotoxic CD8⁺ T cells [10]. An efficient T cell-mediated immune defense against cancer requires both, the recognition of tumor antigens by antigen-presenting immune cells, and the activation of tumor-specific cytotoxic CD8⁺ T cells. ICI therapy can overcome an exhausted T cell phenotype and re-activate suppressed effector T cells against cancer cells, at best a CD8⁺ T cells response, and thus overcome this type of tumor immune escape [11,12]. However, by disturbing immune homeostasis ICI can also exert a new form of feared serious adverse effects, the so-called immune-related adverse effects (irAE) [13], [14], [15].

The rationale for combining ICI-therapy and radiotherapy is based on the findings that in addition to inducing damage to cellular macromolecules, particularly the DNA, and associated growth arrest or cell death, radiotherapy enhances immunogenicity of cancer cells. Furthermore, radiotherapy acts as immune adjuvants that modulates the local environment and activates local and systemic innate and adaptive immune mechanisms and thereby further supports ICI-mediated immune enhancement [16], [17], [18]. Radiation-induced cell damage, permanent growth arrest (senescence) or cell death were shown to stimulate Type I interferons and other inflammatory cytokines/chemokines: involved pathways include for example, i) the cytosolic recognition of released nuclear and mitochondrial DNA fragments via cGAS-STING (cyclic GMP-AMP synthase/stimulator of interferon (IFN) genes) and subsequent IRF3 (interferon regulatory factor 3) signaling; ii) the release and recognition of damage associated molecular patterns (DAMPs) via pattern recognition receptors, e.g., TOLL-like receptors, and subsequent nuclear factor kappa B (NF-κB) signaling; iii) activation of the inflammasome, or iv) the release of immune mediators from senescent cells [19], [20], [21], [22], [23]. Furthermore, radiotherapy leads to the modulation of peptides, an increase in HLA expression, and the release of tumor antigens upon induction of cell death [24] and thereby further supports more efficient antigen presentation and recognition. Synergistic effects of ICI and radiotherapy may also relate to radiotherapy-induced upregulation of immune checkpoints, e.g., PD-L1 [25].

The combined treatment with radiotherapy and ICI therapy therefore appears to be an optimal tool to defeat cancer, were it not for the individual factors of the patients that influence the efficacy and toxicity of radiotherapy, immunotherapy, or their combination. Radiotherapy-induced immune responses are influenced by physical factors and by tumor intrinsic, environmental and other health factors that can lead to different treatment outcomes between individual patients, including induction of immunosuppression [25] or immune-related toxicities [21,22]. Various reports have revealed that factors influencing antigenicity, adjuvanticity as well as homeostatic feed-back mechanisms can cause tumor immune escape and also drive intrinsic or acquired resistance to ICI [26], [27], [28]. We thus speculate that complex interactions between the biology of individual tumors (e.g., a defective DNA damage response or disturbed DNA repair) [29,30], physical aspects of radiotherapy (e.g., dose, fractionation, volume, and quality) [31], alterations in innate immune sensing, multifaceted metabolic and immune adaptations [25,[32], [33], [34], as well as environmental and patient-specific health factors will determine individual sensitivity of tumors to immune enhancement induced by radiotherapy and combined radioimmunotherapy. These parameters will also influence the induction of adverse effects of radiotherapy in normal tissues, and might be critical for altered toxicity rates or new toxicities upon combined or sequential radioimmunotherapy in sensitive patients.

Adverse effects of radiotherapy in the lung and heart are dose limiting for radiotherapy and combined modality treatments [35]. According to a recent metaanalysis and systematic literature review the risk of severe pneumonitis grade 3-5 in NSCLC patients from chemoradiation ranges from 3.62% to 7.85% [36]. In the PACIFIC trial, the most frequent adverse events leading to discontinuation of durvalumab were pneumonitis or radiation pneumonitis in 6.3% and pneumonia in 1.1% of the patients [2]. Those receiving placebo developed pneumonitis or radiation pneumonitis in 4.3% and pneumonia in 1.3% [2]. In patients who received durvalumab consolidation, pneumonitis or radiation pneumonitis of grade 3 or 4 occurred in 3.4% compared to 2.6% of those who received placebo; and pneumonia of grade 3 or 4 occurred in 4.4% compared to 3.8% of those who received placebo [2]. Jabbour et al. report grade 3 or higher pneumonitis in 6.9-8.0% of NSCLC patients of a prospective phase-II-trial undergoing radiochemotherapy with concurrent pembrolizumab [37]. The overall rate of grade 3+ pneumonitis was 8.1% in three consecutive phase-I-trials examining pneumonitis in patients treated with stereotactic body irradiation and ICI [38]. Obviously, the incidence of pneumonitis may vary greatly depending on the therapeutic regimen (Table 1). Thus, concurrent with the development of new treatment options and strategies, questions regarding safety and dose-limiting toxicity, particularly to the lung and heart, must be addressed. A particular concern poses the risk of development of a radiation-/ immune-mediated lung injury, which in the early phase presents like a pneumonitis and in the late phase, is characterized by pulmonary fibrosis.

Table 1.

Similarities and differentiating characteristics of pneumonitis and lung fibrosis after radio(chemo-)therapy compared to concurrent immunotherapy with radio(chemo-)therapy.

	Pneumonitis and lung fibrosis after RCTx	Pneumonitis and lung fibrosis after combined immunotherapy and RCTx
Prevalence	According to a recent metaanalysis and systematic literature review the risk of severe pneumonitis grade 3-5 in NSCLC patients from concurrent chemoradiation ranges from 3.6% to 7.9% [36].	The incidence of grade 3-5 pneumonitis after sequential radiochemotherapy followed by immunotherapy is estimated at 0.4-7% [2,78,79].
	Incidence of pneumonitis grade 3-5 under simultaneous radiochemo-/immunotherapy is reported by several prospective trials between 3.0-11.7% [37,80,81].
Temporal changes	Radiation induced signs of pneumonitis increase after 3 months and stabilize at 1 year [82].	May occur and worsen acute within several days; median time to onset of pulmonary adverse events is reported 2.8 months, with a wide range from several days to 19.2 months [83], [84], [85].
Clinical characteristics	Asymptomatic or non-specific respiratory symptoms such as dyspnea, dry cough, shortness of breath, worsening shortness of breath or cough, possibly fever.  More likely to be confined to RT-field. Imaging findings include at an early-stage ground-glass opacities, patchy consolidation, reticular pattern, late changes show a scarlike pattern. Important classification system of side effects on normal tissues are LENT SOMA scale [86] and CTCAE criteria [87], both allowing assessment of side effects after surgery, chemo- or radiotherapy and after combined treatment modalities.	Asymptomatic or non-specific respiratory symptoms such as dyspnea, dry cough, shortness of breath, worsening shortness of breath or cough, possibly fever. More likely bilateral, and involved more lobes [88]. CT is the recommended imaging modality for the evaluation of IR-pneumonitis [14]. Imaging findings are described according to American Thoracic Society and European Respiratory Society classifications of interstitial pneumonias [89], [90], [91], [92], [93].  LENT SOMA scale and CTCAE criteria may be used for grading of irAE.
Treatment	Mainstay of therapy are systemic corticosteroids; consider supportive treatment with antitussive therapy, oxygen, and antibiotics, consider pneumocystis prophylaxis for patients with risk factors [94], [95], [96]. Treatment escalation for non-responders with grade 3+ pneumonitis should follow symptom orientated institutional guidelines. A severity dependent dosage is discussed in [97].	Corticosteroids; consider oxygen and pneumocystis prophylaxis; for grade 3+ pneumonitis: permanently discontinue ICI-treatment, if no improvement after 48 hours after corticosteroid initiation may add infliximab [56] or cyclophosphamide, in corticosteroid-refractory pneumonitis follow consensus guidelines for escalation [55,56]. Supportive treatment alike chemoradiation-induced pneumonitis.
Long-term consequences	Risk of local lung fibrosis and worsening pulmonary function	Risk of ubiquitous pulmonary fibrosis and worsening pulmonary function

From the biological point of view both, radiotherapy and immunotherapy, risk evoking harmful effects in the lung tissue that are immune associated, namely radiation-induced lung injury (RILI) and immunotherapy-related lung injury (IRLI), respectively [10,21,39].

The mechanisms driving RILI have been intensely studied in the past: Radiotherapy-induced damage to parenchymal cells, vasculature, and/or stroma is followed by recruitment and activation of immune cells, subacute radiation pneumonitis, and potentially, chronic inflammation and fibrosis [21,40]. On the molecular level, ionizing radiation causes DNA damage and reactive oxygen species generation in irradiated cells and tissues resulting in cell death or senescence, release of DAMPs (e.g., HMGB1, calreticulin, ATP), and of various cytokines/chemokines (e.g., TNFα, TGFβ, IL-1β, IL-6, IL-8, IFNs, Ccl-2), to evoke and promote tissue inflammation involving similar pathways as described above [21,41,42].  The initiated inflammatory response can lead to an acute radiation pneumonitis, and in case of a chronic scenario result in immune deviation, an abnormal wound healing response, and pulmonary fibrosis involving diverse signaling pathways, and cell types (for a detailed review see [10]).

In contrast, the mechanisms driving IRLI and the suspected interaction between RILI and IRLI remain still underexplored: ICI therapy dysregulates immune homeostasis, resulting in serious local irAE such as ICI pneumonitis [43], or systemic untargeted inflammation or autoimmunity as described for ICI therapy in melanoma, lung, kidney, and other cancers [43], [44], [45], [46]. ICI-related irAE involve enhanced cytokine secretion (e.g., IL-1β, IL-6, TNF-α, TGF-β, IL-10, IFN-γ, IL-17) [40,43,[47], [48], [49]. Moreover, antigens shared by tumor and normal lung tissue can be recognized by reactivated T cells, thus leading to cytotoxic effects in normal tissues and tissue damage. The resulting release of DAMPs as well as release of dsDNA and activation of the cGAS/STING pathway can further fuel inflammatory conditions. Other studies highlight, that preexisting autoantibodies as well as epitope spreading from neo- and tumor-antigens in normal tissues contribute to irAE (e.g., pneumonitis), especially in lung cancer patients, thus causing misdirected immune responses in neighboring normal tissues and distant organs [50], [51], [52]. In contrast, further studies are needed to gain a better understanding of the role of IRLI in the development of pulmonary fibrosis.

Cell damage caused by radiotherapy and the immunotherapy-induced immune boost both result in common endpoints, namely the release of several cytokines/chemokines and the induction of damage-associated signaling pathways (for a review see [10]). Consequently, overlapping mechanisms of pulmonary toxicity are suspected, that may result in life-threating acute tissue responses [53,54]. However, the complex immune interactions are still inadequately understood and further studies are needed to determine how to reach equilibrium in order to amplify antitumor effects and minimize adverse effects upon sequential or concurrent combined treatment. For optimizing the therapeutic gain of radio(chemo)immunotherapy and our understanding of the involved processes, it will be important to reveal the diverse aspects of the immune status individually for each patient.

Diagnosis and potential biomarkers

Currently, diagnosis of pneumonitis is difficult, primarily made by exclusion of an infectious focus, repetitive blood cultures, clinical assessment, oxygen saturation, pulmonary function tests and changes observed on imaging. Particularly, drawing a differential diagnosis between RILI and IRLI in clinical routine is challenging. Table 1 summarises similarities and discrepancies between RILI and IRLI. Depending on clinical severity, the initial treatment of "adverse effects after either radiotherapy (RILI) or concurrent immunotherapy with RCTx (IRLI)" involves corticosteroids and the avoidance of the underlying cause, particularly for grade III and grade IV toxicity. While the initial treatment comprises systemic corticosteroids, if there is no improvement within 48 hours after corticosteroid initiation in grade 3+ pneumonitis follow consensus guidelines for toxicity management [55,56]. More and more studies support the effort to diagnose a pneumonitis with the help of biomarkers and to differentiate various types of pneumonitis by cytokine levels as well as the lymphocytic landscape [43]. Upregulation of pro-inflammatory and downregulation of the counterregulatory anti-inflammatory cellular mechanisms is identified as an important contributor to the development of lung tissue changes [43].

Prior to initiation of treatment, it is critical to perform a thorough clinical assessment and to evaluate pulmonary and cardiac function, a clinical assessment, which is mandatory anyway before starting treatment, and an interdisciplinary tumor board evaluation. Besides clinical examination, functional lung and cardiac tests, blood serum tests and imaging, bronchial brushes and washes and bronchioalveolar lavage (BAL), if clinically indicated, may give additional information on the patient`s innate pulmonary lymphocytic landscape. Furthermore, sputum cultures and BAL, if indicated, allow the exclusion of infectious diseases. Advanced imaging techniques such as computed-tomography (CT), ⁶⁸GaFAPI- and ¹⁸FFDG-positron emission tomography (PET)/CT allow a differentiation of other underlying conditions and tumor progression.

Previous chemotherapy and/or radiotherapy, as well as previous immunotherapy, may be a serious contraindication to such complex treatment regimens, depending on the anatomic region and previous treatment history. Inadequate patient compliance (e.g., symptomatic psychiatric disorders) or severe weight loss of more than 10% in the last six months are additional factors that may lead to high treatment-related toxicities. Any concurrent chemotherapy, biologics, or hormonal therapy for cancer treatment, immunosuppression other than corticosteroids to treat an adverse event, or recurrence of an adverse event with re-challenge are always critical issues that need to be discussed in an experienced interdisciplinary setting.

The lung dose required to induce tumor-cell death can be a non-negligible factor in lung tissue changes [57], especially in a diseased lung. It is reported that the mean lung dose, V5, V10, V13 and V20 (amount of radiation dose delivered to the percentage of lung) are closely related to the development of pneumonitis and fibrosis [58,59].

A special focus must also be laid on asymptomatic and subclinical interstitial lung disease. According to Lee et al. the risk of radiation-mediated pneumonitis grade ≥ 2, ≥ 3, or ≥ 4 was significantly higher in patients with preexisting, interstitial lung changes irrespective of grade (grade 2, 15.6% to 46.7%, p=0.03; grade 3, 4.4% to 40%, p=0.002; grade 4, 4.4% to 33.3%, p=0.008) [60]. Preexisting, interstitial lung changes and elevated TGFß1, interleukins 1, 6, 8, and 10, Krebs von den Lungen-6 (KL-6) and surfactant proteins (SP) are reported to be associated with the development of a clinically relevant RILI [57,61].

Besides radiation, a whole range of factors can contribute to the development of pneumonitis. Similarly, to a radiation mediated pneumonitis, Cho et al. described that the presence of preexisting interstitial lung disease odd ratio (OR), 6.03; 95% confidence interval (CI), 1.19–30.45; p = 0.030 was associated with a higher incidence of ICI-related pneumonitis [62]. Real-world data suggests an increased incidence of 19 % and an increased mortality of immune-related pneumonitis in NSCLC patients treated with ICI [39]. Particularly increased TGF-α and IL-6 levels seem to be associated with immune-mediated pneumonitis [63,64].

Henceforth, a particular concern that must be considered before initiating radioimmunotherapy is a medical history of preexisting lung or airway disease. Idiopathic pulmonary fibrosis, pneumonitis (including drug-induced), organizing pneumonia (i.e., bronchiolitis obliterans, cryptogenic organizing pneumonia, etc.), or evidence of active pneumonitis on screening chest CT represent all changes which may lead to a multiple risk excess of treatment-related lung injury [57,60,65].

The medical history of allogeneic organ transplantation, stem cell transplantation, active or previous documented autoimmune or inflammatory disease (including inflammatory bowel disease e.g, colitis or Crohn's disease, diverticulitis, or rheumatic diseases such as systemic lupus erythematosus, sarcoidosis syndrome or Wegener's syndrome [granulomatosis with polyangiitis], Graves' disease, rheumatoid arthritis, pituitary inflammation, uveitis, etc.) are all medical conditions that must be closely monitored to avoid "irAE" introduced above instead of immune related adverse effects [66], [67], [68], [69], [70], [71], [72], [73], [74], [75], [76]. The incidence of pneumonitis seems to be closely linked to HLA-B*35 and DRB1*11, alleles associated to autoimmune diseases [77]. Recent work further indicates that serum and BAL fluid (BALF) cytokine levels, as well as the lymphocytic landscape, may help to differentiate various types of pneumonitis.

Thus, these medical conditions may pose relevant implications for the treatment with ICI of patients with NSCLC [77].

While the pretreatment phase focuses on identifying and inclusion of patients who may profit from this complex treatment, during treatment, it is essential to guard the patient from immune-related, radiation-related, or combined immune- and radiation-related toxicities. Respiratory gating, high-quality image-guidance of image-guided radiation therapy (IGRT), intensity modulated radiotherapy (IMRT) and its advanced form volumetric modulated arc therapy (VMAT), are nowadays essential features and indispensable in modern radiotherapy of lung cancer. These make radiotherapy a gentle, non-invasive form of treating not only locally advanced and inoperable tumors with steep dose gradients to organs-at-risk. In addition, adaptive radiotherapy (ART) is the latest innovation that allows online onboard plan adjustment for best possible dose reduction to organs at risk with optimal target coverage.

Daily performed cone-beam computed tomography (CBCT), essential for precise radiotherapy, gives information beyond plain positioning accuracy. Clinically relevant lung tissue changes such as dystelectasis, effusion or new infiltrates may be diagnosed on a low dose CBCT, as the lung represents a high contrast tissue. Daily CBCTs which allow a cross-sectional imaging without overlays are also a key factor in the early diagnosis of treatment-induced lung tissue changes. Thus, accurate monitoring of daily CBCTs was, is and will become more and more essential, overall, for lung cancer patients undergoing concurrent radioimmunotherapy and radiochemoimmunotherapy, for an early identification of immune dysregulation. This important image evaluation must be performed by an experienced radiation oncologist as part of daily image guidance and offline review. In case of clinically important findings such as effusion or dystelectasis further diagnostics must be performed at the discretion of the treating physician.

After completion of treatment, thorough postradiotherapy and oncological follow-up is crucial to exclude late treatment-related adverse events, again by means of a regular thorough clinical assessment, laboratory exams and imaging. Damage to lung tissue, particularly pneumonitis, can result into critical pulmonary fibrosis leading to worsening of lung function. Predisposing factors for the development of pulmonary fibrosis are patient-, treatment-, and tumor-related. The pretreatment related risk factors may eventually evolve in lung tissue injury after successful treatment completion. Considering the above factors, complex treatment protocols and strategies, including radioimmunotherapy and radiochemoimmunotherapy, may provide a safe and effective treatment and open new treatment perspectives for patients with locally advanced NSCLC.

Funding sources

Supported by the Deutsche Forschungsgemeinschaft (DFG) grant numbers GRK1739/2 to VJ and MS, GRK2762/1 to NG, VJ, and MS,. WI 4702/2-1 WI 4702/2-1 to FW, and the Bundesministerium für Bildung und Forschung (BMBF) grant number 02NUK047D to VJ.

CRediT authorship contribution statement

Nika Guberina: Conceptualization, Investigation, Funding acquisition, Writing – original draft, Writing – review & editing. Florian Wirsdörfer: Investigation, Funding acquisition, Writing – original draft, Writing – review & editing. Martin Stuschke: Funding acquisition, Writing – review & editing. Verena Jendrossek: Conceptualization, Investigation, Funding acquisition, Writing – original draft, Writing – review & editing.

Declaration of Competing Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

AstraZeneca- Advisory board function

- Research funding to institution