About 60% of all cancer patients receive radiotherapy (RT) at some point during the course of their disease, and good results in terms of long-term survival and tumor cure are achieved in a variety of tumors by multimodal combinations of surgery, RT, and chemotherapy. Concurrent radiochemotherapy could improve the prognosis of glioma, lung, head and neck, esophageal, cervical, anal, and rectal cancer (1–8) and is part of standard therapy for locally advanced tumors of these entities. Yet, treatment outcome is still unsatisfactory for common forms of cancer with high loco-regional failure rates or frequent development of metastases. Although patient-specific clinical factors may explain some of these failures, it is commonly assumed that biological factors adversely affecting the response of tumor cells to treatment, such as intrinsic radioresistance, tumor promoting mutations, unfavorable gene expression profiles, heterogeneity in radiation responses, or a resistance-promoting microenvironment, significantly contribute to treatment failures (9–14). Acute and late toxicity to normal tissues also limits the radiation dose that can be applied to the tumor, and tolerable doses are often linked to suboptimal tumor control—even accepting side effects that lead to decreased quality of life (15). Normal tissue toxicity also precludes therapy intensification efforts for many locally advanced tumors by the combination with cytotoxic chemotherapy (16–18). As a consequence, there is high interest in improving the therapeutic ratio either by technical and physical innovations in treatment delivery, e.g., intensity-modulated radiation therapy or particle therapy, or by developing effective strategies to prevent or treat the toxic effects of ionizing radiation (IR) in normal tissues without protecting the tumor cells, or to increase intrinsic radiosensitivity of cancer cells without increasing sensitivity of normal tissue cells, respectively.

Dose-limiting side effects in the lung tissue after RT of the thoracic region or total body irradiation in conditioning regimens for hematopoietic stem cell transplantation include inflammatory (pneumonitis) and fibrotic changes (pulmonary fibrosis) (19–21). Radiation-induced damage to the lung tissue leads, like infectious, thermal, or physical damage, to the activation of the immune system. This inflammatory response is needed to orchestrate tissue repair and regeneration in order to restore tissue homeostasis. Depending on the degree of the resulting aseptic inflammation, patients can present with pneumonitis. Radiation-induced pneumonitis can develop at 4–12 weeks after RT with symptoms like fever, chest pain, dry cough, and dyspnea or even respiratory failure in severe cases and occurs in 5–20% of patients with lung or breast cancer (22–24). The pneumonitic phase is characterized by the recruitment of diverse immune cells of myeloid and lymphoid origin and a perpetual cascade of cytokines/chemokines resulting in various degrees of lung inflammation and the described symptoms (Figure 1).

Development of radiation-induced lung fibrosis is mostly observed 6–24 months after RT and may become chronic in patients with a large irradiated lung volume (24). Major symptoms of lung fibrosis are breathing difficulties and subsequent volume loss of the lung (25). Studies from various groups using rodent models and patient samples helped to reveal a complex response of the lung tissue toward irradiation with multiple interactions between resident lung cells including lung-resident mesenchymal stem cells (MSC), locally generated or recruited fibroblasts, and infiltrating immune cells, respectively (26–37). We speculate that a sophisticated network between damaged resident cells (epithelial cells, endothelial cells, and lung-resident MSC), recruited immune cells, and soluble mediators (cytokines, chemokines, growth factors, and proteases) and the resulting environmental changes participate in shaping the observed inflammatory and fibrotic alterations of the lung tissue (Figures 1 and 2).

Among other molecular markers there is evidence from preclinical and clinical studies that T lymphocytes infiltrate the lung to a considerable extent, particularly during the pneumonitic phase at 3–12 weeks post-irradiation, although lymphocyte infiltration was also observed at later time points during the chronic inflammatory/fibrotic phase at 16–30 weeks post-irradiation (30, 35, 38–40). Interestingly, earlier studies described a correlation between the presence of CD4⁺ T cells in the bronchoalveolar lavage of irradiated breast or lung cancer patients and the development of pneumonitis (39, 41, 42). Furthermore, depletion of CD4⁺ T cells during the pneumonitic phase attenuated the development of lung fibrosis upon thoracic irradiation in a murine model (43). These findings suggest a complex role of CD4⁺ T cells in the pathogenesis of radiation-induced lung disease. Antibody-mediated inhibition of the accumulation of CD3⁺ lymphocytes or depletion of CD4 and CD8 T lymphocytes also reduced fibrosis levels in the murine model of pulmonary fibrosis induced by the radiomimetic and DNA-damaging drug bleomycin (BLM) (44, 45). In contrast, the lack of mature T and B lymphocytes in recombination-activating gene 2 (RAG2)-deficient mice exacerbated radiation-induced fibrosis (46). Altogether, these findings highlight that lymphocytes play a complex role in DNA damage-induced lung disease and suggest that depending on the disease stage and the environmental conditions, shaped by the tissue response to the damage, specific lymphocyte subpopulations exert either beneficial or adverse effects (Figure 1). We propose that a disturbed balance between tissue inflammation and repair processes participates in the development of radiation-induced pulmonary fibrosis as it has been described for other fibrotic diseases and that lymphocytes are involved in these processes (47). Nevertheless, it remains to be demonstrated whether lymphocytes directly contribute to radiation-induced lung disease or only modulate disease progression. Furthermore, it remains to be explored whether, besides the myeloid compartment, innate lymphoid cells (ILC) might contribute to radiation-induced fibrosis. Finally, the mechanisms driving radiation-induced lymphocyte deviation remain to be defined.

Lymphocytes are characterized as white blood cells that are homogeneous in appearance but that have various functions. They include T cells, B cells, and ILC among them conventional natural killer (NK) cells. T cells, ILC, and B cells are responsible for the production of cytokines and antibodies (B cells), whereas NK cells can induce direct cell-mediated killing of virus-infected cells and tumor cells. Here, we will focus on a potential role of B and T lymphocytes as well as ILC.

The different major subpopulations of T lymphocytes include CD8⁺, CD4⁺ T cells, NK T cells, and γδT cells. CD8⁺ T cells comprise cytotoxic T cells or cytolytic T cells. They control and eliminate intracellular pathogens and tumor cells and can further differentiate into CD8⁺ memory cells (48). γδT cells express a T cell receptor differing from the conventional αβT cells. The function of γδT cells is poorly understood, but current knowledge implies a role in immunoregulation in pathogen and allergen responses (49). NK T cells are a unique subpopulation of lymphocytes that are mainly involved in innate immunity and will not be further discussed in the present review.

CD4⁺ T cells comprise T_H1 and T_H2 subpopulations. Furthermore, advances in immunology have led to the characterization of newly appreciated CD4⁺ T cell effector populations that regulate the immune response such as interleukin (IL)-17-producing T cells (T_H17 cells), T cells with regulatory function [regulatory T cells (T_reg)], IL-9-secreting T_H9 cells, IL-22-dominant T_H22 cells, and B cell-interacting follicular helper T cells (T_FH), thus revising established paradigms (50–58).

The secretion of interferon (IFN)-γ and the directed elimination of intracellular pathogens characterize a T_H1 response. In contrast, T_H2 responses are shaped by the cytokines IL-4 and IL-13, supporting the defense against parasites, and moreover contribute to the generation of antibodies (59).

T_H17 cells preferentially produce IL-17A-F and play a role in inflammatory processes such as autoimmune diseases and the defense against extracellular pathogens. T_H17 cells further produce the cytokines IL-21, IL-22, and IL-23, which exert strong pro-inflammatory effects (60). T_H17 cells are induced by IL-6, IL-21, and transforming growth factor beta (TGF-β), a potent regulator of lung homeostasis as well as in pathologies (61, 62).

Another important subpopulation of T cells are T_reg. T_reg show a suppressive capacity, control immune reactions, and inhibit exaggerated inflammation (60, 63). T_reg exist as natural occurring T_reg (nT_reg) and induced T_reg (iT_reg). Murine thymus-derived nT_reg are CD4⁺/CD25⁺ and express the transcription factor FoxP3 (murine cell marker), whereas in humans not all T_reg express FoxP3. Therefore, T_reg in humans are mainly characterized via the marker profile CD4⁺/CD25^hi/CD127^low (64). T_reg show their suppressive capacity by secreting anti-inflammatory cytokines like IL-10 and TGF-β, which can induce cell cycle arrest or apoptosis in T effector cells.

Recent studies also revealed a small population of CD8⁺ T_reg at steady state; these CD8⁺ T_reg are present in the human and the murine system, and they express the marker CD25 as well as FoxP3. Churlaud et al. showed that CD8⁺ T_reg are highly suppressive and responsive to IL-2 (65), but further studies are needed to uncover the origin and the role of CD8⁺ T_reg in health and pathologies.

B lymphocytes represent the second heterogeneous group of lymphocytic cells. B cells originate from the bone marrow, mature in the spleen, and differentiate in the lymph nodes into germinal center cells after contact with antigens and T cells (66, 67). Besides their capacity for antibody secretion, they have functions in antigen presentation and secretion of diverse cytokines (68, 69).

The newly identified group of the lymphoid cells, namely ILC, plays an important junction between innate and adaptive immunity. Like other lymphoid cells, they originate from a common lymphoid progenitor in the bone marrow. But in contrast to their relatives, they lack (RAG)-dependent rearrangement of antigen receptors as well as phenotypical markers of myeloid and dendritic cells (70, 71). ILC have been characterized by their expression pattern of the master transcription factors (T-bet, GATA3, and RORγt) and specific cytokines that usually define T cell subpopulations. Based on this categorization, three different subpopulations (ILC1, ILC2, and ILC3) have been defined as follows: (i) ILC1 cells include conventional NK cells and T-bet⁺/IFN-γ-producing cells; (ii) ILC2 cells show GATA3 expression, and they secrete T_H2 cytokines, IL-4, IL-5, IL-9, and IL-13 in response to IL-25 and IL-33; and (iii) ILC3 cells express the transcription factor RORγt, and they release the cytokines IL-17, IL-22, granulocyte–macrophage colony-stimulating factor (GM-CSF) as well as lymphotoxins. ILC act on tissue homeostasis and tissue remodeling; moreover, they participate in regulating immune responses to inflammation and infection in tissues like liver, lymph nodes, and mucosal barriers like gut and lung (72, 73). These characteristics make it highly likely that ILC also modulate the progression of radiation-induced lung inflammation and fibrosis (see Role of Lymphocytes in the Defense of the Lung Tissue).

In the lung, T cells are found in relatively high numbers in the mucosa, in the intraepithelial part, in the underlying lamina propria, and also in the lung parenchyma. T cells of the intraepithelial region express CD8, while CD4 is the dominant surface marker of T cells in the lamina propria. Furthermore, it has been described that CD45RO is expressed on both, CD4 and CD8 T cell subsets, indicating their role as effector and/or memory cells (74). For CD8⁺ T cells it is known that they protect the lung against influenza infection. Nevertheless, there it is believed that CD8⁺ T cells also contribute to lung injury, e.g., during influenza infection due to their cytotoxic effects and the massive production of the pro-inflammatory cytokines IFN-γ and tumor necrose factor (TNF)-α (75), respectively.

Among the CD4⁺ T cells, lung T_H17 cells may not only play a role in neutrophil recruitment and pathogen clearance but also be highly relevant in respiratory inflammatory diseases (76). Moreover, a recent study demonstrated that T_H17 cells also play a role in lung cancer progression: here, the authors analyzed blood samples from patients and described that T_H17 as well as T_reg subsets are involved in the immunopathology of NSCLC (77). In other studies, T_reg were found to counteract the inflammation-induced injury to the airways associated with lung infections as well as the development of atopic diseases. T_reg also play a role in mediating inhalation tolerance and in controlling allergen-specific T cells from activation (74, 78). Interestingly, there is evidence from some studies that T_reg contribute to fibrotic diseases in the lung by promoting a pro-fibrotic microenvironment (79–81).

Also, γδ T cells show functions in the airways. For the lungs it is described that γδ T cells reside in the subepithelium of alveolar and non-alveolar regions (82). Here, they modulate immune responses against allergens and pathogens (49, 83). A study from Simonian et al. also revealed a role in inflammation (hypersensitivity pneumonitis)-induced lung fibrosis. Here, γδ T cells diminished CD4⁺ cell recruitment by the secretion of regulatory IL-22 thereby protecting the lung from fibrosis (84).

However, the role of T_H9, T_H22, and T_FH cells in lung pathologies is still underexplored. So far, current studies hint to a role for T_H22 and T_FH cells in host defense against viruses and bacteria in the lung, whereas T_H9 cells seem to play an important role in asthma (85–87).

Similar to T cell populations, B cells are also present in the parenchymal lung as well as in the conducting airways. In the lamina propria, they act as antibody-secreting plasma cells producing immunoglobulin (Ig)A but may also contribute to local antigen presentation (74, 88). In the lung, B cells can act, for example, as memory B cells producing IgA and IgG neutralizing antibodies that have the ability to protect against pulmonary viral reinfections (89).

The group of ILC is in the focus of current research. ILC have been implicated in immunity of the mucosal barrier in the lung, e.g., in allergic asthma, hyper responsiveness, and viral infection (90–93). The present knowledge on ILC has nicely been summarized in two recent reviews with a focus on their roles in the lung tissue (94, 95). Though these reviews emphasize that the role of ILC in the lung is still poorly characterized, we will highlight some important observations at this point. Among the three ILC subsets defined so far, ILC2 present the main ILC in the murine lung, but with 2–3 × 10⁴ cells per lung and thus 0.4–1% of total lung cells, these ILC seem to be a relatively rare population, at least under physiological conditions (91). But, pathological conditions in the lung are associated with changes in the ILC population (96). In this context, ILC2 and ILC3 seem to play more imported roles than ILC1 during chronic lung disease in both, mice and men (94, 97, 98). For example, in patients with chronic obstructive pulmonary disease (COPD) ILC3 constitute the major population with 60% of all ILC, whereas ILC2 amount to 30% and ILC1 to 10% of the ILC (98). Of note, depletion of ILC2 reduced epithelial integrity, induced epithelial degeneration, and impaired lung functions during influenza virus infection highlighting a protective role of ILC2 for these processes (91).

First reports suggest that ILC may also play an important role during acute pulmonary inflammation. In this context, ILC3 have recently been identified as a major source of IL-17A thereby inducing neutrophil recruitment in a murine model of LPS-induced acute respiratory distress syndrome (99). Furthermore, in eosinophilic crystalline pneumonia, a murine idiopathic type 2 lung inflammation, IL-2 was shown to function as an important activator of ILC2 functions; the authors further highlight a potential cross talk between ILC2 and ILC3 as well as T cells and T_reg during disease pathogenesis (100). Even more important, in a murine model for allergic asthma, pulmonary epithelial cell-derived TGF-β1 and IL-33 contributed to ILC2-mediated responses (101).

Besides their effects on acute pulmonary responses, ILC also impact the development of pulmonary fibrosis [for a recent review, refer to Ref. (102)]. In this context, an interplay between ILC, macrophages, and cells of the adaptive immune system was shown to participate—together with IL-13, IL-25, and IL-33—in the pathogenesis of BLM-induced pulmonary fibrosis in mice (103–105). Moreover, Hams et al. also found elevated levels of IL-25 and ILC2 in the lungs of patients with idiopathic pulmonary fibrosis (IPF) (105, 106), whereas others described a role for ILC2 and ILC3 in fibroblast activation providing a mechanistic link between ILC and fibrotic diseases (102).

Since work from our own laboratory and others implicate T_H17 cells and T_reg in the pathogenesis of radiation-induced lung disease, these observations strongly suggest that ILC, particularly ILC2 and ILC3, might also participate in the cross talk between damaged resident cells, recruited immune cells, and activated fibroblasts and thus modulate the extent of lung inflammation and progression of radiation-induced fibrosis. But, an involvement of ILC in radiation-induced pneumopathy remains to be demonstrated, and potential beneficial or disease-promoting effects during the different disease stages (inflammation, fibrosis) have to be explored.

Radiation therapy is an essential and common approach in cancer treatment. So far, the use of IR in cancer treatment is based on its high potential to induce tumor cell death and to abrogate survival of clonogenic tumor cells. The toxic effects of IR result from the deposition of energy from IR in tumor and normal tissue cells including immune cells. Energy deposition results in damage to cellular macromolecules, particularly cellular DNA, e.g., by direct breakage of chemical bonds within the DNA as well as by the generation of free radicals (107, 108). Among the diverse damaging effects of IR, the induction of DNA double strand breaks is considered as the most toxic lesion in cells.

The hematopoietic compartment is particularly sensitive to IR, for example, blood sample analysis revealed the rapid development of a hematopoietic syndrome in patients exposed to a total body irradiation of 1–2 Gray (Gy), which was characterized by a decline of the hematopoietic compartment (109). Similar to other hematopoietic cells, lymphocytes are particularly sensitive to radiation-induced cell death. Nevertheless, the various lymphocyte subtypes differ in their radiosensitivity: up to now it has been demonstrated that B cells, naive T cells, and NK cells are highly radiosensitive, whereas T memory cells, NK T cells, and T_reg cells are more resistant to the toxic effects of IR (110–114). Moreover, a study from 1995 revealed a higher radiosensitivity of IL-4-producing T_H2 cells compared to T_H1 cells (115). Based on their high radiosensitivity and easy accessibility, blood lymphocytes are frequently used in biodosimetry (116–119).

Further direct effects described in irradiated lymphocytes concentrate on the transcriptional response of these cells to IR, e.g., by gene expression profiling. These studies revealed that a majority of the strongly activated genes are p53 targets, like DNA damage-binding protein 2, the BCL-2-associated gene BAX, and tumor necrosis factor receptor superfamily, member 10b (TNFRSF10B) that are involved in DNA repair and apoptosis regulation (120–122).

However, in addition to these direct or “targeted” effects of IR on lymphocytes such as the induction of cellular stress responses and cell death in lymphocytes within the radiation field, lymphocytes can also mount an indirect response to radiation-induced tissue damage. In this context, “danger signals” released from damaged or dying cells in irradiated tissues result in lymphocyte activation, infiltration into the damaged tissue, and the release of inflammatory mediators (Figure 2).

Of note, high-dose irradiation also efficiently triggers direct tumor cell death and augments innate immune responses and tumor-specific immunity thereby enhancing the local and distant antitumor effects of RT. This was nicely summarized by one of the pioneers in the filed of radioimmunotherpy, Silvia Formenti (123). Accordingly, high numbers of tumor-infiltrating cytotoxic lymphocytes may predict the response of certain tumors to treatments involving RT (105, 124). This is a hot topic in the field of radiation biology and oncology highlighted in other reviews; here, we will concentrate on the contribution of radiation-induced modulation of the lymphocyte compartment to the adverse late effects of IR in the lung.

As nicely highlighted in a recent review on the general effects of IR on T lymphocytes and normal tissue responses, our understanding of the interaction between lymphocytes and radiation-induced tissue damage is still rudimentary (125). Because of the complexity of the involved cellular systems and soluble factors, investigations about the mechanisms underlying radiation-induced adverse late effects, for example, in the lung can only be performed in patients as well as animal models in vivo, particularly rodent models.

There is increasing evidence that lymphocytes play a role in RT-induced adverse late effects in the lung. Thus, these cells or the mediators associated with their pro- or anti-fibrotic function may constitute valuable targets for the prevention or treatment of radiation-induced lung disease. However, so far there is only little knowledge about the use of specific lymphocyte subpopulations or their mediators as potential diagnostic or predictive biomarkers for early or late adverse effects of IR in the lung. Consequently, so far treatment strategies targeting immune cells or associated mediators suspected to participate in disease pathogenesis are only tested in preclinical investigations in mice.

Instead, current treatment options for patients suffering from radiation-induced pneumonitis are limited to the symptomatic administration of anti-inflammatory drugs such as glucocorticoids thought to limit the toxic effects of the overwhelming inflammation by reducing the levels of pro-inflammatory cytokines, chemokines, and growth factors. But, these anti-inflammatory therapies may also indirectly impact on lymphocyte responses, their activation state, or both.

Up to now, mechanistic knowledge about the role of lymphocytes in radiation-induced pneumopathy is still limited, and no reliable diagnostic or predictive biomarkers for beneficial and adverse effects of specific lymphocyte subsets are available to date. Nevertheless, preclinical and clinical investigations indicate that radiation-induced immune changes are important to the outcome of RT and that lymphocytes contribute to the beneficial and adverse effects of IR in tumors and normal tissues, including the lung.

We hypothesize that radiation-induced acute damage to resident cells including progenitor cells of mesenchymal origin and a perpetual cascade of cytokines/chemokines triggers immune cell recruitment and activation to promote tissue repair. However, in cases where the immune response cannot be controlled by anti-inflammatory cells such as T_reg or M2-like macrophages, patients may develop pneumonitis. If this initial damage response is not sufficient to repair the radiation-induced damage, the persistent damage results in chronic inflammation and delayed changes of resident cells, such as epithelial-mesenchymal-transition, endothelial-mesenchymal-transition, activation of MSC, or even chronic (mitotic) cell death of endothelial cells and alveolar epithelial cells. The resulting delayed environmental changes involve among others tissue hypoxia (by chronic endothelial cell loss), chronic inflammation, and chronic increase in fibroblasts and fibrosis-promoting mediators such as adenosine, hyaluronic acid, and TGFβ. These changes act together in the generation/activation of disease-promoting cell phenotypes such as activated myofibroblasts, pathologic T_reg, or M2-like macrophages thereby promoting exaggerated ECM deposition and fibrosis development (Figure 2).

Here, we want to stress that radiation-induced immune changes can be either pro-inflammatory (acute phase) or anti-inflammatory/pro-fibrotic (chronic phase), and that lymphocytes exert distinct functions during radiation-induced pneumopathy (see Figure 1). We speculate that T_reg might be beneficial during radiation-induced pneumonitis due to their suppressive action on pro-inflammatory cells as seen in hypersensitivity pneumonitis (266). In contrast, as outlined above current preclinical studies in diverse murine fibrosis models highlight a potential contribution of T_reg in fibrosis development, and own work supports such a disease-promoting role also for radiation-induced pulmonary fibrosis (35). We therefore assume that these immunosuppressive cells or the environmental factors promoting their recruitment/expansion may constitute promising therapeutic targets to prevent or treat radiation-induced fibrosis. We are aware that the use of specific lymphocyte populations or associated signaling molecules as therapeutic targets is complicated by the fact these cells exert either beneficial or harmful roles. This depends on the disease stage, and potentially patient-specific genetic factors. Consequently this requires the careful definition of optimal treatment schedules to avoid immune cell-associated complications during both, pneumonitis and fibrosis. We also want to point out that recently discovered lymphocyte subsets like ILC as well as signaling pathways like IL-33/ST2 are of interest due to their potential contribution to RT-induced pneumopathy highlighting the need of related studies.

Further important issues that also need to be addressed are (i) the potential high intrinsic radioresistance of thorax-associated solid tumors treated by thoracic RT and (ii) a potential tumor immune escape. Therefore, it is important to consider that any inflammation-modulating or immune cell-targeting strategy for the treatment of radiation-induced pneumopathy may alter the antitumor effect of RT or combined treatment strategies involving immunomodulation or immunoboost; unfortunately, this issue is mostly not addressed by preclinical studies investigating the mechanisms of radiation-induced normal tissue toxicity. Vice versa, a potential increased normal tissue toxicity of such antitumor treatments is not analyzed when studying new combination treatments in preclinical models. Therefore, there is a high need to develop appropriate preclinical models if we want to identify treatment strategies balancing radiation-induced tumor cell clearance and normal tissue protection. Nevertheless, we are convinced that a better understanding of radiation-induced immunomodulation in tumors and normal tissues will offer novel opportunities for widening the therapeutic window by targeting immune cells or immune-associated mediators that promote both, tumor growth/resistance and normal tissue toxicity—of these, TGF-β and adenosine constitute perfect examples.

Taken together, it is important to deepen our knowledge about radiation-induced immune changes, including the modulation of lymphocyte recruitment, proliferation, and/or function of specific lymphocyte subsets during the different stages of radiation-induced lung disease. Further studies are needed to optimize therapeutic strategies for the prevention or treatment of adverse late effects of IR to normal tissues that also take into account the immune repertoire of the respective malignant disease before and after RT, if we aim at protecting the normal tissue without promoting tumor growth and vice versa.

FW and VJ equally contributed to the writing of this review article.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.