The effect of durvalumab consolidation after definitive radiochemotherapy for non-operable stage III non-small cell lung cancer on the dose effect relation for therapy related pulmonary infiltrates as a risk factor for pneumonitis

Introduction

Background

Consolidation therapy with the anti-programmed death-ligand 1 (PD-L1) antibody durvalumab or other immune checkpoint inhibitors is associated with an improvement in progression-free and overall survival in stage III non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC) according to randomised studies (1-3).

Rationale and knowledge gap

According to these randomised studies, the effect of anti-PD-L1 antibody consolidation therapy after definitive radiochemotherapy in NSCLC and SCLC stage III on the risk of treatment related pneumonitis of grade ≤2 is low. Here, either no (2) or no more than 6% increased pneumonitis risks of grade 2 were found after consolidation compared to placebo (3,4). For example, the risk of grade 2 pneumonitis in the ADRIATIC trial was 9.5% after durvalumab consolidation compared to 5.3% in the placebo group (3). In the Pacific trial, the corresponding risks were 15.2% and 9.4% (4). However, in retrospective real life studies on durvalumab consolidation after definitive radiochemotherapy higher rates of grade ≥2 pneumonitis were found. In some cases, it reached up to 50%, on average 23.8% (5).

Several studies analysed the development of lung density changes in follow-up computed tomography (CT) examinations during and after therapy depending on the superimposed dose distribution in the planning CT. Watanabe et al. analysed the dose volume histograms (DVHs) for manually contoured pneumonitis infiltrates (6). The authors found pneumonitis infiltrates at lower mean infiltrate doses for patients with grade 2 than with grade 1 pneumonitis (6). Others investigated the density increases per voxel as a function of the applied cumulative radiation dose and found significant dose effect relationships (7-10). Dependencies on age, smoking status, treated volume and prior thoracic surgery were found (7,9,10). In further studies, the probability of a density change in the lung of more than 80 Hounsfield units (HU) or more than 0.197 g/cm³ was investigated and significant dose dependencies were found (11-14), which were described with the probit model (12-14). Diot et al. found a higher radiation responsiveness after hypofractionation (higher doses of radiation delivered in fewer treatment sessions) than after conventional fractionation (13). Lee et al. found a higher radiosensitivity in lower lobe tumours (14). However, none of the studies examined patient-specific dose-effect relationships for density increase per voxel or within dose bins.

Objective

The main questions of this study are whether durvalumab consolidation alters the dose-response relationship for partial lung infiltrate volumes across different dose ranges. The dose-response relationship describes how the severity or likelihood of a biological effect changes in response to varying radiation doses. In addition, the study will investigate how dose metrics influence the development of lung infiltrates. In addition, the functional form and inter-individual variability of the dose-response relationships will be characterized and their association with the development of clinically apparent grade II pneumonitis will be investigated. We present this article in accordance with the TRIPOD reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2024-1284/rc).

Methods

In order to identify potential predictive factors and a durvalumab effect for the development of therapy associated lung infiltrates, this study examines the dose effect relationships of therapy associated lung infiltrates per dose bin. All treated patients gave their consent to the treatment taking part in the prospective, institutional clinical registry trial (No. 18-8364-BO). The study was conducted in accordance with the principles of the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of University Hospital Essen of University of Duisburg-Essen (No. 23-11186-BO).

Patient cohort

All consecutive patients with inoperable stage III NSCLC treated at a comprehensive cancer center with definitive radiochemotherapy, with and without durvalumab consolidation from a prospective registry are included in the analysis. Patients were treated with volumetric modulated arc therapy (VMAT) or multifield intensity modulated radiotherapy (IMRT) to achieve a conformal dose distribution around the planning target volume (PTV) with good lung sparing. Patients were treated in inspiratory breath hold or gated in exhale during free breathing (15). Treatment was delivered during deep inspiratory breath hold as the preferred method (15). Patients who could not reproduce this manoeuvre precisely were treated during free breathing and radiotherapy was delivered in exhale using a real-time position monitoring (RPM) system (Varian Medical Systems, Palo Alto, CA, USA). The RPM system allows to continuously monitor the respiratory motion and automatically gate the beam in the exhale phase. Daily image guidance was performed on a 6 degrees-of freedom table on a True Beam linear accelerator (Varian, CA, USA). Dose distributions within the lungs were calculated with a clinically used analytical anisotropic algorithm (AAA) with heterogeneity correction and for comparison in 20 randomly selected patients with an Acuros XB algorithm for accurate modelling of radiation transport and dose deposition in heterogeneous media. Absolute dose differences in the lung outside the PTV never exceeded +3.2 or −2.1 Gy in areas of the lung that make up more than 1% of the total lung. On average, the dose differences were −1.31 and +1.53 Gy.

Pulmonary infiltrates due to radiation pneumonitis were contoured in both lungs at the time of maximal expression by a consensus of two expert chest radiologists and radiation oncologists using manual delineation to exclude preexisting fibroses. Additionally, semi-automated conventional contouring was applied such as image thresholding function within certain density thresholds (16). The reviewers were blinded to clinical outcome. All types of pulmonary infiltrates characteristic for fibrotic and non-fibrotic hypersensitivity pneumonitis, occurring during follow-up after therapy were contoured. CT abnormalities indicative for parenchymal infiltrations as areas of differing attenuation, ground glass opacities (GGOs) and mosaic attenuation, at later time points irregular linear opacities, course reticulations and honeycombing indicative for fibrosis were contoured (17). According to Linda et al. 2011 (18), the pneumonitis signs were classified in four pattern, patchy GGOs, patchy consolidation and GGOs, diffuse GGOs as well as diffuse consolidations. These abnormalities were delineated without considering the applied dose distribution for both patients with and without durvalumab consolidation. At a later step, the considered dose response curves for pulmonary infiltrates will be considered in the low and high dose region with or without durvalumab separately (19). Time points during follow-up, at which additional pulmonary infection were diagnosed, were excluded.

Clinical pneumonitis was graded according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0.

Statistical analysis, CT parameters, normalisation, slice thickness, voxel in planning CT

Contoured lung infiltrates outside the PTV were normalized by the total lung volume outside the PTV resulting in the partial volume of lung infiltrates. The partial volume of lung infiltrates per dose step were obtained from the DVHs. DVHs of the lung outside the PTV containing the lung infiltrates and the contoured lung infiltrates outside the PTV were merged in 0.1 Gy dose steps. The infiltrate volume per dose step was normalised by the lung volume per dose step. These data steps were performed in SAS statistical software system (SAS/STAT 15.1, SAS institute Inc., Cary, NC, USA). Each partial infiltrate volume was weighted based on the proportion of lung volume at a specific dose step compared to the total lung volume outside the PTV at that dose. Dose steps (D) of 0.1 Gy were binned to 5 Gy dose bins from x < D < x+5 Gy, for x = 0, 5, 10, (…) 50 Gy as integer multiples of five Gray. A mixed fixed and random effect linear model was used to explain the weighted partial infiltrate volume per dose step data of the patients by the variables dose bin and durvalumab consolidation (yes/no) and a crossed interaction effect between dose bin and durvalumab consolidation as fixed effects. As a random subject effect, the crossed effect between patient and dose bin was introduced, assuming complete independence across subjects. This subject effects models random intercepts for each patient x dose bin combination (Proc Mixed, SAS). This model is termed dose step model. An F-test was used to test whether the random effects have zero variance (20).

In addition, the influence of co-factors related to the total lung radiation dose exposure was examined. The steepness of the dose response of partial infiltrate volumes per 5 Gy dose bins was analysed in dependence of mean lung dose, the partial lung volume that receives at least 5 Gy (V5) or 20 Gy (V20). To do this, the dose steps as well as the cross products of the dose steps and the co-factors were introduced as fixed effects into the model. To relax the assumption of a normal distribution of the random effect parameters, the partial infiltrate volumes per dose step from the different patients were ranked per 5 Gy dose bin and ranks were normalized by the number of patients +1 (21). The same mixed model was than adapted to the rank scores. The same mixed model was than adapted to the normal scores. The predicted mean values per patient and dose bin could be back-transformed to partial volumes of lung infiltrates per dose bin using penalized B splines (Proc Transreg, SAS). A linear quadratic model was applied to the dose data of pulmonary infiltrates, with random intercepts and random coefficients of the linear and quadratic dose terms. To test the effect of durvalumab consolidation, its impact on the intercept and dose-related terms could be added as fixed effects, assumed to be constant for all patients in the study. Heat maps were generated with the procedure SGPLOT from SAS. Mean partial infiltrate volume per dose bin estimates, resulting from the stepwise model, were hierarchially clustered according to a weighted pair group method using centroids (Proc CLUSTER, SAS). Estimated dose response curves parameters from the different patients were ordered according to successive generations of the cluster history. They were joined according to the minimum decrease in the proportion of variance explained by the clusters resulting from joining two clusters. The Akaike information criterion (AIC) was used to compare different dose response models. A difference between the AIC value of a candidate model minus the AIC value of the best model with the smallest AIC value larger than 10 is used to reject the candidate model (22).

Time to pneumonitis grade 2 was used as a clinical end-point. Proportional hazards analysis was performed using dosimetric data, durvalumab consolidation (yes/no) and the dose response data for partial infiltrate volumes per 5 Gy dose bin as explanatory data (Proc PHREG, SAS). Follow-up times of patients without pneumonitis grade 2 were censored at the time of the last available chest CT study during follow-up. All patients with pneumonitis grade 2 had a chest CT at that time. Leave one out cross validation was performed with a SAS macro as described by Rushing et al. (23). This macro is available for download at https://sites.google.com/view/herbpang/sassurvloocv (last assessed 17 December 2024). At each cross validation iteration, the leave-out tests observation was separated according to the median linear predictor from the trainings data set in the high or low risk group.

For sample size determination, the incidence of pneumonitis grade 2 after concurrent radiochemotherapy followed by durvalumab consolidation was estimated to be 35% and after concurrent radiochemotherapy alone 15% (5). The stepwise dose-effect relation at 11 dose-bins had 11 parameters. At least one parameter of the dose-effect relation with pneumonitis grade 2 should be detectable, with an associated hazard ratio between the high and low risk groups as the most important clinical factor identified, i.e., interstitial lung disease. The hazard ratio associated with idiopathic pulmonary fibrosis from the proportional hazards model is about 4 (24). The false discovery rate was controlled at a level of 0.05 (25). In addition, the proportions of patients without durvalumab, compared to those with durvalumab, was selected to be 40:60. The total number of patients needed was estimated to be 90 under these conditions using a one sided log-rank test for comparison of the high and low risks group at a power of 0.80 assuming exponential freedom from pneumonitis curves, a total pneumonitis incidence of 21% over the high- and low-risk groups at 1 year follow-up and considering only one year of follow-up (Proc power, SAS). To allow for incomplete follow-up during the first year, in up to 15% of patients, 100 patients were the target of this study.

Results

Follow-up CT studies of 104 consecutive patients with inoperable stage III NSCLC after definitive radiochemotherapy were analysed. Sixty patients received durvalumab consolidation, 44 patients did not. Patients with durvalumab consolidation were treated between April 2018 and March 2023, and patients without durvalumab consolidation between August 2017 and September 2018, i.e., before durvalumab approval. Only 3 patients started with durvalumab consolidation before durvalumab approval by an early access program and therefore, an effect of patient selection for durvalumab on the group of patients treated with radiochemotherapy alone was tried to be minimized. Patient characteristics are shown in Table 1. The distribution of total radiation doses, received corresponds to that in the Pacific trial (26). Total radiation dose of 50 to <54 Gy, ≥54 to ≤66 Gy and >66 to 72 Gy were received by 2, 95, and 7 patients. No patient received hypofractionation. The reason for a smaller dose was that the lung tolerance was reached, which was assumed to be 20 Gy. The reason for a total dose of 50 to <54 Gy was that accepted lung tolerance was reached, which was assumed to be a mean total lung dose of 20 Gy or a 33% relative volume of both lungs receiving >20 Gy (26). Seven patients who were treated with radiochemotherapy alone without durvalumab consolidation received a total radiation dose between 70 to 72 Gy.

Figure 1 shows the dose response relationships of the partial infiltrate volumes outside the PTV generated as described under materials and methods. As the dose response relationships were complex and patient dependent, a flexible linear mixed fixed and random effect model was used with a random intercept at each 5 Gy dose bin for each patient not assuming a specific functional form of the dose response relationship. This flexible step-wise dose response model described the partial infiltrate volume data considerably better than a linear quadratic model (Figure S1A). The difference between AIC values of the linear quadratic minus the step-wise model is 0.48804 so that the linear quadratic model was rejected as less likely. Nevertheless, the dose response curves for partial infiltrate volumes per dose step showed a good approximation with the linear quadratic dose response (Figure S1B).

Figure 1 Step risk model. Dose partial volumes of pulmonary infiltrates per 0.1 Gy dose step curves from 104 patients with pneumonitis of grade 0–2, as obtained from the differential dose volume histograms for total lung and lung infiltrates both outside the planning target volume for radiotherapy. A step-wise linear mixed model was approximated to these data per 5 Gy dose bin. The fitted step functions were given in the same colour per patient as the pulmonary infiltrates per 0.1 Gy dose step from the dose volume histograms.

There was as significant systematic dependence of the partial infiltrate volume on the respective 5 Gy dose bin as a systematic fixed effect of the step model that applies to all patient and dose bin random effects (P<0.001, F-test, Figure 2). The dependence of the partial infiltrate volumes on the different dose bins as a fixed effect were monotonously increasing with the dose bin number (Table 2). In addition, there was a significant durvalumab effect on the dose response relation of the 104 patients (Figure 2, P<0.001, F-test) that shows a significant interaction with the dose bins (P=0.003, F-test). The main durvalumab effect was present in both patients with subclinical pneumonitis of grade 0/1 (P<0.001, F-test, Figure S1C), and in patients with pneumonitis of grade 2 (P<0.001, F-test, Figure S1D). The interaction effect between durvalumab consolidation and the 5 Gy dose bin was significant for patients with pneumonitis of grade 0/1 (P=0.009, F-test) and there was a trend for patients with pneumonitis of grade 2 (P=0.050, F-test), respectively.

Figure 2 Predicted partial volumes of lung infiltrates per 5 Gy dose bin by the stepwise mixed model in dependence of durvalumab consolidation (yes, blue circles; no, red triangles). The durvalumab effect became significant at P<0.0001 as well as an interaction effect of durvalumab with dose bin (P=0.0036, F-test). Blue circles indicate data from patients with durvalumab consolidation and red triangles from patients without durvalumb.

In addition, we analysed dosimetric factors V5, V20, and mean lung dose as dose modifying factors on the dose response relation controlling for patient and 5 Gy dose bin. There was a significant dose response within the different dose bins characterized by a slope of 0.0602%±0.0018% partial infiltrate volume per Gray (P<0.001, F-test). V5 was the most important factor that modified this slope. The delta of the AIC values decreased for a model with V5 as a dose modifying factor by a delta =−289 in comparison to a model without V5, indicating that the V5 improves the dose effect model. The change in the slope of the dose response relation per 10% of V5 increase was estimated to 0.0223%±0.0013% partial infiltrate volume per Gray. The delta AIC for models with V20 or mean lung dose as dose modifying factors in comparisons to those without the respective parameter were −57 or −34 indicating a smaller improvement of these models.

There was marked heterogeneity in the partial lung infiltrate volumes per 5 Gy dose bin from patient to patient. The hypothesis that the variance components of the random effects are zero could be rejected (P<0.001, F-test). The interindividual heterogeneity of partial infiltrate volumes per 5 Gy dose bin accounted for 71% of the total sum of squared variances of partial infiltrate volumes per dose level, while all systematic fixed effects, the step-wise dose-response relation per dose bin, the durvalumab effect per dose bin and the V5 effect accounted for 27%. Of this 27%, 3.7% was attributable to the durvalumab effect.

Figure 3 shows a heat map of the proportion of pulmonary infiltrates for the different 5 Gy dose bins from 0 to 55 Gy of the different patients ordered according to successive generations of the cluster history with the respective dendrogram on the left side of the heat map. The 17 patients with a pneumonitis grade 2 had significantly higher generation numbers than the patients with pneumonitis of grade 0/1 (P<0.001, Wilcoxon test).

Figure 3 Heatmap of partial infiltrate volumes per 5 Gy dose bin from the stepwise mixed model for the different patients order according to the history of cluster joining. patients were hierarchically clustered according to a weighted pair group method using centroids. The resulting dendrogram is displayed left from the heat map.

To relax the assumption of a normal distribution of the random effect parameters, the mixed model was adapted to the ranked partial infiltrate volumes per 5 Gy dose bin and normal scores were computed for the ranks. Figure S2 shows the rank estimates for the different patients and different 5 Gy dose bins. Again, there was a significant durvalumab main effect (P<0.001, F-test) towards higher partial infiltrate volume scores and an interaction effect of durvalumab consolidation with the dose-bins (P<0.001, F-test). The predicted normal scores per patient and dose bin were back-transformed to partial infiltrate volumes per dose bin using penalized B splines. The predicted normal scores were compared with estimates by the mixed linear model directly adapted to the partial infiltrate data. Figure S3 shows a close agreement of the estimates of the partial infiltrate volumes per 5 Gy dose bin by both methods. The Pearson correlation coefficient was r=0.9982 [95% confidence interval (CI): 0.9980–0.9984].

As a separate end-point, we analysed the delivered dose distributions of patients with and without durvalumab consolidation with the stepwise mixed model. There were no differences in the partial lung volumes exposed to doses within the different dose bins between patients with or without durvalumab consolidation (Figure S4).

In a further step, we analysed the times to pneumonitis grade 2 in dependence on parameters from the dose response relation for the proportion of infiltrates per dose bin, dosimetric factors of the treatment plans, as V5, V20, and mean lung dose, the mean HU of the pulmonary infiltrates as given in Table 1, and durvalumab consolidation (yes/no), by multivariable proportional hazards analysis. The most likely generalizable Cox model using leave-one-out cross-validation according to the AIC criterion was a one parameter model with the selected variable partial infiltrate volume in dose bin 35–40 Gy in 94% of the leave-one-out cycles as a characteristic of the intermediate dose range sensitivity leading to pulmonary infiltrates. The cross-validated times to pneumonitis grade 2 curves were clearly separated for the high and low risk groups (P<0.001, log rank test) (Figure 4). The hazard ratio was 8.4 (95% CI: 1.9–36.6). The regression coefficient for the partial infiltrate volume in the 35–40 Gy dose bin lung volume from the Cox model was 7.013+1.884 using the whole data set. Therefore, the hazard ratio for pneumonitis grade 2 increases by 7.013% per 1% increase in the partial infiltrate volume in that dose bin.

Figure 4 Cross validated Kaplan Meier time to event curves for the high and low risk group according to the best three parameter subset classifiers based on likelihood score statistic, selecting the partial infiltrate volume data form dose bins from 35–40 Gy, and the durvalumab consolidation therapy effect. There was a significant difference between the high risk and low risk group (P=0.0007, log rank test).

The median time to pneumonitis grade 2 was 148 days, and 80% of all pneumonitis grade 2 events were observed later than 183 days after start of treatment. Only three patients without pneumonitis grade 2 had a shorter follow-up time to the last available chest CT smaller than 183 days. Median follow-up time with available imaging studies of patients without pneumonitis grade 2 was 1,037 days. Therefore, follow-up was considered mature.

Discussion

Treatment related pneumonitis induced by thoracic radiochemotherapy in patients with locally advanced, unresectable stage III NSCLC remains a limiting factor in treatment efficacy and potentially a severe adverse and fatal event. Durvalumab consolidation has become a main pillar within multimodality treatment after radiochemotherapy. The Pacific trial investigated the effect of durvalumab after radiochemotherapy and underlined the potential of immunotherapy implementation associated with superior survival rates. However, radiation pneumonitis was the most frequent unfavorable event leading to discontinuation of the study regimen (6.3 % vs. 4.3%) (26). This potential overlapping pulmonary toxicity has become a challenge as both conventional radiochemotherapy and immunotherapy can contribute to the develpoment of pneumonitis. Thus, investigations of risk factors for identifying patients at risk for the development of treatment related pneumonitis, particularly of radiotherapy and immune related parenchymal lung changes, is of high importance.

Key findings

The primary aim of our study was to evaluate the durvalumab effect on the development of low grade pneumonitis (sublinical or pneumonitis grade ≤2) in patients with histologically proven NSCLC undergoing definitive radiochemotherapy and consolidation immune checkpoint inhibition therapy. One main finding of this study is that durvalumab increases the proportion of pulmonary infiltrates per dose bin in a dose dependent manner. Furthermore, durvalumab increases the steepness of the pulmonary infiltrate-dose curve in patients with subclinical or grade 2 pneumonitis. Interestingly, durvalumab increased the proportion of infiltrates in each pneumonitis group, subclinical, grade 1 and 2 pneumonitis. These findings suggest that durvalumab sensitizes the lung parenchyma to irradiation induced lung toxicity.

Comparison with similar researches

Likewise, these results support evidence of Shaverdian et al. (27) who found a higher rate of symptomatic pneumonitis in patients treated with durvalumab consolidation compared to patients treated with radiochemotherapy, only (18% vs. 9%; P=0.09). Furthermore, these results are in line with a systematic review and meta analysis from Geng et al. who found that a combination therapy of programmed death protein 1 (PD-1)/PD-L1 inhibition and radiotherapy increased the risk for mild (grade 1–2) pneumonitis (28).

For radiotherapy of lung cancer, there are accepted tolerability limits for lung exposure from the DVHs, such as for the V20 and the mean lung dose (29). However, in the different studies the association with the incidence of grade 2 pneumonitis is only moderate with large scatter in the observed pneumonitis rates per dose bin (30-33). Thus, in large studies, no consistent and clear dependencies were found in the frequency of symptomatic pneumonitis depending on the V20 between 20–35% or mean lung dose between 13.0–19.5 Gy (30,34,35), the range most commonly used in definitive radiochemotherapy of locally advanced lung cancer (36).

Predictive factors in exploratory analyses partly with best cut-off value selection, which lowers the significance of the stated significances (37) are V5 (38), mean lung dose (24), V20 (39-41), and V40 (42). However, these predictive factors leave a large portion of interindividual heterogeneity within pneumonitis patients unexplained.

In this study, pneumonitic infiltrates were contoured at maximum during follow-up and DVHs of the parenchymal changes were compared dose step-wise with DVHs of the lung. The resulting dose effect relationships were monotonically increasing with large inter-individual differences and revealed systematic effects of treatment related factors such as durvalumab consolidation or lung dose exposure. The durvalumab effect is not additive, but there is an interaction effect with the radiation dose, of 2% additional infiltrates at doses of 5–10 Gy, 7% additional infiltrates at doses of 35–40 Gy, and 9% at 45–50 Gy. The durvalumab effect on pulmonary infiltrates may also be shown for patients with subclinical pneumonitis and therefore detects combination effects in asymptomatic patients. In addition to the durvalumab effect, the dose response relation for pulmonary infiltrates was also dependend on the V5 of the delivered dose distribution in lung in this study. The lung V5 was shown to be associated with radiation induced lymphopenia during or up to 2 weeks after radiotherapy alone (43). Lymphopenia at the end of thoracic radiotherapy increases the risk of infections during the first 6 months of follow-up (44). It was also found that lymphopenia during or after up to 1 month after thoracic radiotherapy or radiochemotherapy for lung cancer was associated with an increased risk and severity of pneumonitis (45,46). Irradiation was shown to lead to an increased influx of lymphocytes into lung tissue on the one hand and lymphopenia in the peripheral blood and peripheral lymphoid tissues 2 to 3 months after irradition (45). Thereby, the risk of pulmonary infiltrates and sensitivity to checkpoint inhibitors might be influenced. However, these systematic effects explain only 27% of the total sum of squared deviations of the partial infiltrate volumes per dose step, while over 70% were due to unexplained inter-indiviudal heterogeneity. These inter-individual differences are the most likely reason for the low predictive value of the previously mentioned dosimetric treatment parameters and radiobiologic models.

A first step is to unravel the inter-individual differences, which was the subject of this work. Even if the differences revealed here using the dose effect relationships for partial volumes of pulmonary infiltrates are not detectable before the start of therapy, individual dose effect relationships represent an in vivo endpoint at which biological factors influencing the individual and therapy dependent variables can be tested. A number of potential factors influencing the effect of radiochemotherapy and durvalumab may affect the dose response relationship in partial pulmonary infiltrates.

Apart from treatment related risk factors several patient related risk factors for the development of pneumonitis have been decribed in literature. Patient related risk factors include higher age, interstitial lung abnormality scores (ILAS), pre-existing interstitiial lung disease (ILD), pre-existing pulmonary fibrosis of various origins, especially idiopathic pulmonary fibrosis (IPF), limited lung function before treatment with low forced expiratory volume in 1 second (FEV1) and low partial pressure of oxygen (PaO₂) (30,32,47-56). Patient dependent factors such as, smoking status, pulmonary function, and chronic obstructive lung disease are further factors mediating the development of pneumonitis (57-59). Mediators of grade 2 pneumonitis are also tumor location, particularly in the mid lower and basal lung segments, dose exposure of the heart or the pulmonary ventilation/perfusion distribution (59-64). Even more, with the advent of artificial intelligence (AI), a series of radiomic features of infiltrates are identified, which indicate a different pneumonitis risk per infiltrate volume, as well as the aforementioned clinical characteristics associated with pneumonitis risk (65,66). Factors influencing inter-individual risk for the development of pneumonitis are individual biological biomarkers. Chemokines and cytokine pathways in peripheral blood (66-68), T-regs and the immune cell landscape in the lung (69-71), altered transcriptome profiles in different cell types of the lung (72-74), and activation of CD73 in lung cells are clinical co variables with potential modulation of the course of pulmonary changes (75). Anscher et al. identified that the transforming growth factor β and interleukin 6 as candidates have an impact on the risk of radiation pneumonitis (76).

Further biologic factors determining the observed large interindividual differences in radiation dose response relation remain to be evaluated.

Overall, the dose effect relationships, which already capture the inter-individual differences, have a high predictive value for grade 2 pneumonitis. The expression of infiltrates was dominant in the intermediate dose range of 35–40 Gy. Most clinically significant grade 2 pneumonitis occurs after durvalumab and definitive radiochemotherapy (77). After cross validation, these were the most dominant factors in this study. The durvalumab effect was dominant on infiltrate expression, but in univariate analysis not significant. This shows that durvalumab moderates individual factors that are expressed in the infiltrate types, but these changes are subtle with regard to the inter-individual differences so that a direct visibility did not emerge in the present study results.

In the Pacific trial with 709 patients in the as treated analysis, no statistically significant differences were found in the crude incidence of grade 2 pneumonitis rate with 7.6% with and 4.3% without durvalumab consolidation (P=0.11, exact Fisher test) (77). Only meta-analyses of real world data showed a small effect of durvalumab on grade 2 pneumonitis incidence, which was demonstrated in this study as an effect on partial infiltrate volumes over the 5 Gy dose bins as a systematic effect (5). However, this systematic durvalumab effect of the dose effect relationships was small and explained only 3.7% of the total sum of squared of partial infiltrate volumes per dose bin.

The time interval between the end of radiotherapy and the start of durvalumab consolidation of <14 d appears to increase the risk of grade ≥2 pneumonitis (78). However, severe grade >3 pneumonitis was independent of the interval between the end of radiotherapy and the start of durvalumab consolidation at an interval of 21–41 and ≥42 days (79).

Strengths and limitations

A limitation of this study is the retrospective nature of our study design. Hence, limitations in patient specific records like documentation of symptoms and prescribed therapies in the ambulant setting after completion of radiotherapy might be incomplete and could potentially lead to a missclassification of patients. For mitigations of these limitations we screened all documents and records also from other oncologic departements that are part of University Hospital Essen and external documents that were stored within patient records. Besides, consequent guideline based CT follow-up can be negatively influenced by inadequate patient compliance but also patient morbidity. In our study long follow-up could be documented and therefore provide a mature data basis for our complex statistical analysis.

Explanations of findings

These data confirm that durvalumab sensitizes the lung parenchyma to the development of pneumonitis. These are more susceptible patients who need to be identified for inter-individual heterogeneity: however, both factors, dose regions and durvalumab, leave much of the inter-individual heterogeneity unexplained, especially the biological factors.

Conclusions

The results of this research indicate that dose effect relations for pulmonary infiltrates after radiochemotherapy are sensitive to detect factors systematically influencing radiation dose response, here durvalumab consolidation and delivered lung dose (V5). In conclusion, dose effect relationships for the partial infiltrate volumes reveal inter-individual heterogeneity of dose response and are in turn closely associated with pneumonitis risk. The durvalumab effect and lung radiation exposure are detectable systematic effects on the dose response of infiltrates.

Acknowledgments

We acknowledge the grant of the Deutsche Forschungsgemeinschaft (DFG) (grant No. GRK 2762/1).

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2024-1284/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2024-1284/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2024-1284/prf

Funding: This study was supported by a grant of the Deutsche Forschungsgemeinschaft (DFG) (grant No. GRK 2762/1 to V.J.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2024-1284/coif). T.G. reports payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Astra Zeneca, BMS, Eisai, Ipsen, MSD, and Roche; receives support for attending meetings and/or travel by Merck serono; participated on a Data Safety Monitoring Board or Advisory Board of APOGEPHA, Astra Zeneca, MSD, Eisai, and Ipsen; and has stock or stock options from Bayer AG. Martin Schuler reports funding for consultancy from Amgen, AstraZeneca, Blueprint Medicines, Bristol Myers Squibb, Gilead, GlaxoSmithKline, Immunocore, Johnson & Johnson, MSD, Novartis, Regeneron, Roche, and Sanofi; honoraria for CME lectures from Amgen, Bristol Myers Squibb, GlaxoSmithKline, Johnson & Johnson, MSD, and Roche; and research funding to institution from AstraZeneca, Bristol Myers Squibb, and Johnson & Johnson. B.M.S. reports research grants from Else Kröner-Fresenius Foundation, Deutsche Forschungsgemeinschaft, and PharmaCept GmbH. S.B. reports consulting fees from BD (Becton Dickinson), Livsmed, Lexington Medical, and Karl Storz; speaker fees from AstraZeneca, Bristol Myers Squibb, KLS Martin, Roche, and Johnson & Johnson; funding for studies to institution from Janssen, Bristol Myers Squibb, AstraZeneca, Achilles Therapeutics, and BD (Becton Dickinson). V.J. reports funding from Deutsche Forschungsgemeinschaft (No. GRK 2762/1). The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All treated patients gave their consent to the treatment taking part in the prospective, institutional clinical registry trial (No. 18-8364-BO). The study was conducted in accordance with the principles of the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics committee of University Hospital Essen of University of Duisburg-Essen (No. 23-11186-BO).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Antonia SJ, Villegas A, Daniel D, et al. Overall Survival with Durvalumab after Chemoradiotherapy in Stage III NSCLC. N Engl J Med 2018;379:2342-50. [Crossref] [PubMed]
2. Zhou Q, Chen M, Jiang O, et al. Sugemalimab versus placebo after concurrent or sequential chemoradiotherapy in patients with locally advanced, unresectable, stage III non-small-cell lung cancer in China (GEMSTONE-301): interim results of a randomised, double-blind, multicentre, phase 3 trial. Lancet Oncol 2022;23:209-19. [Crossref] [PubMed]
3. Cheng Y, Spigel DR, Cho BC, et al. Durvalumab after Chemoradiotherapy in Limited-Stage Small-Cell Lung Cancer. N Engl J Med 2024;391:1313-27. [Crossref] [PubMed]
4. Vansteenkiste JF, Naidoo J, Faivre-Finn C, et al. Symptomatic Pneumonitis With Durvalumab After Concurrent Chemoradiotherapy in Unresectable Stage III NSCLC. JTO Clin Res Rep 2024;5:100638. [Crossref] [PubMed]
5. Han C, Qiu J, Bai L, et al. Pneumonitis Risk After Chemoradiotherapy With and Without Immunotherapy in Patients With Locally Advanced Non-Small Cell Lung Cancer: A Systematic Review and Meta-Analysis. Int J Radiat Oncol Biol Phys 2024;119:1179-207. [Crossref] [PubMed]
6. Watanabe S, Ogino I, Shigenaga D, et al. Relationship Between Radiation Pneumonitis Following Definitive Radiotherapy for Non-small Cell Lung Cancer and Isodose Line. In Vivo 2021;35:3441-8. [Crossref] [PubMed]
7. Palma DA, van Sörnsen de Koste J, Verbakel WF, et al. Lung density changes after stereotactic radiotherapy: a quantitative analysis in 50 patients. Int J Radiat Oncol Biol Phys 2011;81:974-8. [Crossref] [PubMed]
8. Phernambucq EC, Palma DA, Vincent A, et al. Time and dose-related changes in radiological lung density after concurrent chemoradiotherapy for lung cancer. Lung Cancer 2011;74:451-6. [Crossref] [PubMed]
9. Ma J, Zhang J, Zhou S, et al. Regional lung density changes after radiation therapy for tumors in and around thorax. Int J Radiat Oncol Biol Phys 2010;76:116-22. [Crossref] [PubMed]
10. Theuws JC, Kwa SL, Wagenaar AC, et al. Dose-effect relations for early local pulmonary injury after irradiation for malignant lymphoma and breast cancer. Radiother Oncol 1998;48:33-43. [Crossref] [PubMed]
11. Rosen II, Fischer TA, Antolak JA, et al. Correlation between lung fibrosis and radiation therapy dose after concurrent radiation therapy and chemotherapy for limited small cell lung cancer. Radiology 2001;221:614-22. [Crossref] [PubMed]
12. Avanzo M, Barbiero S, Trovo M, et al. Voxel-by-voxel correlation between radiologically radiation induced lung injury and dose after image-guided, intensity modulated radiotherapy for lung tumors. Phys Med 2017;42:150-6. [Crossref] [PubMed]
13. Diot Q, Marks LB, Bentzen SM, et al. Comparison of radiation-induced normal lung tissue density changes for patients from multiple institutions receiving conventional or hypofractionated treatments. Int J Radiat Oncol Biol Phys 2014;89:626-32. [Crossref] [PubMed]
14. Lee S, Stroian G, Kopek N, et al. Analytical modelling of regional radiotherapy dose response of lung. Phys Med Biol 2012;57:3309-21. [Crossref] [PubMed]
15. Guberina M, Santiago A, Pöttgen C, et al. Respiration-controlled radiotherapy in lung cancer: Systematic evaluation of the optimal application practice. Clin Transl Radiat Oncol 2023;40:100628. [Crossref] [PubMed]
16. Hlouschek J, König B, Bos D, et al. Experimental Examination of Conventional, Semi-Automatic, and Automatic Volumetry Tools for Segmentation of Pulmonary Nodules in a Phantom Study. Diagnostics (Basel) 2023;14:28. [Crossref] [PubMed]
17. Raghu G, Remy-Jardin M, Ryerson CJ, et al. Diagnosis of Hypersensitivity Pneumonitis in Adults. An Official ATS/JRS/ALAT Clinical Practice Guideline. Am J Respir Crit Care Med 2020;202:e36-69. Erratum in: Am J Respir Crit Care Med 2021;203:150-1 Am J Respir Crit Care Med 2022;206:518. [Crossref] [PubMed]
18. Linda A, Trovo M, Bradley JD. Radiation injury of the lung after stereotactic body radiation therapy (SBRT) for lung cancer: a timeline and pattern of CT changes. Eur J Radiol 2011;79:147-54. [Crossref] [PubMed]
19. Smesseim I, Mets OM, Daniels JMA, et al. Diagnosis and management of pneumonitis following chemoradiotherapy and immunotherapy in stage III non-small cell lung cancer. Radiother Oncol 2024;194:110147. [Crossref] [PubMed]
20. Demidenko E, Sargent J, Onega T. Random effects coefficient of determination for mixed and meta-analysis models. Commun Stat Theory Methods 2012;41:953-69. [Crossref] [PubMed]
21. Conover WJ, Iman RL. Rank Transformations as a Bridge between Parametric and Nonparametric Statistics. The American Statistician 1981;35:124-9.
22. Burnham KP, Anderson DR. Multimodel Inference: Understanding AIC and BIC in Model Selection. Sociological Methods & Research 2004;33:261-304.
23. Rushing C, Bulusu A, Hurwitz HI, et al. A leave-one-out cross-validation SAS macro for the identification of markers associated with survival. Comput Biol Med 2015;57:123-9. [Crossref] [PubMed]
24. Altan M, Soto F, Xu T, et al. Pneumonitis After Concurrent Chemoradiation and Immune Checkpoint Inhibition in Patients with Locally Advanced Non-small Cell Lung Cancer. Clin Oncol (R Coll Radiol) 2023;35:630-9. [Crossref] [PubMed]
25. Benjamini Y, Yekutieli D. The control of the false discovery rate in multiple testing under dependency. Ann Statist 2001;29:1165-88.
26. Antonia SJ, Villegas A, Daniel D, et al. Durvalumab after Chemoradiotherapy in Stage III Non-Small-Cell Lung Cancer. N Engl J Med 2017;377:1919-29. [Crossref] [PubMed]
27. Shaverdian N, Thor M, Shepherd AF, et al. Radiation pneumonitis in lung cancer patients treated with chemoradiation plus durvalumab. Cancer Med 2020;9:4622-31. [Crossref] [PubMed]
28. Geng Y, Zhang Q, Feng S, et al. Safety and Efficacy of PD-1/PD-L1 inhibitors combined with radiotherapy in patients with non-small-cell lung cancer: a systematic review and meta-analysis. Cancer Med 2021;10:1222-39. [Crossref] [PubMed]
29. Marks LB, Bentzen SM, Deasy JO, et al. Radiation dose-volume effects in the lung. Int J Radiat Oncol Biol Phys 2010;76:S70-6. [Crossref] [PubMed]
30. Palma DA, Senan S, Tsujino K, et al. Predicting radiation pneumonitis after chemoradiation therapy for lung cancer: an international individual patient data meta-analysis. Int J Radiat Oncol Biol Phys 2013;85:444-50. [Crossref] [PubMed]
31. Semenenko VA, Li XA. Lyman-Kutcher-Burman NTCP model parameters for radiation pneumonitis and xerostomia based on combined analysis of published clinical data. Phys Med Biol 2008;53:737-55. [Crossref] [PubMed]
32. Bensenane R, Helfre S, Cao K, et al. Optimizing lung cancer radiation therapy: A systematic review of multifactorial risk assessment for radiation-induced lung toxicity. Cancer Treat Rev 2024;124:102684. [Crossref] [PubMed]
33. Zhang XJ, Sun JG, Sun J, et al. Prediction of radiation pneumonitis in lung cancer patients: a systematic review. J Cancer Res Clin Oncol 2012;138:2103-16. [Crossref] [PubMed]
34. Kwa SL, Lebesque JV, Theuws JC, et al. Radiation pneumonitis as a function of mean lung dose: an analysis of pooled data of 540 patients. Int J Radiat Oncol Biol Phys 1998;42:1-9. [Crossref] [PubMed]
35. Gao RW, Day CN, Yu NY, et al. Dosimetric predictors of pneumonitis in locally advanced non-small cell lung cancer patients treated with chemoradiation followed by durvalumab. Lung Cancer 2022;170:58-64. [Crossref] [PubMed]
36. Bradley JD, Paulus R, Komaki R, et al. Standard-dose versus high-dose conformal radiotherapy with concurrent and consolidation carboplatin plus paclitaxel with or without cetuximab for patients with stage IIIA or IIIB non-small-cell lung cancer (RTOG 0617): a randomised, two-by-two factorial phase 3 study. Lancet Oncol 2015;16:187-99.
37. Faraggi D, Simon R. A simulation study of cross-validation for selecting an optimal cutpoint in univariate survival analysis. Stat Med 1996;15:2203-13. [Crossref] [PubMed]
38. Tsukita Y, Yamamoto T, Mayahara H, et al. Intensity-modulated radiation therapy with concurrent chemotherapy followed by durvalumab for stage III non-small cell lung cancer: A multi-center retrospective study. Radiotherapy and Oncology 2021;160:266-72. [Crossref] [PubMed]
39. Hirama N, Yamamoto M, Nagaoka S, et al. Predictors of lung injury during durvalumab maintenance therapy following concurrent chemoradiotherapy in unresectable locally advanced non-small cell lung carcinoma. Thorac Cancer 2023;14:2601-7. [Crossref] [PubMed]
40. Oshiro Y, Mizumoto M, Sekino Y, et al. Risk factor of pneumonitis on dose-volume relationship for chemoradiotherapy with durvalumab: Multi-institutional research in Japan. Clin Transl Radiat Oncol 2021;29:54-9. [Crossref] [PubMed]
41. Saito G, Oya Y, Taniguchi Y, et al. Real-world survey of pneumonitis and its impact on durvalumab consolidation therapy in patients with non-small cell lung cancer who received chemoradiotherapy after durvalumab approval (HOPE-005/CRIMSON). Lung Cancer 2021;161:86-93. [Crossref] [PubMed]
42. Mayahara H, Uehara K, Harada A, et al. Predicting factors of symptomatic radiation pneumonitis induced by durvalumab following concurrent chemoradiotherapy in locally advanced non-small cell lung cancer. Radiat Oncol 2022;17:7. [Crossref] [PubMed]
43. Kuncman Ł, Pajdziński M, Smółka K, et al. Early lymphocyte levels and low doses radiation exposure of lung predict lymphopenia in radiotherapy for lung cancer. Front Immunol 2024;15:1426635. [Crossref] [PubMed]
44. Terrones-Campos C, Ledergerber B, Forbes N, et al. Prediction of Radiation-induced Lymphopenia following Exposure of the Thoracic Region and Associated Risk of Infections and Mortality. Clin Oncol (R Coll Radiol) 2023;35:e434-44. [Crossref] [PubMed]
45. Zhou P, Chen L, Yan D, et al. Early variations in lymphocytes and T lymphocyte subsets are associated with radiation pneumonitis in lung cancer patients and experimental mice received thoracic irradiation. Cancer Med 2020;9:3437-44. [Crossref] [PubMed]
46. Friedes C, Iocolano M, Lee SH, et al. The effective radiation dose to immune cells predicts lymphopenia and inferior cancer control in locally advanced NSCLC. Radiother Oncol 2024;190:110030. [Crossref] [PubMed]
47. Tsujino K, Hashimoto T, Shimada T, et al. Combined analysis of V20, VS5, pulmonary fibrosis score on baseline computed tomography, and patient age improves prediction of severe radiation pneumonitis after concurrent chemoradiotherapy for locally advanced non-small-cell lung cancer. J Thorac Oncol 2014;9:983-90. [Crossref] [PubMed]
48. Kashihara T, Nakayama Y, Ito K, et al. Usefulness of Simple Original Interstitial Lung Abnormality Scores for Predicting Radiation Pneumonitis Requiring Steroidal Treatment After Definitive Radiation Therapy for Patients With Locally Advanced Non-Small Cell Lung Cancer. Adv Radiat Oncol 2021;6:100606. [Crossref] [PubMed]
49. Yamaguchi T, Shimizu J, Hasegawa T, et al. Pre-existing pulmonary fibrosis is a risk factor for anti-PD-1-related pneumonitis in patients with non-small cell lung cancer: A retrospective analysis. Lung Cancer 2018;125:212-7. [Crossref] [PubMed]
50. Yamaguchi S, Ohguri T, Matsuki Y, et al. Radiotherapy for thoracic tumors: association between subclinical interstitial lung disease and fatal radiation pneumonitis. Int J Clin Oncol 2015;20:45-52. [Crossref] [PubMed]
51. Ozawa Y, Abe T, Omae M, et al. Impact of Preexisting Interstitial Lung Disease on Acute, Extensive Radiation Pneumonitis: Retrospective Analysis of Patients with Lung Cancer. PLoS One 2015;10:e0140437. [Crossref] [PubMed]
52. Goodman CD, Nijman SFM, Senan S, et al. A Primer on Interstitial Lung Disease and Thoracic Radiation. J Thorac Oncol 2020;15:902-13. [Crossref] [PubMed]
53. Kashihara T, Nakayama Y, Okuma K, et al. Impact of interstitial lung abnormality on survival after adjuvant durvalumab with chemoradiotherapy for locally advanced non-small cell lung cancer. Radiother Oncol 2023;180:109454. [Crossref] [PubMed]
54. Lee YH, Kim YS, Lee SN, et al. Interstitial Lung Change in Pre-radiation Therapy Computed Tomography Is a Risk Factor for Severe Radiation Pneumonitis. Cancer Res Treat 2015;47:676-86. [Crossref] [PubMed]
55. Kim H, Hwang J, Kim SM, et al. Risk factor analysis of the development of severe radiation pneumonitis in patients with non-small cell lung cancer treated with curative radiotherapy, with focus on underlying pulmonary disease. BMC Cancer 2023;23:992. [Crossref] [PubMed]
56. Inoue A, Kunitoh H, Sekine I, et al. Radiation pneumonitis in lung cancer patients: a retrospective study of risk factors and the long-term prognosis. Int J Radiat Oncol Biol Phys 2001;49:649-55. [Crossref] [PubMed]
57. Kong FM, Wang S. Nondosimetric risk factors for radiation-induced lung toxicity. Semin Radiat Oncol 2015;25:100-9. [Crossref] [PubMed]
58. Chen F, Niu J, Wang M, et al. Re-evaluating the risk factors for radiation pneumonitis in the era of immunotherapy. J Transl Med 2023;21:368. [Crossref] [PubMed]
59. Vogelius IR, Bentzen SM. A literature-based meta-analysis of clinical risk factors for development of radiation induced pneumonitis. Acta Oncol 2012;51:975-83. [Crossref] [PubMed]
60. Palma G, Monti S, Pacelli R, et al. Radiation Pneumonitis in Thoracic Cancer Patients: Multi-Center Voxel-Based Analysis. Cancers (Basel) 2021;13:3553. [Crossref] [PubMed]
61. Defraene G, van Elmpt W, Crijns W, et al. Regional variability in radiation-induced lung damage can be predicted by baseline CT numbers. Radiother Oncol 2017;122:300-6. [Crossref] [PubMed]
62. Farr KP, Kallehauge JF, Møller DS, et al. Inclusion of functional information from perfusion SPECT improves predictive value of dose-volume parameters in lung toxicity outcome after radiotherapy for non-small cell lung cancer: A prospective study. Radiother Oncol 2015;117:9-16. [Crossref] [PubMed]
63. Baschnagel AM, Flakus MJ, Wallat EM, et al. A Phase 2 Randomized Clinical Trial Evaluating 4-Dimensional Computed Tomography Ventilation-Based Functional Lung Avoidance Radiation Therapy for Non-Small Cell Lung Cancer. Int J Radiat Oncol Biol Phys 2024;119:1393-402. [Crossref] [PubMed]
64. Midroni J, Salunkhe R, Liu Z, et al. Incorporation of Functional Lung Imaging Into Radiation Therapy Planning in Patients With Lung Cancer: A Systematic Review and Meta-Analysis. Int J Radiat Oncol Biol Phys 2024;120:370-408. [Crossref] [PubMed]
65. Zhang Z, Wang Z, Yan M, et al. Radiomics and Dosiomics Signature From Whole Lung Predicts Radiation Pneumonitis: A Model Development Study With Prospective External Validation and Decision-curve Analysis. Int J Radiat Oncol Biol Phys 2023;115:746-58. [Crossref] [PubMed]
66. Liu X, Shao C, Fu J. Promising Biomarkers of Radiation-Induced Lung Injury: A Review. Biomedicines 2021;9:1181. [Crossref] [PubMed]
67. Moore ZR, Huang X, Lobaugh S, et al. Biomarkers associated with pulmonary exacerbations in a randomized trial of nintedanib for radiation pneumonitis. Radiother Oncol 2024;196:110320. [Crossref] [PubMed]
68. Wang S, Campbell J, Stenmark MH, et al. Plasma Levels of IL-8 and TGF-β1 Predict Radiation-Induced Lung Toxicity in Non-Small Cell Lung Cancer: A Validation Study. Int J Radiat Oncol Biol Phys 2017;98:615-21. [Crossref] [PubMed]
69. Suresh K, Naidoo J, Zhong Q, et al. The alveolar immune cell landscape is dysregulated in checkpoint inhibitor pneumonitis. J Clin Invest 2019;129:4305-15. [Crossref] [PubMed]
70. Guo T, Zou L, Ni J, et al. Regulatory T Cells: An Emerging Player in Radiation-Induced Lung Injury. Front Immunol 2020;11:1769. [Crossref] [PubMed]
71. Läubli H, Koelzer VH, Matter MS, et al. The T cell repertoire in tumors overlaps with pulmonary inflammatory lesions in patients treated with checkpoint inhibitors. Oncoimmunology 2018;7:e1386362. [Crossref] [PubMed]
72. Melms JC, Biermann J, Huang H, et al. A molecular single-cell lung atlas of lethal COVID-19. Nature 2021;595:114-9. [Crossref] [PubMed]
73. Curras-Alonso S, Soulier J, Defard T, et al. An interactive murine single-cell atlas of the lung responses to radiation injury. Nat Commun 2023;14:2445. [Crossref] [PubMed]
74. Zhang M, Lan H, Jiang M, et al. NLRP3 inflammasome mediates pyroptosis of alveolar macrophages to induce radiation lung injury. J Hazard Mater 2025;484:136740. [Crossref] [PubMed]
75. Wirsdörfer F, de Leve S, Cappuccini F, et al. Extracellular Adenosine Production by ecto-5'-Nucleotidase (CD73) Enhances Radiation-Induced Lung Fibrosis. Cancer Res 2016;76:3045-56. [Crossref] [PubMed]
76. Anscher MS, Kong FM, Andrews K, et al. Plasma transforming growth factor beta1 as a predictor of radiation pneumonitis. Int J Radiat Oncol Biol Phys 1998;41:1029-35. [Crossref] [PubMed]
77. Naidoo J, Vansteenkiste JF, Faivre-Finn C, et al. Characterizing immune-mediated adverse events with durvalumab in patients with unresectable stage III NSCLC: A post-hoc analysis of the PACIFIC trial. Lung Cancer 2022;166:84-93. [Crossref] [PubMed]
78. Harada D, Shimonishi A, Saeki K, et al. Early administration of durvalumab after chemoradiotherapy increased risk of pneumonitis in patients with locally advanced non-small cell lung cancer. Asia Pac J Clin Oncol 2023;19:e111-7. [Crossref] [PubMed]
79. Yang Z, Zhong W, Luo Y, et al. The timing of durvalumab administration affects the risk of pneumonitis in patients with locally advanced non-small cell lung cancer: a systematic review and meta-analysis. BMC Cancer 2023;23:962. [Crossref] [PubMed]