Pre-Existing Interstitial Lung Abnormalities Are Independent Risk Factors for Interstitial Lung Disease during Durvalumab Treatment after Chemoradiotherapy in Patients with Locally Advanced Non-Small-Cell Lung Cancer

Simple Summary Interstitial lung disease (ILD) is a life-threatening toxicity caused by chemoradiotherapy and durvalumab; however, the risk factors for ILD during durvalumab treatment after chemoradiotherapy have not been established in non-small-cell lung cancer patients. We examined whether interstitial lung abnormalities (ILAs) could be risk factors for ≥grade-two ILD during durvalumab treatment. The prevalence of ILAs in 148 patients before durvalumab treatment was 37.8%, and the multivariate analysis revealed ILAs as independent risk factors for ≥grade-two ILD. Attention should be paid to the development of ILD during durvalumab treatment in patients with ILAs. Abstract Introduction/Background: Chemoradiotherapy (CRT) followed by durvalumab, an immune checkpoint inhibitor, is the standard treatment for locally advanced non-small-cell lung cancer (NSCLC). Interstitial lung disease (ILD) is a life-threatening toxicity caused by these treatments; however, risk factors for the ILD have not yet been established. Interstitial lung abnormalities (ILAs) are computed tomography (CT) findings which manifest as minor interstitial shadows. We aimed to investigate whether ILAs could be risk factors for grade-two or higher ILD during durvalumab therapy. Patients and Methods: Patients with NSCLC who received durvalumab after CRT from July 2018 to June 2021 were retrospectively enrolled. We obtained patient characteristics, laboratory data, radiotherapeutic parameters, and chest CT findings before durvalumab therapy. Results: A total of 148 patients were enrolled. The prevalence of ILAs before durvalumab treatment was 37.8%. Among 148 patients, 63.5% developed ILD during durvalumab therapy. The proportion of patients with grade-two or higher ILD was 33.8%. The univariate logistic regression analysis revealed that older age, high dose-volume histogram parameters, and the presence of ILAs were significant risk factors for grade-two or higher ILD. The multivariate analysis showed that ILAs were independent risk factors for grade-two or higher ILD (odds ratio, 3.70; 95% confidence interval, 1.69–7.72; p < 0.001). Conclusions: We showed that pre-existing ILAs are risk factors for ILD during durvalumab treatment after CRT. We should pay attention to the development of grade-two or higher ILD during durvalumab treatment in patients with ILAs.

Durvalumab, an anti-PD-L1 antibody, is an immune checkpoint inhibitor (ICI). The PACIFIC study, a randomized phase III trial, compared durvalumab and a placebo as maintenance therapy for locally advanced unresectable NSCLC patients after curative chemoradiotherapy (CRT). The primary endpoints were progression-free survival and overall survival, which were better in the durvalumab group than in the placebo group [4][5][6][7]. This favorable result makes durvalumab a standard therapy for locally advanced unresectable NSCLC after curative CRT [5,8].
Interstitial lung disease (ILD) is a major life-threatening adverse event caused by ICIs, chemotherapy, and radiotherapy. A systematic review and meta-analysis showed that the incidence of ICI-induced ILD was relatively higher in NSCLC than in other types of cancers [9,10], and 10% of cases were mortal [11]. The PACIFIC study revealed that the incidence of ILD of any grade in the durvalumab group was 33.9% [4]. Additionally, real-world data showed a higher incidence of ILD (approximately 60-80%) related to the PACIFIC regimen, especially durvalumab [12][13][14][15][16][17][18]. The possible risk factors were radiotherapeutic parameters, such as the volume of the lung parenchyma that received 20 Gy (V20) or the mean lung dose (MLD) [12,13,17]. However, these factors were not significant in other studies [15,16,18]. Therefore, it is necessary to identify the risk factors for ILD during durvalumab treatment.
Interstitial lung abnormalities (ILAs) are defined as subtle or mild parenchymal abnormalities identified in more than 5% of any lung zone on computed tomography (CT) scans in patients not previously clinically suspected of ILD [19,20]. Although ILAs are subclinical findings, their clinical importance is increasing [21]. We have reported that ILAs were risk factors for ICI-induced ILD in patients with NSCLC [22] and even in patients with cancers other than lung cancer [23]. Thus, we hypothesized that ILAs were also risk factors for ILD during durvalumab therapy.
Here, we performed a multicenter retrospective study to investigate whether clinicopathological characteristics and ILAs could be risk factors for ILD during durvalumab treatment.

Study Design and Participants
Consecutive patients with unresectable NSCLC treated with CRT followed by durvalumab between July 2018 and June 2021 were retrospectively enrolled from 10 institutions. The patients who had received systemic corticosteroids before durvalumab treatment were excluded. Information on patient characteristics before durvalumab treatment, including chest CT findings and laboratory data, was obtained. The opt-out method was used in this study to obtain patient consent. This study was approved by the institutional review

Assessment of CT Findings
We investigated the presence of interstitial pneumonia, pre-existing ILAs, and other abnormal findings on CT images taken at the end-inspiration phase in the supine position from the end of CRT before the start of durvalumab treatment. The types of ILAs (Supplementary Figure S1) were determined and categorized as follows: ground glass attenuation (GGA), reticulation, honeycombing, centrilobular nodularity, traction bronchiectasis, nonemphysematous cysts on CT images, and non-dependent abnormalities affecting more than 5% of any lung zone (upper, middle, and lower lung zones were demarcated by the levels of the inferior aortic arch and right inferior pulmonary vein) [20]. The CT images were evaluated by two pulmonologists who did not have information on the characteristics of each patient. First, Readers 1 and 2 independently evaluated all the CT images. CT images with discordant results by the two readers were evaluated by Reader 3, and major opinion was considered the final evaluation of those images.

Diagnosis of ILD during Durvalumab Treatment
The diagnosis of ILD during durvalumab treatment was defined as follows: (1) new abnormal shadows found on chest CT; (2) exclusion of bacterial pneumonia (that did not improve even after administration of antibiotic drugs, or absence of bacterial pathogens detected by sputum culture); (3) exclusion of heart failure (by laboratory data and/or transthoracic echocardiography findings; and (4) exclusion of tumor progression (using laboratory data and version 1.1 of the Response Evaluation Criteria in Solid Tumors). ILD during durvalumab treatment in this study includes both radiation pneumonitis and ICI-induced ILD. ILD was classified according to pneumonitis based on the pneumonitis described in Common Terminology Criteria for Adverse Events (CTCAE) v5.0.

Statistical Analysis
The results are expressed as median (range). Comparisons between the two groups were performed by using Pearson's chi-square or Wilcoxon tests. Univariate and multivariate logistic regression analyses were used to identify risk factors for grade 2 or higher (grade ≥ 2) ILDs. Factors with a p-value less than 0.05 in the univariate analyses were selected for inclusion in multivariable analysis. In addition, V20 in the radiotherapeutic parameters was selected for the multivariate analysis since V20 has been reported as a predictive dose-volume parameter for radiation pneumonitis in several studies [12,13,17,[24][25][26][27][28]. To determine which GGA or reticulation in the ILAs was an independent risk factor for ILD, we generated Models A and B in the multivariate analysis. The optimal cutoff values of continuous variables such as the white blood cell (WBC) count, V5, V20, and MLD were estimated by using receiver operating characteristic curve analysis. The time of onset of ILD was analyzed by using the Kaplan-Meier method. All reported p-values were two-sided. Statistical significance was set at p < 0.05. All statistical analyses were performed by using JMP Pro 16 software (SAS Institute Inc., Cary, NC, USA).

Patient Characteristics
A total of 153 patients were enrolled in this study. A flowchart of the patient selection process is shown in Supplementary Figure S2. Ultimately, 148 patients were included in the analysis. The clinical characteristics and laboratory findings of the patients, as well as information regarding CRT and CT findings before durvalumab treatment are presented in Table 1. Male patients (71.6%), patients with Eastern Cooperative Oncology Group performance status (ECOG-PS) of 0 or 1 (98.0%), and current and former smokers (84.5%) were predominant. The majority of patients (90.5%) had stage III disease.  BMI, body mass index; ECOG PS, Eastern Cooperative Oncology Group performance status; EGFR, epidermal growth factor receptor; ALK, anaplastic lymphoma kinase; PD-L1, programmed death-ligand 1; TPS, tumor proportion score; WBC, white blood cell; LDH, lactate dehydrogenase; CRP, C-reactive protein; CRT, chemoradiotherapy; V5, the volume of lung parenchyma that received ≥5 Gy; V20, volume of lung parenchyma that received ≥20 Gy; IMRT, intensity modulated radiation therapy; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; CT, computed tomography; IP, interstitial pneumonitis; ILAs, interstitial lung abnormalities; GGA, ground glass attenuation.

Incidence and Severity of ILD during Durvalumab Treatment
The occurrence and severity of ILD during durvalumab therapy are shown in Table 2. Among the 94 patients in whom ILD occurred after administration of durvalumab, 50 patients (33.8%) developed grade ≥2 ILD, 54 patients (36.5%) were led to discontinuation of durvalumab treatment, and 40 (27.0%) were administered systemic corticosteroids. ILD occurred within 16 weeks in approximately 90% of patients who developed ILD (Figure 1). We also present a case of grade-three ILD in a patient with ILAs (Supplementary Figure S3).

Comparison of Characteristics between Patients with and without Grade ≥2 ILD
When comparing the baseline characteristics of each group (Table 3), patients with grade ≥2 ILD were significantly older than those without. Radiotherapy parameters such as V5, V20, and MLD were significantly higher in the patient group with grade ≥2 ILD than in that without. The proportion of patients with ILAs on CT was significantly higher in the group with grade ≥2 ILD than in that without (60.0% vs. 26.5%, respectively, p < 0.001). The proportions of patients with GGA and reticulation were significantly higher in the study group with grade ≥2 ILD than in that without (56.0% vs. 13.3%, p < 0.001; 32.0% vs. 14.3%, p = 0.011; respectively).    PD-L1, programmed death-ligand 1; TPS, tumor proportion score; WBC, white blood cell; LDH, lactate dehydrogenase; CRP, C-reactive protein; CRT, chemoradiotherapy; V5, the volume of lung parenchyma that received ≥5 Gy; V20, volume of lung parenchyma that received ≥20 Gy; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; CT, computed tomography; IP, interstitial pneumonitis; ILA, interstitial lung abnormalities; GGA, ground glass attenuation.

Figure 3.
Kaplan-Meier analysis of time to onset of grade ≥2 ILD between patients with ILAs and higher V20, with ILAs or higher V20, and without ILAs and with lower V20. The time to onset of early onset grade ≥2 ILD was significantly shorter in the group with ILAs and higher V20 compared to the group with ILAs or higher V20, log-rank test p < 0.001; the group without ILAs and with lower V20, log-rank test p < 0.001. ILAs, interstitial lung abnormalities; V20, volume of lung parenchyma that received ≥20 Gy; ILD, interstitial lung disease.

Discussion
In this study, the multivariate logistic regression analysis revealed that V20 and ILAs, especially GGA, were independent risk factors for grade ≥2 ILD. To the best of our knowledge, this is the first study to show that pre-existing ILAs are independent risk factors for grade ≥2 ILD during durvalumab treatment. Previous studies showed that the Figure 3. Kaplan-Meier analysis of time to onset of grade ≥2 ILD between patients with ILAs and higher V20, with ILAs or higher V20, and without ILAs and with lower V20. The time to onset of early onset grade ≥2 ILD was significantly shorter in the group with ILAs and higher V20 compared to the group with ILAs or higher V20, log-rank test p < 0.001; the group without ILAs and with lower V20, log-rank test p < 0.001. ILAs, interstitial lung abnormalities; V20, volume of lung parenchyma that received ≥20 Gy; ILD, interstitial lung disease.

Discussion
In this study, the multivariate logistic regression analysis revealed that V20 and ILAs, especially GGA, were independent risk factors for grade ≥2 ILD. To the best of our knowledge, this is the first study to show that pre-existing ILAs are independent risk factors for grade ≥2 ILD during durvalumab treatment. Previous studies showed that the prevalence of ILAs was 4-9% in smokers and 2-7% in non-smokers [29][30][31][32]. Reportedly, ILAs progress to advanced fibrosis and are associated with decreased lung capacities and exercise capacities and high rates of mortality due to respiratory disease and allcause mortality [19,21,29,33,34]. We have already reported that ILAs are risk factors for ICI-induced ILD in NSCLC patients and even in patients with cancers other than lung cancer [22,23,35]. Based on the results of this and previous studies, attention should be paid to the development of grade ≥2 ILD during durvalumab treatment in patients with ILAs.
The finding that ILAs, especially GGA, but not reticulation or honeycombing, is an independent risk factor for grade ≥2 ILD during durvalumab treatment is consistent with that of previous studies [22,23]. Durvalumab exerts antitumor effects by blocking immune tolerance and upregulating lymphocyte activity. GGA reflects the infiltration of lymphocytes into the interstitium of the lungs. In contrast, reticulation reflects slight fibrosis and lymphocytic inflammation, and honeycombing reflects the destruction of the normal alveolar structure [36,37]. These data explain why the presence of GGA in ILAs was associated with ICI during durvalumab treatment in this study. Based on these observations, we should consider ILD development in patients with GGA during durvalumab treatment.
Furthermore, we showed that V20, in addition to ILAs, was an independent risk factor for grade ≥2 ILD. Older age, high V20, high KL-6 level, and smoking history were reported as risk factors for radiation pneumonitis before the approval of durvalumab for CRT [23][24][25][26][27]. Recent studies have shown that a higher V20 or MLD is a risk factor for ILD in patients treated with durvalumab after CRT [12,13,17]. Our study also revealed that V20 was an independent risk factor for ILD. Thus, it is important to focus on V20 in addition to ILAs for predicting severe ILD during durvalumab treatment.
The incidence of ILD (any grade: 63.5%) in our study was higher than that in the PACIFIC study (any grade: 33.9%) [4]. The other real-world data and meta-analysis [12][13][14][15][16][17][18] also showed a higher prevalence of ILD than that in the PACIFIC study. This difference can be explained by the following three factors. First, ethnic differences exist between the PACIFIC study and real-world data. The pooled analysis in a real-world setting showed that the occurrence of ILD during durvalumab treatment after CRT for stage III NSCLC in Asian patients was significantly higher than that in Western patients [38]. The second difference was in the number of elderly patients. The present and previous studies [13,14,18] included a larger number of elderly patients than did the PACIFIC study; this difference might have influenced the frequency of ILD. Finally, MLD and V20 may also influence the prevalence of ILD. In the PACIFIC study, the MLD or V20 was limited to less than 35%. However, several patients in the present and previous studies received radiotherapy that exceeded these limits [12,15,16,18].

Limitations
First, this was a retrospective study. Further prospective studies are required to validate our findings. Second, we did not diagnose ILD pathologically and exclude the influence of concomitant drugs and environment completely. Pathological observations can be valuable in discussing the relationship between CT and pathological findings.

Conclusions
Our study showed that pre-existing ILAs, especially GGA, were significant risk factors for grade ≥2 ILD during durvalumab treatment. This observation is consistent with previous reports from our laboratory. Therefore, significant attention should be paid to the CT findings before durvalumab therapy. We believe that the observations from our study will be useful for the management of durvalumab treatment.
Supplementary Materials: The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/cancers14246236/s1. Figure S1: Examples of ILAs. These are CT images of ILAs: (A) ground glass attenuation, (B) reticulation, (C) honeycombing, (D) centrilobular nodularity, and (E) traction bronchiectasis. ILAs, interstitial lung abnormalities; CT, computed tomography. Figure S2: Flowchart of patient selection. Figure S3: CT images of grade 3 ILD in a patient with ILAs. On the CT findings before durvalumab treatment (A,B), ground glass attenuation in ILAs (arrowheads) was detected in the peripheral fields of both lungs on CT (C). Grade 3 ILD occurred six weeks after durvalumab treatment (D,E). CT, computed tomography; ILD, interstitial lung disease; ILAs, interstitial lung abnormalities. Funding: This research did not receive any specific grants from funding agencies in the public, commercial, or not-for-profit sectors.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Hiroshima University (protocol code, E-1590; and date of approval, 18 April 2019).

Informed Consent Statement:
The opt-out method was used to obtain patient consent in the study.

Data Availability Statement:
The data presented in this study are available to interested researchers upon reasonable requests to the corresponding author based on ethical approval.