Abstract
Background/Aim: Epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs) are the standard treatment therapy for non-small cell lung cancer (NSCLC) with EGFR mutations. However, patients harboring the exon 21 L858R point mutation (L858R) typically have poorer treatment outcomes than those with exon 19 deletions. Our previous findings suggest that among patients with L858R-positive NSCLC, those with programmed death-ligand 1 (PD-L1) expression may have an increased risk of developing interstitial lung disease (ILD) during EGFR-TKI therapy, potentially compromising treatment continuity and prognosis. This study investigated the association between PD-L1 expression and ILD occurrence in patients with EGFR-mutant NSCLC receiving EGFR-TKIs.
Patients and Methods: We retrospectively analyzed patients with EGFR-mutant NSCLC treated with EGFR-TKIs and compared clinical characteristics between those who developed ILD and those who did not. Survival was defined as the interval from initiation of lung cancer treatment to death or last follow-up.
Results: Among 76 patients, 11 (14.5%) developed ILD during treatment. The ILD group had a significantly higher proportion of L858R-positive cases compared with the non-ILD group. Multivariate analysis identified L858R mutation and PD-L1 positivity as independent risk factors for ILD. Overall survival was significantly longer in the non-ILD group than in the ILD group (p=0.001).
Conclusion: Among patients with EGFR-mutant NSCLC undergoing EGFR-TKI therapy, the presence of the L858R mutation and PD-L1 expression are associated with an elevated ILD risk. Enhanced monitoring and individualized, risk-adapted management strategies are warranted to optimize outcomes for high-risk patients.
- Non-small-cell lung cancer
- epidermal growth factor receptor
- interstitial lung diseases
- mutation
- programmed death-ligand 1
Introduction
Epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs) are the standard targeted therapy for non-small cell lung cancer (NSCLC) with EGFR mutations (1, 2). The most common activating mutations are exon 19 deletion (DEL19) and exon 21 L858R point mutation, both conferring EGFR-TKIs sensitivity. However, patients with L858R generally exhibit poorer treatment outcomes than those with DEL19 (3, 4).
Interstitial lung disease (ILD) during EGFR-TKI therapy is a severe adverse event that often necessitates treatment interruption or discontinuation, potentially compromising outcomes and prognosis (5). Established risk factors include male sex, advanced age, poor performance status (PS), smoking history, pre-existing lung disease, and prior radiation therapy (6). Among Japanese patients, genetic predisposition and immune response characteristics have also been implicated in ILD pathogenesis (7, 8).
Our previous study identified a higher incidence of ILD in programmed death-ligand 1 (PD-L1)-positive L858R cases (9). In this study, we examined a broader EGFR-mutant cohort to validate these findings and further assess PD-L1 expression as an ILD risk factor.
Patients and Methods
We retrospectively reviewed the data of patients with EGFR-mutant NSCLC who received EGFR-TKIs for stage IIIB, stage IV, or postoperative recurrence (July 2012 and December 2023). Patients were excluded if they had non-adenocarcinoma histology, uncommon EGFR mutations, or who were lost to follow-up within 7 days after EGFR-TKI initiation. Patients were categorized into two groups: the ILD group, consisting of patients who developed ILD after EGFR-TKI therapy, and the non-ILD group, consisting of patients who did not develop ILD (Figure 1).
Study flow diagram. A total of 85 patients with stage IIIB, stage IV, or postoperative recurrent non-small cell lung cancer who received EGFR-tyrosine kinase inhibitor (TKI) therapy were initially identified. One patient with squamous cell carcinoma harboring an L858R mutation, four patients with exon 18 G719X mutations, and three patients with exon 20 insertion mutations were excluded. In addition, one patient who was referred to another hospital within 7 days after initiating EGFR-TKI therapy was excluded because the patient’s interstitial lung disease (ILD) status could not be assessed. Consequently, 76 patients were included in the final analysis.
Clinical data included age, sex, smoking history, PS, histological subtype, clinical stage, Krebs von den Lungen-6 (KL-6), percent vital capacity (%VC), percent diffusing capacity for carbon monoxide (%DLCO), and treatment details. ILD severity was evaluated according to the Common Terminology Criteria for Adverse Events (CTCAE), version 5.0. This study was approved by the Institutional Review Board of Kanazawa Medical University Hospital (approval number: C106). The requirement for informed consent was waived due to the retrospective nature of the study.
Immunohistochemistry. PD-L1 expression was assessed by immunohistochemistry (IHC) using the PD-L1 IHC 22C3 pharmDx kit (Dako, Glostrup, Denmark) following the manufacturer’s instructions (10).
EGFR mutation analysis. Tumor specimens from primary or metastatic lesions were analyzed using Cycleave-PCR before 2021 and the Oncomine Dx Target Test (Ion Torrent PGM Dx Sequencer, Thermo Fisher Scientific, Norristown, PA, USA) from 2022 onward. Cytological specimens were excluded, and all analyses were performed by SRL Laboratories (Hachioji, Tokyo, Japan), a certified commercial laboratory.
Statistical analysis. Statistical analyses were performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). Categorical variables with expected frequencies >5 were analyzed using the chi-square test, and those with expected frequencies <5 were analyzed using Fisher’s exact test. Continuous variables were compared using the unpaired t-test. Logistic regression was used for multivariate analysis.
Overall survival (OS) was defined as the interval from initiation of lung cancer treatment to death or last follow-up and was analyzed using the Kaplan-Meier method, with the log-rank test for group comparisons. A p-value <0.05 was considered statistically significant.
Results
Patient background. A total of 85 patients with EGFR mutation-positive NSCLC were identified. We excluded patients with non-adenocarcinoma histology (squamous cell carcinoma harboring L858R, n=1), with uncommon EGFR mutations (exon 18 G719X, n=4; exon 20 insertion, n=3), and who were lost to follow-up within 7 days after EGFR-TKI initiation, in whom ILD status could not be assessed (n=1). Consequently, 76 patients (11 in the ILD group and 65 in the non-ILD group) were included in the final analysis (Figure 1).
Patient characteristics are shown in Table I. Of the 76 patients with EGFR mutation-positive NSCLC, 35 patients (46.1%) had an exon 21 L858R point mutation, and 41 (53.9%) had DEL19. EGFR-TKI regimens comprised osimertinib in 34 patients (33 as monotherapy, one with platinum-based chemotherapy), gefitinib in 24 (all monotherapy), erlotinib in 12 (10 monotherapy, two with ramucirumab), and afatinib in six. ILD developed in 11 patients (14.5%), all of whom received EGFR-TKI as first-line therapy.
Comparison of patient characteristics between the interstitial lung disease (ILD) and non-ILD groups.
Clinical characteristics of the ILD group. Patients in the ILD group tended to be older (p=0.057) and had a significantly higher proportion of L858R-positive cases than the non-ILD group (p=0.005). A significantly greater proportion of non-responders was observed in the ILD group (p=0.018), and the median EGFR-TKI treatment duration was shorter (143.7 days vs. 584.0 days, p<0.001). No significant differences were noted between groups regarding sex, PS, smoking history, PD-L1 expression, treatment regimen or initiation period, pulmonary function (%VC, %DLCO), or KL-6 levels. The incidence of ILD by EGFR mutation subtype was 25.7% (9/35) for L858R and 4.9% (2/41) for DEL19.
Severity and rechallenge. The characteristics of patients who developed ILD are shown in Table II. ILD severity was Grade 5 in two patients, Grade 4 in one, Grade 3 in three, and Grade ≤2 in five. All Grade ≤2 cases improved with EGFR-TKI discontinuation alone. Rechallenge was attempted in only one case (Case 11), with therapy switched from osimertinib to gefitinib. In the remaining cases, EGFR-TKI treatment was not resumed.
Characteristics of patients who developed interstitial lung disease (ILD).
ILD risk factors. Univariate analysis identified the L858R mutation as a significant risk factor for ILD [odds ratio (OR)=6.750; 95% confidence interval (CI)=1.349-33.787; p=0.020]. In multivariate analysis, including L858R mutation, age, smoking history, and PD-L1 expression, the L858R mutation remained an independent risk factor for ILD (OR=9.108; 95%CI=1.555-53.340; p=0.014). Negative PD-L1 expression was independently associated with a lower ILD risk (OR=0.085; 95%CI=0.008-0.924; p=0.043) (Table III).
Univariate and multivariate analyses of risk factors for nonresponse in patients with non-small cell lung cancer treated with docetaxel (DTX) +ramucirumab (RAM) therapy.
Survival analysis. The survival curves for patients with and without ILD are shown in Figure 2. Kaplan-Meier analysis demonstrated significantly longer OS in the non-ILD group than in the ILD group (log-rank test, p=0.001).
Overall survival of patients with and without interstitial lung disease (ILD). The median overall survival was 653.0 days in the ILD group and 2,117.0 days in the non-ILD group, with a statistically significant difference (log-rank test, p=0.001).
Discussion
In this retrospective study, we aimed to validate the association between L858R mutation and ILD and assess PD-L1 expression as an ILD risk factor. Our results showed that L858R mutation and positive PD-L1 expression were independent ILD risk factors, an association rarely explored, making our findings clinically relevant.
Regarding PD-L1 expression, previous studies have reported that the administration of EGFR-TKIs following PD-1/PD-L1 inhibitor therapy increases the risk of ILD (11, 12). This suggests that an immune-activated state may lower the threshold for EGFR-TKI-induced lung injury. Furthermore, tumors with high PD-L1 expression exhibit a “T cell-inflamed” tumor microenvironment, characterized by elevated interferon-γ-related gene expression and abundant tumor-infiltrating lymphocytes (13, 14). Consistent with this concept, Sasada et al. reported that the occurrence of immune-related adverse events, including pneumonitis, was positively correlated with PD-L1 expression levels, particularly in patients with high PD-L1 expression (>50%) (15). Such a locally activated immune milieu could predispose patients to drug-induced lung injury upon exposure to EGFR-TKIs. Although our cohort did not include patients previously treated with immune checkpoint inhibitors, the observed association between high PD-L1 expression and ILD risk may reflect a similar underlying immune-activated state. Nevertheless, these interpretations remain speculative, and further prospective histopathological and immunological studies –including paired pre- and post-treatment biopsies or bronchoalveolar lavage assessments– are warranted.
Prior reports have suggested that the L858R mutation is associated with lower responsiveness to EGFR-TKIs compared with DEL19 (3, 4). In our study, the incidence of ILD was significantly higher in patients with L858R mutation than in those with DEL19, and multivariate analysis identified L858R as an independent risk factor for ILD. These findings suggest that patients harboring the L858R mutation may be more susceptible to EGFR-TKI-related lung injury, potentially leading to treatment discontinuation and affecting clinical outcomes. However, the underlying mechanisms remain unclear, and further studies are needed to confirm this association.
Study limitations. First, it is a single-center, retrospective analysis, which may introduce selection bias and rely on incomplete medical records. In addition, all patients were Japanese, among whom the incidence of EGFR-TKI-associated ILD has been reported to be particularly high (16). Therefore, caution should be exercised when generalizing these results to other populations or ethnic groups. Second, the overall sample size was limited, particularly in the ILD group (n=11), which may have reduced statistical power, limited the stability of the multivariate analysis, and increased the risk of overfitting. Thus, the associations reported in this study should be interpreted as hypothesis-generating rather than definitive conclusions. Third, PD-L1 expression data were missing in 34% of patients, which may affect the robustness of the analysis. Fourth, most patients in this study received first- to third-generation EGFR-TKIs; newer agents expected to gain broader clinical use, such as the combination of amivantamab and lazertinib, were not included. Therefore, the applicability of these findings to future treatment regimens remains uncertain. Furthermore, heterogeneity in EGFR-TKI type, administration timing, and use of combination therapies may introduce confounding effects when evaluating treatment efficacy and adverse events. Notably, the incidence of ILD differs among EGFR-TKI generations: first- and second-generation TKIs induce ILD in 3-5% of East Asian patients, whereas osimertinib is associated with a higher ILD incidence of 12-18% (17). This heterogeneity may introduce potential bias when comparing results across studies. In our cohort, the distribution of osimertinib use was similar between patients who developed ILD and those who did not, suggesting that this variability is unlikely to have significantly affected our findings. To overcome these limitations and validate our results, prospective multi-institutional studies are warranted.
Conclusion
We identified the EGFR L858R mutation and positive PD-L1 expression as independent risk factors associated with the development of ILD during EGFR-TKI therapy in patients with EGFR-mutated NSCLC. These findings suggest the value of evaluating PD-L1 expression prior to initiating EGFR-TKI treatment, particularly in patients harboring the L858R mutation. Given the possibility of an elevated ILD risk in patients with high PD-L1 expression, close clinical monitoring is warranted, and individualized treatment strategies should be considered to ensure patient safety and optimize clinical outcomes.
Acknowledgements
None.
Footnotes
Authors’ Contributions
All Authors had full access to the study data and take responsibility for the integrity of the data and accuracy of the data analysis. All the authors have read and approved the submission of the manuscript. Conceptualization, Y. T.; Resources, Y. T., R. A., S. N., T. T., Y. I., I. S., K. Y., M. N., and M. I.; Investigation, R. A., S. N., T. T., Y. I., I. S., K. Y., M. N., and M. I.; Methodology, Y. T. and M. I.; Writing–original draft preparation, Y. T., with support from M. I.
Conflicts of Interest
The Authors report no conflicts of interest in relation to this study.
Funding
This study did not receive financial support from any public, commercial, or nonprofit funding agency.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received October 23, 2025.
- Revision received November 11, 2025.
- Accepted November 12, 2025.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
Jump to section
Related Articles
Cited By...
- No citing articles found.