Reduced FEV1 as Prognostic Factors in Patients With Advanced NSCLC Receiving Immune Checkpoint Inhibitors

Background The aim of study is to investigate the influence of pulmonary function on the prognosis in patients with advanced non-small cell lung cancer (NSCLC) receiving immune checkpoint inhibitors (ICI). Patients and Methods Data were collected retrospectively from 151 patients with stage IV NSCLC who received ICI and completed spirometry before ICI therapy in Taipei Veterans General Hospital between January 2016 and December 2020. The co-primary end points were overall survival (OS) and progression-free survival (PFS) between groups divided by 80% predicted FEV1 since ICI therapy started; the secondary outcomes were objective response rate. Results Among 151 patients enrolled to this study, 67.5% of patients were men, 75.5% were adenocarcinoma, 24.5% had known targetable driver mutation, 33.8% received first-line ICI, and 62.8% received ICI monotherapy. The objective response rate was 24.5% and disease control rate was 54.3%. In multivariable analysis, patient with reduced FEV1 had inferior PFS (FEV1 < 80% vs. FEV1 ≥ 80%, adjusted HR = 1.80, P = 0.006) and OS (FEV1 < 80% vs. FEV1 ≥ 80%, adjusted HR = 2.50, P < 0.001). Median PFS and OS in the preserved FEV1 group (≥80% predicted FEV1) compared to the reduced FEV1 group (<80% predicted FEV1) were 5.4 vs. 2.9 months (HR = 1.76, P = 0.003) and 34.9 vs. 11.1 months (HR = 2.44, P < 0.001), respectively. The other independent prognostic factors of OS include stage IVA disease (adjusted HR = 0.57, P = 0.037), initial liver metastasis (adjusted HR = 2.00, P = 0.049), ICI monotherapy (adjusted HR = 1.73, P = 0.042) and ICI related pneumonitis (adjusted HR = 3 .44, P = 0.025). Conclusions Reduced FEV1 is strongly associated with inferior clinical outcomes in patients with advanced NSCLC treated with ICI.


INTRODUCTION
Lung cancer is the leading cause of death worldwide, most of which is non-small cell lung cancer (NSCLC), accounting for 85% of all cases (1). In recent decades, several oncogenic molecular alterations such as epidermal growth factor receptor (EGFR) mutation, anaplastic lymphoma kinase (ALK) gene rearrangement, have been found and established as the targets of therapy. The management of NSCLC with targetable oncogene is remarkably advanced in past decade (2). However, the outcome of advanced NSCLC without targeted therapy remained dismal. Immune checkpoint inhibitors (ICI) represent another breakthrough achievement in cancer treatment. ICI which targets programmed cell death protein-1 (PD-1) or programmed cell death ligand-1 (PD-L1), have shown improvement of progression free survival (PFS) and overall survival (OS) in clinical trials, compared to chemotherapy in first-line or secondline therapy (3)(4)(5)(6)(7).
Interestingly, recent studies on immunotherapy implicate that COPD is associated with better clinical outcomes in patients with NSCLC treated with ICI (21). Another immunological study suggests pre-existing ILD doesn't impact prognosis in patients treated with first-line pembrolizumab (22). To the best of our knowledge, the influence of reduced pulmonary function in patients with NSCLC treated with ICI is not fully investigated. The aim of study is to investigate the impact of pulmonary function on the prognosis and treatment outcome in patients treated with ICI.

Study Population
This is a retrospectively observational cohort study of patients with advanced NSCLC who received ICI. We identified 296 patients who received ICI between January 2016 and December 2020 from the lung cancer registry in Taipei Veterans General Hospital. Patients with small cell lung cancer (N = 32), or with stage III disease received definite concurrent chemoradiotherapy then followed by ICI (N = 7) were excluded. Then patients who received ICI treatment without undergoing spirometry (N = 95) before ICI, or those who underwent spirometry without meeting the criterions of the American Thoracic Society/European Respiratory Society (N = 11) were excluded. The final study population included 151 patients (Figure 1). Our study was carried out in accordance with the principles of the Declaration of Helsinki. The Institutional Review Board of Taipei Veterans General Hospital had approved our study (VGHIRB no. 2020-07-046CC) and waived the requirement for informed consent.

Assessments and Data Collection
Data regarding patients' demographics, including age, sex, smoking history, Eastern Cooperative Oncology Group performance status (ECOG PS) (23), and tumor characteristics, such as stage, histology, initial metastatic sites, status of EGFR mutation or ALK, c-ROS oncogene 1 (ROS-1) rearrangement, prior treatments and ICI-related pneumonitis (ICI-pneumonitis) were collected from electronic medical records. PD-L1 expression was assessed in formalin-fixed, paraffin-embedded tumor samples using PD-L1 IHC 22C3 pharmDx Kit (clone 22C3 [DAKO, Carpinteria, CA]), performed on Dako Autostainer Link 48 platform with a validated and automated staining protocol. The tumor proportion score (TPS) was defined as the percentage of viable tumor cells showing partial or complete membrane staining. Spirometry was performed using Vmax 22 (SensorMedics, Yorba Linda, CA), and interpretated according to the recommendations of the American Thoracic Society/European Respiratory Society guidelines (24,25). The FEV 1 (% of predicted), FVC (% of predicted), and FEV 1 /FVC (ratio) from pulmonary function test results were based on pre-bronchodilation values measured before ICI treatment. FEV 1 /FVC<0.7 is classified as airflow obstruction, and 80% of predicted FEV 1 was determined as a universal cut-off value for distinction between the reduced FEV 1 group and preserved FEV 1 group, according to previous studies and the classification of Global Initiative for Chronic Obstructive Lung Disease (GOLD) guideline (20,(26)(27)(28).

Endpoints
The co-primary endpoints of this study were OS, measured from the date of starting ICI to death from any cause or last known date alive, and PFS, measured from the date of starting ICI to the date of initial disease progression, death from any cause, or the last date known to be alive without disease progression. The treatment response and the date of disease progression were confirmed by two authors (Y.L.S. and Y.H.L.) who reviewed the diagnostic imaging and medical record. The secondary endpoint was the objective response rate (ORR), defined as the percentage of patients with a confirmed complete response (CR) or partial response (PR). Treatment response was routinely reviewed every 2-3 months, or when disease progression was highly suspected. The assessment of response was based on Response Evaluation Criteria in Solid Tumors (RECIST, version 1.1) (29).

Statistical Analysis
Categorical data from patients' profile were presented as numbers (%), and compared using Pearson's Chi-square-test and Fisher's exact test, as appropriate. Continuous variables were presented as means with standard deviation or median with interquartile range based on Kolmogorov-Smirnov normality test, then performed Student's t-test or Mann-Whitney U-test, respectively. The Kaplan-Meier method with the log-rank test was used for survival analysis. Hazard ratios (HRs) and 95% CI were calculated using the Cox proportional-hazard model, and multivariable analysis for baseline characteristics of patients and spirometry parameters. Multiple Cox proportional-hazard models were performed using the significant variables (p < 0.10) in the multivariate analysis. All tests were two-sided, and p < 0.05 were considered significant. All analyses were performed using SPSS software (version 25.0, IBM corp., Chicago, IL, USA).

Patient Characteristics and Treatment
Among 151 patients receiving ICI, the mean age was 63.0 years old while started treatment; 102 (67.5%) were men and 83 patients had smoking history (55.0%). The majority of ECOG PS was 1-2 (90 patients, 59.6%), and other patients was 0 (61 patients, 40.4%). There were 58 patients at stage IVA (38.4%) and 93 patients at stage IVB or IVC (61.6%) before ICI, and the proportions of patients with initial distant metastasis of brain, lung, and liver were 22.5, 33.1, and 11.3%, respectively. Most of the histopathological type was adenocarcinoma (114 patients, 75.5%), then followed by squamous cell carcinoma (25 patients, 16.6%). Thirty-three patients presented EGFR mutation (21.9%); 1 patient had ALK rearrangement; 3 patients possessed ROS-1 rearrangement, and 99 patients with EGFR wild type were documented without known driver mutation. The proportions of patients with PD-L1 TPS ≥ 50%, 1-49%, and <1% were 23.2, 17.9, and 19.9%, respectively. However, PD-L1 expression was unavailable in many patients (39.1%). Total 51 patients received first-line ICI therapy (33.8%). Previous-treated patient received ICI had all received chemotherapy before (100 patients), then 89 patients had previously received radiotherapy and 42 patients had lung surgery. There were 96 patients received ICI monotherapy (62.8%) and 55 patients received ICI treatment combined with other therapies, including anti-angiogenesis agent, chemotherapy, or others (37.2%). Total 6 (4%) patients had ICI-pneumonitis. For spirometry data, median FEV 1 was 1.99 liters (84.0% median prediction of FEV 1 ), and the median FVC was 2.65 liters (83.0% median prediction of FVC). The median FEV 1 /FVC ratio was 77.0% ( Table 1). There was significant correlation between percentage of prediction of FEV 1 and FVC in our study (R = 0.910, P < 0.001). Thirty-three patients had FEV 1 /FVC ratio < 0.7, but only 7 patients had physiciandiagnosed COPD in medical record. There are 65 patients with reduced FEV 1    had received radiotherapy and 1 patient was treated as asthma and received inhaled corticosteroid. There was 1 patient cannot be refined detailly based on electrical medical record.

Treatment Response in Patients Receiving ICI
In first period of follow-up, the ORR based on RECIST was 24.5%, and the disease control rate was 54.3%. The demographics distribution of patients, categorized by treatment response, was not significantly different in levels of PD-L1 expression and other variables, except the ECOG PS. The proportion of ECOG PS = 0 in the group with treatment responder (with CR or PR) was higher than that in the non-responder (with stable disease or progressive disease) (P = 0.020) (

Overall Survival in Patients Treated With ICI
Total 82 events of death were record in study cohort. In Kaplan-Meier survival analysis, OS in reduced FEV 1 group was significantly shorter than in preserved FEV 1 group. (Median OS: 11.1 vs. 34.9 months; HR = 2.44, p < 0.001) ( Figure 3A). The subgroup analysis of OS showed that an increased risk of death with reduced FEV 1 was found in most subgroups ( Figure 3B).
Reduced FEV 1 group also presented increased risk of disease progression or death in analysis of key subgroups (Different lines of therapy, patients without driver mutation; detail in Supplementary Data).

DISCUSSION
To date, this is the first study to analyze the treatment outcome of stage IV NSCLC receiving ICI in terms of lung function. Our study demonstrated that reduced FEV 1 , not reduced FEV 1 /FVC ratio, is an independent prognostic factor of the inferior survival outcome in patients with ICI-treated advanced NSCLC, irrespective of ECOG PS, various degrees of distant metastasis, different lines of therapy, ICI-combination therapy and ICI-pneumonitis.
In clinical practice, FEV 1 is a non-invasive tool for evaluating pulmonary function. Patient who has declined FEV 1 might be associated with increased respiratory symptoms, inferior life quality, and mortality (30)(31)(32). Past studies showed FEV 1 worked as a better predictor than FVC in survival (33). A previous research discloses that patients with small cell lung cancer and FEV 1 < 80% had inferior OS compared to those with FEV 1 ≥ 80% (27). According to the American Society of Clinical Oncology guideline, baseline spirometry is recommended for every patient with cancer prior to ICI (34). However, it is not routinely tested in real-world experience. In most clinical settings, spirometry was performed for preoperative evaluation, pre-radiotherapy assessment, evaluating the cause of persisted dyspnea, or follow-up for chronic lung diseases, such as COPD or ILD.
Our study tested the hypothesis that using 80% predicted value of FEV 1 in patients with advanced NSCLC receiving ICIs could meet the prognostic significance in OS and PFS. Previous two studies have reported that 50% predicted value of FEV 1 at initial Frontiers in Medicine | www.frontiersin.org diagnosis of lung cancer is an independent prognostic factor for advanced NSCLC, after adjusting TNM stage and the presence of malignant pleural effusion, suggesting reduced FEV 1 is an important factor in survival prediction (18,19). Additionally, our study focuses on pre-treatment pulmonary functions, considering that initial pulmonary functions would be affected by subsequent treatment such as palliative surgical resection, radiotherapy, or drug-induced lung toxicities. Collectively, these findings suggest that patients with reduced FEV 1 had inferior outcome, and their pulmonary function should be carefully monitored. Lung function forms part of exercise performance and quality of life. Assessment of respiratory symptoms and health related quality of life might be important to clarify the relationship between FEV 1 and mortality.
In an earlier study, patients coexisting with COPD and NSCLC who received pembrolizumab monotherapy had longer PFS and OS than those without COPD (21). Nonetheless, our study did not show significant correlation between COPD and outcome. We used pre-bronchodilator spirometry in this study due to the limited number of patients with postbronchodilator spirometry, which estimate the prevalence of COPD should be careful (35). Surprisingly, total 33 patients had FEV 1 /FVC<0.7 but only 7 patients had physiciandiagnosed COPD in electrical medical records, which indicated COPD might be underestimated. Clinical physicians might overlook COPD or other chronic lung diseases and attribute respiratory symptoms to lung cancer in real-world practice (36). Inappropriate management of comorbidity might be fatal. The importance of pulmonary function tests for improving the clinical practice in comorbidity management of those patients with lung cancer is heightened by our findings in the current study.
In consideration of the ECOG PS, tumor stage, and ICI regimen, reduced FEV 1 is still statistically significant with the risk of death, implicating that reduced FEV 1 might interfere ICI treatment via some ambiguous mechanisms. Reduced FEV 1 could be resulted from chronic lung disease in several pathways, such as recurrent infection, airway inflammation, mucus plugging, structural change of alveoli. Chronic engagement of checkpoint receptors with frequent inflammation and antigenic stimulation would lead to T cell exhaustion. Previous research disclosed that increasing PD-1 expression of CD8 + tumorinfiltrating T lymphocytes is found in patients coexisting with NSCLC and COPD, correlated to the level of reduced FEV 1 (37). Given that T-cell exhaustion is the self-protective mechanism for dysregulation of immune reaction, patients with impaired lung function might be vulnerable to ICI therapy as blockade of PD-1/PD-L1 pathway, causing airway injury, lung function decline, or pneumonitis. Additionally, analysis of circulating inflammatory markers related to ICI treatment such as neutrophil to lymphocyte ratio, LDH or CRP, which are widely used in previous studies, could provide more detailed inflammatory profiles for evaluation in the future investigation (38,39).
Interestingly, in this study, ICI-pneumonitis was also associated with poorer PFS and OS. ICI-pneumonitis would impair pulmonary function seriously, and lead to mortality and morbidity if left untreated. However, our study didn't investigate all types of immune-related adverse events (irAE), which might be related to variable treatment response (40,41). Consequently, reduced FEV 1 before treatment may be potentially exacerbated by ICIs, resulting in deterioration of pulmonary function and poor survival outcome. Based on this study, pretreatment pulmonary function is a potentially immunotherapeutic parameter that should be regularly performed before ICI administration, regardless of previous treatment status. It also provides more information for clinicians regarding comprehensive evaluation of illness. In real-world practice, the prognosis for previous-treated patients remains dismal, the useful parameters are urgently needed clinically for these patients. Further prospective studies for longitudinal investigation of the dynamic changes of pulmonary function before and after ICI are warranted, and could help us elucidate the relationship between pulmonary function and prognosis of patients receiving ICIs.
There are several limitations of in our study. First, this is a retrospective cohort study in a single tertiary medical center, and some relevant data including all types of irAE, cumulative dose of ICI, subsequent treatment after ICI failure, tumor infiltrated lymphocytes, and tumor mutational burden were not available. Besides, there were about 39% of all patients misses the PD-L1 study owing to lack of adequate biopsy tissue or not performed. Those missing data may influence the statistical analysis. Second, our study population was relatively small and had limited generalizability due to a single center experience and lack of validation cohort. Third, we used spirometry test as closely as possible before ICI treatment, but it might remain unsatisfied for the perfect timing of test. Additionally, the follow-up spirometry for lung function decline was unavailable for comprehensive study. Fourth, lung volume test, bronchodilator test and diffusing capacity are not routinely performed in most patients, so the impact of restrictive lung disease or air trapping on the outcome is not fully investigated. Despite the limitations, our data are still representative in the real-world practice and useful for patients who undergo ICI therapy for NSCLC.
In conclusion, reduced FEV 1 with cut-off level of 80% predicted value is strongly associated with inferior outcomes in patients with advanced NSCLC treated with ICIs. Regular follow-up of spirometry might facilitate more accurate prediction of prognosis, thereby assisting in optimal decision-making in patients with ICI treatment. The detailed pathophysiology regarding the influences of reduced FEV 1 on patient's prognosis remains to be elucidated, and further perspective study is warranted.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Institutional Review Board of Taipei Veterans General Hospital (VGHIRB no. 2020-07-046CC). Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

AUTHOR CONTRIBUTIONS
Y-LS: conceptualization, methodology, software, formal analysis, writing-original draft, and visualization. C-IS: conceptualization, methodology, formal analysis, investigation, resources, and data curation. C-LC, H-CH, and K-TC: investigation, resources, and data curation. C-HC: investigation, resources, data curation, and funding acquisition. Y-MC: investigation resources, supervision, project administration, and funding acquisition. Y-HL: conceptualization, formal analysis, investigation, resources, data curation, writing-review and editing, supervision, project administration, and funding acquisition. All authors contributed to the article and approved the submitted version. The remaining 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.
Publisher's Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Copyright © 2022 Shen, Shen, Chiang, Huang, Chou, Chiu, Chen and Luo. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.