Abstract:

Background and purpose: In clinical studies, checkpoint inhibitor-related pneumonitis (CIP) ranks first among the causes of death in programmed cell death protein-1 (PD-1)/programmed death ligand-1 (PD-L1) inhibitor immune-related adverse events. Real-world CIP situations lack extensive population reports. This study aimed to understand the incidence of CIP in the real world of lung cancer in China, and to summarize its characteristics, treatment status and outcomes. Methods: This study retrospectively collected the essential clinical information of patients with an initral diagnosis of lung cancer who received treatment with immune checkpoint inhibitors (ICIs) at the First Affiliated Hospital of Guangzhou Medical University and Shanghai Pulmonary Hospital between January 2019 and September 2021. For patients with CIP, we also collected the time of its onset, grade, treatment regimen and outcome. The analyses of incidence, patient’s characteristics and the risk factors of CIP in overall and subgroup were carried out. Moreover, we analyzed the outcomes of patients treated with immunosuppressive therapy. Results: A total of 2 031 patients with immunotherapy were enrolled, with a CIP incidence rate of 7.2% (147/2 031), a severe CIP rate of 2.6% (52/2 031) and a mortality rate of 0.4% (9/2 031). The rate of severe grade in the population with CIP was 35.4%, and the mortality rate was 6.1% (9/147). Compared with non-CIP patients, more CIP patients were male, older (>65 years), with combination therapy, and on first- and second-line immunotherapy in advanced treatment. In subgroup analyses, the incidence of CIP was higher in men, the elderly (>65 years), squamous cancer, combination therapy, anti-PD-1 inhibitors, and first- and second-line therapy in advanced treatment. The median onset time of CIP in the real world was 148 days, with a double-peak characteristic, that was, 60-90 days and 150-210 days after immunotherapy were both the peak time periods for CIP onset. The incidence of CIP was also influenced by seasonality, with a high incidence in autumn and winter. All treated patients used corticosteroids as first-line treatment; the immunosuppressive treatment rate of CIP in this study was 76.2%. After treatment, 97.9% of mild CIP patients and 81.2% of severe CIP patients had a good prognosis, and 17.3% of severe CIP patients died due to CIP. Conclusion: In the real world, the incidence of CIP for lung cancer patients was 7.2%, incidence of severe CIP was 2.6%, and mortality rate was 0.4%; the incidence of severe disease in the population with CIP was 35.4%, and mortality rate was 6.1%. The median onset time of CIP was characterized by a double peak, and incidence of CIP was higher in autumn and winter. Men, the elderly, squamous cancer patients, patients on combination therapy, patients who used anti-PD-1 inhibitors, and patients with advanced treatment had higher incidence of CIP. Most patients with CIP had good outcomes after immunosuppressive therapy.

Key words: Lung cancer, Checkpoint inhibitor-related pneumonitis, Immune-related adverse event, Immune checkpoint inhibitor, Epidemiology, Real-world research