Objective To investigate the potential risk factors for immune checkpoint-related pneumonia (CIP) in non-small cell lung cancer (NSCLC) patients treated with immune checkpoint inhibitors (ICIs), to identify high-risk patients with CIP at an early stage. Methods A total of 728 NSCLC patients treated with ICIs at the First Affiliated Hospital of Zhengzhou University from January 2020 to June 2023 were retrospectively selected, including 47 cases with CIP (CIP group), and 141 matched NSCLC patients without CIP (control group). Clinical data, laboratory tests, and CT images before the first immunotherapy were collected for all patients in two groups. The FACT medical imaging software was utilized for quantitative emphysema assessment in patients' CT scans. Univariate and multivariate logistic regression analyses were conducted to identify the risk factors associated with CIP. Receiver operating characteristic (ROC) curves were plotted to evaluate the predictive value of these factors for CIP occurrence in NSCLC patients. Results Among the 47 CIP patients, 40(85.1%) were male, with 25(53.0%) aged between 41 and 65 years. Grade 3 pneumonia according to the Common Terminology Criteria for Adverse Events (CTCAE) was found in 28(59.6%) cases, characterized by a predominant reticular radiographic pattern. Multivariate logistic regression analysis showed that a low albumin level (OR=0.889, 95%CI 0.808-0.979, P=0.017), targeted therapy (OR=9.204, 95%CI 1.678-50.486, P=0.011), anti-angiogenic therapy (OR=0.056, 95%CI 0.020-0.161, P<0.001), and a high percentage of low attenuation area (LAA%) (OR=1.227, 95%CI 1.053-1.430, P=0.009) were significant factors influencing CIP occurrence. The combined predictive model using these four factors showed an area under the ROC curve of 0.888 (95%CI 0.838-0.939), with a sensitivity of 0.894 and a specificity of 0.801 for predicting CIP in NSCLC patients. Conclusions Low serum albumin, first-line targeted therapy, and high LAA% are identified as risk factors for CIP occurrence, while anti-angiogenic therapy is a protective factor. The predictive model based on these four variables effectively predicts the risk of CIP in NSCLC patients.