Pulmonary Complications of CAR T-cell Therapy (CAR-T PULMOX)

Background and Rationale Chimeric Antigen Receptor (CAR) T-cell therapies are emerging as a revolutionary treatment for hematological malignancies. However, this treatment carries a non-negligible clinical risk of pulmonary complications, which present as varied syndromes. These range from acute disorders, such as Cytokine Release Syndrome (CRS) with respiratory distress (ARDS), to interstitial lung diseases, as well as opportunistic and community-acquired infections related to treatment-induced immunosuppression. A better understanding of these pulmonary complications is necessary to identify predictive and risk factors. This knowledge is essential to guide the development of potential prevention strategies.

Objectives The primary objective of this study is to describe the early and late pulmonary complications associated with CAR T-cell therapies. The primary endpoint is the incidence and nature of pulmonary complications occurring early (≤30 days post-reinjection) and late (>30 days up to 2 years) after treatment.

Secondary objectives include:

• Investigating associations between patient characteristics or follow-up data and the onset of complications through exploratory analyses.
• Describing the prevalence of risk factors.
• Describing complication rates according to per-procedure data (e.g., pre-CAR-T lymphocyte count, duration of aplasia).
• Analyzing complication incidence by calendar periods of reinjection to study the evolution of practices.
• Describing survival curves based on different variables. Methods This is a descriptive, retrospective cohort study. The study will analyze data from all adult patients (expected n=266) who received CAR T-cell therapy at the Hematology Department of Lyon Sud Hospital (CHLS) and who have at least two years of follow-up data collected.

Patient data from January 9, 2017, to January 1, 2025, will be collected from medical records. Statistical analysis will include comparisons of continuous variables (Wilcoxon or Student's t-test) and categorical variables (Chi-squared or Fisher's exact test). Survival will be represented using Kaplan-Meier curves and compared with the Log-Rank test. Univariate and multivariate logistic regression models will be used to identify associations, with a Bonferroni correction applied for multiple tests.

Expected Outcomes The findings are expected to help identify predictive factors for pulmonary complications. This may lead to modified recommendations for pre-CAR-T cell assessments and adaptations to patient follow-up protocols.