Figure 1.
Study flowchart and disproportionality analysis of CAR-T−related PML adverse events (AEs). (A) Reports without a defined primary suspected drug for a given AE and suspected duplicated reports were excluded. (B) Disproportionality analysis of CAR-T products compared to the full FAERS database (ie, “main analysis”) and sensitivity analyses using restricted comparator groups of cancer patients and hemato-oncological patients. Results of disproportionality analysis by CAR-T product and following restriction to NHL patients are also presented. A positive lower bound of the information component 95% credibility interval (IC₀₂₅ > 0) and a lower limit of the ROR 95% confidence interval above 1 are the conventional thresholds for significant signal detection.