Supplementary Methods, Tables 1 - 19, Figures 1 - 5 from Identification of Predictive Biomarkers for Cytokine Release Syndrome after Chimeric Antigen Receptor T-cell Therapy for Acute Lymphoblastic Leukemia

This is the authors' abstract. We don't add key points for this paper.

Supplementary methods, statistical methods, results, tables, and figures. Supplementary Table 1. CRS grading system. Supplementary Table 2. Samples collected and time points on the trials. Supplementary Table 3. HLH diagnostic criteria. Supplementary Table 4. Timing of severe CRS relative to CRS onset and time from infusion. Supplementary Table 5. Patient demographics and baseline characteristics. Supplementary Table 6. Co-morbid infections. Supplementary Table 7. Median (range) for baseline cytokine values comparing normal subjects and total study population, children on the trials, and adults on the trials. Supplementary Table 8. Median (range) baseline cytokine values comparing baseline disease burden. Supplementary Table 9. Median (range) one-month peak cytokine values in children and adults comparing CRS grades (0-3 vs 4-5). Supplementary Table 10. Time points of target cytokine assessment listed as day since CTL019 infusion, with the acceptable windows provided by the study protocols. Supplementary Table 11. Median (range) day 1-3 peak cytokine values for those with severe (grade 4-5) and without severe (grade 0-3) CRS for total study population, and split into adults and children. Supplementary Table 12. Summary of all predictive models generated by discovery cohort and validated. Supplementary Table 13. Clinical details on the validation cohort. Supplementary Table 14. Therapies used for CRS. Supplementary Table 15. Tocilizumab use by grade and age, as well as, clinical effects of tocilizumab on CRS. Supplementary Table 16. Lee CRS grading scale. Supplementary Table 17. Davila CRS grading scale. Supplementary Table 18. Reclassification of CRS severity by dividing grade 3 into 3a and 3b. Supplementary Table 19. Summary of top predictive models with alternate determination of CRS severity. Supplementary Figure 1. CRP and ferritin are poor early predictors of severe CRS. Supplementary Figure 2. Early increases in IFNγ and sgp130 after CTL019 infusion are associated with development of severe CRS. Supplementary Figure 3. Severity of CRS correlates moderately with CAR T cell expansion but early expansion of CAR T cells does not predict CRS severity. Supplementary Figure 4. Cytokine profiles predict CRS. Supplementary Figure 5. Alternate CRS prediction models.