Analysis of Factors Predicting Treatment Response of 254 Patients Who Received CD19-Targeted CAR-T Cell Therapy for Relapsed/Refractory (R/R) B-Cell Acute Lymphoblastic Leukemia (B-ALL)

Backgrounds Chimeric antigen receptor (CAR) T cell therapy targeting CD19 has demonstrated 80-90% complete remission (CR) rate in R/R B-ALL; however, 10% to 20% of patients have no response and the potential etiology and risk factors remain unclear. Patients and Methods From Apr. 2017 to Mar. 2019, 254 patients with R/R B-ALL received treatment at our single center with a second-generation CD 19 CAR-T from 5 companies (NCT03173417; ChiCTR-ONC-17012829; NCT02546739; ChiCTR1800016541; and NCT03671460). The targeted median infused CAR-T cell dose was 3x105/kg. For all patients, screening was undertaken for 58 types of B-ALL related gene mutations and 49 congenital immune deficiency gene mutations using next generation sequencing (NGS) method. The patient and CAR-T product characteristics were analyzed to see if any factors affected the initial CR rate. Univariate analysis comparing CR rates between subgroups were carried out using chi-squared tests. Logistic regression model was adopted for multivariate analysis. Results On D30, 90.6% (230/254) patients achieved CR, and minimal residual disease (MRD)-negative CR was 89.4% (227/254). Univariate analysis (Table 120% bone marrow (BM) blasts had a worse outcome than those with BM blasts≤20% (79.17% vs. 97.47%, p0.001). There were no differences in CR between patients age 1-14 yr and 14 yr, patients with or without extramedullary disease (EMD), or patients receiving different chemotherapies between enrollment and CAR-T cells infusion on top of fludarabine and cyclophosphamide (FC) conditioning regimen. Cytokine release syndrome (CRS) and neurotoxicity had no effect on CR rate.There was a trend towards a lower CR rate in patients who received donor-derived CAR-T versus patient-derived CAR-T products (61.54% vs. 92.12%, p = 0.081). A lower CR rate was observed in patients who received CD28 costimulatory domain versus 4-1BB CAR-T products (77.27% vs. 91.70%, p=0.053). There was no difference in CR rate between patients infused with cell dose ≥3x105/kg vs. cell dose 3x105/kg and 2x104/kg. No CR difference was observed from different manufacturing companies. Moreover, 125/254 cases had available data for analysis on subsets of CAR-T cells including central memory T cells (Tcm), naive T cells and T memory stem cells (Tscm). Each group was divided into high- or low-level groups based on the median level. The CR rate was 92.19% vs. 88.52% in high and low Tcm groups, 92.31% vs. 88.33% in high or low naive group, and 89.80% vs.90.24% in high and low Tscm groups, respectively. No difference in CR between the groups were observed. Multivariate analysis was performed to investigate factors affecting achievement of CR (Table 3). Female group had lower odds of achieving CR after CAR-T therapy compared to male group OR=0.232(95%CI 0.082-0.652). Compared to patients without TP53 mutation or only with other gene mutations, patients with TP53 mutation were more likely to have a low CR rate after CAR-T treatment OR=4.511(95%CI 1.295-16.895). Patients with BM blast 20% had lower odds of achieving CR compared to those with BM blast ≤20% OR=0.095 (95%CI 0.022-0.260). Treatment with CAR-T with CD28 as a costimulatory domain resulted lower odds of CR compared to treatment with CAR-T with 4-1BB OR=7.141 (95%CI 1.722-29.612). Patients who developed severe neurotoxicity (grade 2-4) post CAR-T had higher odds of achieving CR compared to patients with grade 0-1 neurotoxicity OR=34.796(95%CI 3.232-374.659). Conclusions In this analysis, we have observed that significant risk factors for not achieving CR after CD-19 CAR-T therapy include female gender, BM blasts more than 20%, a positive TP53 mutation, treatment with CD28 co-stimulatory domain vs 4-1BB CAR-T product, and mild as opposed to severe CAR-T related neurotoxicity. Disclosures No relevant conflicts of interest to declare.