Factors associated with manufacturing failure of commercial CD19 CAR-T cell products for large b cell lymphoma (LBCL).

Abstract

7044 Background: As CD19 chimeric antigen receptor T-cell (CAR-T) therapy has shown curative potential in patients (pts) with relapsed/refractory (R/R) LBCL, indications for treatment are expanding. The reported manufacturing failure rate was 1-13% in clinical trials, but there is poor understanding of out-of-specification (OOS) products and CAR-T products on 2nd apheresis. This study aims to characterize factors associated with successful manufacturing. Methods: This study includes pts with R/R LBCL who underwent apheresis for CAR-T product between 12/2017 and 2/2023. Baseline variables were obtained at time of apheresis. OOS product is defined as not meeting eligibility for a commercial product but suitable for infusion on expanded access protocol. 2nd apheresis was performed due to initial manufacture failure. Progression-free survival (PFS) is the time from CAR-T conditioning to disease progression or death, whichever occurs first. Overall survival (OS) is defined as time from conditioning to death or last follow-up. A chi-square test or Fisher’s exact test was used for comparison of 2 categorical variables. Results: 319 pts underwent apheresis; 261(78%) were axicabtagene ciloleucel, 26(8%) were tisagenlecleucel and 47(14%) were lisocabtagene maraleucel 284(85%); products were in-specification (in-spec) on 1st apheresis, 8(2%) in-spec on 2nd apheresis, 27(8%) OOS, and 15(4%) manufacturing failures. Comparison of variables are in the table. OOS rates were 10/250(4%) for axicabtagene, 4/23(17.4%) for tisagenlecleucel and 13/46(28.3%) for lisocabtagene. Reason for OOS product were low viability:14, high interferon:4, internal manufacturing issue:2, CD4 lineage purity:2, high transduction:1, low transduction:2, and low vector copies:2. No association was found with clinical factors and OOS. Factors associated with OOS were advanced stage (P=0.012), IPI score >3 (p=0.001), extranodal involvement (P=0.012) and older age (p= 0.002). In-spec product on 1st apheresis had greater OS (p= 0.011) and PFS (p=0.043) than 2nd apheresis. Conclusions: Factors predictive of OOS product include older age, advanced disease, extranodal involvement and high IPI. Pts requiring 2nd apheresis had worse survival. Lisocabtagene had higher likelihood for OOS product. Outcomes for OOS products are not well described. Failure to manufacture an eligible product is a barrier for LBCL pts to receive CAR-T therapy. [Table: see text]