ALBERT

Description: Introduction: Liso-cel is an investigational, anti-CD19, defined composition (4-1BB) chimeric antigen receptor (CAR) T cell product administered at a target dose of CD4+ and CD8+ CAR T cells. Liso-cel manufacturing process design includes controls that minimize between-lot variability, enabling robust CAR T cell generation across heterogeneous patient populations and disease indications. Characterization of liso-cel includes measurements of cell health, phenotype, and function. To demonstrate the robustness of the manufacturing process for which a contributor of variation is variability in incoming patient material, we developed a statistical method leveraging canonical correlation analysis (CCA) and lasso regression for predicting CAR T cell composition from measurements of cell health and phenotype in incoming patient T cells. These methods may also improve our understanding of donor variability effects on CAR T cell quality. Methods: CAR T cells were manufactured from autologous leukapheresis material in the TRANSCEND NHL 001 (NCT02631044) clinical trial. CCA and lasso models were constructed from 34 starting material attributes and 101 CD4 and CD8 clinical drug product attributes from 119 patients. CCA was implemented using prospective meta-analysis and telefit packages, and lasso regression was implemented using the glmnet package, both in R v3.5. Predictive accuracy was assessed for both methods using ten-fold cross validation. Results: CCA simultaneously found linear combinations of incoming patient T cell attributes and linear combinations of drug product attributes such that their correlation was maximized with an option of evoking a sparsity "penalty" to reduce model complexity by down-weighting (regularizing) attributes with small, independent effects. This approach enabled us to identify "meta-features" of primary components of incoming T cells strongly correlated with those of CAR T cells. Meta-feature 1 indicated that proportions of naïve CD4 T cells in starting T cell material were highly correlated with frequencies of naïve-like CD4 and CD8 CAR T cells post manufacturing (Figure 1). Meta-feature 2 revealed that naïve and central memory CD4 and CD8 T cell proportions in starting materials were correlated with naïve and central memory CD8 CAR T cells. Meta-feature 3 indicated that effector CD4 T cell proportions measured phenotypically in starting material were correlated with CD4 and CD8 CAR T cell effector functions, including antigen-specific cytokine production. Lastly, meta-feature 4 suggested that effector CD8 T cell proportions in starting material were correlated with CD8 CAR T cell effector functions. Because penalized CCA identified primary components of features correlated between incoming patient T cell material and manufactured CAR T cells, it can predict multiple attributes simultaneously, but with reduced capacity to most effectively predict a single attribute of interest. Hence, we implemented the lasso regression method that performs both variable selection and regularization to enhance the predictive accuracy of single attributes one at a time. Lasso regression models predict subsets of CAR T cell attributes more accurately than CCA and identify which starting T cell attributes are most important for prediction at the expense of having less power for predicting drug product attributes with limited relevant individual features in starting material. CCA achieved prediction accuracies up to an R2 of 42% for predicting CD4+ CAR+ naïve-like T cells (P=0.008), whereas lasso regression achieved up to an R2 of 67% for the same CAR T cell attribute (P=6×10-275). Both methods perform best at predicting classically naïve and TEMRA T cell compositions. Using CCA and lasso, we achieved nominally significant predictions for 53 of the 101 CAR T cell attributes using only 34 starting material attributes as input; the residual variation in the CAR T cell attributes independent of starting material composition was likely due to other patient or process variables. Conclusion: The application of statistical learning approaches to CAR T cell characterization data can enable us to predict CAR T cell characteristics that are directly related to donor-to-donor variability in incoming T cell material. These methods may allow us to develop adaptive manufacturing processes to improve treatment outcomes of autologous cellular therapies. Disclosures Jiang: Juno Therapeutics, a Celgene Company: Employment, Equity Ownership. Mashadi-Hossein:Celgene Corporation: Employment, Equity Ownership. Yost:Juno Therapeutics, a Celgene Company: Employment, Equity Ownership. Teoh:Juno Therapeutics, a Celgene Company: Employment, Equity Ownership. Larson:Juno Therapeutics, a Celgene Company: Employment, Equity Ownership. Hause:Juno Therapeutics, a Celgene Company: Employment, Equity Ownership.

Description: †These authors contributed equally. Introduction Ide-cel, an anti-BCMA CAR T cell therapy, has demonstrated promising efficacy in the phase I CRB-401 trial in relapsed/refractory multiple myeloma (MM) (objective response rate, 85%; median progression-free survival [PFS], 11.8 months [95% CI: 6.2, 17.8]), but a subset of patients failed to respond and the duration of response varied across patients (Raje et al, N Engl J Med. 2019). A systematic examination of patient, product, and post-infusion correlates of overall and long-term response could offer biological insight into heterogeneous efficacy as well as provide biomarkers to guide post-CAR T disease management and future CAR T manufacturing and patient enrollment efforts. Soluble BCMA was of particular interest due to its expression on malignant and healthy plasma cells and its role as a composite measure of disease burden in MM. Methods We performed a retrospective analysis of 33 patients from the phase I CRB-401 study of ide-cel. The concentrations of ten immune-related factors in the blood (GMCSF, IFN-γ, IL-10, IL-1b, IL-2, IL-6, IL-8, MCP-1, TNF-α) and soluble BCMA were measured by ELISA before and after infusion with ide-cel along with 290 ide-cel CAR T-cell drug product attributes measured by flow cytometry and immunoassays. The absolute concentrations and fold-changes from baseline were assessed for correlation with overall and long-term response using univariate and multivariate (random forests) approaches. Results Several CAR T-cell drug product covariates nominally associated with longer PFS included reduced senescence phenotype in CD4 CAR T-cells and increased IL-2 and TNF-α production (P 〈 0.05). Pre-infusion levels of soluble BCMA correlated with serum monoclonal protein (M-protein) levels in 20 of 33 patients for whom M-protein levels were measurable (ρ = .49; P = 0.03) and with concentrations of the involved free light chain (ρ = .59; P = 0.005) in 23 of 33 patients with measurable levels. Our investigation of soluble BCMA levels in patients achieving a partial response (PR) or better confirmed significant decreases in soluble BCMA levels relative to nonresponders (NR) as early as seven days post-infusion (median reduction of 50% for ≥ PR vs. median increase of 27% for NR, P = 0.02). The fold-change in soluble BCMA 1 month after infusion stratified patients who achieved a PR or better from those who did not (P = 0.0001). Notably, patients who maintained a response to ide-cel for ≥ 18 months (M18 R) experienced a greater depth of clearance of soluble BCMA at month 2 (median concentration of 1835 ng/L for M18 R vs. 6299 ng/L for M18 NR, P = 0.002). The induction of IL-6 and TNF-α in blood on days 1-9 post-infusion was also higher in patients with a PR or better in response to ide-cel (e.g. IL-6 median fold change increase at day 2 of 2.9 for ≥ PR vs. 0.7 for NR, P = 0.001), consistent with an active inflammatory response and higher levels of CAR T expansion. Conclusions These data from CRB-401 identify candidate drug product attributes and soluble factors that correlate with response to ide-cel and potentially MM-directed cellular therapies in general. These data suggest that changes in soluble BCMA may be a robust biomarker of both early and durable responses to ide-cel and the depth of clearance of soluble BCMA at 2 months post-infusion may identify patients at risk of progression before standard markers of myeloma progression have emerged. Further molecular characterization of drug product attributes, including CyTOF and RNA sequencing, is ongoing to identify additional biomarkers associated with clinical outcomes following ide-cel treatment. These data will help inform future strategies to improve the efficacy of ide-cel and validation in a larger cohort is ongoing. Disclosures Thompson: Celgene Corporation: Employment, Equity Ownership. Jiang:Juno Therapeutics, a Celgene Company: Employment, Equity Ownership. Campbell:Celgene Corporation: Employment, Equity Ownership. Fuller:Celgene Corporation: Employment, Equity Ownership. Kaiser:Celgene Corporation: Employment. Mashadi-Hossein:Celgene Corporation: Employment, Equity Ownership. Rytlewski:Adaptive Biotechnologies: Equity Ownership; Juno Therapeutics, a Celgene Company: Employment, Equity Ownership. Martin:Celgene Corporation: Employment, Equity Ownership. Finney:bluebird bio Inc.: Employment. Kleinsteuber:bluebird bio Inc.: Employment, Equity Ownership. Alonzo:bluebird bio Inc.: Employment, Equity Ownership. Pandya:bluebird bio Inc.: Employment. Agarwal:Celgene Corporation: Employment, Equity Ownership. Hege:Celgene Corporation: Employment, Equity Ownership, Patents & Royalties; Arcus Biosciences: Membership on an entity's Board of Directors or advisory committees; Society for Immunotherapy of Cancer: Membership on an entity's Board of Directors or advisory committees; Mersana Therapuetics: Membership on an entity's Board of Directors or advisory committees. Raje:Merck: Consultancy; Takeda: Consultancy; Janssen: Consultancy; Celgene Corporation: Consultancy; Amgen Inc.: Consultancy; Bristol-Myers Squibb: Consultancy. Munshi:Celgene: Consultancy; Oncopep: Consultancy; Amgen: Consultancy; Janssen: Consultancy; Takeda: Consultancy; Abbvie: Consultancy; Adaptive: Consultancy. Hause:Juno Therapeutics, a Celgene Company: Employment, Equity Ownership. OffLabel Disclosure: ide-cel /bb2121 is an investigational agent and not yet approved in the US

Description: Introduction: Identifying prior therapy exposures that affect the patient or their peripheral blood mononuclear cell (PBMC) material is one strategy to optimize outcomes to CAR T cell therapy. Alkylating agents commonly used in multiple myeloma management, such as cyclophosphamide, have been reported to impair the proliferative capacity of T lymphocytes and to blunt their functional activity (Ercolini et al. J Exp Med. 2005;201:1591; Banissi et al. Cancer Immunol Immunother. 2009;58:1627; Litterman et al. J Immunol. 2013;190:6259). In the pivotal phase 2 KarMMa trial (NCT03361748) investigating the BCMA-directed CAR T cell therapy idecabtagene vicleucel (ide-cel, bb2121) in triple-class exposed patients with RRMM, 80% of patients had a history of prior anticancer treatment with ≥1 alkylating agents. In this retrospective analysis, patient and PBMC characteristics associated with time from last dose of alkylating agent(s) until apheresis of PBMCs for CAR T cell manufacture were identified. Methods: PBMCs isolated from patient apheresis material, which serves as starting material for CAR T cell manufacturing, were immunophenotyped by polychromatic flow cytometry for markers associated with T cell differentiation, memory, senescence, and exhaustion. Data from relevant prespecified clinical and exploratory endpoints were collected, and a novel implementation of left-censored time-to-event analysis (Ware et al. Biometrics. 1976;32:459) was used to identify statistically significant relationships between washout time after prior alkylator exposure (encompassing 14 individual drugs) and patient and PBMC variables. Dose intensity of prior alkylators was not considered due to sparse annotations in the patient histories. Optimal cutpoints were identified for each variable that maximized the proportional hazard of receiving an alkylator between patients above and below the cutpoint, and P values were adjusted for testing multiple cutpoints. Relationships were verified by nonparametric correlation, in which alkylator washout was encoded as 1/log(−washout). Results: More recent exposure to an alkylating agent (after diagnosis but before apheresis) was associated with patients receiving more prior therapies per year to manage their disease (hazard ratio [HR]=2.63, ρ=−0.54, P