Rapid Increase in Blood CD19 CAR-T Vector Load during the First 5 Days Post Infusion Is Associated with Severe Icans

Le résumé fourni par la source

Introduction Quantifying chimeric antigen receptor T cell (CART) vector load in blood may predict outcomes and help manage toxicity in patients receiving axicabtagene ciloleucel (axi-cel) treatment for Large B cell lymphoma(LBCL) . High concordance of the Idylla TM platform, a point-of-need, real-time assessment of CART vector load from peripheral blood, with standard droplet digital polymerase chain reaction (ddPCR) method was reported previously (98.6%). Here, we correlate CART vector load measurements with toxicities and clinical outcomes. Methods In our study of 100 subjects at Stanford, Mayo Clinic, and Vanderbilt receiving axi-cel for LBCL, blood samples were collected for ddPCR (copies per cells) and Idylla (estimated copies per uL) vector load measurement to derive pharmacokinetics at pre-lymphodepletion and on days 1, 3, 5, 7, 14, and 28 post-infusion. CART vector load was log-transformed, and data points from days 1 to 5 were linearly interpolated. Patients with fewer than 3 measurements were excluded. A slope representing CART vector load change was calculated with the intercept set to zero using linear models. Predictive analysis used LASSO to assess severe toxicity risk and identify associated key features. Toxicities were graded using ASTCT consensus grading. Progression-free survival (PFS) was analyzed using Kaplan-Meier methods, stratifying patients into high and low groups based on median slope during the first five days post-infusion, with log-rank tests comparing PFS. Results 602 samples were collected across 100 patients. The slopes of change in CAR-T vector load for both Idylla and ddPCR were concordant and showed statistically significant differences in patients with and without severe ICANS (Grade3+), p-values of 0.008 and 0.004, respectively. (Fig 1) Predictive LASSO model achieved a median area under the receiver operating characteristic curve of 0.74 on unseen test sets for predicting severe ICANS. Key predictive features with increased severe ICANS risk, were Day 3 Idylla measurements, ferritin , and LDH elevation. In contrast, prior lines of therapy and hemoglobin levels showed negative coefficients. There was a significant difference in PFS between patients with high and low CAR-T vector load slopes for both Idylla and ddPCR (p < 0.03 and p < 0.06, respectively) (Fig 2). Discussion Standard PCR and FACS methods require technical laboratory skills and have multi-day turnaround times. The Idylla platform offers rapid, automated CART vector load detection in about 90 minutes with only 2 minutes of hands-on time. Our results show that rapid increase in blood CD19 CART vector load during the first 5 days is associated with severe ICANS and is associated with better PFS. Day 3 CART vector load measurement predicts risk of severe ICANS. This suggests that point of need measurement with the Idylla platform could improve toxicity management and PFS in CART therapy .

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.