Digital PCR Assay for Evaluation of CD19-CAR-T Cells In Vivo Expansion in Aggressive B-Cell Lymphoma: Impact on Clinical Outcome

Background: Chimeric antigen receptor (CAR)-T cells therapy has revolutionized hematological treatment mainly in non-Hodgkin lymphoma (NHL) demonstrating unprecedented clinical responses when directed against CD19 combining the antigen-binding properties of antibodies with the effector functions of T cells. In particular, CAR-T cell peak concentrations and area under the curve in the first 28 days after Axicabtagene Ciloleucel (Axi-cel) infusion have been associated with long-term efficacy. Based on a digital PCR assay for sensitive CAR-T cells detection, we aimed to detect the CAR-T product in patients (pts) and associate its kinetics with patient outcome. Methods: A retrospective analysis was conducted involving 28 patients diagnosed with NHL treated with Axi-cel CAR-T products (2019-2023) in the Department of Hematology of the Clinical University Hospital of Valencia, Spain according to the Declaration of Helsinki ethical criteria. 97 peripheral blood mononuclear cell (PBMC) DNA samples were extracted from 3-12 ml EDTA tubes. We performed serial measurements of CAR-T-cell blood levels (7,14, 28 days post-infusion) using a newly developed axi-cel-specific dPCR assay (Bio-Rad,USA) with QIAcuity Digital PCR System and QIAcuitySoftware for data analysis (QIAGEN, Germany). The peak CAR-T-cell blood concentrations (CAR-T-Cmax) and the cumulative CAR-T-cell levels over the first 28 days in PB by area under the curve (AUC CAR-T) were calculated based on CAR-T dPCR values (copies/μL). Response assessment was performed per institutional practice and based on Lugano Criteria. Statistical analyses were performed using SPSS Software (IBM, Germany, v.22). Logistic regression was used to evaluate the relationship between covariate and outcome. Kruskal-Wallis and Dunn's tests were used when 3 subgroups were involved. Survival analyses were conducted according to CAR-T dPCR values. Results: A total of 28 pts (16 [57.1%] females) who received CAR-T cell therapy were included [25 (89%) DLBCL; 3 (11%) PMBCL]. The patient's best overall response assignment (partial or complete) were observed in twenty-two (78,4%) named as responders (R), otherwise six (21,4%) nonresponders (NR) around day 30 of evaluation. CAR-T-CMax ranged from 2.30 to 176.10 (median, 40.16) CAR-T copies/μL, 7 days later of majority expansion day post infusion: R pts had a higher peak expansion of 66.38 copies/µL compared to 32.43 copies/µL although this difference did not reach statistical significance ( p=018). Expansión was faster in R pts (median 7.64 days pot-infusion) versus NR pts (median 12.83, p=0,10) . Quantitative 7-d CAR-T copies/μL, and AUC value from days 0 to 7, showed statistical significance differences post-infusion between R and NR (52.08±32.48 vs 15.51±20.51, p<0.005; 3603.88±2520.02 vs 1291.74±1948.77, p<0.037). As for response assessment, 7-d CAR-T values were able to successfully predict the development of NR pts with an area under the curve of 0.841 (CI 95% 0.66-1.00), a sensitivity of 0.950 and a specificity of 0.625 (Fig 1.). Eleven pts (39.3%) died at a median of 3.6 months (1.7-6.1 months) post-infusion. For survival analyses, a cut-off value of 18.35 7-d CAR-T copies/μL that maximized the Log-Rank statistic, was selected to divide pts into R and NR categories. Median overall survival (OS) was 23.9 [16.3-NA, 95% CI] months and 13.2 [3.30-3.96, 95% CI] months for R and NR according to CAR-T copies/μL 7 days, respectively (Fig.2). Conclusions: Digital PCR for CAR-T evaluation shows that quantitative CAR-T copies/μL could be a potential predictor of short-term response and survival at first 7 days post-infusion value in pts with DLBCL or PMBCL treated with CAR-T cell therapy.