Digital PCR Assays for Precise Quantification of CD19-CAR-T Cells after Treatment with Axicabtagene Ciloleucel

Abstract

Treatment with axicabtagene ciloleucel (Axi-cel) CD19-CAR-T (chimeric antigen receptor T) cells has been approved for refractory/relapsed diffuse large B cell lymphoma (DLBCL) and primary mediastinal large B cell lymphoma (PMBCL). Because treatment success as well as side effects might depend on CAR-T cell expansion in vivo, we aimed at developing digital PCR (dPCR) assays for detection and quantification of CAR-T cells. To this end, we cloned and sequenced the complete cDNA of the CAR construct. We designed different combinations of primers and dual-labeled hydrolysis probes located in various CAR regions. Three combinations were successfully tested on CAR-positive and -negative cells in duplex reactions with a reference gene (REF) to concomitantly assess cell numbers. All assays demonstrated excellent specificity and reproducibility with neglectable inter-assay variations. For all three assays, almost perfect correlation between the two dPCRs (Axi-cel versus REF) was observed, and the limit of detection was one single CAR-transduced cell corresponding to a sensitivity of 0.01% for 100 ng genomic DNA. After cross-validation, we used one assay to monitor Axi-cel CAR-T numbers in patients. CAR-T expansion and contraction followed the expected kinetics with median peak value of 11.2 Axi-cel CAR-T cells/μL at 11.3 days (median). Clinically, we observed only two partial responses (PRs) in the five patients with CAR-T cell peak numbers below median, whereas four of the five patients with comparatively good expansion showed clinical responses (two complete responses [CRs] and two PRs) on day 30. In conclusion, we established a novel dPCR assay for the sensitive detection of transgenic CAR-T cells, which should be very useful in the context of Axi-cel treatment.

Keywords: Axi-cel; Axicabtagene Ciloleucel; CAR monitoring; Chimeric antigen receptor (CAR); Yescarta; digital PCR (dPCR).

Graphical abstract

Figure 1

Digital PCR Assays Facilitate Sensitive and Accurate Quantification of Vector Copies (A) Perfect separation of single- and double-positive droplets in duplex PCR concomitantly detecting Axi-cel and REF sequences as exemplified for assay C. (B) Dilution series (100 ng → 10 pg) of genomic DNA isolated from T cells containing two REF and approximately two vector copies per cell demonstrate excellent correlation of both Axi-cel and REF-directed dPCRs. (C) The presence of additional genomic DNA (100 ng per probe) in the samples had no negative impact on sensitivity or correlation (right plot). Pearson coefficients (r) and corresponding p values are indicated. 10-pg samples corresponding to three REF and Axi-cel copies were consistently positive, indicating a limit of detection (LoD) of 1 copy for both dPCRs (considering Poisson distribution).

Figure 2

Axi-cel Kinetics in Patients (A) Excellent correlation of Axi-cel kinetics in patient #003 independently assessed using assays A, B, and C in duplex PCRs with REF. (B) For male patients (here: patient #003), replacement of the REF by a Y chromosome-specific dPCR in the duplex assay is possible without negative impact on results. (C and D) Individual patients show different engraftment and expansion kinetics of Axi-cel: (C) weak expanders (<11.2 Axi-cel CAR-T cells/μL) and (D) strong expanders (>11.2 CAR-positive Axi-cel T cells/μL). Negative values were set to 0.001%. LoD, limit of detection; r, Pearson coefficient; p, corresponding probability value.