Cellular Kinetics and Anti-Therapeutic Antibody in Relapsed/Refractory B-NHL Patients Treated with JWCAR029

Background:

JWCAR029 is a CD19-directed 4-1 BB CAR T cell product, of which CD4 and CD8 CAR T cells are produced together and transfused in non-fixed ratio. A phase I, single-arm, open label study was conducted to evaluate the safety and efficacy of JWCAR029 in patients (pts) with relapsed or refractory B-NHL. Previously, preliminary data in six pts (Yan et al, Blood 2018 132:4187) showed high response rates and favorable safety profiles of JWCAR029. Herein, we presented the data of the Phase I trial of JWCAR029 (NCT03344367 and NCT03355859) in 29 pts with pharmacokinetics (PK), pharmacodynamics (PD), and anti-therapeutic antibody (ATA) evaluations.

Methods:

Eligible pts received lymphodepletion, with 25mg/m² flu and 250mg/m² cy, followed by a single dose of JWCAR029 at one of four dose levels (DL1, 25×10⁶ cells; DL2, 50×10⁶ cells; DL3, 100×10⁶ cells; DL4, 150×10⁶ cells). Blood samples were collected and analyzed for PK, PD, and ATA at a central lab per protocol defined time points. The existence and duration of CAR T cells (PK) were measured by validated flow cytometry and qPCR assays. CD4 and CD8 subpopulation of CAR+ T cells were detected by cetuximab targeting EGFRt as a marker co-expressed with CAR transgene in fresh peripheral blood. In parallel, batched frozen blood samples collected from each pt were detected for integrated CAR transgene by qPCR at the same protocol defined time points. Plasma ATA against murine CD19 scFv (FMC63) was measured with a validated electrochemiluminescent (ECL) assay.

Results:

As of July 5, 2019, blood samples from 29 pts who received JWCAR029 treatment with a minimum follow-up of 6 M (median, 6 M) were evaluable in the analysis. From DL1 to DL4, median C_max, T_max and AUC_0-28 for JWCAR029 transgene detected by qPCR did not differ among dose levels (Table 1). CD4/CD8 ratio (range, 0.23-5.50) at cryopreserved drug product of JWCAR029 was not associated with best response of CR/PR at 6 M. Greater in vivo expansion was detected by both qPCR and flow cytometry in pts with best response of CR/PR than those with SD/PD at 6 M (Table 1). Higher concentration of CD8+CAR+T cells than CD4+CAR+T cells were detected in PB by flow cytometry for all treated pts (C_max median= 30.6 vs 5.64). At 3 M, 81.5% (22/27) and 48.2% (13/27) pts had detectable CD8+ and CD4+ CAR+ T cells, respectively. Of those pts with detectable CAR+ T cells at 3 M, 70% (14/20) and 35% (7/20) had detectable CD8+ and CD4+ CAR+ cells at 6 M, respectively. Significantly higher C_max and AUC_0-28 were observed in patients with ≥ Grade 1 CRS (C_max median= 85004 vs 16328, P<0.01; AUC median=536543 vs 141731, P<0.01). And relatively higher C_max and AUC_0-28 were found in patients with NT (C_max median= 116112 vs 40391; AUC median=711306 vs 301035).

27.5% of pts (8/29) had detectable ATA in plasma, of which 25% (2/8) pts had pre-existing antibodies before CAR T cell infusion. 6 pts developed antibodies without pre-existing antibodies and were considered treatment-induced. The median time for treatment-induced antibody development was 6 M (range, 3-12). Increasing level of antibodies were detected at median time of 6 M (range, 6-6) for pts who had pre-existing antibodies and were considered treatment-boosted. No significant differences in PK profiles of JWCAR029 transgene levels were found between ATA negative group and treatment-induced ATA positive group (C_max median= 44497 vs 50032; AUC median= 420635 vs 313654; Fig.1). Although the sample size of the treatment-boosted subgroup was small, there was a trend for lower expansion of CAR T cells in pts who had pre-existing ATA than pts who did not develop ATA (C_max median= 3051 vs 44497; AUC median = 16437 vs 420635;Fig.1). In ATA positive subgroup, 100% (8/8) pts responded with CR rate of 75% (6/8). 6 M-response rate was 65.5% (5/8) for ATA positive subgroup and 57.1% (12/21) for ATA negative subgroup. Incidence of ≥ Grade 1 CRS or NT was indistinguishable between ATA positive and negative subgroups, 50% (4/8) in ATA positive vs 57.1% (12/21) negative.

Conclusion:

Preliminary data from JWCAR029 Phase I study has demonstrated that pts with best response of CR/PR at 6 months had a relatively higher CAR T cell expansion. Current data suggested that the prevalence of pre-existing ATA may compromise CAR+T PK profile. No association of the presence or boost of ATA with efficacy or safety of JWCAR029 was observed. Further exploration of ATA and clinical outcomes will be studied in the ongoing pivotal Phase 2 study in 70 pts with B-NHL.

View largeDownload slide

View largeDownload slide

Close modal