Efficacy and Toxicity of CD19-Specific Chimeric Antigen Receptor T Cells Alone or in Combination with Ibrutinib for Relapsed and/or Refractory CLL

Background In relapsed/refractory (R/R) chronic lymphocytic leukemia (CLL) patients (pts) who had previously failed ibrutinib, we observed durable responses after CD19 chimeric antigen receptor-modified T-cell immunotherapy (Turtle CJ et al, JCO, 2017; NCT01865617). The combination of CD19 CAR-T cells with ibrutinib might improve response and decrease toxicity. Objectives To evaluate the efficacy and toxicity of CD19 CAR-T cell immunotherapy alone or in combination with ibrutinib for R/R CLL. Methods Our phase 1/2 study of CD19 CAR-T cell immunotherapy established a preferred regimen for R/R CLL pts of cyclophosphamide and fludarabine (Cy/Flu) lymphodepletion followed by 2 × 10 6 CD19 CAR-T cells/kg (Turtle, JCO, 2017). We then compared outcomes of these pts (No-ibr cohort) with a subsequent cohort receiving Cy/Flu with 2 × 10 6 /kg JCAR014 CAR-T cells with concurrent ibrutinib (420 mg/d) from at least 2 weeks prior to leukapheresis until at least 3 months after JCAR014 infusion (Ibr cohort). Results We treated 19 and 17 pts in the No-Ibr and Ibr cohorts, respectively. Pt characteristics were comparable (Table 1). Median follow-up in responders was 764 and 98 days in the No-ibr and Ibr cohorts, respectively. Administration of ibrutinib with Cy/Flu and CD19 CAR-T cells was well tolerated in most pts; ibrutinib was reduced or discontinued in 6 pts (35%) at a median of 21 days after JCAR014 infusion. In the Ibr cohort 1 pt with grade 2 CRS (Lee, Blood, 2014) developed fatal presumed cardiac arrhythmia and 1 pt developed a subdural hematoma in the setting of trauma and thrombocytopenia. Although the proportions of pts with grade ≥1 CRS were similar between cohorts (76% vs 89%, P = 0.39), the severity of CRS (grade ≥3 CRS: Ibr, 0%; No-Ibr, 26%; P = 0.05) and serum peak IL-8 (P = 0.04), IL-15 (P = 0.003) and MCP-1 (P = 0.004) concentrations were lower in the Ibr cohort. However, we found higher peak CD4+ CAR-T cell counts (P = 0.06) and comparable peak CD8+ CAR-T cell counts (P = 0.29) in blood in the Ibr compared to the No-Ibr cohort. Sixteen pts (94%) and 18 pts (95%) in the Ibr and No-ibr cohorts, respectively, were evaluable for response. We observed a higher proportion of responders (complete response and partial response) by 2008 IWCLL criteria in the Ibr compared to the No-ibr cohort (88% vs 56%, respectively, P = 0.06). In pts with no disease by BM flow cytometry after treatment (N = 23), a higher proportion of pts in the Ibr cohort had no malignant IgH sequences at 4 weeks (83% vs 60%, P = 0.35). Using multivariable logistic regression (Table 2), treatment in the Ibr cohort and a lower pre-treatment SUVmax were associated with a higher probability of response (Ibr cohort, OR = 14.02, P = 0.05; SUVmax, OR = 1.31 per SUV unit decrease, P Conclusion The combination of ibrutinib with CD19 CAR-T cell immunotherapy was well tolerated in most pts, and might decrease the incidence of severe CRS and improve responses in pts with R/R CLL.