CD5 CAR T-Cells for Treatment of Patients with Relapsed/Refractory CD5 Expressing T-Cell Lymphoma Demonstrates Safety and Anti-Tumor Activity
LaQuisa Hill, Rayne H. Rouce, Tyler S. Smith, Lina Yang, Madhuwanti Srinivasan, Huimin Zhang, Silvana Perconti, Birju Mehta, Olga Dakhova, Josalind Randall, Bambi Grilley, Helen E. Heslop, Malcolm K. Brenner, Maksim Mamonkin
Abstract
We describe a Phase I dose escalation study (NCT03081910) of autologous CD5-directed CAR T cells as treatment of relapsed or refractory (r/r) T cell malignancies. CD5 is a pan-T cell marker present in ∼85% of T cell malignancies. Establishing a CAR T platform to target neoplasms of T cell origin has been hindered by shared expression of targetable antigens on both malignant and normal T cells promoting CAR T cell fratricide. We developed a second-generation CD5-specific CAR with CD28 co-stimulation that produces minimal and transient fratricide in T cells. The study was designed to evaluate the safety and feasibility of CD5 CAR T cells in patients with r/r T-cell malignancies as a bridge to allogeneic HSCT. Here we present our findings in patients with r/r non-Hodgkin T cell lymphoma (T-NHL).We have treated 5 patients (62-71 yrs) with CD5+ r/r T-NHL on dose levels 1 and 2 (1 × 107 and 5 × 107 CAR T cells/m2). All patients were heavily pretreated, with a median of 4 (range 2 -18) prior lines of therapy. One patient had previously failed allo-HSCT. All patients were transplant-eligible with an identified donor. Patients received cytoreductive chemotherapy with Cy/Flu followed by infusion of CD5 CAR T cells. We evaluated adverse events, clinical responses, and expansion/persistence of CAR T cells post-infusion.Two patients received CD5 CAR T cells on dose level 1 and 3 on dose level 2. CAR T cells reached peak expansion in peripheral blood (PB) 1-4 weeks after infusion. CD5 CAR T cells were detected in lymph nodes in 2 patients who underwent biopsies following treatment. PB CD3+ cell numbers were transiently reduced after cytoreduction and CAR T cell infusion, but ablation was never complete. CRS occurred in 3/5 patients (all at DL2). Grade 1 CRS was observed in 2 patients, and 1 patient had Grade 2 CRS and Grade 2 neurotoxicity, which resolved after treatment with tocilizumab and supportive care, respectively. Two patients experienced prolonged cytopenias at 6 weeks, one of whom also had reactivations of CMV and BK virus requiring antiviral therapy. On disease re-evaluation 4-8 weeks post-CD5 CAR T cell infusion, 3 of 5 evaluable patients obtained an objective response (1 on DL1 and 2 on DL2).Complete responses (CR) were achieved in 1 patient with angioimmunoblastic T cell lymphoma (AITL) and 1 with peripheral T cell lymphoma (PTCL) (Figure 1). Another patient with extensive AITL was classified as a mixed response due to the appearance of a new PET-avid lesion with resolution of remaining lesions (Figure 2). This patient received a 2nd infusion, proceeded to HSCT and remains in CR 6 mons post-transplant.These results demonstrate that CD5 CAR T cells are safe and can induce clinical responses in heavily treated patients with r/r CD5+ T-NHL without inducing complete T-cell aplasia. Importantly, elimination of malignant T cells by CD5 CAR T cells may allow previously ineligible patients to proceed to HSCT. We describe a Phase I dose escalation study (NCT03081910) of autologous CD5-directed CAR T cells as treatment of relapsed or refractory (r/r) T cell malignancies. CD5 is a pan-T cell marker present in ∼85% of T cell malignancies. Establishing a CAR T platform to target neoplasms of T cell origin has been hindered by shared expression of targetable antigens on both malignant and normal T cells promoting CAR T cell fratricide. We developed a second-generation CD5-specific CAR with CD28 co-stimulation that produces minimal and transient fratricide in T cells. The study was designed to evaluate the safety and feasibility of CD5 CAR T cells in patients with r/r T-cell malignancies as a bridge to allogeneic HSCT. Here we present our findings in patients with r/r non-Hodgkin T cell lymphoma (T-NHL). We have treated 5 patients (62-71 yrs) with CD5+ r/r T-NHL on dose levels 1 and 2 (1 × 107 and 5 × 107 CAR T cells/m2). All patients were heavily pretreated, with a median of 4 (range 2 -18) prior lines of therapy. One patient had previously failed allo-HSCT. All patients were transplant-eligible with an identified donor. Patients received cytoreductive chemotherapy with Cy/Flu followed by infusion of CD5 CAR T cells. We evaluated adverse events, clinical responses, and expansion/persistence of CAR T cells post-infusion. Two patients received CD5 CAR T cells on dose level 1 and 3 on dose level 2. CAR T cells reached peak expansion in peripheral blood (PB) 1-4 weeks after infusion. CD5 CAR T cells were detected in lymph nodes in 2 patients who underwent biopsies following treatment. PB CD3+ cell numbers were transiently reduced after cytoreduction and CAR T cell infusion, but ablation was never complete. CRS occurred in 3/5 patients (all at DL2). Grade 1 CRS was observed in 2 patients, and 1 patient had Grade 2 CRS and Grade 2 neurotoxicity, which resolved after treatment with tocilizumab and supportive care, respectively. Two patients experienced prolonged cytopenias at 6 weeks, one of whom also had reactivations of CMV and BK virus requiring antiviral therapy. On disease re-evaluation 4-8 weeks post-CD5 CAR T cell infusion, 3 of 5 evaluable patients obtained an objective response (1 on DL1 and 2 on DL2). Complete responses (CR) were achieved in 1 patient with angioimmunoblastic T cell lymphoma (AITL) and 1 with peripheral T cell lymphoma (PTCL) (Figure 1). Another patient with extensive AITL was classified as a mixed response due to the appearance of a new PET-avid lesion with resolution of remaining lesions (Figure 2). This patient received a 2nd infusion, proceeded to HSCT and remains in CR 6 mons post-transplant. These results demonstrate that CD5 CAR T cells are safe and can induce clinical responses in heavily treated patients with r/r CD5+ T-NHL without inducing complete T-cell aplasia. Importantly, elimination of malignant T cells by CD5 CAR T cells may allow previously ineligible patients to proceed to HSCT. Figure 1, Figure 2.Figure 2(A) Week 4 post-treatment PET/CT showing flare vs persistent disease and (B) 8 week post-treatment PET/CT showing significant interval resolution of disease with one new FDG avid lesion in left medial thigh in patient with AITL.View Large Image Figure ViewerDownload Hi-res image Download (PPT)