Introduction

Axi-cel is an autologous anti-CD19 CAR T-cell therapy approved by the US FDA 10/18/2017, for the treatment of adults with relapsed or refractory (r/r) large B-cell lymphoma (DLBCL), primary mediastinal large B-cell lymphoma (PMBCL), transformed lymphoma (tFL), and high-grade B-cell lymphoma (HGBCL) who have failed at least two prior systemic lines of therapy. In the pivotal ZUMA-1 trial, 108 patients (pts) with r/r DLBCL were treated with axi-cel: the best overall response rate (ORR) was 82% and complete response (CR) rate was 58%. At a median follow-up of 15.4 months, 42% of the pts had ongoing remission (Neelapu and Locke et al. NEJM 2017). Grade 3 or higher cytokine release syndrome (CRS) by Lee criteria and neurologic events (NEs) occurred in 13% and 31% of the pts, respectively. Here, we evaluated the real world outcomes of pts treated with standard of care axi-cel under the commercial FDA label.

Methods and Results

Seventeen US academic centers contributed data to this effort independently of the manufacturer. As of 6/30/2018, 211 pts were leukapheresed with intention to manufacture commercial axi-cel. Of these, 165 (78%) pts completed axi-cel infusion as of 6/30/18 and a further 23 (11%) pts are scheduled for axi-cel infusion in July 2018. Of the 23 remaining pts, 7 (3%) received axi-cel therapy on ZUMA-9 expanded access trial (NCT03153462) due to non-conforming cell therapy product, 15 (7%) pts died before axi-cel infusion (14 from lymphoma progression and 1 from sepsis) and 1 (1%) patient attained CR from bridging therapy and was not infused.

Safety was evaluable in 163 pts receiving axi-cel. Grade ³3 CRS and NEs occurred in 7% and 31% of pts. Tocilizumab was administered in 62% of pts and 57% received corticosteroids. Outside of lymphoma progression, 3 deaths occurred post-axi-cel; 1 due to HLH, 1 due to systemic candidiasis, and a third due to septic shock. There were no grade 5 NEs observed.

Response assessment was done for pts infused with axi-cel and who were re-staged at day 30 and/or day 100, or were deemed to have clinical progression. Of 112 pts evaluable at day 30, ORR was 79% with 50% CR, 29% PR, 6% SD and 14% with PD. Of 39 pts evaluable at day 100, 59% of pts had ongoing response (CR 49%, PR 10%).

At the time of abstract submission, more detailed patient characteristics and treatment course data were available in 134/165 pts infused. Median age was 59 (range 21-82) with 57% male. Performance status (PS) was ECOG 0-1 (81%), ECOG 2 (16%) and ECOG 3 (3%). By histology, 61% of pts had DLBCL including HGBCL, 31% had tFL and 8% PMBCL. Thirty-one percent had a prior autologous stem cell transplant. Bridging therapy between apheresis and infusion was given in 56% of patients, the majority of which consisted of chemotherapy. Sixty-six of 134 (49%) would not have met eligibility criteria for ZUMA-1 at the time of leukapheresis. Common criteria that would have made these patients ineligible for ZUMA-1 included ECOG PS >1 (n = 22), platelets <75k (n = 17), GFR <60 (n =12), active DVT/PE (n = 13), liver enzyme abnormalities (n = 10), a history of CNS lymphoma (n = 8), recent checkpoint inhibitor therapy (n = 7), ejection fraction <50% (n = 4), prior CD19 CAR T therapy (n = 4), and prior allogeneic transplant (n=2). Median time from leukapheresis to start of conditioning chemotherapy was 21 days and median time from leukapheresis to axi-cel infusion was 26 days. Median hospitalization period was 14 days.

Conclusions

This multicenter retrospective study …