CD30-Chimeric Antigen Receptor (CAR) T Cells for Therapy of Hodgkin Lymphoma (HL)

Almost all HL and some NHL express the CD30 antigen, and monoclonal antibodies (mAb) targeting CD30 (e.g. brentuximab) produce objective antitumor responses. However, mAb have limited bio-distribution and their benefits may be short-lived. We therefore expressed the antigen binding domain of a CD30 mAb as part of a chimeric antigen receptor (CAR) on T cells, coupled to the CD28 and ζ chain endodomains. We have previously published results of a phase 1 study of autologous CD30.CAR-T cells (CD30.CARTs) infused in patients with relapsed/refractory CD30 + HL or NHL without preceding cytoreductive chemotherapy (Ramos et al. , J Clin Invest 2017). Of 6 patients with relapsed active HL, 1 entered complete remission (CR) that has lasted more than 3 years, and 3 had transient stable disease. We have now incorporated cytoreductive chemotherapy (cyclophosphamide and fludarabine) prior to CD30 CAR T infusion to discover if we can enhance the in vivo expansion and anti-tumor activity of these CD30.CARTs, (RELY-30, NCT02917083). Our preliminary results suggest a substantial improvement in efficacy. We have manufactured CD30.CARTs for 18 patients using retroviral transduction. Cell products comprise >98% T cells with a final transduction efficiency of 97.6%±1.8%. Ten relapsed/refractory HL patients have received these CD30.CARTs under the RELY-30 trial, 7 of whom had relapsed or progressed after treatment with brentuximab. Three patients have been treated on dose level (DL) 1 (2 × 10 7 CD30.CAR + T cells/m 2 ), 6 on DL2 (1 × 10 8 ) and 1 on DL3 (2 × 10 8 ). All patients received preceding lymphodepleting chemotherapy (cyclophosphamide 500 mg/m 2 and fludarabine 30 mg/m 2 daily for 3 days). CART infusions were associated with grade 1 cytokine release syndrome in 4 of the patients, and a transient maculopapular rash in 6 of the patients, starting approximately one week after administration. The molecular signal from CARTs, assessed by Q-PCR in peripheral blood, peaked at 1-2 weeks following infusion, but dropped progressively after 4 weeks and decreased to the limit of detection by 6 months. The signal level was dose dependent, with a peak average of 19,371 copies/mg DNA in patients treated on DL2 versus 7,132 copies/mg for DL1; this expansion was 45 and 119-fold higher, respectively, than patients receiving the same CART dose without prior cytoreduction in our previous trial. Nine patients have been evaluated at 6 weeks after infusion. Six have had a CR lasting up to >9 months, while 3 patients had disease progression. Hence, infusion of CD30.CARTs after cytoreductive chemotherapy is well tolerated at the doses used. Inclusion of cytoreduction pre-infusion substantially improves CD30.CART expansion and appears associated with superior anti-tumor activity in relapsed patients (6/9 CR versus 1/6 CR, P  = 0.04).