Safety and Efficacy of Letetresgene Autoleucel (lete-cel; GSK3377794) Alone or in Combination with Pembrolizumab in Relapsed and Refractory Multiple Myeloma

Background: Outcomes remain poor for patients with relapsed and refractory multiple myeloma (RRMM) progressing on conventional therapy (proteasome inhibitors, IMiDs, anti-CD38 antibodies, BCMA targeting agents, corticosteroids). Lete-cel, an autologous T-cell therapy, targets NY-ESO-1/LAGE-1a+ tumors using a genetically modified, high-affinity T-cell receptor. NY-ESO-1 and LAGE-1a are immunogenic cancer/testis antigens frequently overexpressed in MM and are linked to poor clinical outcomes. Additionally, PD-1 expression, which may limit adaptive immune response, has been observed previously in RRMM patients following treatment with lete-cel (Stadtmauer et al. Blood Adv, 2019; 3: 2022-2034).

This open-label, pilot study evaluated the safety and efficacy of lete-cel +/- pembrolizumab (pembro) in patients with RRMM.

Study design and methods: Key eligibility criteria include: age ≥18 yr; HLA-A*02:01; A*02:05, and/or A*02:06; NY-ESO-1+ and/or LAGE-1a+; protocol-required prior regimens; specified washouts from prior therapy; no active/chronic/intercurrent illness. Following lymphodepletion (LD), patients received lete-cel (Arm 1) or lete-cel + pembro (Arm 2). Planned pembro dosing was 200 mg/dose Q3 weeks (wks) starting at Wk 3. Primary endpoint was safety and tolerability. Key efficacy endpoint was investigator-assessed overall response rate (ORR) by International Myeloma Working Group uniform response criteria for MM (2016); response was assessed Q3 wks from Wk 3 to Wk 24, then Q6 wks to Wk 72, then every 3 months (mo) to disease progression/death/withdrawal. NY-ESO-1/LAGE-1a expression was assessed by qRT-PCR on myeloma cells. Transduced cell kinetics were assessed by qPCR of transgene vector copies in DNA from peripheral blood mononuclear cells.

Results: Six patients (all male; median age 63 yr) were enrolled; 3 per arm. All had prior systemic anti-cancer therapy; 3 patients had received ≥5 prior regimens. Five of 6 patients received systemic anti-cancer therapy between leukapheresis and LD. All received reduced LD due to age and, in some patients, renal impairment. Patients in each arm received similar numbers of transduced T cells. Each of the 3 patients in Arm 2 received a median of 3 pembro doses (range: 3-4 doses). Start of pembro dosing was delayed to Wk 6 in 2 patients due to ongoing toxicities.

There were no Grade 5 AEs. Grade 3/4 T-cell related AEs were reported in 3 patients, and all patients had Grade 3/4 LD-related AEs. Hematopoietic cytopenias were the most common treatment-emergent and treatment-related Grade 3/4 AE, occurring in all patients. All cytopenias were reported to have resolved for 4 patients or to have improved to Grade 1 at final patient follow-up for 2 patients. Three patients had cytokine release syndrome (Arm 1: 1 patient, Grade 2; Arm 2: 2 patients; 1 Grade 1 and 1 Grade 2); all patients recovered. There was 1 event of graft vs host disease (GvHD skin; Grade 1) and, in a separate patient, 1 event of Immune Effector Cell-Associated Neurotoxicity Syndrome (ICANS) (Grade 1). Both events resolved.

All patients had reduction in tumor burden. Arm 1 (lete-cel alone) had an ORR of 33.3% (1 CR, 2 SD) and median progression-free survival (PFS) of 2.79 mo, while Arm 2 (combination) had an ORR of 66.7% (1 VGPR, 1 PR, 1 SD) and median PFS of 2.78 mo. Time to response for all responders was 3 weeks. Pembro dosing for the 2 Arm 2 responders began at Wk 6. Duration of response in each responder was 2.1 mo. Overall survival data are not mature. Two of 3 responders exhibited clearance of antigen positive myeloma cells in the bone marrow for up to 6 weeks after lete-cel infusion. T cell kinetics trended toward higher peak expansion (Cmax) and area under the curve (AUC) over the first 28 days post-dose (AUC0-28d) in responders vs. non-responders. Serum cytokine profiles in relation to response and CRS will be discussed.

Conclusions: A single lete-cel infusion was associated with antitumor activity in 6/6 heavily pretreated RRMM patients, including 1 CR, 1 VGPR, 1PR. Both responses in Arm 2 occurred prior to pembro initiation. The associated safety profile was manageable and consistent with that seen in other lete-cel studies. Responders showed a trend toward higher Cmax and AUC0-28d as compared to non-responders. The study was closed in November 2020 due to protracted enrollment. This study (208470; NCT03168438) was funded by GlaxoSmithKline. Submission support was provided by Fishawack Health.