Introduction

Azacitidine (AZA) is a DNA methyltransferase inhibitor and cytotoxic agent with survival benefits in higher-risk MDS and AML. We have demonstrated that AZA rapidly induces apoptosis of MM cells, and induces synergistic killing when combined with lenalidomide (LEN). This, coupled with development of an oral AZA formulation (CC-486), provides rationale for pursuing this combination in MM. The cereblon [CBRN]-Ikaros/Aiolos-IRF4/c-MYC signalling pathway has been implicated in the mechanism of action of cytotoxicity for IMiD®compounds. This pathway, as well as immune markers (CD8, PD-1) may represent potential biomarkers for predicting outcomes/response to IMiD® compounds.

Aims:

In R/R myeloma patients who previously failed a LEN-containing regimen: to determine the maximum tolerated dose (MTD) of CC-486 in combination with Rd; to characterise safety/tolerability; to assess efficacy: overall response rate (ORR), progression free survival (PFS), overall survival (OS).

To correlate pre-treatment MM protein expression (CRBN, Ikaros, Aiolos, IRF4, c-MYC), as well as whole marrow expression of CD8/PD-1 with response and survival (PFS/OS) in patients receiving CC-489 plus Rd.

Methods:

Phase Ib, single centre, 3 x 3 dose escalation study. LEN 25mg d1-21 and dexamethasone 40mg d1, 8, 15, 22 of 28 day cycle were combined with escalating doses of CC-489: initial dose 100mg for d1-14, increasing by either 7 days or 50mg/cohort, to a maximum of 200mg d1-21. Dose limiting toxicity (DLT) was assessed during cycle 1. Treatment continued until toxicity/progression.

Single- and dual-IHC assays were performed on pre-treatment trephines (Single: CD8, PD-1; Dual: CD138 paired with CRBN, Aiolos, Ikaros, IRF4, c-MYC). In samples with >10% CD138, presence and sub-cellular localisation of each stain was documented, and given an H score: product of % of tumour positive and intensity of staining (0-3). Single CD8 and PD-1 IHC assays were reported as positive cells/mm2. Samples were grouped according to quartiles (Group1: lower quartile, Group2: mid-range, Group3: upper quartile) and analysis was performed using SAS statistical platform v9.4.

Results:

22 patients commenced therapy (F=10, M=12), median age 67yrs (50-82yrs). Median prior lines of therapy: 5 (2-8), including 16 ASCT and 2 prior allogeneic transplant. All had failed LEN (R/R=18, primary refractory (PrimR)=4), 10/22 received and failed pomalidomide (POM) (R/R=4, PrimR=6). All had received bortezomib (R/R=10, PrimR=10), 18/22 were both bortezomib and LEN refractory.

CC-486 dose reached was 200mg d1-21, with no DLTs observed at time of reporting.

One patient died due to unrelated causes prior to end cycle 1, therefore was not evaluable for response. ORR (≥PR) was 43% (9/21): 8 PR, 1 VGPR. Of the remaining patients, 2 achieved MR, 5 SD, and 5 PD. (clinical benefit rate (≥MR) = 52%). Median time to best response in patients with ≥MR: 2.5m (1-3.7m). Responses were seen in cohorts: [100mg d1-14 (3PR=3), 100mg d1-21 (VGPR=1, PR=3), 150mg d1-21 (PR=2, MR=2)]. Median time on study: 3m; responders (≥MR): 6.3m (2.5-15m), non-responders: 1.7m (0.9-8m). Median PFS 3m, median OS 15m. One patient remains on study.

3/6 patients treated with LEN in prior 1-2 treatment lines responded (PR=2, VGPR=1), 1/6 had SD. 1/10 patients treated with POM in prior 1-2 treatment lines responded (PR), with 2 achieving MR and 3 SD.

Patients with a lower expression of cMYC and IRF4 had superior PFS compared with patients with higher expression (cMYC: Group1 vs Group3 p=0.052; IRF4: Group1 vs Group2 p=0.04). Patients with lower numbers of CD8+ T-cells had better PFS than those with higher (p=0.069). There was no association between degree of expression of CBRN, Ikaros, Aiolos or PD-1 and survival (PFS/OS).

As CRBN expression increased, patients were more likely to respond (≥PR) (p=0.07) and patients with low PD-1 expression were more likely to respond (≥ MR) (Group1 vs Group3 p=0.05).

Conclusion

CC-489 combined with Rd is well tolerated and effective with durable responses in a subset of heavily pre-treated R/R MM patients, including those who recently failed IMiD® compound therapy, suggesting that CC-489 may overcome drug resistance. IHC may have utility in identifying subsets of patients more likely to respond to CC-489 and Rd (CRBN, PD-1) and predict survival (cMYC, IRF4 and CD8). This trial has been expanded to include other sites and less heavily pre-treated patients.

Disclosures

Thakurta:Celgene: Employment, Equity Ownership. Wang:Celgene: Employment. Guzman:Celgene: Employment. Cuoto:Celgene: Employment. Ren:Celgene: Employment, Equity Ownership.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution