Introduction: The outcome for patients who relapse after allogeneic stem cell transplantation (allo-SCT) remains extremely poor and the development of novel salvage regimens represents a major unmet clinical need. Intensive chemotherapy is not only frequently ineffective in this population but also associated with substantial toxicity which substantially limits its utilization in practice. Both azacitidine (AZA) and lenalidomide (LEN) represent promising salvage agents post allograft not only because they demonstrate significant anti-tumor activity as monotherapy in AML and MDS but also because they have the potential to augment the graft-versus-leukemia effect. AZA is well tolerated post-transplant and results in a reduced risk of graft-versus-host disease (GVHD) in animal models but is associated with only modest clinical activity in patients who relapse after allo-SCT. LEN, however, results in excessive rates of severe GVHD when administered post-transplant which has limited its assessment as a salvage therapy in this setting. Reasoning that AZA may reduce the GVHD risk associated with LEN administration we evaluated the safety and efficacy of combined LEN and AZA therapy in patients with relapsed AML post allo-SCT. Patients and Methods: 29 patients with morphological evidence of disease relapse at a median of 10 months after allo-SCT were treated with escalating doses of LEN (0 mg, 2.5mg, 5 mg, 10 mg, 15 mg, 25 mg, 35 mg) in combination with AZA (75 mg/m2) x 7 days in 28 day cycles. 24 patients were transplanted for AML and 5 for MDS using a matched sibling (n = 13) or unrelated donor (n=16). Patients received a median of 3 cycles of LEN/AZA combination therapy.The maximum tolerated dose (MTD) of LEN/AZA combination therapy was estimated using a modified continual reassessment method (CRM) based on occurrence of dose-limiting toxicities observed within the first cycle. Clinical response was assessed after every 3rd cycle using modified Cheson criteria. The frequency of CD3+ T cells and their expression of PD1, LAG3, and TIM3 exhaustion markers was analyzed in all samples by flow cytometry. T cell activation status was determined by PMA and ionomycin stimulation followed by cytokine profile analyzing bead-array flow cytometry. Results: Using an efficient CRM model a dose of 25 mg LEN (days 10-30) in combination with AZA 75 mg/m2 x 7 days ( days 1-5, 8-9) was established as the MTD. Only one patient developed Grade 3-4 GVHD during LEN/AZA combined therapy. Major clinical responses were observed in 7 of the 15 patients who received three or more cycles of combined LEN/AZA treatment: CR/CRi n=6, PR n=1. Acquisition of a major clinical response was associated with prolonged overall survival (OS): median OS 17.3 versus 3.2 months (p<0.001). Increased levels of PD1 and LAG3 expression were present on CD3+ T cells at baseline compared with healthy controls (p=0.0025 and p=0.0177 respectively) consistent with a T cell exhaustion phenotype. T cells demonstrated impaired IFN-γ/TNF-α production at relapse and this was not reversed during LEN/AZA therapy. Conclusion: High doses of LEN can be safely delivered post-allograft when co-administered with AZA. Assessment of the MTD in this very high risk patient population was accelerated by the adoption of a CRM design which coped effectively with the challenges of cohort variation and early patient drop out. LEN/AZA combination therapy demonstrates promising clinical activity supporting further prospective evaluation as a novel salvage strategy post-transplant. Preliminary data supports combined LEN/AZA therapy exerting an anti-leukemia effect through a PD1 independent pathway.


Cavenagh: Amgen: Honoraria; Janssen: Honoraria; Celgene: Honoraria. Christina: Celgene: Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution