Background

No abstract

Purpose: Azacitidine (AZA) is a novel therapeutic option in older patients with acute myeloid leukemia (AML), but its rational utilization is compromised by the fact that neither the determinants of clinical response nor its mechanism of action are defined. Co-administration of histone deacetylase inhibitors, such as vorinostat (VOR), is reported to improve the clinical activity of AZA, but this has not been prospectively studied in patients with AML.Experimental Design: We compared outcomes in 259 adults with AML (n ¼ 217) and MDS (n ¼ 42) randomized to receive either AZA monotherapy (75 mg/m 2 Â 7 days every 28 days) or AZA combined with VOR 300 mg twice a day on days 3 to 9 orally. Next-generation sequencing was performed in 250 patients on 41 genes commonly mutated in AML. Serial immunophenotyping of progenitor cells was performed in 47 patients.Results: Co-administration of VOR did not increase the overall response rate (P ¼ 0.84) or overall survival (OS; P ¼ 0.32). Specifically, no benefit was identified in either de novo or relapsed AML. Mutations in the genes CDKN2A (P ¼ 0.0001), IDH1 (P ¼ 0.004), and TP53 (P ¼ 0.003) were associated with reduced OS. Lymphoid multipotential progenitor populations were greatly expanded at diagnosis and although reduced in size in responding patients remained detectable throughout treatment.Conclusions: This study demonstrates no benefit of concurrent administration of VOR with AZA but identifies a mutational signature predictive of outcome after AZA-based therapy. The correlation between heterozygous loss of function CDKN2A mutations and decreased OS implicates induction of cell-cycle arrest as a mechanism by which AZA exerts its clinical activity.

Reduced-intensity conditioning (RIC) allogeneic hematopoietic cell transplantation is increasingly considered for patients with chronic lymphocytic leukemia (CLL). To investigate the impact of in vivo T cell depletion with alemtuzumab on the incidence of graft-versus-host disease (GVHD), nonrelapse mortality (NRM), progression-free survival (PFS), and overall survival (OS), we retrospectively analyzed the outcomes of 62 consecutive CLL patients conditioned with fludarabine and melphalan at 4 institutions. For GVHD prophylaxis, 41 patients (cohort 1) received alemtuzumab and cyclosporin; and 21 patients (cohort 2) received cyclosporin plus methotrexate or mycophenolate. Donors were 50 siblings and 12 unrelated volunteers. Twenty-two (36%) patients received donor lymphocyte infusions (DLI), 20 (49%) from cohort 1 and 2 (10%) from cohort 2 (P=.002). Grade III-IV acute GVHD (aGVHD) was observed in 20% and 38% of patients from cohorts 1 and 2, respectively (P=.14). Extensive chronic GVHD (cGVHD) was observed in 10% and 48% of patients from cohorts 1 and 2, respectively (P=.03). There was a trend toward a higher viral infection rate in cohort 1 compared to cohort 2 (68% versus 43%, P=.062), but the incidence of cytomegalovirus (CMV) reactivation was not significantly different. The 3-year OS, PFS, NRM, and relapse rates were 65%, 39%, 28%, and 32%, respectively, for cohort 1; and 57%, 47%, 34%, and 20%, respectively, for cohort 2 (P=.629, P=.361, P=.735, and P=0.112, respectively). In conclusion, both methods of GVHD prophylaxis were equivalent in terms of survival. The administration of alemtuzumab led to reduced cGVHD, possibly improving quality of life.

RIC allo-HCT is a reasonable option for patients with high-risk CLL. However, these results require confirmation before the procedure can be recommended outside clinical trials.