Azacytidine and Vorinostat in Patients with Chronic Lymphocytic Leukemia (CLL) Diagnosed with Therapy-Related Myelodysplastic Syndromes/Acute Myeloid Leukemia (t-MDS/AML)

Background: There is evidence of a leukemogenic effect of purine analogues, mainly when combined with DNA-damaging agents. Various series report an approximately 5% rate of t-MDS/AML in patients treated with a fludarabine-based regimen. Patients are generally old, and old age is associated with worse outcomes. To date, there is no established standard therapy recommendation for this group of patients, and results of prospective treatment evaluations are scarce. Aim: To determine the characteristics and treatment outcomes of patients with CLL and t-MDS/AML. Methods: We analyzed a group of 6 patients with newly diagnosed t-MDS/AML who were treated in our institution from September 2011 to July 2014. Patients were enrolled in a phase II trial of azacitidine (75 mg/m² IV daily x 5 days) in combination with vorinostat (200 mg orally three times daily x 5 days) (Arm A) or azacitidine alone (Arm B), with courses repeated every 3-8 weeks. This trial was designed to uniformly treat patients not eligible for other leukemia protocols due to comorbidities. Results: At baseline, all patients had an underlying diagnosis of CLL that was in remission or minimally active. The median percentage of CLL bone marrow involvement was 10% (range 0-60%), and ALC was 0.9 K/uL (range 0.2-9.79). The median number of prior CLL treatments was 2 (range 0-3). All patients had previously received fludarabine-based regimens. The median time from chemotherapy to t-MDS/AML diagnosis was 10 years (range 4-10). All patients were male, and the median age was 72 years (range 52-72). All patients had 3-line cytopenia, with median WBC 2.3 K/uL (range 0.8-8.9), ANC 0.55 K/uL (range 0-1.22), hemoglobin 9.6 G/DL (range 8.3-10.4), platelets 39 K/uL (range 6-80), and bone marrow blast percentage of 5% (range 1-18%). The karyotype was complex in all patients. Molecular studies showed that 3 patients had TP53 gene mutations. Five patients received treatment in Arm A, and only 1 patient was randomized to Arm B. Patients received a median of 4 cycles (range 2-7) and remained in the study for a median time of 216 days (range 86-329) before progression. None of the patients achieved remission, but stable disease was observed in 5 out of 6 patients. At the time of this analysis, 4 patients are dead and 2 are still alive: one discontinued treatment because of prolonged myelosuppression and is receiving best supportive care, and the other is recovering from cycle number 4 of treatment. The median survival in the group from the time of treatment initiation was 10.1 months and from the time of study discontinuation was 3.1 months. Further therapy was attempted in 3 patients without response. Conclusion: This is a group of patients with poor prognostic features. Azacytidine and vorinostat have been previously reported to be a safe combination (Garcia-Manero et al. ASH 2010, abstract 604) and may constitute a reasonable treatment alternative. Further prospective studies involving larger numbers of patients are required.

WBC: White blood cells, Hb: hemoglobin, Plt: Platelets, BM: Bone Marrow, CLL: Chronic Lymphocytic Leukemia, F: Fludarabine, C: Cyclophosphamide, R: Rituximab, B: Bendamustine, MEDI-551: anti-CD19 antibody, NR: no response, SD: stable disease.