A Phase II Study of Decitabine In Advanced Chronic Myelomonocytic Leukemia (CMML)

CMML, now classified among MDS/MPN, is prognostically heterogeneous. We (JCO 1988 6:1417, Blood 1996 88:2480) and others found prognosis in CMML to depend on both “MDS factors” (% marrow blasts, cytopenias, karyotype) and “MPN” factors, ie splenomegaly (SMG), WBC count, extramedullary disease (EMD). Treatment of advanced CMML is difficult. The hypomethylating agents Azacitidine and Decitabine (DAC) have shown efficacy in CMML, but in prognostically heterogeneous cohorts. We conducted a phase II trial of DAC in a well defined cohort of advanced CMML.

To be included, CMML (according to WHO) should have: if WBC<13G/l an IPSS≥1.5; if WBC≥13G/l, two of the following criteria: marrow blasts≥5%, Hb<10g/dl, plts<100G/l, abnormal cytogenetics, SMG >5cm below costal margin, (SMG>5cm), EMD, based on our previous prognostic factors. Pts received DAC 20mg/m2/d IV for 5 days every 28 days for at least 3 cycles. Response criteria were based on IWG 2006 for pts with WBC <13G/L and for pts with WBC > 13 G/L also included evolution of WBC, SMG and EMD (Blood 1996 88:2480).

Between Nov 2008 and June 2009, 41 pts were included in 16 centers, of whom 39 completed at least one cycle and were considered evaluable for response (the 2 other pts died from septic shock before and during the first cycle respectively). Median number of cycles was 9 (range 1–17). Median age was 71 years (range 54–88), M/F:30/9. Seventeen pts had CMML 1 and 22 had CMML 2 (including pts with up to 29% marrow blasts). Median WBC count was 29.5G/l (range 4.1–147.3), median blood monocytes 3G/l (range 1.05–95.7), and median BM blasts 10% (1-29). Nine pts had WBC<13G/l and 30 WBC≥13G/l. Abnormal karyotype was found in 18 (46.2%) pts, including +8 and -7 in 7 and 1 case, respectively. 15 pts (38.6%) had SMG>5cm and 8 (20.5%) EMD involving skin (n=5) and lymph nodes (n=3). Overall Response Rate (ORR,) was 38.6% with 4 (10.3%) CR, 8 (20.5%) marrow CR and 3 (7.7%) Stable Disease (SD) with HI. 1 CR pt received allo SCT. 18 (46%) pts had SD without HI and 6 (15.4%) pts progressed to AML. 8 (36.4%) of the 22 RBC transfused pts became RBC transfusion independent. 3/10 (30%) pts with plt<50G/l reached plt>100G/l. Four pts had cytogenetic response (3 CR and 1 PR) exclusively among pts with +8. Median peripheral monocyte count decreased from 4.8G/l to 0.3G/l after 3 cycles of DAC among responders. SMG disappeared in 6/15 (40%) pts and EMD in 6/8 (75%) pts. 16 pts were receiving Hydroxyurea (HY) at inclusion which could be stopped in 12 of them. With a median follow up of 10 months (1-18), 7 pts had died from progression (n=3), sepsis (n=3) and unrelated cause (n=1). Overall Survival (OS) estimate was 60% at 2 years (median not reached). By comparison, in our previous trial of HY in CMML where inclusion criteria were the same, 2 year survival was 43% (median OS 20 months; Blood 1996 88:2480), for a median follow up of 11 months (range1-43). The only factor associated with response to DAC was WHO subtype, CMML 2 pts showing significantly better ORR (30.8% vs 7.7% in CMML 1; p=0.041).There was no difference in survival between CMML 1 and 2 pts. Treatment with DAC was generally well tolerated with, except for usual grade 3/4 cytopenias, and grade 3 fatigue (n=1).

DAC is active in advanced CMML and safe in these elderly pts. A possibly better survival than with HY will have to be confirmed in randomized trials. Correlative genetic studies identifying markers potentially predicting response and survival for DAC are currently underway in our lab.

No relevant conflicts of interest to declare.