Predictive Factors for Response and Survival in Patients with Myelodysplastic Syndromes (MDS) after Hypomethylating Agent (HMA) Failure: Primary Resistance (PriRes) Vs. Secondary Resistance (SecRes)

Introduction:

The outcomes of patients with MDS after HMA failure are poor without effective therapy. The aim of this study is to evaluate predictive factors for response and survival after HMA failure.

Methods:

We reviewed 149 patients who received 2nd-line therapy after HMA failure at our tertiary care center between 1/2010 and 5/2014. Patients were assigned to receive either decitabine (DAC) or azacitidine (AZA) in equal numbers before this study in the context of a prospective clinical trial. Of 149 patients, 23 patients proceeded to allogeneic stem cell transplant, 15 patients had progression to acute myeloid leukemia (AML) at HMA failure, and 18 patients were not evaluable. The remaining 93 patients are the focus of this analysis. Response data were collected according to IWG 2006 criteria. Patients were divided into two categories: PriRes or SecRes. PriRes was defined as lack of any response to HMA therapy, and SecRes was defined as loss of response. There were 4 groups of 2nd-line treatment: the other HMA (AZA or DAC), cytarabine, clofarabine, and investigational agents. Overall survival (OS) was defined as time from the date of first 2nd-line treatment to the date of last follow-up or censorship date. Progression-free survival (PFS) was defined as the time from the date of first 2nd-line treatment to the date of progression to AML. Univariate (UVA) and multivariate analysis (MVA) were performed to analyze the response to second-line therapies and survival.

Results:

Ninety-three patients were evaluable; 30 (32%) had PriRes and 63 (68%) had SecRes. Thirty-four (37%) received AZA, and 59 (63%) received DAC as the first HMA. Median follow-up was 16 months, and median age was 65 years (20-85). Baseline patient characteristics did not differ significantly between groups. Median duration of AZA and DAC were 5.6 months (0.2-49.2) and 7.2 months (0-38.4), respectively (p= 0.1). Primary resistance was observed in 14 patients (41%) receiving AZA and 16 patients (27%) receiving DAC (p= 0.163). Median duration of HMA therapy in the PriRes and SecRes groups were 1.1 months (0-5.8) and 10.5 months (0.2-49.2), respectively (p< 0.001). Of 63 patients in SecRes, 43 (68%) achieved CR after HMA, 10 (16%) CRp, 4 (6%) PR, and 6 (10%) HI. Complex karyotype was observed in 3 patients (10%) in PriRes and 19 (30%) in SecRes (p= 0.032). RAS mutation was observed in 4 patients (13%) in PriRes and 4 (6%) in SecRes. One patient (3%) in PriRes had FLT3-D835 mutation. No FLT3-ITD, NPM1, or JAK2 mutations were detected in this study. Of the 30 patients in PriRes, 2 (7%) received the other type of HMA, 5 (17%) cytarabine, 6 (20%) clofarabine, and 17 (57%) other treatments. Of the 63 patients in SecRes, 9 (14%) received the other HMA, 11 (17%) cytarabine, 19 (30%) clofarabine, and 20 (32%) others. Of 30 patients in PriRes, OIR was achieved in 9 (29%), CR in 1(3%), CRp in 2 (7%), HI in 5 (17%), and SD/PD in 18 (60%). Of 63 patients in SecRes, OIR was observed in 18 (29%), CR in 11 (18%), CRp in 2 (3), PR in 2 (3%), HI in 3 (5%), and SD/PD in 36 (57%). The response rate did not differ significantly between groups (p= 0.739). By UVA for response, only transition to the other HMA was a better prognostic factor for response (p= 0.019; hazard ratio [HR], 0.201, 95% confidence interval [CI] 0.053-0.764). Both patients in PriRes that switched to the other HMA (1 patient, AZA to DAC; 1 patient, DAC to AZA) had HI. Of 9 patients receiving the other HMA in SecRes (4 patients, AZA to DAC; 5 patients, DAC to AZA), in the DAC to AZA group, 1 (25%) achieved CR, 2 (50%) SD, and 1 (25%) died, and in the AZA to DAC group, 4 (80%) achieved CR and 1 (20%) had unknown outcome. Median PFS from date of first 2nd-line treatment in PriRes and SecRes were 5.8 months and 1.8 months (p=0.178), respectively. Median OS from date of first 2nd-line treatment in PriRes and SecRes were 10.4 months and 9.8 months (p =0.995), respectively. By MVA for survival, absence of response after 2nd-line therapy [p= 0.001; HR, 3.076; 95%CI, 1.610-5.877], presence of complex karyotype [p=0.046; HR, 1.849; 95%CI, 1.012-3.378], and 2nd-line therapy with clofarabine [p=0.025; HR, 1.894; 95%CI, 1.085-3.304] were poor prognostic factors.

Conclusion:

Outcome of patients after HMA failure is generally poor. Patients who were previously refractory to HMAs seem to respond to 2nd-line therapies and had similar survival. This finding needs to be confirmed by other prospective studies.