Expression Of CD25 Antigen On CD34+ Cells Is An Independent Predictor Of Survival In Late Stage MDS Patients Treated With Azacitidine

Expression of CD25 on blasts of AML patients has been shown to hold independent prognostic value for both survival and response to induction therapy. Patients with MDS-related AML have generally higher CD25 expression from de novo AML patients, but though there is paucity of a serviceable biomarker of outcome in MDS patients treated with azacytidine, the prognostic value of CD25 has not yet been investigated.

Bone marrow samples of 61 patients with intermediate-2/high risk IPSS, high/very high WPSS and low blast count AML were obtained before and 15 days (D15) after the initiation of treatment. All patients received azacytidine in a non clinical trial setting at an initial dose of 75mg/m2 SC for 7 days on 28-day cycles. CD25 expression was assessed by 4-color flow cytometry on total CD34+ blasts, committed progenitors (Lin-CD38+CD34+) and leukemic stem cells (LSC, Lin-CD38-CD34+). Positivity was defined as a CD25 expression of ≥ 20%. Statistical comparisons were done by ÷2, one-way ANOVA and paired or unpaired t-test as appropriate, and survival with Kaplan-Meier analysis and log-rank test. Overall survival (OS) was defined as the time from azacytidine initiation to death from any cause and event-free survival (EFS) as the time from diagnosis to disease progression, relapse or death. Multivariate survival analysis was based on Cox’s proportional hazards model using a backward stepwise selection procedure with entry and removal criteria of p=0.05 and p=0.10, respectively.

As shown in table 1 the cohorts of CD25- and CD25+ patients were well balanced for most known predictive factors and characteristics, except sex. Compared to CD25+ patients the CD25- ones have significantly longer OS (16.2 vs 8.8 months, respectively, p=0.04) and EFS (12.8 vs 6.66 months, p=0.04) in univariate analysis. Multivariate analysis confirmed the independent predictive power of CD25 for OS and EFS (p=0.006 and p=0.009, respectively), whereas heavy transfusion requirements (p=0.003 and p=0.002) and age>75 (p=0.02 and p=0.02) were also independent predictors. The average expression of CD25 in CD34+ blasts of all patients was 21.6%±24%. Compared to committed progenitors, LSCs displayed higher expression (19.4%±23.7% vs 24.1%±28.2%, respectively, p=0.027). Interestingly, on D15 CD25 was downregulated in LSCs (p=0.03) but remained stable in committed progenitors (p=0.8, n=18), indicating a particular sensitivity of the CD25+ subset of LSCs in azacytidine.

Table 1

Patient characteristics. N/A: not applicable/not available.

.	CD25- (n=36) .	CD25+ (n=25) .	p-value .
Age	72,5 (53,4-83.5)	72,9 (52-81.7)	0.2
>65	31(86%)	17(68%)
<65	5(14%)	8(32%)
Sex			0.027
Male	20(55%)	21(84%)
Female	16(45%)	4(16%)
WHO classification			0.28
RCMD	2(5%)	0(0%)
RAEB-I	0(0%)	1(4%)
RAEB-II	17(47%)	13(52%)
CMML-II	6(16%)	7(28%)
Low blast count - sAML	11(30%)	4(16%)
Baseline blood counts
Hemoglobin (g/dl)	8,6(6.1-10.6)	8,8(6.8-11.5)	0.78
ANC(x 109/L)	1,1(0.04-13.4)	2,8(0.08-23)	0.07
Platelets (x 109/L)	66(9-383)	50(11-181)	0.42
IPSS			0.6
Intermediate-2	17(47%)	9(36%)
High	8 (22%)	6(24%)
N/A	11(31%)	10(40%)
WPSS			0.22
High	15(42%)	8(32%)
Very high	6(16%)	7(28%)
N/A	15(42%)	10(40%)
IPSS-R			0.8
Intermediate	3(8%)	1(4%)
High	14(39%)	8(32%)
Very high	17(48%)	12(48%)
N/A	2(5%)	6(24%)
IPSS-R Cytogenetic risk			0.72
Good	18(50%)	10(40%)
Intermediate	8(22%)	6(24%)
Poor	5(14%)	6(24%)
Very poor	4(11%)	2(8%)
N/A	1(3%)	1(4%)
PB blasts			0.5
Present	19(53%)	13(52%)
Absent	14(39%)	11(44%)
N/A	3(8%)	1(4%)
BM blasts			0.2
>15%	18(50%)	9(36%)
≤15%	18(50%)	16(54%)
Transfusions ≥ 4 per month			0.48
Yes	23(4%)	17(4%)
No	13(4%)	8(4%)
Response			0.4
CR & PR	12(33%)	4(16%)
Hematologic improvement	5(14%)	4(16%)
Stable disease	8(22%)	5(20%)
Failure	11(31%)	12(48%)

.	CD25- (n=36) .	CD25+ (n=25) .	p-value .
Age	72,5 (53,4-83.5)	72,9 (52-81.7)	0.2
>65	31(86%)	17(68%)
<65	5(14%)	8(32%)
Sex			0.027
Male	20(55%)	21(84%)
Female	16(45%)	4(16%)
WHO classification			0.28
RCMD	2(5%)	0(0%)
RAEB-I	0(0%)	1(4%)
RAEB-II	17(47%)	13(52%)
CMML-II	6(16%)	7(28%)
Low blast count - sAML	11(30%)	4(16%)
Baseline blood counts
Hemoglobin (g/dl)	8,6(6.1-10.6)	8,8(6.8-11.5)	0.78
ANC(x 109/L)	1,1(0.04-13.4)	2,8(0.08-23)	0.07
Platelets (x 109/L)	66(9-383)	50(11-181)	0.42
IPSS			0.6
Intermediate-2	17(47%)	9(36%)
High	8 (22%)	6(24%)
N/A	11(31%)	10(40%)
WPSS			0.22
High	15(42%)	8(32%)
Very high	6(16%)	7(28%)
N/A	15(42%)	10(40%)
IPSS-R			0.8
Intermediate	3(8%)	1(4%)
High	14(39%)	8(32%)
Very high	17(48%)	12(48%)
N/A	2(5%)	6(24%)
IPSS-R Cytogenetic risk			0.72
Good	18(50%)	10(40%)
Intermediate	8(22%)	6(24%)
Poor	5(14%)	6(24%)
Very poor	4(11%)	2(8%)
N/A	1(3%)	1(4%)
PB blasts			0.5
Present	19(53%)	13(52%)
Absent	14(39%)	11(44%)
N/A	3(8%)	1(4%)
BM blasts			0.2
>15%	18(50%)	9(36%)
≤15%	18(50%)	16(54%)
Transfusions ≥ 4 per month			0.48
Yes	23(4%)	17(4%)
No	13(4%)	8(4%)
Response			0.4
CR & PR	12(33%)	4(16%)
Hematologic improvement	5(14%)	4(16%)
Stable disease	8(22%)	5(20%)
Failure	11(31%)	12(48%)

View Large

Figure 1

View largeDownload slide

(A) OS and EFS according to CD25 positivity status. (B) OS and EFS according to transfusion requirements.

Figure 1

View largeDownload slide

(A) OS and EFS according to CD25 positivity status. (B) OS and EFS according to transfusion requirements.

Close modal

Collectively, our findings reveal an independent prognostic role for CD25 in MDS patients treated with azacytidine. In addition, the differential expression and epigenetic modulation of CD25 in the LSC compartment support the investigation of therapeutic strategies using monoclonal antibody targeting combined with epigenetic agents.