Leukemia (2009) 23, 649–655 & 2009 Macmillan Publishers Limited All rights reserved 0887-6924/09 $32.00 www.nature.com/leu ORIGINAL ARTICLE

The CDX2 is aberrantly expressed and associated with an inferior prognosis in patients with acute lymphoblastic leukemia

S Thoene1,2,4, VPS Rawat1,2,4, B Heilmeier1, E Hoster1, KH Metzeler1, T Herold1, W Hiddemann1,2,NGo¨kbuget3, D Hoelzer3, SK Bohlander1,2, M Feuring-Buske1,2 and C Buske1,2

1Department of Medicine III, Klinikum Grosshadern, Munich, Germany; 2Clinical Cooperative Group Leukemia, Helmholtz Center Munich for Environmental Health, Munich, Germany and 3Department of Medicine III, Johann Wolfgang Goethe University, Frankfurt/Main, Germany

Molecular characterization of acute lymphoblastic leukemia gene deregulation.11 However, it was shown that besides Hox (ALL) has greatly improved the ability to categorize and Cdx2 regulates other regulatory genes, such as Scl, prognostify patients with this disease. In this study, we show 12 that the proto-oncogene CDX2 is aberrantly expressed in the Gata1 and Runx1. Thus, aberrant expression of CDX2 might majority of cases with B-lineage ALL and T-ALL. High expres- perturb the stem cell regulatory network at different levels. sion of CDX2 correlated significantly with the ALL subtype We now report that CDX2 is aberrantly expressed in 81% of pro-B ALL, cALL, Ph þ ALL and early T-ALL. Furthermore, high adult patients with ALL, and that high expression levels of this expression of CDX2 was associated with inferior overall proto-oncogene predict poor treatment outcome in patients with survival and showed up as a novel and strong risk factor for this disease. ALL in bivariate analysis. Functional analyses showed that overexpression of Cdx2 in murine bone marrow progenitors perturbed genes involved in lymphoid development and that depletion of CDX2 in the human ALL cell line Nalm6 inhibited Materials and methods colony formation. These data indicate that aberrant CDX2 expression occurs frequently and has prognostic impact in Patient samples adult patients with ALL. Bone marrow or peripheral blood samples from 57 newly Leukemia (2009) 23, 649–655; doi:10.1038/leu.2008.355; diagnosed adult patients with ALL were analyzed. Of these, 31 published online 22 January 2009 patients were enrolled between 2000 and 2004 in the protocols Keywords: CDX2; acute lymphoblastic leukemia; HOX genes; prognostic factor 06/99 and 07/03 of the German Multicenter Study Group for Adult ALL (GMALL).13 Sixteen patients were enrolled in other multi- center protocols (GMALL Elderly 12/96, GMALL Elderly 01/03 Introduction and GMALL B-ALL/NHL 02). Ten patients were treated outside a study protocol. Patients gave written informed consent according With intensive risk adapted chemotherapy, stem cell transplanta- to the Declaration of Helsinki. The studies were approved by the tion and targeted therapies, the outcome of adult acute ethics board of the University Frankfurt am Main, Germany. lymphoblastic leukemia (ALL) has improved in the past decades from o10 to 40–50%.1,2 In addition, great progress has been made in understanding the biology of ALL: recently, genome-wide Quantitative PCR analyses of ALL patients showed deletions and mutations of genes Expression analyses were performed by TaqMan qRT-PCR associated with lymphoid development, such as LEF1, TCF, PAX5, using the Applied Biosystems 7900HT fast real-time PCR IKAROS and NOTCH1.3–5 However, for the majority of ALL cases system with pre-designed assays purchased the molecular mechanisms driving the malignant transformation from Applied Biosystems (Assay IDs CDX2: Hs01078080_m1, are unknown. Recently, data from experimental models and gene HOXA7: Hs00600844_m1, HOXA9: Hs00365956_m1, expression profiling in patients with HOXB6: Hs00980016_m1, TATA box binding (TBP): (AML) have identified CDX2 as a powerful oncogene when 4333769F; Applied Biosystems, Foster City, CA, USA). Reac- aberrantly expressed in adult hematopoietic progenitor cells:6–8 tions were run with 1 ml of cDNA containing the equivalent of the ‘caudal related homeobox gene’ CDX2 belongs to the family 50 ng total RNA in a total volume of 20 ml. DCT values were ofso-called‘ParaHoxgenes’,whichalsoincludesCDX1, CDX4 obtained by normalization to the housekeeping gene TBP. For and the GSH2 homeobox gene.9 Normally expression of CDX2 is patients with undetectable DCT values (negative samples), the tightly regulated in the adult organism, with expression in the lowest possible DCT was calculated by subtracting the CT value intestine but no expression in adult hematopoietic tissue.10 It was of TBP from the maximum number of cycles (45 cycles). Note shown that CDX2 is among the most frequent aberrantly expressed that DCT values are inversely correlated to gene expression proto-oncogenes in AML, with up to 89% of AML cases with levels. For the ‘low density assay’ (custom design LDA, array normal karyotype expressing CDX2.8,11 In AML patients, high configuration 7), 92 different genes involved in self-renewal, expression levels of CDX2 were closely associated with HOX proliferation and differentiation were selected. Each sample was run in duplicate and fold expression was calculated using the DDC method after normalizing to b-actin. Correspondence: Dr C Buske, Department of Medicine III, Klinikum T Grosshadern, Marchioninistrasse 15, 81377 Munich, Germany. E-mail: [email protected] 4These authors contributed equally to this work. Retroviral infection and in vitro assays 7 Received 3 November 2008; accepted 19 November 2008; published Primary mouse BM cells were transduced as described earlier. online 22 January 2009 For transduction of Cdx2, cells were co-cultured with irradiated CDX2 in acute lymphoblastic leukemia S Thoene et al 650 (4000 cGy from a 137Cs g-radiation) GP þ E86 Cdx2 producer SCIENCE/hannon.html. Four different shRNAs were obtained cells. Clonogenicity of short hairpin RNA (shRNA) transduced and cloned into the MSCV/LTRmiR30-PIG retroviral vector Nalm6 cells was analyzed in the CFC assay as described earlier (kindly provided by Scott W Lowe, Howard Hughes Medical (Methocult H4434, Stemcell Technologies, Vancouver, British Institute, New York, USA). Short hairpin RNA sequences are Colombia, Canada).14 available on request. Cell lines were retrovirally transduced as described earlier in Ahmed et al.14

Short hairpin RNA Methylation analysis Short hairpin RNA against CDX2 (NM_001265) were designed Quantitative DNA methylation status of the CpG island surround- using the following website http://www.cshl.org/public/ ing the transcription start site of CDX2 (À181 to þ 163)15 was

20.00 20.00

15.00 15.00

10.00 10.00 T T C C ∆ ∆ 5.00 5.00

0.00 0.00 (Normalized to TBP) (Normalized to TBP) -5.00 -5.00 Pro-B B-ALL + Early Thymic c-ALL Ph+ ALL Total Total ALL Burkitt T-ALL T-ALL ALL AML

% Positive patients 100 40 71 70 100 100 81 79 No of patients positive/ total (9/9) (4/10) (5/7) (7/10) (10/10) (11/11) (4657) (91/115)

12

10

8 Low/ absent CDX2 expression 6 High CDX2 expression 4 Number of cases 2

0 c-ALL Ph+ Pro-B B-ALL/ Early Thymic ALL ALL Burkitt T-ALL T-ALL

16.00 14.00 12.00 CDX2 10.00 HOXA7 8.00

T 6.00 HOXA9 C

∆ 4.00 HOXB6 2.00 0.00

(Normalized to TBP) -2.00 -4.00 -6.00 CD34+ MNC c-ALL Thymic Ph+ ALL Early Pro-B B-ALL/ (n=3) (n=3) (n=6) T-ALL (n=6) T-ALL ALL Burkitt (n=5) (n=4) (n=6) (n=3)

Figure 1 Quantitative expression of CDX2 and HOX genes in ALL patients. (a) Expression analysis of CDX2 in different acute lymphoblastic leukemia (ALL) subtypes and acute myeloid leukemia (AML) with normal and abnormal karyotype was performed by qRT-PCR. DCT values were calculated by normalization to the housekeeping gene TATA box binding protein (TBP). Dots represent individual patients, bars indicate the median expression level of CDX2 in the different ALL subgroups (±0.5 Â interquartile range (IQR)). The dashed line indicates the median expression level of CDX2 for all patients measured. Note that DCT values are inversely correlated to expression level. (b) Distribution of patients with low/absent versus high CDX2 expression levels in different ALL subgroups. Patients with no CDX2 expression or expression above the median (DCT of 7.14) were considered as low/absent, patients with CDX2 expression equal to or below the median as high. (c) Expression of HOX genes in CDX2-positive ALL patients and normal healthy individuals. (MNC, mononuclear BM cells, CD34 þ , CD34 þ BM cells) quantified by qRT-PCR. Columns represent average expression levels±s.e.m. The number of patient samples is indicated.

Leukemia CDX2 in acute lymphoblastic leukemia S Thoene et al 651 assessed by pyrosequencing of bisulfite-treated genomic varied substantially between different ALL subgroups: the DNA. After bisulfite treatment of genomic DNA, the region median expression level was 16-fold higher in pro-B ALL of interest was amplified by the primer set CDX2_F (50-TTGGT compared with mature B-ALL and 29-fold higher in early T-ALL GTTTGTGTTATTATTAATAGAGTTTTGTAAATAT-30)andCDX2_R compared with thymic T-ALL (Figure 1a, Table 1). When high (50-biotin-ATCCCAAAACAAACCTCACCATACTA-30). PCR pro- and low CDX2 expression levels were defined as below or duct was immobilized to Streptavidin Sepharose HP beads (GE above the median DCT value, none of the patients with mature Healthcare, Waukesha, WI, USA) followed by annealing to the B-ALL or thymic T-ALL had high CDX2 expression, whereas all sequencing primer. For sequencing three different primers were the patients with pro-B ALL, 8 of 10 patients with cALL and 8 of used, which are available on request. CpG analysis was done 11 patients with Ph þ ALL (10 pre-B/c-ALL and 1 pro-B ALL) with Pyro Q-CpG software (Biotage, Uppsala, Sweden). showed high CDX2 expression (Po0.001) (Figure 1b). Of note, ALL subtypes such as pro-B ALL expressed CDX2 more than four-fold higher than AML cases with normal karyotype, Statistical analysis described as the AML subgroup with highest CDX2 expression Correlations of CDX2 expression level with other patient levels11 (Figure 1a). variables were analyzed using Fisher’s exact test and the Although expression was detectable in CDX2- Mann–Whitney U-test. For this, patients were divided into positive ALL cases, there was no correlation between expression groups with high and low/absent CDX2 expression using the median DCT value (7.14) as cut-off. Data from patients enrolled in the GMALL 06/99 and 07/03 studies were used for correlation 1.0 of CDX2 expression level with different patient parameters as low (median), (n=15) model for overall survival. P-values o0.05 were considered statistically significant. Statistical analyses were performed using 0.4 SPSS (SPSS Inc, Chicago, IL, USA). Overall survival

0.2 Results

CDX2 is aberrantly expressed in the majority of patients 0.0 with acute lymphoblastic leukemia Fifty-seven samples from adult patients with newly diagnosed 0 500 1000 1500 2000 2500 ALL (Table 1) were analyzed for CDX2 expression levels by Time (days) qRT-PCR. In contrast to human normal hematopoietic bone þ 7,8 Figure 3 High expression of CDX2 is associated with poor treatment marrow, CD34 or peripheral blood mononuclear cells outcome in acute lymphoblastic leukemia (ALL). Kaplan–Meier plot (PBMCs), 46 of 57 ALL patients (81%) were positive for CDX2 for overall survival of patients with high versus low/absent CDX2 expression. Among the CDX2-positive cases, expression levels expression levels.

5.00

4.00

3.00

2.00

1.00

0.00 log10

-1.00

-2.00

-3.00

-4.00

-5.00 Kit Il7r Id3 Irf8 Tek Flt3 Tal1 Flt3l Bcl2 Mpo Tcf3 Lef1 Pilra Junb Sfpi1 Actr2 Cdx2 Ddx4 Csf1r Csf3r Emr1 Mycn Lmo2 Cxcr4 Meis1 Gata1 Gata2 Marco Lphn1 Hoxa5 Hoxa7 Hoxa9 Hoxb3 Hoxb8 Ccnd1 Ccnd2 Dusp2 Csf2ra Cd34a Cd34b Epor a Epor b Muc13 Slamf1 Hmga2 Zfp36l1 Abcb1b Hoxa10 Cdkn1a Aldh1a1 Aldh3b1

Figure 2 Differential gene expression in Cdx2-transduced bone marrow. Differential gene expression in Cdx2-transduced-bonemarrow cells. Fold expression levels of different genes in murine progenitor enriched 5-FU bone marrow after ectopic expression of Cdx2 determined 72 h after the end of transduction. Fold expression difference compared with empty vector-transduced bone marrow cells was calculated by averaging three independent experiments. Only genes with significant up- and downregulation are shown.

Leukemia CDX2 in acute lymphoblastic leukemia S Thoene et al 652 Table 1 Patient characteristics

ALL Patient no. Age Gender Karyotype Other relevant CDX2 subtype (years) markers expression

Pre-B/c-ALL 1a 49 F 46,XX 1 2 22 F 48,XX,+X,+7 1 3a 62 F 46,XX 1 4 26 F 46,XX,del(2)(q12q14) 1 5 34 M 44,XY,t(2;10)(p25;q24), 1 del(5)(q13q31),dic(7;17) (p11;p11), dic(9;12)(p11;p11),del(9) (q11), der(20)t(9;20)(q21;p13) 6 48 M 46,XY 1 7 40 M 46,XY 1 8 34 M 46,XY 1 9 25 F 46,XX,+8,dic(9;12)(p13;p11) 1 10 21 M 46,XY 1

Ph+ALLb 11 36 M 46,XY,t(9;22)(q34;q11) BCR/ABL 1 12 82 M 46,XY,t(1;4;11;8)(p11;q33;p12;p11), BCR/ABL 1 t(9;22)(q34;q11) 13 57 M 46,XY,der(9)t(9;22)(q34;q11)t(9;22) BCR/ABL 1 (p11;q11), der(22)t(9;22)(q34;q11) 14 38 M 45,X,-Y,t(9;22)(q34;q11),i(9)(q10), der(16) BCR/ABL 1 t(1;16)(q21;q13) 15 43 M 46,XY,t(9;22)(q34;q11),del(9)(p22) BCR/ABL 1 16 66 M 46,XY,t(9;22)(q34;q11) BCR/ABL 1 17 83 F 46,XX,t(9;22)(q34;q11),t(9;11)(p22;q25) BCR/ABL 1 18 58 F 51,XX,+5,+6,i(7)(q10),+8,t(9;22)(q34;q11), BCR/ABL 1 +15,+der(22)t(9;22) 19 29 F 46,XX,der(7;9)(q10;q10)t(9;22)(q34;q11), BCR/ABL, FLT3 D324 1 der(7;9)(p10;p10)del(7)(p?)del(9)(p?) 20 62 F 46,XX,der(9)t(4;9)(q31;q34)del(9)(p13), del(10) BCR/ABL 1 (q24),der(19)t(19;22)(q13;q11), der(22)t(9;22)(ABL+) 21 72 F 46,XX,t(9;22)(q34;q11) BCR/ABL 1

Pro-B-ALL 22a 77 F 46,XX,t(4;11)(q21;q23) MLL-AF4 1 23 47 F 46,XX,t(4;11)(q21;q23) MLL-AF4 1 24 64 F 46,XX,t(4;11)(q21;q23) MLL-AF4 1 25 40 F 46,XX,t(4;11)(q21;q23) MLL-AF4 1 26 53 F NA MLL-AF4 1 27a 39 M 47,XY,+X,t(11;19)(q23;p13) MLL-ENL 1 28 62 M 46,XY,t(4;11)(q21;q23) MLL-AF4 1 29 34 M 47,XY,+X,t(4;11)(q21;q23) MLL-AF4 1 30 61 F 46,XX,t(4;11)(q21;q23) MLL-AF4 1

B-ALL/Burkitt 31 42 F 46,XX,del(2)(q31),t(8;14)(q24;q32), der(11) IgH-cMYC-Rearr. 0 dup(11)(q21q25)t(2;11)(q31;q25) 32a 54 M 46,XY,dup(7)(q11q36),t(8;14)(q24;q32) IgH-cMYC-Rearr. 0 33 61 F 46,XX,dup(1)(q21q44),der(1)t(1;9)(p36;q?), IgH-cMYC-Rearr. 1 der(3)del(3)(p21)t(1;3)(q25;q25), der(6)t(3;6)(p21;q24), 34 34 F 46,XX,t(8;14)(q24;q32) IgH-cMYC-Rearr. 1 35 80 F 46,XX,dup(1)(q11q44), IgH-cMYC-Rearr. 0 der(2)t(2;11)(p25;q23),t(3;9;14) (q27;p13;q32),t(8;14)(q24;q32),del(11)(q11q23), der(22)t(11;22)(q23;q13) 36 72 F 49,X,del(X)(q24),+7,+8,t(8;14)(q24;q32),ider(13) FLT3 D324, 0 (q10)dup(13q?q?)t(13;14)(q34;q24) IgH-cMYC-Rearr., t(14;18)(q32;q21),der(14)t(X;14)(q26;q24),der(18) IgH-BCL2-Rearr. t(14;18)(q32;q21),+17 37 79 M 46,XY,t(8;14)(q24;q32) IgH-cMYC-Rearr. 0 38 40 M 47-51,XY,dup(1)(q21q44),+2,del(2)(q21),+7, IgH-cMYC-Rearr. 1 t(8;14)(q24;q32),+13,+15,+16 39 27 M 48,XY,+Y,+i(7)(q10),t(8;14)(q24;q32),der(17)t(1;17) IgH-cMYC-Rearr. 0 (q12;p13) [13] 49,XY,+Y,+i(7)(q10), t(8;14)(q24;q32),+der(14)t(8;14) [4] 40 26 M 46,XY,dup(1)(q21q44),t(8;14)(q24;q32), IgH-cMYC-Rearr. 1 inv(13)(p11q21)

Early T-ALL 41 46 M 47,XY,t(10;11)(p12;q21),+13 CALM-AF10 1 42a 70 F 46,XX,t(2;9)(q37;p11),t(3;7)(q27;p11), del(13)(q?) 1 43a 56 F 46,XX,der(5)t(4;5)(q25;q15) 1

Leukemia CDX2 in acute lymphoblastic leukemia S Thoene et al 653 Table 1 (Continued ) ALL Patient no. Age Gender Karyotype Other relevant CDX2 subtype (years) markers expression

44a 34 M 46,XY,t(10;11)(p12;q13),del(12)(p11) CALM-AF10 0 45 27 M 46,XY 1 46 52 F 46,XX 1 47 54 M 46,XY,t(2;14)(p25;q11) 0

Thymic T-ALL 48 39 M 46,XY 1 49 29 M 46,XY,t(1;14)(p34;q11),der(9)t(9;16)(p22;q22), 0 der(16)t(7;16)(?;p13)t(9;16)(p22;q22) 50 62 M NA 1 51 38 M NA 1 52a 22 F 46,XX 0 53 37 M 46,XY,del(6)(q21),i(9)(q10) 1 54 46 M 46,XY,t(5;7)(q35;q22),del(9)(p13) TLX-Rearr. 1 55 24 M 47,XY,t(7;14)(q35;q11),+16 1 56 42 M 46,XY,del(9)(p13),t(10;14)(q24;q11) 1 57 29 M 46,XY FLT3 D324 0 Abbreviations: 0, negative; 1, positive; ABL, Abelson kinase; AF, ALL-1 fused gene; ALL, acute lymphoblastic leukemia; BCR, breakpoint cluster region; CALM, clathrin assembly lymphoid myeloid leukemia protein; CDX2, expression of CDX2; cMYC, c- myelocytomatosis viral oncogene homolog; ENL, eleven-nineteen-leukemia protein; F, female; FLT, FMS-like tyrosine kinase; IgH, immunoglobulin heavy chainlocus; M, male; MLL, mixed-lineage leukemia; NA, not available; Ph+, Philadelphia t(9;22)(q34;q11) positive; TLX, T-cell leukemia homeobox. Patient characteristics, treatment, survival and CDX2 expression status are indicated. aMethylation status of the CDX2 promoter determined. bPh+ALL (n ¼ 11): 10 cases classified as pre-B/c-ALL, 1 case as pro-B-ALL.

Table 2 Methylation analysis of ALL samples

CDX2 Patient no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Average s.d.

Pro-B-ALL + 22 3 3 3 33 7 1 2 0 3 3 2 1 16 11 3 3 3 6 3 40 7 7 0 2 7 10 Early T-ALL + 43 88 80 98 93 92 82 89 81 95 90 91 58 78 93 70 91 91 82 77 100 5 80 74 79 82 19 c-ALL + 1 29 45 24 57 60 17 15 10 38 18 10 12 49 34 24 28 22 21 16 21 80 52 7 18 29 19 Early T-ALL + 42 67 63 78 76 75 62 66 62 80 72 82 60 72 79 59 77 80 70 67 92 35 70 52 56 69 12 Pro-B-ALL + 27 5 4 13 9 5 1 1 0 5 4 3 3 29 36 13 42 40 33 17 51 63 10 0 5 16 18 c-ALL + 3 31 30 39 37 15 2 4 3 3 4 5 4 34 33 25 35 8 33 11 41 8 30 2 19 19 14 Early T-ALL À 44 84 77 94 88 87 79 86 77 91 84 93 71 78 90 69 90 92 83 74 100 25 81 61 76 80 15 B-ALL À 32 7 4 4 8 14 1 2 1 39 39 14 26 42 15 9 14 21 9 9 12 79 13 4 9 16 18 Thymic T-ALL À 52 54 47 60 74 82 70 69 32 79 61 82 64 71 81 62 82 86 74 79 71 11 17 33 66 63 21 Abbreviations: ALL, acute lymphoblastic leukemia; Ph+, philadelphia chromosome t(9;22)(q34;q11) positive. Quantitative analysis of methylation status of CpG islands surrounding the transcription start site of CDX2 from the position À181 to +163 (15) in six ALL patients positive for CDX2 (2x c-ALL, 2x pro-B-ALL, 2x early T-ALL) and three patients negative for CDX2 (1x B-ALL, 1x early T-ALL, 1x thymic T-ALL). The table indicates the methylation per site and the average methylation level of all 24 CpG sites located within the analyzed region.

of CDX2 and expression of the leukemogenic HOXA7 or High CDX2 expression levels are associated with poor HOXA9 genes. There was a correlation between CDX2 and treatment outcome in ALL HOXB6 expression, but of borderline significance (P ¼ 0.048, When we analyzed the expression of CDX2 in different Mann–Whitney U-test) (Figure 1c). risk groups of ALL patients (according to Raff et al.13), We tested promoter hypomethylation as a potential cause most very high risk (VHR) and high risk (HR) patients for aberrant CDX2 expression, but did not detect any showed high CDX2 expression, in contrast to the standard difference in methylation of a region surrounding the transcrip- risk (SR) group (P ¼ 0.007 according to Fisher’s exact test) tional start site of CDX28,15 between CDX2-positive and CDX2- (Table 3). negative ALL cases (Table 2). We next analyzed whether In a univariate analysis, high CDX2 expression levels shRNA-mediated CDX2 depletion would affect the CDX2- were significantly associated with inferior OS (hazard ratio positive human pre-B ALL cell line Nalm6. In the CFC assay, (HR) 4.2, 95% CI: 1.1–15.7) (Figure 3): survival rate two Nalm6 cells transduced with HU_672_CDX2 (inducing a 49% years after diagnosis was 86% of patients with low/absent and reduction of CDX2 expression at the mRNA level by qRT-PCR) 52% of patients with high CDX2 expression levels (P ¼ 0.019, generated 33% fewer colonies compared with empty vector log rank test). In bivariate Cox regression, the CDX2 expression control and the non-active shRNA HU_673_CDX2 (data not level remained a significant risk factor even after adjusting for shown), indicating an important role of CDX2 for the growth the risk factors age and presence of molecular markers. The of this cell line. hazard ratio for CDX2 did not change substantially after Besides Hox genes,11 enforced expression of Cdx2 in normal adjusting for age (HR for CDX2 3.3, 95% CI: 0.86–12.85), murine progenitors significantly upregulated genes involved in presence of molecular aberrations (HR for CDX2 4.0, 95% CI: lymphopoiesis, such as Lef1 (3.9-fold), Tcf3 (13.3-fold), Flt3 1.07–15.13), ALL subtype (HR for CDX2 1.8, 95% CI: 0.36– (4.2-fold) and Id3 (10.2-fold) (Figure 2). 8.96), ALL risk category (HR for CDX2 2.9, 95% CI: 0.69–

Leukemia CDX2 in acute lymphoblastic leukemia S Thoene et al 654 Table 3 Correlation of CDX2 expression levels with clinical patient characteristics

Variable P-value High CDX2 expression % Median value in high vs low/absent CDX2 expression groups

ALL subtype o0.001 c-ALL 8 of 10 80 Ph+ ALL 8 of 11 73 Pro-B ALL 9 of 9 100 Mature B-ALL/Burkitt lymphoma 0 of 10 0 Early T-ALL 4 of 7 57 Thymic T-ALL 0 of 10 0

Menigeosis leucemica 0.343 Not present 12 of 22 55 Present 1 of 4 25

Risk13 0.007 SR 3 of 14 21 HR 8 of 11 73 VHR 5 of 6 83

Age (years) 0.303 48 vs. 40 years Molecular aberrationsa 0.790 Present 11 of 23 48 Not present 18 of 34 53

Gender 0.111 Female 16 of 25 64 Male 13 of 32 41

Karyotype 1.000 52 Normal 6 of 11 55 Aberrant 22 of 43 51

Blasts in BM (in % of cells) 0.047 95 vs. 95% Blasts in PB (in % of WBC) 0.188 95 vs. 90%

Leukocytes at presentation 1.000 52 o30 000/ml11of2250 430 000/ml17of3253 Abbreviations: ALL, acute lymphoblastic leukemia; BM, bone marrow; HR, high risk; PB, peripheral blood; Ph+, Philadelphia chromosome t(9;22)(q34;q11) positive; SR, standard risk; VHR, very high risk; WBC, white blood count. The correlation between different patient characteristics and high CDX2 expression levels is shown.

12.25), presence of karyotype aberrations (HR for CDX2 4.7, advantage for the cell than acquiring aberrant CDX4 expression 95% CI: 0.99–12.45), gender (HR for CDX2 6.2, 95% CI: 1.58– as constitutive expression of Cdx2 seems to be much more 24.24) or leukocytes at presentation (HR for CDX2 3.8, 95% CI: leukemogenic than Cdx4 in a murine BM transplantation 0.97–15.18). model.16 So far the leukemogenic potential of Cdx2 has been linked to its ability to deregulate HOX gene expression: it has been shown Discussion that Cdx2 is able to upregulate expression of 50 located Hox genes known to be leukemogenic and that this ability depends This report describes for the first time that the proto-oncogene on its N-terminal transactivation domain.11 However, it is CDX2 is aberrantly expressed in the majority of patients with B- unclear whether CDX2 can also exert its oncogenic effect by lineage or T-lineage ALL. Thus, it extends our knowledge about other pathways. This might be particularly true in adult patients the deregulation of this that was reported with ALL, in which HOX gene deregulation is comparably rare earlier to be one of the most frequent ectopically expressed compared with AML.5 This could indicate that in ALL the proto-oncogenes in AML,8,11 and underlines that aberrant leukemogenic effect of CDX2 does not depend on HOX gene expression of this homeobox gene is not restricted to myeloid perturbation to the same extent as in AML, and that CDX2 malignancies. This aberrant expression of CDX2 in human mediates its oncogenic potential through different gene net- leukemias stands in clear contrast to highly related genes such as works in ALL. When we analyzed gene transcription after CDX4 or CDX1, which were not expressed in our series of AML retroviral induction of aberrant Cdx2 expression in murine BM patients or were detected only in a minor fraction of patients progenitor cells, we observed that besides Hox genes also with AML.11,16 Furthermore, analyses of CDX4 expression in our factors involved in lymphoid development such as Lef1 were series of ALL samples did not show expression of this ParaHox upregulated. Recently, we showed that aberrant expression of gene (data not shown). The reason for this is unknown. But data this factor caused AML and also B-lineage ALL in transplanted from experimental murine models indicate that acquiring mice.17 Therefore, the aberrant CDX2 expression might initiate aberrant CDX2 expression might provide a greater growth or support myeloid as well as lymphoid leukemogenesis,

Leukemia CDX2 in acute lymphoblastic leukemia S Thoene et al 655 probably depending on the differentiation stage of the cell 4 Mullighan CG, Goorha S, Radtke I, Miller CB, Coustan-Smith E, initially affected by CDX2 overexpression. The underlying Dalton JD et al. Genome-wide analysis of genetic alterations in mechanism for aberrant CDX2 expression in human leukemias acute lymphoblastic leukaemia. Nature 2007; 446: 758–764. is unknown: we did not detect any difference in promoter 5 Paulsson K, Cazier JB, Macdougall F, Stevens J, Stasevich I, Vrcelj N et al. Microdeletions are a general feature of adult and methylation of the CDX2 gene between CDX2-positive and adolescent acute lymphoblastic leukemia: unexpected similarities CDX2-negative ALL cases. The same was reported for patients with pediatric disease. Proc Natl Acad Sci USA 2008; 105: with AML, where also no mutation in the coding region or gene 6708–6713. amplification was detected.8,15 6 Chase A, Reiter A, Burci L, Cazzaniga G, Biondi A, Pickard J et al. That aberrant CDX2 expression has functional relevance in Fusion of ETV6 to the caudal-related homeobox gene CDX2 in ALL was supported by the observation that shRNA-mediated acute myeloid leukemia with the t(12;13)(p13;q12). Blood 1999; 93: 1025–1031. CDX2 depletion compromised the clonogenic growth potential 7 Rawat VP, Cusan M, Deshpande A, Hiddemann W, Quintanilla- of the CDX2-positive human pre-B ALL cell line Nalm6. Martinez L, Humphries RK et al. Ectopic expression of the Furthermore, high expression of CDX2 was mostly found in homeobox gene Cdx2 is the transforming event in a mouse model ALL patients with a high-risk profile. In line with this, high of t(12;13)(p13;q12) acute myeloid leukemia. Proc Natl Acad Sci expression of CDX2 was associated with a significantly inferior USA 2004; 101: 817–822. OS. Remarkably and despite the limited patient number 8 Scholl C, Bansal D, Dohner K, Eiwen K, Huntly BJ, Lee BH et al. The homeobox gene CDX2 is aberrantly expressed in most cases of available for this analysis, CDX2 expression level remained a acute myeloid leukemia and promotes leukemogenesis. J Clin significant and strong risk factor even after adjusting for the risk Invest 2007; 117: 1037–1048. factors age and presence of molecular markers in bivariate 9 Brooke NM, Garcia-Fernandez J, Holland PW. The ParaHox gene analyses. cluster is an evolutionary sister of the Hox gene cluster. Nature Altogether, these results show that aberrant expression of 1998; 392: 920–922. CDX2 is a frequent event in adult patients with ALL. Delineating 10 Suh E, Traber PG. An intestine-specific homeobox gene regulates proliferation and differentiation. Mol Cell Biol 1996; the underlying molecular mechanisms of aberrant CDX2 16: 619–625. expression and characterizing factors mediating CDX2-induced 11 Rawat VP, Thoene S, Naidu VM, Arseni N, Heilmeier B, Metzeler leukemogenesis will help to understand important steps in K et al. Overexpression of CDX2 perturbs HOX gene expression in leukemogenesis, which are common in both lymphoid and murine progenitors depending on its N-terminal domain and is myeloid leukemias. closely correlated with deregulated HOX gene expression in human acute myeloid leukemia. Blood 2008; 111: 309–319. 12 Wang Y, Yabuuchi A, McKinney-Freeman S, Ducharme DM, Ray Acknowledgements MK, Chawengsaksophak K et al. gene deficiency compro- mises embryonic hematopoiesis in the mouse. Proc Natl Acad Sci USA 2008; 105: 7756–7761. We thank Bianka Ksienzyk and Nicole Behm for their excellent 13 Raff T, Gokbuget N, Luschen S, Reutzel R, Ritgen M, Irmer S et al. technical assistance, and the members of the animal facility at the Molecular relapse in adult standard-risk ALL patients detected by Helmholtz Center Munich for their excellent breeding and prospective MRD monitoring during and after maintenance maintenance of the animals. VPSR and CB and their work were treatment: data from the GMALL 06/99 and 07/03 trials. Blood supported by a grant of the DFG (SFB 684 project A7), MF-B by 2007; 109: 910–915. 14 Ahmed F, Arseni N, Glimm H, Hiddemann W, Buske C, Feuring- the Deutsche Krebshilfe (70-2968-Fe I to MFB) and the DFG (SFB Buske M. Constitutive expression of the ATP-binding cassette 684 project A8), SKB by the DFG (SFB 684 project A6), CB, MF-B transporter ABCG2 enhances the growth potential of early human and SKB by the Bundesministerium fu¨r Bildung und Forschung hematopoietic progenitors. Stem Cells 2008; 26: 810–818. (NGFN2 Grant 01GS0448) and DH by the Deutsche Krebshilfe 15 Yuasa Y, Nagasaki H, Akiyama Y, Sakai H, Nakajima T, Ohkura Y (70-2657-Ho2). et al. Relationship between CDX2 gene methylation and dietary factors in gastric cancer patients. Carcinogenesis 2005; 26: 193–200. 16 Bansal D, Scholl C, Frohling S, McDowell E, Lee BH, Dohner K References et al. Cdx4 dysregulates Hox gene expression and generates acute myeloid leukemia alone and in cooperation with Meis1a 1 Pui CH, Evans WE. Treatment of acute lymphoblastic leukemia. in a murine model. Proc Natl Acad Sci USA 2006; 103: N Engl J Med 2006; 354: 166–178. 16924–16929. 2 Gokbuget N, Hoelzer D. Treatment of adult acute lymphoblastic 17 Petropoulos K, Arseni N, Schessl C, Stadler CR, Rawat VP, leukemia. Hematol Am Soc Hematol Educ Program 2006; 1: 133–141. Deshpande AJ et al. A novel role for Lef-1, a central transcription 3 Armstrong SA, Look AT. Molecular genetics of acute lymphoblastic mediator of Wnt signaling, in leukemogenesis. J Exp Med 2008; leukemia. J Clin Oncol 2005; 23: 6306–6315. 205: 515–522.

Leukemia