Non-classical karyotypic features in relapsed childhood B-cell precursor acute lymphoblastic leukemia

Cancer Genetics and Cytogenetics 189 (2009) 29e36

Non-classical karyotypic features in relapsed childhood B-cell precursor acute lymphoblastic leukemia Lea A. Wehrlia, Julia Braunb, Luisa Nobile Buettic, Nicole Hagleitnerd, Heinz Hengartnere, Thomas Ku¨hnef, Sonja Lu¨erg, Hulya Ozsahinh, Maja Beck Popovici, Felix K. Nigglia, David R. Bettsj, Jean-Pierre Bourquina,* b

a Division of Oncology, University Children’s Hospital, Steinwiesstrasse 75, CH-8032 Zurich, Switzerland Biostatistics Unit, Institute of Social und Preventive Medicine, University of Zurich, Hirschengraben 84, CH-8001 Zurich, Switzerland c Department of Pediatrics, Ospedale Regionale di Locarno, Via all ’Ospedale 1, CH-66001 Locarno, Switzerland d Department of Pediatrics, Children’s Hospital, Kantosspital, CH-6000 Lucerne 16, Switzerland e Division of Oncology, Children’s Hospital, Claudiusstrasse 6, CH-9006 St. Gallen, Switzerland f Division of Hematology/Oncology, University Children’s Hospital, Ro¨mergasse 8, CH-4005 Basel, Switzerland g Division of Hematology/Oncology, University Children’s Hospital, Inselspital, CH-3010 Bern, Switzerland h Division of Hematology/Oncology, University Children’s Hospital, Rue Willy-Donze´ 6, 1211 Geneva 14, Switzerland i Division of Hematology/Oncology, University Children’s Hospital, CHUV, Rue du Bugnon 46, 1011 Lausanne, Switzerland j National Centre for Medical Genetics, Our Lady’s Children’s Hospital, Crumlin, Dublin 12, Ireland

Received 30 April 2008; accepted 9 October 2008

Abstract

Karyotype analysis of acute lymphoblastic leukemia (ALL) at diagnosis has provided valuable prognostic markers for treatment stratification. However, reports of cytogenetic studies of relapsed ALL samples are limited. We compared the karyotypes from 436 nonselected B-cell precursor ALL patients at initial diagnosis and of 76 patients at first relapse. We noticed a relative increase of karyotypes that did not fall into the classic ALL cytogenetic subgroups (high hyperdiploidy, t(12;21), t(9;22), 11q23, t(1;19), !45 chromosomes) in a group of 29 patients at relapse (38%) compared to 130 patients at presentation (30%). Non-classical cytogenetic aberrations in these 29 patients were mostly found on chromosomes 1, 2, 7, 9, 13, 14, and 17. We also describe six rare reciprocal translocations, three of which involved 14q32. The most frequent abnormalities were found in 9p (12/29 cases) and were associated with a marked decrease in the duration of the second remission, but not of the probability of 10-year event-free survival after relapse treatment. From 29 patients with nonclassical cytogenetic aberrations, only 8 (28%) had been stratified to a high risk-arm on the first treatment protocol, suggesting that this subgroup might benefit from the identification of new prognostic markers in future studies. Ó 2009 Elsevier Inc. All rights reserved.

1. Introduction Karyotype is an independent predictor of outcome in precursor-B acute lymphoblastic leukemia (pre-B-ALL) [1,2]. In about 30% of childhood ALL, high hyperdiploidy (HHD, O50 chromosomes) can be detected and is associated with good prognosis [3,4]. In contrast, hypodiploidy (!45 chromosomes) and near-haploidy (23e29 chromosomes) are associated with worse outcome [5,6]. Among the most frequent structural chromosomal abnormalities, * Corresponding author. Tel.: þ41 44 266 7304; fax: þ41 44 266 7171. (J-P. Bourquin). E-mail address: [email protected] (J.-P. Bourquin). 0165-4608/09/$ e see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.cancergencyto.2008.10.002

t(12;21)(p12;q22), resulting in the ETV6/RUNX1 fusion transcript in 25e30% of B-lineage ALL, is associated with a favorable prognosis despite a notable increase of late relapses [7]. The t(9;22)(q34;q11.2), resulting in the BCR/ ABL1 fusion transcript [8,9], and rearrangements involving the MLL gene on chromosome 11q23 [10] are associated with adverse outcome in pre-B-ALL. In addition, abnormalities of the short arm of chromosome 9 [abn(9p)] have been reported in approximately 11% of pre-B ALL [11], which was shown to be a poor prognostic factor in two large pediatric ALL studies [11,12]. Most cases with abn(9p) appear to involve two linked cyclin-dependent kinase inhibitor genes at 9p21, CDKN2A, and CDK2B [13,14]. More recent studies trying to correlate specific loss

30

L.A. Wehrli et al. / Cancer Genetics and Cytogenetics 189 (2009) 29e36

of CDKN2A with outcome did not confirm deletion of this gene as an independent prognostic marker of poor outcome [15,16]. Using fluorescence in situ hybridization (FISH), a region of intrachromosomal amplification of chromosome 21 (iAMP21) was found to provide an additional relevant adverse prognostic marker at diagnosis [17,18]. While progress has been made to understand the impact of genetic lesions in childhood ALL at diagnosis, as exemplified by a recent genome-wide analysis [19], less common recurrent abnormalities are not well categorized, and limited cytogenetic data has been reported for relapsed ALL [20,21]. Currently recognized factors predictive of survival after first relapse of childhood ALL include the site of relapse and the duration of first remission. Extramedullary relapse and early relapse were shown to be unfavorable factors [22e25]. This prompted us to review the cytogenetic data obtained for relapsed ALL samples in our laboratory, with a focus on unusual cytogenetic features, which we found to occur more frequently in relapsed ALL.

supplemented with 20% fetal bovine serum (Invitrogen) and 1% penicillin/streptomycin (Invitrogen). In all cases, at least four different cultures were set up, with culture times ranging from 4 hours to 96 hours. Colcemid (Invitrogen) was added before harvesting. Karyotypic analysis was performed on Giemsa-stained metaphases and described according to ISCN 2005 [26]. In cases defined as normal, a minimum of 20 metaphases were fully analyzed. FISH analysis was performed for all patients at initial presentation, or at relapse if presentation data were not available, to rule out the presence of t(11;21) and t(9;22), according to the manufacturer’s recommendations (Abbott Molecular/Vysis, Des Plaines, IL). FISH evaluation for MLL rearrangement was performed with a break-apart rearrangement probe (Vysis) at 11q23 in cases with normal karyotypes or when an abnormality was suspected on chromosome 11 by conventional cytogenetics. A minimum of 200 cells was analyzed per probe set (interphase and metaphase combined). 2.3. Statistical analysis

2. Materials and methods 2.1. Patients We included cytogenetic data from 436 pre-treatment preB-ALL samples at diagnosis and 76 samples at first relapse, which were sent to our laboratory from 8 pediatric centers in Switzerland for analysis over a 12-year period from 1994 to 2006. All patients were younger than 18 years. Of the 76 patients with relapsed ALL, 29 patients had an abnormal clone that was not classifiable in the categories HHD (O50 chromosomes), t(12;21)(p13;22) [ETV6/ RUNX1], t(1;19)(q23;p13) [TCF3-PBX1], t(9;22)(q34;q11) [BCR-ABL1], 11q23 [MLL], hypodiploidy and near-haploidy (!45 chromosomes), and normal karyotype (Fig. 1). We named this subgroup ‘‘cases with non-classical cytogenetic aberrations.’’ Four additional cases with nonclassical aberrations were categorized as ‘‘unknown’’ (Fig. 1) and excluded from further analysis because, due to insufficient material, we could not rule out the presence of t(12;21) or t(9;22). For this subgroup, clinical data were obtained from chart review by the same investigator (L.W.; Table 1). Two additional cases with isolated central nervous system (CNS) relapse presenting with non-classical cytogenetic features were ruled out from this study because cytogenetic data were not available at relapse. Paired presentation cytogenetic data were available in 23/29 cases with non-classical aberrations (Table 2) and in 22 out of the remaining 47 relapse cases (data not shown). 2.2. Cytogenetics and FISH analysis Cytogenetic investigation was performed on either bone marrow aspirations or peripheral blood samples using standard techniques. Briefly, the cells were cultured in RPMI 1640 with Glutamax 1 (Invitrogen, Basel, Switzerland)

Statistical analysis was performed with the SPSS package (version 15.0 for Windows, SPSS, Inc., Chicago, IL). The 29 patients with non-classical cytogenetic aberrations were further subdivided into two subgroups with and without abn(9p) at the time of first relapse. The event-free survival (EFS1) was defined as the time from initial diagnosis until first relapse. The second event-free survival (EFS2) was defined as the time from first relapse until the next subsequent event (second relapse or death from any cause). Overall survival (OS) was defined as the time from the first relapse until death. Patients who did not have a second relapse or died were censored out at the date of their last follow-up. EFS1, EFS2, and OS, including median and 95% confidence intervals (95% CI), were estimated with Kaplan-Meier life tables. Tests for differences of the two, groups 9p and abn(9p), were performed using the log-rank test. P values smaller than 0.05 were considered significant. Cox regression analysis was performed to determine the prognostic significance of several factors in relation to EFS2. Univariate analysis was performed to estimate potential predictors of outcome, including the white blood cells count (WBC) at diagnosis (!50 vs. O50 109/L), the duration of time from diagnosis to first relapse, the risk stratification for the initial treatment at first diagnosis (high risk vs. standard risk), consolidation with bone marrow transplantation (BMT) after the first relapse, and presence of abn(9p).

3. Results 3.1. Distribution of cytogenetic groups at initial presentation and first relapse To identify cytogenetic subgroups with a potential prognostic significance, we first compared the karyotypes in an

L.A. Wehrli et al. / Cancer Genetics and Cytogenetics 189 (2009) 29e36

Presentation n = 436 Normal 7%

31

Relapse n = 76 Unknown 5%

HHD 30.5%

Non-classical 30%

Normal 4%

HHD 21%

Non-classical 38% t(12;21) 20%

< 45 Chr. 1.5% 11q23 1.5%

t(9;22) 2%

t(1;19) 3.5%

t(12;21) 24%

t(1;19) 2.5%

< 45 Chr. 1.5% 11q23 4%

t(9;22) 4%

Fig. 1. Distribution of chromosomal abnormalities in childhood precursor-B-cell ALL. The relative frequencies of chromosomal aberrations found at presentation (436 cases) and at the time of first relapse (76 cases) are shown according to defined cytogenetic subgroups as described in Materials and methods (the ‘‘Patients’’ section).

unselected series of 76 relapsed pre-B-ALL and 436 cases at initial diagnosis that were analyzed in our laboratory over a period of 12 years (Fig. 1). Samples were available in all these cases and karyotype analysis was always performed successfully. From the 76 relapse cases, 45 had karyotype analysis at diagnosis. Clonal cytogenetic abnormalities were observed in 93% of cases at initial presentation and 96% of cases at first relapse. As expected, cytogenetic abnormalities commonly associated with a more favorable outcome were less frequent in relapsed cases. This included cases with HHD (30.5% at diagnosis vs. 21% at relapse) and cases with t(12;21) (24 vs. 20%). In contrast, cases with 11q23 abnormalities increased from 1.5 to 4%, and cases with t(9;22) from 2 to 4%. We also noticed a relative increase of patients with non-classical cytogenetic aberrations at relapse (29/76 patients, 38%) compared to the distribution observed at diagnosis (30%). This subgroup of 29 patients was analyzed in more detail to detect cytogenetic abnormalities of potential clinical relevance (Table 1). From 29 cases, 28 were bone marrow relapses, combined in 5 cases with CNS involvement and in 2 cases with bilateral testicular relapses. One patient had a bilateral testicular relapse without bone marrow involvement by morphology. The majority of patients were male (17/29). Practice varied depending on the treatment center: 19 patients were treated at initial presentation on protocols from the European ALL-BFM (Berlin-Frankfurt-Mu¨nster) study group, 9 patients on protocols from the North American Pediatric Oncology Group (POG), and one patient on the international Interfant-99 protocol. Out of 29 patients with non-classical cytogenetic aberrations at relapse, only 8 (28%) were predicted to be at high risk on the basis of current stratification criteria on their protocol and received the most intensive chemotherapy regimen available. Of note, the new stratification strategy based on the molecular assessment of minimal residual disease, which appears to

be the most powerful predictor of outcome overall on ALL-BFM-2000, did not identify any of the five patients in this subgroup as high-risk cases (Table 1). 3.2. Characterization of a subgroup of patients with non-classical cytogenetic aberrations At relapse, abnormal features were identified in the subgroup of 29 patients with non-classical cytogenetic aberrations on most chromosomes, except for chromosomes 19, 22, and Y (Fig. 2; Table 2). The majority of aberrations observed were unbalanced structural events that would typically result in partial chromosomal deletions. The short arm of chromosome 9 (9p) was most frequently involved with 11 cases (38%) showing partial deletion (Fig. 2; Table 2). In one case, a secondary reciprocal translocation, t(1;9)(p13;p13), which involves the 9p region, was detected only at relapse (case 25). Monosomy 9 was observed in case 1 at relapse and appears to be a secondary event. There was no evidence to suggest a relationship between 9p aberrations and any other chromosomal event. Among cases without abnormalities in 9p, case 5 was shown to have iAMP21 by FISH at first presentation (data not shown). Of note, a second patient had iAMP21 in the initial karyotype analysis but was excluded from this study because his relapse was isolated to the CNS. In 23/29 cases, for which paired diagnosis and relapse samples were available, clonal evolution was a major feature. Two cases, cases 2 and 29, showed no evidence of karyotypic evolution. We have not found evidence for a 9p deletion as a result of clonal evolution. In five cases with paired karyotype data, rare reciprocal translocations were identified both at diagnosis and relapse (Table 3). It is interesting to note that these rearrangements involved 9p21 (case 28), and most likely the immunoglobulin heavy chain locus, with a breakpoint at 14q32 (cases 13 and 18). Interestingly, the second patient with isolated CNS

L.A. Wehrli et al. / Cancer Genetics and Cytogenetics 189 (2009) 29e36

32

Table 1 Clinical characteristics of 31 patients with childhood B-ALL Patients without abn(9p) Case

Age at Dx (yr)

Year of Dx

WBC at Dx (109/L)

Treatment protocol at Dx

Risk group at Dx

EFS1

Location at Relapse

Treatment at first Relapse

BMT

EFS 2

OS

Cause of Death

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

14 5 11 12 7 7 /12 9 10 1 /12 8 10 15 6 2 7 5 1

1997 1997 2003 1994 2000 2000 2000 2002 1993 2003 2003 1996 2000 1991 1995 1995 2001

175.0 90.0 282.5 9.1 25.3 3.4 13.9 440.0 144.0 13.3 29.8 3.8 112.0 5.8 6.9 9.3 6.2

BFM / 1995 BFM / 1995 BFM / 2000 BFM / 1990 BFM / 1995 Interfant / 1999 POG / 1996 POG / 1999 BFM / 1990 POG / 1999 BFM / 2000 BFM / 1995 POG / 1999 BFM / 1990 BFM / 1995 BFM / 1990 POG / 1999

HR HR MR SR MR HR SR HR MR SR MR MR HR SR MR SR SR

12 30 16 35 25 35 49 25 33 10 18 27 38 93 59 77 24

BM BM BM BM BM, testes BM BM BM BM BM, CNS Testes BM BM BM BM BM BM, CNS

BFM REZ 1996 BFM REZ 1996 BFM REZ 2002 BFM REZ 1996 BFM REZ 1996 BFM REZ 2002 POG / 1994 POG / 1994 BFM REZ 1995 COG / 2001 BFM REZ 2002 BFM REZ 1996 POG / 1994 BFM REZ 1996 AML BFM 1998 BFM REZ 1996 COG / 2002

No Yes No No Yes No No Yes No Yes Yes Yes No No Yes No No

3 94 1 23 18 50 40 10 141 7 38 5 19 15 90 20 4

3 94 1 30 29 50 45 14 141 7 38 14 28 16 90 23 7

Leukemia Alive Leukemia Leukemia Leukemia Alive Leukemia Leukemia Alive TRM Alive Leukemia Leukemia Leukemia Alive Leukemia Alive

with abn(9p) 8 2000 1 1993 7 1994 15 1999 9 1998 5 1996 12 2001 11 1996 5 1997 7 1994 3 2003 1 2001

3.0 15.8 35.0 325.0 26.5 12.3 25.9 4.3 415.4 1.8 6.9 35.9

BFM / 1995 BFM / 1990 BFM / 1990 POG / 1994 POG / 1996 POG / 1994 BFM / 2000 POG / 1994 BFM / 1995 BFM / 1990 BFM / 2000 BFM / 2000

MR MR MR HR SR HR MR HR MR SR SR MR

20 19 30 34 29 73 58 35 7 28 24 35

BM, BM, BM BM BM BM BM BM, BM BM, BM BM

BFM REZ 1996 BFM REZ 1990 BFM REZ 1996 COG / 2001 POG / 1994 BFM REZ 1996 BFM REZ Pilot 2002 POG / 1994 BFM REZ 1996 BFM REZ 95/96 BFM REZ 2002 BFM REZ 2002

No Yes No Yes Yes No Yes No No No Yes No

5 11 1 5 9 64 17 18 4 129 7 26

9 19 1 5 9 64 17 24 5 129 7 26

Leukemia Leukemia Infection TRM TRM Alive Alive Leukemia Leukemia Alive Leukemia Leukemia

Patients 18 19 20 21 22 23 24 25 26 27 28 29

CNS CNS

testes CNS

Abbreviations: Dx, diagnosis; WBC, white blood cell count; BM, bone marrow; BMT, bone marrow transplantation for the treatment of first relapse; CNS, central nervous system; BFM, ALL treatment protocol of the Berlin-Frankfurt-Mu¨nster study group, the number after the slash refers to the year of the protocol version; BFM REZ, treatment protocol for relapsed ALL of the Berlin-Frankfurt-Mu¨nster study group, the numbers refer to the year of the protocol version; POG, treatment protocol of the north American Pediatric Oncology Group, the four-digit numbers after the slash refer to the year of the protocol version; COG, Childrens Oncology Group, the four numbers after the slash refer to the year of the protocol version; SR, standard-risk group; MR, intermediate-risk group; HR, high-risk group; TRM: transplant-related mortality. EFS1, EFS2, and OS are as defined in Materials and methods.

relapse who was excluded from this study presented with a t(7;14)(q32;q32), also involving 14q32. 3.3. Treatment outcome after first relapse In the subgroup with non-classical cytogenetic aberrations, the median EFS2 was 18 months [standard deviation (SD) 6.0, 95% CI 6.22e29.79] and the median OS, which was defined as the time from the first relapse until death, was 23 months (SD 6.8, 95% CI 9.71e36.29). At 10 years, the probability of EFS2 was 28.5% (SD 8.8). Of the 15 patients who relapsed a second time, only 1 survived. Median follow-up time of the nine survivors that are still in complete remission from the time point of their first relapse is 64 months (17e41 months). Given that two large pediatric studies identified 9p abnormalities as an adverse prognostic factor for outcome after treatment with first line ALL protocols [11,12], a result

challenged by others [15,16], we evaluated next the association of abn(9p) with worse outcome for the treatment of relapsed ALL. The subgroups with or without abn(9p) were comparable with respect to age, WBC at diagnosis, risk stratification in the initial treatment protocol at diagnosis, and duration of first remission (EFS1), which are factors that could confound our analysis (Table 1). Kaplan-Meier survival analysis for EFS1, one of the most relevant prognostic factors for the treatment of relapsed ALL, resulted in a median EFS1 of 28 months (SD 4.33; 95% CI 19.51e36.49) for abn(9p) versus 33 months (SD 3.29; 95% CI 26.55e39.45) for patients without abn(9p), which was not significant (P50.332). In contrast, event-free survival analysis from the time of first relapse (Fig. 3) showed that patients with abn(9p) relapsed earlier, after a median EFS2 time of 7 months (SD 3.46; 95% CI 0.21e13.79), versus patients without abn(9p), who relapsed after a median time of 20 months (SD 3.43; 95% CI 13.28e26.72).

L.A. Wehrli et al. / Cancer Genetics and Cytogenetics 189 (2009) 29e36

33

Table 2 Karyotype at presentation and relapse Case

Presentation

First relapse

1

46,XX,add(12)(p1?)[cp4]/46,idem,add(13)(q3?)[cp2]/ 45,XX,der(2)?inv(2)(p2?q2?),e15[cp3]/46,XX[1] 46,XY,del(13)(q14.1q31)[8]/46,XY[2] 46,XX,add(13)(p11),inc[5]/46,XX[15] ND

44,XX,der(2)?inv(2)(p2?q2?),del(4)(p14p16),-9, -15[2]/43~44,idem,þ1~4mar[cp5]/46,XX[3] 46,XY,del(13)(q14.1q31)[9]/46,XY[1] 46,XX[22] 47~48,XY,þX,del(13)(q1?q21),add(14)(q32),e21,þ2~3mar[cp5]/ 46,XY[5] 46,XY,add(5)(q31),e17,e21,2~4mar[cp3]/46,XY[8]

2 3 4 5 6 7

46,XY,add(5)(q31)[1]/46~47,XY,add(5)(q31),þ10,þ11, -21,þr,þ0~2mar[cp6]/46,XY[2] 46,XX,add(2)(p2?)[4]/46,idem,add(12)(p12)[2]/46,XX[5] 48~50,XX,þ5,þdel(6)(q1?q2?),e7,add(7)(p15),þ21c,þ22, þmar[cp4]/47,XX,þ21c[15]

8

46,XY,t(15;18)(q26;q11)[10]/46,XY[1]

9

ND

10

46,XY,t(3;13)(q13;q12)[6]/46,idem,psu dic(?;5)(?;p15)[3]/46,XY[1]

11

46,XY,add(2)(q32),add(5)(q31),add(14)(q1?)[6]/46,XY[4]

12 13 14

46,XY[20] 48,XY,þX,þ5,t(14;20)(q32;q13.1)[10]/46,XY[1] ND

15 16 17 18

ND 47,XY,þ21[6]/46,XY[4] 48,XX,þX,þ21[9]/46,XX[1] 46,XY,t(2;14)(q21;q32),del(9)(p21),del(12)(p13)[7]/46,idem, del(9)(p23)[2]/46,XY[1] ND 46,XX,del(9)(p1?),del(18)(q12)[4]/46,XX[6]

19 20

21 22 23 24 25 26 27 28

29

46,X,inv(X)(p21.1q26),der(7)del(7)(p13)add(7)(q34), del(9)(p13)[3]/46,XX[10] 46,XY,der(?8)t(?8;19)(?;q13)add(?8)(q24),del(9)(p22), add(19)(q13)[cp9]/46,XY[2] ND 46,XX,del(9)(p22p24)[6]/46,XX[4] 47~48,XY,del(13)(q12q22)[cp6]/46,XY[5] 46,XY,add(2)(p22),add(9)(p11)[8]/46,XY[2] 45,XY,der(7;9)(q10;q10)[7]/46,XY[5] 47,XY,t(1;12)(q21;q13),þ8,t(8;9)(q24;p21),del(9)(p21)[2]/ 50,sl,þX,þ14,þ22[4]/51,sdl1,þ17[2]/52~53,sdl2, þ21[cp4]/46,XY[1] 46,XX,del(9)(p13.1)[10]

46,XX,add(2)(p2?)[6]/46,XX[5] 46~47,XX,-4,þ5,þ6,del(6)(q1?q2?),-7,add(9)(q34),þ21c,þ3~4mar,inc [cp3] /47,XX,þ21c[17] 46,XY,add(5)(q3?), e7,add(12)(p13),t(13;20)(q12;q13),t(15;18)(q26;q11),þmar[cp7] /46,XY[3] 46,XY,del(11)(q14q25)[3]/47,sl,þ14[2]/47,sdl1,add(14)(q31) [4]/46,XY[1] 46,XY,t(3;13)(q13;q12),der(10;17)(p10;q10),der(16)t(10;16)(q21;q11) [7]/46,XY[1] Right testis46,XY,t(1;10)(q25;q24)[2]/46,idem,del(9)(q12q22) [4]/46,XY,i(17)(q10)[cp2]/46,XY[2] Left testis46,XY,i(17)(q10)[cp3]/46,XY[7] 46,XY,del(11)(q23)[4]/46,XY[30] 48,XY,þX,þ5,del(7)(p15),t(14;20)(q32;q13.1)[10]/46,XY[1] 46,XX,del(6)(q22~23),add(17)(q22)[3]/46,idem,del(12)(p13)[2]/ 46,XX[5] 46,XX,del(13)(q12q14),inv(14)(q11q22)[10] 47,XY,þ21[5]/48,idem,þ19[4]/46,XY[2] 48,XX,þX,del(7)(p13p22),þ21[7]/46,XX[4] 46,XY,t(2;14)(q21;q32),del(9)(p21),inv(9)(p23q12)[9]/46,XY[1] 45,XX,dic(9;20)(p11;q11)[8]/45,idem,del(9)(p21)[4]/46,XX[4] 46,X,-X,del(9)(p13),del(18)(q21),þ1~4mar[cp4]/ 46,idem,add(7)(q2?),del(10)(q22),del(11)(q22),e12,e14, e17,del(17)(q21),edel(18)(q21)[cp7]/46,XX[3] 46,X,inv(X)(p21.1q26),der(7)del(7)(p13)add(7)(q34),del(9)(p13) [3]/45~47,idem,add(3)(q27)[cp3]/46,XX[10] 47,X,eY,þder(5)t(1;5)(q21;q31),e8,del(9)(p13),þ13,þmar[cp7]/ 46,XY[4] 46,XY,t(1;8)(q42;q13),del(7)(p13),idic(9)(p11)[8]/46,XY[2] 46,XX,del(9)(p22p24)[6]/46,idem,add(6)(q2?)[3]/46,XX[3] 48,XY,t(1;9)(p13;p13),þ5,þ10,der(17)t(1;17)(q12;p11.1)[8]/46,XY[2] 45,XY,der(2)add(2)(p22)del(2)(q14q21),e7,add(9)(p11)[5]/46,XY[5] 45,XY,der(7;9)(q10;q10)add(7)(q34)[9]/46,XY[1] 48,XY,þX,t(1;12)(q21;q13),þ8,t(8;9)(q24;p21),del(9)(p21)[6]/48,idem, del(14)(q21q32)[4] 46,XX,del(9)(p13.1)[4]/46,XX[6]

Note: Cases 1e17, without abn(9p); cases 18e29, with abn(9p). Abbreviations: ND, no data.

Accordingly, the median OS time from the time of first relapse was 9 months for patients with abn(9p) and 29 months for patients without abn(9p). The difference in EFS2 probability was not significant (log-rank P 5 0.178) with 22.2% EFS2 at 10 years for patients with abn(9p) (SD 12.8%) and 32.9% EFS2 for patients without abn(9p) (SD 12.1%). In the univariate analysis, the duration of first remission was most closely associated with outcome after relapse treatment, but the Cox regression analysis did not show

clear significance. (P 5 0.065). As expected, the presence of abn(9p) was not significantly associated with worse outcome (P 5 0.187), but taken together with the marked decrease in EFS2 time for patients with abn(9p), this result may suggest a trend, given the small number of patients included in this analysis. Other variables, such as elevated WBC at diagnosis (P 5 0.397), consolidation with BMT after the first relapse (P 5 0.325), and inclusion in the high-risk treatment arm during the initial treatment

L.A. Wehrli et al. / Cancer Genetics and Cytogenetics 189 (2009) 29e36

34

Relapse

Number of appearance

Presentation 12

12

10

10

8

8

6

6

4

4

2

2

0

0

2

2

4

4

6

p q

6 2

4

6

8 10 12 14 16 18 20 22 X Y

2

4

6

8 10 12 14 16 18 20 22 X Y

Chromosome

Chromosome

Fig. 2. Distribution of uncommon abnormalities by chromosome in 29 patients with relapsed precursor-B-cell ALL. Frequencies of chromosome abnormalities found at presentation and at relapse for the subgroup with non-classical cytogenetic aberrations. The number of cases showing an aberration on a particular chromosome is plotted with bars above the horizontal line for the short arm and below the horizontal line for the long arm. If several aberrations were recorded on the same arm of a chromosome for the same patient, they were counted only once.

4. Discussion Given the limited cytogenetic data available for relapsed acute lymphoblastic leukemia, we initiated a retrospective analysis of relapsed precursor-B ALL (pre-B-ALL) cases that were processed in our laboratory within the last decade. We focused on a group of 29 patients with non-classical cytogenetic aberrations that were represented in increased proportion at relapse. The outcome of this group was poor and comparable to results that have been reported for unfavorable prognostic groups after treatment of bone marrow relapse [27]. The fact that 21/29 patients (72%) were not treated on the high-risk arm of their respective treatment protocols suggests that current risk factors do not detect a significant subgroup of patients with non-classical cytogenetic aberrations that might benefit from more intensive treatment. We therefore reviewed carefully the cytogenetic data at diagnosis and relapse for the presence of characteristic lesions.

We detected 9p abnormalities in 80 cases in a total of 436 cases at diagnosis (18%, data not shown) and 18 cases of a total of 76 cases at first relapse (24%, data not shown). Karyotypes with abn(9p) were more frequent in the subgroup of patients with non-classical cytogenetic aberrations at relapse, (12/29 cases, 41%), compared to 6 of the remaining 47 cases (13%) at relapse (Table 4 and data not shown). The prognostic significance of abn(9p) in childhood ALL is still controversial. Two large pediatric studies reported 9p abnormalities in 11% and 8% of presentation cases, respectively, which were associated with adverse prognosis [11,12]. Detection of abn(9p) by single-nucleotide polymorphism arrays in pre-B-ALL was more sensitive. The most frequent minimally deleted region included the CDKN2 gene in 1.0

0.8

EFS 2 (%)

protocol (P 5 0.581), did not influence EFS2. The results obtained for overall survival were similar (data not shown). Our collective results indicate that the presence of abn(9p) is associated with a shorter duration of second remission, which implies that abn(9p) at relapse may be associated with more aggressive subtypes of the disease, acknowledging the limitations due to our small sample size.

0.6

0.4

Table 3 Rare reciprocal chromosome translocations identified in paired presentation and relapse cases

0.2

Case

Translocation

Cause of death

0.0

8 10 13 18 28

t(15;18)(q26;q11) t(3;13)(q13;q12) t(14;20)(q32;q13.1) t(2;14)(q21;q32) t(1;12)(q21;q13), t(8;9)(q24;p21)

Leukemia TRM Leukemia Leukemia Leukemia

Abbreviation: TRM, transplant-related mortality.

non-abn(9p)

abn(9p)

0

20

40

60

80

100

120

140

Months from first Relapse Fig. 3. Event-free survival after relapse in children with ALL according to the presence or absence of abn(9p). Probability of EFS2 for precursorB-ALL cells with abn(9p) (n 5 12 cases) and without abn(9p) [non-abn(9p), n 5 17]. Tick marks indicate censoring based on the last follow-up visit.

L.A. Wehrli et al. / Cancer Genetics and Cytogenetics 189 (2009) 29e36 Table 4 Abn(9p) in 76 cases with relapsed ALL Subgroup

abn(9p) / total

HHD t(12;21) t(9;22) 11q23 !45 Chr. Normal Unknown Non-classical cytogenetic aberrations

1/16 4/15 0/5 0/3 0/1 0/3 1/4 12/29

Note: Abbreviations as defined in the ‘‘Patients’’ section of Materials and methods.

9p21.3 in 34% of pre-B-ALL patients [19], but no correlation with outcome was reported. A similar frequency was reported for the loss of the CDKN2 locus in pre-B-ALL by FISH [15] and by determination of gene dosage and methylation status of CDKN2 and neighboring genes using PCR [16], which in both studies was not associated with worse outcome. Given the relatively small size of our subgroup with non-classical abnormalities, a conclusive outcome analysis was not possible. The much shorter duration of second remission in patients with non-classical aberrations, including abn(9p), is strongly suggestive of a more aggressive ALL subtype. Thus, the significance of 9p abnormalities warrants further investigation. Among abnormalities that were detectable in paired diagnosis and relapse cases with non-classical cytogenetic aberrations, we noticed six reciprocal chromosomal translocations (Table 3). None of these patients were long-term survivors. It is interesting to note that three translocations involved 14q32, the site of the immunoglobulin heavychain locus (IGH). Recurrent IGH locus translocations are increasingly being reported in childhood ALL and may constitute relevant leukemogenic features leading to the dysregulation of genes such as the CCAAT/enhancer binding protein beta (CEBPB), as we suspect in case 13 (Table 3), and the helix-loop-helix inhibitor protein ID4 [28,29]. A t(2;14)(q21;q32), similar to the translocation identified in case 18 (Table 3), has been cloned in nonHodgkin lymphoma, resulting in a juxtaposition of the BCL-6 gene next to the IGH enhancer [30]. To our knowledge, neither the t(2;14)(q21;q32) nor the t(7;14)(q32;q32) have been reported in childhood ALL. New prognostic factors are needed to improve treatment outcome for this group of patients. The most recent approach to stratify patients for treatment intensity based on the in vivo response to chemotherapy by assessment of the minimal residual disease (MRD) [31] may constitute a valuable alternative to identify those patients who are at higher risk in the subgroup with non-classical cytogenetic aberrations. However, in all cases where MRD was included for risk stratification (five patients treated on the ALL-BFM-2000 study protocol, Table 1), none of these relapsed patients could be predicted to belong to the high-risk group. This observation

35

indicates that risk stratification based on MRD may not substitute the need for additional cytogenetic risk factors to guide treatment decisions for a relevant subgroup in pre-BALL. One example of a new cytogenetic marker of potential clinical relevance is iAMP21, which was shown to be associated with adverse outcome in childhood ALL [17,18]. Interestingly, two of the relapsed patients with non-classical cytogenetic aberrations did show iAMP21 at first presentation by FISH, illustrating that thorough cytogenetic analysis at diagnosis can still provide relevant information to guide future treatment. Thus, systematic high-resolution cytogenetic studies should be considered in an effort to identify new prognostic markers for this subgroup of ALL patients with non-classical cytogenetic aberrations.

Acknowledgments We thank Karin Hartmann and Lukas Hegi for assistance with chart review and Joelle Tchinda for reviewing the manuscript. References [1] Mrozek K, Heerema NA, Bloomfield CD. Cytogenetics in acute leukemia. Blood Rev 2004;18:115e36. [2] Harrison CJ, Foroni L. Cytogenetics and molecular genetics of acute lymphoblastic leukemia. Rev Clin Exp Hematol 2002;6:91e113. [3] Secker-Walker LM, Chessells JM, Stewart EL, Swansbury GJ, Richards S, Lawler SD. Chromosomes and other prognostic factors in acute lymphoblastic leukaemia: a long-term follow-up. Br J Haematol 1989;72:336e42. [4] Trueworthy R, Shuster J, Look T, Crist W, Borowitz M, Carroll A, Frankel L, Harris M, Wagner H, Haggard M, Mosijczuk A, Pullen J, Steuber P, Land V. Ploidy of lymphoblasts is the strongest predictor of treatment outcome in B-progenitor cell acute lymphoblastic leukemia of childhood: a Pediatric Oncology Group study. J Clin Oncol 1992;10:606e13. [5] Heerema NA, Nachman JB, Sather HN, Sensel MG, Lee MK, Hutchinson R, Lange BJ, Steinherz PG, Bostrom B, Gaynon PS, Uckun F. Hypodiploidy with less than 45 chromosomes confers adverse risk in childhood acute lymphoblastic leukemia: a report from the children’s cancer group. Blood 1999;94:4036e45. [6] Gibbons B, MacCallum P, Watts E, Rohatiner AZ, Webb D, Katz FE, Secker-Walker LM, Temperley IJ, Harrison CJ, Campbell RH, Nash R, Broadbent V, Chessels JM. Near haploid acute lymphoblastic leukemia: seven new cases and a review of the literature. Leukemia 1991;5:738e43. [7] Seeger K, Adams HP, Buchwald D, Beyermann B, Kremens B, Niemeyer C, Ritter J, Schwabe D, Harms D, Schrappe M, Henze G. TEL-AML1 fusion transcript in relapsed childhood acute lymphoblastic leukemia. The Berlin-Frankfurt-Munster Study Group. Blood 1998;91:1716e22. [8] Secker-Walker LM, Prentice HG, Durrant J, Richards S, Hall E, Harrison G. Cytogenetics adds independent prognostic information in adults with acute lymphoblastic leukaemia on MRC trial UKALL XA. MRC Adult Leukaemia Working Party. Br J Haematol 1997;96:601e10. [9] Arico M, Valsecchi MG, Camitta B, Schrappe M, Chessells J, Baruchel A, Gaynon P, Silverman L, Janka-Schaub G, Kamps W, Pui CH, Masera G. Outcome of treatment in children with Philadelphia chromosome-positive acute lymphoblastic leukemia. N Engl J Med 2000;342:998e1006.

36

L.A. Wehrli et al. / Cancer Genetics and Cytogenetics 189 (2009) 29e36

[10] Pui CH, Chessells JM, Camitta B, Baruchel A, Biondi A, Boyett JM, Carroll A, Eden OB, Evans WE, Gadner H, Harbott J, Harms DO, Harrison CJ, Harrison PL, Heerema N, Janka-Schaub G, Kamps W, Masera G, Pullen J, Raimondi SC, Richards S, Riehm H, Sallan S, Sather H, Shuster J, Silverman LB, Valsecchi MG, Vilmer E, Zhou Y, Gaynon PS, Schrappe M. Clinical heterogeneity in childhood acute lymphoblastic leukemia with 11q23 rearrangements. Leukemia 2003;17:700e6. [11] Heerema NA, Sather HN, Sensel MG, Liu-Mares W, Lange BJ, Bostrom BC, Nachman JB, Steinherz PG, Hutchinson R, Gaynon PS, Arthur DC, Uckun FM. Association of chromosome arm 9p abnormalities with adverse risk in childhood acute lymphoblastic leukemia: A report from the Children’s Cancer Group. Blood 1999;94:1537e44. [12] Hann I, Vora A, Harrison G, Harrison C, Eden O, Hill F, Gibson B, Richards S. Determinants of outcome after intensified therapy of childhood lymphoblastic leukaemia: results from Medical Research Council United Kingdom acute lymphoblastic leukaemia XI protocol. Br J Haematol 2001;113:103e14. [13] Bertin R, Acquaviva C, Mirebeau D, Guidal-Giroux C, Vilmer E, Cave H. CDKN2A, CDKN2B, and MTAP gene dosage permits precise characterization of mono- and bi-allelic 9p21 deletions in childhood acute lymphoblastic leukemia. Genes Chromosomes Cancer 2003;37:44e57. [14] Strefford JC, Worley H, Barber K, Wright S, Stewart AR, Robinson HM, Bettney G, van Delft FW, Atherton MG, Davies T, Griffiths M, Hing S, Ross FM, Talley P, Saha V, Moorman AV, Harrison CJ. Genome complexity in acute lymphoblastic leukemia is revealed by array-based comparative genomic hybridization. Oncogene 2007;26:4306e18. [15] van Zutven LJ, van Drunen E, de Bont JM, Wattel MM, Den Boer ML, Pieters R, Hagemeijer A, Slater RM, Beverloo HB. CDKN2 deletions have no prognostic value in childhood precursorB acute lymphoblastic leukaemia. Leukemia 2005;19:1281e4. [16] Mirebeau D, Acquaviva C, Suciu S, Bertin R, Dastugue N, Robert A, Boutard P, Mechinaud F, Plouvier E, Otten J, Vilmer E, Cave H. The prognostic significance of CDKN2A, CDKN2B and MTAP inactivation in B-lineage acute lymphoblastic leukemia of childhood. Results of the EORTC studies 58881 and 58951. Haematologica 2006;91: 881e5. [17] Attarbaschi A, Mann G, Panzer-Grumayer R, Rottgers S, Steiner M, Konig M, Csinady E, Dworzak MN, Seidel M, Janousek D, Moricke A, Reichelt C, Harbott J, Schrappe M, Gadner H, Haas OA. Minimal residual disease values discriminate between low and high relapse risk in children with B-cell precursor acute lymphoblastic leukemia and an intrachromosomal amplification of chromosome 21: the Austrian and German acute lymphoblastic leukemia Berlin-Frankfurt-Munster (ALL-BFM) trials. J Clin Oncol 2008;26:3046e50. [18] Moorman AV, Richards SM, Robinson HM, Strefford JC, Gibson BE, Kinsey SE, Eden TO, Vora AJ, Mitchell CD, Harrison CJ. Prognosis of children with acute lymphoblastic leukemia (ALL) and intrachromosomal amplification of chromosome 21 (iAMP21). Blood 2007;109:2327e30. [19] Mullighan CG, Goorha S, Radtke I, Miller CB, Coustan-Smith E, Dalton JD, Girtman K, Mathew S, Ma J, Pounds SB, Su X, Pui CH, Relling MV, Evans WE, Shurtleff SA, Downing JR. Genome-wide analysis of genetic alterations in acute lymphoblastic leukaemia. Nature 2007;446:758e64.

[20] Heerema NA, Palmer CG, Weetman R, Bertolone S. Cytogenetic analysis in relapsed childhood acute lymphoblastic leukemia. Leukemia 1992;6:185e92. [21] Vora AJ, Potter AM, Anderson LM, Lilleyman JS. Frequency and importance of change in blast cell karyotype in relapsing childhood lymphoblastic leukemia. Pediatric hematology and oncology 1994;11:379e86. [22] Gaynon PS, Qu RP, Chappell RJ, Willoughby ML, Tubergen DG, Steinherz PG, Trigg ME. Survival after relapse in childhood acute lymphoblastic leukemia: impact of site and time to first relapseethe Children’s Cancer Group Experience. Cancer 1998;82:1387e95. [23] Chessells JM, Veys P, Kempski H, Henley P, Leiper A, Webb D, Hann IM. Long-term follow-up of relapsed childhood acute lymphoblastic leukaemia. Br J Haematol 2003;123:396e405. [24] Einsiedel HG, von Stackelberg A, Hartmann R, Fengler R, Schrappe M, Janka-Schaub G, Mann G, Hahlen K, Gobel U, Klingebiel T, Ludwig WD, Henze G. Long-term outcome in children with relapsed ALL by risk-stratified salvage therapy: results of trial acute lymphoblastic leukemia-relapse study of the Berlin-FrankfurtMunster Group 87. J Clin Oncol 2005;23:7942e50. [25] Malempati S, Gaynon PS, Sather H, La MK, Stork LC. Outcome after relapse among children with standard-risk acute lymphoblastic leukemia: Children’s Oncology Group study CCG-1952. J Clin Oncol 2007;25:5800e7. [26] Shaffer LG, Tommerup N. ISCN 2005: An International System for Human Cytogenetic Nomenclature (2005). Basel: S. Karger AG, 2005. [27] Gaynon PS. Childhood acute lymphoblastic leukaemia and relapse. Br J Haematol 2005;131:579e87. [28] Akasaka T, Balasas T, Russell LJ, Sugimoto KJ, Majid A, Walewska R, Karran EL, Brown DG, Cain K, Harder L, Gesk S, Martin-Subero JI, Atherton MG, Bruggemann M, Calasanz MJ, Davies T, Haas OA, Hagemeijer A, Kempski H, Lessard M, Lillington DM, Moore S, Nguyen-Khac F, Radford-Weiss I, Schoch C, Struski S, Talley P, Welham MJ, Worley H, Strefford JC, Harrison CJ, Siebert R, Dyer MJ. Five members of the CEBP transcription factor family are targeted by recurrent IGH translocations in B-cell precursor acute lymphoblastic leukemia (BCP-ALL). Blood 2007;109: 3451e61. [29] Russell LJ, Akasaka T, Majid A, Sugimoto KJ, Loraine Karran E, Nagel I, Harder L, Claviez A, Gesk S, Moorman AV, Ross F, Mazzullo H, Strefford JC, Siebert R, Dyer MJ, Harrison CJ. t(6;14)(p22;q32): a new recurrent IGHa translocation involving ID4 in B-cell precursor acute lymphoblastic leukemia (BCP-ALL). Blood 2008;111:387e91. [30] Walker LC, Morrison MJ, Parfitt R, Crossen PE. Translocation (2;14) associated with complex rearrangements of the Ig heavy chain in non-Hodgkin lymphoma. Cancer Genet Cytogenet 2001;128: 137e40. [31] Flohr T, Schrauder A, Cazzaniga G, Panzer-Grumayer R, van der Velden V, Fischer S, Stanulla M, Basso G, Niggli FK, Schafer BW, Sutton R, Koehler R, Zimmermann M, Valsecchi MG, Gadner H, Masera G, Schrappe M, van Dongen JJ, Biondi A, Bartram CR. Minimal residual disease-directed risk stratification using real-time quantitative PCR analysis of immunoglobulin and T-cell receptor gene rearrangements in the international multicenter trial AIEOP-BFM ALL 2000 for childhood acute lymphoblastic leukemia. Leukemia 2008;1e12.