Detailed clonality analysis of relapsing precursor B acute lymphoblastic leukemia: implications for minimal residual disease detection

Leukemia Research 25 (2001) 1033– 1045 www.elsevier.com/locate/leukres

Detailed clonality analysis of relapsing precursor B acute lymphoblastic leukemia: implications for minimal residual disease detection Ai-Hong Li a, Richard Rosenquist a, Erik Forestier b, Jack Lindh c, Go¨ran Roos a,* a

Department of Medical Biosciences, Pathology, Umea˚ Uni6ersity, 90187 Umea˚, Sweden b Department of Clinical Sciences, Pediatrics, Umea˚ Uni6ersity, 90187 Umea˚, Sweden c Department of Radiation Sciences, Oncology, Umea˚ Uni6ersity, 90187 Umea˚, Sweden Received 16 January 2001; accepted 7 April 2001

Abstract Genetic instability has important implications for detection of minimal residual disease (MRD) when the target is a clonal genetic marker revealed at diagnosis. A successful MRD detection approach requires a stable marker and for lymphoid leukemias clonal rearrangements of immunoglobulin (Ig) and T cell receptor (TCR) genes are commonly used. In the present study, Ig heavy chain (IgH) and TCR (g and d) genes were studied in 18 consecutive, relapsing precursor-B ALL patients. At least one clonal rearrangement was found in all cases at presentation (IgH 94%, TCRg 39% and TCRd 28%). An altered rearrangement pattern between diagnosis and relapse was demonstrated in 14 patients (78%). At least one stable molecular target was found in 13 out of 18 cases (72%). Clonal differences between diagnostic and relapse samples were explained by: (1) loss of original rearrangements; (2) VH to DJH joining; (3) VH gene replacement; (4) appearance of new rearrangements. In two cases with apparently new IgH gene rearrangements at relapse extended sequencing of the diagnostic samples revealed minor clonal rearrangements identical to the relapse clones. Interestingly, one patient displayed instability on both the IgH and TCR gene loci, whereas a stable Igk rearrangement was found at presentation and relapse. These data show that clonal diversity is common in precursor-B ALL and strongly suggest that MRD detection should include multiple gene targets to minimize false-negative samples. Even so, five of our 18 relapse cases (28%) lacked stable clonal markers and should have been unsuitable for MRD detection. © 2001 Elsevier Science Ltd. All rights reserved. Keywords: Precursor B ALL; Clonality; Ig/TCR gene rearrangement; MRD

1. Introduction Immunoglobulin (Ig) and T cell receptor (TCR) gene rearrangements have been widely used as markers of clonality in human lymphoid neoplasm. In B-lineage acute lymphoblastic leukemia (ALL), clonal rearrangements have been detected in the majority of cases at the Ig heavy chain (IgH) gene level (80 – 96%), whereas Abbre6iations: ALL, acute lymphoblastic leukemia; Ig, immunoglobulin; IgH, immunoglobulin heavy chain; IgL, immunoglobulin light chain; PCR, polymerase chain reaction; MRD, minimal residual disease; TCR, T cell receptor. * Corresponding author. Tel: +46-90-7851801; fax: + 46-907852829. E-mail address: [email protected] (G. Roos).

clonality was present to a lesser extent at the TCR gene loci [1–4]. These gene rearrangements provide unique clonal markers for tracing residual leukemic cells during chemotherapy. Recently, several groups have reported that detection of minimal residual disease (MRD) at certain time points during treatment has prognostic implications for the clinical outcome in childhood ALL [5–10]. The presence of MRD at the end of induction chemotherapy and before consolidation treatment can predict a forthcoming relapse [8,10]. Therefore, it has been proposed that treatment strategies for ALL should be designed according to established MRD-based risk groups to reduce the relapse hazard [5,8,11]. Moreover, Panzer-Grumayer et al. [12] reported that childhood ALL patients with negative or

0145-2126/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 5 - 2 1 2 6 ( 0 1 ) 0 0 0 7 2 - 8

1034

A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045

low levels of MRD at day 15 (during induction therapy) had a better prognosis than patients with higher levels of MRD. A variety of cellular features, such as the expression of gene fusion transcripts, immunophenotype and antigen receptor gene rearrangements, can be used to detect MRD in patients with ALL [9,13]. Most molecular genetic techniques utilize PCR for amplification of tumor-specific junctional regions of rearranged IgH, TCRg or TCRd genes to detect MRD with a high sensitivity (one target cell per 103 – 106 cells) [14,15]. However, these target genes may be altered or additional rearrangements may become dominant (‘clonal evolution’) during the course of disease [5,14– 16]. Several groups have shown that clonal Ig or TCR rearrangements were altered in a large proportion (10 – 44%) of B lineage ALL cases between presentation and relapse using Southern blotting or PCR analysis [3,17 –20]. Nucleotide sequence analysis of Ig and/or TCR gene rearrangements has provided insights into the molecular mechanisms responsible for changed rearrangement patterns. Such mechanisms can be: (1) continuing rearrangement of IgH gene by VH addition to a preexisting D– JH complex (VH to D – JH joining); (2) VH –VH gene replacement; (3) an ‘open-and-shut mechanism’ where N-sequences are added or excluded in the VH –N –DH junction; (4) generation of new clones/subclones; (5) disappearance of the original (sub)clones [19– 25]. Ongoing alterations might lead to false negative results when using junctional regions of the IgH or TCR gene rearrangements as targets for MRD detection. In the present study, we have analyzed the clonal rearrangements of multiple gene targets (IgH and TCRg/d for all cases, supplemented with Ig light chain genes for selected cases) in 18 relapsing precursor B ALL patients at presentation and relapse. Altered rearrangement patterns between presentation and relapse were observed at a high frequency (78%) and were due to various molecular mechanisms. In 72% of the cases, at least one gene target was found to be stable and could be used for MRD detection whereas five out of 18 cases (28%) lacked stable clonal markers and thus would have been inappropriate for MRD detection.

2. Material and methods

2.1. Patient material Bone marrow (BM), peripheral blood (PB) or lymph node (LN) samples were obtained from 18 patients (nine children and nine adults) who presented with precursor B ALL between 1991 and 1996 at the University Hospital, Umea˚ , and who subsequently relapsed. Tumor samples were collected at diagnosis and relapse

in all cases. From one patient (case 1), a testis tumor sample taken at a second relapse was obtained. Patients were selected on the basis of availability of fresh frozen material for further molecular analysis. The diagnosis of precursor B ALL was based on standard morphologic evaluation of bone marrow smears, cytochemical stains, flow cytometric immunophenotyping and karyotyping.

2.2. DNA preparation High molecular weight DNA was extracted according to the manufacture’s instructions (Nucleon II, Scotlab, UK). The DNA was precipitated with ethanol and dissolved in ultra pure water. The quality of DNA was controlled by PCR amplification of a 536 bp fragment of the b-globin gene for cases with no clonal PCR bands using Ig and TCR consensus primers.

2.3. PCR amplification 2.3.1. IgH- and TCR (k, l) -PCR The primers used in the present study to identify IgH, TCRg and TCRd (Vd2 –Dd3) gene rearrangements, PCR conditions and the method used for detection of PCR products have been detailed previously [4]. PCR products showing one distinct band after polyacrylamide gel electrophoresis were interpreted as monoclonal. Polyclonality was characterized by DNA fragment length heterogeneity seen as broad bands or a smear on the gels. 2.3.2. Ig light chain (IgL) -PCR The Vk and Vl family-specific primers used hybridize to the framework region 1 (FR1) of the IgL gene. PCR amplification with Vk or Vl primers together with Jk or Jl primers was performed in separate reactions resulting in a product of approximately 350 bp. The primers used were selected according to Ku¨ ppers and coworkers [26,27]. Amplification of Vk and Vl gene rearrangements was carried out in a 50 ml reaction volume containing 100–500 ng of genomic DNA, 50 mmol/l KCl, 10 mmol/l Tris –HCl pH 8.4, 1.5 mmol/l MgCl2, 0.01% gelatin, 0.125 mM of each primer, 2.5 mmol/l dNTP, and 5 mCi [h-32P]dCTP (specific activity of 3000 Ci/ mmol, Amersham, UK). After denaturation at 95°C for 5 min, 2.5 U Taq polymerase (Boehringer Mannhein, Germany) was added at 70°C, followed by 34 cycles of 90 s at 95°C, 30 s at 61°C, 80 s at 72°C for Vk gene amplification and 34 cycles of 30 s at 95°C, 30 s at 61°C, 30 s at 72°C for Vl gene amplification. For all amplifications the program ended with 5 min of incubation at 72°C. All amplifications were performed in a programmable thermal controller 100 (MJ Research Inc, USA).

A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045

Detection and analysis of PCR products were performed as described for IgH gene rearrangements [4].

2.4. Cloning and DNA sequencing Selected PCR products were gel-purified, cloned by ligation into either the pT7 Blue (R) vector (Novagen, USA) or the pGEM-T vector (Promega, USA), and transformed into either Escherichia coli cells (Novagen, USA) or JM 109 competent bacteria (Promega, USA). After blue-white screening, 6– 20 white colonies were picked randomly and cultured in LB medium. Plasmid DNA was extracted by plasmid miniprep technique (Qiagen, Germany) and sequenced using BigDye terminator cycle sequencing reaction kit (Perkin-Elmer, Applied Biosystems). Sequencing reactions were analysed on an automatic sequencing system (ABI377, Applied Biosystems). Nucleotide sequences were analysed using Lasergene software (DNASTAR, Madison, WI). Consensus sequences were established by sequencing at least three colonies for each PCR product. The derived Ig gene and TCR (g and d) gene sequences were identified by comparison with the corresponding germline genes using Genbank (EMBL), BLAST or FASTA searches. For identification of D genes comparison was made with published germline genes according to criteria as previously described [20].

2.5. Criteria for detection of clonal di6ersity Criteria used to classify a leukemia as showing clonal diversity between diagnosis and relapse were: (1) change in the number of clonal PCR products, (2) band size difference by single strand conformation polymorphism (SSCP) analysis and (3) utilization of different VH family genes.

3. Results

3.1. Clonal rearrangements of Ig and TCR genes detected by PCR amplification All 18 patients in the study demonstrated at least one or several clonal rearrangements of the IgH and/or TCR genes at diagnosis (IgH 94%, TCRg 39% and TCRd 28%). Concomitant clonal rearrangements of IgH and TCR genes were found in ten of 18 patients (56%). The characteristics of the Ig and TCR gene rearrangements detected are summarized in Table 1. Fourteen out of 18 patients (78%) displayed an altered pattern of the clonal rearrangements between diagnosis and relapse (IgH 71%, TCRg 13% and TCRd 67%). In four patients (case 1, 2, 4 and 16), differences in both the Ig and TCR gene loci were demonstrated. The

1035

appearance of new rearrangements was found at relapse in eight patients for Ig genes and in two patients for TCR. In nine patients, one or more of the original rearrangements were lost at relapse. Two adult patients (case 10 and 13) showed no clonal rearrangement at relapse for any gene target, despite an intact DNA as judged by amplification of the control b-globin gene. All together, at least one stable molecular target was found in 13 out of 18 patients (72%). Only four patients (case 8, 11, 14 and 17) presented unchanged rearrangement patterns. In five patients, none of the original rearrangements could be detected at relapse.

3.2. Clonal characterization deduced from nucleotide sequencing Cases with altered patterns of gene rearrangements (from diagnosis to relapse) were selected for nucleotide sequencing and the data revealed that four molecular events were involved in our patients (see Table 1): VH to DJH joining rearrangement (or VH to JH joining), VH –VH gene replacement, loss of original rearrangement and appearance of de novo rearrangement, as described in earlier studies [20,25,28]. Some of these cases are shortly described below.

3.2.1. Case 1 At diagnosis, VH3, VH4, VH6 rearrangements were present and at least 16 different IgH rearrangements were identified, 15 of which displayed a common DJH region. Nine different VH germline genes were found to be attached to the identical DJH region. One additional VH3 rearrangement was also present. At a first BM relapse, a new VH6 rearrangement was shown using the DK1 –JH6 complex seen at diagnosis and none of the original rearrangements was detected (as previously reported in Ref. [20]). In a later testis relapse, a diagnostic VH3 rearrangement reappeared. One identical TCRd rearrangement was shown in the BM and testis relapses and from this rearrangement clone specific primers were selected and used for analysis of the diagnostic samples. As can be seen in Fig. 1 the same clonal pattern appeared in two of the three diagnostic samples using this sensitive approach. An additional TCRd rearrangement was shown in the testis sample. These data demonstrated a very complex and unusual clonality pattern at diagnosis making it impractical to choose proper targets for MRD evaluation. Furthermore, only in the testis relapse sample one diagnostic clone (VH3) reappeared. A schematic illustration of this case is given in Fig. 2. 3.2.2. Case 4 Two diagnostic VHDJH rearrangements were undetectable at relapse. Regarding TCRg, three different clones were identified at diagnosis and only one of these

1036

Table 1 Characteristics of Ig gene and TCR gene rearrangements in 18 relapsing precursor B ALL patients Patient

Children 1

2 3

5 6

7 8 9 Adults 10 11 12 13 14

7-0

0-5 11-8 4-6 2-0 5-10

10-11 3-0 10-11

26-0 30-0 63-0 52-0 64-0

Diagnosis

Time to relapse (months)

Material

Ig genes

TCR genes

IgH(FR1, 2, 3)

Igk

Igl

TCRg

TCRd

Stable molecular targets

Molecular events based on sequence data

PreB PreB-relapse 1 PreB-relapse 2

25 50

BM, LM, PB BM Testis

V3, V4, V6 V6 V3

− − nd

− − nd

− − −

− + +2b

Yes (IgH)

VH to DJH joininga Loss and gain of rearrangements Loss and gain of rearrangements

PreB PreB-relapse

20

BM BM

V2, FR2 V4, V6, FR3

nd −

nd −

− −

+ −

No

VH to DJH joining Loss and gain of rearrangements

PreB PreB-relapse

40

BM, PB BM

V2, V3, FR2 V2, V3, FR2

nd nd

nd nd

− +

− −

Yes (IgH)

PreB PreB-relapse

72

BM BM

V3, FR2, FR3 −

k1 k1

− l3

+2b +2bc

+ −

Yes (Igk, TCRg)

PreB PreB-relapse

27

BM, PB BM

V5 V4

− −

− −

− −

− −

No

BM, LN

V3, V4, FR2, FR3 V3b, V4b, FR2, FR3

nd

nd

−

+3b

nd

nd

−

+3b

PreB

c

VH–VH replacement Yes (IgH-FR2,3, TCRd)

PreB-relapse

68

BM, LN

PreB PreB-relapse

52

BM, PB BM, PB

V1, V3, V6, FR3 nd V1, V6, FR3 nd

nd nd

− −

− −

Yes (IgH)

PreB PreB-relapse

39

BM BM

V3, FR2, FR3 V3, FR2, FR3

nd nd

nd nd

− −

+3b +3b

Yes (IgH, TCRd)

PreB PreB-relapse

30

BM BM

V2 V2, V6, FR2

nd nd

nd nd

+4b +4b

− −

Yes (IgH, TCRg)

PreB PreB-relapsee

13

BM, LN, PB BM

V1, V3 −

nd −

nd −

− −

− −

No

PreB PreB-relapse

37

BM BM

V2 V2

nd nd

nd nd

+2b +2b

− −

Yes (IgH, TCRg)

PreB PreB-relapse

16

BM BM

FR2 −

nd nd

nd nd

+ +

− −

Yes (TCRg)

PreB PreB-relapse

10

BM BM

FR3 −

nd nd

nd nd

− −

− −

No

27

BM BM

V2 V2

nd nd

nd nd

+ +

− −

Yes (IgH, TCRg)

PreB PreB-relapse

Loss and gain of rearrangements

Loss and gain of rearrangements VH to DJH/VH to JH joining

Gain of rearrangement

A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045

4

Age (years-months)

Table 1 (Continued) Patient

16 17 18

23-0

72-0 39-0 59-0

Diagnosis

Time to relapse (months)

PreB

Material

BM

PreB-relapse

30

BM

PreB PreB-relapse

8

BM BM

PreB PreB-relapse PreB PreB-relapse 1 PreB-relapse 1

Ig genes

TCR genes

Stable molecular targets

Molecular events based on sequence data

IgH(FR1, 2, 3)

Igk

Igl

TCRg

TCRd

V4, V6, FR2, FR3 V1, V3, V6b, FR2

nd

nd

−

−

nd

nd

−

−

Yes (IgH-FR2)

Loss and gain of rearrangements

− −

− l3

− l3

− −

+ +b

No

Loss and gain of rearrangements

d

23

PB BM

V3, V5, FR3 V3, V5, FR3d

nd nd

nd nd

+ +

− −

Yes (IgH, TCRg)

5 6

BM BM BM

V1, V3, FR2 V1, V3, FR2 V1, V3, V5, FR2, FR3

nd nd nd

nd nd nd

+ + +

− − −

Yes (IgH, TCRg)

Gain of rearrangement

Ig, immunoglobulin; IgH, immunoglobulin heavy chain; FR, framework region; TCR, T cell receptor; Igk, immunoglobulin kappa; Igl, immunoglobulin lambda; preB, precursor B cell; BM, bone marrow; PB, peripheral blood; LN, lymph node, nd, not done. a Previously reported in Ref. [20] b Changes in SSCP pattern c Bands with unequal intensity d Large clonal PCR fragment. Sequence data showed that the JH primer had annealed to the JH4 gene instead of the JH3 gene which resulted in an amplification of a non-coding region (327 bp), a phenomenon that we have previously described in Ref. [4]. e Fifty percent of the lymphoblastic cells expressed CD13 and CD15.

A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045

15

Age (years-months)

1037

1038

A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045

Two stable IgH rearrangements (VH1 and VH6) were identified at diagnosis and relapse, whereas nucleotide sequence analysis of an altered VH3 rearrangement suggested a mechanism of VH to DJH and VH to N–JH joining rearrangements (D gene deleted, see Fig. 3).

Fig. 1. In case 1 no clonal similarity was found between diagnosis and relapse using TCRd consensus primers (Table 1). PCR amplification and SSCP (single strand conformation polymorphism) analysis using relapse clone specific TCRd primers showed however positivity in PB and BM samples from diagnosis. Lane A: relapse sample from BM. Lanes B, C and D: diagnostic samples from PB, LN and BM, respectively

rearrangements reappeared at relapse. Thus, only one out of five clonal IgH/TCR gene sequences at diagnosis was found in the relapse sample. Interestingly, a stable IgL(k) rearrangement was found which could have served as an appropriate MRD target (Fig. 2).

3.2.3. Case 6 In this case three TCRd rearrangements were preserved at relapse and could theoretically have been useful as MRD targets. Concerning the IgH gene targets altered patterns seen on the SSCP gel using VH3/JH and VH4/JH primers could be explained by ongoing VH –VH replacements for this patient (Fig. 3). 3.2.4. Case 7

3.2.5. Case 15 An original VH6 rearrangement was undetected at relapse whereas new VH1, VH3 and VH6 clones appeared. The new clones were undetectable at diagnosis using clone specific primers chosen from the relapse clones. A VH4 rearrangement could be amplified by two sets of primers (VH4/JH and FR2/JH) at diagnosis, but by only one set (FR2/JH) at relapse. A homogeneous sequence using the DP70, DK4, and JH4 genes was identified at both diagnosis and relapse, but no mutations in the FR2 and JH primer regions. This indicated mutations in the VH4 primer region in relapse sample. 3.2.6. Case 16 Diagnostic clonal patterns of TCRd and Igl were completely changed in the relapse sample. At relapse, none of 15 individual TCRd sequences was related to the original TCRd rearrangement. Furthermore, a new Igl rearrangement was found at relapse, which was undetectable in the diagnostic sample by clone specific amplification, suggesting that a new clone had evolved. 3.2.7. Case 18 Stable VH1, VH3 and TCRg rearrangements were demonstrated together with a new, seemingly unrelated VH5 rearrangement at relapse, indicating that all the

Fig. 2. A schematic illustration of the clonality patterns in two relapsing precursor B ALL cases. Case 1: two clonal IgH gene rearrangements reappeared individually at different time points after diagnosis. *: Nine different VH germline genes (DP38, DP47, DP50, DP58, DP63, DP70, DP74, DP78 and COS 8) were found to be attached to the identical DJH region (previously reported in Ref. [20]). §: One TCR d rearrangement was found at both relapses (BM and testis) which could be amplified in the diagnostic samples using relapse clone specific primers. An additional TCR d rearrangement was only shown in the relapse sample from testis. Case 4: diagnostic IgH rearrangements were absent at relapse and only one Igk and one TCRg rearrangement were preserved from diagnosis to relapse.

A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045 Fig. 3. Nucleotide sequences of IgH rearrangements in cases 6 and 7. n: Number of colonies sequenced. D-sequences are underlined. Nucleotide identity is indicated by a dash. Case 6: one VH3 rearrangement composed of the COS 8, the DXP’1 and JH6 segments was identical at diagnosis and relapse. However, two VH4 segments (HCAC and DP72) from 12 colonies in the diagnostic BM sample and two other VH4 segments (DP63 and DP71) from 10 colonies in the relapse LN sample replaced the VH3 segment (COS 8). The N(6 – 8 bp) –DXP’1–JH6 complex was identical. The other VH3 rearrangement utilising the DP48, DN4 and JH4 germline elements was similar in diagnostic and relapse samples except one point mutation (C “ T) in the D–JH junction at relapse. Moreover, at relapse the VH3 (DP48) segment was replaced by two VH4 segments (VH443 and DP71) and the N(3 – 5 bp) –DN4–JH4 region remained unchanged. Sequence data in this case thus revealed a mechanism of VH –VH replacement. Case 7: one VH1 (DP25) and three VH3 gene segments (DP35, DP49 and DP58) were attached to a common D21 –9–JH6 region. The DP35 germline gene used in four identical colonies was attached to a 19 nucleotides long homologous sequence in the latter part of the VH –JH junction region but no obvious D-sequence could be determined. These data suggested a mechanism of VH to DJH and VH to JH joining. 1039

1040 A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045 Fig. 4. Schematic illustration of VH gene clonal patterns in samples from case 2. (A) Clonal IgH sequences in diagnostic and relapse BM samples, including extended sequence data on the VH4 and VH6 genes at diagnosis. n; number of colonies sequenced. D-sequences are underlined. (B) Two unrelated VHDJH rearrangements were identified at diagnosis. At relapse three previously undetected rearrangements (VH6/DXP’1/JH6, VH6/DN4/JH6 and DP66/D21– 9/JH4) were observed. By extended sequencing of the diagnostic sample one relapse rearrangement (VH6/DXP’1/JH6) was found in two out of 15 colonies and the remaining 13 colonies showed different VH germline genes (DP70, DP79, H435, VHVI) attached to two identical D –JH rearrangements (DXP’1–JH6 and DN1–JH4) with preserved D –N–JH junctional regions.

A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045

Fig. 4. (Continued)

1041

1042

A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045

Fig. 5. A schematic illustration of the clonal evolution in case 5. Minor clonal rearrangements at diagnosis were found to cause the relapse.

diagnostic IgH and TCRg gene targets would have been appropriate markers for MRD detection.

3.3. Minor clonal rearrangement at diagnosis as a source for later relapse In selected cases with apparently new IgH gene rearrangements at relapse we tested if the relapse clone could be found in the diagnostic tumor sample using extended sequencing. For this purpose the FR1 gene family of interest was amplified. The often diffuse or hardly visible band in the agarose gel was cut out, cloned and transformed into bacteria and at least 15 colonies were sequenced.

3.3.1. Case 2 At diagnosis two different VH2 rearrangements and at relapse one VH4 and two VH6 rearrangements were found. No distinct bands by radioactive PCR amplification using VH4/JH and VH6/JH primers were observed at diagnosis, but the sequence data showed that two out of 15 colonies (13%) were identical to the one of relapse VH6 rearrangements. For VH4 no identical sequence

was found. An interesting finding by extended sequencing was that VH to DJH joining rearrangement was involved in two diagnostic clones and one of these clones was retained with same DJH segment at relapse, whereas the other clone disappeared. For this case, sequence data and a schematic illustration are given in Fig. 4.

3.3.2. Case 5 A VH5 rearrangement was shown at diagnosis but a polyclonal pattern appeared by nucleotide sequencing of eight colonies. Two unrelated VH4 rearrangements were identified at relapse. Extended sequencing of VH4 rearrangements in the diagnostic sample demonstrated that five out of 20 colonies (25%) were identical to one of the VH4 clones (H434/three fused DH genes/JH5) and one colony (5%) was identical to the second relapse clone (DP66/D21 –9/JH4). The remaining 14 colonies showed unrelated, polyclonal sequences. The extended sequence analysis is illustrated as a schema in Fig. 5. The data suggested that apparently ‘new’ rearrangements at relapse were present as minor clonal rearrangements at diagnosis.

A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045

4. Discussion In the present study diagnostic samples from all the 18 ALL patients presented at least one clonal IgH or TCR rearrangement (IgH 94%, TCRg 39% and TCRd 28%). This is in line with previous studies showing a very high frequency of detectable clonal rearrangements in ALL [3,4,29,30]. The analysis of matched diagnostic and relapse samples from these patients showed a frequent clonal diversity in IgH (71%) and TCRd loci (67%) whereas clonal TCRg genes were altered in only 13%. In accordance with this finding Beizhuizen et al. observed that the incidence of clonal evolution at the TCRd gene locus in pre-B ALL patients was higher than at the TCRg locus analysed by Southern blot analysis [18]. The more stable TCRg target can be explained by the absence of D segments and, on average, fewer non germline encoded nucleotide additions [14,17]. This study and those by Beizhuizen, Macintyre, and Cave suggested that whenever possible, the use of TCRg as clonal marker is desirable for large scale MRD monitoring in combination of TCRd and/or IgH genes. As described in earlier studies [19– 22,24,25,31,32], different molecular mechanisms can account for clonal evolution in progressive/relapsing ALL. Firstly, the loss of one or more rearrangements at relapse could be interpreted as a loss of malignant clones eradicated by chemotherapy or could be due to mutations in the primer sequences of these specific genes. Secondly, in three patients homology within the diversity gene coding regions and the absence of N-region homology at VH – D junction were in accordance with a mechanism of VH to DJH joining [22]. Thirdly, VH – VH gene replacement was observed in one case showing different VH genes joined to an identical DJH region, leaving a variable degree of N-region homology at VH –D junction. Fourthly, appearance of new rearrangements was found at relapse. Wasserman et al. [28] proposed that the finding of new clones during chemotherapy, especially taken together with the disappearance of an original clone, may result from a marked growth advantage for the new clones or from other molecular changes in the new clone. In one case (case 16), completely changed TCRd and Igl rearrangements between diagnosis and relapse suggested no relation between these leukemic cell populations. In addition, no amplifiable gene targets at relapse in two adult patients (case 10 and 13) could also be explained by the appearance of new clones. In case 10, the relapse was due to a biphenotypic leukemia showing expression of CD13 and CD15 not detected at diagnosis. For case 13 the absence of amplifiable clones at relapse could be explained by a mismatch between primers and target sequences or by a partial DJH rearrangement.

1043

Despite the heterogeneous rearrangement patterns, at least one stable molecular target was traced in 72% of our patients. A certain degree of clonal stability is supported by previous reports indicating that even relapses occurring more than 5– 10 years after diagnosis are due to re-emergence of the original leukemic clone [33,34]. Vora et al. [33] suggested that a late clinical recurrence could arise when there was either a diminished immune surveillance for maintaining minimal residual leukemia in limbo or a return to active cell cycle triggered by the acquisition of new mutations within the dormant clone. Studies of IgH and TCR gene rearrangements for MRD detection have suggested that the use of different combinations of PCR primers could detect markers present only in a subpopulation of blasts including those expressed by a minority of the cells [14]. Case 4 showed heterogeneity on the IgH and TCR gene targets but also a single Igk rearrangement that remained identical, indicating one common leukemic population characterized by its monoclonal Igk rearrangement. Similarly, one case was reported to have one identical Igk gene rearrangement throughout the course of leukemia, including three relapses, together with the presence of multiple IgH rearrangements and two TCRd rearrangements [35]. The authors assumed that leukemic clones were derived from a common progenitor with one Igk rearrangement and with the IgH and TCRd genes in germline configuration. Moreover, Kubagawa et al. [36] characterized four EBVtransformed human fetal bone marrow (FBM) pre-B cell lines that synthesized k L chain, but no m H chains. Except such rare cases of human leukemia reported [25,37], these findings are unusual events compared with the predominant rearrangement sequence (H chain “ k L chain “ l L chain hierarchy). Interestingly, in diagnostic samples from two patients minor clonal IgH rearrangements could be detected by extended sequencing, which were identical to the clones found at relapse. Using relapse clone specific primers, only one out of three cases showed positivity at diagnosis. These findings support the hypothesis that apparently new rearrangements observed at relapse indeed can originate from an undetected diagnostic clone [34]. Standard MRD approaches would not have detected the ‘relevant’ clone in these patients. In summary, our study showed that changes in rearrangement patterns on multiple gene loci occurred at a high frequency (78%) in precursor B-ALL patients between presentation and relapse due to different molecular mechanisms. Thirteen out of 18 patients (72%) demonstrated at least one stable clonal target, but five cases (28%) lacked this feature. This finding has obvious implications in a situation when routine MRD screenings are included as a stratifying parameter in the therapy protocols. One important question to answer is

1044

A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045

if this high fraction of cases unsuitable for MRD detection can be tolerated in the clinical setting. Previous studies have demonstrated a predictive value for MRD detection in ALL but it must be emphasized that the patient studied were selected based on the clonality pattern at diagnosis [5,18,24,38]. The patients studied here represented all relapsing ALL cases at our hospital during one time period except cases where diagnostic or relapse samples were missing. It is also important to stress that in order to achieve that at least 70% of future patients are analyzed for MRD properly as many gene targets as possible should be studied, and allele specific detection only based on direct sequencing of the diagnostic sample is likely to be insufficient.

Acknowledgements This work was supported by grants from the Children’s Cancer Foundation, the Swedish Cancer Research Foundation, and the Lions’ Cancer Research Foundation. We would like to acknowledge the skilful technical assistance of Anita Lindstro¨ m. Conception and design: G Roos, A Hong Li, R Rosenquist, E Forestier, J Lindh. Analysis and interpretation of data: G Roos, A Hong Li, R Rosenquist, E Forestier. Drafting the article: G Roos, A Hong Li. Critical revision of the article for important intellectual content: R Rosenquist, J Lindh. Final approval of article: G Roos, A Hong Li, R Rosenquist, E Forestier, J lindh. Provision of study materials or patients: R Forestier. Administrative, technical or logistic support: A Hong Li. Collection or assembly of data: G Roos, A Hong Li, R Rosenquist.

References [1] Taylor JJ, Rowe D, Williamson IK, Christmas SE, Proctor SJ, Middleton PG. Detection of T-cell receptor gamma chain V gene rearrangements using the polymerase chain reaction: application to the study of clonal disease cells in acute lymphoblastic leukemia. Blood 1991;77:1989 –95. [2] Potter MN, Steward CG, Oakhill A. The significance of detection of minimal residual disease in childhood acute lymphoblastic leukaemia. Br J Haematol 1993;83:412 – 8. [3] Steward CG, Goulden NJ, Katz F, et al. A polymerase chain reaction study of the stability of Ig heavy-chain and T-cell receptor delta gene rearrangements between presentation and relapse of childhood B-lineage acute lymphoblastic leukemia. Blood 1994;83:1355 –62. [4] Li AH, Rosenquist R, Forestier E, et al. Clonal rearrangements in childhood and adult precursor B acute lymphoblastic leukemia: a comparative polymerase chain reaction study using multiple sets of primers. Eur J Haematol 1999;63:211 – 8. [5] Campana D, Pui CH. Detection of minimal residual disease in acute leukemia: methodologic advances and clinical significance. Blood 1995;85:1416 –34 (see comments).

[6] Campana D, Coustan-Smith E. Detection of minimal residual disease in acute leukemia by flow cytometry. Cytometry 1999;38:139 – 52. [7] van Dongen JJ, Szczepanski T, de Bruijn MA, et al. Detection of minimal residual disease in acute leukemia patients. Cytokines Mol Ther 1996;2:121 – 33. [8] van Dongen JJ, Seriu T, Panzer-Grumayer ER, et al. Prognostic value of minimal residual disease in acute lymphoblastic leukaemia in childhood. Lancet 1998;352:1731 –8 (see comments). [9] van Dongen JJ, Macintyre EA, Gabert JA, et al. Standardized RT-PCR analysis of fusion gene transcripts from chromosome aberrations in acute leukemia for detection of minimal residual disease. Report of the BIOMED-1 Concerted Action: investigation of minimal residual disease in acute leukemia. Leukemia 1999;13:1901 – 28. [10] Cave H, van der Werff ten Bosch J, Suciu S, et al. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia. European Organization for Research and Treatment of Cancer — Childhood Leukemia Cooperative Group. New Engl J Med 1998;339:591 – 8 (see comments). [11] Roberts WM, Zipf TF, Kitchingman GR, Tubergen DG, Estrov Z. Monitoring residual disease in acute lymphoblastic leukemia: therapeutic implications. Cytokines Mol Ther 1995;1:65 –9. [12] Panzer-Grumayer ER, Schneider M, Panzer S, Fasching K, Gadner H. Rapid molecular response during early induction chemotherapy predicts a good outcome in childhood acute lymphoblastic leukemia. Blood ;95 2000;95:790 – 4 (published erratum appeared in Blood May 15;95(10):3010). [13] Coustan-Smith E, Behm FG, Sanchez J, et al. Immunological detection of minimal residual disease in children with acute lymphoblastic leukaemia. Lancet 1998;351:550 – 4 (see comments). [14] Cave H, Guidal C, Rohrlich P, et al. Prospective monitoring and quantitation of residual blasts in childhood acute lymphoblastic leukemia by polymerase chain reaction study of delta and gamma T-cell receptor genes. Blood 1994;83:1892 – 902. [15] Roberts WM, Estrov Z, Kitchingman GR, Zipf TF. Detection of minimal residual disease in all: biology, methods, and applications. Cancer Treat Res 1996;84:149 – 66. [16] Macintyre E, d’Auriol L, Amesland F, et al. Analysis of junctional diversity in the preferential V delta 1-J delta 1 rearrangement of fresh T-acute lymphoblastic leukemia cells by in vitro gene amplification and direct sequencing. Blood 1989;74:2053 – 61. [17] Macintyre EA, d’Auriol L, Duparc N, Leverger G, Galibert F, Sigaux F. Use of oligonucleotide probes directed against T cell antigen receptor gamma delta variable-(diversity)-joining junctional sequences as a general method for detecting minimal residual disease in acute lymphoblastic leukemias. J Clin Invest 1990;86:2125 – 35. [18] Beishuizen A, Verhoeven MA, van Wering ER, Hahlen K, Hooijkaas H, van Dongen JJ. Analysis of Ig and T-cell receptor genes in 40 childhood acute lymphoblastic leukemias at diagnosis and subsequent relapse: implications for the detection of minimal residual disease by polymerase chain reaction analysis. Blood 1994;83:2238 – 47. [19] Marshall GM, Kwan E, Haber M, et al. Characterization of clonal immunoglobulin heavy chain and T cell receptor gamma gene rearrangements during progression of childhood acute lymphoblastic leukemia. Leukemia 1995;9:1847 – 50. [20] Rosenquist R, Thunberg U, Li AH, et al. Clonal evolution as judged by immunoglobulin heavy chain gene rearrangements in relapsing precursor-B acute lymphoblastic leukemia. Eur J Haematol 1999;63:171 – 9. [21] Kitchingman GR. Immunoglobulin heavy chain gene VH –D junctional diversity at diagnosis in patients with acute lymphoblastic leukemia. Blood 1993;81:775 – 82.

A.-H. Li et al. / Leukemia Research 25 (2001) 1033–1045 [22] Steenbergen EJ, Verhagen OJ, van Leeuwen EF, von dem Borne AE, van der Schoot CE. Distinct ongoing Ig heavy chain rearrangement processes in childhood B-precursor acute lymphoblastic leukemia. Blood 1993;82:581 – 9. [23] Steenbergen EJ, Verhagen OJ, van Leeuwen EF, van den Berg H, von dem Borne AE, van der Schoot CE. Frequent ongoing T-cell receptor rearrangements in childhood B-precursor acute lymphoblastic leukemia: implications for monitoring minimal residual disease. Blood 1995;86:692 – 702. [24] Choi Y, Greenberg SJ, Du TL, et al. Clonal evolution in B-lineage acute lymphoblastic leukemia by contemporaneous VH– VH gene replacements and VH – DJH gene rearrangements. Blood 1996;87:2506 –12. [25] Davi F, Gocke C, Smith S, Sklar J. Lymphocytic progenitor cell origin and clonal evolution of human B-lineage acute lymphoblastic leukemia. Blood 1996;88:609 –21. [26] Kuppers R, Zhao M, Rajewsky K, Hansmann ML. Detection of clonal B cell populations in paraffin-embedded tissues by polymerase chain reaction. Am J Pathol 1993;143:230 –9. [27] Kuppers R, Willenbrock K, Rajewsky K, Hansmann ML. Detection of clonal lambda light chain gene rearrangements in frozen and paraffin-embedded tissues by polymerase chain reaction. Am J Pathol 1995;147:806 –14. [28] Wasserman R, Yamada M, Ito Y, et al. VH gene rearrangement events can modify the immunoglobulin heavy chain during progression of B-lineage acute lymphoblastic leukemia. Blood 1992;79:223 – 8 (see comments). [29] Deane M, Pappas H, Norton JD. Immunoglobulin heavy chain gene fingerprinting reveals widespread oligoclonality in B-lineage acute lymphoblastic leukaemia. Leukemia 1991;5:832 – 8. [30] Forestier E, Nordenson I, Lindstrom A, Roos G, Lindh J. Simultaneous immunoglobulin/T-cell receptor gene rearrange-

[31]

[32]

[33] [34]

[35]

[36]

[37]

[38]

1045

ments and multiclonality in childhood acute lymphoblastic leukemia. Acta Paediatr 1994;83:319 – 26. Kleinfield R, Hardy RR, Tarlinton D, Dangl J, Herzenberg LA, Weigert M. Recombination between an expressed immunoglobulin heavy-chain gene and a germline variable gene segment in a Ly 1+ B-cell lymphoma. Nature 1986;322:843 – 6. Reth M, Gehrmann P, Petrac E, Wiese P. A novel VH to VHDJH joining mechanism in heavy-chain-negative (null) pre-B cells results in heavy-chain production. Nature 1986;322:840 –2. Vora A, Frost L, Goodeve A, et al. Late relapsing childhood lymphoblastic leukemia. Blood 1998;92:2334 – 7. Lo Nigro L, Cazzaniga G, Di Cataldo A, et al. Clonal stability in children with acute lymphoblastic leukemia (ALL) who relapsed five or more years after diagnosis. Leukemia 1999;13:190 – 5. Stolz F, Panzer S, Fischer S, Panzer-Grumayer ER. Oligoclonal immunoglobulin heavy-chain and T-cell receptor delta rearrangements persist in a recurrent acute lymphoblastic leukemia with one immunoglobulin kappa rearrangement as a clonal marker. Mod Pathol 1999;12:819 – 26. Kubagawa H, Cooper MD, Carroll AJ, Burrows PD. Lightchain gene expression before heavy-chain gene rearrangement in pre-B cells transformed by Epstein-Barr virus. Proc Natl Acad Sci USA 1989;86:2356 – 60. Beishuizen A, Hahlen K, Hagemeijer A, et al. Multiple rearranged immunoglobulin genes in childhood acute lymphoblastic leukemia of precursor B-cell origin. Leukemia 1991;5:657 –67. Green E, McConville CM, Powell JE, et al. Clonal diversity of Ig and T-cell-receptor gene rearrangements identifies a subset of childhood B-precursor acute lymphoblastic leukemia with increased risk of relapse. Blood 1998;92:952 – 8.