Prognostic significance of minimal residual disease in infants with acute lymphoblastic leukemia treated within the Interfant-99 protocol

Transcription

1 ORIGINAL ARTICLE (2009) 23, & 2009 Macmillan Publishers Limited All rights reserved /09 $ Prognostic significance of minimal residual disease in infants with acute lymphoblastic leukemia treated within the Interfant-99 protocol VHJ Van der Velden 1,14, L Corral 2,3,14, MG Valsecchi 4,14, MWJC Jansen 1,14, P De Lorenzo 2,3,4, G Cazzaniga 2,3, ER Panzer-Grümayer 5,6, M Schrappe 7, A Schrauder 7, C Meyer 8, R Marschalek 8, LL Nigro 9, M Metzler 10, G Basso 11, G Mann 6, ML Den Boer 12, A Biondi 2,3, R Pieters 12,13 and JJM Van Dongen 1 1 Department of Immunology, Erasmus MC, University Medical Center, Rotterdam, The Netherlands; 2 Pediatric Clinic, San Gerardo Hospital, University of Milan-Bicocca, Milan, Italy; 3 M Tettamanti Research Center, Monza, Italy; 4 Department of Clinical Medicine and Prevention, Interfant Trial Data Center, University of Milan-Bicocca, Monza, Italy; 5 Children s Cancer Research Institute, Vienna, Austria; 6 St Anna Children s Hospital, Vienna, Austria; 7 Department of Pediatrics, University Hospital Schleswig-Holstein, Campus Kiel, Kiel, Germany; 8 Institute of Pharmaceutical Biology/ZAFES/DCAL, University of Frankfurt, Frankfurt/Main, Germany; 9 Center of Pediatric Hematology Oncology, Azienda Policlinico University of Catania, Catania, Italy; 10 Hospital for Children and Adolescents, University of Erlangen, Erlangen, Germany; 11 Department of Pediatrics, University of Padova, Padova, Italy; 12 Department of Pediatric Oncology/Hematology, Erasmus MC-Sophia Children s Hospital, Rotterdam, The Netherlands and 13 Dutch Childhood Oncology Group, The Hague, The Netherlands Acute lymphoblastic leukemia (ALL) in infants younger than 1 year is a rare but relatively homogeneous disease (B80% MLL gene rearranged, B70% CD10-negative) when compared with childhood and adult ALL. Several studies in children and adults with ALL have shown that minimal residual disease (MRD) status is a strong and independent prognostic factor. We therefore evaluated the prognostic significance of MRD in infant ALL. Ninety-nine infant patients treated according to the Interfant-99 protocol were included in this study. MRD was analyzed by real-time quantitative PCR analysis of rearranged immunoglobulin genes, T-cell receptor genes and MLL genes at various time points (TP) during therapy. Higher MRD levels at the end of induction (TP2) and consolidation (TP3) were significantly associated with lower disease-free survival. Combined MRD information at TP2 and TP3 allowed recognition of three patients groups that significantly differed in outcome. All MRD-high-risk patients (MRD levels X10 4 at TP3; 26% of patients) relapsed. MRD-low-risk patients (MRD level o10 4 at both TP2 and TP3) constituted 44% of patients and showed a relapse-rate of only 13%, whereas remaining patients (MRDmedium-risk patients; 30% of patients) had a relapse rate of 31%. Comparison between the current Interfant-06 stratification at diagnosis and the here presented MRD-based stratification showed that both stratifications recognized different subgroups of patients. These data indicate that MRD diagnostics has added value for recognition of risk groups in infant ALL and that MRD diagnostics can be used for treatment intervention in infant ALL as well. (2009) 23, ; doi: /leu ; published online 12 February 2009 Keywords: acute lymphoblastic leukemia; infant; minimal residual disease; Ig/TCR gene rearrangements; MLL; outcome Introduction Acute lymphoblastic leukemia (ALL) in infants (aged o12 months) is a rare but relatively homogeneous disease when compared with childhood or adult ALL, characterized by Correspondence: Professor JJM Van Dongen, Department of Immunology, Erasmus MC, University Medical Center Rotterdam, Dr Molewaterplein 50, GE Rotterdam 3015, The Netherlands. j.j.m.vandongen@erasmusmc.nl The data reported in this manuscript are original, have not been published previously and have not been submitted for publication elsewhere. 14 These authors contributed equally to this work. Received 5 December 2008; accepted 12 December 2008; published online 12 February 2009 MLL gene rearrangements (B80%), CD10-negative immunophenotype (B70%), immature rearrangement patterns of immunoglobulin (Ig) and T-cell receptor (TCR) genes (B65%), and high tumor load at presentation (X cells/l; B60%). 1 4 Infant ALL thereby is biologically different from ALL in older children. 5,6 Also clinically, infant ALL comprises a distinct subgroup, characterized by an event-free survival (EFS) of around 47%, which is in contrast to the EFS in older children with ALL (approximately 80%). 7 The clinical outcome of infants with ALL is dependent on several characteristics. In the recent Interfant-99 protocol, it was shown that MLL rearrangements, very high white blood cell count ( cells/l at diagnosis) and age o6 months were independently associated with inferior outcome. 8 On the basis of these three risk factors, infant ALL patients could be divided into three groups, which differed significantly in 4-year EFS: patients with germline MLL (74.0; 20% of patients); patients with all three risk factors (19.8; 20% of patients); and remaining patients (44.8; 60% of patients). 8 Several studies have shown that detection of minimal residual disease (MRD) is a strong and independent prognostic factor in the heterogeneous groups of childhood (41 year of age) and adult ALL In particular, analysis of the kinetics of tumor load reduction appears to be highly informative to identify patients with a high chance of relapse and those who will likely remain in complete remission. 9 Whether MRD also has prognostic significance in infant ALL is not yet known. Furthermore, MRD analysis in infant ALL may be hampered by the immature Ig/TCR gene rearrangement pattern and the high frequency of oligoclonality. 3 However, in these patients the rearranged MLL gene can serve as MRD-PCR target, if the exact DNA breakpoint region in the MLL gene has been identified To evaluate whether MRD has prognostic significance in infant ALL, we analyzed MRD by real-time quantitative-pcr analysis of rearranged Ig/TCR and MLL genes in infant ALL patients treated according to the Interfant-99 protocol. Materials and methods Patients and treatment protocol Ninety-nine patients were included in this study, based on the availability of DNA material at diagnosis as well as at least two

2 1074 follow-up time points (TP). All patients were enrolled into the Interfant-99 protocol 8 and their parents gave written consent according to the Declaration of Helsinki. This study was approved by the Ethics Committee of all collaborating institutions and registered at ClinicalTrials.gov, number NCT , and at controlled-trials.com, number ISRCTN Cell sampling Bone marrow samples were obtained at diagnosis, during induction (day 15, TP1; n ¼ 47), after induction (TP2; n ¼ 84), after MARAM (TP3; n ¼ 60), after OCTADD (TP4; n ¼ 38), after VIMARAM (TP5; n ¼ 11), after maintenance 1 (TP6; n ¼ 6) and after maintenance 2 (TP7; n ¼ 5) (Figure 1a). Samples were shipped to the central MRD-PCR laboratories (Vienna (Austrian patients), Monza (Italian patients) and Rotterdam (all other patients). Detection of MRD Ig/TCR gene rearrangements were detected in the diagnostic specimens as described previously On the basis of the junctional region of the identified rearrangements, patientspecific primers were designed and tested for sensitivity and specificity. 20,22,24 27 At diagnosis, MLL rearrangements were evaluated by split signal FISH and/or PCR analysis of fusion gene transcripts. 3 For patients with MLL rearrangements, the break point at the DNA level was analyzed as described previously ,28 30 Patientspecific primers were designed and used in combination with MLL probes for real-time quantitative-pcr analysis. Due to the large variation in break point locations, most patients required the design of specific oligonucleotides. 31 Real-time quantitative- PCR data were analyzed according to the guidelines of the European Study Group on MRD detection in ALL (ESG-MRD- ALL). 32 If analysis of the same bone marrow sample, with two or three different PCR targets, resulted in different tumor-load estimations, the highest MRD level was assumed to be most accurate and used for all subsequent data analysis. 9,14 Statistical analysis To compare Interfant-99 patients with and without MRD evaluations with respect to known prognostic features, w 2 -tests were performed. Disease-free survival (DFS) was defined as the time from first complete remission (or from TP2 or TP3 for MRD-based subgroups) to last follow-up or to relapse, death in complete continuous remission or second malignancy. At each time point, all patients evaluated for MRD were by definition Figure 1 (a) Treatment scheme of the Interfant-99 protocol with minimal residual disease (MRD) sampling time points. (b) MRD positivity at the different follow-up time points (all patients). (c) MRD positivity at the different follow-up time points for MLL-positive (left panel) and MLL-negative patients (right panel).

3 failure free from diagnosis to the considered time point. Three patients resistant to induction who were evaluated for MRD at TP1 or TP2 were excluded from the analyses. DFS curves were computed according to the Kaplan Meier estimator, and their standard errors (s.e.) according to Greenwood formula. Outcome in subgroups was compared by the log-rank test. Analysis of prognostic relevance of MRD was conducted with the Cox model and Wald test. All tests were two sided. Analysis were performed using SAS Results Patients Out of the 478 patients enrolled in the Interfant-99 protocol, 99 patients were included in the MRD study. These 99 patients were all B-lineage ALL with no significant differences in age, gender, MLL status, CD10 distribution and 5-year DFS as compared with the patients not studied for MRD. The median follow-up time was 58 months (range: 1 102). In 83 patients (84%), at least one MRD-PCR target reached a quantitative range of p10 4. From the remaining patients, 11 had a quantitative range of and five had a quantitative range of In these 16 patients, MRD data were only taken into account if quantitative MRD data were obtained (that is, MRD level within quantitative range of the assay). 32 Comparison between MLL and Ig/TCR targets In total, 149 Ig/TCR targets and 49 MLL targets were used for MRD analysis. A quantitative range of at least 10 4 was reached for 74 Ig/TCR targets (52%) and for 35 MLL targets (71%). A sensitivity of at least 10 4 was reached in 134 Ig/TCR targets (94%) and 47 MLL targets (96%). In 40 patients, both Ig/TCR and MLL rearrangements were used as MRD-PCR target. As MLL gene rearrangements are related to the oncogenic process, such rearrangements are likely present in all leukemic cells. In contrast, Ig/TCR gene rearrangements (particularly IGH gene rearrangements) in infant ALL are often oligoclonal and may hence only be present in a subset of the leukemic cells. 3 We therefore analyzed whether MRD data obtained by Ig/TCR gene rearrangements were comparable with MRD data obtained by MLL gene rearrangements. Concordant results (defined as less than threefold difference between the two approaches) were obtained in 80 out of 123 samples (65%) (Figure 2). In 23 samples (19%) seemingly discrepant MRD data were observed: in 8 samples (7%) MRD levels obtained by both targets were outside the quantitative range of the assay (positive versus negative MRD results) and in 15 samples (12%) MRD data could be quantified by one target, but could only be scored positive by analysis of the other target. In the latter samples, MRD levels were always just around the quantitative range of the assay. The discrepant MRD data obtained in these 19% of samples therefore do not reflect true differences in MRD level but rather reflect minor variations around and below the limit of the quantitative range. In six samples (5%), MRD data obtained by Ig/TCR rearrangements moderately exceeded the MRD level obtained by analysis of MLL rearrangements by more than factor three (range: 3.5- to 5.7-fold). Finally, in 14 samples (11%), MRD data obtained by Ig/TCR rearrangements were clearly lower than MRD data obtained by MLL rearrangements, indicating that only a subclone was monitored by Ig/TCR rearrangement analysis (range of difference in MRD level: 3.9- to fold). Figure 2 Comparison of minimal residual disease (MRD) results obtained by real-time quantitative (RQ)-PCR analysis of MLL gene rearrangements with RQ-PCR analysis of Ig/TCR gene rearrangements. Kinetics of MRD positivity during treatment The frequency of MRD-positive patients as well as the level of MRD positivity decreased with time (Figure 1b). After induction therapy (TP2), about 40% of patients still had high (10 3 to o10 2 ) or very high (X10 2 ) MRD levels, while after consolidation therapy (TP3) approximately half of patients became MRD-negative. At later TP MRD could be detected in only a minority of patients and MRD levels generally were o10 4. Patients with MLL-rearrangements showed a significant slower reduction in tumor load as compared with MLL-germline patients (Figure 1c). Prognostic significance of MRD levels at a single TP Out of 46 patients with available data at day 15 (TP1), the vast majority had very high (29) or high (6) MRD levels, resulting in 18 relapses overall. Of the 11 remaining patients, two patients were MRD-negative and remained disease-free, whereas out of the nine patients with low-to-moderate MRD levels, one suffered an off-therapy relapse. The MRD level at TP2 (after induction) significantly discriminated outcome: the higher the MRD levels, the lower the DFS (Po0.0001, Figure 3a). One MLL-germline patient had MRD levels at TP2 and relapsed; all other MLL-germline patients had lower MRD levels and remained in remission. Also at TP3 (after MARAM), the MRD level significantly influenced DFS (Po0.0001, Figure 3b). In fact, all patients with moderate to very high MRD levels (X10 4 ) at TP3 relapsed, indicating that MRD levels X10 4 at this TP allow recognition of extremely poor-prognosis patients. All evaluable MLL-germline patients had negative of very low MRD levels (o10 4 ) at TP3 and all remained in remission. At TP4 (after OCTADD), MRD positivity (14 out of 38 patients) was associated with a high relapse rate (71%), whereas out of the 24 MRD-negative patients 4 relapsed (17%). At later TP, too few samples were available for reliable analysis. Of note, the single patient being MRD positive at the end of therapy (TP7) relapsed few weeks later, whereas the MRD-negative

4 1076 Figure 3 Prognostic significance of minimal residual disease (MRD) levels at TP2 (a) and TP3 (b) as shown by Kaplan Meier estimates of diseasefree survival. Table 1 Patients by MRD levels at TP2 and TP3 (in parenthesis, number of relapses) MRD category at TP2 a MRD category at TP3 a Overall Negative Low (o10 4 ) Moderate (10 4 to o10 3 ) High (10 3 to o10 2 ) Very high (X10 2 ) Negative 13 (2) 1 (0) 0 (0) 0 (0) 0 (0) 14 (2) Low (o10 4 ) 6 (1) 4 (0) 1 (1) 0 (0) 0 (0) 11 (2) Moderate (10 4 to o10 3 ) 6 (2) 4 (3) 1 (1) 0 (0) 1 (1) 12 (7) High (10 3 to o10 2 ) 4 (0) 2 (0) 2 (2) 2 (2) 2 (2) 12 (6) Very high (X10 2 ) 0 (0) 0 (0) 1 (1) 2 (2) 2 (2) 5 (5) Overall 29 (5) 11 (3) 5 (5) 4 (4) 5 (5) 54 (22) Abbreviations: Ig, immunoglobulin; MRD, minimal residual disease; RQ-PCR, real-time quantitative-pcr; TCR, T-cell receptor. a The MRD level at the various follow-up time points was determined by RQ-PCR analysis of MLL and/or Ig/TCR gene rearrangements. If analysis of the same bone marrow sample, with two or three different PCR targets, resulted in different tumor-load estimations, the highest degree of MRD was assumed to be most accurate and used data analysis. patients all remained in complete remission at 3, 4, 4.5 and 5 years after stop of therapy. As the absence of MLL rearrangements is associated with good prognosis (4-year EFS 74.0%), 8 we also evaluated the prognostic significance of MRD after exclusion of 22 and 17 MLL-germline patients at TP2 and TP3, respectively. DFS curves of MLL-positive patients only showed the same pattern as the total group of patients, both at TP2 and TP3 (data not shown). Prognostic significance of MRD levels at TP2 and TP3 As the kinetics of tumor load reduction may provide more accurate prognostic significance than MRD analysis at a single TP, 9 we grouped patients according to the combined MRD results at TP2 and TP3. On the basis of the combinations of MRD levels, the distribution of patients and relapses (Table 1), three risk groups could be distinguished. An evident MRD-high-risk group was defined by an MRD level at TP3X10 4. This MRD-high-risk group consisted of 26% of patients and all patients relapsed. The remaining patients could be separated in an MRD-low-risk group (defined by MRD levels o10 4 at both TP) and an MRD-medium-risk group (all remaining patients). In the relatively large MRD-low-risk group (44% of patients) only 13% relapsed, whereas in the MRD-medium-risk group (30% of patients) 31% relapsed (Figure 4). These MRD-low-risk, MRD-medium-risk and MRD-high-risk groups contained nine, two and none MLL-negative patients, respectively. If only patients with MRD-negativity (instead of MRDo10 4 ) were considered MRD-low-risk (comparable with current childhood ALL definitions), 9,14 the MRD-low-risk group consisted of 24% of patients with 15% relapses, whereas in the MRDmedium-risk group (50% of patients) 22% of patients relapsed. These MRD-low-risk and MRD-medium-risk groups contained six and five MLL-negative patients, respectively. Comparison between MRD-based stratification and Interfant-06 stratification On the basis of the results of the Interfant-99 protocol, infants enrolled in the recently started Interfant-06 protocol are

5 Table 2 Patients by MRD-based risk group stratification and Interfant-06 stratification (in parenthesis, number of relapses) a 1077 MRD Stratification Interfant-06 stratification Overall Low risk Medium risk High risk MRD-low risk 9 (0) 9 (3) 4 (0) 22 (3) MRD-medium risk 2 (0) 10 (4) 3 (1) b 15 (5) b MRD-high risk 0 8 (8) 5 (5) 13 (13) Overall 11 (0) 27 (15) 12 (6) 50 (21) Abbreviations: CCR, complete continuous remission; MRD, minimal residual disease. a Four patients were excluded because their MLL status was not known. b One death in CCR. Figure 4 Prognostic significance of minimal residual disease (MRD) levels at TP2 and TP3 as shown by Kaplan Meier estimates for diseasefree survival. Patients were considered MRD-high-risk if the MRD level at TP3 was X10 4 ; patients were considered MRD-low-risk if MRD levels were o10 4 at both time points; all remaining patients were considered MRD-medium-risk. stratified as low-risk (MLL-germline), high-risk (MLL rearrangement, age o6 months at diagnosis and WBC X /l) or medium-risk (all remaining patients). 8 The relationship between this Interfant-06 stratification at diagnosis and the here presented MRD-based stratification shows that the two stratifications recognize different subgroups of patients (Table 2). It is noteworthy that an MRD level X10 4 at TP3 dissects patients with dismal prognosis in both medium-risk and highrisk group according to Interfant-06 stratification. This factor is significant when analyzed in a Cox model where both stratifications are included. The MRD-high-risk group, as compared with the MRD-medium-risk group, is associated with a 6.4-fold increase of hazard of relapse (95% confidence interval , Po0.001). This finding should be seen with caution due to low numbers that can also explain the lower than expected number of relapses in the 12 patients who, according to Interfant-06 stratification, are at high risk with an expected 20% EFS. 8 Discussion Although some reports described MRD data in infants with ALL, this study is by far the largest and the first multi-center prospective study. Several factors were crucial to perform this study: a large-scale international treatment protocol (Interfant- 99) allowing analysis of similarly treated infants with ALL, largely standardized laboratory protocols and standardized guidelines for MRD real-time quantitative-pcr data interpretation. 8,32,33 These protocols and guidelines were developed over the past years within the I-BFM-SG MRD Task Force and the ESG-MRD-ALL. 32,33 Infant ALL is characterized by high frequencies of MLL gene rearrangements, 1,3 and as such rearrangements most likely are stable during the disease course they potentially are excellent MRD targets. The recent development of new methods to characterize MLL break points at the DNA level allows the routine application of MLL rearrangements as MRD-PCR targets. 17,18,28 30 In fact, MLL rearrangements are preferred over Ig/TCR rearrangements as MRD-PCR targets, as MLL rearrangements are assumed to be present in the total leukemic clone, which is in contrast to the frequently oligoclonal Ig/TCR rearrangements in infant ALL. 3 Consequently, MRD data obtained by analysis of MLL rearrangements may provide a more accurate estimation of the level of MRD. Indeed, our data show that in approximately 10% of samples, MRD data obtained by analysis of Ig/TCR rearrangements result in an underestimation of the actual MRD load as determined by MLL analysis. Furthermore, the quantitative range of MLL rearrangements reaches at least 10 4 in over 70% of cases as compared with approximately 50% of Ig/TCR targets, which most likely is related to the frequently subclonal Ig/TCR rearrangements. We therefore recommend the use of MLL rearrangements as first choice targets for MRD analysis in MLL-positive ALL. Grouping patients according to the MRD level at TP2 (postinduction) resulted in a broad gradient of DFS values. This TP particularly seems to be suitable for recognition of low-risk patients. In contrast, MRD diagnostics at TP3 (post-maram) particularly recognizes high-risk patients with high accuracy: all patients with MRD levels X10 4 relapsed. Our data are in contrast to those reported by Li et al., 4 who did not find any difference in MRD levels post-induction between those patients who did or did not relapse. However, in that study only nine infant ALL patients were monitored. The use of MRD at TP3 may be limited due to the short time to relapse, which may require intensification already early during treatment, that is, after induction. Within the on-going Interfant-06 protocol, two strategies of post-induction treatment intensification are therefore being compared. Although MRD data at individual TP already had prognostic significance, an even more accurate risk group definition could be made based on combined MRD results at TP2 and TP3. Most importantly, such stratification allowed recognition of an evident MRD-high-risk group (26% of patients) with 100% relapses that may profit from alternative or experimental treatment strategies. 5 In addition, an MRD-low-risk group of

6 % of patients with 13% relapses and a remaining MRDmedium-risk group of 30% of patients with 31% relapses could be identified. Despite the large group of MRD-low-risk patients (almost half of patients), only 14% of all relapses occurred in this group and these patients therefore do rather well using the current treatment. Of note, in this group both MLL-germline patients and patients with MLL-rearrangements were present. In the MRD-medium-risk group, 38% of all relapses occurred and these patients consequently might profit from alternative or intensified therapy. If an alternative definition was used for the MRD-low-risk group (MRD negative at TP2 and TP3), comparable data were obtained, although the groups defined by this definition differed with respect to the relative sizes of the MRDmedium-risk and MRD-low-risk groups. Given the small sample sizes of the MRD-medium-risk and MRD-low-risk groups and the modest difference in DFS, future studies are required to confirm that these two groups indeed are different. Whereas MRD negativity at early TP in childhood ALL is associated with a very low relapse rate (o10%), 9 two out of 13 infants who were MRD-negative at both TP2 and TP3 (15%) relapsed after VIMARAM during maintenance therapy (remission duration 9 and 18 months). These two patients did not appear to have special characteristics (one 7 months old MLL- AF10-positive girl and one 2 months old MLL-AF4-positive girl, both with a good prednisone-response and white blood cell count o cells/l at diagnosis) and had no apparent protocol deviations. False-negative MRD results were excluded by confirming the presence of the applied MRD-PCR targets in the relapse specimen and by short tandem repeat genotyping excluding sample change (data not shown). Therefore, although patient numbers in our study are still limited, it might be that analysis of MRD in infant ALL may not allow recognition of a very good risk group. Comparison between the risk group stratification of the Interfant-06 protocol and the MRD-based risk group stratification suggests that the MRD-based classification is more useful to identify the patients with high risk of relapse: 13 out of 21 relapses were comprised in the MRD-high-risk group, whereas the Interfant-06-based high-risk group contained only 6 out of 21 relapses. This finding should be seen with caution due to low numbers that can also explain the lower than expected number of relapses in the 12 patients who, according to Interfant-06 stratification, are at high risk with an expected 20% EFS. 8 Furthermore, this comparison suggests that the 11 MLL-germline patients might not need MRD diagnostics, because none of these patients relapsed. However, in the total Interfant-99 protocol, the MLL-germline patients reached a 4-year EFS of only 74%. 8 Furthermore, the single MLL-germline patient with high MRD levels (410 2 ) at TP2 relapsed; TP3 was not available. Therefore, it might well be that relapses in the MLL-germline patients can be predicted by MRD diagnostics; this is currently under investigation in the Interfant-06 protocol. The MRD results of the Interfant-99 protocol indicate that MRD diagnostics has added value for recognition of risk groups in infant ALL and that MRD diagnostics can also be used for treatment intervention in the rather homogeneous group of infant ALL. Apparently, MRD diagnostics has independent prognostic value, simply because it measures in vivo therapy efficacy, which represents a complex combination of parameters different from the classical clinical and laboratory characteristics at diagnosis. As several adaptations have been introduced in the early treatment phases of the current Interfant- 06 protocol, the value of MRD in Interfant-06 has to be shown by the currently ongoing MRD study in which standardized MRD diagnostics is performed at the national level, according to the guidelines and quality control system of the European Study Group on MRD in ALL. 32 Acknowledgements We gratefully acknowledge Patricia Hoogeveen, Maaike de Bie, Susi Fischer, Eva Csinady, and Jessica Buijs-Gladdines for excellent technical assistance and Marieke Comans-Bitter for preparing the figures. We thank all clinicians of the collaborative group for taking and sending in patient samples. This study was supported by the Dutch Cancer Foundation (KWF Grant EUR ; MWJCJ, VHJvdV), Fondazione Città Della Speranza, Fondazione Cariparo, AIRC, Murst (ex 40%) (GB), Österreichische Kinderkrebshilfe, the Research Program Genome Research for Health of the Austrian Ministry of Education, Science and Culture (GEN-AU Child, GZ /1-VI/1/2005), OENB (ERP-G), Wilhelm-Sander Foundation (MM), the Deutsche Krebshilfe (research grant ; RM), IBISCUS and AIRC (LLN), Fondazione Tettamanti, Fondazione Cariplo, AIRC and MIUR (GC). This study was performed on behalf of the Interfant-99 study group (coordinator: R Pieters), which is composed of AIEOP (G de Rossi, A Biondi; Italy), ANZCHOG (R Suppiah; Australia, New Zealand), Argentina (M Felice), BFM-A (G Mann; Austria), BFM-G (M Schrappe; Germany), COALL (G Janka-Schaub; Germany), CWPGH (J Stary; Czech Republic), DCOG (R Pieters; the Netherlands), DFCI consortium (L Silverman; USA), EORTC-CLCG (A Fester; France, Belgium, Portugal), FRALLE (F Mechinaud; France), Hong Kong (CK Li), NOPHO (L Hovi; Scandinavian countries), PINDA (M Campbell; Chile), PPLLSG (T Szczepañski; Poland), SJCRH (JE Rubnitz; USA), UKCCSG (I Hann, A Vora; United Kingdom). References 1 Biondi A, Cimino G, Pieters R, Pui CH. Biological and therapeutic aspects of infant leukemia. Blood 2000; 96: Armstrong SA, Staunton JE, Silverman LB, Pieters R, den Boer ML, Minden MD et al. MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia. Nat Genet 2002; 30: Jansen MW, Corral L, van der Velden VH, Panzer-Grumayer R, Schrappe M, Schrauder A et al. Immunobiological diversity in infant acute lymphoblastic leukemia is related to the occurrence and type of MLL gene rearrangement. 2007; 21: Li A, Goldwasser MA, Zhou J, Armstrong SA, Wang H, Dalton V et al. Distinctive IGH gene segment usage and minimal residual disease detection in infant acute lymphoblastic leukaemias. Br J Haematol 2005; 131: Stam RW, den Boer ML, Pieters R. Towards targeted therapy for infant acute lymphoblastic leukaemia. Br J Haematol 2006; 132: Mann G, Cazzaniga G, van der Velden VH, Flohr T, Csinady E, Paganin M et al. Acute lymphoblastic leukemia with t(4;11) in children 1 year and older: the big sister of the infant disease? 2007; 21: Schrappe M, Reiter A, Zimmermann M, Harbott J, Ludwig WD, Henze G et al. Long-term results of four consecutive trials in childhood ALL performed by the ALL-BFM study group from 1981 to Berlin-Frankfurt-Munster. 2000; 14: Pieters R, Schrappe M, De Lorenzo P, Hann I, De Rossi G, Felice M et al. A treatment protocol for infants younger than 1 year with acute lymphoblastic leukaemia (Interfant-99): an observational study and a multicentre randomised trial. Lancet 2007; 370: van Dongen JJM, Seriu T, Panzer-Grumayer ER, Biondi A, Pongers-Willemse MJ, Corral L et al. Prognostic value of minimal residual disease in acute lymphoblastic leukaemia in childhood. Lancet 1998; 352:

7 10 Eckert C, Biondi A, Seeger K, Cazzaniga G, Hartmann R, Beyermann B et al. Prognostic value of minimal residual disease in relapsed childhood acute lymphoblastic leukaemia. Lancet 2001; 358: Goulden N, Bader P, Van Der Velden V, Moppett J, Schilham M, Masden HO et al. Minimal residual disease prior to stem cell transplant for childhood acute lymphoblastic leukaemia. Br J Haematol 2003; 122: Bruggemann M, Raff T, Flohr T, Gokbuget N, Nakao M, Droese J et al. Clinical significance of minimal residual disease quantification in adult patients with standard-risk acute lymphoblastic leukemia. Blood 2006; 107: van der Velden VHJ, Hochhaus A, Cazzaniga G, Szczepanski T, Gabert J, van Dongen JJM. Detection of minimal residual disease in hematologic malignancies by real-time quantitative PCR: principles, approaches, and laboratory aspects. 2003; 17: Flohr T, Schrauder A, Cazzaniga G, Panzer-Grumayer R, van der Velden V, Fischer S et al. 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. 2008; 22: Coustan-Smith E, Behm FG, Sanchez J, Boyett JM, Hancock ML, Raimondi SC et al. Immunological detection of minimal residual disease in children with acute lymphoblastic leukaemia. Lancet 1998; 351: Burmeister T, Marschalek R, Schneider B, Meyer C, Gokbuget N, Schwartz S et al. Monitoring minimal residual disease by quantification of genomic chromosomal breakpoint sequences in acute leukemias with MLL aberrations. 2006; 20: Meyer C, Kowarz E, Schneider B, Oehm C, Klingebiel T, Dingermann T et al. Genomic DNA of leukemic patients: target for clinical diagnosis of MLL rearrangements. Biotechnol J 2006; 1: Meyer C, Schneider B, Reichel M, Angermueller S, Strehl S, Schnittger S et al. Diagnostic tool for the identification of MLL rearrangements including unknown partner genes. Proc Natl Acad Sci USA 2005; 102: van Dongen JJM, Langerak AW, Bruggemann M, Evans PAS, Hummel M, Lavender FL et al. Design and standardization of PCR primers and protocols for detection of clonal immunogloulin and T-cell receptor gene recombinations in suspect lymphoproliferations. 2003; 17: Szczepanski T, van der Velden VHJ, Hoogeveen PG, De Bie M, Jacobs DCH, Van Wering ER et al. V{delta}2-J{alpha} gene rearrangements are frequent in precursor-b-acute lymphoblastic leukemia but rare in normal lymphoid cells. Blood 2004; 103: Szczepanski T, Willemse MJ, van Wering ER, van Weerden JF, Kamps WA, van Dongen JJM. Precursor-B-ALL with D(H)-J(H) gene rearrangements have an immature immunogenotype with a high frequency of oligoclonality and hyperdiploidy of chromosome ; 15: Verhagen OJ, Willemse MJ, Breunis WB, Wijkhuijs AJ, Jacobs DC, Joosten SA et al. Application of germline IGH probes in real-time quantitative PCR for the detection of minimal residual disease in acute lymphoblastic leukemia. 2000; 14: Pongers-Willemse MJ, Seriu T, Stolz F, d Aniello E, Gameiro P, Pisa P et al. Primers and protocols for standardized detection of minimal residual disease in acute lymphoblastic leukemia using immunoglobulin and T cell receptor gene rearrangements and TAL1 deletions as PCR targets: report of the BIOMED-1 CONCERTED ACTION: investigation of minimal residual disease in acute leukemia. 1999; 13: Bruggemann M, van der Velden VH, Raff T, Droese J, Ritgen M, Pott C et al. Rearranged T-cell receptor beta genes represent powerful targets for quantification of minimal residual disease in childhood and adult T-cell acute lymphoblastic leukemia. 2004; 18: van der Velden VHJ, Wijkhuijs JM, Jacobs DC, van Wering ER, van Dongen JJM. T cell receptor gamma gene rearrangements as targets for detection of minimal residual disease in acute lymphoblastic leukemia by real-time quantitative PCR analysis. 2002; 16: van der Velden VHJ, Willemse MJ, van der Schoot CE, Hahlen K, van Wering ER, van Dongen JJM. Immunoglobulin kappa deleting element rearrangements in precursor-b acute lymphoblastic leukemia are stable targets for detection of minimal residual disease by real-time quantitative PCR. 2002; 16: van der Velden VH, de Bie M, van Wering ER, van Dongen JJ. Immunoglobulin light chain gene rearrangements in precursor-bacute lymphoblastic leukemia: characteristics and applicability for the detection of minimal residual disease. Haematologica 2006; 91: Langer T, Metzler M, Reinhardt D, Viehmann S, Borkhardt A, Reichel M et al. Analysis of t(9;11) chromosomal breakpoint sequences in childhood acute leukemia: almost identical MLL breakpoints in therapy-related AML after treatment without etoposides. Genes Chromosomes Cancer 2003; 36: Metzler M, Brehm U, Langer T, Viehmann S, Borkhardt A, Stanulla M et al. Asymmetric multiplex-polymerase chain reaction - a high throughput method for detection and sequencing genomic fusion sites in t(4;11). Br J Haematol 2004; 124: Robinson BW, Slater DJ, Felix CA. BglII-based panhandle and reverse panhandle PCR approaches increase capability for cloning der(ii) and der(other) genomic breakpoint junctions of MLL translocations. Genes Chromosomes Cancer 2006; 45: Reichel M, Gillert E, Angermuller S, Hensel JP, Heidel F, Lode M et al. Biased distribution of chromosomal breakpoints involving the MLL gene in infants versus children and adults with t(4;11) ALL. Oncogene 2001; 20: van der Velden VH, Cazzaniga G, Schrauder A, Hancock J, Bader P, Panzer-Grumayer ER et al. Analysis of minimal residual disease by Ig/TCR gene rearrangements: guidelines for interpretation of real-time quantitative PCR data. 2007; 21: van der Velden VH, Panzer-Grumayer ER, Cazzaniga G, Flohr T, Sutton R, Schrauder A et al. Optimization of PCR-based minimal residual disease diagnostics for childhood acute lymphoblastic leukemia in a multi-center setting. 2007; 21: