Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Serial minimal residual disease (MRD) analysis as a predictor of response duration in Philadelphia-positive acute lymphoblastic leukemia (Ph+ALL) during imatinib treatment


Patients with refractory or relapsed Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph+ALL) rarely have prolonged responses to salvage therapy, including imatinib, resulting in a short opportunity for potentially curative stem cell transplantation. To identify minimal residual disease (MRD) parameters predictive of imminent relapse, we quantitated Bcr-Abl expression by real-time PCR in peripheral blood (PB) and bone marrow (BM) of 24 Ph+ALL patients after achieving a complete response and MRD minimum. The ratio of Bcr-Abl and glyceraldehyde-3-phosphate dehydrogenase copies, magnitude of increase and velocity of increase were evaluated regarding subsequent time intervals to relapse, death or censoring. High Bcr-Abl levels 5 × 10−4 in PB (n=23) and 10−4 in BM (n=18) were significantly associated with short time periods to relapse. Bcr-Abl increases >2 logarithmic units (log) in PB, but not in BM preceded short-term relapse. The velocity of Bcr-Abl increases predicted response duration in PB (cutoff: 1.25 log/30 days) and BM (0.6). Bcr-Abl level and velocity of increase in BM as well as magnitude of increase in PB correlated with remaining periods of survival and predicted relapse within 2 months in nine of 10, 10 of 11 and four of four patients, respectively. Thus, these MRD parameters may guide timing and intensity of therapeutic modifications.


In total, 60–80% of patients with Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph+ALL) achieve a complete remission (CR) in response to multiagent induction chemotherapy, with a median remission duration of only 9 months and overall survival of less than 10% at 3 years.1,2,3,4,5,6 Allogeneic stem cell transplantation (alloSCT) is potentially curative, but only patients who are in CR at the time of transplant are likely to experience long-term disease-free survival (DFS).7,8,9,10,11,12 Following relapse, aggressive salvage chemotherapy has limited efficacy and substantial morbidity and may contribute to the high transplant-related mortality commonly observed in these patients. Imatinib (GlivecR) is a tyrosine kinase inhibitor that is selective for the Abl, Kit and PDGFR kinases and exerts substantial but usually brief antileukemic activity in the majority of patients with relapsed or refractory Ph+ALL.13,14 Single-agent imatinib is well tolerated and has been used successfully as salvage therapy prior to alloSCT.15 Accordingly, all patients with a compatible donor who are considered eligible for alloSCT should be transplanted as soon as possible. In practice, with a median time to progression on imatinib of only 2 months overall13,14 and of 5.4 months in patients reaching a complete hematologic response (CHR), a considerable number of patients relapse before actually undergoing alloSCT. We have shown previously that these patients have a dismal outcome,15 whereas the probability of DFS in patients who are transplanted while still in CR is 51% at 12 months. As a consequence, all patients on imatinib who cannot be transplanted in a timely manner, that is, within approximately 6 weeks of initiating salvage therapy, should be considered for additional or alternative therapies to avoid relapse, although this may entail considerable toxicity. The ability to distinguish patients at risk of imminent relapse from those likely to derive continued benefit from imatinib monotherapy would therefore provide a rational basis for treatment decisions involving intensification of therapy prior to planned alloSCT, or for the use of alternative experimental strategies in patients not eligible for SCT.

Quantitative PCR techniques able to detect a single leukemia cell in a background of 104–106 normal cells have revealed that the level of residual leukemia correlates with the probability of relapse in childhood and adult Ph−ALL patients.16,17,18,19,20,21 High MRD levels were also associated with early relapse in pediatric Ph+ALL patients.6,22 Increases in residual leukemia levels, as documented by conversion from negative to positive PCR results, were shown to precede relapse in Ph+ALL patients by 1.5 weeks to 6 months.23,24,25 Bone marrow transplantation (BMT) was shown to be more effective in accomplishing negative PCR results than conventional chemotherapy in Ph+ALL.7,26 Detection of Bcr-Abl transcripts in 36 Ph+ALL patients after BMT was associated with an increased risk of relapse.27 In addition, increasing Bcr-Abl levels after BMT were indicative of imminent relapse.28 Specific criteria for interpreting MRD results in Ph+ALL patients receiving imatinib therapy remain to be established.

We have shown previously in cohorts of Ph+ALL patients that the absolute level of Bcr-Abl transcripts measured in PB and BM samples during the first 2–4 weeks of imatinib treatment correlated significantly with the duration of response to imatinib.29 In individual patients, however, the early reduction of Bcr-Abl transcripts does not necessarily predict the kinetics with which a resistant leukemic clone becomes the predominant cell population during subsequent imatinib therapy. Therefore, it is of practical relevance to determine whether the sensitivity and reproducibility of serial quantitative PCR are sufficient to define and detect clinically relevant MRD increments that are predictive of relapse, and whether such an analysis is clinically feasible. Moreover, it was unclear whether PB and BM were equivalent sources of cell samples for MRD analysis. The present analysis was therefore conducted to identify thresholds for the Bcr-Abl level, magnitude of increase and speed of increase that may guide clinical decisions to initiate additional or alternative treatments.

Materials and methods

Patients and treatment design

We studied 56 patients in total who were enrolled in two successive clinical phase II studies of imatinib (CSTI571 109 and CSTI571 114) designed to determine the safety and efficacy of imatinib in patients with relapsed or refractory Ph+ALL.14,15 A CHR was defined as a reduction of marrow blasts to less than 5% with no blasts in peripheral blood (PB) and hematopoietic recovery with absolute neutrophil counts 1.5 × 109/l and platelet counts 100 × 109/l and no evidence of extramedullary disease. A complete marrow response (CMR) was a reduction of marrow blasts to less than 5% and of peripheral blasts to 0% with no evidence of extramedullary disease, but incomplete hematopoietic recovery. Complete response (CR) means CHR or CMR. Partial response (PR) was defined as reduction of bone marrow (BM) blasts to 6–25%. Relapse was defined as disease recurrence with BM blasts exceeding 5% or reappearance of PB blasts in a patient who had achieved a CHR or CMR. Patients were considered refractory to imatinib, if there was no elimination of peripheral blasts or extramedullary disease and/or a failure to reduce marrow blasts to <25%. Minimal residual disease (MRD) results of the initial phase of therapy with imatinib describing the degree of reduction have been published previously.29 Of the 56 patients, 16 were refractory or with a PR and therefore not of interest for further MRD analysis. Since the present study intended to examine further MRD courses during imatinib therapy by monthly BM and PB samples, 10 of the remaining 40 patients with CHR or CMR were excluded from evaluation due to early transfer to SCT (28–67 days after starting imatinib therapy). Four of the remaining 30 patients suffered short-term relapse before or at 2 months of imatinib therapy (29, 46, 55 and 64 days after start), thus not permitting predictive MRD analysis with a schedule of monthly sample collection. Another two patients were excluded from analysis since there was no regular sample collection or sufficient sensitivity at scheduled monthly time points (first patient: sensitivity too low at day 28; only valid sample at day 51; relapse at day 85; second patient: sensitivity too low at day 28; only valid sample at day 56, relapse at day 70). Therefore, 24 patients were evaluable, of whom 23 provided PB samples and 18 BM samples.

Upon initiation of imatinib, two patients were in CR (MRD+), one in PR, five refractory to chemotherapy, 10 in first and six in second relapse. An alloSCT had been performed in 14 patients (matched related n=10; matched unrelated n=4). The minor (p185Bcr-Abl; e1a2) and major (p210Bcr-Abl) Bcr-Abl transcripts were identified in 18 and six patients, respectively (b2a2: n=4; b3a2: n=2).

Cell samples and real-time PCR for Bcr-Abl and GAPDH

BM aspirates and PB samples were collected in EDTA immediately prior to starting imatinib therapy, 2 weeks, 4 weeks and subsequently monthly thereafter. Bcr-Abl quantification was performed as described previously.29 Briefly, mononuclear BM and PB cells were separated by Ficoll-Hypaque density gradient centrifugation and aliquots of viable cells were cryopreserved in liquid nitrogen. RNA was extracted by Ambion's total RNA extraction kit per supplier's instructions. cDNA was synthesized from 1 to 5 μg RNA according to standard conditions. Plasmid standard titrations with defined copy numbers for Bcr-Abl and GAPDH (housekeeping gene) were analyzed simultaneously with patient samples. TaqMan PCR was conducted in duplicate reactions employing ABI PRISM 7700 (PE Biosystems, Weiterstadt, Germany) with standard conditions (50°C for 2 min, 95°C for 10 min and 45 cycles at 95°C for 15 s and 60°C for 1 min). In order to amplify m-Bcr-Abl (e1a2), 5 μl of template was used in 50 μl reaction mixtures consisting of the primer a2-F (IndexTermCAGACCCTGAGGCTCAAAGTC) at 200 nM and the primer rALL-TB (IndexTermGCAAGACCGGGCAGATCT) at 200 nM and the break point-specific probe ALL12-FAM (IndexTermCCGCTGAAGGGCTTCTGCGTCTCC) labelled with FAM at the 5′end and TAMRA at the 3′ end at 200 nM final concentration. MgCl2 was used at 5 mM and other reagents were added as per the supplier's instructions (Core Reagents Kit, PE Biosystems). In order to amplify M-Bcr-Abl (b2a2) primers, a2-F and b2-1R (IndexTermGCATTCCGCTGACCATCAA) were used in combination with break point-specific probe CML22-FAM (IndexTermCCGCTGAAGGGCTTCTTCCTTATTG) and 6 mM MgCl2 concentration. There is a one-mismatch polymorphism in 30% of the patients expressing b2a2.30 Our probe does not map to this polymorphic sequence exactly. Although this mismatch to our probe may impair PCR efficiency in polymorphic patients compared to patients with the predominant allele, this mismatch does not impair intraindividual sequential analysis and comparison as carried out in this study. M-Bcr-Abl (b3a2) was amplified by primers a2b3-F (IndexTermGAGTTCCAACGAGCGGCTT) and b3-1R (IndexTermTCATCGTCCACTCAGCCACT) and break point-specific probe CML32-FAM (IndexTermCCGCTGAAGGGCTTTTGAACTCTG) at 4.5 mM MgCl2. For normalization, GAPDH housekeeping gene expression was analyzed using a predeveloped assay by PE Biosystems. Copy numbers of Bcr-Abl transcripts were calculated by plasmid standard curves, normalized by GAPDH housekeeping gene transcripts and expressed as Bcr-Abl/GAPDH ratios (relative Bcr-Abl levels). It was required to have at least 105 GAPDH copies in a sample to consider a negative PCR result valid. The sensitivity of the method permitted detection of one Bcr-Abl positive leukemic blast of a patient sample in 105 Bcr-Abl negative background cells (BM or PB mononuclear cells of normal donors) as titrated in six patient samples of PB and BM, respectively.

Algorithms for serial MRD evaluation

The present study examined consecutive MRD levels in PB and BM samples collected during imatinib therapy from 24 Ph+ALL patients subsequent to their achieving a relative minimum of Bcr-Abl transcripts. This minimum was defined as the lowest measured relative Bcr-Abl level prior to an increase or identical value. The first Bcr-Abl level that was at least 1 log higher than the minimum was used for statistical analysis of an association with the remaining time of remission or survival. If there was no increase of more than 1 log, the highest value after the minimum was used for analysis. Increased Bcr-Abl levels were only considered if the sample was obtained at least 7 days prior to hematologic relapse. The magnitude of increase was determined for each patient by calculating the log ratio between the increased Bcr-Abl level (as defined above) and the minimum. The velocity of MRD increases was defined as the maximum Bcr-Abl increase per time, expressed as log increase per month (30 days). The starting point of this parameter may be at, or anytime after, the minimum and prior to the increased (ie >1 log) Bcr-Abl level.

Statistical analysis

The Bcr-Abl level, increase and velocity of increase as defined above were examined in relation to the time interval between collection of the index sample (PB or BM) and relapse/censoring or death/censoring by Kaplan–Meier plots and log rank tests using Graph Pad Prism software (San Diego, CA, USA). Logarithmic changes of Bcr-Abl levels in relation to negative PCR results were calculated by substituting the value of the particular sensitivity of the PCR reaction for the negative result. Fisher's exact probability test was employed to detect whether Bcr-Abl level, increase or velocity of increase below the respective threshold were significantly associated with a longer time interval to relapse (>60 days) vs values above the respective threshold with a shorter time interval to relapse. Patients who did not relapse within the time period of 60 days, but were observed for less than 60 days, had to be excluded from the analysis.


Response duration

The median time to progression in the cohort of 24 Ph+ALL patients who achieved a CHR or CMR was 5.5 months (range: 2–25). Five of the 24 patients were in continuous CR at the time of most recent evaluation (740+, 515+, 245+, 215+ and 70+ days after starting imatinib). One of these five patients died of septicemia after 70 days. Of the 24 patients, 19 relapsed (one in central nervous system) a median time of 144 days after starting imatinib (range: 53–448 days). Of 24 patients, 11 were alive at the time of most recent evaluation with a median interval from starting imatinib to death or censoring of 244 days (75–746+).

Bcr-Abl levels

Relative Bcr-Abl levels in relation to housekeeping gene GAPDH were determined in serial PB (23 pts.) and BM (18 pts.) samples from the 24 Ph+ALL patients in CR subsequent to achieving a MRD minimum with imatinib therapy. The median of the minimal Bcr-Abl/GAPDH levels in PB was 3.4 × 10−7 (Figure 1a). At the time when increases of Bcr-Abl were evaluated, the median Bcr-Abl/GAPDH level was 2.9 × 10−5, and the median of individual increases was 1.4 log compared to the minimum. The median interval from this time point to relapse was 64 days (range 9–581+).

Figure 1

Bcr-Abl/GAPDH levels in PB (a) and BM (b) at MRD minimum (best) and at the time of evaluation of increase (incr.). Paired samples are connected by lines and negative PCR results are depicted by empty diamonds. At best MRD response, seven samples were negative in PB and two samples in BM, while at the time of evaluation of increase one sample was negative in PB and BM, respectively. The median of individual changes of Bcr-Abl/GAPDH levels was +1.4 log in PB and +1.1 log in BM.

The median minimal Bcr-Abl/GAPDH level in BM was 3.4 × 10−6 at the minimum and 1.4 × 10−4 at the time when increases were assessed (Figure 1b). The median of individual increases was 1.1 log at this time. The subsequent median time to relapse was 38.5 days (7–499+).

Relationship between Bcr-Abl level and subsequent interval to relapse

In PB, Bcr-Abl/GAPDH levels below 5 × 10−4 were associated with significantly (P=0.035) longer subsequent time periods to relapse (median: 83 days; range: 24–581+) than levels above 5 × 10−4, with a median time to relapse of only 27 days (range: 9–150). In BM, a slightly lower threshold than in PB distinguished two prognostic groups: patients with a Bcr-Abl/GAPDH level below 1 × 10−4 relapsed after a median of 81 days (range: 33–499+), compared with 26 days (range: 7–150) in patients with a level above 1 × 10−4 (P=0.005) (Figure 2a). Interestingly, the threshold of 10−4 in BM was also significantly prognostic for the remaining time of survival (P=0.017), since it dinstinguished a group (n=8) with no death during a median follow-up of 242 days from a poor prognostic group (n=10), in which all patients died with a median interval to death of 126 days.

Figure 2

Kaplan–Meier plots of Bcr-Abl/GAPDH levels in BM (a), Bcr-Abl/GAPDH log increases in PB (b) and velocity of Bcr-Abl/GAPDH increases in BM (c) in regard to time interval from sample collection to relapse are shown. Most discriminatory thresholds for Bcr-Abl levels (10−4), increases (2 log) and speed of increase (0.6 log/month) were used to determine the prognostic potential of these parameters. In addition, a Kaplan–Meier plot of velocity of increases in BM in relation to the time interval from sample collection to death or censoring is depicted (d).

We were interested whether these thresholds could be used to indicate the probability of relapse within 60 days of the most recent MRD analysis, as this time period is usually long enough for implementation of other or additional therapeutic interventions. A threshold Bcr-Abl/GAPDH level in BM (n=17) of 10−4 was indeed highly predictive of relapse within the subsequent 60 days, with a 90% probability (9/10 patients). Conversely, five of seven patients with levels <10−4 did not experience relapse within 60 days of the last MRD analysis (P=0.018) (Table 1). Apparently, levels <10−4 are less likely to predict sustained remission than levels 10−4 are able to identify subsequent relapse.

Table 1 Bcr-Abl/GAPDH levels after best MRD response in BM of Ph+ALL patients allocated prognostically into groups with low (<10−4) and high (10−4) MRD levels in relation to interval from sampling to relapse (<60 or 60 days)

Notably, Bcr-Abl levels in PB were not associated with significantly different rates of relapse within 60 days with regard to a threshold of 5 × 10−4.

Relationship between Bcr-Abl increase and subsequent interval to relapse

The relation between the Bcr-Abl/GAPDH log increase above the minimum and the individual patient's subsequent time to relapse was examined by the Kaplan–Meier plots and log rank tests. An increase of 2 log in PB was the threshold allowing the most significant discrimination (P=0.0004) between patients belonging to different prognostic groups. Bcr-Abl increases of less than 2 log were associated with a median subsequent time to relapse of 83 days, while greater log increases correlated with a median of 22 days (Figure 2b). The differences in log increases of Bcr-Abl were independent of time periods that had elapsed since the lowest MRD level. The median MRD levels determined at the time of best molecular response did not differ between the patients with an early or later relapse.

These prognostic patient groups were also significantly different (P=0.0007) with regard to remaining survival periods, with median intervals of 312 and 47 days, respectively (increase <2 log and 2 log).

Analysis of Bcr-Abl/GAPDH log increases in PB with regard to occurence of relapse within 60 days was performed in 22 patients. All four patients in whom an increase of 2 log in PB was observed relapsed within 60 days. Conversely, 13 of the 18 patients with no or maximum increases of less than 2 log remained in remission for at least 60 days (Table 2).

Table 2 Bcr-Abl/GAPDH logarithmic increases in relation to best response in PB of Ph+ALL patients allocated prospectively into groups with small or missing (<2 log) and strong (2 log) MRD increases in regard to interval from sampling to relapse (<60 or 60 days)

In contrast, Bcr-Abl/GAPDH increases in BM were not significantly linked to prognostic patient groups with respect to time to relapse and survival.

Velocity of Bcr-Abl increase in relation to interval to relapse

Since Bcr-Abl levels as well as increments of Bcr-Abl levels were prognostic indicators of the remaining duration of response and survival, we addressed the question whether the speed of increase was also a useful parameter for estimating the time interval remaining prior to disease progression. We defined the velocity of increase as the maximum Bcr-Abl/GAPDH increment per month following the minimum. In contrast, the previously described parameter ‘Bcr-Abl increase’ was not defined in relation to the time period.

In PB, the velocity of increase discriminated significantly (P=0.004) between patients at high and low risk of imminent relapse, with a threshold of 1.25 log/month. While a velocity <1.25 was associated with a median interval to relapse of 157 days, a value 1.25 was associated with a median of only 39 days. However, this parameter did not predict relapse within 60 days in PB samples significantly (Fisher's exact test) despite significant discrimination in Kaplan–Meier plots.

In BM, the threshold providing the greatest significant discriminatory power was 0.6 (P=0.003) (Figure 2c), lower than in PB. The median time between assessment of the velocity of Bcr-Abl/GAPDH increases to relapse was 157 days (slope <0.6) vs 33 days (slope 0.6). Moreover, this cutoff significantly (P=0.006) discriminated a group (n=7) with no death within a median follow-up time of 272 days from a group (n=11) with a median interval to death of 120 days (Figure 2d).

The cutoff 0.6 log/month for speed of increase in BM samples yielded a significant (P=0.003) discrimination between patients relapsing within 60 days (10 of 11 patients) and those without relapse within 60 days (six of seven patients) (Table 3).

Table 3 Bcr-Abl/GAPDH velocity of increase (log increase per 30 days) after best response in BM of Ph+ALL patients allocated prognostically into low (<0.6) and high (0.6) MRD risk groups in relation to interval from sampling to relapse (<60 or 60 days)

Although Kaplan–Meier curves revealed different predictive cutoffs for the velocity of increase in PB and BM, there was no statistically significant difference in PB and BM samples paired per patient. Therefore, the kinetics of Bcr-Abl increases prior to relapse do not appear to be different in these two compartments.


Most clinical trials examining the therapeutic role of imatinib in Ph+ALL to date have focused on patients with advanced disease. Initial antileukemic activity in this setting is pronounced but resistance occurs rapidly; thus, a major application of imatinib is as temporally limited salvage therapy preceding alloSCT, particularly in view of its favorable safety profile. The median response duration in patients treated with imatinib therapy is only 2.2 months, with only 18% of the patients experiencing more prolonged responses exceeding 6 months. As the results of alloSCT are reasonably good only in patients who are in CR at the time of transplantation,15 imatinib-treated patients who are likely to relapse before undergoing alloSCT will require additional chemotherapy despite the additional treatment-related toxicity. Conversely, imatinib-treated patients with a high probability of sustaining a CR until alloSCT can be performed could conceivably be spared the morbidity and mortality associated with additional salvage chemotherapy.

We investigated whether serial quantitative MRD analysis, performed throughout imatinib treatment in patients having achieved at least a CMR, could be used to predict the probability of relapse within a clinically relevant time period and to guide therapy in a pre-emptive manner. High Bcr-Abl/GAPDH levels and rapid increases in BM were significantly associated with shorter subsequent response duration and survival periods. Pronounced Bcr-Abl/GAPDH increases in PB were significantly linked to short remaining intervals to relapse and death. These parameters predicted actual relapse within 60 days in nine of 10, 10 of 11 and four of four patients, respectively, but were less accurate in excluding relapse within this period, when values were smaller than the cutoff. Our data show that both Bcr-Abl levels and dynamic parameters, that is, magnitude and the speed of increase are prognostic indicators that significantly discriminate different response groups regarding remaining remission and survival periods.

These results considerably extend our previous studies in which we attempted to predict the likelihood of early relapse in 56 Ph+ALL patients by quantitative real-time PCR analysis.29 This study was limited to the phase of MRD decrease during initial imatinib treatment. The MRD level at a certain time point is able to identify short-term responders. However, initial MRD assessment cannot predict when relapse is actually imminent in prolonged responders. Therefore, the present approach was chosen to test whether serial frequent MRD analysis is able to identify individual patients with molecular progression and significant risk of hematologic relapse.

One report described that detection of Bcr-Abl transcripts at any time after SCT was associated with a very high risk of early relapse.27 Other investigators used the conversion from negative to positive RT-PCR as a marker of disease progression and pending relapse in Ph+ALL.24,25 In one small study of five patients employing quantitative PCR, increases of >2 log in PB of two patients was closely followed by relapse while patients without such pronounced increase remained in CR for prolonged periods.28 However, absolute quantitative parameters reflecting MRD levels and dynamics have not yet been established statistically to predict probability of relapse within a certain time period.

The fluctuation of Bcr-Abl levels may not only be attributed to real biological changes in the patient due to a variable proportion of residual leukemic blasts but also to technical features of the method. Although a relatively low interassay variability has been found under specific experimental conditions,28,31 the interassay variance may be substantially higher in the setting of a multicenter trial, in which sample shipment may impair RNA quality. Furthermore, the thresholds that were identified in our study need to be validated in future prospective trials. All these issues need to be considered whenever treatment decisions are to be based on MRD.

Our data demonstrate that the kinetics of Bcr-Abl changes (velocity of increase) are important. Although the optimal threshold for increase per time in BM and PB differed (0.6 vs 1.25), there was no significant difference in paired PB and BM samples. This is of interest since the kinetics of Bcr-Abl decrease during the initial treatment phase (4 weeks) with imatinib differed significantly with PB levels declining much faster than BM levels.29 The steepness of Bcr-Abl increases reflecting the fast net proliferation of leukemic blasts may be a sensitive indicator of emerging resistant clones and pending relapse. Secondary resistance may evolve rapidly during imatinib treatment32 and appear to be caused predominantly by selection of resistant clones harboring a point mutation in the Abl domain.33,34,35,36,37,38,39 However, other mechanisms of Bcr-Abl resistance and Bcr-Abl-independent mechanisms of resistance have been found as well.40 Frequent MRD analysis is therefore important to detect the emergence of resistant proliferating clones early to facilitate clinical interventions prior to overt relapse. Since assessment of velocity of Bcr-Abl increase may miss protracted slow increases over long periods, the other parameters, for example, Bcr-Abl level and magnitude of increase, are also useful.

Although BM samples were shown to be the more discriminatory and reliable material regarding Bcr-Abl level and speed of increase compared to PB, the latter material was more predictive regarding Bcr-Abl increase. Moreover, PB samples have the advantage of easy access. It is not clear why increases in PB are significantly associated with subsequent time spans to relapse whereas increases in BM are not. It may be speculated that more frequent sampling of PB may increase the likelihood to detect significant increases compared to BM samples. In conclusion, BM samples are to be considered the golden standard in MRD diagnostics of Ph+ALL, but PB samples may be of additional value by supplementing in between time points of BM aspiration.

In summary, the present study demonstrated the predictive function of Bcr-Abl level, increase and velocity of increase in BM and PB after reaching best MRD response to imatinib in Ph+ALL patients.


  1. 1

    Maurer J, Janssen JW, Thiel E, van Denderen J, Ludwig WD, Aydemir U et al. Detection of chimeric BCR-ABL genes in acute lymphoblastic leukaemia by the polymerase chain reaction. Lancet 1991; 337: 1055–1058.

    CAS  Article  Google Scholar 

  2. 2

    Westbrook CA, Hooberman AL, Spino C, Dodge RK, Larson RA, Davey F et al. Clinical significance of the BCR-ABL fusion gene in adult acute lymphoblastic leukemia: a Cancer and Leukemia Group B Study (8762). Blood 1992; 80: 2983–2990.

    CAS  Google Scholar 

  3. 3

    Hoelzer D . Acute lymphocytic leukemia in adults. In: Hoffman R, Benz EJ, Shattil SJ, Furie B, Cohen HJ, Silberstein LE, McGlave P (eds). Hematology, Basic Principles and Practice. 3rd edn. Philadelphia, PA: Churchill Livingstone, 2000. pp 1089–1105.

    Google Scholar 

  4. 4

    Faderl S, Kantarjian HM, Thomas DA, Cortes J, Giles F, Pierce S et al. Outcome of Philadelphia chromosome-positive adult acute lymphoblastic leukemia. Leuk Lymphoma 2000; 36: 263–273.

    CAS  Article  Google Scholar 

  5. 5

    Radich JP . Philadelphia chromosome-positive acute lymphocytic leukemia. Hematol Oncol Clin North Am 2001; 15: 21–36.

    CAS  Article  Google Scholar 

  6. 6

    Gleissner B, Gokbuget N, Bartram CR, Janssen B, Rieder H, Janssen JW et al. Leading prognostic relevance of the BCR-ABL translocation in adult acute B-lineage lymphoblastic leukemia: a prospective study of the German Multicenter Trial Group and confirmed polymerase chain reaction analysis. Blood 2002; 99: 1536–1543.

    CAS  Article  Google Scholar 

  7. 7

    Stockschläder M, Hegewisch-Becker S, Kruger W, Tom Dieck A, Mross K, Hoffknecht M et al. Bone marrow transplantation for Philadelphia-chromosome-positive acute lymphoblastic leukemia. Bone Marrow Transplant 1995; 16: 663–667.

    PubMed  Google Scholar 

  8. 8

    Dunlop LC, Powles R, Singhal S, Treleaven JG, Swansbury GJ, Meller S et al. Bone marrow transplantation for Philadelphia chromosome-positive acute lymphoblastic leukemia. Bone Marrow Transplant 1996; 17: 365–369.

    CAS  PubMed  Google Scholar 

  9. 9

    Sierra J, Storer B, Hansen JA, Bjerke JW, Martin PJ, Petersdorf EW et al. Transplantation of marrow cells from unrelated donors for treatment of high-risk acute leukemia: the effect of leukemic burden, donor HLA-matching, and marrow cell dose. Blood 1997; 89: 4226–4235.

    CAS  Google Scholar 

  10. 10

    Snyder DS, Nademanee AP, O'Donnell MR, Parker PM, Stein AS, Margolin K et al. Long-term follow-up of 23 patients with Philadelphia chromosome-positive acute lymphoblastic leukemia treated with allogeneic bone marrow transplant in first complete remission. Leukemia 1999; 13: 2053–2058.

    CAS  Article  Google Scholar 

  11. 11

    Martin TG, Gajewski JL . Allogeneic stem cell transplantation for acute lymphocytic leukemia in adults. Hematol Oncol Clin North Am 2001; 15: 97–120.

    CAS  Article  Google Scholar 

  12. 12

    Cornelissen JJ, Carston M, Kollman C, King R, Dekker AW, Lowenberg B et al. Unrelated marrow transplantation for adult patients with poor-risk acute lymphoblastic leukemia: strong graft-versus-leukemia effect and risk factors determining outcome. Blood 2001; 97: 1572–1577.

    CAS  Article  Google Scholar 

  13. 13

    Druker BJ, Sawyers CL, Kantarjian H, Resta DJ, Reese SF, Ford JM et al. Activity of a specific inhibitor of the BCR-ABL tyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblastic leukemia with the Philadelphia chromosome. N Engl J Med 2001; 344: 1038–1042.

    CAS  Article  Google Scholar 

  14. 14

    Ottmann OG, Druker BJ, Sawyers CL, Goldman JM, Reiffers J, Silver RT et al. A phase 2 study of imatinib in patients with relapsed or refractory Philadelphia chromosome-positive acute lymphoid leukemias. Blood 2002; 100: 1965–1971.

    CAS  Article  Google Scholar 

  15. 15

    Wassmann B, Pfeifer H, Scheuring U, Klein SA, Gokbuget N, Binckebanck A et al. Therapy with imatinib mesylate (Glivec) preceding allogeneic stem cell transplantation (SCT) in relapsed or refractory Philadelphia-positive acute lymphoblastic leukemia (Ph+ALL). Leukemia 2002; 16: 2358–2365.

    CAS  Article  Google Scholar 

  16. 16

    Brisco MJ, Condon J, Hughes E, Neoh SH, Sykes PJ, Seshadri R et al. Outcome prediction in childhood acute lymphoblastic leukaemia by molecular quantification of residual disease at the end of induction [see comments]. Lancet 1994; 343: 196–200.

    CAS  Article  Google Scholar 

  17. 17

    Cave H, van der Werff ten Bosch J, Suciu S, Guidal C, Waterkeyn C, Otten J 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. N Engl J Med 1998; 339: 591–598.

    CAS  Article  Google Scholar 

  18. 18

    van Dongen JJ, 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: 1731–1738.

    CAS  Article  Google Scholar 

  19. 19

    Brisco J, Hughes E, Neoh SH, Sykes PJ, Bradstock K, Enno A et al. Relationship between minimal residual disease and outcome in adult acute lymphoblastic leukemia. Blood 1996; 87: 5251–5256.

    CAS  PubMed  Google Scholar 

  20. 20

    Foroni L, Coyle LA, Papaioannou M, Yaxley JC, Sinclair MF, Chim JS et al. Molecular detection of minimal residual disease in adult and childhood acute lymphoblastic leukaemia reveals differences in treatment response. Leukemia 1997; 11: 1732–1741.

    CAS  Article  Google Scholar 

  21. 21

    Bruggemann M, Droese J, Bolz I, Luth P, Pott C, von Neuhoff N et al. Improved assessment of minimal residual disease in B cell malignancies using fluorogenic consensus probes for real-time quantitative PCR. Leukemia 2000; 14: 1419–1425.

    CAS  Article  Google Scholar 

  22. 22

    Brisco MJ, Sykes PJ, Dolman G, Neoh SH, Hughes E, Peng LM et al. Effect of the Philadelphia chromosome on minimal residual disease in acute lymphoblastic leukemia. Leukemia 1997; 11: 1497–1500.

    CAS  Article  Google Scholar 

  23. 23

    Miyamura K, Tanimoto M, Morishima Y, Horibe K, Yamamoto K, Akatsuka M et al. Detection of Philadelphia chromosome-positive acute lymphoblastic leukemia by polymerase chain reaction: possible eradication of minimal residual disease by marrow transplantation. Blood 1992; 79: 1366–1370.

    CAS  PubMed  Google Scholar 

  24. 24

    Mitterbauer G, Fodinger M, Scherrer R, Knobl P, Jager U, Laczika K et al. PCR-monitoring of minimal residual leukaemia after conventional chemotherapy and bone marrow transplantation in BCR-ABL-positive acute lymphoblastic leukaemia. Br J Haematol 1995; 89: 937–941.

    CAS  Article  Google Scholar 

  25. 25

    Preudhomme C, Henic N, Cazin B, Lai JL, Bertheas MF, Vanrumbeke M et al. Good correlation between RT-PCR analysis and relapse in Philadelphia (Ph1)-positive acute lymphoblastic leukemia (ALL). Leukemia 1997; 11: 294–298.

    CAS  Article  Google Scholar 

  26. 26

    Saffroy R, Lemoine A, Brezillon P, Frenoy N, Delmas B, Goldschmidt E et al. Real-time quantitation of bcr-abl transcripts in haematological malignancies. Eur J Haematol 2000; 65: 258–266.

    CAS  Article  Google Scholar 

  27. 27

    Radich J, Gehly G, Lee A, Avery R, Bryant E, Edmands S et al. Detection of bcr-abl transcripts in Philadelphia chromosome-positive acute lymphoblastic leukemia after marrow transplantation. Blood 1997; 89: 2602–2609.

    CAS  PubMed  Google Scholar 

  28. 28

    Mitterbauer G, Nemeth P, Wacha S, Cross NC, Schwarzinger I, Jaeger U et al. Quantification of minimal residual disease in patients with BCR-ABL-positive acute lymphoblastic leukaemia using quantitative competitive polymerase chain reaction. Br J Haematol 1999; 106: 634–643.

    CAS  Article  Google Scholar 

  29. 29

    Scheuring UJ, Pfeifer H, Wassmann B, Bruck P, Atta J, Petershofen EK et al. Early minimal residual disease (MRD) analysis during treatment of Philadelphia chromosome/Bcr-Abl-positive acute lymphoblastic leukemia with the Abl-tyrosine kinase inhibitor imatinib (STI571). Blood 2003; 101: 85–90.

    CAS  Article  Google Scholar 

  30. 30

    Saussele S, Weisser A, Muller MC, Emgi M, La Rosee P, Paschka P et al. Frequent polymorphism in BCR exon b2 identified in BCR-ABL positive and negative individuals using fluorescent hybridization probes. Leukemia 2000; 14: 2006–2010.

    CAS  Article  Google Scholar 

  31. 31

    Eder M, Battmer K, Kafert S, Stucki A, Ganser A, Hertenstein B . Monitoring of BCR-ABL expression using real-time RT-PCR in CML after bone marrow or peripheral blood stem cell transplantation. Leukemia 1999; 13: 1383–1389.

    CAS  Article  Google Scholar 

  32. 32

    Hofmann WK, de Vos S, Elashoff D, Gschaidmeier H, Hoelzer D, Koeffler HP et al. Relation between resistance of Philadelphia-chromosome-positive acute lymphoblastic leukaemia to the tyrosine kinase inhibitor STI571 and gene-expression profiles: a gene-expression study. Lancet 2002; 359: 481–486.

    CAS  Article  Google Scholar 

  33. 33

    Branford S, Rudzki Z, Walsh S, Grigg A, Arthur C, Taylor K et al. High frequency of point mutations clustered within the adenosine triphosphate-binding region of BCR/ABL in patients with chronic myeloid leukemia or Ph-positive acute lymphoblastic leukemia who develop imatinib (STI571) resistance. Blood 2002; 99: 3472–3475.

    CAS  Article  Google Scholar 

  34. 34

    von Bubnoff N, Schneller F, Peschel C, Duyster J . BCR-ABL gene mutations in relation to clinical resistance of Philadelphia-chromosome-positive leukaemia to STI571: a prospective study. Lancet 2002; 359: 487–491.

    CAS  Article  Google Scholar 

  35. 35

    Keeshan K, Mills KI, Cotter TG, McKenna SL . Elevated Bcr-Abl expression levels are sufficient for a haematopoietic cell line to acquire a drug-resistant phenotype. Leukemia 2001; 15: 1823–1833.

    CAS  Article  Google Scholar 

  36. 36

    Hofmann WK, Jones LC, Lemp NA, de Vos S, Gschaidmeier H, Hoelzer D et al. Ph(+) acute lymphoblastic leukemia resistant to the tyrosine kinase inhibitor STI571 has a unique BCR-ABL gene mutation. Blood 2002; 99: 1860–1862.

    Article  Google Scholar 

  37. 37

    Gorre ME, Mohammed M, Ellwood K, Hsu N, Paquette R, Rao PN et al. Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science 2001; 293: 876–880.

    CAS  Article  Google Scholar 

  38. 38

    Hochhaus A, Kreil S, Corbin A, La Rosee P, Lahaye T, Berger U et al. Roots of clinical resistance to STI-571 cancer therapy. Science 2001; 293: 2163a.

    Article  Google Scholar 

  39. 39

    Barthe C, Cony-Makhoul P, Melo JV, Mahon JR . Roots of clinical resistance to STI-571 cancer therapy. Science 2001; 293: 2163a.

    Article  Google Scholar 

  40. 40

    Hochhaus A, Kreil S, Corbin AS, La Rosee P, Muller MC, Lahaye T et al. Molecular and chromosomal mechanisms of resistance to imatinib (STI571) therapy. Leukemia 2002; 16: 2190–2196.

    CAS  Article  Google Scholar 

Download references


The present study was supported by grants from the BMBF Competence Network ‘Acute Leukemias’ Grant No. 01G19971, the German Genome Research Network (NGFN) and the Adolf Messer Stiftung, Germany.

We are indebted to S Kriener, MD for the pathological review of marrow histologies, to Anja Binckebank for coordinating the study and to Anja Goodwin, Rabia El Kalaäoui, Heike Nürnberger, Martine Pape, Holger Thüringer and Sandra Wagner for their excellent technical assistance.

Author information



Corresponding author

Correspondence to U J Scheuring.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Scheuring, U., Pfeifer, H., Wassmann, B. et al. Serial minimal residual disease (MRD) analysis as a predictor of response duration in Philadelphia-positive acute lymphoblastic leukemia (Ph+ALL) during imatinib treatment. Leukemia 17, 1700–1706 (2003).

Download citation


  • acute lymphoblastic leukemia
  • Bcr-Abl
  • Philadelphia chromosome
  • tyrosine kinase
  • real-time PCR
  • minimal residual disease
  • imatinib (Glivec)

Further reading


Quick links