Resources and information for parents of children with cancer . . . by parents of children with cancer.

this page last modified

Cave article

"Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia." (19)

This paper comes out of the Laboratoire de Biochimie Genetique, Hospital Robert Debre, Paris, France. Helene Cave and colleagues report the results of EORTC (European Organization for Research and Treatment of Cancer) protocol 58881 which is a Berlin-Frankfurt-Munster (BFM) protocol, very similar to POG protocols as used in North America. A total of 246 children diagnosed with ALL were enrolled in the MRD study between 1989 and 1996; results were compared with 654 children treated during the same time period at CLCG (Childhood Leukemia Cooperative Group) centers that did not participate in the MRD arm of the study although the treatment was identical.

Bone marrow samples from newly diagnosed children were analyzed first for the presence of rearrangements in the DNA which codes for TCR-gamma and TCR-delta proteins by using appropriate PCR primers. If a discrete PCR DNA product was not found, arrangements in the IgH coding region were sought in a similar manner. The PCR DNA products found by either (or both) method(s) was extracted from the electrophoresis gel and sequenced. Once the sequence of a child’s leukemic clone DNA in this particular region was known, small DNA pieces, or oligomers, with this sequence were machine-synthesized. These oligomers are called "patient-specific DNA probes".

During treatment, at specific treatment times, samples of each child’s bone marrow were analyzed for MRD. For each test they used 2x105 mononuclear bone marrow cells from which they purified the DNA. First they used the appropriate primers and PCR to prepare PCR product, then they analyzed this product by hybridizing it to the patient-specific DNA probes. This method insures that the PCR product indeed comes from a cell exactly like the child’s leukemia cells (or, clone) at diagnosis, to rule out false-positive results. It also detects very low levels of PCR product, levels which could not be seen on a gel. The competitive PCR protocol included the rigors of serial dilutions to correctly correlate the amount of product with the amount of leukemic DNA. They report that their median level of detection was 5 blasts per 100,000 mononuclear cells.

Of the 246 children entered into the study, only 178 could be monitored for MRD. Various reasons account for this: 4 did not go into remission, 16 did not send the appropriate bone marrow samples, 25 had no rearrangements in either the IgH or TCR coding regions, and 23 could not have patient-specific probe made because of technical reasons. Only 75% could be monitored.

The table summarizes the correlations between MRD (absence, presence, and degree) and relapses. An MRD value of "absent or present" is indicated by a positive signal on PCR; the degree or MRD is obtained from sensitivity studies. The numbers of total patients at the different time points differ because of protocol differences, e.g., some did not have delayed intensification.

residual disease test results # of patients/total # # with relapses relative risk
after induction therapy blank blank blank
absent 88/151 7 1.0
present 63/151 25 5.7
residual blasts <10-3 109/133 10 1.0
residual blasts 10-3 to <10-2 9/133 2 2.3
residual blasts >10-2 15/133 11 16.0
after consolidation therapy blank blank blank
absent 95/127 8 1.0
present 32/127 15 7.3
residual blasts <10-3 110/118 11 1.0
residual blasts >10-3 8/118 6 15.3
after interval therapy blank blank blank
absent 108/130 11 1.0
present 22/130 11 7.3
residual blasts <10-3 118/127 13 1.0
residual blasts >10-3 9/127 7 18.5
after delayed intensification therapy blank blank blank
absent 107/123 14 1.0
present 16/123 10 9.2
residual blasts <10-3 114/119 17 1.0
residual blasts >10-3 5/119 5 22.0


According to the authors, the risk of relapse was markedly increased in patients with over 1 in 100 residual leukemic cells at the end of induction therapy. After consolidation, interval, and delayed intensifications, the risk of relapse increased at the cut off of over 1 in 1000 residual leukemia cells.

The authors did a careful statistical analysis of all the data, comparing the MRD study arm (178 patients) with the control group (over 600 patients). According to the authors, stratified and multivariate analyses showed that the presence or absence of MRD is a significant prognostic factor even when traditional factors such as white blood cell count at diagnosis and age were taken into consideration.

Note that at each time point some patients relapsed even though MRD was not detected in their bone marrow samples. Cave and colleagues interpret this as meaning that it is likely that these relapses resulted from the emergence of a malignant subclone. Also note that some patients showed MRD at each time point and did not relapse.

The authors suggest that in the future, "if the detection of residual leukemia is to be used in clinical practice to identify patients with a high probability of early relapse, two conditions must be met: the analysis should be performed as early as possible, and the laboratory technique should be simple and rapid so that treatment can be tailored to the adjusted assessment of risk."

previous | next | contents | bibliography

General Disclaimer

These pages are intended for informational purposes only and are not intended to render medical advice. The information provided on Ped Onc Resource Center should not be used for diagnosing or treating a health problem or a disease. It is not a substitute for professional care. If you suspect your child has a health problem, you should consult your health care provider.

© copyright 1998-2018 by Patty Feist

contact webmaster/ped-onc home/last updated 7/05