Detection of minimal residual disease (MRD) in childhood acute lymphoblastic leukemia (ALL); a comparative and feasibility study to design the best protocol for MRD-based treatment intervention
2000 - 12 / 2004
In 1998 the first three reports of large-scale clinical MRD studies in childhood ALL have become available, including the extensive study of the International BFM Study Group (I-BFM-SG). These studies clearly show the prognostic value of MRD information for predicting relapse. In the I-BFM-SG study we could recognize MRD-based risk groups of substantial size with highly significant differences in relapse rates. This risk group classification was based on the kinetics of tumor reduction during the first months of follow-up, as measured at two early consecutive time points, preferably using two different PCR targets per patient with sensitivities of at least 10 4. This information provides new openings for treatment stratification with treatment reduction for low-risk patients and treatment intensification for high-risk patients. In 1996 the Dutch Childhood Leukemia Study Group (DCLSG) decided to start a new treatment protocol (ALL-9) with the same backbone as the successful DCLSG ALL-6 protocol, which was less toxic than the two preceding BFM-like protocols (ALL-7 and ALL-8) and had a high survival rate of about 81%. No MRD-based treatment intervention could be included in this new ALL-9 protocol, because the protocol was started before the results of the large-scale MRD studies became available and because the MRD results of the I-BFM-SG study could not be extrapolated to the less toxic ALL-9 protocol. Therefore, it is important to evaluate the effectiveness of the ALL-9 treatment protocol via a large-scale MRD study in order to assess the kinetics of tumor load reduction during the first phases of treatment. Furthermore, we will evaluate whether the leukemic cells fully disappear during maintenance treatment and are also absent in most patients after treatment. The results of this 'early clinical study' will bring the DCLSG in the privileged position of having the choice between MRD-based stratification via an intensive ALL-8-like protocol (BFM backbone) or via a less toxic ALL-9-like protocol (ALL-6 backbone). This early clinical study has the following six aims: 1. MRD analysis in about 350 ALL-9 patients by use of real-time quantitative PCR (RQ-PCR) analysis of immunoglobulin (Ig) and T cell receptor (TCR) gene rearrangements as well as fusion gene transcripts of chromosome aberrations. Based on the extensive experience and using the available standardized laboratory protocols, it will be possible to detect, identify, and select at least two leukemic clone-specific Ig/TCR gene rearrangements in about 95% of all childhood ALL. Fusion gene targets will be detectable in 35-40% of cases. MRD in the above-mentioned clinical studies was monitored with semi-quantitative methods. The importance of the kinetics of MRD regression prompted us to develop quantitative MRD techniques by RQ-PCR using the TaqMan technology. 2. Comparison of fusion gene transcripts and Ig/TCR gene rearrangements as PCR targets. In patients with fusion gene transcripts, we will compare the RQ-PCR results of the Ig/TCR targets and fusion gene targets with respect to sensitivity and quantification of MRD levels. This comparison should answer the question whether fusion gene targets can replace the Ig/TCR gene targets and thereby avoid the sequencing of junctional regions and usage of clone-specific probes or primers. 3. Comparison of MRD monitoring via BM and PB sampling. The frequent BM sampling during follow-up is a traumatic experience for children with ALL. Therefore, we will evaluate to what extent PB sampling can replace BM sampling and whether this is possible at all time points during follow-up. 4. Feasibility study of prospective MRD analysis in about 60 patients. In the last phase of the study we will perform the MRD analyses as if it would concern a fully prospective study. 5. Clinical evaluation of the ALL-9 MRD results and comparison with the I-BFM-SG study. The prognostic value of MRD information in the two treatment protocols will be compared. 6. Design of an MRD-based intervention treatment protocol, DCLSG ALL-10. In the last phase of the proposed 'early clinical study' (i.e. in 2003) all relevant MRD data will be available for decision making and for the design of a new treatment protocol with MRD-based stratification, whether it is an ALL-8-like protocol or an ALL-9-like protocol. In conclusion, MRD detection appears to be the most important prognostic factor for predicting relapse in childhood ALL. In previous years we have developed standardized laboratory techniques that are needed for implementation of this new diagnostic approach for treatment stratification. However, before this can be done the clinical information of the proposed study is needed. The results of this study will not only be relevant for childhood ALL; the technical strategies and logistics are of direct relevance for MRD studies in adult ALL patients as well as in patients with acute myeloid leukemia.