Validation of MRD Quantification By Flow Cytometry for Pediatric BCP ALL Relapsed Patients Treated on the Intreall Protocol

At acute lymphoblastic leukemia (ALL) relapse, about 40% of children can be salvaged with intensified multi-agent, high dose chemotherapy and in very high risk patients, with additional stem cell transplantation (SCT). To improve on this, the International BFM Study Group has led a consortium of 19 countries to develop the world’s largest trial for relapsed ALL (IntReALL). Standard risk patients will be randomized to receive the ALL-REZ BFM 2002 or UK ALL R3 therapy and post induction will be randomised to receive the additional targeted anti-CD22 drug, Epratuzumab, during consolidation, to clear residual disease. Children with end of re-induction MRD positive bone marrow will undergo SCT following consolidation. Both ALL-REZ BFM 2002 and UK ALL R3 used MRD PCR-based quantification of clonal Ig/TCR rearrangements, with different cut offs (10^-3 for ALL-REZ BFM 2002 and 10^-4for UK ALL R3) and this is the reference assay for IntReALL. However, flow MRD may also play a role for patients without PCR targets. Flow MRD relies on the discrimination of leukaemic blasts from hematogones and this can be hampered depending on the degree of haematopoietic regeneration which varies depending on the treatment protocol and is especially important after intensive induction treatment in relapsed protocols.

Thus, prospective MRD quantification of patients entered onto the UKALLR3 and ALL-REZ BFM 2002 clinical trials was performed by a standardised, quality assured, 4-8colour Flow MRD assay in end of re-induction bone marrow aspirates, by laboratories in the IBFM FLOW consortium (n=221). Flow MRD in both treatment protocols was classed as a prospective biological add on study and not used for clinical decision making. Median MRD levels were 0.026 +/-9.9% SD for BFM versus 0.027+/-18% SD for UK protocols, with comparative MRD positivity rates of 45% versus 54%, respectively. Comparison with MRD levels as assessed by molecular analysis of antigen receptor gene rearrangements was performed in 170 samples (BFM,128; UK R3, 42). The Spearman rank correlation of all samples was 0.90 (p<0.0001) for patients treated on the BFM protocol, compared to 0.82 (p<0.0001) for those on UK ALL R3. Risk category concordance was 88% (ALL-REZ BFM) and 88% (UKALLR3). For the 21 discordant samples, 5 were MRD positive by flow but negative by PCR and 17 were negative  by flow and positive by PCR. When analysing the accuracy, with which flow MRD classified specimens identically as PCR, the sensitivity of flow MRD in the ALL-REZ BFM protocol was 81% (cut off 0.1% ) and in UK ALL R3 was 79% (cut off 0.01%). Specificity values were 93% versus 100%, respectively. 

Although sample processing and quantification of MRD differ between PCR and FC MRD, in both re-induction protocols, there was good correlation of MRD levels assessed by flow cytometry and PCR, validating the use of Flow MRD as a method of choice in patients without PCR targets in the IntReALL trial. Flow MRD also has the advantage of enabling levels of CD22 to be assayed on MRD cells, prior to treatment with Epratuzumab.

This research has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no 278514 – IntReALL”, Deutsche Kinderkrebsstiftung for its funding support of the ALL-REZ BFM 2002 clinical trial and the minimal residual disease studies by PCR and the Deutsche Jose Carreras Leukämiestiftung for support of the international principal investigator, Leukaemia and Lymphoma Research and North East Children's Cancer  Research Fund, NT 13462-4, NV15-28525A, NV15-26588A, UNCE 204012.