Mathematical Modeling and Comparisons of Relapse Patterns in AML with 11q23/MLL translocations and Normal Karyotype AML with MLL-PTD

Abstract

Abstract 750▪FN2▪This icon denotes a clinically relevant abstract INTRODUCTION:Close monitoring of minimal residual disease (MRD) in acute myeloid leukemia (AML) patients in complete remission (CR) holds the promise of detecting relapses in time to initiate cytoreduction prior to a full-blown hematological relapse. However, to implement MRD measurements in protocols as well as clinical practice, key parameters such as optimal sampling intervals and sampling tissues (peripheral blood (PB) vs. bone marrow (BM)) are necessary. We have previously defined these for PML-RARA+, CBFB-MYH11+, RUNX1-RUNX1T1+, NPM1 mutated AML and for AML with over-expression of the WT1 gene (Ommen, et al, BLOOD, 2010). However, the important subgroups with 1) rearrangement of the MLL gene 11q23 translocations (t(11q23)) and 2) partial tandem duplication of the MLL-gene (MLL-PTD) have not been studied, yet. AIM: Optimal sampling intervals and relapse detection rates were investigated in the two AML subgroups with t(11q23) and MLL-PTD. Median time from MRD positivity to hematological relapse should be defined. PATIENTS AND METHODS: 1,321 MRD samples from 258 patients were available for analysis. Based on CR samples from patients who did not experience a relapse we initially established a cut-off level for MRD conversion above which hematological relapse was inevitable (MLL-PTD: 45 patients, 150 samples and t(11q23): 55 patients, 212 samples). Based on MRD courses from 21 pts with MLL-PTD experiencing 24 relapses (CR samples: n=149) and 16 pts with t(11q23) with 17 relapses (CR samples: n=100) we analyzed relapse kinetics using the model we have previously published (Ommen, et al, BJH 2008). This model exploits the information from both positive and negative samples taken prior to relapse. The model establishes the relationship between sampling intervals and the likelihood of discovering a patient positive prior to hematological relapse on the one hand, and the median time from first positive sample to hematological relapse (HR) on the other. RESULTS: We observed strikingly different relapse kinetic patterns in the AML subgroups studied here. Thus, MLL-PTD levels were found to be very informative in some cases with conversion to MRD positivity occurring up to 10 months prior to HR (conversion 4–10 months before relapse in 5/24 relapses, but only 45% (5/11) of patients where bone marrow status could be evaluated 2 month before HR were qPCR positive. Based on this data a 3-monthly BM sampling schedule yield only mediocre relapse detection rates (60%) but in the cases where a positive sample is obtained, relative long time passes before hematological relapse occurs (2 months). Conversely, in the cases with the different t(11q23) markers (t(9;11): n=7, t(6;11): n=5, t(10;11): n=2 and t(11;19): n=2) relapses were correlated with very homogenous relapse kinetics. None of the t(11q23) relapses could be detected 3 months before HR. On the other hand, 80% (4/5) of relapses had converted to negativity as assed by quantitative 2 months before HR. This translated into a 3-monthly BM sampling schedule yielding relapse detection rates of 75% with a median time to HR of 1.1 months. Compared to the relapse kinetics of previously published AML subtypes, the in vivo growth of t(11q23) leukemic cells (leukemic clone median doubling time 9 days, range 4.5–15 days) is faster than both the in vivo growth of CBFB-MYH11 relapses (doubling time 36 days, range 7.4–175 days) (P<0.01) and NPM1c+FLT3-ITD- relapses (doubling time 15 days, range 2.2–33 days) (P=0.07) (Ommen, et al, Blood 2010). CONCLUSION: These data add considerably to our knowledge of molecular relapse kinetics in AML by adding t(11q23) cases and those with MLL-PTD to those already described in the literature. They can be utilized for statistical power calculations when designing clinical trials and allow for cost-benefit calculations regarding MRD monitoring implementation in clinical decision-making in the individual patient. Disclosures:Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Alpermann:MLL Munich Leukemia Laboratory: Employment. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Schnittger:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.