Early Blast Clearance Evaluation after Induction Chemotherapy for Acute Myeloid Leukemia By Multiparameter Flow Cytometry and WT1-RNA Quantification: A Single Center Experience

Introduction: in acute myeloid leukemia (AML) the response to treatment is evaluated upon full recovery of peripheral blood counts by bone marrow (BM) assessment using morphology and cytogenetics, if appropriate number of metaphases is obtained. Real time quantitative PCR (RQ-PCR) and multiparametric flow citometry (MFC) are sensitive techniques to assess minimal residual disease (MRD) mostly used to refine risk stratification and to guide therapy. Evaluation of response to induction chemotherapy (CT) during aplasia (around day 14 from start of induction) has shown a significant prognostic impact, the presence of residual disease predicting a worse prognosis. All reported studies on early BM blast clearance evaluation rely on morphology as a single technique. Unfortunately, at day 14 BM is often not evaluable for cytology and morphologyc blast count. Unpredictability and intra-observer variability must also be considered when assessing blast count by morphology alone. No studies analysing early BM blast clearance by means of MFC and/or PCR have been published and the correlation between the two assays is unknown in this setting. Moreover, it is to be established if early intensification (around day 15 from start of induction) with a second cycle of CT could increase the CR rate in pts with disease persistence in the BM evaluated at day 14.

Matherial and methods: we retrospectively evaluated data of 23 newly diagnosed AML pts who received induction CT at our center between 02/2009 and 05/2014, and for whom analysis of BM both at day 14 and after hematologic recovery (around day 28) was performed. The aim of our study was to define the prognostic value of early MRD quantification by MFC (d14-LAIP) and WT1 quantification by PCR (d14-WT1) in predicting the response to induction. Firstly we compared d14-LAIP and d14-WT1 to identify the more sensitive and specific assay in predicting the response to induction, in particular for cases not evaluable for morphologic blast count. Then we compared the outcome of pts who received or did not receive an early intensification for persistence of disease at d14 BM evaluation.

Results: 20 pts received the 3+7 induction regimen, 3 pts the ICE induction regimen. After BM evaluation at day 14, 7 pts received early reinduction CT (FLAG-IDA regimen) starting at day 16 (median), 16 pts did not receive further therapy before BM evaluation at day 28. Overall CR rate was 70% (16 pts), PR/NR 30% (7 pts), TRM 4% (1 pt). At day 14, leukemic blast percentage was not evaluable by morphology in 8 (35%) cases due to marrow aplasia, in 1 (4%) case blasts were <5%, in 14 (61%) cases blasts were ≥5%; by MFC (23 cases), in 9 (39%) cases blasts were ≤2%, in 14 (61%) were >2%; by PCR (17 cases), in 4 (24%) cases WT1 was <250 cp/10e4 ABL, in 13 (76%) WT1 was ≥250 cp/10e4 ABL. At day 28 BM evaluation, of the 8 pts with aplastic marrow at day 14, 6 (75%) were in CR and 2 (25%) in PR/NR, of the 14 pts with blasts ≥5% at day 14, 9 (64%) were in CR and 5 (36%) in PR/NR. Analysis of the paired results from nadir to recovery revealed that d14-LAIP 2% had a positive predictive value (PPV) of CR at day 28 of 71% and negative predictive value (NPV) of 89%, with a sensitivity of 83% and a specificity of 80%, d14-WT1 250 had a PPV of 50% and NPV of 25%, with a sensitivity of 57% and a specificity of 20%. The d14-LAIP analysis was strongly associated with CR after induction (p 0.034). Considering the 14 pts with blasts ≥5% at day 14, 6/7 (86%) who received early reinduction CT and 3/7 (43%) who did not, obtained the CR at day 28 (p ns).

Discussion: in our series d14-LAIP proved to be more predictive of response after induction CT than d14-WT1. Although one-third of pts had an early morphologic response not evaluable due to marrow aplasia MFC proved to be a useful assessment tool with a 100% applicability. Moreover, correlation between D14-LAIP and d14-WT1 was > 90%. Our data confirm the prognostic value of day 14 BM evaluation and suggest that MRD detection, also in aplasia, could drive early reinduction CT which probably could increase the CR rate, without significantly clinical complications. Anyway, this last point must be confirmed with a larger study.