Abstract
Abstract 4813
Accurate cytogenetic and molecular diagnosis of acute myeloid leukemia help to establish the prognosis and to select the best postconsolidation treatments. This genetic characterization is currently being performed by means cytogenetic techniques and molecular methods mainly based in PCR, capillary electrophoresis and Sanger sequencing. Recently, large and homogeneous series of AML patients have been analyzed by means expression arrays. These studies have helped to standardize the method and they are becoming an alternative to traditional methods in terms of simplicity and time.
To assess the usefulness of the AMLprofiler™ in genetic stratification of de novo AML enrolled in a cooperative Group. To investigate the applicability of the array diagnostic platform in multicenter trials. To compare the AMLprofiler™ with the standard approach.
Twenty one consecutive de novo AML cases enrolled in the Spanish CETLAM protocol were included in the study. Patients were recruited during 2012. Cytogenetic analysis was performed in each participating center and the results were sent to the HSCSP. Complete immunophenotype and molecular studies were performed centrally at the HSCSP. Purified DNA was employed to test FLT3-ITD and TKD mutations, NPM1, CEBPa and MLL. For each FLT3-ITD positive case, the allelic ratio was calculated. Given that FLT3 mutations are not detected by the AML-profiler™, this determination was common to both approaches. Purified RNA was used to assess the presence of the AML1-ETO and CBFb-MYH11 rearrangements and to test RNA integrity. We used the AMLprofiler™ assay which detects chromosomal aberrations (t(8;21), t(15;17), inv(16), mutations (CEBPA dm, NPM1 A/B/D) and genetic expresion of BAALC and EVI1. Only those cases fulfilling the quality required by the manufacturer were hybridized the AMLprofiler™array. Based on this requirement one case was ruled out.
Twenty cases were analyzed with the AMLprofiler™ and with the conventional methods (21 intended/20 hybridized). There were 4 cases with FLT3-ITD. One case had a t(8;21) and it has been succesfully detected by both methods. There were 4 inv(16), in one case the RNA quality was not satisfactory and was not hybridized, in the remaining 3 cases the AMLprofiler detected the genetic lesion including one case with a cryptical translocation and in one case with a typical inv(16) but with a complex CBFb-MYH11 transcript. In 8 cases a NPM1 mutation was detected by molecular methods, in all NPM1 A, B or D, the AMLprofiler™ gave concordant results, one case with a non-ABD mutation was not detected. When considering technician times, AMLprofiler™ compared favorably with the conventional molecular biology techniques (mean times: 2958 min vs 3422 min). This figure does not take into consideration the cytogenetic workload performed in each participating center.
The AMLprofiler™ could be easily used to stratify de novo AML patients enrolled in multicenter trials. It provides accurate results and requiring less time than the currently employed methods in the CETLAM group. The AMLprofiler™ seems to be specially useful in the core binding-factor leukemias. The most important limitations are related to the need of an excellent RNA sample and the presence of rare NPM1 mutations.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
