Poster Session II: Acute myeloid leukemia - Biology - translational research
The current standard for morphologic complete remission in acute myeloid leukemia (AML) is less than 5% myeloblasts, but mounting data show this criterion is not sufficiently sound. Alternative methods, such as quantitative reverse-transcription polymerase chain reaction (RT-qPCR), are widely used to detect molecular responses, but it relies on the initial detection of a fusion transcript, or overexpressed gene. Deeper knowledge of the clonal dynamics of AML could potentially be of clinical utility. A wide scope testing technology is required in order to address the molecular heterogeneity of AML. We reasoned that an appropriate Next Generation Sequencing (NGS) panel could be a useful tool to provide personalized molecular monitoring in patients diagnosed as or progress to AML.
The aim of this study is to evaluate the clinical utility of NGS panel in the prognostic and treatment monitoring in patients diagnosed as or progress to AML.
We studied the genomic alterations of 19 AML cases (13 de novo, and 6 secondary to a preexisting MN) during disease follow-up; 11 of these patients received hematopoietic stem cell transplantation (HSCT). The 67 samples were tested with our custom Pan-Myeloid Panel (48 genes, SOPHiA GENETICS). Samples were provided by the Biobank of the University of Navarra and were processed following SOP approved by the Ethical and Scientific Committee of the University. Libraries were pair-end sequenced on a Miseq sequencer (Illumina). Sequencing data were analyzed by two geneticists with expertise in hematological malignancies.
Sequencing data identified genomic clonal markers with clinical utility (i.e. diagnostic, prognostic, and/or predictive value) in 89,5% of cases. In patients not receiving HSCT (n = 8), NGS was useful to classify them in two genetic profiles: those achieving molecular complete remission (mCR) (n = 2) (Figure 1A), and those not responding to treatment and undergoing disease progression (n = 6) (Figure 1B). In the last group, NGS identified pathogenic variants in DDX41, DNMT3A, IDH1, JAK2, NRAS, SRSF2, U2AF1 genes. In patients receiving HSCT (n = 11), NGS also classified patients in two groups: those clearing pathogenic variants upon HSCT (n = 5) (Figure 1C), and those with persisting variants, not achieving mCR (n = 6) (Figure 1D). Again, NGS identified initial clones harboring pathogenic variants, like KRAS, that appeared in the 66% of the patients after HSCT failure. Also mutations in CBL, DNMT3A, FLT3, JAK2, KRAS and SRSF2 genes are present in this cohort of patients.
In two cases, NGS either did not detect any clinically relevant variant, or it detected variants only after disease progression; in these two cases an NGS panel was insufficient, and therefore more comprehensive studies are needed (e.g. exomes). Of note, NGS data detected clones harboring pathogenic variants in two patients with negative minimal residual disease (MRD), as measured by flow cytometry (Figure 1D), indicating that NGS could complement current gold standard follow-up method in some instances.
A 48-gene panel NGS has been useful for molecular diagnosis, treatment follow-up, and relapse detection in nearly 90% of the AML cases included in our study (17 of 19). NGS was also useful for following mutational clearance and/or clonal evolution in 12 of 19 patients (63%), including cases undergoing HSCT. According to our data, NGS could be of clinical utility for routine diagnosis and follow-up in an elevated proportion of AML patients, even complementing immunophenotypic techniques for MRD monitoring in some instances.