HEMATOLOGIA TRASLACIONAL-BIOLOGIA MOLECULAR
Contributions
Type: Publication Only
Background
AML is a highly heterogeneous pathology with a variety of subtypes stratified function of cytogenetic and molecular markers. To date, the major prognosis criteria is cytogenetics, however the 40-50% of de novo patients have normal karyotype. Recent sequencing techniques such as NGS allow us to identify the mutational profile, to decide the diagnosis and risk stratification of patients and to personalize therapy.
Aims
To identify the mutational profile of AML and to correlate with the patient's survival data.
Methods
A cohort of 65 AML patients who were treated with intensive chemotherapy was selected. Mediam age at diagnosis was 52 years (range: 18-79), male:female ratio 30:35; 53 patients had de novo AML, 10 s-AML following MDS or MPN, and 2 t-AML; median leukocytes was 36.8x109/L. Cytogenetic results were available in 64 cases and categorized according to MRC (Blood 2010): favorable in 6 (9.4%), intermediate in 44 (68.8%), and adverse in 14 (21.9%). We performed targeted gene sequencing by NGS (Ion Torrent Proton System–Life Technologies) using a panel of 33 genes implicated in leukemia prognosis; in addition, FLT3 internal tandem duplication (ITD) was detected by GENSCAN and NPM1 mutation was detected by qPCR.
The discrete variables of patients with and without analyzed gene mutations were compared using the X2 test. Kaplan-Meier survival curves and Long-rank test were used for estimation of survival and difference between groups
Results
On average, 90,8% of the target sequence showed a mean depth coverage around 1300. We discovered 140 non-synonymous mutations which 118 were somatic single nucleotide variants (SNVs) and 22 small insertions and deletions (Indels) in coding regions. No mutations were detected in 5 samples, 14 samples presented 1 mutation, and the others 46 samples present 2 or more mutations. TET2 was the most frequently mutated gene (33,8%), followed by DNMT3A (26,2%), NPM1 (26,2%), FLT3 (21,5%), NRAS (20%), FLT3-ITD (18,5%), KMT2A (16,9%), IDH2 (13,8%), RUNX1 (13,8%), EPOR (12,8%), KRAS (12,8%), ASXL1 (9,2 %), IDH1 (9,2 %) and the others genes had a frequency below 8%.
Mutations in ASXL1 (p=0.005), KRAS (p=0.033), JAK2 (p=0.061) and ZRSF2 (p=0.019) were associated with secondary AML. DNMT3A (p<0.001) and ZRSF2 (p=0.005) mutations with NPM1 mutations. DNMT3A and ETV6 mutations with FLT3 ITD AML (p=0.041 and 0.029 respectively). AML patients who present TP53 or TET2 mutations showed a statistically non-significant correlation with death (p=0.103 and 0.062 respectively), in contrast IDH2 mutations were more frequent among alive patients (p=0.061). RUNX1 mutations were associated with female gender (p=0.023), and these and TP53 mutations were associated with refractoriness (p=0.009 and p=0.001, respectively). EZH2 and TET2 mutations were associated to AML patients > 55yrs (p=0.011 and 0.025 respectively). In contrast to KRAS mutations that were detected mainly in patients <55yrs (p=0.090).
Median RFS was significantly longer in female (28.5 vs 10.8months, p=0.017) and also in patients <55 years (18.9 vs 10.8 months, p=0.026). Median OS was significantly longer in patients <55 years (18.9 vs 11.8 months, p=0.038). No significant differences in RFS in function of analyzed gene mutations were found, but we observed that the median OS was significantly shorter in mutated TET2 patients (8.4 vs 18.8 months, p=0.049).
Summary
NGS is a useful technique to classify groups in AML with biological and prognosis implications. These results support the role of TET2 as an important poor prognostic biomarker in AML.
This study was funded by Instituto Carlos III (PI13/02387)
Keyword(s): Acute myeloid leukemia, Mutation
Session topic: Publication Only
Type: Publication Only
Background
AML is a highly heterogeneous pathology with a variety of subtypes stratified function of cytogenetic and molecular markers. To date, the major prognosis criteria is cytogenetics, however the 40-50% of de novo patients have normal karyotype. Recent sequencing techniques such as NGS allow us to identify the mutational profile, to decide the diagnosis and risk stratification of patients and to personalize therapy.
Aims
To identify the mutational profile of AML and to correlate with the patient's survival data.
Methods
A cohort of 65 AML patients who were treated with intensive chemotherapy was selected. Mediam age at diagnosis was 52 years (range: 18-79), male:female ratio 30:35; 53 patients had de novo AML, 10 s-AML following MDS or MPN, and 2 t-AML; median leukocytes was 36.8x109/L. Cytogenetic results were available in 64 cases and categorized according to MRC (Blood 2010): favorable in 6 (9.4%), intermediate in 44 (68.8%), and adverse in 14 (21.9%). We performed targeted gene sequencing by NGS (Ion Torrent Proton System–Life Technologies) using a panel of 33 genes implicated in leukemia prognosis; in addition, FLT3 internal tandem duplication (ITD) was detected by GENSCAN and NPM1 mutation was detected by qPCR.
The discrete variables of patients with and without analyzed gene mutations were compared using the X2 test. Kaplan-Meier survival curves and Long-rank test were used for estimation of survival and difference between groups
Results
On average, 90,8% of the target sequence showed a mean depth coverage around 1300. We discovered 140 non-synonymous mutations which 118 were somatic single nucleotide variants (SNVs) and 22 small insertions and deletions (Indels) in coding regions. No mutations were detected in 5 samples, 14 samples presented 1 mutation, and the others 46 samples present 2 or more mutations. TET2 was the most frequently mutated gene (33,8%), followed by DNMT3A (26,2%), NPM1 (26,2%), FLT3 (21,5%), NRAS (20%), FLT3-ITD (18,5%), KMT2A (16,9%), IDH2 (13,8%), RUNX1 (13,8%), EPOR (12,8%), KRAS (12,8%), ASXL1 (9,2 %), IDH1 (9,2 %) and the others genes had a frequency below 8%.
Mutations in ASXL1 (p=0.005), KRAS (p=0.033), JAK2 (p=0.061) and ZRSF2 (p=0.019) were associated with secondary AML. DNMT3A (p<0.001) and ZRSF2 (p=0.005) mutations with NPM1 mutations. DNMT3A and ETV6 mutations with FLT3 ITD AML (p=0.041 and 0.029 respectively). AML patients who present TP53 or TET2 mutations showed a statistically non-significant correlation with death (p=0.103 and 0.062 respectively), in contrast IDH2 mutations were more frequent among alive patients (p=0.061). RUNX1 mutations were associated with female gender (p=0.023), and these and TP53 mutations were associated with refractoriness (p=0.009 and p=0.001, respectively). EZH2 and TET2 mutations were associated to AML patients > 55yrs (p=0.011 and 0.025 respectively). In contrast to KRAS mutations that were detected mainly in patients <55yrs (p=0.090).
Median RFS was significantly longer in female (28.5 vs 10.8months, p=0.017) and also in patients <55 years (18.9 vs 10.8 months, p=0.026). Median OS was significantly longer in patients <55 years (18.9 vs 11.8 months, p=0.038). No significant differences in RFS in function of analyzed gene mutations were found, but we observed that the median OS was significantly shorter in mutated TET2 patients (8.4 vs 18.8 months, p=0.049).
Summary
NGS is a useful technique to classify groups in AML with biological and prognosis implications. These results support the role of TET2 as an important poor prognostic biomarker in AML.
This study was funded by Instituto Carlos III (PI13/02387)
Keyword(s): Acute myeloid leukemia, Mutation
Session topic: Publication Only