HEMATOLOGIE CLINIQUE
Contributions
Type: Oral Presentation
Presentation during EHA20: From 13.06.2015 16:00 to 13.06.2015 16:15
Location: Room C1
Background
Up to 50% of lower risk MDS respond to ESA. While response can be predicted by serum(s) EPO level and transfusion dependence (TD), it is unknown if molecular findings are also prognostic.
Aims
We studied prognostic factors of erythroid response (HI-E, IWG 2006 criteria) to ESA in 90 lower risk MDS treated with ESA in France (GFM), Italy (FISM) and Germany (GMDS) with a focus on molecular mutations.
Methods
We studied prognostic factors of erythroid response (HI-E, IWG 2006 criteria) to ESA in 90 lower risk MDS treated with ESA in France (GFM), Italy (FISM) and Germany (GMDS) selected on the availability of all the following: sEPO, serum ferritin (SF), marrow slides for review (especially of dysplasia). In 79 of them, mutational analysis of 39 genes recurrently mutated in myeloid malignancies was performed on marrow cell DNA collected before ESA treatment, using an AmpliSeq approach (Life Technologies), and confirmed by Sanger method when the variant allele frequency was found upper than 20%.
Results
Median age was 74 years, with 56% males, 26% RBC TD patients, median Hb level 9.5 (range 6.2-11) in non RBC-TD patients, median sEPO level 59 U/l (range 9-402), median SF 411 µg/l (range 11-1856), WHO diagnosis : 21% RA; 24.4% RAEB-1; 24.6% RARS; 24.4% RCMD; 2% del(5q); 3.6% MDS-U. IPSS was low (45.3%) and int-1 (54.7%), IPSS R: very low (6.6%), low (59.7%) and int (33.7%). 85% pts had dyserythropoiesis, 62% dysgranulopoiesis, 67% dysmegakaryopoiesis. HI-E, was 64.5% and median response duration 19.8 months.
Clonal and subclonal mutations were found in: SF3B1 (40.5% patients), TET 2 (35.4%), ASXL1 (31.4%), DNMT3A (20.2%), U2AF1 (10.1%), SRSF2 (8.8%), IDH1/IDH2 (7.4%), RUNX1 ( 2.5%), STAG2 (11.4%), NRas (3.7%), KRas (3.7%), BCOR (3.7%), Jak2, PHF6, and CBL in 1 pt each. 86.1% patients had at least one mutation (median number 2,range 0-6). As previously described by others, SF3B1, DNMT3A and TET2 co-occurred frequently. ASXL1 and DNMT3A seemed to be mutually exclusive and SRSF2 never co-occurred with SF3B1.The number of mutations was not correlated with the degree of dysplasia. By univariate analysis, male gender, sEPO level >100U/L, IPSS-R (int vs very low and low) were correlated with worse HI-E. Patients with ≤ 2 mutations had better HI-E (OR: 0.32, p=0.02), but none of the most frequent mutations (SF3B1, ASXL1, TET2, DNMT3A), age, karyotype and importance of dysplasia had any impact on HI-E. In an adjusted logistic regression model, male gender, sEPO level>100U/L and RBC- TD, but not number of mutations, were significantly correlated with poorer HI-E.
Summary
Presence of >2 mutations predicted worse HI-E to ESA in lower risk MDS. However, in multivariate analysis, mutational analysis lost its prognostic value compared to more conventional factors, especially RBC-TD and sEPO level.
Keyword(s): EPO
Session topic: MDS Clinical
Type: Oral Presentation
Presentation during EHA20: From 13.06.2015 16:00 to 13.06.2015 16:15
Location: Room C1
Background
Up to 50% of lower risk MDS respond to ESA. While response can be predicted by serum(s) EPO level and transfusion dependence (TD), it is unknown if molecular findings are also prognostic.
Aims
We studied prognostic factors of erythroid response (HI-E, IWG 2006 criteria) to ESA in 90 lower risk MDS treated with ESA in France (GFM), Italy (FISM) and Germany (GMDS) with a focus on molecular mutations.
Methods
We studied prognostic factors of erythroid response (HI-E, IWG 2006 criteria) to ESA in 90 lower risk MDS treated with ESA in France (GFM), Italy (FISM) and Germany (GMDS) selected on the availability of all the following: sEPO, serum ferritin (SF), marrow slides for review (especially of dysplasia). In 79 of them, mutational analysis of 39 genes recurrently mutated in myeloid malignancies was performed on marrow cell DNA collected before ESA treatment, using an AmpliSeq approach (Life Technologies), and confirmed by Sanger method when the variant allele frequency was found upper than 20%.
Results
Median age was 74 years, with 56% males, 26% RBC TD patients, median Hb level 9.5 (range 6.2-11) in non RBC-TD patients, median sEPO level 59 U/l (range 9-402), median SF 411 µg/l (range 11-1856), WHO diagnosis : 21% RA; 24.4% RAEB-1; 24.6% RARS; 24.4% RCMD; 2% del(5q); 3.6% MDS-U. IPSS was low (45.3%) and int-1 (54.7%), IPSS R: very low (6.6%), low (59.7%) and int (33.7%). 85% pts had dyserythropoiesis, 62% dysgranulopoiesis, 67% dysmegakaryopoiesis. HI-E, was 64.5% and median response duration 19.8 months.
Clonal and subclonal mutations were found in: SF3B1 (40.5% patients), TET 2 (35.4%), ASXL1 (31.4%), DNMT3A (20.2%), U2AF1 (10.1%), SRSF2 (8.8%), IDH1/IDH2 (7.4%), RUNX1 ( 2.5%), STAG2 (11.4%), NRas (3.7%), KRas (3.7%), BCOR (3.7%), Jak2, PHF6, and CBL in 1 pt each. 86.1% patients had at least one mutation (median number 2,range 0-6). As previously described by others, SF3B1, DNMT3A and TET2 co-occurred frequently. ASXL1 and DNMT3A seemed to be mutually exclusive and SRSF2 never co-occurred with SF3B1.The number of mutations was not correlated with the degree of dysplasia. By univariate analysis, male gender, sEPO level >100U/L, IPSS-R (int vs very low and low) were correlated with worse HI-E. Patients with ≤ 2 mutations had better HI-E (OR: 0.32, p=0.02), but none of the most frequent mutations (SF3B1, ASXL1, TET2, DNMT3A), age, karyotype and importance of dysplasia had any impact on HI-E. In an adjusted logistic regression model, male gender, sEPO level>100U/L and RBC- TD, but not number of mutations, were significantly correlated with poorer HI-E.
Summary
Presence of >2 mutations predicted worse HI-E to ESA in lower risk MDS. However, in multivariate analysis, mutational analysis lost its prognostic value compared to more conventional factors, especially RBC-TD and sEPO level.
Keyword(s): EPO
Session topic: MDS Clinical