Prognostic Impact of Monosomal Karyotype in Patients with Myelodysplastic Syndrome and Abnormal Karyotype. A Report From the Spanish Group of MDS (GESMD)

Abstract 1724

Cytogenetic abnormalities (CA) are the most important prognostic factor in patients with myelodysplastic syndromes (MDS). Monosomal karyotype (MK) defined as the presence of at least two autosomal monosomies or one monosomy plus other structural CA, has been associated with poor prognosis in acute myeloid leukemia (AML) but its significance in MDS remains unclear. The aim of our study was to analyze the prognostic impact of MK in adult patients with MDS and CA.

Patients from Spanish Registry of MDS diagnosed with MDS by WHO 2008 criteria and with CA detected by conventional cytogenetics have been included in the study. Statistical analysis consisted of Kaplan-Meyer univariate analysis (UA) including all known variables associated with prognosis in MDS and a Cox-regression multivariate analysis (MA) in which we included only those variables with a P<0.1 in UA.

There were 1054 patients, 478 (45.3%) women. The median age was 71 (range 16–96) years. Median follow-up for survivors was 24.1 (range 0–210) months. There were 609 (57,8%) refractory anemia (RA) and 445 (42.8%) refractory anemia with excess of blasts (RAEB). The IPSS was low, intermediate-1, intermediate-2 and high in 18.3%, 39.3%, 25%, and 10.6%, respectively (6.7% were unclassifiables). Complex karyotype (CK) and MK was observed in 203 (19.3%) and 172 (16.3%) patients, respectively. Patients with MK showed worse prognostic characteristics than those without MK: More Intermediate-2 and high risk patients (50% vs. 20% and 30% vs. 6.7%; P<0.001); more frequent CK: (87.2% vs. 6%; P<0.001), higher median bone marrow (BM) blast count (7% vs. 3%; P<0.001) and lower hemoglobin (Hb) level (89 vs. 96 g/L; P<0.001).

Median OS for the whole group was 32.7 months. In the UA, the variables associated with lower OS were: Male sex, RAEB subtype (vs. RA), higher IPSS, CK, MK, older age, higher peripheral blood (PB) and bone marrow (BM) blast percentage and lower Hb, platelets and neutrophil count. In the MA the variables associated with lower OS were: Age>60 years (HR 1.7; P<0.001), CK (HR 2.19; P<0.001), higher BM blast (HR 1.07; P<0.001), Hb<100 g/L (HR 1.788; P<0.001) and platelet count <100×10⁹/L (HR 1.62; p<0.001). MK did not reach statistical significance in the MA (HR 1.45; P=0.059).

In the group of patients with CK the UA showed that the presence of MK was associated with a trend to lower OS (Log rank 2.8, P=0.092) while the presence of ≥4 vs. 3 CA was associated with lower OS (Log rank 7.5; P=0.006). Other variables associated with lower OS in UA in this subset of patients were: RAEB (vs RA), presence of abnormalities of 5 and/or 7 chromosome, higher IPSS group, older age, higher BM blast percentage, and lower Hb and platelet count. In MA the variables associated with lower OS in patients with CK were: age>60 years (HR 1.734, P=0.008), RAEB vs. RA (HR 1.542, P=0.02) Hb <100 g/L (HR 2.1, P<0.001) platelets <100×10⁹/L (HR1.886; P=001) and the presence of abnormalities of both 5 and 7 chromosomes (HR 2.058; P=0.001). The presence of MK did not reach statistical significance in the MA.

At last follow-up, 221(21%) patients had shown AML evolution at a median of 9 months (0–125 months) for a 1 and 4 years probability of AML evolution of 14.2% (95% CI 11.8–16.6) and 28.6% (95 CI: 24.8–32.4). The variables associated with higher risk of AML evolution in the UA were: RAEB, higher IPSS, CK, MK, higher PB and BM blast percentage and lower hemoglobin, platelet and neutrophil count. In multivariate analysis the only variables that retained statistical significance were BM blasts (HR 1.13; P<0.001) and CK (HR 3.2; P<0.001).

In conclusion, our study shows that the presence of MK is highly associated with CK and other high-risk features of MDS. In our population, the MK was not an independent risk factor for OS while the presence of CK was the main risk factor associated with poorer OS in MDS patients with abnormal karyotype.