AbstractAbstract 1882Evidence exists for differences between eastern and western type MDS. However, for the development of the IPSS Score, 6 of the participating study groups were European or US, while only one group originated from Japan.In order to find out more about the differences in MDS patients (pts) from separate ethnic groups we performed an analysis of 1048 pts from Dusseldorf, Germany (G), 242 pts from Seoul, Korea (K), and 143 pts from Saitama, Japan (J). Initially, we compared clinical characteristics. Then we evaluated the influence of important prognostic factors. For all pts included at least the parameters needed to determine the IPSS had to be available. For survival analyses pts who received induction chemotherapy or allogeneic transplantation were excluded.Median age was 66 yrs in Germany, 54 in Korea, and 68 in Japan. The distribution of gender, with a male preponderance, was comparable in all cohorts. Median blast count was <5% in all groups. Median cell counts were significantly lower in the Japanese and Korean, when compared to the German pts. However, although Asian pts tended to have more severe cytopenias regarding all 3 lineages, they also had a significantly higher percentage of refractory cytopenias with unilineage dysplasia (21% (K) and 36% (J) compared to 7% (G)). MDS with del5q was significantly more frequent in German pts (9 vs 0 vs 1%) as well as RCMD (40 vs 22 vs 18%), while the frequency of RAEB I/II was comparable. Due to the high incidence of pancytopenia Asian pts rarely had a low risk IPSS (2% (K), 12% (J) vs 27% (G)), while the int I risk group was larger and int II and high risk groups were of almost the same size. Regarding karyotype (KT) risk according to IPSS, the intermediate risk group was smaller, and the low risk group was larger in Asian pts.To evaluate the relevance of different prognostic factors we first compared the whole cohorts of not intensively treated pts. Since the result was all factors analyzed were important in German while many were not in Asian pts we decided to look at a smaller subgroup of German pts to compare similar sizes. Pts were chosen by chance to avoid any kind of bias by matching. The results are presented in table 1. Already survival was very different in the 3 groups: 31 months (ms) (G) vs 43 ms (K) vs 157 ms (J). In univariate analysis in German pts variables with significant influence on survival were Hb, sex, age, LDH, platelets, blast count, KT, and IPSS, while in the Korean and Japanese pts the 3 different cytopenias and LDH had no influence on survival (except a borderline influence of Hb in Japanese pts). Very striking was the importance of gender in the 2 Asian countries. In both men lived for a median of only about 2 years, comparable to German pts, while women had a very long survival with median not reached. Regarding KT risk, the intermediate risk group had in the Asian countries a significantly longer survival than in Germany (nr vs 24 ms), possibly due to a high frequency of +8. The IPSS divided, again in both Asian countries, 2 different risk groups, but not 4. Multivariate analysis identified Hb, platelets, blasts, LDH, age, and KT as independent risk factors for German pts, compared to ANC, blast count, and age in Korean, and blast count plus KT in Japanese pts.It is established that survival of Asian MDS pts is longer than that of European, but it remains unclear why there is such a strong difference between Korean and Japanese pts although the former are younger and most other features are very comparable between the two. The most striking features of Asian MDS are the strong influence of gender, as well as the lesser importance of cytopenias compared to European pts. This, together with a slight difference in the survival according to KT, leads to an inferior separation of risk groups by the IPSS. The most important factor with equal relevance in all MDS pts remains blast count.Table 1:Germany n = 250Korea n = 224Japan n = 139survival/ms psurvival psurvival pall pts3143157Hemoglobin </≥10 g/dl24 vs 520,0236 vs 73ns145 vs nr0,01Sex m/w24 vs 390,00517 vs nr<0.0000530 vs nr<0.00005Age </≥ 60 years61 vs 240,0004100 vs 16<0.00005207 vs 1010,01LDH </≥ 240 U/l21 vs 520,000116 vs 40ns145 vs 174nsPlatelets </≥ 50000/μl17 vs 47<0.0000533 vs 53ns98 vs 207nsANC </≥1000/μl31 vs 33ns40 vs. 52ns101 vs 174nsKT -risk42 vs 24 vs 10<0.0000553 vs nr vs 120,001207 vs nr vs 22<0.00005Blast count </≥ 5%61 vs 22<0.00005100 vs 16<0.00005207 vs 20<0.00005IPSS82 vs 31 vs 23 vs 8<0.00005nr vs 73 vs 16 vs 12<0.0000598 vs nr vs 21 vs 22<0.00005 Disclosures:No relevant conflicts of interest to declare.
Read full abstract