Multivariate Evaluation of the Prognostic and Therapeutic Relevance of Cytogenetics in a Merged European-American Cohort of 3860 Patients with MDS.

This study investigated the influence of cytogenetic findings on survival in MDS, after accounting for other known prognostic parameters and type of treatment. We identified 3210 patients with a known therapy regimen from our large database, including 3860 patients with MDS and secondary AML following MDS. This data pool was collected in the framework of a cooperative project merging data from the German-Austrian MDS study group (55% of pts.) and the MD Anderson Cancer Centre (MDA), Houston, USA (45% of pts.). Median age of all study patients was 65.9 years, the female-male ratio 1:1.53. After accounting for age, sex, % marrow blasts, de novo vs. primary MDS, treatment (supportive care, AML-type, and epigenetic therapy) and site of treatment(Germany/Austria vs.MDA) we looked at the effect of cytogenetic subgroups on survival. Survival data was available for 95.8% of all pts. The distribution of main cytogenetic categories was as follows: normal (1.727, 45%), 5q- (208, 5%), 5q- +1 abnormality (73, 2%), −7/7q- not complex (222, 6%), +8 not complex (203, 5%), complex with −5/−7 (627, 17%), other complex (195, 5%), 20q- not complex (69, 2%), and other not complex (475, 13%). Chemotherapy (c) was applied in 32.9% while 67.1% were treated with supportive care (sc) or epigenetic therapy. The cytogenetic prognosis according to IPSS-criteria was favourable in 1704 (53.1%) and unfavourable in 839 (26.13%) of the 3210 study patients. Multivariate analysis revealed the following non-cytogenetic parameters as unfavourable: age >60, male sex, blasts >5%, secondary MDS, AML-like chemotherapy, MDA as site (due to overrepresentation of high risk MDS). After accounting for these variables the following relative Hazard ratios (HR of 1.0 for normal karyotype) were calculated allowing a relative ranking of cytogentic findings: 5q-: 0.93, 5q- +1: 1.06, other not complex: 1.11, 20q-: 1.31, +8: 1.65, complex without −5/−7: 1.76 −7/7q-: 2.09, complex with −5/−7: 3.88. The effect of cytogenetics in pts. with either supportive care (sc) or chemotherapy (c) measured as relative HR (see above) were: 20q- (sc: 1.25, c: 1.51), 5q- (sc: 0.93, c: 0.73), other not complex (sc: 1.07, c: 1.04), −7/7q- (sc: 2.10, c: 2.09), +8 (sc: 2.05, c:1.54), complex without −5/−7 (sc: 2.23, c: 1.75), complex with −5/−7 (sc: 4.78, c: 3.77) showing that cytogenetics remains it’s prognostic relevance independent from the therapy applied. In Kaplan-Meier-analyses median survival in pts. showing favourable karyotypes was 37.9 months with supportive treatment as compared to 26.4 months with chemotherapy (p<0.01). We observed no therapy-related survival differences with regard to patients of the unfavourable cytogenetic subgroup. The occurrence of complex abnormalities (n=670) was associated with an identical median survival comparing the therapy groups (c: 7.0, sc: 7.1 months). We also investigated groups with distinct abnormalities (5q-, −7/7q-, +8, 20q-) and found significant differences in survival between the therapy groups: Patients with 5q- and 20q- benefited from supportive care while −7/7q- was related with a better survival when chemotherapy was applied. Patients with trisomy 8 showed no differences in survival. Regarding these data, chemotherapy response, expressed by survival, is closely associated with cytogenetics. Complex abnormalities always show a dramatically reduced outcome, independent from the therapeutic strategy.