Myelodysplastic syndromes (MDS) represent a heterogeneous group of clonal hematopoietic disorders, characterized by ineffective hematopoiesis resulting in cytopenias and a highly variable clinical course. Cytogenetics are routinely used as one of the diagnostic, prognostic and therapeutic markers in the clinical management of MDS (Greenberg et al. Blood 120:2454,2012). Although karyotyping is generally considered as the gold standard in the cytogenetic characterization of MDS, 40-60% of the patients exhibit a normal karyotype. Intrinsically, the resolution of karyotyping is limited by its capacity to detect only those copy number changes that are microscopically visible (5-10 Mb in size). In contrast, microarray-based genomic profiling analyses allow a genome-wide detection of copy number alterations (CNAs), down to 100 kb in size, and regions of copy neutral loss of heterozygosity (CNLOH). Such analyses also overcome some of the other major limitations of karyotyping such as low success rate due to inadequate metaphase yield and/or poor banding quality. We have compared karyotyping and fluorescence in situ hybridization (FISH) with microarray-based genomic profiling with respect to the detection yield for genetic abnormalities in bone marrow samples from lower risk MDS patients.

We used the HOVON89 study-cohort, a prospective phase II randomized multicenter study to assess the efficacy of lenalidomide with or without erythropoietin and granulocyte-colony stimulating factor in patients with low/intermediate-1 risk MDS; www.trialregister.nl; NTR1825; EudraCT nr.: 2008-002195-10. Inclusion target of the study is 200 low/intermediate-1 risk MDS patients (134 enrolled, inclusion ongoing). Data regarding cytogenetics, FISH and microarray were obtained in a fully blinded fashion for 68 MDS patients. For microarray-based genomic profiling we used the recently launched high resolution CytoScan HD Array (Affymetrix) platform. The following interpretation criteria were applied: (i) the threshold for CNAs was set at >5 Mb, (ii) inclusion of <5 Mb CNAs segments that coincide with known cancer genes as reported on www.sanger.ac.uk and (iii) the threshold for CNLOH was set at >10 Mb and to telomere. Karyotyping and interphase FISH were performed using standard cytogenetic methods. In all 4 patients where karyotyping revealed no metaphases interphase FISH was performed.

Thirty-six of the 68 (53%) MDS patients had an abnormal microarray profile. Of interest, in 13 of these 36 patients no abnormalities were observed by karyotyping and/or FISH. All these abnormalities observed by microarray only, involved focal <5 Mb CNAs (containing e.g. the TET2, RUNX1 and DNMT3A genes) and regions of CNLOH (coinciding with 1p, 2p, 4q, 7q, 11p, 11q, 12q, 14q and 18q), which are all out of the scope of karyotyping and FISH. All CNAs identified by karyotyping and FISH were also observed by microarray-based genomic profiling, including a case with loss of 5q in 5% of the cells and loss of 7q in 9% of the cells as observed by interphase FISH and another case with loss of 5q in 3 of 20 the analyzed metaphases by karyotyping. These observations demonstrate the high sensitivity of the CytoScan HD array platform for the identification of CNAs in (small) subclones. As expected balanced translocations such as t(3;3)(q21;q26) and t(2;14)(q37;q22) present in 2 of the patients in our cohort were not identified by microarray-based genomic profiling. This study will be extended to all patients to be included in the HOVON89 trial.

In conclusion, we demonstrate that in the present cohort of 68 patients, microarray-based genomic profiling allows the identification of almost all copy number abnormalities also observed by karyotyping and FISH. In addition, we show that microarray-based genomic profiling allows the detection of potential prognostic relevant abnormalities (focal CNAs and CNLOH) which would have remained undetected by karyotyping and FISH. The predictive and/or prognostic value of these novel CNAs and CNLOH will be evaluated within the ongoing prospective clinical HOVON89 trial in lower risk MDS.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.