Background and Purpose

Myelodysplastic syndrome (MDS) is a group of heterogeneous disorders characterized by ineffective hematopoiesis and risk of leukemia transformation. The exact pathophysiology of MDS remains incompletely defined. Endoplasmic reticulum (ER) stress response is induced when environmental stimuli and metabolic changes of cells cause derangement of normal ER functions and accumulation of unfolded proteins within it. ER stress response has been shown to be activated in many types of cancer including leukemia. Its activation is thought to confer survival benefit to cancer cells and is related to resistance to anticancer therapies. However, whether ER stress response plays a role in the pathogenesis or affects the clinical and genetic presentations of MDS is still unknown.

Material and Methods

A total of 176 adult patients with newly diagnosed primary MDS at the National Taiwan University Hospital and 20 healthy marrow donors who had enough cryopreserved bone marrow cells and complete clinical and genetic information for analysis were enrolled. Diagnosis and classification of MDS were made according to the WHO 2008 classification. We used Affymetrix Human Transcriptome Array 2.0 array of RNA to profile the global gene expression. The probe intensity data (CEL files) and probe set summarization files (CHP files) were generated and analyzed by Affymetrix GeneChip Command Console (AGCC) and Transcriptome Analysis Console (TAC) software. We used the Gene Set Enrichment Analysis (GSEA) tool to comprehensively test the roles of the ER stress pathways in MDS.

Results

Among the 176 MDS patients, 42 had refractory cytopenia with unilineage dysplasia (RCUD, 23.9%), 13 refractory anemia with ring sideroblasts (RARS, 7.4%), 36 refractory cytopenia with multilineage dysplasia (RCMD, 20.5%), 9 refractory cytopenia with multilineage dysplasia and ringed sideroblasts (RCMD-RS, 5.1%), 32 refractory anemia with excess blasts-1 (RAEB1, 18.2%), and 44 refractory anemia with excess blasts-2 (RAEB2, 25.0%). Patients with each subtype of MDS had significantly different expression in more than 8000 probe sets of coding genes compared with healthy donors. In contrast, the differences were much less remarkable among the subgroups of MDS, implying a common theme of global gene expression profile among subtypes of MDS and an intrinsic genetic difference between MDS and normal controls. The differentially expressed probe sets between MDS patients and healthy donors were explored by database search and literature review. Among the perturbed biological pathways, we focused on the ER stress response for further study. GSEA revealed a significant dysregulation of ER stress-associated genes in MDS (GSEA enrichment score=0.58; P=0.035; Figure 1), indicating a crucial role of ER stress in the pathogenesis of MDS. We noted 17 genes which were up-regulated in MDS patients in our array data, including BAG5, ATG17, BECN1, CALR, CANX, TRAF2, EIF2S1, PKR, SIAH1, SIAH2, S1P, S2P, EDEM3, PDIA6, GRP94, AMPK, and XBP-1 . On the contrary, GADD34, which inhibits ER stress response, was down-regulated in the MDS patients.

By univariate analysis, 10 (BAG5, CALR, CANX, GRP94, EDEM3, PDIA6, SIAH1, S1P, S2P, and XBP1) of the 17 up-regulated genes in MDS patients were associated with shorter overall survival, and 7 (CALR, CANX, S2P, EDEM3, PDIA6, XBP1, and TRAF2)of them were associated with shorter time to leukemia transformation (all with P<0.05). We used the sum of the expression levels of the five genes with the strongest associations with shorter overall survival, BAG5, CALR, PDIA6, EDEM3, and XBP1 (Figure 2), to establish an "ER stress score". Higher ER stress score was associated with shorter overall survival (Hazard ratio = 2.208, P= 0.003) and shorter time to leukemia transformation (Hazard ratio = 3.176, P= 0.004) in multivariate analysis independent of other well known prognostic factors. (Figure 3A, 3B, and Table 1). The prognostic significance of the score was also confirmed in another 31 MDS patients. (P=0.046, Figure 3C)

Conclusion

Several important genes related to the ER stress response have higher expression levels in MDS patients compared with healthy donors and their up-regulations are associated with shorter overall survival and time to leukemia transformation. This finding indicated that deregulation of ER stress response can predict MDS prognosis and it may be involved in the pathogenesis of MDS.

Disclosures

Tang: Celgene International Sàrl: Research Funding. Yao: Celgene International Sàrl: Research Funding. Li: Celgene International Sàrl: Research Funding. Huang: Celgene International Sàrl: Research Funding. Chen: Celgene International Sàrl: Research Funding. Ko: Celgene International Sàrl: Research Funding. Hsu: Celgene International Sàrl: Research Funding. Wu: Celgene International Sàrl: Research Funding. Lin: Celgene International Sàrl: Research Funding. Tsai: Celgene International Sàrl: Research Funding. Tien: Celgene International Sàrl: Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution