The Association of Hif-1α Expression with Primary Resistance to Azacytidine in Patients with High Risk MDS — Preliminary Data of the Hellenic MDS Study Group

Introduction : Azacitidine (AZA) is the standard of care for patients with high-risk myelodysplastic syndromes (MDS) (IPSS intermediate-2 or high) and patients with AML not candidate for treatment with intensive chemotherapy. Nevertheless, approximately 40% of patients fail to respond, whereas even responders will inevitably relapse. Although some clinical parameters and genetic mutations have been reported to correlate weakly with favorable AZA response, the exact mechanisms underlying primary azacitidine resistance are largely unknown and there is no approach to circumvent it. Several reports have demonstrated the significance of hypoxia in the regulation of both physiological and malignant hematopoiesis. In MDS, Hif-1α has been proposed to participate in the development of the disease, while its up-regulated expression has been correlated with poor prognosis.

Aims : The objective of the current study was to elucidate the expression level and clinical significance of Hif-1α in both AZA responders and non-responders patients with high risk MDS, before and after treatment.

Methods : Bone marrow samples from 44 de novo MDS patients, 3 CMML patients and 4 AML (40 M/11 F) with a median age of 78 (60-89) and 10 healthy donors were collected. MDS patients were classified according to WHO as RCMD (7/43), RAEB I (9/43) and RAEB II (28/43) and according to IPSS as low (3/43), intermediate-1 (4/43), intermediate-2 (20/43) and high (17/43). All patients received AZA treatment at the dose of 75mg/m² x7 days SC. Low and inter-1 MDS patients received azacitidine as off label therapy. Bone marrow samples from 10 MDS [RCMD (1/15), RAEB I (3/15), RAEB II (6/15)], 1 CMML and 2 AML of the aforementioned patients were also collected 6 months after AZA therapy [median age of 80 (62-89), 11 M/2 F]. BM mononuclear cells were isolated from patients before and 6-months after treatment using the Ficoll-paque method, followed by RNA extraction using TRIzol reagent and cDNA preparation using Superscript II reverse transcriptase. Hif-1α expression was estimated by real time PCR TaqMan gene expression assay, using the appropriate primers and probes. Relative gene expression was calculated by comparative threshold cycle (ΔΔCt) method. β-actin was used as a housekeeping gene. The Mann-Whitney U test was performed for statistical analysis of the results.

Results : Out of the 51 patients examined, 25 responded to AZA treatment (including CR, PR and HI) while 26 failed to respond. Using Rt-PCR we found that the ΔΔCt ratio of Hif-1α/β-Actin median expression for control samples was 0,82, for responders to AZA treatment 1,36, while for non-responders 0,83. Interestingly, in MDS patients, a trend towards increasing Hif-1α expression, although not statistically significant, was observed across IPSS risk groups, with Hif-1α median expression varying from 1,09 in low risk group to 1,12 in intermediate-1, 1,31 in intermediate-2 and 1,63 in high risk group of patients.Moreover, 10 responders and 3 non-responders were also examined after treatment. The ΔΔCt ratio of Hif-1α/β-Actin median expression after 6-months therapy presented an increasing trend for both responders [from 1,427 to 1,474 (3,29% increase)] and non-responders [from 0,425 to 0,598 (40,7% increase)], with more than a 10-fold difference in the up-regulation of Hif-1α after therapy between the two groups.

Conclusions : Our data indicate that Hif-1α expression gradually increases across IPSS groups, suggesting differences among MDS patients' IPSS risk groups in terms of hypoxia signalling, and an association of Hif-1α with a more aggressive disease. Our data also suggest that patients with increased pretreatment expression of Hif-1α, which has been previously associated with poor prognosis, seem to better respond to AZA therapy. Most importantly, our findings indicate that AZA non-responders seem to present with up-regulated Hif-1α expression after therapy, which might be associated with their failure to respond to azacitidine. These early stage observations imply that the molecular mechanisms underlying patients' response to AZA treatment might be associated with hypoxia and Hif-1α expression and although they need to be further confirmed in a larger number of MDS patients in order to establish a statistical significance, they appear very encouraging for better understanding primary resistance to AZA, Hif-1α and hypoxia- wise, and for improving current MDS therapy.