IS PI-PLC 1 EXPRESSION A RELIABLE PREDICTOR OF RESPONSE IN LOW RISK MDS PATIENTS TREATED WITH LOW-DOSE AZACITIDINE? RESULTS OF A PROSPECTIVE MULTICENTRIC STUDY

Background. The use of hypomethylating agents significantly modified the therapeutic approach to MDS patients, primarily in high-risk MDS patients. In low-risk MDSs the use of AZA hypomethylating agent is less understood. Epigenetic regulation of Phosphoinositide-Phospholipase C (PI-PLC) beta1 promoter and key molecules involved in the nuclear inositide signalling pathways seems to play an important role for the activity of AZA demethylating therapy. Aims. We prospectively evaluated the efficacy and safety of AZA low-dose in Low or Int-1 risk MDS patients who were symptomatic and/or unresponsive to previous treatments. In the same study we evaluated the molecular effects of AZA on PI-PLCbeta1 promoter methylation, in order to investigate a possibly correlation with the response to the drug. Methods. AZA was administered at a dose of 75 mg/mq/daily s.c for 5 consecutive days every 28 days for a total of 8 cycles. Final response was checked at the end of the 8th course. PI-PLC beta1 gene expression was evaluated on peripheral blood samples from patients at baseline, and monthly until the 8th cycle of AZA administration. Results. Between September 2008 and February 2010, 32 MDS patients with IPSS risk Low- or Int-1 were enrolled into the study. Most patients had a normal karyotype (63%) by metaphase cytogenetics, were BRC transfusion-dependent (81%), receiving a median of 4 units/mo, and were previously unresponsive to treatment including ESAs (69%). Twenty-six patients (81%) completed the treatment plan (8 cycles). The Overall Response Rate after the 8th cycle was 58% (15/26 pts) whereas 42% of patients maintained a stable disease; no patient progressed towards a high-risk MDS or AML. Five (19%) patients reached a complete remission whereas 10 (38%) achieved an hematological improvement. Transfusion independent was achieved in 8/26 patients (31%). The median duration of the response was 10 months; five patients maintain their response, that is CR in 2 cases (+24 and +30 months) and HI-E in 3 cases (+14, +25,+26 months) without any treatment or supportive therapy. No clinical or hematologic factor showed a correlation with the response. PI-PLCbeta1 gene expression was quantified in MDS patients at baseline and during AZA treatment. Results were calculated as a percentage ratio of PI-PLCbeta1 expression during AZA treatment compared to baseline. All but one patients (14/15) who achieved an hematologic response during treatment with AZA showed a statistically significant increase in PI-PLCbeta1 mRNA expression (P<0,001). In all patients the increase of PI-PLCbeta1 levels anticipated the clinical response obtained at the 8th cycle. Conclusion. The current results of our study showed that Aza low-dose schedule may be a safe and effective treatment for low risk MDS pts and may induce durable responses in some cases. The positive and significant correlation between the hematologic response and the PI-PLC 1 expression indicate that PI-PLC 1 could be a realiable marker of response to AZA in LowRisk MDS patients.