High risk MDS (Myelodysplastic Syndrome) patients will be treated with SGI-110 after Allogeneic Stem Cell Transplantation in the hypothesis that SGI-110 maintenance given early after HSCT can prevent relapse without increasing non-relapse mortality translating in an improved disease-free survival.

Allogeneic stem cell transplant (HSCT) is the only curative treatment in patients with intermediate-2 and high risk patients (according to classical IPSS) but approximately 30% of patients relapse and 30% of patients die from non-relapse complications after HSCT. Risk factors for post-transplant outcome are related to the patient itself (age, comorbidity), the disease risk and transplant characteristics (higher relapse in patients receiving a reduced intensity conditioning regimen and in those receiving a T-cell depleted graft).

The risk of post-transplant relapse is however particularly high (> 60-70%) in patients with very poor cytogenetics according to the revised IPSS, patients with monosomal karyotype, and patients with TP53 mutation. Taking into account that these patients also have non-relapse mortality, expected post-transplant survival is very poor, less than 15% and more often 10%. It has been reported that 30 to 35% of those high risk patients respond to hypomethylating agents (HMA) but they have very short remission duration, less than 5 months in median. A recent study reported a prospective, uncontrolled trial including 84 patients with MDS, AML patients receiving Decitabine (DAC). The authors highlight that the response was better in patients with unfavorable cytogenetics and that TP53 clones was cleared after treatment. The cytogenetics was no more a prognostic factor suggesting that DAC has improved survival especially in high-risk patients who had an 11.6-month median survival. This study suggests that DAC is particularly encouraging in high-risk patients. Guadecitabine (SGI-110) is a novel hypomethylating dinucleotide of Decitabine and deoxyguanosine resistant to degradation by cytidine deaminase. Safety and tolerance of SGI-110 in patients with MDS has been reported and this drug is now considered as a potential treatment in patients with AML or MDS. The concept of post-transplant maintenance therapy with one HMA in AML and MDS has been studies by several teams and there are 2 prospective trials exploring escalating dose in 5-azacytidine (AZA) and DAC. a group has reported that DAC maintenance was safe and that there was no dose limiting toxicities with the highest dose tested at 15 mg/m2/day 5 days every 6 weeks from day 50 post-transplant. A phase II trial, the RICAZA study, has tested a maintenance HMA early after transplant from day 40. 37/51 pre-screened patients could receive AZA and only 10% experienced complications. Two-year OS was 50%. HMA induces leukemic differentiation and re-expression of tumor or viral associated genes that had been epigenetically silenced. At high dose, cell die from apoptosis triggered by DNA synthesis arrest and at low doses, cells survive but change their gene expression to favor differentiation. Several groups have demonstrated effects of HMA on T cell-mediated anti-tumor activity which might promote graft-versus-leukemia or MDS effect. In another hand, HMA have been reported to increase the frequency of Tregs after HSCT and lower acute GVHD which might lower non-relapse mortality. Regarding GVHD, acute GVHD should be prevented due to the higher non-relapse mortality associated with acute GVHD. In contrast, several studies have highlighted the benefit of chronic GVHD on relapse risk justifying immunotherapy, donor lymphocyte infusion (DLI) later after HSCT to prevent relapse. The therapeutic strategy combining pre-emptive HMA in combination with DLI has been tested in a prospective study, the RELAZA trial, based on CD34 chimerism.

Taken together, these studies provide a rationale for the early administration of DMA, ie: SGI 110, associated with late DLI after HSCT for AML and MDS. The hypothesis is that SGI 110 maintenance given early after HSCT can prevent relapse without increasing non-relapse mortality translating in an improved disease-free survival. This hypothesis will be tested in the higher risk patients, especially those with TP53 for whom relapse risk is higher than 50%.

Inclusion Criteria:

• Patients aged from 18 to 70 years
• MDS or AML with unfavorable genetics defines as follow:
• 4 cytogenetic abnormalities or more or
• 3 cytogenetic abnormalities and TP53 or
• 3 cytogenetic abnormalities and monosomal karyotype or
• Mutations involving EVI1
• Marrow blast < 20% for and non-proliferative disease
• AML patients should have received chemotherapy before transplant
• A donor is available (HLA matched or mismatched)
• Contraception in women < 50 years and for men at least the first six months after transplant and 3 months after the last dose of guadecitabine"

Exclusion Criteria:

• Karnofsky less than 70%
• Cancer in less than 2 years before inclusion or cancer not in remission the last 2 years before inclusion (except in situ cancer or baso cellular cancer)
• Cardiac failure with EF < 50%
• Creatininemia level > 150 µmol/L
• Liver enzyme > 3 N
• Conjugated bilirubinemia > 25 µmol/L
• MDS occurring in a patients with Fanconi anemia or congenital dyskeratosis
• Proliferative disease in patients no in remission: WBC> 15 G/L or use of continuous cytotoxic to maintain WBC < 15G/L
• Proliferative AML: hyperleucocytosis > 15 G/L, blast count higher than 10% or lower than 10% for less than 6 weeks
• No contraception
• Pregnant women or breastfeeding women