A Phase II Add-on Study of Vorinostat (VOR) in Higher Risk Myelodysplastic Syndrome with Failure of Hypomethylating Agents (HMA): The GFM Azavor Study

Background: Azacitidine (AZA) is the current standard of care for patients treated for higher risk MDS, but 40-50% patients do not respond and most responders eventually relapse. Median survival after AZA failure is only 5 months and no standard of care is defined for this population. Preclinical studies and positive results of phase I-II trials support a synergistic effect of the histone deacetylase (HDAC) vorinostat (VOR) and AZA in terms of response, although no survival advantage of the combination has as yet been demonstrated. We hypothesized that adding VOR to AZA in patients with primary or secondary AZA resistance could rescue response and prolong survival.

Methods: inclusion criteria inGFM AZAVOR study (NCT 01748240) were: 1/IPSS int 2 or high risk MDS at the time of initiation of AZA 2/treatment with at least 6 cycles of AZA and either failure to achieve any response or loss of response (per IWG2006 criteria) 3/a maximum of 3 months between AZA failure and inclusion with no other treatment in between. Patients received VOR 300mg bid from day 3 to day 9 of each cycle. AZA was given at standard 75mg/m2/d day 1 to 7 or at the maximum previously tolerated dose in case of dose reduction. Patients were evaluated after 6 cycles and responding patients treated until progression. The trial used a two-stage design, and accrual was to be stopped if less than 3 responses were seen in the first 14 evaluable patients.

Results 21 patients were included between march 2013 and September 2014. Nineteen patients were treated (1 patient died and 1 progressed before treatment). Median age was 72 years. All pts had higher risk MDS and had received a median of 6 cycles of AZA before entering the trial. The median number of AZA+ VOR cycles administered was 3 (range: 1-12). No unexpected SAEs were seen, and the most common AEs were infection, thrombocytopenia, GI toxicities, and fatigue. After 6 cycles of treatment, only 2 patients (11%) achieved response (1 erythroid hematological improvement, 1 partial remission), , which, per protocol, triggered the stop of accrual. At last follow-up, 18 patients were off study and one patient was still on treatment. Nine patients stopped treatment because of progression (42%), 4 stopped treatment for lack of response (21%), 2 stopped treatment because of intolerance (11%), 1 patient stopped at his request (5%), and 1 patient died of complications of cytopenias while on treatment (5%). Median overall survival was 13 months.

Conclusion This is the first report of an add-on study in high risk MDS, a strategy that may be useful for the early evaluation of drugs for which synergy with AZA is expected. Our results show that the proposed regimen of AZA +VOR can be used safely. However, the observed response rate was not above the "background" response rate expected from AZA alone continuation in a comparable patient population, indicating that the addition of VOR cannot reverse resistance to AZA.