A Randomised Assessment of Vosaroxin Monotherapy and Vosaroxin Combined with Low Dose Ara-C Versus Low Dose Ara-C in Older Patients with Acute Myeloid Leukaemia

The older patient with AML represents an important subgroup who are often not considered fit for traditional “3+7” type chemotherapy. Although demethylation agents and low dose Ara-c (LDAC) are superior to best supportive care, other options have so far failed to improve survival which is typically 3-4 months. Much of the problem could be attributed to chemoresistance of the leukaemic clone.

Vosaroxin is a first in class anti-cancer quinolone which by intercalating into DNA has topoisomerase II inhibitory properties and can by-pass resistance mediated by P-glycoprotein or P53. This results in less free radical production, which has been associated with anthracycline toxicity. In preliminary studies in relapsed/refractory patients >70 yrs, remission rates of up to 30% were observed, with a low 30 day mortality of <10%. In preclinical testing we documented synergy with Ara-C. It was therefore considered to be a candidate treatment in our “Pick a Winner” trial design either alone or combined with LDAC.

Two parallel studies. Study 1 (02/2012-11/2013) 104 untreated patients with AML or high risk MDS (>10% marrow blasts) equally randomised between Vosaroxin and LDAC. Study 2 (06/2012-08/2013) 104 patients equally randomised between Vosaroxin + LDAC vs LDAC. Treatment plan was for 4 courses of allocated therapy with the option to extend for patient benefit.

Study 1: Median age 75 (60-89); 8/104 had MDS, 31/104 had secondary AML; the risk categories were 3 favourable, 37 intermediate and 64 poor. Study 2 Median age 75(60-91); 12/104 had MDS, 27/104 had secondary AML; 1 fav, 41 int and 62 poor. There were no significant differences between randomised arms. There was no cardiac/renal co-morbidity restriction to entry in either study. The median number of courses delivered (investigational vs LDAC) Study 1(1.0 vs 2.5); Study 2(1.0 vs 2.0). Vosaroxin was 75mg/m² IV infusion on days 1 and 4. LDAC as 20mg twice a day SC days 1-10. Toxicities NCICTC v.3.

RESULTS:

Study 1:28% had a complete marrow response (CR 15%; CRi 13%), with no difference between LDAC (CR 16%; CRi 14%) and Vosaroxin (CR 15%; Cri 11%). The 60 day mortality was greater in the Vosaroxin arm (38% vs 20%; HR 2.16 (1.05-4.43) p=0.04), with increased oral and GI toxicity, more resource use, antibiotics days and hospitalisation. Of those who did not respond, survival on the Vosaroxin arm was worse (1.9 vs 4.9mo: HR 1.68(1.04-2.71), p=0.03). In the responders the OS from response was better on LDAC (17.5 vs 6.9 mo, p=0.04). There was a trend for a better RFS in the LDAC arm (30% vs 8%). Because of the increased early mortality and inferior RFS the OS at 12 mo on vosaroxin was worse than LDAC (12% vs 31%; HR 1.94(1.26-3.00), p=0.003) with a median OS of 3.2 vs 9.0 mo respectively. Cause of death (Vosa vs LDAC: resistant 21 vs 23; infection 10 vs 8; recurrent disease 6 vs 5; cardiac 3 vs 0; pulmonary 1 vs 1; haemorrhage 1 vs 0; renal 0 vs 1; multiple 4 vs 3; other/unknown 2 vs 1).

Study 2:36% response (CR 22%; CRi 14%), no significant difference between LDAC (CR 20%; CRi 14%) and LDAC+ Vosaroxin (CR 25%; CRi 13%). The day 60 mortality was greater on the combo arm (36% vs 18% HR 2.12 (1.01-4.45), p=0.05). Of the non-responders those on the combo arm had worse survival (1.5 vs 4.1 mo; HR 1.92 (1.1-3.31), p=0.02). Survival of responders was not significantly different (median not reached vs 10.1 mo). No relapsed patient survived beyond 9 mo, although survival was better in the LDAC compared to the combo arm (8.8 vs 2.7 mo, p=0.05). The inferior outcome for patients on the combo arm who did not achieve CR or relapsed resulted in non-significantly worse OS (9.6 vs 3.1 mo HR 1.30 (0.81-2.07), p=0.3). Cause of death (Vosa+LDAC vs LDAC: resistant 16 vs 19; infection 9 vs 10; infection+haemorrhage 1 vs 0; recurrent disease 4 vs 4; cardiac 2 vs 0 other cancer 1 vs 0; renal 1 vs 0; other/unk 3 vs 1). In exploratory subgroup analysis we found no evidence of any subgroup that showed a survival benefit. According to the rules of Pick a Winner there had to be evidence of a superior remission rate; while this was passed the increased 60-day mortality and non-significantly worse survival (p=0.10 at time of DMEC meeting) led to a recommendation for closure.

Conclusion: In spite of the attributes of Vosaroxin, we failed to find benefit in unselected patients classified as unfit for intensive chemotherapy, either as monotherapy, or in combination with LDAC.

Acknowledgments: We are grateful to Sunesis for supporting this Investigator Initiated Study.