Is Arsenic Trioxide (ATO) Required in the Treatment of High Risk Newly Diagnosed APL? Analysis of a Randomized Trial (APL2006) By the French Belgian Swiss APL Group

Background:In standard risk APL, ATRA+ATO combinations (without CT) are at least as effective as classical ATRA + anthracycline based chemotherapy (CT) while being less myelosuppressive (Lo Coco, NEJM 2014, Burnett, Lancet Oncol, 2015). In high risk APL (WBC> 10G/l), it is still unclear if CT can be avoided or greatly reduced, but addition of ATO to ATRA + CT reduces relapses (Powell, Blood 2010). In a randomized trial (APL2006 trial) in high-risk APL patients (pts) who received ATRA + CT induction treatment, we evaluated the addition of ATO to CT during consolidation.

Methods: Between 2006 and 2015, newly diagnosed APL pts <70 years with WBC> 10 G/L, after an induction of ATRA 45mg/m2/d until CR with Idarubicin (Ida) 12 mg/m²/dx3 and AraC 200mg/m2/dx7, were randomized for consolidation between CT and CT+ATO. The CT group (standard group) received a first consolidation with Ida 12 mg/m²/dx3 and AraC 200mg/m2/dx7, a second consolidation with Ida 9 mg/m²/dx3 and AraC 1g/m2/12h x4d, and 2-year maintenance with intermittent ATRA and continuous 6 MP + MTX. The CT+ATO group received the same treatment except that ATO 0.15 mg/Kg/d d1 -25 was added during both consolidation courses. After a first interim analysis in Sept 2010, based on 81 pts, AraC was deleted from consolidation cycles of the CT+ ATO group. The primary endpoint was EFS from CR achievement.

Results: 211 pts <70 years with WBC> 10 G/L were included (after the exclusion of 8 diagnostic errors) in 58 centers. 95.7% achieved CR, 3.3% had early death and 1% resistant leukemia. 193 pts were randomized for consolidation, 97 in the CT and 96 the CT+ ATO groups. Pre-treatment characteristics were well balanced between the 2 groups. 7 pts (3 CT vs 4 CT+ATO) had relapsed (5-year cumulative incidence of relapse (CIR) of 2.5% vs 3.9%; p= 0.39) and 9 pts had died in CR :7 (7.8%), 2 (5.1%), 0 (0%) in the CT, CT (with AraC) + ATO, CT (without AraC) + ATO groups respectively (p=0.04). Causes of death in CR were bleeding (n=5), infection (n=2), previous cancer relapse (n=2). One patient in the CT+ATO arm developed AML/MDS.

5-year OS was 93% vs 94% (p=0.56) and 5-year EFS was 89% vs 93% (p=0.47) in the CT and CT+ATO groups, respectively. Omission of AraC (after the amendment) in the CT+ATO group did not increase CIR (5 year CIR 5.3% with and 3.3% without AraC, p=0.57). In the CT, CT (with AraC) + ATO, CT (without AraC) + ATO groups respectively, median time to ANC>1 G/L after consolidation 2 was 22, 25 and 19 days (p<0.0001), and median time to platelets>50G/l was 24, 26 and 20 days (p<0.0001). Similarly, median duration of hospitalization after the first and the second consolidation courses were 33, 34, 29 d (p<0.0001) and 31, 32, 28 d (p=0.0005), respectively.

Conclusion: Very high CR rates and very few relapses are now obtained in high risk APL on a large multicenter basis using classical ATRA and CT (including AraC) for induction and consolidation, and maintenance treatment, but at the expense of 5 to 7% deaths in CR. Addition of ATO to this regimen did not further reduce relapses, and added some myelosuppression if CT contained AraC. However, if ATO was added and AraC omitted from consolidation cycles, relapses were not increased, while myelosuppression and deaths in CR were reduced. ATO therefore appears useful in high risk APL. We are currently comparing, in the international APOLLO trial, the "classical" ATRA-CT approach and a prolonged ATO-ATRA regimen with only 2 days of Ida during induction treatment, hoping to further reduce myelosuppression and possibly relapses.