Long Term Follow up from the NCRI AML17 Trial of Attenuated Arsenic Trioxide and ATRA Therapy for Newly Diagnosed and Relapsed Acute Promyelocytic Leukaemia

Background: Two randomized studies have reported comparison of the "chemo-free" combination of ATO+ATRA with Anthracycline/ ATRA (AIDA) in APL. The GIMEMA-AMLSG-SAL trial, using a daily schedule improved both survival and relapse risk in patients with newly diagnosed, low or intermediate-risk APL. (Lo Coco et al., NEJM 2013; 369: 111-21). The NCRI AML17 trial using an attenuated dosing schedule in all risk groups resulted in a very low risk of relapse, significantly better EFS but no significant survival advantage

Aims: The NCRI AML17 Trial compared AIDA vs the chemo-free combination of ATO (in an attenuated schedule) + ATRA. Here we present long term survival results for randomized patients and for 24 patients who received the same chemo-free schedule of ATO + ATRA after relapsing from the AIDA arm of the trial.

Methods: From May 2009 to October 2013, 235 patients aged >16 who entered the AML17 trial with molecularly confirmed APL were randomized to either ATRA+ATO (8 week induction 0.3mg/kg d1-5 w1, 0.25mg/kgx2/w w2-8, followed by 4 consolidation courses of 0.3mg/kgx2 w1, 0.25mg/kgx2/ w2-4 (63 ATO doses) OR AIDA schedule: Idarubicin (Ida)12mg/m² d2,4,6,8 + ATRA to d60) (induction) then Ida 5mg/m² d1-4 + ATRA d1-15 (Course 2); Mitox. 10mg/m² d1-4 + ATRA d1-15 (course 3); Ida 12mg/m² d1 + ATRA d1-15 (Course 4). ATRA was 45mg/m²/d. Maintenance was not given. High risk patients could receive Gemtuzumab Ozogamicin (d1,6mg/m²). Another 70 patients were treated in AML17 with AIDA after closure of the randomization. 25/189 patients relapsed post AIDA and 24 were treated with a median of 4 cycles (range 1-5) of ATRA + ATO, 11/24 were later transplanted. Follow-up is complete to 1 January 2016.

Results: The median age was 47y (16-77); 57 had WBC>10x10⁹/L (27 AIDA, 30 ATRA+ATO) and 49 (24 AIDA, 25 ATRA+ATO) were >60y. The early results of the randomization for newly diagnosed patients have been reported (Burnett et al. Lancet Oncol. 2015, 16 (13):1295). 91% entered morphological CR with no significant difference in CR rate between the arms (Chemo-free 94%, Chemo 89%; OR 0.54 (0.21-1.34), p= 0.18). The OS at 4 years was 93% (Chemo-free) vs 89% (Chemo), HR 0.60 (0.26-1.42) p= 0.2, but EFS was significantly superior with ATRA+ATO (91% vs 70%, HR 0.35 (0.18-0.68) p=0.002).

With a longer median follow-up of 53.4 months 5-year survival is now 93% (chemo-free) v 87% (chemo) (HR 0.61 (0.27-1.35) p=0.2). A significant reduction in relapse (2% vs 16% at 5 years, HR 0.19 (0.09-0.45) p=0.0005) translates to continuing significant RFS benefit for the chemo-free approach (96% vs 82%; HR 0.30 (0.13-0.67) p=0.004). This is seen both in low risk (95% vs 86% HR 0.45 (0.17-1.20) p=0.11) and high risk disease (100% vs 69% HR 0.10 (0.02-0.46) p=0.003), p=0.11 for heterogeneity.

25 patients relapsed following AIDA therapy, of whom 1 died in frank relapse before treatment could be initiated and 24 (5 with concomitant CNS involvement) were treated with the attenuated ATO + ATRA schedule. Of these 16 were treated at molecular relapse, reflecting the value of centralised MRD monitoring as part of the trial protocol. All 24 patients achieved molecular CR post ATO +ATRA. Eleven patients were transplanted (8 autograft, 3 allograft) including 4 of the 5 patients with CNS disease. 3 patients have had a second molecular relapse after ATO + ATRA salvage (1 transplanted and 2 not transplanted). The 3-year overall survival post-relapse is 96% with the only other death occurring post-transplant after 37 months.

Summary/Conclusion: The combination of ATO + ATRA continues to show a very low risk of relapse irrespective of risk group resulting in significantly better RFS compared to AIDA and excellent survival but no survival benefit has emerged primarily because of effective salvage interventions for AIDA-treated patients with most patients treated at molecular relapse. Molecular monitoring for t(15;17) is therefore essential to optimize therapy with AIDA. For patients treated with frontline ATO+ATRA molecular monitoring is of questionable value once achievement of molecular CR has been documented, but molecular surveillance remains important in those with relapsed disease. The attenuated AML17 ATO dosing approach is effective both upfront and in patients relapsing post AIDA and these results question the role of transplantation as consolidation in patients achieving molecular CR2 with ATO + ATRA who do not have CNS disease at relapse.