Azacitidine in Older Patients with Acute Myeloid Leukemia (AML). Results from the Expanded International E-Alma Series (E-ALMA+) According to the MRC Risk Index Score

Background

Treatment of older (>60y) patients (pts) with AML remains challenging. Azacitidine (AZA) showed a trend towards a better overall survival (OS) in older pts with AML vs. conventional care (Dombret H, Blood. 2015) in the AZA-AML-001 trial. However, comparison between AZA and intensive chemotherapy (IC) was limited by the low number of subjects randomized to AZA or IC. The Medical Research Council (MRC) and the Leukemia Research Foundation (LRF) developed and validated a 1y survival risk index score in a large series of older AML pts (N=2843; AML11 and AML14 trials) treated with IC and non-intensive chemotherapy (NI), identifying 3 groups with different risk estimates (Wheatley K, B J Haematol, 2009).

Aim

To assess the impact of AZA as front-line therapy in older AML pts unfit for IC from different national registries who served as the basis for the European ALMA score (Spain, Italy, France and Austria), now expanded with the cases from PETHEMA AML Registry, and stratified by the MRC/LRF risk score, and to display a tentative nonrandomized comparison between IC, NI and AZA within those risk strata.

Methods

Retrospective analysis of 456 AML pts who received front-line AZA within drug label, off-label or compassionate named patient programs. Cytogenetic, age, white blood count (WBC), PS and AML type categorized pts as good, standard and poor-risk groups, as stated by Wheatley's score.

Results

Baseline characteristics of pts and distribution across MRC/LRF risk categories are shown in Table 1. After accounting for the above 5 parameters, 323 pts (71%) were categorized as poor-risk (as compares to 31% and 37%, respectively, from the MRC/LRF AML11 and AML14 trials). The poor-risk subset included pts with age >75y (61%), adverse karyotype (45%), secondary AML (62%), WBC >10e9/L (32%) and PS³3 (6%). After 28 months median follow-up (95% CI: 25-31), median OS for the whole series was 9.7 months (95% CI: 8.5-11). Median OS for the good, standard and poor-risk groups was 15.1, 12.4 and 8.5 months, respectively (p=0.001. Fig 1), while 1y OS estimates were 64%, 51% and 36%, respectively. A comparison of these estimates with those of MRC/LRF trials are also shown in Table 1. As regard 1y OS, AZA therapy seems to confer a clinical benefit similar to that of IC, and superior to NI chemotherapy.

Conclusion

Although retrospective and non-matched, this is the largest comparison between AZA and IC in older AML pts. AZA therapy seems to confer a clinical benefit similar to IC, that in the poor-risk subset (n=323) might be even more relevant.

Table 1.  Baseline characteristics and 1y OS (N=456)

Parameter			N (%)
Age (median)			75 (56-93)
PS			0-1: 345 (76) 2: 88 (19.5) >2: 19 (4.5)
WCB (>109/L)			103 (25)
Therapy-related/AHD1 AML			220 (48)
Cytogenetic according to MRC2			Inter: 245 (53.7) Adv: 150 (32.9) NA: 61 (13.4)
Risk categories			Good: 24 (5.3) Standard: 109 (23.9) Poor: 323 (70.8)
1 year Overall Survival (%)
	AML113	AML14I	AML14NI	AML14NIA	E-ALMA+
n	1071	1137	275	NA	456
Therapy	IC	IC	HU + LDAC +/- ATRA	LDAC +/- Mylotarg	AZA
MRC/LRF Risk group
Good	53	60	25	36	64
Standard	43	48	33	42	51
Poor	16	30	10	14	36

¹Antecedent hematologic disorder

²Cytogenetic: Good: t(8;21), inv(16); Adverse: -5, -7, del(5q), abn(3q), complex (5 or more). Intermediate: Normal karyotype and all other abnormal karyotypes. NA: Not available.

³AML11 trial: 2 induction courses (cytarabine and anthracycline, ± thioguanine or etoposide). If CR: One or 3 consolidation courses ±I FN maintenance.

⁴AML14 included 2 schedules: Intensive (AML14I: 2 induction courses with daunorubicin and cytarabine. If CR: One consolidation course) and non-intensive (AML14NI: Hydroxicarbamide with LDAC, ±ATRA or AML14NIA: LDAC vs LDAC plus Mylotarg)

Figure 1. Survival probability