Reduced-intensity transplantation in older patients with acute myeloblastic leukemia: NCRI AML16 trial

Treatment of older patients with acute myeloid leukemia (AML) remains a challenge as these patients are more vulnerable and often have more aggressive malignancies than younger patients. The incidence of AML increases with age, with more than half of patients diagnosed when aged >65 years. Current treatments for older patients have not shown the improvements in survival that have been achieved in younger patients. A benefit of reduced-intensity conditioning (RIC) transplantation over chemotherapy has been reported in patients aged 40–59 years. Thus, with the superior potential of RIC regimens, this therapy has been increasingly applied to older patients with AML.

The National Cancer Research Institute (NCRI) AML16 trial (NCT00454480; EUDRACT 2005-002847-14) is a randomized, multi-arms phase II/III trial comparing different combination chemotherapy regimens and RIC transplantation in older patients (>60 years) with AML and high-risk myelodysplastic syndromes (MDS). Early outcomes showed that induction chemotherapy with gemtuzumab ozogamicin (GO) reduced relapse risk and improved survival in older patients. The latest results, comparing RIC transplantation with chemotherapy in older patients, were recently published in Haematologica by Russell, et al.¹ Below, we summarize the key findings.

Methods and patients

A total of 932 patients who were in remission after Course 1 and without favorable-risk cytogenetics were eligible to undergo RIC transplantation from either a matched sibling or matched unrelated donor (MUD). Of the 144 RIC transplants performed, 52 were from matched sibling donors and 92 were from a MUD. The other 788 patients received chemotherapy alone. Overall, 83% of patients had intermediate-risk cytogenetics and 17% of patients had adverse-risk cytogenetics. The median time from complete remission to transplant was 110 days. The median follow-up for survival from remission was 60 months.

The study design and patient characteristics are detailed in Figure 1 and Table 1, respectively.

AML, acute myeloid leukemia; CR, complete remission; CRi, incomplete CR; FLAMSA, fludarabine, amsacrine, and cytarabine; GO, gemtuzumab ozogamicin; HLA, human leukocyte antigen; MDS, myelodysplastic syndromes; MUD, matched unrelated donor; PR, partial remission; RIC, reduced-intensity conditioning; TBI, total body irradiation.
*Adapted from Russell, et al.¹

Table 1. Baseline characteristics*

HCT-CI, Hematopoietic Cell Transplantation Comorbidity Index; ITD, internal tandem duplication; MDS, myelodysplastic syndromes; Q1, first quartile; Q3, third quartile; WHO, World Health Organization; WT, wild type. *Adapted from Russell, et al.1 †All tests are the ꭓ2-test (unless otherwise stated). ‡Wilcoxon rank-sum test. §Mantel-Haenszel test for trend.
Characteristic, % (unless otherwise stated)	Overall (n = 932)	No allograft (n = 788)	Allograft (n = 144)	p value†
Median age, years (Q1–Q3)	65 (63–68)	66 (63–68)	64 (62–66)
60–65 years	51	48	69	<0.001
66–70 years	49	52	31	<0.001‡
Sex
Male	58	57	65	0.062
Female	42	43	35
WHO performance status
0	62	60	76	<0.001§
1	33	35	23
2	3	3	1
3	2	2	1
Diagnosis
De novo	74	76	65	0.012
Secondary	14	14	17
High-risk MDS	12	10	18
Cytogenetics
Intermediate	83	83	84	0.868
Adverse	17	17	16
Unknown, n	204	171	33
Risk group
Good	49	48	58	0.013§
Standard	29	29	27
Poor	22	23	15
HCT-CI score
0	51	50	50	0.016§
1 or 2	34	34	34
≥3	15	16	9
FLT3-ITD
WT	82	81	90	0.105
Mutant	18	19	10
Fail, n	9	6	3
Unknown, n	506	422	84
NPM1
WT	73	72	83	0.087
Mutant	27	28	17
Fail, n	14	11	3
Unknown, n	535	447	88
Donor type
Sibling	—	—	36	—
MUD	—	—	64

Results

Outcomes after RIC transplants

• In the 144 transplant recipients, the 5-year overall survival (OS) after remission was 37%; this did not differ significantly between those with sibling (44%) or unrelated (34%) donors (p = 0.200) (Table 2).
• The 5-year non-relapse mortality did not differ significantly between siblings (28%) and MUD (36%) transplant recipients (p = 0.318).
• When stratified by Wheatley risk group, there was no significant difference in OS between sibling and MUD transplants in terms of good (51% vs 42%), standard (49% vs 31%), or poor (0% vs 13%) risk groups (p = 0.7).
• Of note, the incidence of severe, absent, mild, and moderate chronic graft-versus-host disease at 12 months was only 12%, 38%, 26%, and 24%, respectively.

Table 2. Survival estimates*

CI, cumulative incidence; CR1, first complete remission; MUD, matched unrelated donor; OS, overall survival. *Adapted from Russell, et al.1 †The methodology used to compare transplant vs “no transplant” was: all patients entering CR start in the “no transplant” group, with no patients in the transplant group; when patients were transplanted, they would change to the allograft group at the time of transplantation. ‡Relapse as a competing risk. §Death as a competing risk.
Outcome, %	Transplant			No transplant† (n = 932)
	All transplanted patients (n = 144)	Sibling donor (n = 52)	MUD (n = 92)
OS after CR1	37	44	34	20
Relapse-free survival	32	42	31	13
CI of death in Remission 1‡	34	28	36	10
CI of relapse§	34	30	33	77
Survival by Wheatley risk group
Good	45	51	42	26
Standard	36	49	31	21
Poor	12	0	13	7

Comparison of RIC transplantation with chemotherapy

• The OS was significantly greater with the RIC transplantation compared with the chemotherapy-only group (37% vs 20%; hazard ratio [HR], 0.67; p < 0.001) (Table 2).
• Relapse-free survival was significantly improved by transplantation (32% vs 13%; HR, 0.56; p < 0.001).
• The 5-year cumulative incidence of death in remission was 34% in the RIC group versus 10% in the chemotherapy-only group (HR, 5.02; p < 0.001, with relapse analyzed as a competing risk).
• The 5-year relapse risk was greatly reduced by transplantation (34%) versus chemotherapy alone (77%) (HR, 0.30; p < 0.001, with death analyzed as a competing risk).
• When stratified by Wheatley risk groups, there was consistent benefit for RIC transplantation across good (45% vs 26%), standard (36% vs 21%), and poor (12% vs 7%) risk groups.
• Furthermore, a survival benefit associated with RIC was observed in patients with FLT3 internal tandem duplication, or NPM1 mutations.
• In addition, when stratified by post-course 1 flow minimal residual disease (MRD) status, including MRD-negative, MRD-positive, MRD unknown, and no complete remission (CR)/incomplete CR (CRi) post-course 1, benefits were again observed for patients who underwent transplantation (HR, 0.68; p < 0.001).

Conclusion

Here, we summarized the most recent data from the NCRI AML16 trial. Russell, et al. conducted an in-depth comparison between RIC transplantation and chemotherapy in older patients with AML (aged 60–70 years) in remission and without favorable disease. The results showed overall improvements in all transplant patients compared with chemotherapy alone, including improved 5-year OS, reduced 5-year relapse risk, and consistent benefits across risk groups with RIC, suggesting that RIC transplantation is a promising option for older patients with AML lacking favorable-risk cytogenetics. Furthermore, Russell, et al. concluded that their findings point to the need for more precise risk stratification, such as detailed genomic analysis, to select older patients, particularly in intermediate-risk patients with gene mutations typical of secondary AML, for transplantation.