Impact of MRD status and number of remissions on HSCT outcomes

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is one of the few curative options for patients diagnosed with acute myeloid leukemia (AML).¹ Transplantation can be performed after first relapse or in the first complete remission (CR) period if a patient’s relapse risk is deemed high enough to outweigh the risk of non-relapse mortality. It is also a viable option for patients who achieve a second CR period after relapse.¹ To further refine pre-treatment assessment, a large proportion of patients who undergo HSCT in the first CR are intermediate or adverse genetic risk according to the European LeukemiaNet (ELN) 2017 risk classification. Measurable residual disease (MRD) assessment has increasingly become part of the evaluation process prior to treatment and has allowed clinicians to assess a patient’s response to different therapy options with greater precision.¹

Jentzsch et al. recently published a study in Blood Advances that evaluated the currently used genetic risk stratification models and MRD status in predicting patient outcomes when performing HSCT in the first or second remission periods.¹ We summarize the findings below.

Study design

A total of 580 patients diagnosed with AML who had their first allo-HSCT between January 1999 and November 2020 were retrospectively evaluated. Cytogenetic analysis was performed at diagnosis, including the mutation status of CEBPA, NRM1, and internal tandem duplication (ITD) in FLT3 (FLT3-ITD). Assessment of MRD was performed up to 28 days before HSCT using remission samples. The median follow-up time was 3.9 years.

Characteristics of patients receiving HSCT in first and second remissions

The median age at HSCT was 59.6 years, with 72% of patients having received nonmyeloablative or reduced-intensity conditioning, and 28% myeloablative conditioning. Patients undergoing HSCT in the second CR or CR with incomplete peripheral cell count recovery (CRi) period (n = 120), compared with patients transplanted in first CR/CRi (n = 460), had an increased likelihood of the following:

• De novo AML (p = 0.002)
• NPM1 mutation (p = 0.03)
• Normal karyotype (p < 0.001)
• FLT3-ITD mutation (p < 0.001)

Patients who underwent HSCT in the second CR/CRi also had elevated levels of hemoglobin (p = 0.01), white blood cell counts (p = 0.006), and bone marrow blasts (p = 0.03) when compared with baseline levels. On the other hand, patients in this group had a lower likelihood of having a monosomal or complex karyotype. Furthermore, fewer cycles of chemotherapy were administered prior to HSCT to patients in their second CR/CRi, who also more often received grafts from human leukocyte antigen (HLA)-mismatched donors compared with patients in their first CR/CRi receiving other types of grafts. Full patient characteristics are shown in Table 1.

Table 1. Patient characteristics*

AML, acute myeloid leukemia; BM, bone marrow; CR, complete remission; CRi, complete remission with incomplete peripheral cell count recovery; ELN, European LeukemiaNet; HLA, human leukocyte antigen; HSCT, hematopoietic stem cell transplantation; ITD, internal tandem duplication; MRD, measurable residual disease; WBC white blood cell. *Adapted from Jentzsch, et al.1
	All patients (N = 580)	HSCT in first CR/CRi (n = 460)	HSCT in second CR/CRi (n = 120)	p value
Characteristics at diagnosis
Sex, n (%)				0.68
Male	308	242 (53)	66 (55)
Female	272	218 (47)	54 (45)
Disease origin, n (%)				0.002
Secondary/treatment-related AML	195	169 (37)	26 (22)
De novo AML	385	291 (63)	94 (78)
Hemoglobin, g/dL, median (range)	8.9 (3.2–15.7)	8.7 (3.2–15.7)	9.5 (4.3–14.9)	0.01
WBC, ×109/L, median (range)	5.9 (0.1–385)	5.4 (0.1–385)	17.1 (0.6–366)	0.006
BM blasts, %, median (range)	51 (0–95)	50 (0–95)	61 (22–95)	0.03
ELN 2017 risk group, n (%)				0.15
Favorable	112	89 (25)	23 (34)
Intermediate	147	123 (34)	24 (36)
Adverse	168	148 (41)	20 (30)
Normal karyotype, n (%)				<0.001
Absent	288	253 (59)	35 (34)
Present	247	178 (41)	69 (66)
Complex karyotype, n (%)				<0.001
Absent	445	346 (84)	99 (100)
Present	67	67 (16)	0 (0)
Monosomal karyotype, n (%)				<0.001
Absent	459	359 (87)	100 (98)
Present	58	56 (13)	2 (2)
NPM1, n (%)				0.03
Wild type	328	280 (77)	48 (65)
Mutated	108	82 (23)	26 (35)
FLT3-ITD, n (%)				<0.001
Wild type	345	298 (81)	47 (63)
Mutated	96	68 (19)	28 (37)
CEBPAbiallelic, n (%)				0.08
Wild type	344	292 (99)	52 (95)
Mutated	7	4 (1)	3 (5)
Characteristics at HSCT
Median age at HSCT, years (range)	59.6 (16.3–76.8)	60.3 (16.3–76.0)	59.6 (19.2–76.8)	0.43
Treatment cycles prior to HSCT, n (%)				<0.001
1	157	72 (16)	85 (71)
2	311	280 (61)	31 (26)
≥3	111	107 (24)	4 (3)
Blood count regeneration at HSCT, n (%)				0.41
CR	484	387 (84)	97 (81)
CRi	96	73 (16)	23 (19)
MRD status at HSCT, n (%)				0.10
MRD−	167	133 (58)	34 (47)
MRD+	133	95 (42)	38 (53)
Donor type, n (%)				0.02
Matched related	124	109 (24)	15 (13)
Unrelated, HLA matched	329	258 (56)	71 (59)
HLA mismatched	112	80 (17)	32 (27)
Haploidentical	14	12 (3)	2 (2)

 

Patient outcomes

By number of remissions

Patients transplanted in the second CR/CRi period recorded a significantly higher cumulative incidence of relapse (CIR; p < 0.001) and shorter relapse-free survival (RFS; p = 0.002) compared with patients receiving HSCT in the first CR/CRi. However, overall survival (OS) and non-relapse mortality were similar in both groups. Multivariable analysis revealed the number of remissions remained a significant prognostic factor for CIR and RFS (both p < 0.001), although not for OS, following adjustment for ELN 2017 genetic risk and the MRD status at HSCT. Moreover, for patients undergoing HSCT in the first CR/CRi, more intensive conditioning regimens were associated with better OS (p < 0.001). In contrast, patients in the second CR/CRi who remained in remission for longer than 1 year before first relapse demonstrated lower CIR (p < 0.001) and longer RFS (p = 0.001).

By MRD status

In both groups of patients, MRD positivity prior to HSCT was a significant prognostic factor for higher CIR (both groups, p < 0.001) and shorter RFS (first CR/CRi, p = 0.002; second CR/CRi, p = 0.04,). However, OS was similar across patients who were MRD positive or MRD negative who underwent HSCT in first or second CR/CRi.

By ELN risk

The distribution of ELN genetic risk groups was similar between patients transplanted in first and second remission. However, in all risk groups, patients who received HSCT in second remission had a higher CIR compared with patients transplanted in first remission. The frequency of ASXL1, RUNX1, and TP53 mutations were similar across each of the risk groups.

Patients classified as intermediate risk

• Patients who underwent HSCT in the second CR/CRi had an increased likelihood of de novo AML compared with patients who underwent HSCT in first CR/CRi (p = 0.04).
• Patients who were transplanted in second remission had a shorter OS (p = 0.05) compared with patients who were transplanted in first remission.
• MRD positivity influenced CIR and RFS; both of which were comparable between remission groups.
• OS was similar between patients who underwent HSCT in first or second remission (p = 0.30 vs p = 0.20).

Patients classified as adverse risk

• Patients who underwent HSCT in the second CR/CRi period were less likely to have a monosomal or complex karyotype (both p < 0.001), but more likely to have a normal karyotype (p < 0.001) and FLT3-ITD mutations (p = 0.01).
• Survival rates (OS and RFS) were similar in both the first and second remission groups.

Patients classified as favorable risk

• MRD positivity had a larger impact on CIR in those undergoing HSCT during the second CR/CRi compared with the first (CIR, p = 0.003 vs p = 0.21; RFS, p = 0.02 vs p = 0.10; OS, p = 0.30 vs p = 0.07, respectively). MRD positivity influenced CIR and RFS; both of which were comparable between remission groups. However, OS again did not differ significantly between patients with intermediate risk who underwent HSCT in first or second remission (p = 0.20 vs p = 0.30).

Conclusion

Transplantation during second remission was associated with a higher risk of relapse than in the first remission, regardless of ELN genetic risk. The authors also noted that, within the group of patients with ELN intermediate genetic risk, MRD positivity had a great prognostic value, with patients who were MRD positive and underwent transplantation during the second remission having the poorest outcomes. They therefore recommended avoiding transplantation in these patients.

Limitations to this study included its retrospective nature and the lack of data availability, in particular MRD status during chemotherapy and post-HSCT.

Although MRD status at HSCT still remains highly relevant, therapies targeting specific molecular aberrations should be considered prior- and post-HSCT, especially in the group of patients who are MRD positive and in second remission, in order to improve outcomes.