Fludarabine/treosulfan for allo-HCT in patients with MDS

While allogeneic hematopoietic cell transplantation (allo-HCT) is potentially curative for patients with high-risk myelodysplastic syndromes (MDS), most patients with MDS are older and have comorbidities, rendering them ineligible for standard myeloablative conditioning (MAC) due to high nonrelapse mortality (NRM). Reduced intensity conditioning (RIC) regimens have allowed for transplantation in older patients, although evidence demonstrating their association with reduced NRM comes from retrospective studies that employed purposive sampling, and evidence supporting the use of one regimen over another is inconclusive.

Treosulfan is a bifunctional alkylating agent with durable myeloablative and immunosuppressive characteristics allowing consistent hematopoietic cell engraftment with a lower risk of graft-versus-host disease (GvHD), making it a suitable component of the conditioning regimen given, prior to allo-HCT. The combination of fludarabine and treosulfan (FT) has demonstrated lower risk of GvHD and NRM in patients with acute myeloid leukemia (AML). Shimoni et al.¹ retrospectively compared outcomes following FT versus other RIC and MAC regimens in patients with MDS, which was recently published in the British Journal of Haematology and summarized herein.

Study design

This was a retrospective cohort study using patient data from the Chronic Malignancies Working Party (CMWP) database of the European Society for Blood and Marrow Transplantation (EBMT). Eligibility criteria for patients included de novo or secondary MDS and had undergone an allo-HCT from a human leukocyte antigen (HLA)-matched sibling or an unrelated donor between 2000‒2016. Patients were divided into three groups (FT, RIC, and MAC) based on dose intensity as reported by the participating center.

The primary outcome was overall survival (OS), defined as the day of allo-HCT until death of any cause or last follow-up. Secondary outcomes included:

• NRM: Death without prior disease recurrence
• Relapse mortality
• Acute GvHD (aGvHD) and chronic GvHD (cGvHD)

Results

Baseline characteristics

A total of 1,722 patients with MDS, either from HLA-matched siblings (n = 747) or unrelated donors (n = 975), were included. The median age was 56 years (range, 18‒76) and 60% of the patients were male (Table 1).

• Patients in the FT and RIC group were older compared with those in the MAC group (p < 0.001), but other disease characteristics were well balanced between the three groups (Table 1).
• The most common regimens included high-dose busulfan/cyclophosphamide (37%) and fludarabine/high-dose busulfan (32%) for MAC and fludarabine with reduced-dose busulfan (82%) for RIC. FT regimens included total doses of 30 g/m² (10%), 36 g/m² (17%), 42 g/m² (44%) or other/missing (29%).

Table 1. Baseline characteristics*

allo-HCT, allogeneic hematopoietic cell transplantation; AML, acute myeloid leukemia; BM, bone marrow; CMV neg/neg, donor and recipient are CMV negative; CR, complete response; F → M, female donor to male recipient; FT, fludarabine and treosulfan; MAC, myeloablative conditioning; PB, peripheral blood; PR, partial response; RA, refractory anemia; RAEB, refractory anemia with excess blasts; RIC, reduced-intensity conditioning; SC, stem cell. *Adapted from Shimoni et al.1 †Values in bold are statistically significant. ‡Calculated from known data, data missing (<10%).
Characteristic, % (unless otherwise stated)	Total (n = 1,722)	FT (n = 367)	RIC (n = 687)	MAC (n = 668)	p value†
Median age, years (range)	56 (18–76)	59 (18–76)	59 (19–75)	51 (18–74)	<0.001
Disease classification at diagnosis
RA	35	39	35	34	0.46
RAEB1	22	21	23	21
RAEB2	30	25	30	33
Missing	13	14	13	12
AML transformation before allo-HCT	23	20	22	25	0.11
Status at allo-HCT
Untreated	34	45	26	36	<0.001
CR/PR	29	26	36	25
Refractory	36	29	39	39
Cytogenetics
Good	17	16	18	15	0.49
Intermediate	14	13	15	15
Poor	14	14	14	12
Secondary‡	21	23	20	20	0.38
Time from diagnosis to allo-HCT, months (range)	9.7 (0.1–370)	9.4 (0.3–214)	10.6 (0.2–207)	9.0 (0.1–370)	0.007
Donor
Sibling	43	41	39	49	0.001
Unrelated	57	59	61	51
SC source
PB	88	91	94	78	<0.001
BM	12	9	6	22
In vivo T-cell depletion	65	65	79	50	<0.001
F → M	20	21	25	18	0.51
CMV neg/neg‡	25	22	29	23	0.02
Median follow-up, months (range)	64 (1–171)	75 (1–169)	60 (2–171)	64 (1–137)	<0.001
Median year of allo-HCT, range	2009 (2000–2016)	2009 (2002–2016)	2010 (2002–2016)	2009 (2000–2016)	<0.001

OS and NRM

• The median follow-up amongst surviving patients (n = 850) was 64 months and 75, 60, and 64 months after FT, RIC, and MAC, respectively (p < 0.001).
• The 5-year OS rate across all groups was 44%, and 50%, 43%, and 43% after FT, RIC, and MAC, respectively (p = 0.03). Factors predicting OS using univariate analysis are presented in Table 2.
• The major causes of death included disease recurrence (n = 384), GvHD (n = 165), infection (n = 161), organ toxicities (n = 73), and second malignancies (n = 16).
• Improved OS rates were consistent in the multivariate analysis (hazard ratio [HR], 0.72; p = 0.01) for FT.
  • However, lower survival rates were predicted for advanced age (HR, 1.41; p < 0.001), transformation to AML before allo-HCT (HR, 1.43; p < 0.001), secondary MDS (HR, 1.19; p = 0.05), poor cytogenetics (HR, 1.34; p = 0.004), and earlier years of allo-HCT (≤2009, HR, 1.28; p = 0.001).
  • No significant differences were observed in the 5-year OS after conditioning with different dose intensity of FT (p = 0.68).
  • OS improved in all three groups and the improvement associated with FT was mostly observed in patients ≥55 years of age (p = 0.001) compared with younger patients (p = 0.52).
  • OS with FT was also more prominent with active disease at allo-HCT and in patients with poor cytogenetics.
• The 5-year NRM rate was 30% across all groups, and 30%, 27%, and 34% after FT, RIC, and MAC, respectively (p = 0.008) (Table 2).
  • Higher NRM was predicted by MAC (HR, 1.44; p = 0.001), advanced age (HR, 1.51; p < 0.001), allo-HCT from an unrelated donor (HR, 1.21; p = 0.06) and earlier years of allo-HCT (≤2009, HR, 1.21; p = 0.05) in the multivariate analysis.

Table 2. Univariate analysis of factors predicting the 5-year survival and NRM after allo-HCT*

AML, acute myeloid leukemia; allo-HCT, allogeneic hematopoietic transplantation; CMV, cytomegalovirus; CR, complete response; FT, fludarabine and treosulfan; MAC, myeloablative conditioning; NRM, nonrelapse mortality; OS, overall survival; PR, partial response; RA, refractory anemia; RAEB, refractory anemia with excess blasts; RIC, reduced-intensity conditioning. *Adapted from Shimoni et al.1 †Values in bold are statistically significant.
Factor	Alive (n)	OS Cumulative incidence	p value†	NRM (n)	NRM Cumulative incidence	p value†
All patients	850	44	—	485	30	—
Conditioning regimen
FT	194	50	0.03	104	30	0.008
RIC	331	43	170	27
MAC	325	43	211	34
Age
≥55 years	421	39	<0.001	292	33	0.03
<55 years	429	51	193	27
Gender
Male	499	44	0.22	312	32	0.03
Female	351	45	173	27
Disease classification at diagnosis
RA	338	51	0.001	183	33	0.09
RAEB1	155	36	114	33
RAEB2	246	43	126	26
Missing	111	44	62	32
AML transformation before allo-HCT
Yes	158	35	<0.001	108	29	0.04
No	692	47	377	31
Status at allo-HCT
Untreated	310	48	0.006	183	34	0.05
CR/PR	257	45	125	27
Refractory	283	40	177	30
Cytogenetics
Good	147	51	<0.001	87	31	0.36
Intermediate	117	45	65	27
Poor	84	33	60	26
Missing	502	45	273	32
Secondary
Yes	137	36	0.04	88	30	0.81
No	614	46	351	31
Time from diagnosis to early allo-HCT
Early	428	44	0.95	226	28	0.07
Donor
Sibling	382	45	0.26	186	27	0.01
Unrelated	468	44	299	33
CMV neg/neg
Yes	215	49	0.07	115	30	0.82
No	585	43	342	30
Year of allo-HCT
≤2009	404	41	0.006	268	32	0.14
>2009	446	48	217	29

Relapse mortality

• FT, RIC, and MAC showed relapse rates of 25%, 38%, and 25%, respectively (p < 0.001).
• FT (HR, 0.55; p <0.001) and MAC (HR, 0.61; p < 0.001) predicted a lower relapse rate compared with RIC in the multivariate analysis.
  • Higher relapse rate was predicted by advanced age (HR, 1.22; p = 0.06), transformation to AML before allo-HCT (HR, 1.93; p < 0.001), poor cytogenetics (HR, 1.87; p < 0.001), treatment-refractory disease (HR, 1.59; p < 0.001) and earlier years of allo-HCT (≤2009, HR, 1.42; p = 0.001).

GvHD

• A total of 455 patients developed Grade 2–4 aGvHD with a cumulative incidence of 30% at 1 year and 633 patients developed cGvHD with a cumulative incidence of 43% at 2 years after allo-HCT.
  • The cumulative incidences for aGvHD were 26%, 24%, and 34% after FT, RIC, and MAC, respectively (p = 0.001), and 46%, 41%, and 43% after FT, RIC, and MAC, respectively (p = 0.56) for cGvHD.
  • 350 patients of the 633 developed extensive cGvHD with cumulative incidences of 31%, 25%, and 30% after FT, RIC, and MAC, respectively (p = 0.18).
• Deaths related to GvHD occurred in 9%, 10%, and 18% of patients in the FT, RIC, and MAC groups, respectively (p = 0.007).

Conclusion

This large cohort study in patients with MDS demonstrated that the OS with FT was comparable to the OS seen with RIC and MAC, and FT was associated with low relapse rates, like MAC, and low NRM rates, like RIC. The current study had certain limitations in that it was a retrospective study with limited data and no data on comorbidities in the patients. Treosulfan was approved by the European Medicines Agency (EMA) in June 2019 as a conditioning treatment prior to hematopoietic progenitor cell transplantation, and FT may become the preferred regimen for allo-HCT in patients with MDS.