Conditioning regimens prior to allo-HSCT: Updates from the 48th Annual Meeting of the EBMT (2022)

The standard myeloablative conditioning regimen (MAC) for patients with acute myeloid leukemia (AML) undergoing allogeneic hematopoietic stem cell transplantation (allo-HSCT) is busulfan (BUS) or total body irradiation (TBI) in combination with cyclophosphamide (Cy). More recently, fludarabine (Flu) has been investigated as an alternative to cyclophosphamide to lessen regimen toxicity. Treosulfan, an analogue to busulfan, has also been investigated as an alternative to improve efficacy.

At the 48th Annual Meeting of the European Society for Blood and Marrow Transplantation (EBMT), the safety and efficacy of conditioning combinations were evaluated for patients undergoing allo-HSCT, using data from the Acute Leukemia Working Party (ALWP) of the EBMT. We summarize results from four key talks below.

TBI plus fludarabine versus busulfan plus fludarabine¹

Ryszard Swoboda presented a retrospective study¹ comparing a combination of fludarabine with the standard myeloablative regimens of either TBI at 12 Gy (FluTBI12) or intravenous busulfan (FB4) before allo-HSCT (Figure 1).

Allo-HSCT, allogeneic hematopoietic stem cell transplantation; AML, acute myeloid leukemia; CR1, first complete remission; CR2, second complete remission; FB4, busulfan intravenous at a dose of 12.8 mg/kg (4 days) plus fludarabine; FluTBI12, fludarabine with total body irradiation at a dose of 12 Gy; HLA, human leukocyte antigen; MSD, matched sibling donor; UD, unrelated donor.
*Adapted from Swoboda et al.^1
†Patients were exactly matched for cytogenetic risk groups, disease status at allo-HSCT (CR1/CR2), donor type (MSD, UD), and stem cell source (peripheral blood/bone marrow).

Results

• Patient characteristics are summarized in Table 1; there were no significant differences apart from serological status with patients receiving FB4 more frequently positive for cytomegalovirus (p = 0.0004).

Table 1. Patient characteristics*

CR1, first complete remission; FB4, busulfan intravenous at a dose of 12.8 mg/kg (4 days) plus fludarabine; FluTBI12, fludarabine with total body irradiation at 12 Gy; HLA, human leukocyte antigen. *Adapted from Swoboda et al.1
Characteristic	FluTBI12 (n = 109)	FB4 (n = 213)
Median age, years (range)	41 (18.2–61.8)	41.4 (18.7–67.4)
Male/Female, %	61.5/38.5	62.4/37.6
Cytogenetic risk group, %
Standard	8.3	8
Intermediate	41.3	42.3
High	11.9	11.7
Unknown	38.5	38
In CR1 at transplantation, %	78	78.9
Median year of transplantation (range)	2017 (2009–2020)	2016 (2009–2020)
Donor type, %
HLA-identical sibling	32.4	27.7
10/10 HLA-matched unrelated	54.1	51.4
9/10 HLA-mismatched unrelated	13.5	20.8

• Survival outcomes are shown in Table 2. Notably, no significant differences were reported between the conditioning regimens for any of the efficacy and toxicity parameters.

Table 2. Efficacy and toxicity results of FluTBI12 versus FB4*

FB4, busulfan intravenous at a dose of 12.8 mg/kg (4 days) plus fludarabine; FluTBI12, fludarabine with total body irradiation at 12 Gy; GvHD, graft-versus-host disease. *Adapted from Swoboda et al.1
Parameter, %	FluTBI12 (n = 109)	FB4 (n = 213)	HR (95% CI)
Relapse incidence	19.2	29.4	1.55 (0.9–2.66)
Nonrelapse mortality	15.6	10.9	0.62 (0.34–1.15)
Leukemia-free survival	65.2	59.8	1.1 (0.73–1.67)
Overall survival	69.7	72.1	0.96 (0.61–1.52)
GvHD-free, relapse-free survival	48.5	48.5	0.95 (0.67–1.36)
Acute GvHD grade
2–4	17.8	24.4	1.49 (0.86–2.6)
3–4	7.5	6	0.83 (0.34–2.07)
Chronic GvHD	41.8	34.3	0.77 (0.52–1.15)
Extensive chronic GvHD	16.4	15.9	0.96 (0.52–1.78)

In summary, FluTBI12 and FB4 produced comparable efficacy and toxicity in adult patients receiving allo-HSCT. Swoboda et al.¹ highlighted that the regimen choice should be guided based on patient history, availability of TBI, and experience of the centers using MAC.

Fludarabine plus TBI versus cyclophosphamide plus TBI²

Sebastian Giebel presented results from a retrospective study² comparing the safety and efficacy of FluTBI12 versus cyclophosphamide plus TBI at 12 Gy (CyTBI12).

The study design is depicted in Figure 2.

allo-HSCT, allogeneic hematopoietic stem cell transplantation; AML, acute myeloid leukemia; CR1, first complete remission; CR2, second complete remission; CyTBI12, cyclophosphamide plus total body irradiation at 12 Gy; FluTBI12, fludarabine plus total body irradiation at 12 Gy; MSD, matched sibling donor; UD, unrelated donor.
*Adapted from Giebel et al.^2
†Patients were exactly matched for cytogenetic risk groups, disease status at allo-HSCT (CR1/CR2), donor type (MSD, UD), and stem cell source (peripheral blood/bone marrow).

Results

Patient characteristics are shown in Table 3. The only significant difference was found between the median year of transplantation (p < 0.0001).

Table 3. Patient characteristics*

CR1, first complete remission; CyTBI12, cyclophosphamide plus total body irradiation at 12 Gy; FluTBI12, fludarabine with total body irradiation at 12 Gy; HLA, human leukocyte antigen. *Adapted from Giebel et al.2
Characteristic	FluTBI12 (n = 109)	CyTBI12 (n = 109)
Median age, years (range)	38 (18–69)	40 (18–60)
Cytogenetic risk group, %
Standard	8	8
Intermediate	35	35
High	13	13
Unknown	44	44
In CR1 at transplantation, %	78	78
Median year of transplantation (range)	2016 (2009–2020)	2012 (2009–2020)
Donor type, %
HLA-identical sibling	37	32.5
10/10 HLA-matched unrelated	51	54
9/10 HLA-mismatched unrelated	11	14

There was no significant difference in leukemia-free survival (LFS), overall survival (OS), relapse incidence, nonrelapse mortality (NRM), chronic graft-versus-host disease (GvHD), or GvHD-free/relapse-free survival (GRFS) between conditioning regimens (Figure 3).

aGvHD, acute graft-versus-host disease; cGvHD, chronic graft-versus-host disease; CyTBI12, cyclophosphamide plus total body irradiation at 12 Gy; LFS, leukemia-free survival; FluTBI12, fludarabine plus total body irradiation at 12 Gy; OS, overall survival; NRM, nonrelapse mortality.
*Adapted from Giebel et al.²

There was a significant difference found in acute GvHD (aGvHD) incidence with patients receiving CyTBI12 experiencing higher rates of aGvHD (p = 0.005).

In summary, the data demonstrate that fludarabine is a valid substitute for cyclophosphamide in TBI containing conditioning regimens, with encouraging LFS and OS rates and, importantly, a lower incidence of Grade 2–4 aGvHD.

Treosulfan + fludarabine versus busulfan + fludarabine³

Treosulfan is an analogue of busulfan that has shown favorable outcomes when substituted for its counterpart in patients with AML/myelodysplastic syndromes (MDS) at mixed doses. However, conditioning with FB4 (fludarabine at 150 or 160 mg/m² and busulfan at 12.8 mg/kg) was superior to FT14 (fludarabine at 150 or 160 mg/m² and treosulfan at 42 g/m²) in younger patients (<55 years) with AML in complete remission. Eleni Gavriilaki presented results from a retrospective study³ comparing FB4 with FT14 in patients with relapsed/refractory AML. The study design is summarized in Figure 4.

Allo-HSCT, allogeneic hematopoietic stem cell transplant; FB4, fludarabine at 150 or 160 mg/m² and busulfan at 12.8 mg/kg; FT14, fludarabine at 150 or 160 mg/m² and treosulfan at 42 g/m².
*Adapted from Gavriilaki et al.³

Results

Patient characteristics are shown in Table 4.

Table 4. Patient characteristics*

AML, acute myeloid leukemia; FB4, fludarabine-busulfan; FT14, fludarabine-treosulfan; HSCT, hematopoietic stem cell transplant; MSD, matched sibling donor; UD, unrelated donor. *Adapted from Gavriilaki et al.3
Characteristic	FT14 (n = 113)	FB4 (n = 233)
Median age, years (range)	58.1 (21.2–75.9)	52.4 (18.5–70.7)
Median follow-up, months	32.93	24.91
Adverse cytogenetics, %	23	16.7
Time diagnosis to HSCT, median in months (range)	6.9 (0.5–147.2)	5.9 (0.3–66.2)
Cell source, %
Bone marrow	3.5	9
Peripheral blood	96.5	91
Diagnosis, %
De novo	69	78.5
Secondary AML	31	21.5
Type of donor, %
MSD	31	44.6
UD	69	55.4
Median dose of fludarabine, mg/m2	150	160

• The follow-up period and time of diagnosis to HSCT had a trend toward significant difference between the two cohorts.
• There was a significant difference in patient age (p = 0.0006), diagnosis (p = 0.054), type of donor (p = 0.015), and median dose of fludarabine between the groups (p < 0.0001).

Survival outcomes

• There was no significant difference in the cumulative incidence of aGvHD (FT14, 8.5% vs FB4, 9%), chronic GvHD (FT14, 10.7% vs FB4, 10.9%), and NRM (FT14, 20.8% vs FB4, 22.6%).
• In a multivariate analysis, treosulfan was demonstrated to be associated with improved LFS, relapse incidence, OS, and GRFS (Table 5).

Table 5. Multivariate analysis*

FB4, fludarabine-busulfan; FT14, fludarabine-treosulfan. *Adapted from Gavriilaki et al.3
Factor	Relapse		Leukemia-free survival		Overall survival		GvHD-free, relapse-free survival
	HR (95% CI)	p value	HR (95% CI)	p value	HR (95% CI)	p value	HR (95% CI)	p value
FB4 vs FT14	1.83 (1.16–1.9)	0.01	1.52 (1.1–2.1)	0.012	1.65 (1.13–2.41)	0.009	1.48 (1.04–2.12)	0.03
Age (per 10 years)	1.04 (0.89–1.21)	0.61	1.13 (1–1.27)	0.049	1.17 (1.02–1.33)	0.023	—	—

In line with the previous subset analysis in active AML, treosulfan was associated with superior OS compared with FB4. Notably, the dose of treosulfan used in this study was higher than currently recommended (FT10). Further retrospective studies will be needed to confirm the optimal conditioning regimen.

Treosulfan- versus thiotepa-based conditioning regimen for haploidentical transplant⁴

A treosulfan-based regimen was previously shown to reduce NRM in patients with AML undergoing allo-HSCT from matched sibling donors (MSD) and unrelated donors (UD). Francesco Saraceni presented results from a retrospective study⁴ investigating whether this improvement was observed for patients undergoing transplant from haploidentical donors, which does not have a standard conditioning regimen. Treosulfan-based (Treo) or thiotepa-busulfan-fludarabine (TBF) conditioning were compared as the preparatory regimens (Figure 5).

CR1, first complete remission; CR2, second complete remission; GRFS, GvHD-free, relapse-free survival; GvHD, graft-versus-host disease; haplo-HSCT, haploidentical hematopoietic stem cell transplantation; MAC, myeloablative conditioning; RIC, reduced intensity conditioning; TBF, thiotepa-busulfan-fludarabine; Treo, treosulfan.
*Adapted from Saraceni et al.⁴

Results

Patient characteristics for patients included in the matched-pair analysis are shown in Table 6.

Table 6. Patient characteristics*

CR1, first complete remission; RIC, reduced intensity conditioning; TBF, thiotepa-busulfan-fludarabine; Treo, treosulfan. *Adapted from Saraceni et al.4
Characteristic	Treo (n = 142)	TBF (n = 142)
Median age, years (range)	58 (18.1–75.7)	58 (21.2–72.2)
Median year of transplant (range)	2019 (2012–2020)	2019 (2014–2020)
CR1 at transplant, %	82	82
RIC, %	47	47
Stem cell source, %
Bone marrow	13	13
Peripheral blood	87	87
Cytogenetic risk, adverse, %	16	16

Primary and secondary endpoints

• Engraftment was comparable between the two conditioning regimens (95% for both).
• The 2-year primary and secondary endpoint outcomes are summarized in Figure 6.

aGvHD, acute GvHD; cGvHD, chronic GvHD; GRFS, GvHD-free, relapse-free survival; GvHD, graft-versus-host disease; LFS, leukemia-free survival; NRM, nonrelapse mortality; OS, overall survival; TBF, thiotepa-busulfan-fludarabine; Treo, treosulfan.
*Adapted from Saraceni et al.⁴

• There was no significant difference in any of the 2-year survival outcomes between the conditioning regimens.
  • When stratifying by conditioning intensity, there was again no significant difference in NRM, relapse incidence, LFS, and OS in patients receiving myeloablative treatment nor reduced intensity conditioning (RIC).
• Finally, the causes of death are shown in Table 7.
  • As expected, occurrence of original disease, GvHD, and infection were the most common causes of death in both cohorts.
  • Notably, no deaths from veno-occlusive disease were reported in the Treo cohort.

Table 7. Causes of death*

GvHD, graft-versus-host disease; SOS, sinusoidal obstruction syndrome; TBF, thiotepa-busulfan-fludarabine; Treo, treosulfan; VOD, veno-occlusive disease. *Table from Saraceni et al.4
Cause of death, %	Treo (n = 142)	TBF (n = 142)
Original disease	41	31
GvHD	10	31
Infection	31	12
SOS/VOD	0	6
Cardiac toxicity	3	3
Other	16	17

In summary, treosulfan and TBF based conditioning regimens produced similar outcomes for patients undergoing haploidentical allo-HSCT. Treosulfan-based conditioning may be a valid alternative in AML, although further studies are needed.

Conclusion

Data from the above studies presented at the 48th Annual Meeting of the EBMT, firstly demonstrate the validity of substituting busulfan with fludarabine in standard conditioning regimens in a relapsed/refractory and complete remission setting; secondly, they demonstrate that the substitution of busulfan with treosulfan led to improved survival outcomes for patients in complete remission when combined with fludarabine. Finally, treosulfan-containing regimens are a valid alternative for patients undergoing haploidentical transplant with similar survival outcomes compared with TBF-based conditioning.