Expert OpinionSecondary acute myeloid leukemia (sAML) comprises a heterogenous group of diseases. It is most often derived from conditions such as myelodysplastic syndrome (MDS) and myeloproliferative neoplasms (MPNs), though there is a significant population with therapy-related AML (t-AML) related to prior treatment for a hematological lymphoid malignancy. Patients with t-AML typically have poor outcomes due to the pretreatment received, and/or an older age at diagnosis. Allogeneic hematopoietic stem cell transplant (allo-HSCT) is potentially curative for these patients, though relapse remains an issue. The allo-HSCT outcomes of patients who developed a t-AML after treatment of a B-cell malignancy have not been well studied, and the potential impact of choice of conditioning regimens on post-transplant outcomes has not been discerned.
Katie S. Gatwood, Vanderbilt University Medical Center, Nashville, US, and colleagues conducted a multicenter, retrospective study using the Acute Leukemia Working Party (ALWP) of the European Society for Blood and Marrow Transplantation (EBMT) registry. They aimed to evaluate the impact of myeloablative (MAC) versus reduced intensity conditioning (RIC) regimen on allo-HSCT outcomes in patients with t-AML following a lymphoid malignancy.
Patient characteristics
The authors analyzed data of adult patients with sAML (n= 549) who had previously been treated for a lymphoid malignancy and had received their first allo-HSCT between 2000–2016. The prior malignancies included acute lymphoblastic leukemia (ALL), chronic lymphocytic leukemia (CLL), lymphoma and plasma cell dyscrasias. Patient characteristics at baseline are shown in Table 1.
Table 1. Baseline characteristics by conditioning regimen
|
Conditioning regimen |
Myeloablative (MAC) |
Reduced intensity (RIC) |
|
N |
258 (47%) |
291 (53%) |
|
Median age at transplant (years) |
47.8 |
55.9 |
|
Median time from diagnosis to transplant (months) |
4.7 |
4.7 |
|
Previous diagnosis: · ALL · CLL · Lymphoma · Multiple myeloma (MM) |
· 25 (9.7%) · 29 (11.2%) · 194 (75.2%) · 10 (3.9%) |
· 15 (5.2%) · 36 (12.4%) · 219 (75.3%) · 21 (7.2%) |
|
AML cytogenetics · Favorable · Intermediate · Adverse · Missing |
· 8 (3.1%) · 73 (28.3%) · 73 (28.3%) · 104 (40.3%) |
· 9 (3.1%) · 96 (33%) · 68 (23.4%) · 118 (40.6%) |
|
Disease status at allo-HSCT · Active · Complete remission (CR) 1 · CR2 |
· 80 (31%) · 161 (62.4%) · 17 (6.6%) |
· 92 (31.6%) · 181 (62.2%) · 18 (6.2%) |
|
Donor type · Matched sibling (MSD) · Unrelated (URD) · Haploidentical · Cord blood transplant |
· 93 (36%) · 126 (48.8%) · 25 (9.7%) · 14 (5.4%) |
· 90 (30.9%) · 174 (59.8%) · 15 (5.2%) · 12 (4.1%) |
Efficacy
The efficacy results are shown in Table 2, by total cohort, and by conditioning regimen. Patients receiving RIC had a lower risk of non-relapse mortality (NRM), improved leukemia-free survival (LFS), and superior overall survival (OS, Table 3) in multivariate analysis. The choice of conditioning regimen did not significantly impact relapse incidence though.
Table 2. Efficacy results for the total cohort and by conditioning regimen and multivariate analysis
|
|
Total cohort, % 95% CI |
MAC, % 95% CI |
RIC, % 95% CI |
p value |
|
Two-year LFS |
31.7 27.5–35.9 |
27.9 22–33.8 |
35.1 29.2–41 |
0.055 |
|
Two-year RI |
39.1 34.8–43.4 |
38.6 32.3–44.9 |
39.6 33.7–45.5 |
0.82 |
|
Two-year OS |
37.4 33–41.8 |
34.2 27.9–40.5 |
40.2 34.1–46.3 |
0.074 |
|
Two-year NRM |
28.9 25–33 |
33.3 27.4–39.4 |
25.3 20.2–30.6 |
0.04 |
|
Two-year graft-versus-host disease (GvHD)-free relapse-free survival (GRFS) |
22.8 19–26.6 |
19.8 14.5–25.1 |
25.5 20.1–30.9 |
0.148 |
Table 3. Factors significantly associated with outcomes in multivariate analysis
|
|
HR |
95% CI |
p value |
|
LFS |
|
|
|
|
Conditioning regimen (MAC vs RIC) |
0.67 |
0.52–0.85 |
0.001 |
|
Prior autologous HSCT (yes) |
1.3 |
1.01–1.67 |
0.04 |
|
Cytogenetics (adverse vs favorable) |
3.15 |
1.35–7.37 |
0.008 |
|
Active disease at transplant vs CR1 |
1.68 |
1.31–2.56 |
< 0.001 |
|
CBT vs MSD |
0.9 |
0.51–1.61 |
0.04 |
|
Donor (female to male) |
1.35 |
1.03–1.77 |
0.028 |
|
OS |
|
|
|
|
Conditioning regimen (MAC vs RIC) |
0.69 |
0.53–0.89 |
0.004 |
|
Cytogenetics (intermediate vs favorable) |
3.56 |
1.01–11.76 |
0.037 |
|
Cytogenetics (adverse vs favorable) |
6.61 |
2–21.85 |
0.002 |
|
Active disease at transplant vs CR1 |
1.57 |
1.2–2.04 |
0.001 |
|
RI |
|
|
|
|
Active disease at transplant vs CR1 |
2.25 |
1.62–3.13 |
< 0.001 |
|
Karnofsky performance status (KPS, < 80% vs ≥ 80%) |
0.46 |
0.29–0.72 |
0.001 |
|
NRM |
|
|
|
|
Conditioning (MAC vs RIC) |
0.58 |
0.4–0.83 |
0.003 |
|
Cytogenetics (adverse vs favorable) |
4.64 |
1.05–20.54 |
0.043 |
|
KPS (< 80% vs ≥ 80%) |
0.4 |
0.24–0.66 |
< 0.001 |
|
Donor (female to male) |
1.521 |
1.02–2.27 |
0.04 |
|
GFRS |
|
|
|
|
Conditioning regimen (MAC vs RIC) |
0.79 |
0.62–0.99 |
0.045 |
|
Cytogenetics (adverse vs favorable) |
2.82 |
1.29–6.19 |
0.02 |
|
Active disease at transplant vs CR1 |
1.66 |
1.3–2.13 |
< 0.001 |
|
KPS (< 80% vs ≥ 80%) |
0.47 |
0.34–0.66 |
< 0.001 |
|
Donor (female to male) |
1.32 |
1.02–1.71 |
0.037 |
Rates of graft-versus-host disease (GvHD) are shown in Table 4. Unrelated donor transplant was associated with higher rate of grade II–IV aGvHD (HR: 1.67, 1.06–2.63, p= 0.027) and a KPS of > 80% was associated with lower rate of grade III–IV aGvHD (HR: 0.45, 0.2–1, p= 0.049).
Table 4. Cumulative incidence of GvHD for total cohort and by conditioning regimen
|
|
Total cohort, %, 95% CI |
MAC, %, 95% CI |
RIC, %, 95% CI |
p value |
|
Grade II–IV acute GvHD (aGvHD) at day 100 post-transplant |
30.6 26.6–34.6 |
32.7 26.9–38.7 |
28.6 23.3–34.1 |
0.2 |
|
Grade III–IV aGvHD at day 100 post-transplant |
13.7 10.9–16.8 |
15.37 11.1–20.1 |
12.3 8.7–16.5 |
0.26 |
|
Chronic GvHD (cGvHD) at two-years |
27 23–31.1 |
23.7 18.3–29.6 |
30.1 24–36.4 |
0.16 |
|
Extensive cGvHD at two-years |
12.8 9.9–16 |
11.4 7.5–16.1 |
14 10–18.6 |
0.38 |
Other variables associated with poor outcomes in the total cohort were; older age, adverse cytogenetics and active disease at time of transplant.
Deaths
In total, 171 patients receiving MAC and 174 patients receiving RIC died. The main causes (> 10% of patients) were:
Given as MAC vs RIC
- Relapse: 40.8% vs 45.9%
- Infection: 22.6% vs 19.8%
- GvHD: 20.1% vs 16.9%
Conclusion
This study has some limitations, including the retrospective nature, potential for selection bias regarding intensity of conditioning, a lack of molecular characterization of AML subtype, a lack of analysis by disease risk, and missing cytogenetic data in around 40% of patients.
In summary, this analysis supports the use of allo-HSCT with RIC for patients with sAML following a prior lymphoid malignancy since patients treated with RIC regimens had a lower risk of NRM and improved LFS, OS and GFRS.
