Allogeneic hematopoietic stem cell transplantation in patients older than 65 years with acute myeloid leukemia and myelodysplastic syndrome: a 15-year experience

Allogeneic stem cell transplantation (allo-HCT) is the only curative option for intermediate and high-risk adult acute myeloid leukemias (AML) and myelodysplastic syndromes (MDS). Despite the median age of occurrence of these diseases (>65y) [1], the majority of patients older than 65y had historically been excluded from this potentially curative option for both changes in tumor biology (conferring treatment resistance) and patient character- istics (decreasing allo-HCT tolerance) that translated into lower survival rates. Understanding which older patients are likely to bene ﬁ t from allo-HCT versus low-intensity therapies or supportive care is critical. We here report our 15-year experience of allo-HCT in patients diagnosed with AML or MDS and older than 65y with remarkable results in terms of survival, transplant related mortality, relapse and graft-versus -host disase (GvHD) incidence. Between January 2005 2019, 90 consecutive adult patients aged >65y received an allo-HCT San Raffaele Milan. Patients' and disease characteristics are described in Supplementary Data 1. Patients' median age 68.29y 76.53). Donors ’ median age 38y (18 – 74): 64y (55 – 74) for sibling, 39y (20 – 58) for haplo, 30y – 50) for matched unrelated donors (MUD) and 32y (18 – 52) for mismatched unrelated donors (MMUD). Conditioning regimens and GvHD prophylaxis are described in Supplementary Data and GvHD without increasing NRM. Multicentric and larger studies with similar platforms are needed to con ﬁ rm our results.

Allogeneic stem cell transplantation (allo-HCT) is the only curative option for intermediate and high-risk adult acute myeloid leukemias (AML) and myelodysplastic syndromes (MDS). Despite the median age of occurrence of these diseases (>65y) [1], the majority of patients older than 65y had historically been excluded from this potentially curative option for both changes in tumor biology (conferring treatment resistance) and patient characteristics (decreasing allo-HCT tolerance) that translated into lower survival rates. Understanding which older patients are likely to benefit from allo-HCT versus low-intensity therapies or supportive care is critical. We here report our 15-year experience of allo-HCT in patients diagnosed with AML or MDS and older than 65y with remarkable results in terms of survival, transplant related mortality, relapse and graft-versus -host disase (GvHD) incidence.
The novelties of our experience compared to previous and contemporary literature on patients allografted older than 65y are several. First, the majority of our patients received a treosulfanbased conditioning regimen and a rapamycine-based GvHD prophylaxis. The myeloablative properties of the treosulfan/ fludarabine conditioning regimen have already been demonstrated [2][3][4] and our data confirms prompt engraftment, with fulldonor hematopoietic chimerism, as well as a low relapse incidence, even in an older population. Second, the intensity of conditioning regimens used in our patients is remarkably different from all previous reports on older patients so far. Indeed, in half of our patients we attempted to increase the intensity of conditioning regimen with a second alkylating agent or with TBI. This could be a possible explanation for the lower relapse incidence in our patient, despite the high percentage of high/very high risk diseases. Moreover, it has been reported that patients with advanced disease performed better after transplant if treosulfan was administered as part of their conditioning regimen [4]. Third, almost all patients received in-vivo T-cell depletion. NRM was not affected by increased intensity or by in vivo T-cell depletion strategies; this may be a result of the low toxicity profile of treosulfan-based conditioning regimen [5], and of the use of PTCy for T-cell in-vivo depletion strategy [6, 7]. Both day-100 and 3-y NRM were in line with previous reports on leukemic/myelodysplastic allografted patients, even younger [8,9].
The nearly exclusive use of PBSC did not increase the incidence of aGvHD II-IV, and the incidence of grade III-IV was similar to that of previous reports in older patients [10]. These results confirm the efficacy of a rapamycine-based GvHD prophylaxis and may support the more favorable cytokine production due to treosulfan use compared to busulphan [4] . Of note, most of the extensive cGvHD resolved and patients restored a good quality of life withdrawing all IST.
These encouraging results in terms of NRM, RI, and GvHD, resulted in a high OS and DFS at 3 years in our older population, far better than the first reports about allo-HCT in the elderly [8,11]. In contrast to previous data [11] but according to Ustun et al.
[12], we did not find any difference in transplant outcomes in patients younger or older than 70 years. Both KPS and HCT-CI confirmed to be useful for patients selection, but they are not superimposable. Patients diagnosed with AML resulted in a reduced survival after transplant, mostly as a result of higher numbers of death from infections. Therefore, it should be suggested to proceed to transplantion as soon as possible if the patient has an indication for allo-HCT and according to his fitness. Based on our data, the use of a matched donor with PTCy as back-bone for GvHD prophylaxis seems to be the best choice for improving survival in the elderly.
Our results confirm that patient age should not be the only criteria for excluding AML/MDS patients older than 65y from allo-