Haploidentical Related Donor versus Matched Sibling Donor Allogeneic Hematopoietic Stem Cell Transplantation for Acute Myeloid Leukemia and Myelodysplastic Syndrome Aged Over 50 Years: A Single-Center Retrospective Study.

BACKGROUND Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a potentially curative therapeutic option for patients with acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS). Increasing data supports the utility of haploidentical related donor (HID) HSCT in fit older patients and resulting in improvement of outcomes. This study compared the outcomes of acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS) patients age ≥50 years underwent haploidentical related donor (HID) or matched sibling donor (MSD) allogeneic hematopoietic stem cell transplantation (allo-HSCT).METHODS We retrospectively studied 38 patients with AML/MDS aged ≥50 years who underwent HID transplantation and compared their outcomes with 55 similarly aged patients who underwent MSD transplantation.RESULTS The 100-day cumulative incidence of II-IV° acute graft-versus-host disease (GVHD) were 34.2 ± 7.7% and 23.6 ± 5.7%, respectively, in HID and MSD groups (P = 0.189), and III-IV° acute GVHD were similar between two groups (5.3% and 7.3%, respectively, P=0.700). The 2-year cumulative incidence of limited and extensive chronic GVHD was not statistically different in HID and MSD groups(22.8 ± 10.8% vs. 18.2 ± 6.0% and 18.3 ± 10.4% vs. 22.1 ± 6.8%, P = 0.890 and P=0.424, respectively). The 2-year cumulative incidences of relapse (29.5±10.3% and 20.7 ±6.1%, P=0.458), 2-year overall survival (58.5±9.7% and 67.9±6.8%, P=0.373), 2-year transplant-related mortality (17.3±6.4% and 15.0±5.3%, P=0.717), 2-year progression free survival (56.8±9.7% and 64.6±7.4%,


Evaluation points and definitions
This study mainly focused on engraftment, GVHD, relapse, transplant-related mortality (TRM), overall survival (OS), and progression-free survival (PFS). TRM was estimated as death without evidence of leukemia recurrence. PFS was defined as survival in continuous complete remission without hematological relapse. aGVHD and cGVHD were graded according to the guidelines in the literature(16).

Statistical analysis
Our study data were analyzed on June 30, 2019. Comparisons of categorical variables were made by means of chi-squared and Fisher exact tests for small numbers. Differences between numerical variables were calculated by means of a 2sample t test. The incidence of time-dependent variables was estimated by the Kaplan-Meier method. A Cox regression model was used to analyze prognostic factors for relapse, PFS, TRM and OS. Numerical variables were analyzed as categories based on their values being below or above the median of the entire cohort. All statistical tests were two-sided, and a P-value less than 0.05 was considered significant. A multivariate analysis was performed using the Cox proportional hazards model. Variables were included in the multivariate model if they were conceptually important or if they approached or attained statistical significance by univariate analysis. All data analysis was performed with SPSS 24.0 (SPSS, IBM, USA).

Patient Clinical and Transplant Characteristics
A total of 93 AML or MDS patients aged ≥50 years after allo-HSCT were enrolled in this retrospective study, including 38 patients who underwent HID allo-HSCT and 55 patients who underwent MSD allo-HSCT. The median age of the patients was 58  Table 1. Significant differences were noted in the donor age, the stem cell source and the family relationship between recipients and donors between both groups. There were no significant differences in patient age, sex, sex match, disease status, conditioning regimen, hematopoietic cell transplantation comorbidity index (HCT-CI) score, time of follow-up, or dose of nucleated cells between the two groups. The patients' clinical and transplant characteristics are shown in Table 1.

Engraftment
All patients achieved hematopoietic reconstitution except one patient in the HID group who died of graft failure. Neutrophil reconstruction occurred in the HID group at a median of 12 d (range, 9-18) and in the MSD group at a median of 12 d (range, 8-22) (P=0.458). Platelet reconstruction in the HID and MSD groups occurred at a median of 13 d (range, 10-53) and 13 d (range, 8-91), respectively (P=0.333).

GVHD
The 100-day cumulative incidences of grade II-IV aGVHD were 34.2 ± 7.7% and 23.6 ± 5.7% in the HID and MSD groups, respectively (P = 0.189). The incidences of grade III-IV aGVHD were 5.3 ± 3.6% and 7.3 ± 3.5% in the HID and MSD groups, respectively (P = 0.700). One patient died of grade IV gut aGVHD in the MSD group, while no patients died of aGVHD in the HID group. The 2-year cumulative incidences of limited and extensive cGVHD were 7.3 ± 5.0% and 9.2 ± 6.4% and 14.8 ± 5.2% and 18.0 ± 5.8% in the HID and MSD groups, respectively (P = 0.890 and P=0.424, respectively). One patient died of cGVHD in the HID group, and three died of cGVHD in the MSD group (2.63% vs. 5.45%, P=0.642). The incidences of aGVHD and cGVHD are shown in Figure 1 and Figure 2. Sex mismatches were significantly associated with a higher risk of grade II-IV aGVHD and cGVHD (HR 2.400, CI 1.013-5.402, P=0.032 and HR 2.275, CI 1.022-5.011, P=0.042, respectively) in the multivariate analysis ( Table 3).

Infections
Patients in the HID group had significantly higher rates of CMV DNAemia than patients in the MSD group (p = 0.0005). The incidence of other major infectious complications, including sepsis, CMV disease, EBV DNAemia and invasive fungal infection, were similar between the 2 groups ( Table 2).

Relapse and Survival
The median follow-up was 13.

Discussion
Here, we report the outcomes of HID versus MSD allo-HSCT for patients with outcomes, including incidences of TRM and GVHD as well as relapse and survival.
Traditionally, allo-HSCT in the elderly has had a higher risk of TRM because of the patients' frequent comorbidities and poor performance status. Reports from main transplant centers for older patients have shown 2-year TRM rates ranging from 7% to 35% (2,7,(17)(18)(19). Some studies have reported that a myeloablative conditioning regimen (MAC) is associated with higher TRM rates than a nonmyeloablative conditioning (NMAC) or a reduced-intensity (RIC) conditioning regimen(20-23). In the present study, the 2-year cumulative incidences of TRM in the HID and MSD groups ranged from 15% to 17.3%. A report from Seattle of 1,055 patients undergoing allo-HSCT showed that the 2-year rate of TRM was 14%, 21%, and 41% for patients with HCT-CI scores of 0, 1 to 2 and 3 or more, respectively(24). A reasonable interpretation of the relatively lower rate of TRM is that our patients had relatively lower HCT-CT scores than the patients in other studies. In this report, only 18.4% of the patients in the HID group and 16.3% of the patients in the MSD group had HCT-CI scores ≥3. Additionally, 11 patients (28.9%) in the HID group and 19 patients (34.5%) in the MSD group received a BuF MAC. This may also be one of the reasons for the lower TRM. We and other researchers have shown that patients who receive a BuF MAC have a lower TRM incidence than those who receive BuCy(25, 26). Whether HID transplants have a higher rate of TRM than MSD transplants is currently under study. A growing number of studies show that there is no difference between HID and MSD transplants in terms of TRM, including in elderly patients(17-19, 27, 28). Similar results were obtained in this study.
Relapse is a major cause of failure in patients undergoing allo-HSCT. Many factors influence relapse, such as donor resources, disease status at transplants, patient age, conditioning regimen, and so on(29-31). In terms of donor resources, some studies showed that HID transplantation had a stronger GVL effect than MSD transplantation, decreasing the rate of relapse(14, 32, 33). Other studies suggested that there was no difference in the relapse rate between two donor resources(13, 17, 18, 34, 35). In the present study, there was no difference in relapse between the two groups. In multivariate analysis, we found that disease status at transplant was an independent risk factor for relapse, similar to other reposts(36, 37).
GVHD is the most common transplant-related complication that affects the outcomes of transplant patients(38, 39). Traditionally, HID transplantation was associated with a higher incidence of GVHD than MSD transplantation. Over the last decade, great improvements have been made in prophylaxis for GVHD in HID transplantation, especially the use of T-cell depletion in vivo by means of Cy or ATG(17, 32, 40). A growing amount of evidence has shown that the incidences of GVHD in HID transplantation are not different from those in MSD transplantation, especially in terms of cGVHD(14, 17, 27, 40-42). In the ATG protocol, our previous results showed that the incidences of grade II-IV aGVHD were higher for HID transplantation than for MSD transplantation, but the incidences of severe aGVHD and cGVHD were comparable between the two groups(14, 32, 42). In this study, HID transplantation was associated with a trend of higher incidences of grade II-IV aGVHD than those seen in MSD transplantation. In multivariate analysis, we found that a sex-mismatched donor (female donor/male recipient) was significantly associated with a higher risk of aGVHD and cGVHD than a sex-matched donor.
Several other studies have also reported the same results(43-47).
As we know, ATG as GVHD prophylaxis are associated with an increased incidence of infections, especially fatal viral infections(48-50). In the present study, although HID transplantation was associated with much higher incidences of CMV DNAemia than MSD transplantation, the number of deaths caused by viral diseases and other infectious diseases did not differ between the two groups. A possible explanation for these results might be the extensive experience at the study centers in effectively managing infectious diseases, resulting in many patients with CMV or other infections avoiding TRM.
The limitations of this study are the relatively small number of patients and the retrospective nature of this single-center study. We were not able to perform subgroup analyses, which would have been informative in some specific settings, such as in terms of advanced cytogenetics and/or older age. The sample size limitations in the oldest age group may have masked smaller differences in outcomes. Selection bias may also have influenced inferences from the data. It is possible that the older patients included in this study were a highly selected group with lower HCT-CI scores.

Conclusion
The present data showed similar outcomes in patients aged 50 years and older who underwent HID transplantation compared to MSD transplantation at our institution.