Session: 723. Allogeneic Transplantation: Long-term Follow-up, Complications, and Disease Recurrence: Poster I
Hematology Disease Topics & Pathways:
Research, Elderly, Clinical Practice (Health Services and Quality), Clinical Research, Real-world evidence, Human, Study Population
This retrospective study included elderly patients (≥60 years old) who underwent haploidentical HSCT at twelve transplant centres in Zhejiang Province, China, between April 2016 and November 2023.
The EASIX scores (lactate dehydrogenase × creatinine/platelets) were calculated at 0 to 30 days before initiating the conditioning regimen (EASIX-pre), at day 0 post-HSCT (EASIX-d0),at day 7 post-HSCT (EASIX-d7), at day14 post-HSCT(EASIX-d14), at day 28 post-HSCT (EASIX-d28), at day 56 post-HSCT(EASIX-d56) and at day 100 post-HSCT (EASIX-d100). In order to reduce bias, the EASIX scores were converted to a log2 form of analysis (log2-EASIX).
The median age of all patients (n=164) was 63.5 years (range: 60.0-71.9 years). Acute myeloid leukaemia (n=71, 43%) and myelodysplastic syndromes (n=54, 33%) were the most common baseline diagnoses. All patients underwent haploidentical HSCT, and 91% (n=150) of patients received reduced-intensity conditioning. At day+100, the cumulative incidences of I–IV aGVHD and transplantation-associated thrombotic microangiopathy (TA-TMA) were 28.7% and 3.7%, respectively. The incidence of cGVHD at two year was 30.6%. With a median follow-up period of 14 months, the 2-year OS, NRM, and relapse incidence were 54.7%,28.8%, and 18.8%, respectively.
EASIX scores exhibited a rapid increase after HSCT and reaching a peak on day +7 , followed by a slight decline until day +28, but remained above the pre-HCT baseline.
The log2-EASIX-PRE(continuous variable) was significantly associated with worse OS (univariate analysis;[HR],1.25;95% CI,1.07 to 1.47;p= 0.005) and higher NRM (univariate analysis; [HR],1.32; 95% CI, 1.08 to1.62; p = 0.007), but not associated with relapse (univariate analysis; [HR], 0.98; 95% CI, 0.78 to 1.23; p = 0.86). Similarly, log2-EASIX-d0 ([HR], 1.25; P =0.021),log2-EASIX-d7 ([HR],1.40;P =0.015),log2-EASIX-d14 ([HR], 1.40;P =0.003), log2-EASIX-d21 ([HR], 1.42; P<0.001),log2-EASIX-d28 ([HR], 1.66; P<0.001), log2-EASIX-d56 ([HR], 1.54; P<0.001)and log2-EASIX-d100 ([HR],1.83;P<0.001) were also found to be predictive of higher NRM.
The most discriminating EASIX cut-off value before transplantation for risk of OS and NRM was 4.25 on the original scale(2.09 on the log2 scale). Patients with EASIX-PRE≥4.25 exhibited lower OS (2 years,24.3% vs 59.3%; P<0.001) and higher NRM (2 years, 54.3%vs 24.5%; P=0.001) compared to patients with EASIX-PRE<4.25.In multivariate analyses, we were able to validate the cut-off and found that EASIX-PRE≥4.25 was associated with more than threefold increased risk for OS(HR=3.72, 95%CI 1.74 to 7.94, p<0.001)and NRM (HR=4.04, 95%CI 1.86 to 8.79,p<0.001).Furthermore, EASIX-PRE≥4.25 was related to an increased risk of death from infection (HR=2.34, 95%CI 1.09 to 5.03,p=0.029).
The occurrence of cGVHD was found to be associated with log2-EASIX-d0([HR], 0.82; P=0.039), log2-EASIX-d21([HR], 0.78; P=0.003) and log2-EASIX-d28([HR], 0.76; P=0.004) .Nevertheless, no correlation was observed between log2-EASIX at any time point and the occurrence of aGVHD and TA-TMA.
EASIX is a readily accessible dynamic prognostic score that accurately predicts OS and NRM in elderly patients undergoing haploidentical HSCT at various time points.Notably, the optimal pre-transplantation EASIX cut-off value can be employed as a pre-transplantation assessment in elderly patients undergoing haploidentical HSCT, irrespective of the existing scoring system.
Disclosures: No relevant conflicts of interest to declare.