P1374: EASIX ASSESSED AT 1 YEAR AFTER ALLOGENIC HEMATOPOIETIC CELL TRANSPLANTATION PREDICTS THE ONSET OF SUBSEQUENT NON-RELAPSE MORTALITY

Abstract

Background: Delayed non-malignant complications after allogeneic hematopoietic cell transplantation (allo-HCT) affect quality of life and increase the risk of premature death over 10 fold. Although about half of all deaths after allo-HCT are due to the primary disease, non-relapse mortality (NRM) is also a major issue for long-term follow-up. Transplant endothelial activation and stress index prior to allo-HCT (EASIX-pre) were reported to be predictive markers for NRM after allo-HCT. However, risk factors of late NRM after over 1 year from allo-HCT, which is caused by organ damage affected by allo-HCT, have not been elucidated. In the present study, we evaluated late NRM after allo-HCT by assessing EASIX 1 year after allo-HCT (EASIX-1year). Aims: The purpose of this study was to clarify the usefulness of EASIX-pre and EASIX-1year as predictive markers of NRM and overall survival (OS). Methods: We included 210 patients with hematological disease who underwent a first allo-HCT between 2006 and 2019 at our facility. We evaluated EASIX-1year in 102 (53 males) patients who were alive after 1 year without relapse. The patients consisted of acute myeloid leukemia (n = 44), acute lymphoblastic leukemia (n = 24), myelodysplastic syndrome (n = 14), malignant lymphoma (n = 8), and other diseases (n = 12). The median age of the patients was 40 years (range: 16-66 years). Fifty-seven patients received myeloablative conditioning,77 patients received total body irradiation, and 44 patients developed chronic graft-versus-host disease. The HCT-comorbidity index (HCT-CI) risk groups consisted of low-risk (n = 51), intermediate-risk (n = 28), and high-risk (n = 23). EASIX was calculated as lactate dehydrogenase (U/I) × Cre (mg/dL) / Plt (nL). EASIX-pre was evaluated 7-10 days before conditioning, and EASIX-1year was evaluated within 1 month of 1 year after allo-HCT. NRM and OS that developed 1 year after allo-HCT were defined as late NRM and late OS, respectively. Analyses were performed using landmark analysis. Late NRM and late OS assessments were performed based on the cutoff values from the receiver operating characteristic curve. The study was approved by the Ethical Committee of our facility. Results: Median EASIX-pre was 0.98 (0.12–24.1). The C-statistics of EASIX-pre for 1-year NRM and OS were 0.579 and 0.698 (both cutoff values: 1.108), and those for late NRM and late OS were 0.561 and 0.591 (cutoff value: 0.595 and 0.766), respectively. Univariate analysis revealed that a high EASIX-pre value was associated with higher 1-year NRM (18.6% vs. 8.0%, p<0.05), and significantly lower 1-year OS (56.3% vs. 82.8%, p<0.01). However, a high EASIX-pre value was not significantly associated with both late NRM and late OS (5-year NRM 3.4% vs. 0%, p=0.24, 5-year OS 88.9% vs. 96.9%, p=0.06). HCT-CI was not an indicator of late NRM (5-year NRM 10.2 % vs. 0%, p=0.29). Median EASIX-1year was 0.98 (0.15–21.8). The C-statistic of EASIX-1year for late NRM and late OS were 0.63 and 0.663 (cutoff value: 0.595 and 1.159), respectively. Univariate analysis revealed that a high EASIX-1year value was significantly associated with both late NRM and late OS (5-year NRM 10.1% vs. 1.3%, p<0.05, 5-year OS 88.7% vs. 95.1%, p 0.05). Multivariate analysis was performed by adjusting age, donor source, HCT-CI, chronic graft-versus-host disease, and conditioning. A high EASIX-1year was extracted as a risk factor for late NRM (Hazard ratio 4.26, p<0.05). Summary/Conclusion: This study showed that EASIX-1year is a useful prognostic indicator for late NRM occurring after over 1 year after allogeneic HCT.