AUTHORS
Jing Dong, PhD, Shahram Arsang-Jang, PhD2, Maryam Rafati, MD, PhD3, Tao Zhang, PhD4, Paul L. Auer, PhD5, Kristine Jones, PhD6, Jia Liu7, Stephen R. Spellman, MBS8, Yung-Tsi Bolon9, Wael Saber, MD10, Shahinaz Gadalla, MD, PhD11
AFFILIATIONS
1 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
2 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
3 National Cancer Institute, Bethesda, United States
4 CIBMTR® (Center for International Blood and Marrow Transplant Research), NMDP, Minneapolis, United States
5 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
6 National Cancer Institute, Bethesda, United States
7 National Cancer Institute, Bethesda, United States
8 CIBMTR® (Center for International Blood and Marrow Transplant Research), NMDP, Minneapolis, United States
9 CIBMTR® (Center for International Blood and Marrow Transplant Research), NMDP, Minneapolis, United States
10 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
11 National Cancer Institute, Bethesda, United States
ABSTRACT
Background: Mitochondrial dysfunction has been implicated in hematologic malignancies and transplant-related complications, yet the contribution of mitochondrial genomic variation to allogeneic hematopoietic cell transplantation (allo-HCT) outcomes remains poorly understood. We previously demonstrated that recipient mitochondrial DNA (mtDNA) variants in genes including MT-CYB and MT-ND5 are associated with post-transplant outcomes in patients with myelodysplastic syndromes (MDS). However, independent validation of these associations and evaluation of mitochondrial genomic determinants in other myeloid malignancies are lacking. In this study, we sought to validate previously identified prognostic mtDNA associations in an independent MDS cohort, identify novel mitochondrial genomic determinants of allo-HCT outcomes in myelofibrosis (MF), and determine whether integration of mtDNA variation improves risk prediction beyond established clinical prognostic models.
Methods: We performed ultra-deep mtDNA sequencing in 295 MDS patients enrolled in the Genomic Studies in Blood and Marrow Transplantation (GS-BMT) cohort and 228 MF patients from the Myelofibrosis Etiology and Transplant Outcome (MF-ETO) Study. GS-BMT and MF-ETO are collaborative initiatives between the NCI and the CIBMTR designed to investigate genomic determinants of transplant outcomes. mtDNA sequences were analyzed using the MToolBox pipeline. We evaluated the associations of mtDNA variation with post-transplant outcomes, including overall survival (OS), relapse-free survival (RFS), relapse, and transplant-related mortality (TRM). The contribution of mtDNA variation to outcome prediction beyond clinical risk factors was assessed using random survival forest models.
Results: The mean mtDNA sequencing depth was 21,067×, with greater than 99.9% of mitochondrial genome bases covered. Among MDS patients, the median age was 65.6 years, 65% were male, and 63% received grafts from 8/8 HLA well-matched unrelated donors. According to the Revised International Prognostic Scoring System (IPSS-R), 27.5% of patients had poor or very poor risk disease. The median follow-up was 95.9 months among survivors. A total of 2,004 mtDNA variants were identified in MDS. In independent validation analyses, previously reported associations were confirmed, including MT-ND4 with OS (P = 2.20 ×10-3) and MT-tRNA with relapse (P = 1.18 ×10-4) after Bonferroni correction. In addition, gene-based Sequence Kernel Association Test (SKAT) analyses identified significant associations of MT-CO1 with relapse (P = 1.24 ×10-3) and MT-CO3 and MT-tRNA with TRM (P = 2.05 ×10-3 and 1.62×10-5, respectively), further supporting the prognostic relevance of mitochondrial genomic variation in MDS.
In the MF cohort, the median age was 57 years, 84.2% of patients had intermediate-1 or intermediate-2 Dynamic International Prognostic Scoring System (DIPSS) risk, and 51% received myeloablative conditioning regimens. A total of 1,718 mtDNA variants were detected, of which 45.9% were shared with the MDS cohort, suggesting both common and disease-specific mitochondrial genomic features across myeloid malignancies. Higher mtDNA variant burden was associated with inferior OS (HR, 1.03; P=0.04). Gene-based analyses identified MF-specific associations involving OS with MT-CO3 (P = 7.63×10-4); relapse with MT-ND1 (P = 1.79×10-3) and MT-ND4 (P = 2.01×10-3); and TRM with MT-CYB (P=1.21×10-3), MT-ND4 (P=1.18×10-3), MT-CO3 (P=5.67×10-4), MT-ATP6 (P=2.07×10-3) and MT-tRNA (P=8.57×10-5). Shared prognostic associations between MDS and MF included OS with MT-ND2 (P=9.26×10-5), RFS with MT-ND4 (P=2.99×10-4), and relapse with MT-tRNA (P=3.55×10-4), highlighting mitochondrial pathways influencing transplant outcomes across myeloid diseases. Incorporation of mtDNA variation into predictive models improved discrimination of post-transplant outcomes beyond established clinical prognostic models in both MDS and MF.
Conclusions: This study provides independent validation of mitochondrial genomic determinants of allo-HCT outcomes in MDS and demonstrates, for the first time, that mitochondrial genomic variation contributes to transplant outcome heterogeneity in MF. These findings establish mtDNA variation as a complementary genomic dimension of transplant risk assessment and suggest that mitochondrial genomics may enhance risk stratification beyond conventional clinical models in patients with myeloid malignancies undergoing allo-HCT.
DISCLOSURES
Jing Dong, PhD: No relevant conflicts of interest to declare.