Abstract
Introduction: With the advent of modern multi-drug chemotherapy regimens, several aggressive hematologic malignancies now have excellent outcomes. However, long-term follow-up studies show increased incidences of secondary malignancies, lifestyle diseases, and other features of accelerated aging.
Aging is a complex, pervasive, and progressive phenomenon typified by characteristic molecular and cellular changes. Over the past two decades, a new unifying theory of aging has been shown to best explain the aging phenotype - the “Information Theory of Aging” proposes that aging in eukaryotes is due to loss of transcriptional networks and epigenetic information over time. Human DNA methylation studies demonstrate this accrual of epigenetic dysregulation over time which strongly corroborates with parameters of aging, while genomic sanctity is broadly maintained by high-fidelity, endogenous repair mechanisms.
This raises the concern that standard chemotherapeutic drugs, despite their life-saving action, may affect the epigenetic landscape via DNA methylation modification, resulting in an accelerated aging phenotype in keeping with the prevailing information theory of aging.
Our objective was to study these methylation changes in survivors of hematologic malignancies.
Methods: In this observational cohort study, we conducted a comprehensive analysis of blood methylation profiles in survivors of Hodgkin's Lymphoma (HL), Acute Promyelocytic Leukemia (APL), and Acute Myeloid Leukemia (AML); all 30-50 years of age, disease-free for >5 years from treatment completion (n=10 in each group). Age-matched normal individuals were tested as controls. Whole-genome methylation patterns were assessed using Illumina's Infinium MethylationEPIC v2.0 BeadChips and represent current methylation profiles of nucleated myeloid/lymphoid cells and their progenitors. A comparison of methylation profiles against controls hypothetically represents epigeno-toxicity of three different treatment regimens, differing in drugs, dosages, and durations.
Results: Methylation profiles in the cancer survivor cohorts were found to be different from age-matched controls. We demonstrated that levels of aberrant DNA methylation were higher among cancer survivors; observing an average of 2.34%, 6.93%, and 10.68% higher methylation in the lowest quantile of the 935K CpG sites on the BeadChip array in HL, APL, and AML survivors, respectively. To understand if these methylation changes were associated with epigenetic age acceleration (EAA), we used the PCGrimAge and DunedinPACE epigenetic clock tools. Indeed, when applying the DunedinPACE tool, EAA corresponded to increased methylation noted above. When normalized against healthy controls, HL, APL, and AML survivors showed an age increase of 13.8, 23.8 and 28.8 months on PCGrimAge, respectively.
Of the differentially methylated CpGs (DMCs) identified, the HL, APL, and AML cohorts had 211, 2085, and 4248 DMCs of significance (p<0.0001), respectively. Principal Component Analyses of the top DMCs showed a clear separation between these cohorts. A gene ontology pathway analysis of the most involved DMCs in the AML survivor cohort compared to normal controls showed enrichment in embryonic/tissue morphogenesis, neuronal migration and stem cell proliferation - processes potentially pointing to cellular de-differentiation. All these findings taken together demonstrate a signal towards significant, dose-dependent methylation change in these individuals, in keeping with accelerated epigenetic aging.
Conclusions: Our findings suggest that different hematological malignancies and their chemotherapies are associated with varying long-term epigenetic changes and EAA in normal hematopoietic progenitors. The implications underscore an unmet need for further research and potential changes in treatment algorithms on this basis.
