Presenter(s)
A. Le1, K. McGreevy2, M. Carter1, E. Rietdorf3, A. J. Chang4, and J. B. Weidhaas5; 1UCLA, Los Angeles, CA, 2MiraDx, Los Angeles, CA, 3UCLA Department of Radiation Oncology, Los Angeles, CA, 4Department of Radiation Oncology, University of California, Los Angeles, Los Angeles, CA, 5Department of Radiation Oncology, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA
Purpose/Objective(s):
PROSTOX is a germline microRNA variant (mirSNP) based signature predicting late GU adverse events (AEs) for prostate cancer patients receiving either stereotactic body radiotherapy (SBRT, PROSTOX Ultra) or conventionally/moderately hypofractionated radiation (PROSTOX Standard). To further investigate the genetic basis of late GU AEs based on these signatures, we performed a baseline and longitudinal mRNA and miRNA sequencing study of patient-derived lymphoblastoid cell lines (LCLs) from patients with known PROSTOX scores, treatments and outcomes.
Materials/Methods:
LCLs were irradiated with 5Gy and total RNA collected at 0 (baseline), 6, 24 and 48 hours post-irradiation. Expression was modeled using linear mixed-effects models accounting for repeated measures and time, and gene set enrichment analysis (GSEA) was used to identify GO Biological Process pathways differing between PROSTOX High- and Low-risk groups.
Results:
Thirty-seven genes (20 long noncoding RNAs [lncRNAs] and 17 protein-coding) and 1,459 GO Biological Process pathways were significantly associated with PROSTOX Ultra risk (High- versus Low, adj p < 0.05) at baseline and for the first 48-hours post-irradiation. Enriched pathways prominently involved RNA-related processes (mRNA splicing, RNA processing, post-transcriptional regulation), chromatin remodeling, DNA repair, and cell cycle control. A single lncRNA had a significantly different trajectory and 15 enriched pathways over the time-period of the analysis. No individual miRNA showed a significant global or time-dependent difference after multiple-testing correction. However, target-based GSEA of the top 14 miRNAs identified 18 enriched GO processes (adj p < 0.05); 8 (44%) of these were also enriched in the mRNA analysis, including regulation of the cell cycle, intrinsic apoptotic signaling in response to DNA damage, and global negative regulation of translation and gene expression. This concordance suggests that PROSTOX Ultra-associated differences in miRNA regulation propagate to downstream mRNA expression within the same biological circuits.
Conclusion:
These findings support the hypothesis that the mirSNP-based PROSTOX Ultra signature identifies a pre-existing, multi-layered regulatory state in which miRNA variation helps set global mRNA expression patterns, leading to unique radiation responses. Ongoing work is investigating differences between PROSTOX Standard High- and Low risk LCLs. In addition, validation work is ongoing in prostate cancer patient biopsy samples to confirm concordance as well as to allow evaluation of any maintained long-lasting gene expression differences between PROSTOX subgroups.