Abstract P3-05-06: Genome-wide DNA methylation analysis identifies novel biomarkers associated with risk of relapse beyond oncotype DX recurrence-score risk assessment within HR+/HER2- early-stage breast cancer patients

Abstract

Abstract Background: Early-stage, node-negative, hormone receptor positive (HR+) HER2 negative (HER2-) breast cancer (BC) patients comprise a subset with good prognosis. Based on gene expression panels, these patients may skip chemotherapy and complete adjuvant endocrine therapy as an individual genomic basis assessment. However, up to 15% of patients categorized as non-high risk by Oncotype DX (Recurrence Score, RS≤25) may experience disease relapse. The field of cancer epigenomics is emerging as a dynamic tool to complement prognostic expression analyses and to provide further insight into the biological underpinnings of early HR+/HER2- BC. Methods: A genome-wide DNA methylation profiling in tumor tissue samples, from node-negative HR+/HER2- BC patients with available Oncotype DX RS data, was analyzed using the Illumina MethylationEPIC 850K BeadChip. We evaluated all differential methylation patterns independently of their position to establish a bioinformatic analytical pipeline. The online tools KM Plotter and ROC Plotter were used to assess the impact of survival outcomes (5-year relapse-free survival, RFS) and response to therapy of the candidate genes. Individual patient data from TCGA was used to analyze the correlation between methylation and mRNA levels in HR+/HER2- BC patients. Statistical significance was defined by p<0.05; no multiple testing correction was applied. Results: 31 patients were evaluated (mean age 51.4, SD 13.3). Four samples were discarded after quality control analysis. Sixteen tumor patients presented a RS ≤25, of which 6 patients had experienced relapse with a median follow-up of 7.1 years. Of the 773,193 loci studied, significant differential methylation was observed between the relapse and non-relapse groups at 19665 cgs (2,5%). Genes were ranked according to differential methylation patterns, and those with Delta ≥|0.25|were selected. Ninety-nine genes met these criteria. Of them, 85 (86%) were hypomethylated and 14 (14%) were hypermethylated in the relapse group. We studied methylation-expression correlations and the impact on RFS in independent datasets. Among the hypomethylated genes, VAC14 (HR 0.71, CI 0.63-0.82, p 9.7e-7), NDUFS6 (HR 0.68, CI 0.59-0.76, p 1.5e-8), and C7orf50 (HR 0.74, 0.6-0.91, p 0.0059) were found significantly associated with a worse RFS and exhibited a negative correlation between methylation levels and gene expression. Among the hypermethylated genes, MFSD7 (HR 1.31, CI 1.15-1.49, p 6.1e-5), PLA2G3 (HR 1.15, CI 1.09-1.40, p 0.0016), and FGFR2 (HR 1.27, CI 1.11-1.45, p 3.6e-4) were significantly associated to a better RFS and showed a strong negative correlation. Additionally, NDUFS6 mRNA levels were found to be increased in patients with worse response to adjuvant chemotherapy (p 0.012, 1.5x higher; ROC 0.733), while C7orf50 expression levels was associated with a better outcome following adjuvant ET (p 0.012, 1.3x higher, ROC 0.795). Conclusion: In this exploratory study, 6 differentially methylated genes were identified to discriminate patients with non-high RS who may be at a higher risk of relapse. An effort is ongoing towards the development of an analytical pipeline including a more extensive annotation of both experimental and validation cohorts. Citation Format: Jesus Fuentes-Antrás, Vanesa García-Barberán, Nicolas Costa-Fraga, Fernando Moreno, Aida Bao-Caamano, Aitor Rodríguez-Casanova, Alfonso López de Sá, Alicia De Luna, Igor Lopez-Cade, Carmen Ramirez-Ruda, Alejando Pascual, Pedro Perez-Segura, Balázs Győrffy, Alberto Ocaña, Angel Diaz-Lagares, Jose Angel Garcia-Saenz. Genome-wide DNA methylation analysis identifies novel biomarkers associated with risk of relapse beyond oncotype DX recurrence-score risk assessment within HR+/HER2- early-stage breast cancer patients [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P3-05-06.