Abstract 6255: Personalized medicine for DCIS

Abstract Background: In 2023, 55,720 women will be diagnosed with non-invasive ductal carcinoma in situ (DCIS) in the United States. We evaluated RNA sequencing (RNA-Seq), clinical and social factors that drive subsequent breast events in a DCIS study population enriched for Black women (30%). Methods: RNA was extracted from macrosections of archival formalin-fixed paraffin-embedded (FFPE) DCIS tissue from Resource Archival Human Breast Tissue (RAHBT) from the St. Louis Breast Tissue Registry at Washington University School of Medicine and the Dana Farber Cancer Institute patient cohorts. Tissues were prepared by microsection of 5 uM whole section from unstained slides (Fisher). RNA was extracted using Qiagen RNeasy FFPE Kits from macrosectioned DCIS specimens. Purified total RNA was evaluated for quality and quantity using the Agilent Fragment Analyzer RNA assay. Bulk RNA sequencing was piloted in DCIS case-control samples using the SMART-seq and Hybrid Capture platforms at the DFCI Molecular Biology Core Facilities. Raw reads were aligned to the GRCh38/hg38 reference genome using STAR. Preliminary quality control analysis of a curated gene list showed higher quality data using the Hybrid Capture technology. Libraries for Hybrid Capture were pooled in 8plex reactions and hybridized to Twist Exome 2.0 probes for target enrichment. The final library pools were sequenced on an Illumina NovaSeq. Cases were defined as women diagnosed with DCIS in 1999 or later with subsequent invasive breast cancer, local DCIS recurrence, or lobular CIS recurrence. Controls were defined as women diagnosed with DCIS in 1999 or later without subsequent breast events during a similar follow-up time. The R package DESeq2 was used for data analysis. Results: The mean age at DCIS diagnosis was 56.3 (Std Dev 10.23). The mean follow-up time was 147.8 months. Hybrid Capture RNA-Seq files are stored and analyzed from the Harvard Medical School O2 Cluster using R programming (version 4.2.2, DEGreport). Quality control analyses demonstrated that the majority of the samples sequenced by Hybrid Capture had over 20 million total reads. Principal Component Analyses (PCA) and hierarchical clustering analyses of 19,821 protein-coding genes were conducted on the 143 DCIS samples sequenced using Hybrid Capture. PC1 was race (9%). A significant association was observed for follow-up time (PC3, 4.03%). First, we explored the association of the candidate gene ERBB2. The mean transcript abundance for ERBB2 was 11784.37 (case-control p-value=0.08). Preliminary analysis of selected genes of interest did not show statistically significant differences between cases and controls after FDR p-value adjustment. Differential gene expression analyses are ongoing. Conclusion: This study demonstrates the feasibility of generating high-quality RNA-seq data from diverse, archival DCIS samples. Citation Format: Aditi Hazra. Personalized medicine for DCIS [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6255.