Abstract 2172: Tumor break load is a biologically relevant characteristic of genomic instability with prognostic value for patients with localized colorectal cancer

Abstract

Abstract Background: Genomic instability is a common feature of colorectal cancer (CRC). More than 80% of CRCs exhibit chromosomal instability (CIN) and 15-20% microsatellite instability (MSI). MSI results in a high ‘tumor mutational burden’ (TMB) and is associated with relatively good prognosis in localized CRC. However, for the large group of CIN CRC patients with localized disease it remains unclear who has a relatively good prognosis and who is at high risk of disease recurrence. CIN is characterized by excessive somatic copy number aberrations (SCNAs), which is often summarized as the ‘fraction genome altered’ (FGA). In addition, CIN is also characterized by excessive chromosomal rearrangement structural variants (SVs). We quantitated the chromosomal breaks as the ‘Tumor Break Load’ (TBL) to investigate the biological impact and clinical relevance of chromosomal breakage in colorectal cancer. Aim: To investigate the biological impact and clinical relevance of TBL in CRC. Data and Methods: We used SCNA, RNA-Seq, and Disease-Free Survival (DFS) data from 633 CRC samples from The Cancer Genome Atlas project. For validation purposes, we used SCNA and DFS data from an independent series of 72 CRC samples. TBL was calculated as the sum of SCNA-associated chromosomal breaks. We used the 25% and 75% quantiles of the TBL distribution as predefined high and low TBL groups to train an RNA expression-based classifier that identifies TBL-associated expression phenotypes as an indicator of biological impact. Results: Within the group of microsatellite stable (MSS) CRC samples there was a large variety in TBL (median TBL of 54 [1-308]). TBL was poorly correlated with TMB (R2: 0.06) and FGA (R2: 0.07). Using RNA expression data as a phenotypic representation of genomic instability, MSI and MSS CRC samples could be classified with high accuracy (AUC: 0.99, p < 0.01), confirming the well-known impact of MSI on tumor biology. Using the same methodology within the group of MSS samples, the TBL could also be classified with high accuracy (AUC: 0.88; p < 0.01). Moreover, high TBL was associated with poor DFS in stage II-III localized MSS CRC (HR: 6.1; p = 0.007). This result was validated in an independent data series of 72 untreated stage II-III localized MSS CRC patients (HR: 3.0; p < 0.02). Conclusion: TBL is a characteristic of chromosomal instability that is largely independent from TMB and FGA, with significant impact on tumor biology. Our data indicate that the TBL may be used as a valuable prognostic biomarker for DFS in patients with localized colorectal cancer. Citation Format: Soufyan Lakbir, Jaap Heringa, Gerrit A. Meijer, Sanne Abeln, Remond J. Fijneman. Tumor break load is a biologically relevant characteristic of genomic instability with prognostic value for patients with localized colorectal cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2172.