Chromosomal instability (CIN) is a major driver of tumor progression and treatment resistance in many cancers. CIN is characterized by ongoing chromosome missegregation, generating copy number heterogeneity that provides a substrate for natural selection. Although CIN has been well studied in model systems, the evolutionary dynamics and genomic impact of CIN is underexplored in primary tumors and metastases. Newly developed single cell genomics assays allow for the measurement of the whole genomes of single cells obtained from patient tissues, thus providing a direct view into the evolution of tumor genomes undergoing CIN. To study the single-cell and spatial variation of chromosomally unstable tumors, we performed multi-region analysis of 128 tumor samples in a cohort of 45 patients with high-grade serous ovarian cancer (HGSOC), an archetype of high CIN cancers. We performed single-cell whole genome sequencing (scWGS) and profiled a total of 70,644 single cells from 63 pre-treatment samples obtained across 43 patients. For each patient, we computed total and allele specific copy number per cell, and inferred the evolutionary history of each tumor, including clonal and subclonal whole genome doubling (WGD) events and cell-specific whole chromosome or arm level copy number changes, indicative of missegregations. To determine whether missegregated chromosomes fail to become part of the primary nucleus (PN) upon mitotic exit and become enclosed in a micronucleus (MN), we used high-resolution cGAS immunofluorescence (IF) microscopy to profile site-matched FFPE tissues and quantify spatially-resolved rates of MN formation and rupture across 128 site-matched samples. A majority of tumors were characterized by early clonal WGD (61%, 22/36 cases). The remaining patients exhibited subclonal WGD subpopulations averaging 10% of the patient’s cells in total. To understand the interplay between recent WGD and CIN across the cohort, we subset all patients cells into non-WGD, subclonal WGD, and clonal WGD and computed missegregation rates for these subsets. Non-WGD cells showed the lowest levels of missegregation. Subclonal WGD exhibited the highest rates of missegregation indicative of CIN following a recent WGD event. Clonal WGD exhibited intermediate levels of missegregation indicative of stabilization following a historical WGD event. CIN following WGD often resulted in a reduction in ploidy relative to the ancestral WGD cell. Using IF, we quantified the frequent breakdown of the MN membrane, and estimated spatially-resolved CIN rates by tracking localization of cGAS to the MN, which revealed elevated rates of MN formation and rupture. Through multimodal single cell measurements of HGSOC tumors, we were able to characterize genomic complexity from individual cell division errors in HGSOC, and shed light into how the transient processes governing ovarian cancer cell evolution impact the genomes of cell populations.
Citation Format: Ignacio Vazquez-Garcia, Florian Uhlitz, Marc J. Williams, Jun Li, Hongyu Shi, Samuel Freeman, Matthew Myers, MSK SPECTRUM Consortium, Britta Weigelt, Dmitriy Zamarin, Andrew McPherson, Samuel F. Bakhoum, Sohrab P. Shah. Transient chromosomal instability as a driver of ovarian cancer genome evolution [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 102.