Deciphering functional differences in the mechanisms driving focal oncogene amplification

Extrachromosomal DNA (ecDNA) is one of the key mechanisms driving somatic copy number amplification of oncogenes in cancer. Cancer Grand Challenges team eDyNAmiC has focused on identifying the unique biology of ecDNA, and its role in mediating the pathology of a multitude of cancers. 

Despite their importance, ecDNAs are not the only mechanism resulting in focal oncogene amplification. Breakage-Fusion-Bridge (BFB) cycles—first described by Barbara McClintock nearly 80 years ago in irradiated maize cells—have been identified as a significant driver of genomic instability in cancer. 

Importantly, BFBs are chromosomal, and are predicted to follow Mendelian inheritance during cell division, unlike ecDNAs, and therefore, may differ in their impact on treatment resistance and patient survival. Previously, methods to distinguish BFBs from ecDNA have been limited. In this paper, we develop a powerful new tool for detecting BFBs and distinguishing them from ecDNAs, confirming the hypothesis that BFBs confer different biological and clinical features, thus demonstrating the need to distinguish these types of gene amplification.

To address this gap, the team developed OM2BFB, a novel computational method designed to detect and reconstruct BFB structures using BioNano Optical Genome Maps (OGM). Unlike traditional sequencing-based approaches, OGM provides high-resolution, long-range structural information, enabling more accurate identification of BFB cycles. 

OM2BFB was rigorously validated using cytogenetic experiments, including metaphase and interphase fluorescence in situ hybridization (FISH), in human cancer cell lines and patient-derived xenografts. OM2BFB was also used to benchmark the accuracy of short-read sequencing methods, providing us with an opportunity to apply it to 1,538 whole-genome sequencing samples to systematically map the landscape of BFB-driven focal amplifications across diverse cancer types, and in many cancer cell lines. 

Further, by integrating functional datasets— chromatin conformation, drug treatment response, gene expression profiles, and patient outcomes—OM2BFB provided new insights into the structural and functional difference between BFBs and ecDNAs as contrasting methods of oncogene amplification

BFB amplifications differ from ecDNA in key structural and functional aspects, starting with location. BFB-driven amplifications are typically observed by cytogenetics methods as homogeneous staining regions (HSRs) on native chromosomes, whereas ecDNAs exist as circularized extrachromosomal elements. 

Unlike ecDNA, which has tremendous copy number heterogeneity due to random segregation, BFB-driven amplifications are more stable and uniform across cells. They are less able to modify copy numbers, leading to lower gene expression variability and reduced adaptability to environmental change. 

The tumor microenvironment of BFB containing tumours is significantly different from that of those with ecDNA. The regular, foldback driven structure of BFBs reveal they have lower levels of regulatory rewiring and enhancer hijacking compared to ecDNA. 

Finally, BFB positive tumours show delayed resistance to targeted therapies compared to ecDNA positive tumors, suggesting that cancers with BFB-driven oncogene amplification may respond differently to treatment. Together, these findings highlight that BFB cycles represent a distinct oncogene amplification mechanism, with critical implications for cancer biology and therapeutic strategies.

Each step in my academic journey has been instrumental in shaping my scientific interests and aspirations. I pursued my Bachelor's degree in Computer Science from 2013 to 2018, which laid the foundation for my interest in computational biology. 

Currently, I am a PhD student in the Computer Science Department at the University of California, San Diego, under the supervision of Vineet Bafna, where I focus on bioinformatics and cancer genomics. My broader research interests include algorithm development for genomic structural variant reconstruction and chromosomal rearrangement analysis, and their impact on cancer. 

As a member of team eDyNAmiC, we are dedicated to developing computational tools that empower the scientific community to analyse genomic data effectively. My primary focus is on creating computational methods for analysing amplicons, improving our understanding of oncogene amplification mechanisms.

I joined Paul Mischel’s lab at Stanford University as a postdoctoral fellow, after I received my PhD at the University of Hong Kong studying gynaecological cancers. 

As part of team eDyNAmiC, we are adopting an interdisciplinary approach to advance our knowledge on a huge problem in cancer biology posed by ecDNAs. I am mainly interested in the mechanistic role played by ecDNAs in fostering genome instability and aberrant transcription events.

I received my PhD from the University of Rochester in New York and have been conducting my research in Jean Zhao’s lab at Dana-Farber Cancer Institute/Harvard Medical School in Boston. My research journey has been centered on unravelling the intricate molecular mechanisms of cancer, with a particular emphasis on breast cancer, breast cancer brain metastasis, and primary brain tumors. 

I am dedicated to advancing our understanding of cancer biology and developing novel therapeutic strategies. Recently my research has centered on the roles of HER2 in breast cancer brain metastases, and the impact of ecDNA on oncogene amplification. Through these studies, I aim to contribute to the development of more effective cancer treatments.

---------------------------------

Through Cancer Grand Challenges team eDyNAmiC is funded by Cancer Research UK and the National Cancer Institute, with generous support to Cancer Research UK from Emerson Collective and The Kamini and Vindi Banga Family Trust.