Evolution and clinical significance of complex genome rearrangements

PROJECT SUMMARY Focal amplifications and structural variants arise from ongoing processes of genomic instability, a hallmark of cancer. They often affect prevalent cancer driver genes and were the first classes of clinically actionable genomic alterations for targeted therapies. Despite their prevalence and clinical significance, their place in cancers' recur- rent evolutionary patterns has been relatively underexamined compared to point mutations (SNVs) and simple copy number alterations (CNAs). The temporal order, or relative `timing', of SNVs and low-level CNAs (< 10 copies) in tumor evolution has been characterized in thousands of tumors across dozens of tumor types. This has provided insight into their role in tumorigenesis and progression and revealed opportunities for early cancer screening. Moreover, these evolutionary reconstructions may inform efforts in precision medicine by revealing recurrent drivers that are often acquired early or late in cancer progression, upon treatment, at recurrence, and in metastases. Recent work from our lab has significantly improved temporal resolution for evolutionary recon- structions of low-level CNAs during clonal evolution. Here, we will build on this work and develop evolutionary reconstruction approaches bespoke to focal amplifications and structural variants and map the evolutionary tra- jectories of complex genome rearrangements in pan-cancer datasets and two novel cohorts of rare tumors in which these alterations play a significant role. In Aim 1, we will develop two separate, complementary approaches for evolutionary reconstruction of amplifications and complex rearrangements. Aim 1A will produce Amplicon Gain And Velocity Estimation (AGAVE) which will time when, and how quickly amplifications occurred during clonal evolution. Aim 1B will produce Reconstruction Of Structural Evolution (ROSE) which will assemble partially ordered chronologies of structural variants formed during clonal evolution. We hypothesize that the evolution- ary patterns of amplifications and rearrangements provide insight into their causes and consequences. To investigate this hypothesis, we will use these methods, and others, to map trajectories of structural evolution in thousands of primary and metastatic tumors across dozens of tumor types. In Aim 2A, we will perform a pan- cancer analysis of structural evolution in 8,000+ primary and metastatic tumors from the TCGA, ICGC, and Hartwig cohorts. In Aim 2B, we will investigate the role of potential structural evolution in the tumorigenesis, progression, and differential survival of spontaneous and radiation-associated undifferentiated pleomorphic sar- coma. Lastly, in Aim 2C, we will reconstruct the evolution of rearrangements and amplifications in small cell lung cancer metastases sampled at autopsy, identifying potential roles for these alterations in metastasis and thera- peutic resistance in this highly lethal, frequently metastatic cancer. These studies will produce the most comprehensive portraits of tumor evolution to date, address a long-standing absence of complex ge- nome rearrangements from models of tumor evolution, and elevate our understanding of their roles in tumorigenesis, progression, and metastasis.