https://www.sciencedaily.com/releases/2018/05/180529153055.htm

Cancer researchers have come to understand tumors not as lumps of identical cells, but rather as diverse, dynamic populations unto themselves. And, like individuals within animal populations, cells within tumors compete with one another, some thriving, some failing.

In a new study, researchers from the University of Pennsylvania have crafted a mathematical model to understand the dynamics at play as cancerous tumors grow and spread. Setting their model into action, Jimmy Qian, a rising senior in the Vagelos Scholars Program in the Molecular Life Sciences, and Erol Akçay, an assistant professor of biology in the School of Arts and Sciences, were able to explain a somewhat paradoxical observation, that mutations that lead to metastasis -- the spread of cancer to sites distant from the primary tumor -- often arise early, rather than late, in a tumor history.