Trevor Graham Barts Cancer Institute Carcinogenesis is fundamentally an evolutionary process; establishing the prognosis for a cancer therefore requires predicting the future course of tumour evolution. The same is true in premalignant conditions: the risk of developing cancer is determined by how the premalignant lesion is evolving. The level of diversity within a population largely determines how fast that population will evolve. If there is no diversity natural selection cannot operate, whereas diverse populations are likely to contain well-adapted individuals that can prosper in changing environments. Consequently, quantification of within-tumour diversity is likely to be a proxy-measure of the rate of the underlying evolutionary process that drives carcinogenesis. In this talk, I will describe how we have measured within-tumour diversity, both genetically and phenotypically, and used these measures to successfully determine prognosis in both established cancers and in premalignant lesions. I will focus particularly on lung cancer and the premalignant condition Barrett’s Oesophagus. Building upon this work, I will present the case for the quantification of within-tumour diversity as a universal biomarker for cancer prognosis.