The somatic evolution that drives neoplastic progression and therapeutic resistance has made it difficult to develop reliable biomarkers in cancer. Most biomarkers have been based on the presence, absence or level of a gene or gene product. However, the spatial and temporal heterogeneity of neoplasms, coupled with the fact that there are many different sets of mutations that can generate a cancer, make most biomarkers poorly correlated with the future behavior of the neoplasm. An alternative to measuring the products of somatic evolution is to measure the evolutionary process itself. We, and others, have shown that measures of the genetic diversity within a neoplasm predict progression to malignancy as well as overall survival. Measures of the parameters of evolution as well as the selective pressures on the neoplasms should be predictive across virtually all cancers. Furthermore, once we have measures of the rate of evolution, we can use those measures to develop cancer prevention interventions that slow somatic evolution.