Main

Tomasetti C, Vogelstein B. Science 2015; 347: 78–81

This header is predicated on the high correlation (r = 0.81) between the number of divisions of self-renewing cells in that tissue and the lifetime risk of that cancer. The investigators argue only a third of cancers are as a consequence of environmental factors or hereditary. 'Machine learning methods were employed to classify tumors based only on this score' – extra risk score (see www.sciencemag.org/content/347/6217/78/suppl/DC1). For R-tumours (R for replicative with stochastic effects influencing risk) that comprise the majority of tumours including pancreatic islet, osteosarcoma and head and neck cancer but head and neck cancer only marginally, 'primary prevention measures (altered lifestyles or vaccines)...are not likely to be very effective.' For these tumours, secondary prevention including early detection should be the major focus. In contrast, primary prevention may have a major impact on that minority classified as D-tumours (D for deterministic such as HPV-16 head and neck, and lung smokers). This paper moves the emphasis away from a life-style victim-blaming model for oncogenesis.