© Vicky Askew

The possibility that cancer could have a genetic basis was recognized in the early twentieth century (see Milestone 2). In addition, evidence for a clonal origin of tumours had emerged, as had views of cancer as a multistep process. Leslie Foulds, for example, early on described cancer as a “dynamic process advancing through stages that are qualitatively different”, progressing from precancerous stages to increasingly invasive and metastatic stages.

Yet, the now prevailing concept of Darwinian evolution and the stepwise progression of tumours was perhaps most convincingly articulated by Peter Nowell in 1976. His article incorporated the idea of cancer being caused by multiple mutations or 'hits' (see Milestone 9) into a general framework of tumour development and progression, through the accumulation and selection of genetic changes.

Nowell concluded that the first step results in cell proliferation that is “unrestrained to some degree”, allowing for a selective growth advantage. While also acknowledging the potential role of epigenetic alterations (see Milestone 19), he suggested that, as the result of acquired genomic instability in the expanding cell population, rare subvariants endowed with an extra selective advantage could emerge. Sequential rounds of clonal selection would produce tumour-cell populations with more aggressive phenotypes. Support for this concept came from the observation that advanced solid tumours often had a greater degree of aneuploidy than early stage lesions, and from the discovery of specific chromosomal changes that developed during the clinical progression of leukaemias.

Nowell discussed the mechanisms underlying genomic instability, such as DNA-repair defects or mitotic errors (see Milestones 2 and 22), and noted that diverse agents that cause cancer, such as ionizing radiation and viruses, can induce genetic changes and might contribute to the initial changes as well as the subsequent alterations.

Nowell wrote “it would be helpful if we could associate specific chromosomal alterations with particular aspects of tumour suppression”. However, at the time, few consistent changes had been noted, with the exception of the famous Philadelphia chromosome (see Milestone 10). Although Nowell anticipated similarities between different tumours, he also recognized that these would be difficult to identify amongst the multitude of evolutionary steps, and that variations due to different selection pressures were likely.

Subsequent years saw the identification of a number oncogenes and tumour-suppressor genes that were altered in human cancer. In an influential paper in 1990, Eric Fearon and Bert Vogelstein amalgamated these findings together with the idea of clonal evolution into a coherent molecular model of multistep tumorigenesis.

Focusing on colorectal cancer, the authors noted the clonal nature of the disease, and the consistent occurrence of mutations in the KRAS oncogene and the allelic loss of known or candidate tumour-suppressor genes, including p53 (also known as TP53 ). Although certain changes were preferentially associated with specific stages of disease progression, the authors documented a multitude of chromosomal and other changes, such as frequent DNA hypomethylation of specific regions. They therefore considered the total accumulation of changes, rather than their sequence, as most important for tumour progression. They also concluded that five or more genetic alterations were probably required for the development of carcinomas, with fewer changes needed for benign tumorigenesis.

This model of cancer evolution through the accumulation of mutations in both oncogenes and tumour-suppressor genes, and the stepwise selection of more malignant tumour-cell populations, has since been widely adopted and generalized to all common forms of cancer. At a time when we are beginning to see the basic understanding of the molecular changes underlying tumorigenesis translated into the development of targeted therapies (see Milestone 24), it is well worth noting the foresight of Nowell in suggesting that individual differences in the genetic and biological changes in each tumour might warrant personalized therapies.