ABSTRACT
A century ago (essentially based on the concept of Virchow) the cell was considered a basic element of many diseases, including cancer. Hundreds of types and subtypes of human tumors are still classified according to the cellular (cytological/histological) characteristics. However, in the past few decades we have moved a dimension further and have begun to understand the molecular machineries operating behind the cell functions as well as some of the molecular changes resulting in different diseases. Although it may sound reductionist, cancer is currently defined as genetic disease of the somatic cells, stating that somatic mutations are what all cancers have in common. Keeping in mind that the precise nature of these mutations and the cell types in which they take place must be understood in each individual type of cancer, our considerations should revolve around three main issues:
Our estimation, deriving from the pioneering work of Knudsen (the two-hit model) and partly from epidemiological studies on age dependence of the incidence of cancer in various populations, is that the number (n) of gene errors required is probably in the range of 2-5. There is no a priori reason why n should be the same for all tumors – even those with the same histological type or subtype. It is known that many gene errors identified in various cancers may appear in normal cells without known functional significance. It has also been demonstrated (especially in work from Vogelstein’s laboratory) that the morphological changes from a normal to a malignant structure can be – at least in certain cases – characterized by the gradual accumulation of gene errors. The sequence of somatic mutations may reflect the interplay of inherited factors and acquired factors in carcinogenesis. The inheritance of a mutated suppressor gene will decrease the number (n) of somatic mutations required to cause cancer by 1.
