People likely have wondered about the cause of cancer for centuries. Its name derives from an observation by Hippocrates more than 2,300 years ago that the long, distended veins that radiate out from some breast tumors look like the limbs of a crab. From that observation came the term karkinoma in Greek, and later, cancer in Latin.
With the work of Hooke in the 1600s, and then Virchow in the 1800s, came the anderstanding that living tissues are composed of cells, and that all cells arise as direct descendants of other cells. Yet, this understanding raised more questions about cancer than it answered. Now scientists began to ask from what kinds of normal cells cancer cells arise, how cancer cells differ from their normal counterparts, and what events promote the prolif eration of these abnormal cells. And physicians began to ask how cancer could be prevented or cured.
more about cancer
Clues from epidemiology. One of the most impor tant early observations that people made about cancer was that its incidence varies between dif ferent populations. For example, in 1775, an extraordinarily high incidence of scrotal cancer was described among men who worked as chimney sweeps as boys. In the mid-1800s, lung cancer was observed at alarmingly high rates among pitch blende miners in Germany. And by the end of the 19th century, using snuff and cigars was thought by some physicians to be closely associated with cancers of the mouth and throat.
These observations and others suggested that the origin or causes of cancer may lie outside the body and, more important, that cancer could be linked to identifiable and even preventable causes. These ideas led to a widespread search for agents that might cause cancer. One early notion, prompted by the discovery that bacteria cause a variety of important human diseases, was that cancer is an infectious disease. Another idea was that cancer arises from the chronic irritation of tissues. This view received strong support with the discovery of X-rays in 1895 and the observation that exposure to this form of radiation could induce local ized tissue damage, which could lead in turn to the development of cancer. A conflicting view, prompted by the observation that cancer sometimes seems to run in families, was that cancer is hereditary.
Such explanations, based as they were on frag mentary evidence and incomplete understanding, helped create the very considerable confusion about cancer that existed among scientists well into the mid-twentieth century. The obvious question facing researchers—and no one could seem to answer it—was how agents as diverse as this could all cause cancer. Far from bringing science closer to understanding cancer, each new observa tion seemed to add to the confusion.
Yet each new observation also, ultimately, contributed to scientists' eventual understanding of the disease. For example, the discovery in 1910 that a defined, submicroscopic agent isolated from a chicken tumor could induce new tumors in healthy chickens showed that a tumor could be traced simply and definitively back to a single cause. Today, scientists know this agent as Rous sarcoma virus, one of several viruses that can act as causative factors in the development of cancer. Although cancer-causing viruses are not prime
agents in promoting most human cancers, their intensive study focused researchers' attention on cellular genes as playing a central role in the development of the disease.
Likewise, investigations into the association between cancer and tissue damage, particularly that induced
by radiation, revealed that while visible damage sometimes occurs, something more subtle happens in cells exposed to cancer-causing agents. One clue to what happens came from the work of Herman Muller, who noticed in 1927 that X-irradiation of fruit flies often resulted in mutant offspring. Might the two known effects of X-rays, promotion of cancer and genetic mutation, be related to one another? And might chemical carcinogens induce cancer through a similar ability to damage genes?
Support for this idea came from the work of Bruce Ames and others who showed in 1975 that com
pounds known to be potent carcinogens (cancercausing agents) generally also were potent muta
gens (mutation-inducing agents), and that compounds known to be only weak carcinogens were only weak mutagens. Although scientists know today that many chemicals do not follow this correlation precisely, this initial, dramatic association between mutagenicity and carcinogenic ity had widespread influence on the development of a unified view of the origin and development of cancer.
Finally, a simple genetic model, proposed by Alfred Knudson in 1971, provided both a compelling explanation for the origins of retinoblas toma, a rare tumor that occurs early in life, and a convincing way to reconcile the view of cancer as a disease produced by external agents that damage cells with the observation that some cancers run in families. Knudson's model states that children with sporadic retinoblastoma (children whose parents have no history of the disease) are genetically normal at the moment of conception, but experience two somatic mutations that lead to the development of an eye tumor. Children with familial retinoblastoma (children whose parents have a history of the disease) already carry one mutation at conception and thus must experience only one more mutation to reach the doubly mutated configuration required for a tumor to form. In effect, in familial retinoblastoma, each retinal cell is already
primed for tumor development, needing only a second mutational event to trigger the cancerous state. The difference in probabilities between the requirement for one or two mutational events, happening randomly, explains why in sporadic retinoblastoma, the affected children have only one tumor focus, in one eye, while in familial retinoblastoma, the affected children usually have multiple tumor foci growing in both eyes.
Although it was years before Knudson's explanation was confirmed, it had great impact on scientists' understanding of cancer. Retinoblastoma, and by extension, other familial tumors, appeared to be linked to the inheritance of mutated versions of growth-suppressing genes. This idea led to the notion that cells in sporadically arising tumors might also have experienced damage to these crit ical genes as the cells moved along the path from the normal to the cancerous state.
