CANCER may spread from one part of the body to another by pure chance and not just as a result of key genetic mutations, according to new research.

Physicists from Dundee ­University, working alongside counterparts based at Arizona State University, believe their work has the potential to spark "game-changing" developments in the way the spread of cancerous tumours is understood.

The conventional view of the process, known as metastasis, has been that only "specialist" cells from the primary tumour can cause the growth of new tumours elsewhere in the body. The cells have the ability to perform a number of different tasks, such as invading local tissue, entering, surviving in and leaving the ­bloodstream, and colonising new tissue environments.

The view explains the ­inefficiency of metastasis and why it often takes years to cause death in most patients, as it is highly improbable that a cell will possess all of the genetic mutations required to carry out all of the functions.

However, using statistical models and theories on probability, the researchers found large numbers of common cancer cells that flow freely in the blood stream can also cause metastasis by pure chance.

Study co-author Professor ­Timothy Newman, from Dundee University, said: "If we use a ­military metaphor, a key mission can be accomplished using either a handful of highly trained special forces - in this case the specialist cells - or a huge number of untrained infantry - the common cancer cells - in which case a ­handful of ordinary soldiers will, by sheer luck, be successful. If one could magically observe the early growth of a metastasis, we show there would be no way of telling from the growth dynamics whether the tumour was seeded by a special forces cell or a lucky infantryman."

The study, which has been published in the Physical Biology journal, concludes that successful metastatic growth from common cells, although rare, would proceed extremely rapidly.

It could spur new ways of ­thinking about cancer research, demonstrating that statistical physics may be as fundamental as complex genetics when studying the occurrence and treatment of the disease.

Mr Newman added: "Our research is an example to the cancer research community that sometimes one needs to pause and step back from the genetic details of cancer and carefully consider in parallel other approaches and paradigms.

"Genetics is undoubtedly important in cancer, but not exclusively so, and there are equally fundamental concepts at higher levels which underpin cancer progression. Perhaps physicists, and others from outside the cancer research area, can help provide more insights along these lines, which may be game-changing."

The researchers also used very crude physiological data to estimate that the rare events caused by common cells would lead to semi-stable metastases in the size range of about 50 cells.

The finding was significant as growths of that size, although too small to observe using medical imaging in human patients, have previously been observed in experiments on mice and zebrafish.

The physicists believe it may be possible to find them in future through fine examination of human tissue.

The study's co-author, Dr Luis Cisneros, added, "If we allow ourselves to consider the role of randomness, then we open the door to perceiving surprising effects of the statistical fluctuations that may not be expected by naive reasoning."