Scientists think they have found a gene combination which allows the deadly spread of cancer around the body.
Fruit flies were engineered to develop tumours
While the findings are currently restricted to fruit flies, the team from Yale University in the US say similar genes may exist in humans.
Cancers that spread, or metastasise, are some of the hardest to treat.
The research, published in the journal Science, suggests there might one day be a way of blocking the spread of cancer by targeting one of the genes involved.
The reason why certain cancer cells are able to leave their original tumour and migrate to other sites around the body are still poorly understood.
However, scientists know that finding ways to stop this happening could potentially save many lives from the disease.
Many believe that metastasis has a genetic root - with mutant genes in the tumour cells somehow circumventing normal safeguards which stop them growing uncontrollably then "breaking out" to take hold elsewhere.
However, with the likelihood that a combination of genes, rather than just one, is responsible for this, finding the culprits within the vast human genome is like finding a needle in a haystack.
Instead, the Yale scientists turned their attention to Drosophila melanogaster a simple fruit fly that has a far smaller genome.
They say that despite their apparent obvious differences, fruit flies and humans share many of the same genes, so it could be possible to study tumour spread in flies - and apply those findings in man.
Other teams have already revealed a gene called "Ras v12" - present in both flies and humans - which, when mutated, is thought to play a significant role in helping the tumour flourish in some types of cancer.
However, this alone does not seem to produce cancer spread.
The US researchers launched a screening programme - creating flies which have the mutant Ras v12 genes, then interfering with other key genes to see if they could find the combination which led to metastasis.
They found a number of genes which normally play a role in the normal running of the cell, which, if inactivated in a tumour cell, produced flies with cancers that spread copiously.
Blocking the problem
The study is more evidence of the guilt of the Ras gene in metastatic cancer.
The more that researchers understand about the possible genes involved in cancer spread, the more chance there is of a method being created to interrupt it.
In this case, the researchers believe that there may be a clear chemical "pathway" within the cell which could influence the behaviour of many of the genes which might partner Ras V12 and cause problems.
This in theory means that a drug might be found that can activate this pathway.
Dr Wen Jiang, from the Metastasis Research Group at the University of Wales College of Medicine, said that the research was "an interesting piece of work".
"They have created a new model to investigate the genetic events during the non- invasive to invasive phase of a tumour.
"So far, this has been extremely difficult to study in humans, due largely to the limitation of capturing the right moment and correct
"It has provided some early information that
metastatic events and growth events may be distinct at genetic level.
"The scientific community yet has to fully understand the genetic
basis of this switch between non-invasive and invasive."