Blocking a specific enzyme could be enough to check the spread of cancer in the human body, researchers say.
Tests were carried out on genetically-modified mice
Researchers believe their work may lead to new life-saving cancer treatments and help cut the amount of chemotherapy required to treat cancer effectively.
The spread of cancer was blocked in six out of seven mice bred to not have the enzyme urokinase plasminogen activator, and the mice did not suffer without it.
The study, by Copenhagen University, is in the International Journal of Cancer.
Cancer cannot spread unless uPA is present, but the body does not need the enzyme to function normally.
This was confirmed by the latest research, which showed mice genetically modified to lack uPA did not suffer in any way from not having the enzyme.
Researcher Dr Morten Johnsen said: "This means that we should be able to block the enzyme and thereby check the spread of the cancer without causing the strong side-effects for the patients that we see with other forms of therapy today."
The next step is to develop a medication that works on mice. Only then will testing the medication on humans be considered.
Dr Torben Skovsgaard, a cancer specialist at Copenhagen's KAS Herlev Hospital, said the research looked "very promising".
He said: "If the clinical trials also show that it is possible to halt the spread of the tumour, then the research team has found the key to stopping the spread of cancer in the body, which is the cause of most of the deaths among cancer patients today."
Kate Law, head of clinical trials at Cancer Research UK, said: "It is true that death from cancer is largely a result of the cancer spreading from the original site to other parts of the body.
"Any therapy that prevented this spread would be a major step forward in the treatment of cancer.
"Long experience should, however, make us cautious in using encouraging laboratory research to begin talking about patient benefit.
"It takes many years for any finding in the laboratory to translate into patient benefit and results in the past have often been disappointing."