Queensland researchers believe a protein found in up to half of all aggressive brain cancers could explain why recurrence rates are so high.
Around 1000 Australians die every year from glioblastoma multiforme (GBM) tumours, the most common form of brain cancer in adults.
Sufferers have an average survival rate of 12 to 15 months, and despite surgical removal of tumours, chemotherapy and radiation treatments, the cancer almost always returns.
"Long-term survival rate is about 2 to 3 per cent of people two years and beyond," said Dr Bryan Day from the Queensland Institute of Medical Research's brain cancer research centre.
"The outcome at present really is quite terrible for people."
Institute team leader, Professor Andrew Boyd, discovered the EphA3 protein in leukaemia cells in 1992, and he and Dr Day have proved it is present in up to 50 per cent of GBM cases.
"What we've shown is that this protein that sits on the cell surface of brain tumours is essentially responsible for making them more likely to be tumorigenic," said Dr Day.
Tumorigenic means capable of or tending to form tumours.
Dr Day said cells with the EphA3 protein seem to stay in the brain despite surgery and chemotherapy treatments, meaning they build a resistance.
"There is a mixed population of cells in any tumour, and some will be sensitive [to treatments], some won't, and the ones that are resistant tend to be more aggressive," said Dr Day.
"These treatments in a way actually kill the easy-to-kill tumour cells, leaving a purified population of aggressive tumour cells, that when they come back, are highly aggressive, highly infiltrative and usually fatal."
The QIMR team said having this target meant they could investigate using an already-existing antibody called IIIA4 to turn the EphA3 cells in on themselves, destroying their ability to promote cancerous cell growth.
So far, trials for its use in leukaemia treatment have proved positive.
"Essentially it's been proven as safe for use in humans," said Dr Day.
"We're hopeful within two or three years of conducting a clinical trial – if we were starting from scratch it would probably be a decade or so."