Drugs that prevent the development of treatment-resistant prostate cancer may be in prospect following the discovery of a new protein.
The molecule, GPR158, is linked to a biological process that plays a critical role in the way the disease stops responding to standard hormone therapies.
Patients with raised levels of the protein were more likely to experience a recurrence of prostate cancer, said scientists.
They believe GPR158 could provide a target for new prostate cancer drugs.
The molecule belongs to a family of cell-surface proteins called G-protein coupled receptors (GPCRs) that promote prostate cancer growth.
It was discovered fortuitously while researchers were searching for new drug targets for the eye disease glaucoma.
Each year around 40,000 men are diagnosed with prostate cancer in the UK and 10,000 die from the disease.
Prostate cancer starts off responding to drugs that prevent male hormones fuelling tumour growth. But eventually it becomes resistant to this form of treatment after which progress to death is often rapid.
US lead scientist Dr Nitin Patel, from the University of Southern California, said: "When a prostate cancer tumour is in its early stages, it depends on hormones called androgens to grow.
"Eventually it progresses to a more lethal form, called castration-resistant prostate cancer (CRPC), and is resistant to drugs that block androgen receptors. We found that GPR158, unlike other members of the GPCR family, is stimulated by androgens, which in turn stimulates androgen receptor expression, leading to tumour growth."
The team also found that GPR158 is associated with a process called neuroendocrine transdifferentiation (NED) which plays a critical role in the development of hormone therapy-resistant cancer.
The research is reported in the online journal Public Library of Science ONE.
Dr Matthew Hobbs, deputy director of research at Prostate Cancer UK, said: "There's still so much we don't know about advanced prostate cancer. Working out why the disease stops responding to treatment over time is one of the big unanswered questions.
"Finding these answers could hold the key to developing new treatments to save thousands of men dying from prostate cancer every year.
"The findings revealed in this paper provide us with another clue, but we've still got a long way to go."