media release by Tania Ewing
For decades it has been thought that AML growth is driven by a sub-population of immature cancer cells called ‘leukaemia stem cells’, which lose their cancerous properties when they mature. Hence there has been growing international interest in developing therapies aimed at forcing immature cancer cells to ‘grow up’. Using genetically engineered mouse models and human AML cells, Associate Professor Dickins and his Monash-led team have found that maturation of AML cells is not unidirectional as originally thought, but can instead be reversible.
Each year in Australia over 1000 people are
diagnosed with acute myeloid leukaemia (AML), an aggressive blood cancer. Less
than one third of AML patients survive 5 years beyond diagnosis. Researchers
from Monash University have discovered a key reason why this disease is so
difficult to treat and therefore cure.
The study, led by Associate Professor Ross Dickins from the Australian Centre for Blood Diseases, is published today(TBC) in the prestigious journal Cell Stem Cell. The paper identifies an
important new concept relevant to clinicians involved in the diagnosis and
treatment of AML patients.
AML is characterised by an overproduction of immature
white blood cells that fail to mature properly. These leukaemia cells crowd the
bone marrow, preventing it from making normal blood cells. In turn this causes
anaemia, infections, and if untreated, death. Acute myeloid leukemia (AML)
remains a significant health problem, with poor outcomes despite chemotherapy
and stem cell transplantation.
For decades it has been thought that AML growth is driven by a sub-population of immature cancer cells called ‘leukaemia stem cells’, which lose their cancerous properties when they mature. Hence there has been growing international interest in developing therapies aimed at forcing immature cancer cells to ‘grow up’. Using genetically engineered mouse models and human AML cells, Associate Professor Dickins and his Monash-led team have found that maturation of AML cells is not unidirectional as originally thought, but can instead be reversible.
The team, which also includes researchers from
the Walter and Eliza Hall Institute of Medical Research and several international
collaborators, found that even mature AML cells can ‘turn back the clock’ to
become immature again. This plasticity means that even mature AML cells can
make a major contribution to future leukaemia progression and therapy
resistance. The discovery has significant implications for the way that AML is
treated, according to Associate Professor Dickins.
“The AML field has
traditionally accepted a model where leukemia maturation is a one-way street,”
he said. “By demonstrating reversible leukaemia maturation, our study raises
doubts around therapeutic strategies that specifically target just leukaemia
stem cells. It highlights the need to eradicate all tumour cells irrespective
of maturation state.”
These new findings are important because several new drugs
that force leukaemia cells to mature have now entered the clinic. The study by
Dickins and colleagues will help clinicians re-shape their thinking to find
ways to eradicate mature leukaemia cells along with their immature
counterparts.
No comments:
Post a Comment
Thankyou for your comment. We moderate all messages and may take a little time to review your comment. Please email inquiries to ccs.comms@monash.edu.