Diffuse low-grade gliomas (LGG) are primary brain tumors. After a slow growth, they evolve to high-grade gliomas, resulting into death. These tumors are very diffuse, thus difficult to treat. A better knownledge of them could allow to cure them or, failing that, to optimize treatments. We studied the growth of LGG with a simple mathematical model, which led us to speculate (i) that they arise in adolescence, (ii) that the age of the tumor at diagnosis can be calculated easily, and (iii) that the growth rate is an important prognostic factor. This last prediction is consistent with clinical observations. To test this spatial model, we have quantitatively cha- racterized biopsy tissues of human LGG, particularly the presence of edema. The microscopic analysis of these data underpins the idea that edema is the cause of the abnormality seen on T2-weighted MR imaging. To take this new result into account, we have incorporated edema into the initial model as a consequence of the presence of tumor cells. This model helps explain the long decay of the tumor radius for tens of months after radiation therapy : as tumor cells become less numerous, drainage of the edema becomes predominant. This model, which has only three free parameters, has been validated thanks to clinical data from twenty patients.
