Van Houtte P. Institut Jules Bordet Université Libre Bruxelles
Brain metastases are a major problem in the management of small cell lung cancer with an incidence rising to 50% at 2 years. PCI has been showed through randomized trials and meta-analysis to decrease the incidence of brain metastases: the landmark analysis of Auperin showed a 25% reduction in brain metastases leading to a 5 % survival benefit at 3 years for patients in complete response to initial therapy. The major concern with PCI is the risk of inducing neurologic symptoms especially cognitive functions and an impaired quality of life. To-day, the question is MRI monitoring and surveillance an alternative to PCI. We should remember some main limitations of the different randomized trials: CNS staging was mainly performed by CT and not MRI and radiosurgery was not so common as to-day. Furthermore, all trials did not require a brain staging before the PCI randomization. MRI was showed to be more efficient in detecting brain metastases compared to CT both at the initial diagnosis and in the follow-up of patients. At the time of brain relapse, the classical treatment was a whole brain radiation and to-day, there are data supporting the use of radiosurgery in selected patients. There is no randomized trial comparing MRI surveillance versus PCI but some series suggesting the feasibility of such an approach challenging the PCI dogma. Quality of life and an economic analysis should be part of phase III trial.