Patterns of failure after proton therapy in medulloblastoma.

Proton therapy centers and the pediatric oncology community will welcome the article by Sethi et al (1) published in the March 1, 2014, issue of this journal. The technical data regarding overor underlapping of adjacent treatment fields are useful to prevent further tumor recurrences at these sites. Also, it is reassuring that an excess tumor recurrence risk was not found just beneath themeningesof thewidest portion of the head, including after low, highly fractionated doses of 23.4Gy (relative biological effect [RBE] Z 1.1) and below, which provides nearly 70% of the population studied. This finding does need to be confirmed in larger numbers of patients. The higher-than-expected “spine only” relapse rate found by the authors is a cause for concern. This finding is compatible with the point made previously (2) that tumor control may be reduced by adopting a tumor RBE of 1.1 if a highly radiation-sensitive tumor (with a high a/b ratio) has an RBE in the range of 1.03 to 1.06. Agreement that the RBE for these highly radiation-sensitive tumors is probably less than 1.1 represents an advance, although modeling predictions are prone to many uncertainties including variations in the input parameters as well as interpatient variations. This must be balanced with the realistic possibility that central nervous system (CNS) tissues may have an RBE of greater than 1.1, such as 1.2 (2, 3). The likely RBE values for important late reacting CNS tissue has not been estimated by experiment nor by inference from clinical data sets, and this is an urgent requirement. Only in some late reacting tissues (lung, lens, and skin) have RBEs been determined, using high doses per fraction in the 9to 12-Gy