LAUSR

We read with great interest the recently published article entitled “Patterns of failure in pediatric medulloblastoma and implications for hippocampal sparing” by Baliga et al. The study provides valuable insights into the patterns of failure in pediatric medulloblastoma, a common type of brain tumor that affects children. The authors also revealed that hippocampal‐avoidance (HA) volumetric‐modulated arc therapy and HA intensity‐modulated proton therapy techniques to reduce dose to the hippocampus are dosimetrically feasible. We congratulate the authors for their contribution to this scarce field. However, we would like to highlight some points of concern. Because of the close proximity of the hippocampus to the lateral ventricles, hippocampus sparing may result in under‐coverage of the whole ventricular system. As the authors noted, this means a significant shortfall for achieving cerebrospinal fluid sterilization. The authors noted that they set the dose constraints to the hippocampus according to Radiation Therapy Oncology Group (RTOG) trial 0933 but allowed higher doses to maintain tumor coverage. As a result of this tolerance, their results showed that both HA volumetric‐ modulated arc therapy and HA intensity‐modulated proton therapy plans were associated with significantly reduced mean dose to hippocamps compared with the standard non‐HA plans, but still above RTOG 0933 recommendations (i.e., D100% >9 grays[Gy]). Meanwhile, these mean dose levels were calculated in the standard‐ risk patients who received relatively low‐dose craniospinal irradiation (23.4 vs. 36 Gy). It is also reported that there is a remarkable reduction in the percentage of the ventricular volume receiving the craniospinal prescription dose. As such, it can be interpreted that the effort to lower the dose to the hippocampus is futile. Furthermore, it seems unrealistic to assume that an inadequate hippocampal dose reduction will have a clinical contribution. Although isolated perihippocampal recurrences reportedly have occurred at a low rate, it should be discussed how rational it would be to waive cerebrospinal fluid sterilization for the likely clinically meaningless expectation. Although several publications suggest less strict dose constraints for the hippocamps than the RTOG 0933 study, similar handicaps exist based on them. Taken all together, HA techniques may be useful in standard‐risk patients who need low‐dose craniospinal irradiation; however, especially in high‐risk patients who need a high dose of 36 Gy, HA seems risky within the current technological possibilities. Another point we wanted to criticize in this study is the reported pattern of failure. In the current study, no patient experienced hippocampal recurrence, and nonisolated perihippocampal relapse was observed in two patients (8%). In contrast to this finding, in our series, there were patients who had isolated or nonisolated