359P - The standard and high dose chemotherapy dose intensity and their effect on the survival of medulloblastoma using nation-wide treatment protocol

Date 10 September 2017
Event ESMO 2017 Congress
Session Poster display session
Topics Cytotoxic agents
Central Nervous System Malignancies
Biological therapy
Presenter Ju Yeon Lim
Citation Annals of Oncology (2017) 28 (suppl_5): v109-v121. 10.1093/annonc/mdx366
Authors J.Y. Lim1, L.M. Cha2, S.M. Hahn3, C. Suh4, D. Kim5, C.J. Lyu3, J.W. Han1
  • 1Department Of Pediatric Hemato-oncology, Yonsei Cancer Center, 03722 - SEOUL/KR
  • 2Department Of Pediatric Dentistry, Yonsei University College of Medicine, 03722 - SEOUL/KR
  • 3Department Of Pediatric Hemato-oncology, Yonsei Cancer Center, SEOUL/KR
  • 4Department Of Radiation Oncology, Yonsei Cancer Center, 03722 - SEOUL/KR
  • 5Department Of Neurosurgery, Yonsei University College of Medicine, 03722 - SEOUL/KR



Medulloblastoma is the most common type of childhood brain tumors. We conducted Korea’s nation-wide protocol based treatment for medulloblastoma, and adopted tandem high dose chemotherapy for high risk disease. Here we present the result of treatment in Yonsei Cancer Center using the protocol and elucidate dose-response relationship.


The patient diagnosed and treated in Yonsei Cancer Center were reviewed retrospectively, from 2006 to 2015. We excluded the patients less than 3 years old and over 30 years old. Dose intensity (DI) was calculated as actual dose level/planned dose level divided by chemotherapy treatment duration. Ind-DI was defined as induction chemotherapy DI and HDCT-DI was as high dose chemotherapy DI. The protocol was composed of 2 cycles of neoadjuvant chemotherapy and 32.4Gy of craniospinal radiotherapy (CSRT) and 50.4 Gy of total tumor dose. After the radiotherapy, 4 cycles of chemotherapy and tandem high dose chemotherapy was done.


Among total 39 patients, 16 were standard risk (SR) and 23 were high risk (HR). The 5 year overall survival (OS) was 92% for SR, and 67% for HR. Disease specific survival (DSS) for HR was 75%, and therefore 8% was treatment related mortality (TRM). The 5Y OS between M0 and M1 status was not statistically different in HR. The ind-DI did not affect in 5Y OS for SR and HR. Ind-DI was strongly correlated with HDCT-DI. The 5Y-OS for HDCT-DI70%. All TRMs developed in HDCT-DI>90%. Therefore, the reduction of HDCT-DI in the protocol was suggested from this analysis. The responsiveness of induction chemotherapy (complete remission and partial remission) was predictive marker for 5Y-OS (P = 0.026). The complete response at the first HDCT was also predictive for 5Y-OS (P = 0.009).


The result from the nation-wide protocol was acceptable but high dose intensity was the cause of treatment related mortality. Now the dose of HDCT was reduced to achieve the ideal survival rate from the protocol.

Clinical trial identification

Legal entity responsible for the study

Jung Woo Han




All authors have declared no conflicts of interest.