Local delivery of mitoxantrone in recurrent glioblastoma multiforme patients: Is there any by benefit?

A. Boiardi, A. Silvani, G. Broggi

Research output: Contribution to journalArticlepeer-review


The outcome of recurrent glioblastoma multiforme (GBM) patients is disappointing, rarely exceeding six months. In the effort to better control local recurrence of GBM, we have systemically repeated local injections of mitoxantrone through an Ommaya/Rickham reservoir positioned into the surgically created cavity. 276 recurrent GBL patients managed at the neurological Institute between 2003 and 2005 entered the study. The whole group of patients was homogeneous being treated at first diagnosis after surgery with chemotherapy (CDDP + BCNU) associated with external beam conformational radiotherapy, total dose 60 Gy. At tumor recurrence, all patients had been treated with temozolomide until tumor evolution. Patients with tumor respectable without risk of adjunctive neurological deficit, underwent a second surgery with or without positioning of Rickham/Ommaya reservoir. 276 recurrent GBM have been divided into three subgroups: A. 161 patients treated only with systemic II line chemotherapy (CHT); B. 50 patients reoperated-on + systemic CHT; C. 65 patients reoperated on + local CHT through Ommaya reservoir + systemic CHT. After reoperation the area of surgically created cavity with residual tumor mass no larger than 3 cm was identified in 32/50 patients with positioned Ommaya reservoir. Results: A. six-month progression free survival (6mPFS) 43%; six-month survival time (6mST) 39.3%; median survival time (mST) 5 months (4-6) and 25% of patients 9 months; B. 6mPFS 64.1%; 6mST 64%; mST 8 months (6-10) and 25% of patients 12 months; C. 6mPFS 70.7% ; 6mST 87.7%; mST 11 months (9-13) and 25% of patients 18 months; A. vs B. vs C. log-rank (p-value <0.001); B. vs. C. log-rank test = 0.041. In a multivariate analysis the weight of the different therapeutic approaches was evaluated according to age and the duration of PFS before recurrence of the tumor. Second tumor debulking was effective on survival reducing of 36% the risk of death (Hazard Ratio; HR = 0.64; 0.46-0.89), but the most significant favorable prognostic factor on survival was the local delivery of mitoxantrone which reduced the risk of death to 50% (HR = 0.50; 0.38-0.68) Data were matched with the group treated only with systemic chemotherapy. Toxicity: hematological toxicity (Eastern Cooperative Oncology Group) grade 3-4 leucopenia occurred in 8.6 and 6.5% respectively; grade 3-4 thrombocytopenia in 10.2 and 7.9% respectively. The impact of neurological toxicity was very low in group: short-lived partial Jackson seizures in 18.5%; headache and light drowsiness lasting no more than 6 hours in 7.6%; transient worsening in a focal deficit for 3-4 days in 6.1%; local infection was the most important side effects in 12.3% of cases treated by locoregional CHT. Conclusion: repeated local delivery of mitoxantrone seems to be effective in controlling local recurrence of the tumor. Results are encouraging but we are well aware that they need to be confirmed in a randomized study.

Original languageEnglish
Pages (from-to)87-96
Number of pages10
JournalRivista Medica
Issue number4
Publication statusPublished - Dec 2007


  • Locoregional chemotherapy
  • Mitoxantrone
  • Ommaya reservoir
  • Recurrent glioblastoma
  • Rickham reservoir

ASJC Scopus subject areas

  • Medicine(all)

Fingerprint Dive into the research topics of 'Local delivery of mitoxantrone in recurrent glioblastoma multiforme patients: Is there any by benefit?'. Together they form a unique fingerprint.

Cite this