Randomized Prospective Comparison of Stereotactic Radiosurgery (SRS) Followed by Conventional Radiotherapy (RT) with BCNU to RT with BCNU Alone for Selected Patients with Supratentorial Glioblastoma Multiforme (GBM): Report of RTOG 93-05 Protocol.

Reviewer: Heather Jones, MD
Last Modified: October 9, 2002

Presenter: L. Souhami
Presenter's Affiliation: McGill University, Montreal, QC, Canada
Type of Session: Scientific


    GBM continues to prove a most challenging tumor to treat. In spite of aggressive therapy local recurrence remains a significant problem. RTOG 9305 is a prospective randomized trial designed to evaluate upfront SRS followed by RT with BCNU compared to RT with BCNU. This study was designed to evaluate survival, patterns of failure, toxicity and quality of life between the two arms.

Materials and Methods

  • Patients were enrolled between February 1994 and September 2000; 186 pts were available for analysis in this phase III trial.
  • Eligible pts had histological proof of a unifocal (less than or equal to 4 cm) GBM.
  • Treatment Arms Pts were randomly assigned to receive: Arm 1) 60 Gy/ 30 fractions/2 Gy once daily plus BCNU 80 mg/m2 days 1, 2, 3 of RT then q 8 weeks for a total of 6 cycles (Arm 1), Arm 2 SRS (15-24 Gy) followed by the same RT and BCNU (Arm 2).
  • The two arms were well balanced for the following pretreatment patient characteristics: age, KPS, RTOG RPA class and pre/post surgery tumor size
  • The Spitzer Quality of Life index was used to measure general quality of life (QOL) and the Mini Mental Status Examination (MMSE) was used for cognitive functioning.


  • The median follow-up time at analysis was 44 months,
  • The median survival for Arm 1 was 14.1 months and 13.7 months for arm 2, p = 0.5328.
  • The 2-year survival rate was for Arm 1 22% and 16% for Arm 2
  • The 3-year survival rate was 18 % for Arm 1 and 8% for Arm 2
  • There was no difference in survival between the arms when analyzed by RTOG RPA classes III or IV (median survivals of 14.7 and 14.2 for Arms 1 and 2, respectively).
  • The was no difference in overall survival between patients treated with Gamma Knife or Linac radiosurgery
  • Compliance was very good for conventional RT and chemotherapy, 18% of the SRS pts had unacceptable deviations.
  • The patterns of failure were similar between the arms with more than 90% of the pts presenting a component of local failure.
  • QOL deterioration (Spitzer index) at the end of therapy, compared to baseline, was similar between the arms (40% on Arm 1 and 50% on Arm 2, p = 0.4606).
  • There was no difference in quality-adjusted survival between the two arms.

Author's Conclusions

    This trial demonstrated that the use of upfront SRS followed by RT and BCNU does not lead to an improved survival or changes in patterns of failure for patients with supratentorial GBM. There was no difference in general QOL and cognitive functioning between the arms.

Clinical/Scientific Implications

    Glioblastoma Multiforme (GBM) is associated with a high rate of local recurrence after primary therapy and a high mortality rate. In spite of aggressive treatment with surgery, radiotherapy and chemotherapy, the median reported survival is less then 1 year. One of the major hindrances to increasing the dose of radiation is normal brain tissue tolerance. It was hoped that SRS would provide a safe way to dose escalate, spare normal tissue and result in improved outcomes. Unfortunately, this would appear not to be the case, as RTOG 9305 is a negative study; with indications that the no SRS arm might have a slightly better outcome in terms of survival rates in this population of patients.

Oncolink's ASTRO Coverage made possible by an unrestricted Educational Grant from Ortho Biotech.