A Phase III Prospective Randomized Trial Comparing Radiosurgery With Versus Without Whole Brain Radiotherapy for 1-3 Newly Diagnosed Brain Metastases

Reviewer: Arpi Thukral, MD
Last Modified: September 23, 2008

Presenter: Eric Chang, MD
Presenter's Affiliation: MD Anderson Cancer Center, Texas
Type of Session: Plenary

  • Brain metastases occur in 20% to 40% of all patients with cancer, and there are approximately 170,000 new cases of brain metastases per year reported in the United States. They generally are associated with a poor prognosis and can lead to a poor quality of life.
  • Whole brain radiation therapy (WBRT) has been a mainstay oftreatment for patients with multiple brain metastases. Although the acute adverse effects of WBRT are generally limited, the long-term side effects associated with neurocognitive decline are serious and likely permanent. To reduce morbidity associated with WBRT, treatments options which are more focal are being sought for use.
  • Recently, stereotactic radiosurgery (SRS), which delivers high-dose, focal radiation, has become a more frequently used treatment for those patients who have limited intracranial disease (1-3 brain metastases). SRS is also commonly used in combination with WBRT in an attempt to treat both visible tumor and potential micrometastases.
  • There has been an ongoing debate in the literature as to whether immediate whole brain radiaiton following SRS is indicated for patients with 1-3 brain metastases. Those who argue for WBRT claim that it improves disease control; those who argue against it state that it may cause inferior neurocognitive functioning in the long run.
  • The long-term risks of serious and permanent toxic effects of the addition of WBRT, including cognitive deterioration and cerebellar dysfunction, are poorly understood.
  • The purpose of this study was to compare survival and rates of local and distant brain recurrence in a prospective randomized trial of patients with 1-3 brain metastases treated with either radiosurgery and whole brain radiation therapy or radiosurgery alone. An additional major goal of the study was to formally assess and compare neurocognitive changes and quality of life between the two treatment groups.
Materials and Methods
  • This is a phase III prospective trial where patients with 1-3 brain metastases who received SRS as initial treatment were randomized to immediate WBRT (within 3 weeks of SRS) versus observation,
  • The primary endpoint evaluated was cognitive decline in learning and memory.
  • Secondary endpoints included overall survival and local and distant brain recurrence rates.
  • Patients were stratified by: 1) RPA class I vs. II; 2) 1 or 2 metastases vs. 3 metastases; and 3) histology.
  • Inclusion Criteria:
    • Age 18 years and older at time of registration
    • Recursive partitioning class (RPA) I or II
    • 1 to 3 newly diagnosed brain metastases
    • Must be eligible to have all lesions treated by SRS as determined by the radiation oncologist on the basis of location and size.
    • MRI with contrast must be performed within 1 month of registration
  • Exclusion Criteria
    • Prior WBRT
    • Prior resection of brain metastasis, SRS or Gamma knife
    • > 3 brain metastases present on MRI
    • Lymphoma, small cell lung cancer, leukemia, germ cell tumors
    • Leptomeningeal disease
    • Unknown primary
    • RPA Class III (i.e. KPS < 70)
  • SRS was delivered according to RTOG 90-05 guidelines. WBRT was given within 3 weeks for those patients randomized to that group. Patients were given 30 Gy in 12 fractions of 2.5 Gy daily to minimize toxicity.
  • Patients who did not receive whole brain radiation and developed recurrent disease were to receive treatment based on the number of lesions.
  • All patients underwent initial neurocognitive testing at baseline with the Hopkins Verbal Learning Test.
  • For the first 18 months, patients underwent neurocognitive testing at months 1, 2, 4, 6, 9, 12, 15, and 18. They then underwent testing every 9 months.
  • A Bayesian statistical design was followed and the study was planned to accrue 90 patients with an 80% power to detect a 30% difference.
  • Between January 2001 and March 31, 2008, 58 patients were enrolled on this study.
  • There were 30 patients in the SRS group and 28 patients in the SRS+WBRT group. Patient characterisitics were comparable between the 2 groups. Median follow up was 7.5 months.
  • The mean probability of neurocognitive decline in the SRS group was 23% vs. 49% in the WBRT group at approximately 4 monhts after treatment. This was significant with a 96% confidence.
  • An independent data monitoring committee stopped the trial after interim results showed the high statistical probability (96.4 percent) that patients randomized to SRS alone would continue to perform better cognitively.
  • The rate for 1-year local control for the SRS alone group was 67% vs. 100% for the WBRT group, (p=0.012). Distant brain tumor control was also significantly worse for the SRS alone group.
  • In terms of median survival, the authors report a significant survival difference of 15.2 months in the SRS group vs. 5.6 months in the WBRT group, with a p value of 0.003.
Author's Conclusions
  • The authors concluded that there is twice the risk of cognitive decline seen in patients who received WBRT vs. those who did not.
  • In addition, they state that there is a clear separation seen in terms of overall survival between the two groups, with the SRS alone group having a much higher overall survival.
  • The authors noted that this study provides level I evidence that SRS alone could be the new standard of care for patients with up to 3 brain metastases.
Clinical/Scientific Implications
  • This prospective randomized trial provides valuable information to help further settle the debate as to whether or not immediate WBRT following SRS for patients with limited brain metastases is necessary.
  • This study was well-designed. Patients were stratified well, and a verified instrument was used to testing neurocognitive decline. The groups were evenly balanced, and a sound Bayesian interim monitoring design was used.
  • Previous data and this study have shown a benefit to WBRT in terms of increased local control. However, the results for the association between WBRT and neurocongitive decline differ from previous studies.
  • In a recent similar paper published by Aoyama et. al. in 2006, the authors sought to determine if WBRT combined with SRS results in improvements in survival, brain tumor control, functional preservation rate, and frequency of neurologic death. They concluded that the addition of WBRT improves brain tumor control and increases the time to neurocognitive (NC) decline (16.5 mo until NC decline in the WBRT group vs. 7.6 months in the SRS alone group). Brain tumor control was found to be the most important factor to reduce neurocognitive decline.
  • With the conflicting results of these two studies, it is unclear at this time which factor - CNS tumor recurrence or the toxicity of WBRT - has a worse impact on NC decline.
  • The authors of this study also noted a significant and large absolute survival difference between the groups, favoring the SRS alone group. These results also differ significantly from the Aoyama trial where no survival difference was seen between the groups.
    • The survival data presented by the these authors are hard to believe since a median survival of 15 months is fairly high for patients with brain metastases, and the study was not powered for a survival outcome. But, these results are also diffucult to ignore. With such a small number of patients, these data should be considered hypothesis-generating, and possible reasons for this difference need to be explored.
  • The authors also noted that 27 patients in the SRS group received salvage treatment versus only 3 patients in the WBRT group. This possible difference in aggressiveness of salvage therapy may have been a confounding factor. In addition, there was no mention by the authors of controlling for corticosteroid or chemotharpy use, which also may confound the data.
  • In conclusion, despite poorer overall brain tumor control at 4 months, the patients in this study were found to have better NC function in the SRS alone group. This data is important, but this is a small study, and should not be used to change standard of care. It still needs to be further corroborated by other data, and therefore the debate still continues. We eagerly await data from larger trials being done on this topic, like the NCCTG trial by Dr. Paul Brown.


CancerLand Bookshelf: Chief Complaint: Brain Tumor
by Carolyn Vachani, MSN, RN, AOCN
March 6, 2013

Being OK with Brain Cancer
by Bob Riter
October 22, 2015

Stay informed with the latest information from OncoLink!   Subscribe to OncoLink eNews
View our newsletter archives