Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial

Authors: Andrews DW, Scott CB, Sperduto PW, Flanders AE, Gaspar LE, Schell MC, Werner-Wasik M, Demas W, Ryu J, Bahary JP, Souhami L, Rotman M, Mehta MP, Curran WJ, RTOG
Source:Lancet 2004 May 22; 363: 1665-72

Background

As many as 20-40% of cancer patients with widespread disease (ie: metastatic or systemic disease) have brain metastases. Of these, as many as 30-40% will have just a solitary brain lesion. In general, the prognosis for such patients is poor, and the median survival time on steroid therapy alone is 1 to 2 months. The administration of whole brain radiation therapy (WBRT) to these patients can extend this survival time to about 6 months. Performing neurosurgical resection of lesions before radiation therapy whenever feasible and practical, (ie: cases of one or two surgically approachable masses), can improve this survival even further. This particular study was launched to investigate the combination of WBRT with stereotactic radiosurgery (SRS), which delivers a single fraction of precise, high-dose radiation to intracranial targets using either Gamma Knife or linear accelerator-based systems. Specifically, the authors evaluated the efficacy of WBRT + SRS in prolonging the survival of patients with three or fewer brain metastases.

Materials and Methods

• 55 participating RTOG institutions
• All patients had confirmed systemic malignant disease and Karnofsky performance score > 70
• Eligible patients had a contrast-enhanced MRI scan with one to three brain metastases, with the largest lesion measuring < 4 cm and all other lesions (if any) < 3 cm
• No prior cranial radiation was allowed
• Both unresectable and postoperative patients with residual or distant brain lesions were eligible, as long as the total number of brain lesions was no more than three
• Excluded were patients with: metastases in the brainstem or close to the optic apparatus; systemic treatment within one month of study enrollment
• Patients with newly diagnosed disease and unknown primary sites of cancer were allowed
• Of 333 eligible patients enrolled from 1/1996 to 6/2001, 331 were randomized to one of two treatment arms:
• • 164 to arm 1 = WBRT + SRS
  • 167 to arm 2 = WBRT alone
• Patients were stratified by number of brain metastases (1 vs. 2-3) and presence of extracranial disease (none vs. yes)
• Patients were classified as either RPA class I (controlled at site of primary disease, < 65 years old, brain only site of metastatic disease) or class II (uncontrolled primary site, age > 65 years old, brain plus other sites)
• • RPA classes evenly distributed between two arms
• WBRT consisted of 250 cGy daily for total of 3750 cGy in 3 weeks
• SRS was performed within one week of finishing WBRT
• SRS doses were based on tumor size (as per RTOG 90-05):
  • up to 2 cm = 24 Gy
  • 2 cm but < 3 cm = 18 Gy
  • 3 cm but < 4 cm = 15 Gy
• On arm 1, 133/164 completed treatment
• On arm 2, all 167 completed treatment
• Primary study endpoint was overall survival, secondary endpoints included tumor response, local control rates, and performance status measurements
• Designed to detect 50% improvement in mean survival time for patients getting SRS boost
• Intent-to-treat analysis

Results

All results are WBRT + SRS vs. WBRT alone:

• Mean survival time = 6.5 mos vs. 5.7 mos, p=0.1356
• Univariate analyses revealed only one subgroup patients with single lesions that had statistically significant benefit:
• • Patients with single metastases = 6.5 mos vs. 4.9 mos, p=0.039
  • Patients with multiple metastases = 5.8 mos vs. 6.7 mos, p=0.9776
  • All other subgroups showed no difference on univariate
• Multivariate analysis found RPA class 1 (p<0.0001) and lung tumor primary (p=0.0121) to be significant predictors of favorable outcome
• Performance scores were significantly better at 6 months in the SRS group
• No significant differences in early and late toxicities or neurologic deaths
• Greater radiographic tumor response (p=0.0438) and local control rates (p=0.0132) in SRS group

Discussion

This is the first prospective randomized multicenter trial to examine this important clinical question. The findings of this study demonstrate that the use of whole brain radiation followed by stereotactic radiosurgery boost to the intracranial target results in longer mean survival time for patients with solitary unresectable lesions, compared to whole brain radiation alone. This applies not only to nonsurgical patients, but also to postoperative patients with a single area of residual or inoperable disease. The radiosurgery was well tolerated by patients with no discernible difference in morbidity or mortality between the two arms. Importantly, the addition of SRS after WBRT significantly improved the performance scores of patients in arm 1 compared to arm 2, regardless of number of lesions. Thus, the authors recommend that the addition of stereotactic radiation boost to cranial irradiation should become the standard of care for patients with solitary brain metastases, and be strongly considered in the management of those with two to three lesions.