RAPid SimPLE Targeted Radiation Treatment for Brain Metastases

The aim of the study is to show that rapid, simple targeted radiotherapy to brain metastases with 8 Gy / 1 is non-inferior to 20 Gy / 5 in terms of overall survival for patients with poor prognosis.

Hypothesis For selected patients with limited life expectancy, a single targeted fraction of 8 Gy / 1 to brain metastases will provide equivalent survival as targeted radiotherapy with 20 Gy / 5 to brain metastases. Justification The QUARTZ study randomized patients with brain metastases and poor prognosis between 20 Gy / 5 whole brain radiotherapy (WBRT) and best supportive care. The median survival in both arms of the study was 9 weeks. The quality of life in both arm was also the same. It is known that WBRT significantly diminishes quality of life. The investigators interpretation of the QUARTZ trial is that the unchecked growth of brain metastases in the best supportive care arm diminishes quality of life as much as WBRT. To control brain metastases without giving WBRT, targeted treatment with 20 Gy / 5 to the metastases alone is standard care at BC Cancer. Numerous randomized trials have compared the efficacy of 20-30 Gy / 5-10 fractions and 8 Gy / 1 for palliative treatment of pain from bone metastases. In addition, there is a randomized trial showing that 20 Gy / 5 and 8 Gy /1 are equally effective for the treatment of symptoms from spinal cord compression. This study will test the use of this shorter, one-visit treatment schedule against the standard 5-visit treatment schedule for patients with brain metastases. Research Design Randomized Phase II Non-Inferiority Study. The estimated median survival for this cohort, based upon a prior cohort study from 2012-2016 at BC Cancer is 3 months (13 weeks). The investigators would regard a study median survival of less than the 9-week median survival observed in the QUARTZ trial as unacceptable. Hence, the hazard ratio for the experimental arm and the control arm is: 13 weeks / 9 weeks = 1.44. With one-sided Type I error set at alpha = 0.2 and power = 0.8, the investigators calculated a theoretical sample size of 86 patients. Based on prior experience with clinical trials for patients with brain metastases, a 15% risk of drop-out and loss to follow up is expected. Hence the final sample size will be 100 patients. Stratification by the diagnosis-specific graded prognostic assessment (DS-GPA) ranges of 0.0 - 1.0 and 1.5 - 2.0 and lung cancer versus other histologies will help to ensure baseline characteristics that predict for equal survival are equally distributed in both arms of the study. Statistical Analysis The primary endpoint is overall survival. The study will be considered to be a positive phase II non-inferiority study if the median survival in the experimental arm is within 4 weeks of the median survival in the standard arm.