Identification of an optimal subgroup for treatment evaluation of patients with brain metastases using rtog study 7916

The overall poor prognosis of brain metastases patients has complicated the evaluation of treatment effectiveness in previous clinical trials involving radiation therapy. Therapy has not been seen to alter survival, which is generally short in these patients. Possible benefits of the treatments tested may be better assessed using a favorable group of patients who are at lower risk of dying quickly from cancer. The determination of a patient subgroup having prolonged survival allows for improvement in the design and analysis of subsequent clinical trials. An optimal patient group was identified in an RTOG study (7916) that evaluated two fractionation schedules (30 Gy/10 fractions/2 weeks and 20 Gy/6 fractions/3 weeks) with or without the administration of misonidazole (MISO) in the treatment of brain metastases. A Cox regression model was used to identify the pretreatment characteristics associated with a favorable prognosis for survival: Karnofsky Performance Status (KPS) of 70-100, an absent/ controlled primary tumor, age < 60 years, and metastatic spread limited to the brain. A logistic model confirmed that the odds of surviving at least 200 days depend on these pretreatment characteristics. Patients with all four favorable characteristics constitute 11% of the evaluable study population and have a predicted 200 day survival of 52%. Prognostically favorable subgroups have been identified as patients having at least three of these four favorable characteristics. These patients have predicted probabilities of 200 day survival between 33 and 52%. Conversely, unfavorable subgroups are defined as patients having two or less favorable characteristics. Subsequent verification of these results by a second data set is warranted. The prognostically favorable characteristics have been used to define the patient population in a current RTOG study evaluating accelerated radiation therapy in patients with brain metastases.