Analysis of Late Cognitive Outcomes after Whole Brain Radiotherapy and Radiosurgery for Brain Metastases in the North Central Cancer Treatment Group (NCCTG) N0574 and N107C Clinical Trials.

To investigate predictors of cognitive outcomes after radiotherapy for brain metastases in clinical trials that randomized patients between whole brain radiotherapy (WBRT) and stereotactic radiosurgery (SRS). The trials' cognitive assessments consisted of the three components of the Hopkins Verbal Learning Test – Revised, Trail-Making Tests A and B, and Controlled Oral Word Association Test administered at standardized times. The trial analyses assumed that missed cognitive assessments and declines of one standard deviation on a single test constituted cognitive events. We wished to investigate predictors of late cognitive decline after the subacute period of reversible encephalopathy. The N0574 trial enrolled 213 participants with up to three brain metastases and the N107C trial enrolled 194 participants with a surgical cavity and up to three brain metastases. Eligible patients for our analysis had a baseline and at least one follow-up cognitive assessment six months or more after radiotherapy. The cognitive endpoint was the difference in the mean Z-scores of all cognitive tests between the baseline and last cognitive assessments. The baseline variables were age, trial, number of brain metastases and SRS versus WBRT. In-brain recurrences before the last cognitive assessment were also included in the linear regression model. The percentage of participants with a baseline and at least one follow-up cognitive panel at 6 months or later was 16.0% (34/213) for N0574, 54.6% (106/194) for N107C, and 34.4% (140/407) for the combined cohort. In the combined cohort, the distribution of last cognitive panels was: 21.4%, 12.1%, 22.1%, 28.6% and 15.7% at 6, 9, 12, 15 and 18 months, respectively. In the model, receipt of WBRT (β = -0.25 [95% CI: -1.60, -0.59], p<0.001) was the only significant predictor of cognitive decline. Age (p = 0.06), trial (p = 0.07), number of brain metastases (p = 0.7), and in-brain recurrence (p = 0.2) did not predict late cognitive outcomes. When considering participants' last cognitive assessments, only receipt of WBRT predicted late cognitive decline in the N0574 and N107C clinical trials. [ABSTRACT FROM AUTHOR]