An Exploratory Analysis of the Conditional Neurocognitive Function Failure Risk in Patients Receiving Whole Brain Radiation Therapy for Brain Metastases on NRG Oncology CC001.

Abstract

Purpose: We sought to estimate the conditional risk of development of neurocognitive function failure (NCFF) after whole brain radiation therapy (WBRT) for patients with brain metastases on NRG Oncology CC001. In addition, we aimed to determine if factors prognostic of NCFF at time of treatment remained relevant over time.

Methods and materials: We performed a post hoc analysis of 518 patients enrolled on NRG-CC001 in which patients with brain metastases were randomly assigned to WBRT + memantine or hippocampal avoidant (HA-WBRT) + memantine. Life table method was used to calculate conditional monthly hazard rates and cumulative incidence was used to estimate rates of NCFF. Risk factors associated with NCFF were analyzed using cause-specific multivariable Cox proportional hazards modeling.

Results: The cumulative risk of development of NCFF by 6 months was 64.0% for the entire cohort. The greatest conditional monthly hazard rate of development of neurocognitive toxicity was 2 to 3 months postradiation (0.97; 95% CI, 0.85-1.10); this rate significantly declined and then plateaued to 0.036 (95% CI, 0-0.11) by 8 months posttreatment. For 2-month survivorship without cognitive failure, HA-WBRT (HR, 0.74; P = .033) and age ≤61 years (HR, 0.62; P = .003) continued to be protective against cognitive toxicity. In addition, conditional cumulative incidence of development of NCFF was significantly reduced with HA techniques for patients living ≥2 months free of cognitive dysfunction (P = .047).

Conclusions: Our data highlight that the greatest risk of development of neurocognitive toxicity is within the first 3 months after treatment, and therefore strategies to mitigate toxicities should focus on this initial period. Moreover, the conditional risk of neurocognitive impairment significantly declines the longer patients live with preserved cognition. Importantly, these data can be used to inform patients on how their risks of development of NCFF can change over time.