Quality of Life Is Independently Associated With Neurocognitive Function in Patients With Brain Tumors: Analysis of a Prospective Clinical Trial

Abstract

Purpose: We conducted the first prospective longitudinal study examining the independent association between patient-reported health-related quality of life (hrQoL) (physical, social/family, emotional, functional, and brain cancer-specific) and neurocognitive function (NCF), while controlling for mood symptoms in patients with primary brain tumors.

Methods and materials: Patients with primary brain tumors (n = 59) receiving brain radiation therapy underwent hrQOL (Functional Assessment of Cancer Therapy-Brain), mood (Beck Depression and Anxiety Inventories), and neurocognitive evaluation at baseline and 3, 6, and 12 months postradiation therapy in a prospective clinical trial. Neurocognitive assessments measured attention/processing speed, memory, and executive function, including the Delis-Kaplan Executive Function System Verbal Fluency, Hopkins Verbal Learning Test Revised (HVLT-R), and Brief Visuospatial Memory Test. Subjects underwent neurocognitive, mood, and hrQoL assessments in the same testing session. Multivariable linear mixed-effects models assessed associations between hrQOL and NCF over time, controlling for patient, tumor, and treatment characteristics as well as timepoint-specific patient-reported mood (ie, anxiety and depression symptoms). P values were adjusted for multiple comparisons.

Results: Higher physical hrQoL was associated with better verbal memory (HVLT-R Total Recall, P = .047), and higher functional hrQoL was associated with better executive function (Delis-Kaplan Executive Function System Verbal Fluency Switching Total, P = .009) and verbal memory (HVLT-R Delayed Recall, P = .006). Higher brain tumor-specific hrQoL was associated with better verbal and nonverbal memory (HVLT-R Total, P = .004 and Delayed Recall, P = .030; Brief Visuospatial Memory Test Total, P = .049 and Delayed Recall, P = .049). There was no association between social/family or emotional hrQoL and NCF after controlling for mood.

Conclusions: Higher physical, functional, and brain tumor-specific hrQoL were associated with better executive function and memory among patients with primary brain tumors. Physical and functional impairments are correlated with cognitive performance. Interventions to maximize quality of life after treatment may influence neurocognition and vice versa.

Figure 1.

Mean FACT-Br Total and subscale scores over time Line plots for raw FACT-Br Total and subscale scores over time. Line plots are mean scores over time with error bars representing the 95% confidence intervals. Mean FACT-Br Social scores significantly decrease over time (likelihood ratio test p-value < 0.001).

Figure 2.

Multivariable associations between patient and clinical factors and baseline and longitudinal FACT-Br scores Baseline (a) and longitudinal (b) multivariable associations between patient, tumor, and treatment characteristics and FACT-Br scores. Only significant (p<0.05) associations are shown in color. Blue squares indicate significant positive association; yellow squares indicate significant negative association. Grey squares indicate no significant association.

Figure 3.

Multivariable linear mixed-effects regression of cognitive function on total, physical, functional, and brain tumor-specific hrQoL Multivariable linear mixed-effects analyses of a) FACT-Br Total, b) FACT-Br Physical, c) FACT-Br Functional, and d) FACT-Br Brain as predictors of neurocognitive functioning. Assessments of Attention/Processing (5 tests), Executive Functioning (3 tests), and Memory (4 tests) are shown. Each assessment was investigated with a unique model. β estimates are shown by dot. Whiskers reflect 95% Confidence Interval (CI) of the estimate. Significant associations after correction for multiple comparisons are reflected by a 95% CI that does not cross the 0.0 reference line. * indicates associations that remained significant after correcting for multiple comparisons.

Figure 4.

Multivariable linear mixed-effects regression of cognitive function on social and emotional hrQoL Multivariable linear mixed-effects analyses of a) FACT-Br Social and b) FACT-Br Emotional as predictors of neurocognitive functioning. Assessments of Attention/Processing (5 tests), Executive Functioning (3 tests), and Memory (4 tests) are shown. Each assessment was investigated with a unique model. β estimates are shown by dot. Whiskers reflect 95% Confidence Interval (CI) of the estimate. There were no significant associations between FACT-Br Social/Family and FACT-Br Emotional hrQoL and neurocognition after correcting for multiple comparisons.