Functional Neuroimaging Signatures of Fatigue in Cancer Patients: A Systematic Review Toward Neural Biomarker Identification

Read the full article

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Background Cancer-related fatigue (CRF) is a prevalent and debilitating symptom. Functional neuroimaging has emerged as a key tool to investigate its neurobiological underpinnings. A systematic synthesis is needed to map the current evidence, assess its quality, and guide future research. We aimed to systematically review the existing literature using functional neuroimaging to investigate the neural correlates of fatigue in cancer patients, and to formally assess the quality of the included studies. Methods A systematic search of PubMed, Scopus, and Web of Science was conducted. Studies were included if they used functional neuroimaging to compare brain function in cancer patients with that of a control group/condition or to relate brain function to fatigue measures. Data on study characteristics, demographics, imaging parameters, and findings were extracted. The quality of individual studies was assessed using the Newcastle-Ottawa Scale (NOS), supplemented by an appraisal of neuroimaging-specific sources of bias, and each study was graded according to how directly its imaging findings were linked to fatigue (evidence tiers). Results Eleven reports, derived from no more than nine independent cohorts, met the inclusion criteria. Imaging modalities and analytic approaches (BOLD fMRI, ASL perfusion, fractal scaling), paradigms (resting-state, working memory, planning, verbal recall, phasic alertness) and comparison groups were heterogeneous. Three studies related imaging measures directly to fatigue within cancer patients (Tier 1), five analyzed fatigue indirectly (Tier 2), and three did not demonstrate an imaging–fatigue association (Tier 3). Tier 1 and Tier 2 evidence converged on the default mode network (DMN): fatigue was associated with greater DMN–frontal connectivity, reduced task-related DMN deactivation, and elevated posterior cingulate perfusion, whereas greater connectivity of DMN nodes with the periaqueductal gray and subgenual cingulate cortex characterized non-fatigued survivors. Lower middle frontal gyrus activity correlated with fatigue severity, and pre-treatment executive-network inefficiency predicted post-treatment fatigue. Frontoparietal hyperactivation during working-memory and planning tasks was reported mainly in comparisons defined by cancer or treatment status and was not specifically linked to fatigue. All studies scored 7–9 on the Newcastle-Ottawa Scale, but the neuroimaging-specific appraisal identified small samples, variable motion control, and inconsistent correction for multiple comparisons. Conclusion Functional neuroimaging evidence specific to cancer-related fatigue is limited and heterogeneous, and several findings previously attributed to CRF reflect effects of cancer or its treatment. The most consistent fatigue-linked signal involves dysregulation of the default mode network and its coupling with frontal and brainstem regulatory regions, together with reduced frontal executive efficiency. Emerging longitudinal evidence suggests that neural inefficiency may, in some cases, precede or predict subsequent fatigue. Standardized fatigue phenotyping, analyses based on fatigue-defined contrasts within cancer populations, harmonized imaging paradigms, and adequately powered prospective designs are needed before neural biomarkers or biotypes of CRF can be established.

Article activity feed