Cancer-related fatigue (CRF) is a common and distressing symptom of cancer and its treatments that may persist for years following treatment completion in approximately one-third of cancer survivors. Despite its high prevalence, little is known about the pathophysiology of CRF. Using a comprehensive group of physiological and psychosocial variables, the aim of the present study was to identify correlates of CRF in a heterogenous group of cancer survivors. Ninety-three cancer survivors (51 fatigued, 42 non-fatigued, with grouping based on validated cut-off scores derived from The Functional Assessment of Chronic Illness Therapy - Fatigue scale) completed assessments of performance fatigability (i.e. the change in maximal force-generating capacity, contractile function and capacity of the central nervous system to activate muscles caused by cycling exercise), cardiopulmonary exercise testing, venous blood samples for whole blood cell count and inflammatory markers and body composition. Participants also completed questionnaires measuring demographic, treatment-related, and psychosocial variables. The results showed that performance fatigability (decline in muscle strength during exercise), time-to-task-failure, peak oxygen uptake (V̇O2peak), tumor necrosis factor-α (TNF-α), body fat percentage and lean mass index were associated with CRF severity. Performance fatigability, V̇O2peak, TNF-α and age explained 35% of the variance in CRF severity. Furthermore, those with clinically-relevant CRF reported more pain, more depressive symptoms, less social support, and were less physically active than non-fatigued cancer survivors. Given the association between CRF and numerous physical activity related measures, including performance fatigability, cardiorespiratory fitness, and anthropometric measures, the present study identifies potential biomarkers by which the mechanisms underpinning the effect of physical activity interventions on CRF can be investigated.