Cancer-related fatigue as partial mediator of the effect of a remotely delivered exercise intervention on health-related quality of life: Phase II randomized trial.
Abstract
276 Background: Patients receiving immunotherapy frequently experience cancer-related fatigue, symptom burden, and impaired health-related quality of life (HRQOL). Although exercise can improve patient-reported outcomes, the mechanisms underlying its effects remain unclear; this study evaluated whether reductions in cancer-related fatigue mediated the effects of a remotely delivered exercise intervention on HRQOL. Methods: This secondary analysis of a Phase II randomized controlled trial included adults with cancer receiving immunotherapy. Participants completed the Functional Assessment of Cancer Therapy-General (FACT-G) and Edmonton Symptom Assessment Scale (ESAS) at baseline and were randomized 1:1 to usual care (UC) or a 12-week remotely delivered exercise intervention. The intervention included weekly exercise physiologist-led telehealth sessions, individualized ACSM-based prescriptions, and 30-45 minutes/day of aerobic, resistance, and mobility exercise. Outcomes were reassessed at 12 weeks. Clinically meaningful HRQOL improvement was defined as a ≥7-point increase in FACT-G. Fisher’s exact tests compared responder rates, and mediation analyses used linear regression to evaluate whether changes in fatigue explained the effect of the intervention on FACT-G scores. Results: Seventy patients were randomized (35 UC, 35 intervention); 65 completed the 12-week assessment. Mean age was 67.6 years (SD = 11.4), 46.2% were female, and 76.9% had stage IV disease. Common cancer types included genitourinary (26.2%), gastrointestinal (20.0%), lung (15.4%), and breast (13.8%) cancers. Most participants received immunotherapy alone (64.6%). A clinically meaningful improvement in HRQOL was observed in 75.8% of participants receiving the exercise intervention compared with 6.3% receiving UC (p = 0.001). Exercise participation was associated with significant reductions in fatigue (β = -3.81, p < 0.001) and significant improvements in HRQOL (β = 11.65, p < 0.001). Fatigue improvement was independently associated with HRQOL improvement (β = -1.42, p < 0.001). After adjustment for changes in fatigue, the direct effect of the intervention on HRQOL was attenuated from β = 11.65 to β = 6.24, indicating that approximately 47% of the intervention effect on HRQOL was mediated through reductions in fatigue. Conclusions: A remotely delivered exercise intervention significantly increased the likelihood of clinically meaningful HRQOL improvement among patients receiving immunotherapy. Nearly half of this benefit was mediated by reductions in cancer-related fatigue, suggesting fatigue as a key mechanism underlying exercise-related gains in quality of life. These findings support integrating scalable remote exercise programs into supportive cancer care and targeting fatigue during immunotherapy.