Feasibility and preliminary efficacy of a 12-week remotely delivered exercise program in patients receiving immunotherapy: A phase II randomized controlled trial.

1638 Background: Patients with advanced cancer receiving immunotherapy experience reduced physical function, increased symptom burden, and poorer health-related quality of life (HRQOL). Limited access to structured exercise programs makes technology-supported interventions a potentially scalable alternative. We assessed the feasibility and preliminary efficacy of a 12-week remote exercise program on HRQOL, symptom burden, and fear of cancer recurrence (FCR), with an exploratory economic evaluation. Methods: In this Phase II trial, adult receiving immunotherapy alone or in combination were randomized 1: 1 to a remote exercise intervention or usual care (UC). The study was powered to detect a 10-point difference in Functional Assessment of Cancer Therapy-General (FACT-G) scores (80% power, α=0. 05). The intervention included weekly virtual sessions with an exercise physiologist, individualized Borg-guided exercise prescriptions, video support, and telehealth monitoring 12 weeks (30-45 minutes/day). Outcomes assessed at baseline and 12 weeks included HRQOL (FACT-G), symptom burden (Edmonton Symptom Assessment Scale; ESAS), and FCR (FCR-7). Between-group differences were analyzed using ANCOVA adjusted for baseline values, with effect sizes estimated using Cohen’s d. A preliminary provider-perspective cost-consequence analysis estimated cost per one-point FACT-G improvement. Results: Seventy patients were randomized (35 per arm) ; five died before 12 weeks (3 control, 2 intervention). Baseline characteristics were balanced. At 12 weeks, the intervention group had significantly higher FACT-G scores compared than UC (mean difference = 11. 1; p<0. 001; Cohen’s d=1. 30). Symptom burden was lower in the intervention arm (p<0. 001; Cohen’s d=1. 68), and FCR was reduced (p=0. 014; Cohen’s d=0. 85). Group assignment remained a significant predictor in adjusted ANCOVA models (all p<0. 01). Estimated intervention cost was R7, 023 per participant, or R634 per FACT-G point gained. Conclusions: A 12-week remote exercise intervention produced a clinically meaningful 11-point improvement in HRQOL at an estimated cost of R 634 per FACT-G point, supporting its economic feasibility and scalability. The program reduced symptom burden and FCR. These findings justify larger trials to confirm effectiveness and further evaluate cost-effectiveness. Changes in patient-reported outcomes at 12 weeks. Outcome (scale) Group BaselineMean ± SD 12-weekMean ± SD Mean Change P-value Cohen’s d FACT-G Control (n=32) 87. 3 ± 9. 7 85. 9 ± 10. 2 -1. 4 0. 001 1. 30 Intervention (n=33) 85. 2 ± 9. 1 96. 9 ± 6. 7 +11. 8 ESAS Control (n=32) 17. 3 ± 12. 1 19. 0 ± 12. 8 +1. 7 0. 001 1. 68 Intervention (n=33) 20. 6 ± 9. 6 5. 5 ± 3. 7 -15. 1 FCR-7 Control (n=32) 15. 0 ± 6. 0 15. 4 ± 6. 5 +0. 4 0. 014 0. 85 Intervention (n=33) 16. 2 ± 6. 3 11. 3 ± 3. 8 -4. 9