Optimizing exercise regimens for quality of life in patients with cancer: a network meta-analysis and dose–response study

To systematically evaluate the effects and dose–response relationships of different exercise interventions in improving the quality of life of patients with cancer using a combined Bayesian network meta-analysis and network dose–response meta-analysis approach. Systematic searches were conducted in PubMed, Web of Science, Embase, and the Cochrane Library (up to August 2026). Interventions were ranked using the surface under the cumulative ranking curve (SUCRA), and dose–response relationships were modelled using the Emax model. A total of 149 randomized controlled trials involving 13,004 patients with cancer were included. Network meta-analysis showed that Qigong/Tai Chi, high-intensity interval training (HIIT), and combined aerobic and resistance training were the highest-ranked interventions according to SUCRA. Model-based dose–response analysis suggested a nonlinear association between exercise dose and quality of life, with the model-estimated overall effect peaking at approximately 2000 MET-min/week. Qigong/Tai Chi was associated with significant improvements at lower doses (750 MET-min/week), whereas HIIT and combined aerobic and resistance training showed model-estimated maximum effects at 800 and 2000 MET-min/week, respectively. This study compared exercise modalities and dose–response characteristics, identifying model-derived dose estimates associated with the greatest quality-of-life improvements in patients with cancer. These estimates may inform the scientific application of exercise interventions in cancer rehabilitation but require prospective validation.