Modality‐Specific Biomarkers and Clinical Outcomes in Music‐Based Interventions for Cancer and Palliative Care: A Systematic Review

Music‐based interventions (MBIs) show promise for symptom burden in cancer and palliative care, yet the field lacks systematic evidence linking music therapy (MT) and music medicine (MM) to objective biomarkers and clinical outcomes. Inconsistent acoustic‐parameter reporting, heterogeneous biomarker measurement, and conflation of MT/MM protocols limit reproducibility, and most studies examine biomarkers separately despite the multi‐system physiology underlying symptoms such as anxiety, pain, fatigue, and quality of life. This review characterizes biomarker‐associated outcomes across MT and MM, evaluates candidate biomarkers' feasibility for monitoring MBI response, and examines acoustic parameter reporting to identify gaps and future directions for real‐time, AI‐assisted biofeedback. A 2020 PRISMA‐ and SWiM‐compliant review was conducted across MEDLINE, Embase, Scopus, and Cochrane CENTRAL (January 2000 onward), including cancer or palliative care patients receiving MT or MM with at least one objective biomarker and validated clinical outcome. Given substantial heterogeneity, no meta‐analysis was performed; evidence was synthesized descriptively by biomarker class, outcome domain, and acoustic parameters, with narrative synthesis and vote counting of directional associations. Of 27 included studies ( N = 1415), MT ( k = 14) involved active, relational techniques by credentialed therapists, while MM ( k = 13) comprised passive listening. Cortisol predominated in MT and was linked to anxiety and quality‐of‐life improvement, while HRV was reported comparably across both modalities and vascular/hemodynamic markers predominated in MM. Serotonin, EEG‐derived valence, and inflammatory markers were less studied but showed consistent, more isolated associations with anxiety, pain, and QoL. Acoustic parameters were inconsistently reported, though tempos of 60–80 BPM and nonmetric, continuous‐resonance instrumentation recurred across autonomic studies. Biomarker temporal resolution ranged from milliseconds (EEG) to days (cytokines), with HRV and EEG best suited for real‐time monitoring. MBIs demonstrate clinically meaningful biomarker‐coupled effects, with modality‐specific patterns suggesting differential mechanisms: MT was more associated with cortisol‐linked stress reduction, while HRV‐indexed vagal modulation was observed comparably across both modalities. Serotonin was the only biomarker associated with pain reduction, warranting further investigation of its analgesic role. HRV and EEG appear most feasible for real‐time biofeedback given continuous wearable acquisition, while cortisol remains better suited for pre/post stress‐axis assessment. All findings are hypothesis‐generating given small sample sizes, heterogeneous designs, and incomplete acoustic reporting. Future studies should integrate multi‐system biomarker panels, standardize acoustic reporting, and pursue prospective multi‐biomarker RCTs toward a precision music medicine framework.