Cryotherapy for preventing chemotherapy-induced neuropathy in breast cancer patients treated with paclitaxel: a systematic review and meta-analysis

Background Cryotherapy has shown potential for preventing chemotherapy-induced peripheral neuropathy (CIPN) in breast cancer (BC) patients receiving taxane-based chemotherapy; however, study findings and intervention protocols remain variable. Methods We searched Scopus, PubMed/MEDLINE, ProQuest, ScienceDirect, SAGE Journals, and the Cochrane Library from database inception to July 17, 2025, without date limits, 10-year restrictions, or randomized controlled trial design filters. The protocol was registered in PROSPERO on July 11, 2025 (CRD420251102474) before the final search update. Eligible studies enrolled BC patients receiving paclitaxel or nab-paclitaxel and evaluated cryotherapy as the main preventive intervention, with or without a comparator. Comparative outcomes were synthesized separately from single-arm proportions. Risk of bias was assessed using RoB 2 for randomized studies and ROBINS-I for non-randomized studies, and certainty of evidence was assessed with GRADE. Single-arm proportions were treated as very low-certainty, non-comparative descriptive evidence. Results Fourteen studies, including 12 randomized controlled trials and 2 retrospective cohort studies, were included. In comparative double-arm studies, cryotherapy was associated with lower CTCAE Grade > = 2 neuropathy (RR = 0.45, 95% CI 0.27–0.77, p = 0.003) and lower PNQ grade D or higher total score (RR = 0.24, 95% CI 0.09–0.62, p = 0.003), with low-to-moderate statistical heterogeneity. Sensory neuropathy was lower in the cryotherapy group (RR = 0.35, 95% CI 0.12–0.98, p = 0.046), whereas motor neuropathy was not significantly reduced. Single-arm studies showed a pooled CTCAE Grade > = 2 neuropathy proportion of 0.20, but this estimate had high heterogeneity and represents only very low-certainty, non-comparative descriptive incidence under cryotherapy. The pooled incidences of any adverse event, cold intolerance, and cryotherapy discontinuation due to side effects were 19%, 16%, and 2%, respectively; these tolerability proportions were also non-comparative and highly heterogeneous. Chemotherapy dose reduction (RR = 0.43, p = 0.109) and treatment discontinuation due to neuropathy (RR = 0.22, p = 0.163) were each based on only two studies, had wide confidence intervals, and were underpowered and inconclusive. Single-arm proportions must be interpreted cautiously and should not be interpreted as demonstrating efficacy or safety. Conclusion Comparative evidence suggests that cryotherapy may reduce sensory CIPN in BC patients receiving paclitaxel or nab-paclitaxel. However, single-arm estimates are very low-certainty and descriptive only, tolerability outcomes are heterogeneous, and chemotherapy dose-reduction and discontinuation outcomes are underpowered and inconclusive. These limitations warrant cautious clinical interpretation and further standardized trials.