Acupuncture for Chemotherapy-Induced Peripheral Neuropathy: A Systematic Review and Meta-Analysis with Trial Sequential Analysis.

Introduction

Cancer imposes a major global health burden, with 18.5 million new cases reported in 2023 and incidence projected to increase over the next two decades. 1 Chemotherapy is a mainstay of cancer treatment that improves prognosis but causes diverse toxicities, potentially limiting its clinical benefit. 2 Among these toxicities, chemotherapy-induced peripheral neuropathy (CIPN) is one of the most common and challenging, affecting approximately 68% of patients within 1 month of treatment. 3 , 4 CIPN primarily manifests as hypoesthesia, such as numbness in the extremities, or hyperesthesia, such as tingling. CIPN symptoms can substantially impair patients’ daily activities and quality of life; severe CIPN may also necessitate dose reduction, treatment delays, or chemotherapy discontinuation, potentially compromising therapeutic efficacy and clinical outcomes. 5 , 6 Pharmacological options for CIPN remain limited. Duloxetine is the only drug recommended by current clinical guidelines for established painful CIPN, although the recommendation remains limited in strength. 7 Other agents, including gabapentinoids and tricyclic antidepressants, are frequently used in practice despite limited supporting evidence. Adverse effects such as nausea, dizziness, and somnolence, together with potential drug interactions, may further complicate care for patients receiving multiple anticancer and supportive medications, reinforcing the need for safer therapeutic alternatives.

Acupuncture, a nonpharmacological intervention, has attracted growing attention in supportive oncology care. 8–10 It is widely used to alleviate chemotherapy-related adverse events and has a favorable overall safety profile. 11 , 12 The American Society of Clinical Oncology (ASCO) clinical practice guideline on CIPN 13 indicates that acupuncture may benefit CIPN prevention and treatment, but limitations in the evidence preclude definitive recommendations. As higher-quality evidence on acupuncture for CIPN treatment has emerged, 14–17 a comprehensive and rigorous synthesis is needed to inform clinical practice guidelines. Existing evidence syntheses of acupuncture for CIPN have included heterogeneous acupuncture modalities, complicating interpretation. In clinical practice and oncology guidelines, manual acupuncture (MA) and electroacupuncture (EA) receive the greatest attention; 18 however, to our knowledge, these modalities have not been systematically evaluated specifically for CIPN. In addition, previous syntheses have not adequately addressed the risk of type I error in cumulative meta-analysis or determined whether the accumulated sample size is sufficient to support stable conclusions.

Accordingly, we conducted a systematic review and meta-analysis with trial sequential analysis (TSA) to evaluate the clinical effects of MA and EA for preventing and treating CIPN and to identify factors that may influence these effects, thereby informing clinical practice and future trial design.

Methods

This systematic review and meta-analysis was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (see the PRISMA 2020 checklist). The review protocol was registered in PROSPERO (CRD420251272891).

Data Sources and Search Strategy

Two reviewers (ZFP and JSW) independently searched eight databases—PubMed, Embase, the Cochrane Library, Web of Science (WOS), China National Knowledge Infrastructure (CNKI), China Science and Technology Journal Database (VIP), SinoMed, and WanFang—from database inception to January 2026. The search strategy combined Medical Subject Headings (MeSH) with free-text terms; for example, the PubMed search included the terms “Chemotherapy,” “Neoplasms,” “Peripheral Nervous System Disorders,” “Acupuncture,” and “randomized controlled trial.” No language restrictions were applied; the complete search strategies are provided in Supplementary Tables 1 – 8 . We also manually screened the reference lists of included studies for additional eligible records.

Inclusion Criteria

Eligibility criteria were defined according to the Participants, Interventions, Comparators, Outcomes, and Study design (PICOS) framework as follows:

(1) Participants: Eligible participants had established CIPN or a pathologically confirmed malignant tumor and were scheduled to receive chemotherapy associated with peripheral neurotoxicity (eg, platinum agents, taxanes, or vinca alkaloids).

(2) Interventions: The intervention group received MA or EA, either alone or combined with other treatments. Eligible comparators included sham acupuncture, conventional drug treatment, and other control interventions. Conventional drug treatments were defined as prescription medications used for CIPN, including duloxetine and mecobalamin.

(3) Outcomes: Studies were required to report at least one validated measure of CIPN. Continuous outcomes included the Functional Assessment of Cancer Therapy/Gynecologic Oncology Group-Neurotoxicity (FACT-Ntx), Visual Analog Scale (VAS), nerve conduction study (NCS) parameters, and quantitative sensory testing (QST) outcomes; the binary outcome was the clinical effectiveness rate (ER), as defined in each original RCT.

(4) Study design: Randomized controlled trials (RCTs).

Exclusion Criteria

(1) Studies reporting only unvalidated outcome measures.

(2) Reviews, case reports, animal studies, other non-RCT designs, studies with incomplete data, and duplicate publications.

Study Selection

Records were imported into EndNote 21 and automatically deduplicated. Two reviewers (ZFP and JSW) then independently screened titles and abstracts, manually checked for duplicate records, and assessed potentially eligible full texts for inclusion. Disagreements were resolved through consultation with a third reviewer (ZNG).

Data Extraction

Two reviewers (ZFP and JSW) independently extracted data using a prespecified standardized form. Extracted study characteristics included author names, publication year, country, cancer-related variables, and acupuncture-related variables; outcome data included binary and continuous measures. For binary outcomes, we extracted event counts and total sample sizes; for continuous outcomes, we extracted postintervention means and standard deviations (SDs). For multi-arm studies, the shared control group was split equally across the treatment arms according to the Cochrane Handbook, with its sample size divided by the number of treatment arms to avoid inflated statistical power. For studies reporting only change scores, we reconstructed the variance using an assumed correlation coefficient, as described by Borenstein et al and the Cochrane Handbook for Systematic Reviews of Interventions, 19 , 20 the correlation coefficient r = 0.5. For studies reporting medians and interquartile ranges (IQRs), the Box-Cox method 21 was used to estimate the corresponding means and SDs. For binary outcomes with zero events in either group, a continuity correction of 0.5 was added to each zero cell before calculating the effect estimate and confidence interval. Disagreements during data extraction were resolved through consultation with a third reviewer (ZNG).

Risk of Bias Assessment

The Cochrane Risk of Bias 2 tool (RoB 2) was used to assess risk of bias in the included RCTs. RoB 2 evaluates five domains: the randomization process, deviations from intended interventions, missing outcome data, outcome measurement, and selection of the reported result. Each domain was rated as “low risk of bias,” “some concerns,” or “high risk of bias.” The overall risk-of-bias judgment was derived from the five domain-level assessments. Disagreements between the two reviewers (ZFP and JSW) were resolved through discussion or, when necessary, consultation with a third reviewer (ZNG).

Certainty of Evidence

Using GRADEpro ( https://www.gradepro.org/ ), the GRADE framework 22 was used to assess the certainty of evidence for FACT-Ntx, the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire–Chemotherapy-Induced Peripheral Neuropathy 20-item scale (EORTC QLQ-CIPN20), and ER. Certainty was evaluated across five domains: risk of bias, inconsistency, indirectness, imprecision, and publication bias. Potential upgrading factors were also considered, and certainty was classified as high, moderate, low, or very low.

Data Analysis

Meta-analyses were performed using the “meta” package in R version 4.5.2. Pooled estimates for binary outcomes were expressed as risk ratios (RRs) with 95% confidence intervals (CIs). For continuous outcomes, mean differences (MDs) with 95% CIs were used when measurement units were consistent; otherwise, standardized mean differences (SMDs) with 95% CIs were calculated. Statistical heterogeneity was assessed using the I 2 statistic. A random-effects model was used when I 2 > 50% or P < 0.10; otherwise, a fixed-effect model was applied. To explore heterogeneity, prespecified subgroup analyses were conducted by acupuncture intervention type, intervention purpose, and chemotherapy type; objective outcomes were also subgrouped by measurement site (see Supplementary Table 9 ). Leave-one-out sensitivity analyses were performed to assess the robustness of the pooled estimates. For outcomes including at least 10 studies, Begg’s and Egger’s tests were used to assess small-study effects and potential publication bias.

Trial sequential analysis (TSA) was conducted for FACT-Ntx, EORTC QLQ-CIPN20, and ER to evaluate the stability of the cumulative evidence. TSA combines principles of interim analysis with cumulative meta-analysis and uses a required information size (RIS) and trial sequential monitoring boundaries to control random error and the risk of type I error. A cumulative sample size that reached or exceeded the RIS was considered to provide sufficient information for a stable conclusion. If this threshold was not reached, additional evidence was considered necessary to clarify the result. TSA was performed using TSA version 0.9.5.10 beta ( https://www.ctu.dk/tsa/ ).

Results

Study Selection

The database searches identified 2,588 records, of which 1,095 were removed as duplicates. Of the remaining 1,493 records, 1,310 were excluded after title and abstract screening. The full texts of 183 reports were then assessed for eligibility. Reasons for full-text exclusion are provided in Supplementary Table 10 . Ultimately, 40 studies were included in the review and meta-analysis ( Figure 1 ). 14–16 , 23–59 Figure 1 Flow chart of the PRISMA. The flowchart outlines the identification of studies via databases. Identification phase: Records identified from PubMed (n equals 83), The Cochrane Library (n equals 345), Web of Science (n equals 240), Embase (n equals 156), CNKI (n equals 262), WanFang (n equals 512), CBM (n equals 955), VIP (n equals 35). Records removed before screening: Duplicate records removed (n equals 1095). Screening phase: Records screened (n equals 1493), Records excluded (n equals 1310). Reports sought for retrieval (n equals 183), Reports not retrieved (n equals 4). Reports assessed for eligibility (n equals 179), Reports excluded (n equals 139). Included phase: Studies included in review (n equals 38), Reports of included studies (n equals 40). A flowchart detailing the identification and screening process of studies via databases.

Study Characteristics

The 40 studies, published from 2009 to 2026, included 2,496 participants. Most studies were conducted in China; the remainder were conducted in the United States, Germany, the United Kingdom, or Brazil. Most studies included mixed cancer types (n = 22); among studies restricted by cancer type, breast cancer (n = 9) and gastrointestinal cancers (n = 5) were the most common. The most frequently evaluated neurotoxic chemotherapy classes were taxanes (n = 10) and oxaliplatin (n = 9). Interventions comprised EA alone (n = 12), MA alone (n = 11), EA combined with other treatments (n = 6), and MA combined with other treatments (n = 13). Comparators included no intervention, conventional treatment, wait-list control, rehabilitation exercise, and sham acupuncture. Frequently used acupoints included LI4, ST36, LI11, SP6, LR3, and EX-LE10, as detailed in Table 1 . Table 1 General Characteristics of Included Studies Author Year Country Cancer Type Chemotherapy Sample Size (Event/Control) Intervention Control Frequency Duration Acupoints Outcome A. Molassiotis 59 2019 Hong Kong, China Any type Unspecified 44/43 MA Wait-list 30min/d, 2d/wk 8wk LI4, LI11, PC7, TE5, Ex-UE9, SP6, ST36, LR3, ST41, EX-LE10 BPI, FACT-Ntx, TNSc, NCI-CTCAE V4.0, NCS B. Liu 49 2009 China Any type Fluorouracil/platinum/paclitaxel 30/30 MA Vitamin B 30min/d, 4wk 4wk ST36, SP6, GB34, LI4, LI11, KI16, CV6, CV4 WHO Peripheral Neurotoxicity Grading Scale for Solid Tumors C. C. Huang 53 2021 Taiwan, China Breast cancer Taxane 10/10 MA SA 30min/d, 15d/9wk 9wk CV6, LI11, PC6, LI4, ST36, SP6 BPI-SF, QST C. L. Li 42 2023 China Breast cancer Unspecified 29/29 EA + PB MeCbl 30min/d, 20d/20d 20d LI11, LI4, PC6, EX-UE9, SP10, ST36, LR3, EX-LE10 NCI-CTCAE V4.0, NCS, VAS C. Lu 33 2024 China Breast cancer Taxane 39/38/37/38 EA MeCbl 30min/d, 3d/wk 4wk LI11, TE5, LI4, SI3, EX-UE9, GB34, SP9, SP6, EX-LE10 FACT-Ntx, NCI-CTCAE V5.0, VAS C. Lu 37 2025 China Breast cancer Utidelon 30/30 EA MeCbl 30min/d, 3d/wk 4wk LI11, TE5, LI4, SI3, EX-UE9, GB34, ST36, SP9, SP6, LR3, EX-LE10 PNQ, EORTC QLQ-CIPN20 C. Zhang 48 2022 China Any type Capecitabine 60/60 MA + Vitamin B Vitamin B 30min/d, 2wk 2wk ST36, SP6, GB34, LI4, LI11, KI16, CV6, CV4 NCI-CTCAE V2.0, NCS D. L. Cui 47 2011 China Any type Oxaliplatin 40/40 MA + Moxa + GSH GSH 10min/d 10d 10d LI11, SJ5, SI3, LI4, SJ3, SP9, ST36, GB34, SP6, GB39, KI6 Levi’s scale D. X. Chen 52 2025 China Any type Oxaliplatin 35/35 EA + GSH GSH 30min/d, 7d/wk 2wk ST36, SP6, LI4, LI11 NCS E. G D’Alessandro 57 2022 Brazil Any type Unspecified 15/14 MA Rehabilitation 30min/d, 10d/5wk 5wk LR3, SP3, KI3, HT7, PC7, LU9, SP9 NCI-CTCAE V.2.0, EORTC QLQ-C30, VAS H. Greenlee 27 2016 America Breast cancer Taxane 32/32 EA SA 30min/d, 1d/wk 12wk GB34, ST36, LI4, LI10, Huatuojiaji L3,L5, EX-LE10, Huatuojiaji C5,C7, EX-UE9 BPI-SF, FACT-Ntx, NPS H. K. Hou 39 2011 China Any type Oxaliplatin/paclitaxel 20/20 MA MeCbl 30min/d, 2wk 2wk LI11, PC6, LI4, ST36, SP6, GB34 NCI-CTCAE V3.0 H. X. Wei 41 2014 China Any type Paclitaxel 30/30 EA + Moxa MeCbl 30min/d, 3d/wk 6wk ST36, ST41, LI4, CV4 NCI-CTC V2.0 J. Q. Xu 23 2025 China Any type Platinum/taxane 35/35 MA + MeCbl SA + MeCbl 30min/d, 3d/wk 2wk ST36, SP10, LI4, LR3, LI11, EX-LE10, EX-UE9, RN6 NCI-CTCAE V3.0, EORTC QLQ-CIPN20, NRS, NCS J. Stringer 16 2022 United Kingdom Any type Unspecified 61/59 MA + UC UC 40min/d, 1d/wk 10wk LR3, SP6, ST36, EX-LE10, BL60, EX-UE9, LI4 MYMOP2, NRS, EORTC QLQ-CIPN20 J. Sun 43 2023 China Any type Paclitaxel 30/30 MA + HFW + MeCbl Duloxetine 30min/d, 5d/wk 4wk BL13, BL18, BL20, BL23, PC6, CV6, ST36, SP6 NCS K. Chan 35 2023 Hong Kong, China Colorectal cancer Oxaliplatin 30/30 EA SA 30min/d, 1d/wk 12wk LI4, PC6, LI11, EX-UE9,ST36, SP6, LR3, EX-LE10 FACT-Ntx, NRS, EORTC QLQ-C30 K. L. Lu 26 2020 China Colorectal cancer Oxaliplatin 26/26 MA + HP MeCbl 30min/d, 5d/wk 2wk ST36, LI10, CV6, LI4, LI11, SP10, ST40, LR3, EX-UE9, EX-LE10 WHO toxicity grading criteria, NCS L. Ter 56 2026 Brazil Breast cancer Paclitaxel 24/23 MA SA 25min/d, 1d/wk 8wk Ex-UE9, TE5, LI4, Ex-LE10, KI3, ST36 NPSI, VAS, FACT-taxane L. Wang 32 2022 China Multiple myeloma Unspecified 26/28 MA UC 30min/d, 2wk 2wk EX-LE10, LI4, LR3, SP10, ST25, SP6, ST36 NCS, FACT-Ntx L. Y. He 30 2024 China Gastric cancer Oxaliplatin 28/28 MA + HP + MeCbl MeCbl 30min/d, 7d/3wk 9wk LI11, LI10, SJ5, LI4, EX-UE9, GB34, ST36, SP6, LR3, EX-LE10, GV20, BL17, BL20, BL23 Levi’s scale M. H. Sui 24 2018 China Any type Unspecified 10/10/10 EA/EA + MeCbl MeCbl 30min/d, 3d/wk 3wk ST36, LI4, LI11, GB34, SP6, EX-LE10, LR3 VAS M. Rostock 54 2013 Germany Any type Unspecified 14/13/15/17 EA HB/Vitamin B/Placebo 15min/d, 8 ± 1d/3wk 3wk LR3, SP9, GB41, GB34, LI4, LI11, SI3, HT3 NRS, NCI-CTCAE V.2.0, EORTC QLQ-C30, NCS Nancy Avis 55 2022 America Breast cancer/gastrointestinal cancer Paclitaxel 11/12 MA UC 45min/d, 8d/10wk 10wk No information EORTC QLQ-CIPN20, NCI-CTCAE v5.0, NCS Q. L. Wen 36 2024 China Any type Unspecified 50/50 MA + Moxa + MeCbl MeCbl 30min/d, 5d/wk 4wk LI4, SJ5, LI10, CV6, GB34, LI5, ST36, LI11, LR3, ST40 EORTC QLQ-C30, NCI-CTCAE V3.0 S. Iravani 38 2020 China/Iran Any type Unspecified 20/20 MA Vitamin B + Gabapentin 20min/d, 3d/wk 4wk CV6, GV20, ST36, SP6, LI4, LI11, LR3, EX-LE10, EX-UE9 NRS, NCI-CTCAE v.4.0, NCS T. Bao 14 2020 America Any type Unspecified 27/24/24 EA SA/UC 20min/d, 10d/8wk 8wk LI4, PC6, SI3, LR3, GB42, ST40, EX-LE101, EX-LE103 NRS T. Bao 15 2021 America Any type Unspecified 27/24/24 EA SA/UC 20min/d, 10d/8wk 8wk LI4, PC6, SI3, LR3, GB42, ST40, EX-LE101, EX-LE103 FACT-Ntx T. T. Wang 45 2025 China Any type Paclitaxel 40/40 MA + Moxa MeCbl 10min/d, 7d/3wk 6wk CV4, CV6, CV12, SP10 BPI, Levi’s scale, NCS W. B. Chen 50 2018 China Colorectal cancer Oxaliplatin 40/40 MA MeCbl 30min/d, 5d/wk 2wk ST36, CV6, SP3, LI4, LI11, LR3, EX-UE9, EX-LE10 WHO Peripheral Neurotoxicity Grading Scale for Solid Tumors W. Fu 40 2024 China Any type Unspecified 45/45 EA + MeCbl MeCbl 30min/d, 5d/wk 4wk LI4, LI11, ST36, SP6, LR3 NCS, TCSS, EORTC QLQ-C30 W. I. Zhi 31 2023 America Any type Unspecified 27/24/24 EA SA/UC 20min/d, 10d/8wk 8wk LI4, PC6, SI3, LR3, GB42, ST40, EX-LE101, EX-LE103 QST W. J. Chen 28 2025 China Any type Oxaliplatin 35/36 EA MeCbl 30min/d, 3d/wk 4wk LI11, TE5, LI4, EX-UE9, ST36, SP9, SP6, LR3, EX-LE10 PNQ, EORTC QLQ-CIPN20, EORTC QLQ-C30 X. H. Han 58 2017 China Multiple myeloma Unspecified 49/49 MA + MeCbl MeCbl 30min/d, 8d/4wk 12wk LR3, ST43, GB41, SP6, ST36, SP10, ST25, GV14, GV12, GV11, GV9, BL13, BL17, BL58 VAS, FACT-Ntx, NCS X. H. Liu 29 2023 China Breast cancer Anthracycline/taxane 24/24 WA + exercise UC 30min/d, 5d/wk 4wk LI4, SJ5, PC6, LI11, ST41, LR3, GB40, SP3, SP4, ST45, ST44, ST36, SP6, CV12, CV6 FACT-Ntx, NCI-CTCAE V5.0 Y. M. Li 44 2025 China Breast cancer Taxane 30/30 MA + Moxa + MeCbl MeCbl 30min/d, 20d 20d RN10, RN12, RN6, RN4, ST25, ST24, ST26, SP15 NCI-CTCAE v5.0, Y. P. Tian 25 2011 China Gastric cancer Oxaliplatin 38/38 MA + Moxa Neurotropin 30min/d, 7d/wk 3wk LI4, SJ5, LI5, LI10, LI11, LR3, ST36, GB34, CV6, ST40 Levi’s scale Y. Wu 51 2018 China Any type Oxaliplatin 20/19/19 MA MeCbl/UC 30min/d, 6d/wk NI LI11, PC6, LI4, SP10, ST36, SP6 Levi’s scale Y. Xiang 46 2018 China Any type Unspecified 08/10/10 EA/EA + MeCbl MeCbl 30min/d, 3d/wk 3wk ST36, LI4, LI11, GB34, SP6, EX-LE10, LR3 VAS, global perceived recovery Y. Yao 34 2024 China Breast cancer Taxane 23/25 EA UC 30min/d, 14d/12wk 12wk GV24+, EX-UE9, TE5, LI4, LI11, ST36, SP9, SP6, KI3, LR3, GB39, EX-LE10 EORTC QLQ-CIPN20, NCS, EORTC QLQ-C30 Abbreviations : BPI-SF, Brief Pain Inventory-Short Form; BPI WPS, Brief Pain Inventory Worst Pain Score; EA, electroacupuncture; GSH, Glutathione; HB, Hydroelectric Baths; HP, herbal prescription; MA, manual acupuncture; MeCbl, mecobalamin; NPSI, Neuropathic Pain Symptom Inventory; NPS, Neuropathic Pain Scale; PNQ, Patient Neurotoxicity Questionnaire; SA, sham acupuncture; TCSS, Toronto clinical scoring system; TNSc, Total Neuropathy Score Clinical Version; UC, usual care.

Risk of Bias

Among the 40 included studies, 10 were judged to have a low risk of bias, 15 to raise some concerns, and 15 to have a high risk of bias. The main concerns were unclear reporting of randomization and allocation concealment, unspecified analysis sets for the primary analyses, lack of assessor blinding, and absence of preregistered protocols; the risk-of-bias plot was generated using robvis ( Figure 2 ). 60 Study-level risk-of-bias assessments are presented in Supplementary Figure 1 . Figure 2 Risk of bias graph. A stacked horizontal bar graph showing risk of bias levels across bias domains and overall risk of bias.

CIPN Symptoms

FACT-Ntx

After one study was excluded based on the sensitivity analysis ( Supplementary Figure 2 ), 10 studies were included in the FACT-Ntx analysis. The results of the meta-analysis showed that acupuncture was more effective than the control group in improving FACT-Ntx scores, indicating an improvement in the overall symptoms of CIPN (MD = −1.43, 95% CI [−1.91, −0.95], P < 0.001, I 2 = 0%; Figure 3A ). In the subgroup analysis by acupuncture type ( Supplementary Figure 3A ), MA significantly improved FACT-Ntx scores compared with control (MD = −1.33, 95% CI [−1.87, −0.78], P < 0.001), whereas EA did not (MD = −1.42, 95% CI [−2.96, 0.12], P = 0.07). By intervention purpose ( Supplementary Figure 3B ), therapeutic acupuncture significantly improved FACT-Ntx scores (MD = −1.41, 95% CI [−1.91, −0.92], P < 0.001), whereas preventive acupuncture did not (MD = −1.80, 95% CI [−4.23, 0.63], P = 0.15). By chemotherapy type, acupuncture significantly improved FACT-Ntx scores in the mixed-chemotherapy subgroup (MD = −1.43, 95% CI [−1.93, −0.92], P < 0.001; Supplementary Figure 3C ) but not in the taxane subgroup (MD = −1.34, 95% CI [−3.54, 0.86], P = 0.23). Begg’s ( P = 0.282) and Egger’s ( P = 0.929) tests showed no evidence of publication bias ( Supplementary Figure 4 ). Figure 3 Forest plots of meta-analysis result for CIPN symptoms. Image A shows a forest plot for FACT Ntx with studies comparing acupuncture and control groups. The x-axis represents mean difference, ranging from -5 to 1 and the y-axis lists 10 studies. Most study results are left of 0, with a pooled diamond at -1.43 (95% CI: -1.91 to -0.95), total weight 100%. Image B displays a subgroup forest plot with 3 studies, x-axis from -8 to 0 and a pooled diamond at -3.61 (95% CI: -6.90 to -3.12). Image C includes 9 studies, x-axis from -4 to 2, with a pooled diamond at -0.97 (95% CI: -1.35 to -0.60). Image D features 10 studies, x-axis from -4 to 2, with a pooled diamond at -1.34 (95% CI: -1.96 to -0.71). Image E shows 4 studies, x-axis from -2 to 1, with a pooled diamond at -0.59 (95% CI: -1.45 to 0.28). A forest plot comparing acupuncture versus control for FACT Ntx scores, with pooled effects favoring acupuncture. Notes : ( A ) FACT-Ntx; ( B ) EORTC QLQ-CIPN20; ( C ) NRS; ( D ) VAS; ( E ) BPI-WPS.

EORTC QLQ-CIPN20

After one study was excluded based on the sensitivity analysis ( Supplementary Figure 5 ), four studies were included in the EORTC QLQ-CIPN20 analysis. Acupuncture significantly reduced EORTC QLQ-CIPN20 scores compared with control (MD = −5.01, 95% CI [−6.9, −3.12], P < 0.001, I 2 = 0%; Figure 3B ). In the prespecified analysis by acupuncture type ( Supplementary Figure 6A ), both EA alone and MA combined with other treatments significantly reduced EORTC QLQ-CIPN20 scores (EA: MD = −3.99, 95% CI [−7.23, −0.75], P = 0.02; MA + other: MD = −5.54, 95% CI [−7.86, −3.21], P < 0.001). By chemotherapy type ( Supplementary Figure 6B ), acupuncture significantly improved EORTC QLQ-CIPN20 scores in the mixed-chemotherapy subgroup (MD = −5.54, 95% CI [−7.86, −3.21], P < 0.001). Each remaining chemotherapy-type subgroup contained only one study, with effect estimates in the same direction as that of the mixed-chemotherapy subgroup.

NRS

Nine studies were included in the NRS analysis. Acupuncture significantly reduced numeric rating scale (NRS) scores compared with control (MD = −0.97, 95% CI [−1.35, −0.60], P < 0.001, I 2 = 0%; Figure 3C ). By acupuncture type ( Supplementary Figure 7A ), both MA combined with other treatments and EA alone significantly reduced NRS scores (MA + other: MD = −0.80, 95% CI [−1.34, −0.25], P < 0.001; EA: MD = −0.87, 95% CI [−1.50, −0.24], P = 0.007). In the analysis by intervention purpose ( Supplementary Figure 7B ), eight therapeutic studies showed a significant reduction in NRS scores (MD = −0.99, 95% CI [−1.38, −0.60], P < 0.001). By chemotherapy type ( Supplementary Figure 7C ), acupuncture significantly reduced NRS scores in the mixed-chemotherapy subgroup (MD = −0.99, 95% CI [−1.38, −0.60], P < 0.001). The sensitivity analysis did not materially alter the pooled estimate ( Supplementary Figure 8C ).

VAS

Eleven studies were included in the VAS analysis. Acupuncture significantly reduced VAS scores compared with control (MD = −1.34, 95% CI [−1.96, −0.71], P < 0.001, I 2 = 64%; Figure 3D ). Acupuncture type did not explain the observed heterogeneity ( P = 0.47; Supplementary Figure 9A ). EA combined with other treatments significantly reduced VAS scores (MD = −1.87, 95% CI [−3.06, −0.67], P = 0.002), whereas EA alone and MA alone did not (EA: MD = −0.85, 95% CI [−1.72, 0.03], P = 0.06; MA: MD = −1.97, 95% CI [−4.30, 0.35], P = 0.10). All 11 studies evaluated acupuncture for treatment rather than prevention. Chemotherapy type also did not explain the heterogeneity (P = 0.53; Supplementary Figure 9B ). Acupuncture significantly reduced VAS scores in the mixed-chemotherapy subgroup (MD = −1.41, 95% CI [−1.95, −0.88], P < 0.001) but not in the taxane subgroup (MD = −0.92, 95% CI [−2.39, 0.56], P = 0.22). The sensitivity analysis did not materially alter the pooled estimate ( Supplementary Figure 10 ). Begg’s ( P = 0.815) and Egger’s ( P = 0.172) tests showed no evidence of publication bias ( Supplementary Figure 11 ).

BPI-WPS

Two studies were included in the Brief Pain Inventory Worst Pain Score (BPI-WPS) analysis, which showed no statistically significant difference between acupuncture and control in worst pain severity (MD = −0.59, 95% CI [−1.45, 0.28], P = 0.18, I 2 = 0%; Figure 3E ).

Clinical Effectiveness Rate

Twenty studies were included in the ER analysis. The likelihood of meeting the study-defined effectiveness criterion was significantly higher with acupuncture than with control (RR = 1.74, 95% CI [1.57, 1.93], P < 0.001, I 2 = 48%; Figure 4 ). Acupuncture type did not explain the observed heterogeneity in ER ( P = 0.87; Supplementary Figure 12A ). EA combined with other therapies had the largest effect estimate (RR = 1.95, 95% CI [1.38, 2.76], P < 0.001); the estimates for EA alone, MA alone, and MA combined with other therapies were also statistically significant (EA: RR = 1.77, 95% CI [1.42, 2.22], P < 0.001; MA: RR = 1.64, 95% CI [1.28, 2.09], P < 0.001; MA + other: RR = 1.73, 95% CI [1.50, 1.99], P < 0.001). Intervention purpose explained some of the between-study heterogeneity ( P = 0.02; Supplementary Figure 12B ). Therapeutic RCTs defined effectiveness as a decrease of ≥1 grade on a neuropathy assessment scale or an increase of ≥2 points on a patient-reported outcome scale; preventive RCTs defined effectiveness as the non-occurrence of CIPN or a grade of 0 on a neuropathy assessment scale. The ER was significantly higher with acupuncture in both therapeutic (RR = 1.63, 95% CI [1.46, 1.82], P < 0.001) and preventive trials (RR = 2.35, 95% CI [1.76, 3.14], P < 0.001), with a larger effect estimate in the preventive subgroup. Chemotherapy type also explained some of the heterogeneity in ER ( P = 0.04; Supplementary Figure 12C ). Acupuncture was associated with significantly higher ERs in the taxane, mixed-chemotherapy, and oxaliplatin subgroups (taxane: RR = 2.31, 95% CI [1.77, 3.01], P < 0.001; mixed: RR = 1.58, 95% CI [1.35, 1.86], P < 0.001; oxaliplatin: RR = 1.57, 95% CI [1.35, 1.83], P < 0.001). The sensitivity analysis did not materially alter the pooled estimate ( Supplementary Figure 13 ). The funnel plot was asymmetric, and both Begg’s ( P = 0.019) and Egger’s ( P = 0.029) tests indicated potential publication bias. After five studies were imputed using the trim-and-fill method, the pooled RR decreased from 1.74 to 1.52 ( Supplementary Figures 14 and 15 ). Figure 4 Forest plots of meta-analysis result for clinical effective rate. Forest plot of 20 studies comparing acupuncture vs control. Table lists study names, events/total for each group. Key studies: C. Lu 2024 (87/114 vs 18/38), H. X. Wei 2014 (15/30 vs 3/30), C. L. Li 2023 (26/29 vs 18/29), Y. Yao 2024 (19/23 vs 7/25), W. J. Chen 2025 (29/35 vs 19/36). Graph shows risk ratio with confidence intervals, scale from 0.2 to 14. Acupuncture on left, control on right. Right columns show weight % and risk ratio: 10.6% and 1.61 (1.13-2.29), 1.2% and 5.00 (1.61-15.50), 7.1% and 1.44 (1.06-1.97), 2.6% and 2.95 (1.53-5.69), 7.4% and 1.57 (1.11-2.21). Total acupuncture events: 674, control: 648. Overall weight: 100%, pooled risk ratio: 1.74 (1.57-1.93). Heterogeneity test: tau superscript 2=0.04, chi superscript 2=36.23, df=19, P=0.010, I superscript 2=48%. Overall effect test: Z=10.50, P<0.001. A forest plot comparing acupuncture versus control, with most studies favoring acupuncture.

Nerve Conduction Studies (NCS)

Sensory Nerve Conduction Velocity

Across 33 studies, acupuncture significantly improved sensory nerve conduction velocity (SCV) compared with control (SMD = 0.66, 95% CI [0.45, 0.88], P < 0.001, I 2 = 76%; Figure 5A ). By acupuncture type ( Supplementary Figure 16A ), neither EA alone nor MA alone significantly improved SCV (EA: SMD = 0.46, 95% CI [−0.06, 0.99], P = 0.08, I 2 = 80%; MA: SMD = 0.09, 95% CI [−0.19, 0.37], P = 0.54, I 2 = 0%), whereas both EA and MA combined with other treatments did (EA + other: SMD = 1.55, 95% CI [1.18, 1.91], P < 0.001, I 2 = 34%; MA + other: SMD = 0.76, 95% CI [0.57, 0.94], P < 0.001, I 2 = 54%). Intervention purpose did not explain the heterogeneity in SCV ( P = 0.13; Supplementary Figure 16B ). Acupuncture significantly improved SCV in both therapeutic and preventive trials (therapeutic: SMD = 0.6, 95% CI [0.32, 0.88], P < 0.001, I 2 = 79%; preventive: SMD = 0.87, 95% CI [0.66, 1.08], P < 0.001, I 2 = 33%). Chemotherapy type explained some of the heterogeneity in SCV (P = 0.009; Supplementary Figure 16C ), and acupuncture significantly improved SCV in the mixed-chemotherapy, capecitabine, oxaliplatin, and taxane subgroups (mixed: SMD = 0.51, 95% CI [0.10, 0.93], P < 0.001; capecitabine: SMD = 0.71, 95% CI [0.45, 0.97], P < 0.001; oxaliplatin: SMD = 1.36, 95% CI [0.99, 1.73], P < 0.001; taxane: SMD = 0.74, 95% CI [0.53, 0.94], P < 0.001). Nerve type did not explain the heterogeneity in SCV ( P = 0.24; Supplementary Figure 16D ). Acupuncture did not significantly improve superficial peroneal nerve (SPN) SCV (SMD = 0.46, 95% CI [−0.3, 1.23], P = 0.24, I 2 = 85%) but significantly improved SCV in the common peroneal nerve (CPN), median nerve (MN), tibial nerve (TN), and sural nerve (SN) (CPN: SMD = 0.98, 95% CI [0.52, 1.45], P < 0.001, I 2 = 75%; MN: SMD = 0.64, 95% CI [0.24, 1.04], P = 0.02, I 2 = 83%; TN: SMD = 1.01, 95% CI [0.46, 1.57], P < 0.001, I 2 = 58%; SN: SMD = 0.44, 95% CI [0.13, 0.76], P = 0.006, I 2 = 31%). The sensitivity analysis did not materially alter the pooled estimate ( Supplementary Figure 17 ). The funnel plot was symmetrical, and Begg’s ( P = 0.114) and Egger’s ( P = 0.101) tests showed no evidence of publication bias ( Supplementary Figure 18 ). Figure 5 Forest plots of meta-analysis result for NCS. Image A shows a forest plot comparing acupuncture to control, with most studies favoring acupuncture. The x-axis, labeled ′Standardised Mean Difference,′ ranges from -1 to 3. The pooled result is 0.66 with a 95% confidence interval of 0.45 to 0.88. Image B, with fewer studies, also shows a positive effect, with a pooled result of 0.78 and a 95% confidence interval of 0.49 to 1.08. Image C indicates a positive effect with a pooled result of 0.43 and a 95% confidence interval of 0.23 to 0.63. Image D shows a near-zero effect, with a pooled result of -0.03 and a 95% confidence interval of -0.37 to 0.32. All images have the x-axis labeled ′Standardised Mean Difference,′ ranging from -1 to 3, except Image D, which ranges from -1 to 2. The y-axis lists study names and years. Four plots: acupuncture vs control. Effects positive in A-C, near zero in D. Notes : ( A ) SCV; subscript a:peroneal nerve(PN); b: median nerve(MN); c: tibial nerve(TN); ( B ) MCV; subscript a:peroneal nerve(PN); b: median nerve(MN); c: sural nerve(SN); d superficial peroneal nerve(SPN); e: tibial nerve(TN); ( C ) amplitude; subscript a: sural nerve(SN); b: median nerve(MN);c: tibial nerve(TN); d: Common Peroneal Nerve(CPN); ( D ) OL; subscript a: tibial nerve(TN); b: sural nerve(SN); c: median nerve(MN). The subscript numbers 1, 2, and 3 are only used to differentiate the treatment arms within the same RCT trial.

Motor Nerve Conduction Velocity

Across 22 studies, acupuncture significantly improved motor nerve conduction velocity (MCV) compared with control (SMD = 0.78, 95% CI [0.49, 1.08], P < 0.001, I 2 = 84%; Figure 5B ). Acupuncture type explained some of the heterogeneity ( P < 0.001; Supplementary Figure 19A ), but neither EA alone nor MA alone significantly improved MCV (EA: SMD = 0.75, 95% CI [−0.4, 1.9], P = 0.2, I 2 = 93%; MA: SMD = 0.11, 95% CI [−0.18, 0.4], P = 0.47, I 2 = 0%). Both EA and MA combined with other treatments significantly improved MCV compared with control (EA + other: SMD = 1.75, 95% CI [1.05, 2.45], P < 0.001, I 2 = 80%; MA + other: SMD = 0.77, 95% CI [0.55, 0.98], P < 0.001, I 2 = 60%). By intervention purpose ( Supplementary Figure 19B ), therapeutic acupuncture significantly improved MCV (SMD = 0.87, 95% CI [0.53, 1.21], P < 0.001, I 2 = 86%), whereas preventive acupuncture did not (SMD = 0.5, 95% CI [−0.07, 1.08], P = 0.09, I 2 = 84%); intervention purpose did not explain the heterogeneity ( P = 0.29). Chemotherapy type explained some of the heterogeneity in MCV ( P = 0.02; Supplementary Figure 19C ), and acupuncture significantly improved MCV across the mixed-chemotherapy and single-agent subgroups (mixed: SMD = 0.78, 95% CI [0.28, 1.28], P < 0.001, I 2 = 89%; taxane: SMD = 0.55, 95% CI [0.16, 0.94], P = 0.006, I 2 = 72%; oxaliplatin: SMD = 1.4, 95% CI [1.03, 1.77], P < 0.001, I 2 = 0%; capecitabine: SMD = 1.05, 95% CI [0.5, 1.6], P < 0.001, I 2 = 76%). Nerve type did not explain the heterogeneity in MCV ( P = 0.78; Supplementary Figure 19D ), and acupuncture significantly improved MCV in the MN, CPN, and TN (MN: SMD = 0.88, 95% CI [0.37, 1.4], P < 0.001, I 2 = 89%; CPN: SMD = 0.66, 95% CI [0.25, 1.07], P = 0.002, I 2 = 80%; TN: SMD = 0.84, 95% CI [0.26, 1.41], P = 0.005, I 2 = 69%). The sensitivity analysis did not materially alter the pooled estimate ( Supplementary Figure 20 ). The funnel plot was symmetrical, and Begg’s ( P = 0.844) and Egger’s ( P = 0.855) tests showed no evidence of publication bias ( Supplementary Figure 21 ).

Compound Motor Action Potential Amplitude

Across 14 studies, acupuncture significantly improved compound motor action potential amplitude compared with control (SMD = 0.43, 95% CI [0.23, 0.63], P < 0.001, I 2 = 0%; Figure 5C ). By acupuncture type ( Supplementary Figure 22A ), neither EA alone nor MA alone significantly improved amplitude (EA: SMD = 0.23, 95% CI [−0.2, 0.66], P = 0.29, I 2 = 0%; MA: SMD = 0.28, 95% CI [−0.03, 0.58], P = 0.07, I 2 = 5%), whereas MA combined with other treatments did (SMD = 0.76, 95% CI [0.41, 1.11], P < 0.001, I 2 = 0%). Chemotherapy type explained some of the heterogeneity in amplitude ( P = 0.04; Supplementary Figure 22B ). Acupuncture did not significantly improve amplitude in the taxane subgroup (SMD = −0.09, 95% CI [−0.6, 0.43], P = 0.75, I 2 = 0%) but significantly improved it in the mixed-chemotherapy subgroup (SMD = 0.52, 95% CI [0.3, 0.74], P < 0.001, I 2 = 0%). By nerve type ( Supplementary Figure 22C ), acupuncture did not significantly improve SN or TN amplitude (SN: SMD = 0.39, 95% CI [−0.01, 0.78], P = 0.06, I 2 = 9%; TN: SMD = 0.3, 95% CI [−0.23, 0.83], P = 0.27, I 2 = 0%). In contrast, acupuncture significantly improved MN and CPN amplitude (MN: SMD = 0.39, 95% CI [0.04, 0.74], P = 0.03, I 2 = 0%; CPN: SMD = 0.59, 95% CI [0.2, 0.98], P = 0.003, I 2 = 0%). All included interventions were therapeutic rather than preventive. The sensitivity analysis did not materially alter the pooled estimate ( Supplementary Figure 23 ). The funnel plot showed potential asymmetry: Egger’s test was statistically significant ( P = 0.004), whereas Begg’s test was not ( P = 0.250; Supplementary Figure 24 ).

Onset Latency

Across five studies, acupuncture did not significantly improve onset latency (OL) compared with control (SMD = −0.03, 95% CI [−0.37, 0.32], P = 0.88; Figure 5D ). All five studies evaluated MA for treatment rather than prevention ( Supplementary Figure 25A ). Chemotherapy type did not explain the subgroup difference ( P = 0.58), and acupuncture did not significantly improve OL in either the mixed-chemotherapy or taxane subgroup (mixed: SMD = −0.11, 95% CI [−0.56, 0.34], P = 0.64, I 2 = 0%; taxane: SMD = 0.08, 95% CI [−0.44, 0.61], P = 0.75, I 2 = 0%). Nerve type did not explain the subgroup difference ( P = 0.41; Supplementary Figure 25B ), and acupuncture did not significantly improve OL in either the TN or SN (TN: SMD = −0.19, 95% CI [−0.7, 0.33], P = 0.47, I 2 = 0%; SN: SMD = 0.1, 95% CI [−0.35, 0.56], P = 0.66, I 2 = 0%). The sensitivity analysis did not materially alter the pooled estimate ( Supplementary Figure 26 ).

Quantitative Sensory Testing

The included trials reported four QST outcomes: tactile detection threshold (TDT), vibrational detection threshold (VDT), thermal detection threshold (THDT), and thermal threshold (THT). VDT, THDT, and THT were reported only by W. I. Zhi 2023, with no statistically significant between-group differences. Six studies reported TDT; the pooled analysis showed no statistically significant effect of acupuncture (SMD = −0.27, 95% CI [−0.63, 0.09], P = 0.14; Figure 6 ). By acupuncture type, neither EA nor MA significantly improved TDT compared with control (EA: SMD = −0.17, 95% CI [−0.6, 0.25], P = 0.43, I 2 = 0%; MA: SMD = −0.51, 95% CI [−1.18, 0.16], P = 0.14, I 2 = 0%; Supplementary Figure 27A ). By chemotherapy type ( Supplementary Figure 27B ), acupuncture did not significantly improve TDT in either the mixed-chemotherapy or taxane subgroup (mixed: SMD = −0.17, 95% CI [−0.6, 0.25], P = 0.43, I 2 = 0%; taxane: SMD = −0.51, 95% CI [−1.18, 0.16], P = 0.14, I 2 = 0%). By measurement site ( Supplementary Figure 27C ), acupuncture did not significantly improve TDT in either the hands or feet (hands: SMD = −0.34, 95% CI [−0.85, 0.18], P = 0.2, I 2 = 0%; feet: SMD = −0.21, 95% CI [−0.72, 0.3], P = 0.43, I 2 = 0%). Both trials reporting the additional QST outcomes evaluated therapeutic acupuncture. The sensitivity analysis did not materially alter the TDT estimate ( Supplementary Figure 28 ). Figure 6 Forest plots of meta-analysis result for TDT. A forest plot compares tactile detection thresholds between Acupuncture and Control groups. It includes columns for study details, mean, standard deviation, total participants, weight percentage and Standardised Mean Difference (SMD) with a 95% confidence interval (CI). The effect axis ranges from -2 to 2, labeled with Acupuncture on the left and Control on the right. Six studies are plotted with SMDs and CIs: -0.31 (95% CI: -1.15 to 0.53, weight 18.6%), -0.06 (-0.94 to 0.82, weight 16.9%), -0.71 (-1.68 to 0.25, weight 14.0%), -0.24 (-1.07 to 0.60, weight 18.7%), -0.07 (-0.95 to 0.81, weight 16.9%) and -0.32 (-1.26 to 0.62, weight 14.9%). The total row shows 48 in Acupuncture and 82 in Control, with a pooled SMD of -0.27 (95% CI: -0.63 to 0.09, weight 100%). Heterogeneity is low with tau superscript 2 = 0, chi superscript 2 = 1.25, df = 5, P = 0.94 and I superscript 2 = 0%. The overall effect is Z = -1.47, P = 0.14. A forest plot of tactile detection threshold comparing acupuncture versus control, with pooled effect near zero. Notes : subscript a: feets; b: hands.

Certainty of Evidence

The certainty of evidence for all three outcomes assessed using GRADE—FACT-Ntx, EORTC QLQ-CIPN20, and ER—was rated very low, indicating that the true effects may differ substantially from the estimates. Certainty was downgraded primarily because of serious risk of bias and indirectness; outcome-specific reasons are presented in Figure 7 . Figure 7 GRADE evidence. The table compares acupuncture versus control across three outcomes, all rated with very low certainty. 1. FACT-Ntx: 10 randomized trials show serious risk of bias and indirectness, with no serious inconsistency or imprecision. Sample sizes: acupuncture 311, control 253. Mean difference (MD) is 1.43 lower (range: 1.01 to 0.95 lower). Importance: Important. 2. EORTC QLQ-CIPN20: 8 trials indicate serious issues in bias, inconsistency, indirectness and imprecision. Sample sizes: acupuncture 183, control 184. MD is 3.63 lower (range: 5.97 to 1.28 lower). Importance: Important. 3. Effective rate: 20 trials with serious bias and indirectness, but no serious inconsistency or imprecision. Publication bias is suspected. Acupuncture: 478.56/714 (71.0%), Control: 254.5/648 (39.3%). Relative effect (RR) is 1.74 (range: 1.57 to 1.93). Absolute effect: 291 more per 1,000 (range: 224 to 365 more). Importance: Critical. A table of three outcomes showing acupuncture effects and very low certainty ratings throughout. Notes : ( a ) Trials contributing 76.8% of the pooled weight had approximately 10% missing participant data, with no reported reasons for attrition and no clear specification of the primary analysis population; ( b ) More than 50% of the included studies were at risk of deviations from intended interventions and selection bias; ( c ) Substantial heterogeneity was observed (p < 0.05); ( d ). The confidence interval was wide; ( e ) Begg’test P = 0.019; Egger’test P = 0.029. Abbreviations : Cl, confidence interval; MD, mean difference; RR, risk ratio.

Trial Sequential Analysis

TSA used the following parameters: 61 a type I error (α) of 5%, a type II error (β) of 10%, the DerSimonian–Laird random-effects model, and a constant continuity correction of 0.5 per group for studies with zero events. Between-study heterogeneity was adjusted using the diversity estimate (D 2 ). For FACT-Ntx and EORTC QLQ-CIPN20, the variance and minimal clinically important difference (MCID) were based on values observed and pooled in the current meta-analysis. For ER, the control-group event rate was estimated from the pooled incidence in the current meta-analysis. A 25% relative increase in ER was prespecified as the minimally important effect. For FACT-Ntx, the cumulative Z-curve crossed both the monitoring boundary and the RIS, supporting the stability of the pooled finding ( Figure 8 ). For EORTC QLQ-CIPN20 and ER, the cumulative Z-curves crossed the monitoring boundaries but did not reach the RIS at the prespecified 25% relative-effect threshold. These findings suggest that the pooled effects were unlikely to be explained entirely by random error; however, the information size remained insufficient for EORTC QLQ-CIPN20 and ER. Figure 8 TSA result. The image A showing a line graph titled, TSA is a Two sided graph. The x axis label is, Number of patients left parenthesis Linear scaled right parenthesis, with labeled value 144 and a vertical line labeled, TSA equals 178. The y axis label is, Cumulative Z Score, ranging from negative 8 to 8. Side labels read, Favours Acupuncture and Favours Control. Two dashed horizontal lines are at 2 and negative 2. A blue z curve rises from about 0 at the left to about 3.2 near 144, then to about 4.0, then to about 5.8 and stays near 5.8 to the right. A red monitoring boundary for benefit starts near 8, drops to about 4.5, then to about 2.2; a lower red boundary rises from about negative 8 to about negative 4.0 to about negative 2.2. The image B showing a line graph titled, TSA is a Two sided graph. The x axis label is, Number of patients left parenthesis Linear scaled right parenthesis, with labeled value 367 and a vertical line labeled, TSA equals 485. The y axis label is, Cumulative Z Score, ranging from negative 8 to 8. Side labels read, Favours Acupuncture and Favours Control. Dashed horizontal lines are at 2 and negative 2. The blue z curve rises from about 0 to about 2.0, stays near 2.0, then rises to about 4.0. The red monitoring boundary for benefit declines from about 8 to about 4.5 to about 2.3; the lower red boundary rises from about negative 8 to about negative 4.0 to about negative 2.3. The image C showing a line graph titled, TSA is a Two sided graph. The x axis label is, Number of patients left parenthesis Linear scaled right parenthesis, with labeled value 1303 and a vertical line labeled, TSA equals 2260. The y axis label is, Cumulative Z Score, ranging from negative 8 to 8. Side labels read, Favours Control and Favours Acupuncture. Dashed horizontal lines are at 2 and negative 2. The blue z curve drops from about 0 to about negative 8 and then stays near negative 8 across multiple points. The red boundary above declines from about 8 to about 4.5 and then to about 3.0. The red boundary below rises from about negative 8 to about negative 4.5 and then to about negative 3.0, with the text, Monitoring boundary for benefit. A multi line graph showing cumulative Z score versus number of patients across three subplots. Notes : ( A ) Trial sequential analysis of acupuncture on FACT-Ntx; ( B ) Trial sequential analysis of acupuncture on EORTC QLQ-CIPN20; ( C ) Trial sequential analysis of acupuncture on ER.

Discussion

Summary of Main Findings

This systematic review and meta-analysis of 40 studies involving 2,496 patients found that acupuncture may provide clinical benefits for CIPN; TSA supported the stability of the FACT-Ntx finding but yielded less conclusive evidence for EORTC QLQ-CIPN20 and ER because the required information sizes were not reached. However, GRADE rated the certainty of evidence for the three assessed outcomes—FACT-Ntx, EORTC QLQ-CIPN20, and ER—as very low; therefore, the findings should be interpreted cautiously. For CIPN symptoms, acupuncture significantly improved several validated scale scores compared with control. Specifically, acupuncture significantly improved FACT-Ntx (MD = −1.43), EORTC QLQ-CIPN20 (MD = −5.01), NRS (MD = −0.97), and VAS (MD = −1.34) scores but did not significantly improve BPI-WPS. For ER, acupuncture increased the likelihood of a clinical response (RR = 1.74); after trim-and-fill adjustment, the effect remained statistically significant and in the same direction (RR = 1.52). For objective neurophysiological outcomes, acupuncture significantly improved SCV (SMD = 0.66), MCV (SMD = 0.78), and compound motor action potential amplitude (SMD = 0.43), whereas no significant effects were found for OL or TDT.

The subgroup analyses provided preliminary, hypothesis-generating evidence relevant to clinical decision-making. Acupuncture type may have influenced the effects on FACT-Ntx, SCV, and MCV. EA combined with other treatments yielded larger effect estimates for VAS, ER, SCV, and MCV, consistent with previous findings. 62 Differences by intervention purpose were most apparent for ER, for which preventive trials yielded a larger effect estimate than therapeutic trials. This finding may reflect the smaller number of trials and participants in the preventive subgroup. Among outcomes analyzed by intervention purpose, therapeutic acupuncture significantly improved FACT-Ntx, NRS, VAS, ER, SCV, MCV, and compound motor action potential amplitude, whereas OL and TDT did not improve significantly. Chemotherapy type may have contributed to subgroup differences in ER, SCV, MCV, and compound motor action potential amplitude. For taxane-related CIPN, acupuncture significantly improved ER but not compound motor action potential amplitude. For oxaliplatin-induced neuropathy, acupuncture significantly improved both sensory and motor nerve conduction velocities. In the mixed-chemotherapy subgroup, acupuncture significantly improved ER, SCV, MCV, and compound motor action potential amplitude; however, these subgroup findings do not establish efficacy across all chemotherapy regimens. For SCV, MCV, OL, and TDT, subgroup differences by nerve or measurement site were not statistically significant. These findings may be compatible with a systemic rather than exclusively local effect, but the subgroup analyses cannot establish the underlying mechanism.

The generally high risk of bias across the included studies warrants caution. Beyond the inherent difficulty of blinding acupuncture practitioners, key concerns included unclear reporting of random sequence generation and allocation concealment, lack of outcome-assessor blinding, and absence of prospectively registered protocols in some studies. These limitations may have introduced selection and measurement bias. Several pooled analyses also showed moderate-to-high heterogeneity. Although prespecified multidimensional subgroup analyses were conducted, some heterogeneity remained unexplained. Potential sources include between-study differences in tumor type, treatment regimen, baseline symptom severity, acupuncture protocols (acupoints, treatment courses, and parameters), and NCS equipment and methods.

Implications for Practice

No clinical consensus has been established regarding the optimal acupuncture protocol for CIPN. Our findings showed that MA alone significantly improved FACT-Ntx, whereas MA combined with other treatments significantly improved EORTC QLQ-CIPN20; both scales assess neurosensory symptoms and motor dysfunction in patients with CIPN. EA alone did not significantly improve FACT-Ntx but significantly improved EORTC QLQ-CIPN20. Previous research has also reported limitations of EA for relieving CIPN-related pain. 27 , 63 Previous research has suggested that traditional MA may be more effective for CIPN-related pain. 64 EA combined with other therapies significantly improved VAS scores, and MA combined with other therapies significantly improved NRS scores, both with large effect estimates; however, the supporting studies 24 , 40–42 , 46 , 52 were predominantly conducted in China and may have been limited by small samples, publication bias, and methodological constraints. One study suggested 65 that high-frequency or continuous physical stimulation may increase neuronal excitability and synchronized firing in a pathological nervous system. Electrical-stimulation parameters may therefore influence individual CIPN symptoms differently. 33 , 66

Heterogeneity in EA parameters may have contributed to variation in the efficacy estimates for CIPN. Combining EA with adjunctive therapies, including bloodletting, moxibustion, reduced glutathione, or mecobalamin, may reduce parameter-related variability and produce complementary effects. Li et al reported 67 that, in mice with CIPN, EA may alleviate pain, numbness, and other sensory abnormalities by inhibiting neuronal CCL2 signaling, reducing proinflammatory macrophage infiltration into the dorsal root ganglion, and partially restoring peripheral sensory structures, particularly intraepidermal nerve fibers. Bloodletting and moxibustion have been proposed to regulate local qi and blood and modulate immune inflammation; 68 , 69 reduced glutathione and mecobalamin may exert antioxidant and neuroprotective effects. 70 , 71 These combined approaches may improve CIPN symptoms through complementary effects on inflammation, neural structure, and functional recovery. A clinical study reported 59 that only a small proportion of patients with CIPN showed NCS abnormalities. In some patients, CIPN predominantly affects small nerve fibers (ie, A-δ or C fibers), whereas NCS primarily assesses large nerve fibers and may yield normal results in small-fiber neuropathy. In that study, combined EA significantly improved SCV and MCV compared with control among patients with established NCS abnormalities. The same study suggested that patients with predominantly subjective sensory symptoms, such as pain and numbness, but no marked NCS abnormalities may be more suitable candidates for MA. 59 Conversely, patients with substantial sensory and motor conduction deficits may derive greater benefit from EA combined with other treatments. 17 Although the preventive subgroup yielded larger effect estimates for outcomes with between-group differences, the therapeutic subgroup findings were more consistently statistically significant. Thus, acupuncture may have potential for both preventing and treating CIPN, although these subgroup findings remain exploratory. By chemotherapy type, acupuncture improved ER in taxane-related CIPN and neurophysiological outcomes in oxaliplatin-related CIPN; these subgroup findings may inform future trials but do not establish regimen-specific clinical recommendations.

Acupoint selection is central to traditional clinical acupuncture protocols. Rational selection and combination of acupoints may influence acupuncture efficacy, with individual points serving distinct therapeutic roles in CIPN management. 57 , 72 We summarized and analyzed the most frequently used acupoints ( Figure 9 ). The most frequently used acupoints were Hegu (LI4), Zusanli (ST36), Quchi (LI11), Sanyinjiao (SP6), Taichong (LR3), and local experiential points on the extremities (EX-UE9 and EX-LE10). This pattern suggests that an empirical consensus may be emerging around prescriptions intended to regulate qi and blood circulation and improve distal-limb sensory disturbances. Many frequently used acupoints were located along peripheral nerve pathways or near NCS sites. One RCT indicated that local toe and finger acupoints were particularly important to acupuncture efficacy in CIPN. 17 This finding highlights the potential importance of distal hand and foot points such as EX-UE9 and EX-LE10, which earlier systematic reviews and meta-analyses did not emphasize. 73 Figure 10 summarizes the other acupuncture parameters. Across the 40 included studies, treatment comprised 6–28 sessions (median, 12); one study did not report the number of sessions. Sessions most commonly lasted 30 min (range, 10–45 min). Treatment was most often administered 2–3 times per week (15/40 studies), with an overall range of 1–7 sessions per week; total treatment duration ranged from 1.4 to 12 weeks (median, 4 weeks). Figure 9 Top 15 acupoints by frequency. A circular bar chart showing acupoint frequency. A legend titled, Acupoints, lists LI4, ST36, LI11, LR3, SP6, EX minus LE10, EX minus UE9, GB34, CV6, PC6. Concentric rings are labeled 0, 10, 20, 30, indicating the radial scale from 0 to 30 counts. Ten wedge bars are shown with labels: n equals 32, n equals 31, n equals 26, n equals 23, n equals 23, n equals 19, n equals 15, n equals 13, n equals 11, n equals 8. The longest wedge reaches 32, slightly beyond the 30 ring; the next reaches 31. The remaining wedges reach 26, 23, 23, 19, 15, 13, 11 and 8, decreasing toward the smallest wedge at 8. A circular bar chart showing acupoint frequency counts from 8 to 32. Figure 10 Frequency statistics of acupuncture parameters. A composite of four bar charts: Image A is a horizontal bar chart showing the total number of acupuncture sessions by study author, ranging from 6 to 28 sessions. Image B is a vertical bar chart depicting session duration in minutes, with 30 minutes being the most common duration across 30 studies. Image C is a vertical bar chart showing sessions per week, with 2 to 3 sessions being the most frequent in 15 studies. Image D is a vertical bar chart illustrating treatment duration in weeks, with 4 weeks being the most common duration in 9 studies. A set of bar charts showing acupuncture sessions, session duration, sessions per week and treatment duration.

Future Research

Future research should prioritize rigorously designed, large, multicenter randomized controlled trials to strengthen the evidence base. Such trials could support the development of standardized CIPN acupuncture protocols, including acupoint selection and procedures for MA and EA. Combination therapies and differences among acupuncture modalities warrant further investigation. Our findings suggest that EA combined with other treatments—including pregabalin, mecobalamin, and moxibustion—may produce benefits across several CIPN outcomes despite variation in EA parameters. EA combined with analgesics, moxibustion, or other adjunctive treatments therefore warrants further evaluation. Future studies should also clarify the optimal timing of acupuncture in CIPN management. It remains unclear whether acupuncture is more suitable for prevention or treatment; one study 27 suggested that EA may be more appropriate as a therapeutic intervention than as a preventive adjunct. Future trials could compare the same acupuncture protocol for preventing CIPN and treating established CIPN while evaluating how intervention timing affects long-term outcomes. Finally, we used the Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA) guidelines to evaluate the reporting completeness of the included acupuncture RCTs ( Supplementary Table 11 ). Most trials reported basic acupuncture parameters, such as acupoint names and needle-retention time, but inadequately described the theoretical basis for acupuncture, needle-insertion depth, practitioner background, and treatment setting. Future acupuncture trials for CIPN should follow STRICTA to improve protocol reporting and research reproducibility.

Strengths

This review evaluated the effects of MA and EA on CIPN-specific outcomes, including subjective rating scales and objective neurophysiological measures. The review also incorporated TSA to assess the stability of the meta-analytic findings, providing a complementary perspective on the effects of MA and EA within the existing evidence base. Prespecified subgroup analyses examined intervention type and purpose, the type of chemotherapy associated with CIPN, and objective-outcome measurement sites. These analyses identified potential effect modifiers and sources of heterogeneity and generated hypotheses for tailoring acupuncture strategies to different patient groups.

Limitations

Several limitations should be considered. First, most included trials were conducted in China, potentially limiting generalizability to other populations and healthcare settings. Second, many trials lacked sham-acupuncture controls, making it difficult to distinguish the physiological effects of acupuncture from placebo effects. Third, some studies had a high risk of bias, particularly in randomization, allocation concealment, and outcome-assessor blinding, which may have inflated the treatment-effect estimates. Fourth, some included studies may have inconsistently reported the scoring direction of CIPN measures, particularly the positive and negative scoring conventions of FACT scales. 74 Finally, although the inherent neurotoxicity of individual chemotherapeutic agents is a primary determinant of CIPN risk, 75 , 76 moderate-to-high heterogeneity persisted for some outcomes after subgroup and sensitivity analyses. Unexplored clinical or methodological differences, including variation in control interventions and ER definitions, may have contributed to this heterogeneity.

Conclusion

This systematic review and meta-analysis provides preliminary evidence that acupuncture may benefit patients with CIPN. TSA supported the stability of the FACT-Ntx finding; although the monitoring boundaries were crossed for EORTC QLQ-CIPN20 and ER, neither outcome reached the required information size. Acupuncture type, intervention purpose, and chemotherapy type may influence effect estimates. However, because GRADE rated the certainty of evidence for FACT-Ntx, EORTC QLQ-CIPN20, and ER as very low, higher-quality, geographically diverse randomized controlled trials are needed before MA or EA can be recommended for CIPN prevention or treatment. Further trials should determine the optimal acupuncture modalities, timing, and therapeutic targets and identify patient subgroups that may derive greater benefit. Such evidence could support the development of standardized, personalized acupuncture strategies for different populations with CIPN.

Declaration of Generative AI and AI-Assisted Technologies in the Manuscript Preparation Process

During the preparation of this work, the author(s) used ChatGPT (OpenAI, GPT-5.5) in order to improve the clarity and readability of the language. After using this tool, the author(s) critically reviewed and edited the content as needed and take full responsibility for the content of the published article.

Data Sharing Statement

The original contributions presented in the study are included in the article/ Supplementary Material , further inquiries can be directed to the corresponding author, Shaoyang Cui.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas, took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

Dr Zhifeng Pan reports Support for the manuscript. The authors report Support for the manuscript from The Sanming Project of Medicine in Shenzhen (No. SZZYSM202311011), 2025 Basic and Applied Basic Research Program (Young PhD “Qihang” Project) of Guangzhou Municipal Science and Technology Bureau (No. 2025A04J4197), 2025 Traditional Chinese Medicine Scientific Research Project of Guangdong Provincial Administration of Traditional Chinese Medicine (No. 20251141) and Shenzhen Chinese Medicine Key Specialty Construction Project (No. ZYTS019), during the conduct of the study. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

1. Force LM , Kocarnik JM , May ML , et al. The global, regional, and national burden of cancer, 1990–2023, with forecasts to 2050: a systematic analysis for the global burden of disease study 2023 . Lancet . 2025 ; 406 ( 10512 ): 1565 – 23 . doi: 10.1016/S0140-6736(25)01635-6 PMC12687902 41015051

2. Lustberg MB , Kuderer NM , Desai A , Bergerot C , Lyman GH . Mitigating long-term and delayed adverse events associated with cancer treatment: implications for survivorship . Nat Rev Clin Oncol . 2023 ; 20 ( 8 ): 527 – 542 . doi: 10.1038/s41571-023-00776-9 PMC10211308 37231127

3. Stubblefield MD , Burstein HJ , Burton AW , et al. NCCN task force report: management of neuropathy in cancer . J Natl Compr Cancer Netw . 2009 ; 7 ( 5 ): S1 – S26 . doi: 10.6004/jnccn.2009.0078 19755042

4. Desforges AD , Hebert CM , Spence AL , et al. Treatment and diagnosis of chemotherapy-induced peripheral neuropathy: an update . Biomed Pharmacother . 2022 ; 147 : 112671 . doi: 10.1016/j.biopha.2022.112671 PMC11118018 35104697

5. Speck RM , Sammel MD , Farrar JT , et al. Impact of chemotherapy-induced peripheral neuropathy on treatment delivery in nonmetastatic breast cancer . J Oncol Pract . 2013 ; 9 ( 5 ): e234 – 240 . doi: 10.1200/JOP.2012.000863 23943894

6. Bhatnagar B , Gilmore S , Goloubeva O , et al. Chemotherapy dose reduction due to chemotherapy induced peripheral neuropathy in breast cancer patients receiving chemotherapy in the neoadjuvant or adjuvant settings: a single-center experience . SpringerPlus . 2014 ; 3 : 366 . doi: 10.1186/2193-1801-3-366 PMC4117856 25089251

7. Smith EML , Pang H , Cirrincione C , et al. Effect of duloxetine on pain, function, and quality of life among patients with chemotherapy-induced painful peripheral neuropathy: a randomized clinical trial . Jama . 2013 ; 309 ( 13 ): 1359 – 1367 . doi: 10.1001/jama.2013.2813 PMC3912515 23549581

8. Yang M , Wan Y , Jiang X , et al. Electro-acupuncture promotes accumulation of paclitaxel by altering tumor microvasculature and microenvironment in breast cancer of mice . Front Oncol . 2019 ; 9 : 576 . doi: 10.3389/fonc.2019.00576 PMC6614178 31312613

9. Liu W , He Y , Zhang H , Xu N , Zhou D . Review of the Application and Research of Acupuncture and Moxibustion in the Prevention and Treatment of Cancer . J Oncol Chin Med . 2023 ; 5 ( 1 ): 1 – 6 . doi: 10.19811/j.cnki.ISSN2096-6628.2023.01.001

10. Huang MC , Chang SC , Liao WL , et al. Acupuncture may help to prevent chemotherapy-induced peripheral neuropathy: a randomized, sham-controlled, single-blind study . Oncologist . 2023 ; 28 ( 6 ): e436 – e447 . doi: 10.1093/oncolo/oyad065 PMC10243779 36971468

11. Mao JJ , Liou KT , Baser RE , et al. Effectiveness of electroacupuncture or auricular acupuncture vs usual care for chronic musculoskeletal pain among cancer survivors: the PEACE randomized clinical trial . JAMA Oncol . 2021 ; 7 ( 5 ): 720 – 727 . doi: 10.1001/jamaoncol.2021.0310 PMC7974834 33734288

12. de Valois B , Young T , Zollman C , et al. Acupuncture in cancer care: recommendations for safe practice (peer-reviewed expert opinion) . Support Care Cancer . 2024 ; 32 ( 4 ): 229 . doi: 10.1007/s00520-024-08386-6 PMC10940387 38483623

13. Loprinzi CL , Lacchetti C , Bleeker J , et al. Prevention and management of chemotherapy-induced peripheral neuropathy in survivors of adult cancers: ASCO guideline update . J Clin Oncol . 2020 ; 38 ( 28 ): 3325 – 3348 . doi: 10.1200/JCO.20.01399 32663120

14. Bao T , Patil S , Chen C , et al. Effect of acupuncture vs sham procedure on chemotherapy-induced peripheral neuropathy symptoms: a randomized clinical trial . JAMA Netw Open . 2020 ; 3 ( 3 ): e200681 . doi: 10.1001/jamanetworkopen.2020.0681 PMC7066475 32159808

15. Bao T , Baser R , Lin C , Li Q , Piulson L , Zhi I . Secondary outcomes in a randomized controlled trial of acupuncture versus sham acupuncture and usual care in solid tumor survivors with chemotherapy-induced peripheral neuropathy . Cancer Res . 2021 ; 81 ( 4 ): 1 .

16. Stringer J , Ryder WD , Mackereth PA , Misra V , Wardley AM . A randomised, pragmatic clinical trial of ACUpuncture plus standard care versus standard care alone For Chemotherapy Induced peripheral Neuropathy (ACUFOCIN) . Eur J Oncol Nurs . 2022 ; 60 : 102171 . doi: 10.1016/j.ejon.2022.102171 PMC9592667 35952460

17. Friedemann T , Kark E , Cao N , et al. Acupuncture improves chemotherapy-induced neuropathy explored by neurophysiological and clinical outcomes - the randomized, controlled, cross-over ACUCIN trial . Phytomed . 2022 ; 104 : 154294 . doi: 10.1016/j.phymed.2022.154294 35785559

18. Lyu X , Yuan X , Fan Z , et al. Evidence map of systematic reviews on acupuncture and moxibustion for prevention and management of cancer-related conditions . World J Acupunct-Moxibustion . 2025 ; 35 ( 4 ): 290 – 302 .

19. Higgins J , Thomas J , Chandler J , Cumpston M , Li T , Page M . eds. Cochrane Handbook for Systematic Reviews of Interventions Version 6.5 (Updated August 2024). Version 6.5. Cochrane ; 2024 . Available from: www.cochrane.org/handbook . Accessed September 15 , 2026 .

20. Borenstein M , Hedges LV , Higgins JPT , Rothstein HR . Introduction to Meta-Analysis . 1st . Wiley ; 2009 . doi: 10.1002/9780470743386

21. McGrath S , Zhao X , Steele R , Thombs BD , Benedetti A . DEPRESsion Screening Data (DEPRESSD) Collaboration. Estimating the sample mean and standard deviation from commonly reported quantiles in meta-analysis . Stat Methods Med Res . 2020 ; 29 ( 9 ): 2520 – 2537 . doi: 10.1177/0962280219889080 PMC7390706 32292115

22. Guyatt GH , Oxman AD , Vist GE , et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations . BMJ . 2008 ; 336 ( 7650 ): 924 – 926 . doi: 10.1136/bmj.39489.470347.AD PMC2335261 18436948

23. Xu J , Zhang Q , Xue N , Zhang C , Xu W . The therapeutic effect of acupuncture in the treatment of chemotherapy-induced peripheral neuropathy: a randomized controlled trial . Front Oncol . 2025 ; 15 : 1 – 12 . doi: 10.3389/fonc.2025.1500410 PMC12040932 40308502

24. Sui M , Xu N , Xiang Y , Shu G , Jin D , Yan T . The Influence of Electrical Acupuncture on the Chemotherapy-induced Neuropathic Pain and Quality of Life . Chin J Integrat Med Cardio-Cerebrovascular Dis . 2018 ; 16 ( 10 ): 1331 – 1333 . doi: 10.12102/j.issn.1672-1349.2018.10.06

25. Tian Y , Huang Y , Jia Y . The curative effect of warm acupuncture and moxibustion on peripheral neurotoxicity caused by oxaliplatin . Tianjin J Traditional Chin Med . 2011 ; 28 ( 3 ): 212 – 213 .

26. Lu K , Tang J . The Clinical Study on the Treatment of Oxaliplatin Neurotoxicity with Yiqi Huoxue Prescription and Acupuncture Therapy . Clin J Tradit Chin Med . 2020 ; 32 ( 1 ): 139 – 142 . doi: 10.16448/j.cjtcm.2020.0136

27. Greenlee H , Crew KD , Capodice J , et al. Randomized sham-controlled pilot trial of weekly electro-acupuncture for the prevention of taxane-induced peripheral neuropathy in women with early stage breast cancer . Breast Cancer Res Treat . 2016 ; 156 ( 3 ): 453 – 464 . doi: 10.1007/s10549-016-3759-2 PMC4924571 27013473

28. Chen W , Wu T , Lu C , Shen Q , Li R , Fang J . Observation on the efficacy of electroacupuncture in the treatment of oxaliplatin-induced peripheral neuropathy and its impact on patients’ quality of life . Zhejiang J Integrated Tradit Chin Western Med . 2025 ; 35 ( 3 ): 233 – 238 . doi: 10.3969/j.issn.1005-4561.2025.03.009

29. Liu X , Zhang Y , Huang Y , Chao T . Observation of the Efficacy of Finger Exercises Combined with Warm Needle Acupuncture in the Treatment of Peripheral Neuropathy Induced by Chemotherapy for Breast Cancer . Chin J Phys Med Rehabil . 2023 ; 45 ( 4 ): 355 – 357 . doi: 10.3760/cma.j.issn.0254-1424.2023.04.015

30. He L , Yang J , Bai L , et al. Observation of the Efficacy of Acupuncture Combined with External Washing of Chinese Herbal Medicine for Promoting Blood Circulation and Removing Blood Stasis in the Treatment of Oxaliplatin-Induced Peripheral Neurotoxicity . Mod J Integr Traditional Chinese and Western Med . 2024 ; 33 ( 20 ): 2868 – 2871 . doi: 10.3969/j.issn.1008-8849.2024.20.017

31. Zhi WI , Baser RE , Talukder D , Mei YZ , Harte SE , Bao T . Mechanistic and thermal characterization of acupuncture for chemotherapy-induced peripheral neuropathy as measured by quantitative sensory testing . Breast Cancer Res Treat . 2023 ; 197 ( 3 ): 535 – 545 . doi: 10.1007/s10549-022-06846-3 PMC11218902 36527520

32. Wang L , Xu X , Yao L , et al. Intervention effect of acupuncture and moxibustion in treatment of chemotherapy-induced peripheral neuropathy in patients with multiple myeloma . J Clin Med Pract . 2022 ; 26 ( 2 ): 28 – 30 . doi: 10.7619/jcmp.20212712

33. Lu C , Fen X , Shen Q , et al. Electroacupuncture with different frequencies for paclitaxel-induced peripheral neuropathy: a randomized controlled trial . Chin Acupuncture Moxibus . 2024 ; 44 ( 10 ): 1139 – 1145 . doi: 10.13703/j.0255-2930.20240123-0001 39401811

34. Yao Y , Zhang X , Ge R , Shin IH , Yao C . Electroacupuncture for the prevention of chemotherapy-induced peripheral neuropathy: a randomized controlled trial . Chin Acupuncture Moxibus . 2024 ; 44 ( 12 ): 1388 – 1394 . doi: 10.13703/j.0255-2930.20240222-k0001 39658376

35. Chan K , Lui L , Lam Y , et al. Efficacy and safety of electroacupuncture for oxaliplatin-induced peripheral neuropathy in colorectal cancer patients: a single-blinded, randomized, sham-controlled trial . Acupunct Med . 2023 ; 41 ( 5 ): 268 – 283 . doi: 10.1177/09645284221125421 36325677

36. Wen Q . Efficacy and Safety of Warming Needle Moxibustion in the Treatment of Chemotherapy-induced Peripheral Neuropathy . Med Innovation of China . 2024 ; 21 ( 22 ): 28 – 32 . doi: 10.3969/j.issn.1674-4985.2024.22.007

37. Lu C , Shen Q , Deng D , et al. Effects of Electroacupuncture and Mecobalamin for Utidelon-Induced Peripheral Neuropathy in Breast Cancer Patients: a Randomized Controlled Clinical Trial . J Pain Res . 2025 ; 18 : 3593 – 3608 . doi: 10.2147/JPR.S526405 PMC12276754 40687335

38. Iravani S , Kazemi Motlagh AH , Emami Razavi SZ , et al. Effectiveness of Acupuncture Treatment on Chemotherapy-Induced Peripheral Neuropathy: a Pilot, Randomized, Assessor-Blinded, Controlled Trial . Pain Res Manag . 2020 ; 2020 : 2504674 . doi: 10.1155/2020/2504674 PMC7341378 32676134

39. Hou H , Chen H . Effect of Acupuncture on Chemotherapy-related Peripheral Neurotoxicity and Digestive Tract Adverse Effects of Malignant Tumor Patients: a Clinical Study on 20 Cases . J Traditional Chin Med . 2011 ; 52 ( 23 ): 2031 – 2033 . doi: 10.13288/j.11-2166/r.2011.23.015

40. Fu W . Clinical Study on Transcutaneous Electrical Acupoint Stimulation for the Treatment of Chemotherapy-Induced Peripheral Neuropathy in Cancer Patients . J Phys Med Theory Pract . 2024 ; 37 ( 23 ): 4111 – 4113 . doi: 10.19381/j.issn.1001-7585.2024.23.054

41. Wei H , Deng S , Lu Y , Long J . Clinical Study on the Treatment of Malignant Tumor Patients by Electric Acupuncture Moxibustion Combined with Paclitaxel . Yunnan J Trad Chin Med Mater Med . 2014 ; 35 ( 09 ): 19 – 21 . doi: 10.16254/j.cnki.53-1120/r.2014.09.009

42. Li C , Li X , Lu Y , Bao C , Liu Q , Liu Y . Clinical Study on Electroacupuncture Combined with Pricking and Bloodletting for Chemotherapy-Induced Peripheral Neuropathy in Breast Cancer . Modern Chin Med . 2023 ; 43 ( 4 ): 65 – 69 . doi: 10.13424/j.cnki.mtcm.2023.04.013

43. Sun J , Hu J , Gu X , Zhou J . Clinical Study on Chinese Herbal Fumigation and Wash Combined with Acupuncture for Peripheral Nerve Injury Induced by Albumin-Bound Paclitaxe . Modern Med Health Res Electronic J . 2023 ; 7 ( 12 ): 25 – 27 . doi: 10.3969/j.issn.2096-3718.2023.12.009

44. Li Y , Zhai Y , Hao X , Zhang Y , Ma X . Clinical Study on Abdominal Acupuncture Combined with Moxibustion for Peripheral Neurotoxicity Induced by Taxane Chemotherapy in Breast Cancer Patients . Mod J Integr Traditional Chinese and Western Med . 2025 ; 34 ( 1 ): 74 – 77 . doi: 10.3969/j.issn.1008-8849.2025.01.012

45. Wang T , Xiong Z , Tian F , Song N . Clinical study of warming needle therapy in preventing and treating peripheral neuropathy after albumin-bound paclitaxel plus cisplatin chemotherapy . J Acupuncture Tuina Sci . 2025 ; 23 ( 1 ): 63 – 72 . doi: 10.1007/s11726-025-1480-6

46. Xiang Y , Xu N , Sui M . Clinical observation on electroacupuncture in the treatment of chemotherapyinduced pathologic neuropathyic pain . Chin J Integr Med Cardio-Cerebrovasc Dis . 2025 ; 16 ( 9 ). doi: 10.12102/j.issn.1672-1349.2018.09.004

47. Cui D , Wang L , Fu C . Clinical Observation of Warm Needle Acupuncture in the Treatment of Oxaliplatin-Induced Neurotoxicity . Western J Tradit Chin Med . 2011 ; 24 ( 2 ): 45 – 46 . doi: 10.3969/j.issn.1004-6852.2011.02.022

48. Zhang C , Li S , Mi S . Clinical Observation of Acupuncture Treatment for Peripheral Nerve Damage Following Capecitabine Chemotherapy . Clin Nurs Res . 2022 ; 31 ( 4 ): 46 – 48,186 .

49. Liu B , Zhang L , Xu K , Luo H , Li Y . Clinical Observation of Acupuncture in the Treatment of Peripheral Neurotoxicity Induced by Chemotherapeutic Drugs . Hebei J Trad Chin Med . 2009 ; 31 ( 7 ): 1040 – 1041 . doi: 10.3969/j.issn.1002-2619.2009.07.057

50. Chen W . Clinical Observation of Acupuncture for Peripheral Neuropathy Induced by Oxaliplatin Chemotherapy in Patients with Colorectal Cancer . World Latest Medicine Inform . 2018 ; 18 ( 05 ): 144 – 145 . doi: 10.19613/j.cnki.1671-3141.2018.05.115

51. Wu Y , Zeng Y , Leng Y . Clinical Observation of Acupuncture Prevention and Cure Effect on the Nitrobenzene lnduced Neurotoxicity in the Circuit of chemotherapy Period . Clin J Tradit Chin Med . 2018 ; 30 ( 04 ): 725 – 728 . doi: 10.16448/j.cjtcm.2018.0219

52. Chen D , Qi C , Lu S . Analysis of Clinical Effect of Electroacupuncture Therapy Combined with Conventional Therapy in Treatment of Peripheral Neuropathy Induced by Oxaliplatin Chemotherapy . Chinese Community Doctors . 2025 ; 41 ( 12 ): 52 – 54 . doi: 10.3969/j.issn.1007-614x.2025.12.018

53. Huang CC , Ho TJ , Ho HY , et al. Acupuncture relieved chemotherapy-induced peripheral neuropathy in patients with breast cancer: a pilot randomized sham-controlled trial . J Clin Med . 2021 ; 10 ( 16 ). doi: 10.3390/jcm10163694 PMC8397157 34441990

54. Rostock M , Jaroslawski K , Guethlin C , Ludtke R , Schröder S , Bartsch HH . Chemotherapy-induced peripheral neuropathy in cancer patients: a four-arm randomized trial on the effectiveness of electroacupuncture . Evid-Based Complement Altern Med . 2013 ; 2013 : 1 – 9 . doi: 10.1155/2013/349653 PMC3771477 24066010

55. Avis N . Acupuncture in Reducing Chemotherapy-Induced Peripheral Neuropathy in Participants With Stage I-III Breast Cancer . Available from : https://clinicaltrials.gov/show/NCT03505671 . 2022 .

56. Ter LI , Yamada AMTD , Basilio ANMB , López RVM , Hsing WT . Acupuncture for peripheral neuropathy induced by paclitaxel in early-stage breast cancer: a randomized, parallel, controlled, blinded study in a brazilian oncologic center (PACLILIN study) . Support Care Cancer . 2026 ; 34 ( 2 ): 102 . doi: 10.1007/s00520-026-10315-8 41543601

57. D’Alessandro EG , Nebuloni Nagy DR , De Brito CMM , Almeida EPM , Battistella LR , Cecatto RB . Acupuncture for chemotherapy-induced peripheral neuropathy: a randomised controlled pilot study . BMJ Support Palliat Care . 2022 ; 12 ( 1 ): 64 – 72 . doi: 10.1136/bmjspcare-2018-001542 31256014

58. Han X , Wang L , Shi H , et al. Acupuncture combined with methylcobalamin for the treatment of chemotherapy-induced peripheral neuropathy in patients with multiple myeloma . BMC Cancer . 2017 ; 17 : 40 . doi: 10.1186/s12885-016-3037-z PMC5223334 28068938

59. Molassiotis A , Suen LKP , Cheng HL , et al. A randomized assessor-blinded wait-list-controlled trial to assess the effectiveness of acupuncture in the management of chemotherapy-induced peripheral neuropathy . Integr Cancer Ther . 2019 ; 18 : 1534735419836501 . doi: 10.1177/1534735419836501 PMC6434440 30905173

60. McGuinness LA , Higgins JPT . Risk-of-bias VISualization (robvis): an R package and shiny web app for visualizing risk-of-bias assessments . Res Synth Methods . 2020 . doi: 10.1002/jrsm.1411 32336025

61. Koch JL , Lew CCH , Kork F , et al. The efficacy of fiber-supplemented enteral nutrition in critically ill patients: a systematic review and meta-analysis of randomized controlled trials with trial sequential analysis . Crit Care . 2024 ; 28 : 359 . doi: 10.1186/s13054-024-05128-2 PMC11545523 39511681

62. Yeh ML , Liao RW , Yeh PH , Lin CJ , Wang YJ . Acupuncture-related interventions improve chemotherapy-induced peripheral neuropathy: a systematic review and network meta-analysis . BMC Complementary Med Ther . 2024 ; 24 ( 1 ): 310 . doi: 10.1186/s12906-024-04603-1 PMC11334450 39160496

63. Lu W , Matulonis UA , Dunn JE , et al. The feasibility and effects of acupuncture on quality of life scores during chemotherapy in ovarian cancer: results from a pilot, randomized sham-controlled trial . Med Acupunct . 2012 ; 24 ( 4 ): 233 – 240 . doi: 10.1089/acu.2012.0904 PMC3579196 24761165

64. Zhao Y , Cao W , Zhang H , et al. The efficacy of acupuncture in the treatment of chemotherapy-induced peripheral neuropathy in cancer patients: a systematic review and meta-analysis . J Cancer Surviv . 2025 . doi: 10.1007/s11764-025-01869-3 40830723

65. Klein MM , Treister R , Raij T , et al. Transcranial magnetic stimulation of the brain: guidelines for pain treatment research . Pain . 2015 ; 156 ( 9 ): 1601 – 1614 . doi: 10.1097/j.pain.0000000000000210 PMC4545735 25919472

66. Shen Q , Deng D , Li G , et al. Electroacupuncture frequency for chemotherapy-induced neuropathy in breast cancer: a randomized controlled trial . Oncologist . 2025 ; 30 ( 9 ): oyaf262 . doi: 10.1093/oncolo/oyaf262 PMC12449621 40828899

67. Li Y , Xu R , Chen M , et al. Electroacupuncture alleviates paclitaxel-induced peripheral neuropathy by reducing CCL2-mediated macrophage infiltration in sensory ganglia and sciatic nerve . Chin Med . 2025 ; 20 ( 1 ): 9 . doi: 10.1186/s13020-024-01023-8 PMC11727193 39806462

68. Li J , Hu X , Liang F , et al. Therapeutic effects of moxibustion simultaneously targeting Nrf2 and NF-κB in diabetic peripheral neuropathy . Appl Biochem Biotechnol . 2019 ; 189 ( 4 ): 1167 – 1182 . doi: 10.1007/s12010-019-03052-8 PMC6882806 31209719

69. Zhang C , Li D , Li J , et al. Clinical Study of Mecobalamin Combined with Qishao Rujin Decoction and Bloodletting Therapy in the Treatment of Oxaliplatin-Induced Neurotoxicity . J Oncol Chin Med . 2025 ; 7 ( 6 ): 70 – 76 . doi: 10.19811/j.cnki.ISSN2096-6628.2025.11.009

70. Huang H , Zhao M , Liu X , et al. Glutathione combined with mecobalamin in the treatment of chemotherapy-induced peripheral neuropathy in multiple myeloma: a retrospective clinical study . Ann Palliat Med . 2021 ; 10 ( 12 ): 12335 – 12346 . doi: 10.21037/apm-21-3313 35016427

71. Cascinu S , Catalano V , Cordella L , et al. Neuroprotective effect of reduced glutathione on oxaliplatin-based chemotherapy in advanced colorectal cancer: a randomized, double-blind, placebo-controlled trial . J Clin Oncol . 2002 ; 20 ( 16 ): 3478 – 3483 . doi: 10.1200/JCO.2002.07.061 12177109

72. Yang K , Zhang L , Li B , et al. Analysis of acupoint selection and prescription rules of acupuncture for treatment of stable angina pectoris based on a traditional Chinese medicine inheritance calculation platform: a systematic review . Medicine . 2022 ; 101 ( 49 ): e31466 . doi: 10.1097/MD.0000000000031466 PMC9750679 36626434

73. Li L , Huang Y , An C , et al. Acupuncture in the treatment of chemotherapy-induced peripheral neuropathy: a meta-analysis and data mining . Front Neurol . 2024 : 15 . doi: 10.3389/fneur.2024.1442841 . PMC11565602 39555482

74. Huang HQ , Brady MF , Cella D , Fleming G . Validation and reduction of FACT/GOG-ntx subscale for platinum/paclitaxel-induced neurologic symptoms: a gynecologic oncology group study . Int J Gynecol Cancer . 2007 ; 17 ( 2 ): 387 – 393 . doi: 10.1111/j.1525-1438.2007.00794.x 17362317

75. Staff NP , Grisold A , Grisold W , Windebank AJ . Chemotherapy-induced peripheral neuropathy: a current review . Ann Neurol . 2017 ; 81 ( 6 ): 772 – 781 . doi: 10.1002/ana.24951 PMC5656281 28486769

76. Kishimoto S , Oshima N , Rinker M , Krishna MC , Takebe N . Identification of high-risk drugs related to chemotherapy-induced peripheral neuropathy in cancer therapy evaluation program-sponsored Phase I trials . Eur J Cancer . 2019 ; 115 : 111 – 119 . doi: 10.1016/j.ejca.2019.04.023 PMC8170839 31132741