This study is designed as a single-center, rater-masked, parallel-arm, randomized controlled trial. We will recruit 100 breast cancer survivors diagnosed with insomnia from the Hospital of Chengdu University of Traditional Chinese Medicine. Participants will be randomly allocated in a 1:1 ratio to either the LSM intervention group or the waitlist control (WC) group. The study will span 9 weeks, comprising a 1-week baseline assessment phase, a 4-week intervention phase, and a 4-week post-treatment follow-up phase. Throughout the 4-week treatment period, both groups will receive standard care, with the LSM group additionally undergoing LSM therapy twice per week. This trial has been rigorously designed according to the Consolidated Standards of Reporting Trials (CONSORT) guidelines and will be reported following the 2013 Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) Checklist (see Supplementary material S1). The study flow is depicted in Figure 1.

All enrolled participants will receive standardized care, comprising sleep hygiene education and, where applicable, hypnotic medication interventions. The sleep hygiene education will emphasize adherence to a consistent sleep schedule, optimizing the sleep environment, and avoiding the consumption of alcohol, nicotine, and caffeine for at least several hours before bedtime. Additionally, participants will be encouraged to engage in 180–300 min of low-to-moderate intensity physical activity per week, while avoiding exercise within 3 h of their intended sleep time. Emotional self-regulation strategies will be incorporated to further support sleep hygiene practices. Hypnotic medications, which are optional and not required for all patients, will be prescribed based on individual therapeutic needs and clinical assessments. These medications may include psychotropic drugs such as benzodiazepines, non-benzodiazepine agents, and narcotics. The type, dosage, and frequency of all prescribed medications will be meticulously documented to ensure comprehensive data collection.

In addition to the aforementioned standard care, participants in the LSM group will receive supplemental LSM therapy, administered twice weekly for a total of eight sessions. The LSM procedure will be conducted following standardized protocols established in previous clinical studies (19). The treatment area will extend along the spine from GV14 (Dazhui), located at the lower border of the spinous process of the seventh cervical vertebra, to DU2 (Yaoshu), located at the lower border of the spinous process of the second sacral vertebra, with a 3 cm extension on both sides of the spine. The acupuncture points within this 3 cm range include the 13 governor vessel points as well as 24 bladder meridian points. Detailed information on the specific acupuncture points is provided in Supplementary material S2. The LSM procedure will commence with the patient lying in a prone position, with the back fully exposed, followed by thorough disinfection of the targeted area. A 12 cm-wide, 70 cm-long piece of mulberry bark paper will then be placed along the spine’s midline, after which a trapezoidal ginger paste (base width: 6 cm, top width: 5 cm, height: 3 cm) will be evenly mounded from GV14 to DU2. A cylindrical moxa cone (5 cm in diameter and 3 cm in height) will subsequently be positioned on the ginger paste, aligned along its full length. The moxa cone will be ignited and allowed to burn completely before being replaced once more, ensuring sufficient heat stimulation across the treatment area. Each LSM session is estimated to last approximately 1.5 h.

Figure 2 provides a detailed timeline outlining the schedule for participant enrollment, intervention procedures, outcome assessments, and systematic data collection throughout the study period.

An independent biostatistician will generate the randomization sequence using a stratified block randomization design. Block sizes of 2 or 4 will be used, with stratification based on whether patients are currently receiving endocrine therapy (yes or no). Participants will be assigned to either the LSM or WC groups in a 1:1 ratio. The randomization codes will be sealed in sequentially numbered, opaque envelopes, which will be securely maintained by independent administrative staff. Participants will receive their envelopes in the order of enrollment to determine their group allocation.

Given the distinct nature of LSM therapy, creating a placebo moxibustion is challenging. Consequently, this study will be conducted as an open-label trial, with both participants and acupuncturists aware of their treatment assignments. To mitigate potential bias, blinding will be maintained for outcome assessors, data managers, and statisticians. During assessments, outcome assessors will use CRFs containing participant names, while both participants and assessors will be instructed to avoid discussing treatment allocation. Once assessments are complete, independent administrative staff overseeing randomization will replace participant names with coded labels, and group assignments will be replaced with anonymized labels. The anonymized dataset will then be provided to data managers and statisticians for analysis. Since statisticians will only have access to the anonymized dataset, they will conduct the analysis without knowledge of treatment group assignments, ensuring unbiased interpretation of results. After data analysis is finalized, the identities associated with the codes will be disclosed to the research team for accurate interpretation of the results.

To obtain a preliminary estimate of LSM’s efficacy, a pilot study was conducted with 12 participants assigned to each of the two groups, based on the rules of thumb (29). After the 4-week treatment period, ISI scores decreased by 6.82 ± 4.06 points in the LSM group and 3.24 ± 2.38 points in the WC group from baseline. The observed effect size (Cohen’s d) was 1.08, which, according to Cohen’s criteria (d = 0.2 as small, d = 0.5 as medium, and d = 0.8 as large), indicates a large effect. Based on these results, the sample size was calculated using the “Two-sample t-tests allowing unequal variance” module in PASS software, version 15.0, with a two-tailed alpha of 0.05 and a power of 0.90, while accounting for a 10% expected dropout rate. Previous studies have shown that endocrine therapy is an important risk factor for insomnia in breast cancer patients (9). Therefore, to account for potential heterogeneity in treatment response, the sample size was stratified by endocrine therapy status (yes vs. no). To ensure sufficient statistical power within both strata, the sample size was adjusted to 46 participants per group. Ultimately, considering the importance of using whole numbers for clear communication of results, and to further enhance the statistical power of the study, the sample size per group was rounded up to 50 participants.

The analysis will be conducted based on the intent-to-treat (ITT) population, defined as participants who complete baseline assessments and have at least one follow-up measurement. The Shapiro–Wilk test will be employed to assess the normality of the data distribution. For continuously measured repeated data, such as ISI scores, PSQI scores, PFS scores, FACT-B scores, and hypnotic medication dosages, a mixed-effects model adjusted for baseline values will be used to compare longitudinal changes from baseline across follow-up time points between groups. The model will include fixed effects for endocrine therapy status (yes or no), time, group, and the interaction between time and group, while treating individual participants as random effects. Given that mixed-effects models can accommodate participants with at least one outcome measurement, no imputation will be performed for missing values in the primary analysis (30).

For non-repeated continuous variables, such as serum biochemical markers and gut microbiota abundance, between-group comparisons will be performed using either the t-test or Mann–Whitney U test, depending on data distribution. Categorical variables, including treatment response rates and incidence of adverse events, will be analyzed using chi-square or Fisher’s exact tests. Correlation analyses between ISI scores, differential serum biochemical markers, differential metabolites, and key gut microbiota will be performed using Pearson or Spearman correlation coefficients. Missing values for data analyzed using non-mixed effect models will be managed using multiple imputation methods.

For the primary efficacy outcome (the change in the ISI score from baseline to the end of the 4-week treatment period), subgroup analyses will be conducted based on hypnotic medication use (yes, no) and insomnia severity (mild, moderate, severe) to further explore potential variations in the treatment effects of LSM across different subgroups. Additionally, sensitivity analyses will be performed for the primary efficacy outcome. Specifically, if missing values or non-normally distributed data are present, multiple imputation methods will be used to assess the impact of missing data, and generalized estimating equations will be employed as a non-parametric alternative for non-normally distributed data. Pre-specified per-protocol analyses will also be conducted. The per-protocol set will include all participants who completed the intervention and follow-up according to the study protocol. The results of the primary analysis will be compared with those from these sensitivity analyses to assess robustness.

All statistical analyses will be carried out using SPSS (version 26.0) and R (version 4.3.1). Two-tailed significance testing will be applied, with a p-value threshold of ≤ 0.05 indicating statistical significance.

Before participant enrollment, all study personnel will undergo comprehensive training sessions aimed at ensuring standardized implementation and enhancing the reliability of study outcomes. These sessions will cover key aspects such as study design, eligibility criteria, intervention procedures, outcome measurement, and accurate data documentation in CRFs.

The LSM procedures will be performed by acupuncturists with a minimum of 3 years of moxibustion experience. To minimize inter-practitioner variability, all practitioners will undergo a standardized training program focusing on the detailed application of the LSM procedure, including ginger paste preparation, moxa cone positioning, mulberry bark paper placement, and the treatment area covered along the spine. Practitioners will be required to demonstrate competency in following the treatment protocol through supervised practice sessions before performing the treatment on study participants. To further ensure treatment fidelity, a senior clinician will monitor and audit the LSM procedures periodically throughout the study. Any deviations from the standardized protocol will be addressed promptly to maintain consistency across practitioners and sessions.

Data collection will be conducted using paper CRFs to systematically document study-related information. All paper records will be securely stored in locked cabinets under the supervision of authorized personnel. Electronic data entry will be managed through a double-entry system, conducted by two independent and experienced staff members. Data will be stored in an encrypted research database, accessible only to authorized research personnel. Strict measures will be implemented to ensure data confidentiality and prevent unauthorized access or breaches. The Chengdu University of Traditional Chinese Medicine Research Ethics Committee will perform regular audits to verify data accuracy, monitor trial conduct, assess study progress, and ensure compliance with ethical standards and protocol requirements.