Chinese Herbal Medicine to Alleviate Vasomotor Symptoms in Breast Cancer Survivors: A Study Protocol for a Randomised, Double Blind, Placebo Controlled, Multi Centre, Cross Over Trial

Introduction Vasomotor Symptoms (VMS) have been documented as one of the most common and distressing side-effects after breast cancer treatment. Currently there is no standard conventional medical treatment. VMS, which include hot flushes and night sweats, can be the result of: a) treatmentschemotherapy, ovarian ablation, ovarian suppression and/or endocrine therapy with tamoxifen or aromatase inhibitors; b) menopause occurring naturally with a breast cancer diagnosis; and/or c) menopause symptoms occurring after cessation of hormone replacement therapy upon breast cancer diagnosis. There is promising evidence of the use of a Chinese herbal medicine formula, Shu-Gan-Liang-Xue Decoction (SGLXD), developed by one author, PPL, for VMS in breast cancer survivors prescribed tamoxifen in China. The aim of this study is to evaluate whether SGLXD is an effective and safe treatment option for reducing VMS and improving quality of life in breast cancer survivors in Australia. Methods/Design We will conduct a multi-centre, randomised, double-blind, placebo controlled, cross-over trial. Participants are women, diagnosed with Stage I-III breast cancer, who have completed the acute stage of breast cancer treatment, who satisfy the inclusion/exclusion criteria and desire an intervention for VMS. Participants will be randomly allocated to treatments to create two groups for 12-weeks. After a 4-week wash-out period there will be a second 12-week period for which groups swap treatments. The primary outcome is reduction in frequency of VMS and secondary outcomes are reduction in mean severity of VMS (measured with subjective Hot Flush Daily Diary) and improvement in mean quality-of-life-score (Menopause Specific Quality of Life and Hot Flush Related Daily Interference Scale). Safety will be measured with pathology tests including liver and renal function tests, hormone assays, tamoxifen metabolites, Estrogen Receptor 1 Gene (ESR1), Circulating Tumour Cells (CTC) and adverse events log. At the end of the study focus groups will be conducted with study participants to discuss a range of topics including reasons for joining the study, experience of taking the intervention, effects of the treatment. Transcripts of the groups will be analysed using grounded theory techniques. Ethics and Dissemination This protocol has been approved by the South Western Sydney Local Health District Human Research Ethical Committee (Project number HE/17/013). All participants will receive Participant Information Sheets and verbal information about the trial and will give informed consent before enrolment. The results will be published in peer-reviewed journals and disseminated through conference presentations.


Introduction
The number of women diagnosed with breast cancer is growing worldwide [1,2]. Due to improved detection methods and treatment the number of survivors is also increasing [3]. However, after treatment for the disease, up to 78% of survivors report bothersome vasomotor symptoms [4][5][6][7]. Vasomotor symptoms (VMS) include hot flushes and night sweats. A hot flush is a rushing sensation of heat spreading over the body and often accompanied by perspiration and a change in body temperature and heart rate [8]. VMS can also be accompanied by other physical sensations including palpitations, sweating, skin reddening, chills, tingling sensations in the extremities, dizziness and can disturb sleep leading to fatigue and mood changes [8]. Emotional symptoms can include panic, anxiety, frustration, irritation, depression and annoyance [8]. VMS can be a source of embarrassment and impact on quality of life for the individual can be distressing [9,10]. The higher frequency in breast cancer survivors can be due to age at diagnosis, abrupt discontinuation of Hormone Replacement Therapy (HRT) after diagnosis, chemotherapy, ovarian ablation or estrogen deficiency symptoms by therapy (tamoxifen or aromatase inhibitors) [11].
Studies have shown VMS to be the most frequently reported [6,10] and residual symptom in the post treatment phase [10].
VMS can have a long-lasting negative effect on quality of life, and for some is an indicator for early discontinuation and compliance to life-saving endocrine therapies, putting them at increased risk of recurrence [4,6,[12][13][14][15][16][17]. Many management options for VMS in healthy subjects have limited safety and effectiveness or adverse effects in this group of women. Hormone Replacement Therapy (HRT), the main management option used in healthy menopausal women, is contraindicated due to increased risk of recurrence of the disease [18,19]. Other treatments commonly prescribed include Selective Serotonin Re-Uptake Inhibitors (SSRIs), Serotonin Norepiephrine Re-Uptake Inhibitors (SNRIs), tiboline, gabapentin, clonidine and stellate ganglion block [20][21][22][23].
These interventions can induce negative side-effects or have only moderate effect [23][24][25]. This leaves clinicians with uncertainty about how to treat VMS safely and effectively, after breast cancer [5,[26][27][28]. Given the absence of a safe and effective treatment intervention, many cancer patients turn to some form of complementary medicine [29][30][31][32][33][34].The aim of the study is to a) evaluate the effectiveness of the Chinese herbal medicine (SGLXD) on reduction of VMS and improvement in quality of life in Australian women treated for breast cancer, b) to monitor the safety of SGLXD in this group of women, and c) to explore the feasibility of implementing and evaluating a Chinese herbal medicine intervention within an Australian primary care system (with regard to recruitment patterns, compliance, withdrawal rate).

Rationale for Use of Chinese Herbal Medicine
Chinese herbal medicine is a popular and prevalent therapy worldwide, in individuals diagnosed with cancer to alleviate treatment related side-effects and to improve quality of life [35][36][37][38]. However, clinical evidence on the efficacy and safety of Chinese herbal medicine for VMS in breast cancer survivors outside of China is limited [3,30].
Standardised formulae are when every participant is prescribed the same herbal formula without considering their presenting signs and symptoms [70,71]. The standardized formula makes the results of the trial more generalisable [70]. Data from laboratory studies and clinical trials conducted in China have shown the effectiveness of SGLXD to ameliorate VMS associated with treatment of breast cancer in women and improvement in sleep quality [72][73][74].
SGLXD also decreased serum estrogen and alleviated the effects of tamoxifen on endometrial hypertrophy in vivo [75]. A study found that SGLXD inhibited breast tumour growth by 26.3% in tumour-bearing mice [76]. Another study, by  found the herbal formula and each of the component herbs did not manifest any estrogenic activity and the formula exhibited an antiproliferative effect on MCF-7cell viability at high concentrations [39].
Appropriate safety of SGLXD has been shown in herb-drug safety/interaction studies [39,59,73,[76][77][78][79][80][81][82][83][84][85][86][87][88][89]. This preliminary research has been limited to China, with Chinese breast cancer survivors and in a Chinese hospital setting. Building on these pre-clinical and clinical results we propose a trial to evaluate the clinical effectiveness and safety of SGLXD in Australian women in an Australian primary care setting. As Chinese herbal medicine is culturally ingrained in China [38,90], this study may ascertain feasibility and tolerability in Australian breast cancer survivors and provide important outcomes on the use of CHM in primary care. This novel intervention for VMS may therefore translate to increased compliance to endocrine therapy and improved quality of life for a growing cohort of Australian survivors.

Methods Design
The study design is a multi-centre, randomised, double-blind, placebo-controlled, cross-over clinical trial comparing SGLXD and placebo granules in participants with VMS after breast cancer treatment. The study schema is represented in ( Figure   1). The protocol design is based on the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) Checklist (File 1). The study complies with the principles of the Declaration of Helsinki and Good Clinical Practice guidelines [91,92].

Ethics and Trial Registration
This protocol has been approved by the South Western Sydney Local Health District Human Research Ethical Committee (Project number HE/17/013). All participants will receive Participant Information Sheets and verbal information about the trial and will give informed consent before enrolment. Any amendments to the study protocol will be submitted to the HREC of SWSLHD for approval. The trial has been registered with the Australia and New

Setting
Eighty-four participants will be recruited from three hospitals in South Western Sydney, Australia. All sites are under the governance of the South Western Sydney Local Health District which provided publicly funded health services to over 820,000 people. The hospitals service an area regarded as one of the most culturally diverse in Australia with more than 38% speaking a language other than English at home [93].

Eligibility Criteria and Enrolment
Inclusion/exclusion criteria have been designed to rule out systemic co-morbidities and pharmaceuticals/over the counter remedies that may cause or interfere with vasomotor symptoms.
Women will be eligible if 1) they have been diagnosed with stage I-III breast cancer, 2) completed surgery, chemotherapy and/or radiation, 3) if prescribed an endocrine therapy i.e. tamoxifen or aromatase inhibitor, stable dosage for >2 months prior to the study, 4) unlikely to be changing/discontinuing endocrine therapy during study, 5) with minimum of at least 3 hot flushes/24 hours during the eligibility period prior to randomisation (Table 1).   (Table   1).
Potential participants will be identified via referring oncologists at the participating sites, based on the preliminary inclusion and exclusion criteria, and provided an information sheet outlining the study, their commitments and possible risks involved. They are then directed to the principal researcher who will explain the study and answer any questions. If they agree to join the study and after signed informed consent forms are obtained, they will be further

Am J Biomed Sci & Res
Copy@ Dianna Porter screened to satisfy all inclusion/exclusion criteria, including pathology tests. The study consists of a 2-week eligibility/baseline period, a 12-week treatment period, a 4-week wash-out period and cross-over to the other arm for a 12-week treatment period, followed by a 4-week follow-up period. The duration of the run-in, treatment and follow-up was decided based on a previous study [94]. An outline of procedures is illustrated in (Figure 1).

Sub Study
As an added safety measure, we will be conducting additional pathology testing on a small cohort of women prescribed endocrine disrupting drugs. Additional consent is required from women who volunteer to participate. Sub-study (a) is for women prescribed tamoxifen and sub-study (b) is for women prescribed aromatase inhibitors. The sub-study will conduct tests at the same three timepoints as the standard pathology tests.
A. To ascertain if the herbal medicine has an impact on plasma concentration of tamoxifen metabolites.

B. Acquired Estrogen Receptor 1 gene (ESR1) mutations are
found predominately in women diagnosed with estrogen receptor positive breast cancer and exposed to aromatase inhibitors [95].
Mutations of the ESR1 gene, the gene that encodes the estrogen receptor, have been recognised as a potential mechanism of endocrine therapy resistance and as an experimental predictive biomarker to guide therapeutic decisions [96]. Testing may provide valuable information on whether the herbal medicine impacts ESR1 mutations. Circulating Tumour Cells (CTCs) characteristics may serve as future indicators of therapeutic interventions [97].

Randomisation and Blinding
If all eligibility criteria are confirmed participants will be randomised. Randomisation will be conducted external to the primary research team by a research officer at the NICM Health Research Institute. A computer generated, permuted block randomisation sequence will randomly associate a unique identifier to either active or placebo treatment (1:1 ratio) and for labelling the codes on the pre-prepared participant medication containers.
Permuted block randomisation (14 per block) will ensure that the treatment groups are balanced at the end of every block. Each participant will be assigned a Personal Identification Number (PIN) at baseline and once randomised will also be provided a randomisation number.
Both sets of codes will be documented and stored securely at the NICM and this information will also be secured on the University Research Data Storage system. This can be accessed by the research officer at NICM, when required, for un-blinding. Throughout the study, the participants, doctors, researchers and the data analyst will be blind to the current treatment of participants until after the study is completed.

Trial Interventions
All trial participants will receive Chinese herbal medicine granule (SGLXD) or placebo in powder form (indistinguishable to each group).The herbal medicine and placebo are manufactured as a concentrated herbal extract by Pura Pharm, Hong Kong in accordance with Australian TGA (Therapeutic Goods Administration) Good Manufacturing Practice (GMP) requirements for herbal medicinal products and according to the Product Specification File [98]. The medication will be mixed, cooked, filtered and pressure-spraydried forming granules. For the control group, placebo granules are prepared from therapeutically inert ingredients -calcium hydrogen phosphate (78.2%), soy fibre (19.6%), natural identical liquorice flavour (1%), bitter flavour (0.3%), colours (0.9%).
Dosage and administration instructions for both groups will be identical. Each intervention, in a water-soluble form, will be administered in a standard 10gram dose, orally twice per day for each 12-week intervention period. Each dose is packaged individually in a sealed foil sachet. Participants dissolve the granules in hot water to take as a tea. Interventional Medical Product (IMP) is packaged as 4-weeks supply (i.e. 56 sachets) to be dispensed at each study visit. The package contains the randomisation number and the batch number and no information as to the contents. The placebo arm is packaged identical to the control arm. Participants will be asked to return all IMP sachets at the next scheduled visit, including those that are empty. Unused IMP will be counted and logged to assess compliance.

Primary outcome
The primary outcome is the changes or differences in VMS frequency, as recorded subjectively in the Hot Flush Daily Diary. Frequency will be averaged over a week (week -1, 4, 8, 12, 16, 20, 24, 28, 32) and assessed for change across time and between active and placebo treatment. For each treatment period the change in VMS frequency will be calculated as the final VMS frequency averaged over a week minus the baseline frequency for that treatment.
averaged over a week.

Secondary Outcome
The secondary outcomes of the study will assess the changes or differences across time between the intervention group and the placebo group.

Safety Endpoints
A. Number and kind of adverse events will be monitored at each review visit by a clinician according to Acute Toxicity CT CAE Circulating Tumour Cells (CTC) (<20) will be monitored in a small cohort of women at the same 3 time-points.

Qualitative Data
Focus groups will be conducted at the end of the study for those who initially volunteered and consented to participate in a group discussion These groups of up to 15 participants will run for approximately 30 minutes and will be digitally recorded to obtain qualitative data on participants' subjective experiences of joining the study, including their response to the taste, burden, compliance and effect of taking the intervention. This may provide valuable information on the feasibility of integrating CHM with conventional care.

Assessment Schedules
After a 2-week eligibility period, the study period will run for 2 x 12-week periods with a 4-week wash-out period in between (Table 2). Baseline demographic and clinical data will be collected including age, age at diagnosis, type and stage of cancer, ER positive/negative, node positive/negative, treatment (surgery, chemotherapy, radiation, endocrine therapy), medical history, menopausal status, height, weight, country of birth, current medications, serum hormone levels, liver and renal function test results, TCM diagnosis. Potential participants will fill in the HFDD for 2-weeks prior to randomisation to establish eligibility criteria and baseline data. After randomisation the HFDD will be filled out  (Table 2).
Unused study material collected/ compliance quality of life and can be used for monitoring treatment effects [104,107] In this study they will be used to make baseline assessments as well as effectiveness measures. Both can be completed in approximately 5 minutes.

Safety Monitoring and Emergency Un Blinding
An independent safety monitor has been appointed. The safety monitor has relevant expertise (Bachelor of Medicine, PhD in Pharmacology) and will assess any Serious Adverse Events (SAEs) or Suspected Unexpected Serious Adverse Reaction (SUSARs). Liver and renal function tests and hormone assays will be conducted at baseline, to establish eligibility, and at the end of each intervention for all participants (3 collection times). Hormone assays will include FSH, LH, estrogen, progesterone. Tamoxifen metabolites will be tested, in a small cohort of women (<10) prescribed tamoxifen, at baseline and end of each treatment arm. Circulating Tumour Cells (CTC) and Estrogen Receptor 1 (ESR1) testing will also be conducted in a small cohort of women (<20) prescribed tamoxifen or aromatase inhibitors as a further safety measure.
If any adverse events are reported appropriate treatment will be provided and documented in a case report form, including severity, duration and probable causality. All SAEs and SUSARs must be recorded in detail in the Clinical Report Forms (CRF) and reported to the Chief Investigator and appointed Safety Monitor within 24 hours. Emergency un-blinding will be escalated by the appointed Medical Monitor for a Serious Adverse Event (SAE) or a Suspected Unexpected Serious Adverse Reaction (SUSAR). The TGA will be notified within 7-15 days. Follow-up for SAEs/SUSARs will be provided until resolution of the event.

Sample Size
The sample size calculation was performed using G*Power using data from previous trials [94,108]. Allowing for a 25% drop in frequency of hot flushes due to placebo effect we define minimum clinically important improvement to be twice the size of the placebo effect (50% decrease in frequency) [94,109]. We expect the mean frequency to be 6.0 per day with a standard deviation of 4 at baseline [94]. We estimate the required sample size is 58 to produce 80% power to detect the difference between 25% decrease (placebo) [110] and 50% decrease (treatment), using a two-sided matched pairs t-test with 5% significance level. Assuming an attrition rate of 30% (withdrawal and non-compliance), based on earlier research, a target sample of 84 participants is required [111].

Data Collection and Management
Case Report Forms (CRF) from this study will be in paper and electronic format for all participants. Data will be collected at various time points ( results, Treatment Satisfaction Form. All paper questionnaires will be self-administered and collected by the principal investigator at each visit. All participants will be given a unique Participant

Identification Number (PIN) to maintain confidentiality. All
Clinical Report Forms (CRFs), pathology tests and reports will be identified with the participant's PIN. All trial data will be entered on an electronic data capture system (Redcap-Research Electronic Data Capture) an online secure web-based application initiated by Vanderbilt University [112]. All data collected will be retained for 15 years in the Western Sydney University Research Data Repository where it will be securely archived.

Data Checking
At completion of data collection and all variables and all logical pairs of variables will be subject to descriptive analyses using graphs, frequency counts and summary statistics. This will allow a) identification of unusual or unexpected results for data checking and b) familiarisation with the distributions and associations within the data set. Outcome variables which have severely nonsymmetric distributions will be either transformed or categorised.
At the completion of data checking and correction, the data set will be locked and remain blinded until completion of the primary and secondary endpoint analysis.

Statistical Analysis
The demographic and medical characteristics of participants will be summarised using percentages or means and standard deviations. To address withdrawals, loss to follow-up or noncompliance with the protocol, the study analyses will be conducted on both an Intention to Treat (ITT) and Per Protocol (PP) basis.
The ITT population is defined as all participants who are recruited into the study and receive the treatment sachets. Withdrawing and non-compliant participants shall be encouraged to continue with data collection even if stopping treatment. Where data items are missing, we shall use the last value carried forward method to replace missing data within each 12-week treatment block in the ITT analysis. Participants who have significant deviations from the protocol will be removed from the PP analysis after the completion of the ITT analysis. Any deviation from the protocol, missing data and withdrawals will be fully reported.
Between group differences from baseline to the end of each treatment will be undertaken applying independent samples t tests. Linear mixed models (with participant as the random effect) will be used to analyse changes over time. Results will be reported with a 95% confidence interval. Data will be analysed using Statistical Package for the Social Sciences (SPSS software). All tests will be two-sided, and p-value <0.05 will be considered statistically significant. Analyses are conducted blinded to group allocation. The main outcomes of reducing the frequency and severity of VMS will be analysed. The primary analysis is intention-to-treat principle with baseline frequency and treatment order as predictors. We will also describe changes in quality of life, adherence to IMP, adverse effects, retention of participants. Numbers and reasons for withdrawal will be recorded as these are key feasibility outcomes that may inform a larger trial.

Regulatory approval
The trial has been approved by the Human Research Ethics Committee of South Western Sydney Local Health District and will be conducted in a hospital setting with all participants referred and initially screened by a participating oncologist. The oncology clinical trial units at each site will assist with co-ordination of the study. This study may help establish a precedent for subsequent CHM trials within primary care.

Crossover study
The trial is designed as a cross-over trial which involves two consecutively administered treatments for each participant. A 2-week pre-treatment period is used to establish eligibility and the baseline frequency/severity of VMS. After the first treatment period a wash-out period of 28 days was given to ensure that any residual effect from the first treatment did not linger and effect the second treatment. Advantages of cross-over trials are they are statistically efficient (less participants required to achieve the same level of statistical power). It also reduces between participant variability as each participant serves as their own control (treatment A versus treatment B on the same participant) and is suitable for chronic conditions where we are not looking for a cure but alleviating symptoms and improving quality of life and for separating treatment effects from period effects [113,114].

Safety
Safety of the concentrated herbal granules is an import issue for this study. Pura pharm's manufacturing process adheres to strict regulatory and quality control of Good Manufacturing Process (GMP) [98]. The placebo is also manufactured by Pura Pharm to be similar in taste, colour, smell and appearance and is packaged identically to the intervention. Adverse events will be logged at each visit, by the oncologist or a hospital staff member not otherwise involved in the study, to reduce bias. The appointed medical monitor has authority to stop the trial if there are serious adverse events related to the intervention. Additionally, safety is also monitored with hormone assays, liver and renal function tests on all participants. (Table 2).

Collaborating organizations
This project will bring together practitioners of Chinese A further strength of this study is that it extends knowledge further than all previous studies in the field as it includes the following: A. The current study is a randomised, double blind, placebocontrolled, cross-over study whereas the previous study was observational.
B. While the previous study only recruited participants experiencing vasomotor symptoms as a side-effect of Tamoxifen the current study will recruit participants experiencing vasomotor symptoms for any reason after breast cancer treatment.
In addition to measuring hot flushes the current study will measure quality of life and hot flush related interference in daily activities and conduct focus groups to obtain qualitative data.

Potential limitations of the study
One potential limitation of this study is participants selfreporting their VMS. Subjective measures can be inaccurate especially with reporting of night sweats. However, researchers at the Mayo Clinic conducted several VMS trials and argue that the diaries were reliable and consistent [99]. The second limitation is that each participant receives the same formula. Usually in Chinese medicine everyone receives a diagnosis according to the symptoms and signs at that time and formulae are adjusted accordingly.
Although they all have the same western medical diagnosis, i.e. VMS after breast cancer treatment, we shall conduct a TCM diagnosis to see whether a TCM sub-group or groups respond more effectively than other diagnoses.
Thirdly, due to our inclusion criteria results will only be generalisable to women with enough English language proficiency without co-morbidities. If results from this study are promising, follow-up trials of these other groups may be warranted.
Management of VMS in breast cancer survivors can be challenging for clinicians and so an alternative approach may be an important a therapeutic option. The findings may contribute to achieving a novel intervention for alleviating VMS after breast cancer and, thereby, increased quality of life for a growing number of survivors and may develop future Chinese herbal medicine research methodology.

Trial Status
Recruitment commenced in January 2018 and is ongoing.

Competing Interests
The authors declare that they have no competing interests.
Professor Li developed the CHM formula.

Authors Contributions
DP, XZ, AB and PDS designed the RCT with advice from the participating oncologists. PF was consulted to calculate sample size for the study and for advice on outcomes and data analysis.
DP and the clinical trials units will manage all aspects of the trial including, recruitment, screening, administration, processing and analysis, data analysis, reporting and communication of findings to participants. XZ, AB and PDS will supervise the clinical trial. DP wrote the initial draft of the manuscript and XZ, PL and PF were co-authors. All authors have contributed to, read and approved the final manuscript.