The effect of a helmet type, home-use low-level light therapy device for chemotherapy-induced alopecia: study protocol for a randomized controlled trial

This protocol conforms with the SPIRIT 2013 (Standard Protocol Items: Recommendations for Interventional Trials) statement (see Additional file 1 for the SPIRIT Checklist). There is no public or patient involvement in the design of the protocol. This study is a two-arm, randomized, controlled trial to determine the efficacy of LLLT for the treatment of CIA. Patients with breast cancer at Keelung Chang Gung Memorial Hospital, Taiwan, were screened by a general surgeon or an oncologist to diagnose and determine the severity of chemotherapy-induced alopecia. The diagnosis used the Common Terminology Criteria for Adverse Events [26].

After confirmation of alopecia, patients were referred to the department of Chinese medicine. The eligibility criteria were verified during the prior assessment, which was standardized by the outpatient department. If the patient was eligible and was willing to participate in the study, their full name and phone number was given to one of the study assessors to schedule a baseline assessment. One hundred six eligible participants were given a baseline assessment. At the time of the baseline visit, a consent form was completed by the participants or a legal guardian, as required by the Chang Gung Medical Foundation Institutional Review Board. A total of 106 participants were randomly distributed into two groups, with 53 individuals in each group:

The primary outcome is the difference in the hair counts for a target area between the baseline and the end of week 4. The secondary outcomes include the change in hair count at the end of week 1, week 2, and week 3 and hair width at the end of week 1, week 2, week 3, and week 4 and the change in distress, quality of life, and self-esteem due to chemotherapy-induced alopecia at the end of week 4. The flowchart for the study is shown in Fig. 1 and Table 1.

Ten attending physicians in the Chinese Medicine Department at Keelung Chang Gung Memorial Hospital were involved in patient recruitment. Ten trained nurses assisted the attending physicians in providing patients with information sheets during the initial assessment. A trained research nurse supervised completion of the consent form during the baseline assessment. Five resident physicians conducted baseline and follow-up assessments, using a phototrichogram, determining any adverse effects and helping patients complete the questionnaires. An external statistician randomized the groups and ensured that the allocation was confidential. The eight individuals who were initially involved in the protocol design attended meetings every 3 months to discuss the progress of the study. An independent data monitoring expert and Institutional Review Board members conducted an annual audit. A coordinator was assigned to ensure the integrity of each dataset and two interns participated in collecting data for the study.

The procedures and consent form were approved by the Chang Gung Medical Foundation Institutional Review Board (protocol no. 202200395A3), and the audit was conducted once a year by an independent data monitoring expert and Chang Gung Medical Foundation Institutional Review Board. This study is also registered at ClinicalTrials.gov (NCT05397457) (Table 2).

One hundred six eligible patients were recruited at Keelung Chang Gung Memorial Hospital, Taiwan. Inclusion criteria are as follows: (1) female patients aged between 20 and 75 years with breast cancer; (2) completing chemotherapy no more than 2 weeks previously; (3) receiving chemotherapeutic agents containing taxanes, anthracyclines, or fluoropyrimidine; (4) presence of chemotherapy-induced alopecia that is classified as grade 2 in the Common Terminology Criteria for Adverse Events [26]; and (5) a life expectancy of at least 6 months.

Exclusion criteria include the following: (1) any serious mental illness or history; (2) taking psychotropic drugs; (3) taking any of the following medications for 6 months before initiation of the study: topical minoxidil, spironolactone, or topical calcitriol; (4) receiving scalp cooling during chemotherapy; (5) a medical history of dermatosis, scalp tumor, or melanoma; (6) severe liver or kidney damage; and (7) pregnancy or potential pregnancy.

Eligible patients participated in this study after completing the consent form, which asked if patients agreed to the use of their data, should they choose to withdraw from the trial. Participants’ permission was sought to share the relevant data with project-related individuals, including experts from universities or regulatory authorities. This trial does not involve collecting biological specimens for storage. All participants were closely monitored during the study. There is no anticipated harm for participants in this clinical trial, and no compensation was provided. Post-trial care is described in the following.

A realistic difference was used to calculate the sample size. The calculation uses an 80% statistical power and a significance level (alpha error) of 5%. Similar to the study by Stadler in 2021 [23], the calculated effect size (Cohen’s d) was approximately 0.54, which corresponds to a medium effect. This calculation gives a total sample size of 88 individuals, but for an expected dropout rate of 20%, the revised requirement is 106 participants. All calculations used the G*Power software version 3.1.9.4 [27] to ensure the accuracy and consistency of the methodology. To increase our pool of participants, referrals were taken from general surgeons and oncologists and posters were used for recruitment.

The statistician, who was not involved in participant recruitment, assessment, or treatment, devised the randomization schedule using a computer-generated random number table. Equal numbers of participants were allocated to each one of the two groups to give 53 in each group. The randomization code was only released after a patient had been recruited for the trial, when all the necessary baseline measurements had been completed.

Each envelope in the randomization list contains a randomization code for either the therapy group or the control group. The attending physician opened an envelope to determine the treatment condition to be administered, based on the information provided. The randomization code was hidden from assessors, the coordinator, data collectors, and statisticians, but because the manufacturer cannot provide a sham device that emits a red light without any LD, patients in the control group were not blinded. Assessors, the coordinator, data collectors, and the external statistician were also not blinded, but the assessor performed the assessment objectively. After the baseline assessment, the Chinese medical physician informed the participant of the type, time, and frequency of treatment. The external statistician analyzed data.

A helmet-type LLLT device was used by Hyojin Kim et al. and developed by Won Technology (Daejeon, Korea) [21]. The light sources for the device for this study are laser diodes (LDs) that emit 650 nm (5 mW, 69 units). All diodes run simultaneously through six cycles. Each cycle is composed of 2 min 50 s on and 10 s off so the energy per unit for each light source with Orion-PD ROHS (Ophir Optronics Ltd., Jerusalem, Israel) is 5 mW (± 20%) for LD. The total energy density for the device is 59 mW/cm². Energy fluence is 60.43 J/cm² for 18 min of treatment and 70.50 J/cm² for 21 min of treatment. Participants in the LLLT group used the device three times per week for 4 weeks.

Prior to statistical analysis, the data was imported into a spreadsheet and checked randomly by one of the authors, who was not involved in entering the data into the spreadsheet, to confirm whether there was any error in the transcription of the data. Statistical analysis uses SPSS v.15 for Windows. The baseline demographic and medical characteristics of participants for the study is analyzed using appropriate descriptive statistics for the two groups. To account for potential baseline confounding variables, analyses are adjusted. These variables include demographic and medical characteristics, particularly breast cancer staging and the dose and regimen for chemotherapy.

Univariate analysis is used to determine the association between the two variables, without adjusting for any confounding variables. Multivariable analysis is used to determine the relationship between the exposure and outcome variables while controlling for the effects of the confounding variables. A sensitivity analysis determines the robustness of the results to different modeling assumptions and to the inclusion or exclusion of certain variables. Intention-to-treat (ITT) analysis is used.

The primary outcome, which is the difference in hair count between the baseline and the end of week 4, is analyzed using an independent-sample t-test and the mean, standard deviation (SD), and 95% confidence interval are reported. Generalized estimating equations (GEEs) are used to determine any improvement in hair count and hair width at the end of week 1, week 2, week 3, and week 4. The change in CIA distress, self-esteem, and QoL at the end of week 4 is determined using an independent-sample t-test. Adverse events are analyzed using a two-sided Fisher’s exact test.

The homogeneity of the low-level light therapy effects on the primary outcome is determined using a one-way ANOVA (analysis of variance) with predefined subgroups, including whether the patient has been in menopause, immunohistochemical diagnosis (ER, PR, HER-2), and particularly the dose and regimen for hormone therapy. The study determines whether hormone therapy is related to recovery from chemotherapy-induced alopecia.

There are no reports on the adverse effects of LLLT. The assessors monitored each case to decide whether treatment should be terminated. The project management group also held meetings every three months to review the progress of participants, determine the cause of any withdrawals, determine whether the study flowchart is to be revised, and determine if additional participants were needed to account for withdrawals. In the event of a withdrawal, the assessor contacted the participant by phone to inquire whether any adverse effects were experienced. The assessor also explained the benefits of LLLT and encouraged the participant to continue if the side effects were tolerable. An audit was conducted once a year by an independent data monitoring expert and the Chang Gung Medical Foundation Institutional Review Board. During the audit, in order to objectively assess the risk of the intervention, members of the Chang Gung Medical Foundation Institutional Review Board randomly contacted the participants to confirm whether the participants were well informed by the investigators and whether there were any adverse effects. If there was any doubt, the project was terminated.

Ethics approval was obtained in May 2022 from Chang Gung Medical Foundation Institutional Review Board. The trial was registered on 1 June 2022 (registration number: NCT05397457). Recruitment and training of the assessor began in May 2022. Recruitment of participants began on August 17, 2022. The project will be completed in August of 2025. The statistical analysis will be completed by the end of October 2025.