The Preventive Effects of Boron-Based Gel on Radiation Dermatitis in Patients Being Treated for Breast Cancer: A Phase III Randomized, Double-Blind, Placebo-Controlled Clinical Trial

Breast cancer is one of the most common cancers in women worldwide, with 1.9 million incident cases in 2017; that is to say, it is the second eminent cause of cancer deaths in women with an estimated 268,000 new cases and 41,760 deaths in 2019 [1, 2]. Furthermore, globally the age-standardized incidence rate for breast cancer increased by 17% between 1990 and 2017 [3, 4]. Radiation therapy (RT) is one of the most common treatments for breast cancer [1] as it has been shown to decrease locoregional recurrence and increase patient survival [5-7]. However, one of the most common side effects of RT is radiation dermatitis (RD), which is reported in >90% of breast cancer patients who were treated using RT [8, 9]. Despite the strong likelihood of RD, there are currently no definitive guidelines for the management or prevention of RD [10]. Skincare advice such as steroidal/nonsteroidal topical agents, systemic therapies, modes of radiation delivery, and dressings, has been suggested, but there is not yet sufficient evidence to support any of these interventions [10-12]. Therefore, new treatments for reducing or preventing RD should be tested in order to improve its management.

Boron is a very stable element with wound healing abilities, as reported by previous in vivo and in vitro studies [13-16]. Furthermore, in order to meet the requirements for phases I and II of the Food and Drug Administration guidelines, 1 study examined the safety and preventive effects of boron-based gel (3% sodium pentaborate pentahydrate) on RD using a limited number of samples [17]. Although this study was valuable in terms of the application of boron-based gel to human patients for the first time, the detailed effect on RD was not examined. Furthermore, it was not clear whether randomization effectively balanced the confounding variables between the 2 groups, which is a common problem with small sample sizes [18]. Therefore, the present phase III clinical trial study was conducted to measure the preventive effects of the aforementioned boron-based gel on different types of RD using a randomized, controlled double-blind trial.

This phase III clinical trial used a double-blind parallel-group design with a placebo control group and randomized group allocation. The present phase III clinical trial study was conducted to measure the preventive effects of the aforementioned boron-based gel on different types of RD using a randomized, controlled double-blind trial. The Ethical Committee of Tabriz University of Medical Sciences approved the protocol study (IR.TBZMED.REC.1397.919) and the study was retrospectively registered and approved at IRCT.ir on the 29-04-2019, IRCT20190305042925N1 and ClinicalTrials.gov, NCT04239560.

The study population consisted of female breast cancer patients who were admitted to Shahid Madani Hospital in 2019 and informed consent forms of patients were obtained. Patients who were 18–75 years old and had no previous history of radiotherapy were included in the study. Pregnant women, patients with unknown causes of dermatitis, and those who were not willing to participate in this study were excluded. As there was no similar study to compare our outcomes, we conducted a pilot study with 30 patients. Based upon the incidence of erythema in the pilot study, a sample size of 16 per group was calculated to ensure a power of 0.8 and a 5% risk of type I error.

In this study, which is based on the CONSORT 2010, 384 out of 440 people were included in the study. Out of the 384, 280 people were randomly included in the intervention group and 104 people were included in the control group. With the removal of 62 people from the intervention group, 218 people from this group and with the removal of 18 people from the control group, 86 from this group people finally completed the study. Among the reasons for the decrease or removal in participants were the impossibility of attending sessions for the patient, the death of the patient, and especially the pandemic condition of COVID-19. Given the fact that the sample size was the same in both groups, we used the formula = Rand () in Microsoft Excel to randomize and to increase the precision and balance of the samples in the randomization. We used several blocks per group to account for patient withdrawal from the plan. It will also be provided to the physician to cover the allocation of matte envelopes. In addition, a gel containing an active ingredient in boron and a placebo gel was provided. The placebo was completely indistinguishable, in that it did not contain the substance but was completely identical in terms of the shape and size of the container, as well as odor and color (this action was taken by the pharmaceutical company). The important point is that the patient was told that the gel used for the patient might be medication or placebo. Clinicians and patients were blinded. Also, all patients had a karnofsky performance status above 90 at the beginning of treatment.

However, we increased the sample size to 181 for the intervention group and 76 subjects for the placebo group in order to meet the sample size guidelines for the Food and Drug Administration Phase III Clinical Trial. The sample size was increased to allow for a >20% attrition rate during the study and to increase the effectiveness of randomization in balancing the confounding variables between the 2 groups. As there was no standard treatment for RD in breast cancer patients being treated with RT, increasing the sample size was not ethically questionable.

The aim of the study was fully described to all eligible subjects and informed consent was obtained. Following this, the patients were assigned to the intervention and placebo groups using the random sequence generated by random allocation software. This process was performed by a person who was not otherwise involved in the study. Fifteen minutes before each radiotherapy session, participants in the intervention group were given a Carbopol gel containing 3% sodium pentaborate pentahydrate, while those in the placebo group received Carbopol gel which did not contain any chemical substance. As the gels were applied to the RT target area before each RT session (a total dose of 50 Gy/day for 25 days), there were no compliance problems in this study. All patients had radiotherapy according to the clinical protocol of Shahid Madani and Tabriz Valiasr Hospitals. Gel administration was performed on the target sites of RT application which was determined by a nurse, a radiotherapist, and a radio oncologist. Finally, after 25 days of treatment, the study’s outcomes were examined for all patients in the 2 groups. Dermatitis and its grades, including erythema, dry desquamation, and moist desquamation were the main outcome variables in this study, based upon the Radiation Therapy Oncology Group (RTOG) criteria as follows: grade 0 (no change), grade 1 (faint erythema, dry desquamation, epilation, and/or decreased sweating), grade 2 (tender or bright erythema, moderate edema, and/or patchy moist desquamation), grade 3 (moist desquamation in areas other than skin folds and/or pitting edema), grade 4 (ulceration, hemorrhage, and/or necrosis), and grade 5 (death).

The distributions of the quantitative variables were initially examined using the Kolmogorov-Smirnov test and histogram plots. Those with and without normal distribution were reported as mean ± standard deviation and median ( interquartile range, respectively. Besides, qualitative variables were reported as both numbers and percentages (%). Normally and non-normally distributed variables were compared between the 2 groups using Independent Samples T Tests and Mann-Whitney U tests, respectively. Qualitative variables were compared between 2 study groups using the χ² and Fisher’s exact test if any. Also, the number needed to treat (NNT) was reported along with a 95% confidence interval (CI) to examine the beneficial effect of the intervention. NNT is an absolute effect measure that was interpreted, in the current study, as the number of patients needed to be treated with the intervention (vs. placebo) to prevent one additional outcome of interest (dermatitis, erythema, dry desquamation, and moist desquamation) within a defined period [19, 20]. Finally, multivariable logistic regression was used to identify whether the confounding variables were distributed equally between the 2 groups and to adjust for them if needed.

Initially, 440 patients were assessed for eligibility but 384 patients out of them were randomized into treatment (n = 280) and placebo (n = 104) groups. Of them, 280 and 104 patients received the allocated intervention in the treatment and placebo groups, respectively. As a result of the lost to follow-up and discontinued interventions in both treatment (n = 62) and placebo (n = 18) groups, the numbers of patients analyzed in the treatment and placebo groups were 218 and 86, respectively. The mean age of the patients in the intervention and placebo groups were 50.0 ± 12.1 and 48.1 ± 11.1, respectively, which was not statistically different (p = 0.240). There were no significant (p > 0.05) differences between the 2 groups on most of the factors associated with dermatitis such as specific diet, hormone therapy, contraceptive use, Cyclosporine use, chemotherapy use, folic acid use, vitamin C use, vitamin B12 use, multivitamin use, smoking, antibiotic use, kidney diseases, type II diabetes, breast implantation, and scleroderma; (Table 1). However, at baseline, the body mass index was significantly higher in the placebo than in the intervention group (29.3 [6.6] vs. 27.3 [5.4]; p < 0.01).

The present study found that dermatitis was more common in the placebo group than in the intervention group (98.7% vs. 9.9%; p < 0.001). Therefore, the intervention was found to be highly effective such that 1.1 patients needed to be treated to prevent dermatitis in 1 patient (NNT = 1.1; 95% CI: 1.1–1.2). Figure 1 demonstrates that the boron-gel administered in the breast area of the 2 patients prevented the RD in this area remarkably. Besides, erythema was also found to be more common in the placebo group than in intervention group (96.1% vs. 12.2%; p < 0.001) and again the intervention was found to be effective such that erythema could be prevented in 1 patient by treating 1.2 patients (NNT = 1.2; 95% CI: 1.1–1.3). Dry desquamation was also more prevalent in the placebo group (50% vs. 3.9%; p < 0.001), and the preventive effect of the intervention was again considerable (NNT = 2.2; 95% CI: 1.7–2.9). Although moist desquamation was not a prevalent outcome compared to those mentioned above, it was significantly higher in the placebo group (18.4% vs. 0.6%; p < 0.001). To prevent moist desquamation in 1 patient, 5.6 patients would need to be treated (NNT = 5.6; 95% CI: 3.8–11.0; Table 2). Clinical presentations of RD such as erythema and dry and moist desquamation in 3 patients allocated to the placebo group are illustrated in Figure 2. The skin toxicity was also graded according to RTOG criteria. There were higher numbers of patients with grade 1 and grade 2 skin toxicity in the placebo group than the treatment group (p = 0.01 and p < 0.001, respectively; Table 3).

This is the first phase III clinical trial examining the protective effects of boron-based gel on different types of RD from RT. The intervention was significantly effective in the prevention of dermatitis, erythema, dry desquamation, and moist desquamation, as the NNTs were reasonable.

RD reduces the quality of life, compliance with RT treatment, and can be considered one of the reasons for discontinuing or deferring RT treatment [8]. Several different approaches have been taken to prevent RD, including the mode of RT delivery, topical treatments, skincare, and other treatments. A recent review of 27 RD studies investigating the mode of RT delivery showed that these consisted of different doses and fractionation protocols. The different types of RT examined included: intensity-modulated RT, accelerated partial breast irradiation, prone positioning, radiation of the internal mammary chain, simultaneous integrated boost, concurrent systemic therapy, and tomotherapy. Although the aforementioned RT delivery methods generally demonstrated less RD than from conventional RT, the findings for several of these methods are controversial [21].

A review of 50 studies from 1979 to 2017 showed the effectiveness of various topical treatments for the prevention and treatment of RD in breast cancer patients [21]. In these studies, they placed topical treatments into 3 categories: nonsteroidal creams, steroid treatments, and barrier products. Among nonsteroidal creams, oil-in-water emulsions, aqueous creams, heparinoid creams, recombinant human epidermal growth factor cream, and sucralfate cream were found to be more effective than a placebo. For steroid treatments, emollients, aqueous creams, petrolatum gel, and melatonin based creams were more effective than placebo gels. Barrier products, such as barrier cream, 3M No string Barrier Film, Mepitel, Mepilex Lite Dressing, and silver leaf dressing also showed more preventive efficacy than aqueous creams [21]. Several studies have also suggested the use of supplements to increase the efficacy of the intervention, such as adlay bran extract, curcumin, verum, and glutamine [21]. Skincare measures such as deodorant use and washing during RT have also been studied, but their effectiveness is currently unclear [21]. Other techniques such as photobiomodulation, massage and laser therapy have also been studied, but further research is needed [21].

The previously mentioned studies showed that interventions for the management of RD have been investigated over the last 3 decades and the effectiveness of these treatments was reported using a placebo group [9, 11, 12, 21]. However, the main questions still remain: “why is RD not managed effectively in patients with breast cancer and why isn’t there definitive evidence to support any particular intervention in the prevention or treatment of RD”? One probable answer is that, despite a large number of studies in this area, there have been very few high-quality comparative studies that have unequivocally supported the effectiveness of an intervention for preventing RD [9, 11, 12, 22]. Another possible reason is that previous studies have not distinguished between “statistically significant” and “clinically important” differences between the treatment and placebo groups. As RD occurs in >90% of RT patients, any treatment which results in a slight decrease in RD (e.g., at least 20%) will be statistically significant and therefore considered to be effective. However, researchers and clinicians should carefully consider whether having 70% of the patients with RD, after treatment, is clinically acceptable or not. Clinicians would like to eliminate RD to increase the quality of life and patient compliance with the treatments. Therefore, statistically significant results should be interpreted with some caution and in combination with clinically important differences. The results of our study show that dermatitis decreased to 9.9% (Cf. 98.7% in the placebo group) in the treatment group, which is very favorable from a clinical and patient point of view (i.e., if 1.1 patients is treated with boron-based gel, dermatitis will be prevented in 1 patient). The treatment was not only effective for dermatitis but also resulted in lower erythema (12.2% vs. 96.1%), dry desquamation (3.9% vs. 50%), and moist desquamation (0.6% vs. 18.4%).

This clinical trial utilized a relatively large sample size to allow comparison between 2 groups with regard to less common types of RD, such as moist desquamation. In addition, due to the large sample size, randomization effectively balanced the confounding variables between 2 groups. Moreover, patient compliance in the present study was very high, as the patients used the gel under the supervision of the researchers before RT.

To the best of our knowledge, the present study is the first to report NNT (which considers the clinical importance of an intervention rather than statistical significance) for interventions aimed at preventing RD. However, 1 limitation of the present study is that although patient satisfaction was verbally requested, these data were not systematically recorded. In addition, the long-term outcomes, such as necrosis, could not be studied in the present study.

The boron-based gel was found to have a strong preventative effect on most types of RD measured in women being treated for breast cancer using RT, including dermatitis, erythema, dry desquamation, and moist desquamation. This intervention needs to be included as a preventive measure for patients being treated for breast cancer using RT. Besides, clinical guidelines must be developed and clinicians should be informed about this effective treatment for RT induced RD.

We would like to thank Tabriz University of Medical Sciences and Yeditepe University for their financial support. We would like to thank Yeditepe University Writing Center for their proofreading support.

Written informed consent was obtained from all participants and the study was approved by the Ethical Committee of Tabriz University of Medical Sciences (IR.TBZMED.REC.1397.919).

The authors declare no competing interests. F.S., A.D,. and S.D. have a licensed patent related to this formulation.

The present study was jointly supported by Tabriz University of Medical Sciences, Tabriz, Iran (No. 62036), and Yeditepe University, Istanbul, Turkey. The funding body had no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

F.S., F.S.N., S.K.S., D.A., and E.A. contributed to the study concept and design. M.B.P., H.B., M.M., R.E.Z., A.G.J., B.N., H.S., and A.R.N. contributed to data collection. F.S., F.S.N., M.B.P., H.B., and M.A.-H. contributed to data analyses, manuscript drafting, and interpretation of results. A.D., S.D., and A.Ç.B. commented on the manuscript. All authors read and approved the final manuscript.

Datasets are available through the corresponding author upon reasonable request. Written informed consent was obtained from the patients for publication of this paper and any accompanying images and videos. A copy of the written consent is available for review by the Editor of this journal.