Trial design and participants
This study was a single-center, randomized controlled trial (registered in Clinical Research Information Service, CRIS: KCT0001372 http://cris.nih.go.kr/cris/en/search/search_result_st01.jsp?seq=6665 and date of registration and the date of enrollment of the first participant were 14/01/2015 and 16/02/2015, respectively.) conducted from February 1, 2015 to October 31, 2018 at the Samsung Comprehensive Cancer Center in Seoul, South Korea. We used a 2-phase strategy to recruit study participants with incident (new onset) skin dryness after 1 cycle of chemotherapy. First, we recruited patients 35 to 65 years of age with a postoperative diagnosis of stage I to III breast cancer who were expected to receive doxorubicin plus cyclophosphamide (AC), fluorouracil plus cyclophosphamide and doxorubicin (FAC), or taxane (T) as adjuvant chemotherapy. We excluded patients with severe skin dryness, atopic dermatitis, psoriasis, or infectious skin diseases prior to chemotherapy, as well as patients who were taking steroids, antihistamines, anti-depressants or anticonvulsants. Second, patients who had agreed to participate in the trial prior to chemotherapy were contacted after 1 cycle of chemotherapy and asked about the development of skin dryness to confirm final eligibility.
Randomization and interventions
Patients were randomized in a 1:1:1 ratio to receive either a tailored moisturizer (1.5% pseudo-ceramide content), usual care, or a general moisturizer (water-based). The primary comparison was between the tailored moisturizer and usual care. The secondary comparison was between the tailored moisturized and the general moisturizer. Randomization was conducted centrally by an independent statistician using randomly permuted blocks and stratified by menopause status and Herceptin treatment. The allocation information was concealed in sealed envelopes and stored at the investigator’s premises until the end of the trial. The sealed code could only be broken at the request of the principal investigator due either to justified serious adverse events or in case of an emergency. Code breaks and their reasons were documented in the case report form.
The tailored moisturizer (cream: Primera, AMOREPACIFIC, Korea) contained 1.5% of pseudo-ceramides, a lipid that comprises more than 50% of skin lipids. Patients randomized to the usual care group did not received any products. They were told that they could use any product but were instructed to record the type of products and time of dosing during the study period. The general moisturizer (cream: AMOREPACIFIC, Korea) was a general moisturizing cream with the same color and aroma as the tailored moisturizer. Patients assigned to each moisturizer group were asked to apply the moisturizer twice a day (0.5 ml per application) from the first day of 2nd cycle of chemotherapy until 1 month after completion of chemotherapy. The initial application of moisturizers was performed under the guidance of a clinical researcher. The study products were supplied in bottles containing enough solution for approximately 3 weeks and patients received the products on the first day of each cycle of chemotherapy. Moisturizers could be stored under ambient conditions and no specific storage conditions were required. To assess compliance, patients were asked to return the dispensed bottle. We also asked patients if they had skipped the moisturizer for more than 3 consecutive days.
While it was not possible to blind the assignment of treatment vs. usual care group (control group), assignment to the tailored moisturizer and the general moisturizer groups was blinded. Both moisturizers had same texture, smell, and color, and both were supplied in identical 30 mL plastic bottles. Investigators, outcome assessors, and statisticians were blinded to treatment assignment.
Measurements and outcomes
The primary outcome was patient-reported severe skin dryness on the face at 1 month after completion of chemotherapy. Skin dryness was assessed by the Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE™) 26,27. The PRO-CTCAE question originally included 5 categories that were coded as none (none), moderate (a little or moderate), and severe (severe or very severe). Severe skin dryness was thus defined as severe or very severe dryness using the PRO-CTCAE questionnaire.
The secondary outcomes were two objective skin parameters: cheek surface sebum and water content. Objective skin parameters were measured on the front of the right cheek by trained researchers according to a standard protocol using a Multi-Probe Adapter System (Courage-Khazaka, Germany). Sebum content (approximated to µg/cm2) was measured only once per visit, while water content (in arbitrary units, AU) was measured three times per visit and the values were averaged. During measurements, patients were asked to lay face up on a bed keeping a straight body position in all visits at the room maintained temperature between 20 and 22 °C and humidity between 30 and 40% 28,29. All outcomes were measured at baseline (beginning of the 2nd cycle of chemotherapy), 3 weeks after baseline, and 1 month after completion of chemotherapy.
In addition, we assessed patient-reported distress due to skin dryness and dullness as well as skin dryness and dullness using a 10-point visual analog scale with scores ranging from 0 to 10 (0 = not at all, 10 = very much). Impairment of dermatological quality of life was measured using the Korean version of the Dermatology Life Quality Index (DLQI) 30. The DLQI consists of 10 simple questions to assess quality of life related to skin problems. Each question had four possible answers scored from 0 to 3. The answers to all questions were added to obtain an overall score on a scale from 0 to 30, with higher scores indicating more impairment of dermatological quality of life 30.
Sociodemographic and behavioral information were collected at baseline using a questionnaire. Clinical information were obtained from electronic medical records. Adverse events were classified and graded according to the Common Terminology Criteria for Adverse Events, version 4.0 [12].
Statistical analysis
The sample size for the trial was calculated to address the primary hypothesis that the tailored moisturizer was more effective than usual care (primary comparison) in preventing chemotherapy-associated dryness at 1 month after completion of chemotherapy. We then adjusted the sample size to incorporate an additional comparison with the general moisturizer (secondary comparison). Based on prior experience, we expected that 60% of patients in the usual care group and no more than 30% of patients in the tailored moisturizer group would report severe dryness on the face 1 month after completion of chemotherapy. We adjusted the p-values for two comparisons using the Bonferroni method. To conduct pairwise χ2 tests comparing proportion of severe dryness under Bonferroni adjustment (2.5% level of significance; P <0.025) with 80% power, the sample size of the trial was 58 patients per arm. Since we expected 15% losses to follow-up, we increased the sample size to 68 patients per arm, for a total of 204 randomized participants.
All analyses were conducted using the intention-to-treat principle, whereas study patients were assigned to their randomized group irrespective of compliance with the study intervention. Differences in baseline characteristics among 3 groups were compared using χ2 tests for categorical variables and t-tests for continuous variables.
We used mixed effect models for longitudinal data analysis to model changes over time in sebum, and water content. P-values <0.05 were considered statistically significant except for the primary outcome (P <0.025). All analyses were conducted using Stata 15.0 (Stata Corp.; College Station, TX).
Ethical considerations
This study was conducted in accordance with the Declaration of Helsinki and with local laws and regulations. The study protocol was approved by the Medical Ethical Committee of Samsung Medical Center in Seoul, Korea. Eligible patients were fully informed about the study and volunteered to write informed consent. If during the study a patient no longer wished to participate, for whatever reason, the patient was allowed to withdraw consent at any time. An independent monitoring committee reviewed accumulating safety data throughout the trial.
Protocol changes
We revised exclusion criteria because of changes in cancer treatment practice caused by changes in insurance reimbursement. Starting on February 2nd, 2016, we did not exclude patients who received Herceptin. Starting on August 8th, 2016, we did not exclude patients who received taxanes. In addition, to minimize losses to follow-up before patients attended the clinic-based baseline measurement, from November 11th, 2017, we changed the time of randomization to the baseline study clinic visit.