Validation of the prophylactic efficacy of urea‐based creams on sorafenib‐induced hand‐foot skin reaction in patients with advanced hepatocellular carcinoma: A randomised experiment study

Abstract Background Hand‐foot skin reaction may influence the effectiveness of patients' treatment, patient quality of life, and the economics of health care. An effective prophylactic dermatological cream for preventing sorafenib‐induced hand‐foot skin reaction (HFSR) is yet to be identified. Aim The aim of this study is validated the prophylactic efficacy of urea‐based creams on sorafenib‐induced hand‐foot skin reaction in patients with advanced hepatocellular carcinoma. Methods This was a randomised double‐blind experimental study. A total of 129 patients with advanced HCC were randomly assigned to three groups. The comparison group received best supportive care (BSC), group A received BSC plus a moisturising cream, and group B received BSC plus a 10% urea‐based cream. Incidence of HFSR and cutaneous wetness were assessed 3 days before starting sorafenib and each week after starting sorafenib for 8 weeks. Results No significant difference was observed in the incidence density of sorafenib‐induced HFSK (comparison group/A group, p > .05; comparison group/B group, p > .05). Group B reported significantly better cutaneous wetness of hands in the seventh week after starting sorafenib (p < .05) and of feet during the first 6 weeks (p < .05–.001). Conclusion This study found a nut size amount of a 10% urea‐based cream applied twice a day can maintain patients' cutaneous wetness in the first 6 weeks after starting sorafenib than moisturising‐alone cream. But it cannot reduce the occurrence of HFSR. Thus, the result supports nut‐size dose of the 10% urea‐based cream three times a day may be an appropriate dose to prevent HFSR. Clinical Trail Registration Number: NCT04568330.


| INTRODUCTION
Cancer incidence and mortality worldwide are increasing rapidly and are responsible for the majority of global deaths. [1][2][3] According to estimates by GLOBOCAN 2018 from the International Agency for Research on Cancer in the World Health Organisation, liver cancer is the fifth most commonly diagnosed cancer and the fourth leading cause of cancer death, with 841 080 new diagnoses and 81 631 cancer deaths. 2 Of the pathological types, 85%-90% are hepatocellular carcinoma (HCC), the highest incidence of which is in East Asia. [1][2][3][4] The Sorafenib Hepatocellular Carcinoma Assessment Randomised Protocol (SHARP) demonstrated that sorafenib offers significant benefits in terms of prolonging median overall survival from 7.9 to 10.7 months (hazard ratio in the sorafenib group, 0.69; 95% confidence interval [CI], 0.55-0.87; p < .001) and radiologic progression from 2.8 to 5.5 months (hazard ratio in the sorafenib group, 0.58; 95% CI, 0.45-0.74; p < .001). 5 In the Asia-Pacific population with advanced HCC and Child-Pugh liver function class A, it prolonged median overall survival from 4.2 to 6.5 months (hazard ratio in the sorafenib group, 0.68; 95% CI, 0. 50-0.93; p < .05) and median time to progression from 1.4 to 2.8 months (hazard ratio in the sorafenib group, 0.57; 95% CI, 0. 42-0.79; p < .001). 6 Thus, sorafenib has been recommended as a standard treatment for advanced HCC. [3][4][5][6][7] Sorafenib is an oral multikinase inhibitor that can block the proliferation of tumour cells by targeting the Raf/MEK/ERK signalling pathway. 8 It also exerts an antiangiogenic effect by targeting the receptor tyrosine kinases vascular endothelial growth factor receptor (VEGFR)-2, VEGFR-3 and platelet-derived growth factor receptor (PDGFR)-β. 8,9 Approximately 10%-62% of patients treated with this agent reported a hand-foot skin reaction (HFSR). 5,6,9 HFSR is also called hand-foot syndrome, palmer-plantar erythrodysethesia, acral erythema, and Burgdorf reaction. 9 It is a dose-accumulated side effect, the highest incidence of which has occurred in patients in Asia-Pacific undergoing advanced HCC treatment with sorafenib. 6 The mechanisms of sorafenib-induced HFSR remain unclear. One theory suggests that inhibitions of multikinase, VEGR, and PDGFR may impede vascular repair mechanisms and induce histopathological characteristics, including keratinocyte vacuolar degeneration, confluent keratinocyte necrosis, perivascular or lichenoid lymphocyte-predominant infiltrate, spongiosis, hyperkeratosis, parakeratosis, and epidermal cleavage. [8][9][10][11] HFSR often occurs 2-6 weeks after starting sorafenib and is related to pharmacologic properties such as dose, peak plasma level, total accumulative dose, and schedule of administration. 8,10 The presentation of HFSR is unique with a prodrome comprising dysesthesia on the palms and soles with tingling. 9 This then progresses to a burning pain and is followed by the symmetrical development of erythematous, edematous, and blistering plaques with hyperkeratosis. 9,11,12 The HFSR can have a significant impact on patients' daily living activities, quality of life, and the economics of health care. 12,13 To date, varying forms of the management of sorafenib-induced HFSR have been recommended based on grades of severity. 14,15 These grades are determined by the National Cancer Institute Terminology Criteria for Adverse Event (NCI-CTCAE). 12,14,15 During the pretreatment period, patients should be informed of dose-dependent HFRS and asked to self-report. 13 Health care specialists must also comprehensively evaluate the baseline condition of patients' limbs. 12,13 A method of   contacting health care specialists for the early diagnosis of HFSR should be   available. 15 Under grade I HFSR, patients may experience mild painless erythema, edema, hyperkeratosis, numbness, dysesthesia, paraesthesia, tin-gling, or swelling with no restriction of daily activity. 13,14 Suggested recommendations for management are to maintain the current dose of sorafenib, avoid hot water, use a moisturising cream and a 20%-40% ureabased cream, and wear thick cotton gloves and socks. [13][14][15] When patients develop grade II HFSR with painful peeling, blisters, bleeding, edema, erythema, or hyperkeratosis and their daily activity starts to become limited, a 50% dose reduction of sorafenib is recommended for 7-28 days along with treatment for grade I toxicity with an additional prescription of 0.05% clobetasol ointment and 2% lidocaine, codeine, or pregabalin. 14,15 Interruption of treatment and treatment for grade II HFSR before the severity improves to grade 0 or grade 1 are advised when grade III HFSR develops with symptoms that restrict daily activity, such as refractory painful moist desquamation, ulceration, blistering, or hyperkeratosis. 12,15 Urea can preserve cutaneous hydration and exert keratolytic action, therefore, urea-based creams can improve the hyperkeratotic condition and has been widely used for tyrosine kinase inhibitor related HFSR. 14-16 A randomised controlled trial validated the use of a 10% urea-based cream with best supportive care (BSC) from treatment day 1 compared with BSC with no creams and reported significant prophylactic efficacy in terms of decreasing the grades of sorafenib-induced HFSR in patients with advanced HCC (hazard ratio in the urea-based cream group, 0.457; 95% CI, 0.344-0.608; p < .001) or delaying median time to first occurrence of HFSR (hazard ratio in the urea-based cream group, 0.658; 95% CI, 0.541-0.799; p < .001). 17 The use of an empiric 12.5% urea-based cream for patients with advanced HCC and a hydrocolloid dressing containing ceramide on the feet of patients with renal cell carcinoma can improve sorafenib-induced HRSF. 18 This result indicates that the appropriate prophylactic dose of a urea-based cream and its effectiveness compared with moisturising-alone creams remain unclear. [17][18] The aim of this study was to validate the prophylactic HFSR incidence density and cutaneous wetness of 10% urea-based creams on sorafenib-induced HFSR in patients with advanced HCC.

| Design
This was a randomised double-blind experimental study. The sample size was estimated using G. power software version 3.1, which was established using logistic regression, odds ratio: 3.8, and power: 0.80. 19 The estimated sample size needed to be at least 125. A CON-SORT diagram of patients included in the study is presented in Figure 1. In total, 150 patients received sorafenib with HCC in a general medical centre in Taiwan between 1 January 2014 and 31 December 2014. Of these, 129 were eligible and agreed to participate. Using EXCEL random sampling, they were randomly assigned in a ratio of 1:1:1 to the following treatment groups: BSC alone (comparison group), BSC plus moisturising cream (group A), and BSC plus 10% urea-based cream (group B). Each group initially comprised 43 patients; however, considering deaths and loss to follow-up, four patients were eliminated from the study. Finally, a total of 42, 41, and 42 patients were included in the comparison group, group A, and group B, respectively. A case manager recruited the eligible patients and was also responsible for obtaining informed consent and providing patient education. A research employee was responsible for recording patients' demographic data, providing an unlabelled cream, and ensuring that the previous container of cream was not exhausted.
A medical oncologist or a nurse was responsible for assessing the severity of HFSR and cutaneous wetness of patients. The assessment was conducted 3 days before starting sorafenib treatment and every 7 days after starting sorafenib treatment for 8 weeks. Creams were provided after the assessment. When patients developed HFSR, they were referred to the most appropriate form of management.

| Inclusion and exclusion criteria
This study population comprised patients with the following characteris- (2) allergic history to urea; (3) ulceration, blisters, infective problems on the palms or soles; or (4) who had undergone previous surgery, systematic chemotherapy, or frequent radial ablation were excluded.

| Intervention
The comparison group who received BSC alone were (1) informed of potential presentations of HFSR, (2) asked to wear waterproof gloves before undertaking household work or working with water,

| Outcomes and assessment
Data on confounders and two end-points, incidence density of HFSR and cutaneous wetness, were collected. The confounders included gender, age, number of chronic illnesses, number of metastatic regions as well as levels of white count, haemoglobin, bleeding time, liver enzymes, albumin, and electrolyte. indicates mildly moist skin, and 47% or more indicates moist skin.
When using the scanner, examiners had to confirm that it was fixed to the skin.

| Comparison of confounders
As noted previously, 129 patients were randomly assigned to three groups. Table 1 presents the comparison of confounders between groups. No significant difference was found among the three groups.

| Incidence density of HFSR
The incidence density of HFSR is presented in

| Comparisons of cutaneous wetness
The majority of patients had dry skin on their hands or feet at different time points. The cutaneous wetness was 32.35%-33.31% (as shown in Figure 2 week 4 (p < .05), week 5 (p < .05), and week 6 (p < .001).

| DISCUSSION
The aim of this study was to validate the prophylactic efficacy of 10% urea-based cream on sorafenib-induced HFSR in patients with advanced HCC. The results indicated that BCS plus 10% urea-based cream can significantly improve cutaneous wetness compared with BSC alone or BSC plus moisturising cream. However, no significant difference was observed in the incidence density of HFSR between the three groups.
The characteristics of patients were similar to the global distribution with a majority of elderly people and a gender ratio of 2-3 males to 1 female. 2,6 All followed the recommended guideline of taking oral sorafenib 400mg twice per day to treat advanced HCC with Child-Pugh liver function class A. 3,4,7 Compared with previous studies, the overall incidence of sorafenib-related HFSR (44.8%) fell within the common range of an Asia-Pacific population receiving standard treatment (10.7%-42%). 6,11 The three groups were recruited for a whole year, so cutaneous status influence with varied whether or daily events may cover. The majority of patients, who were elderly with dry or mild dry skin, were available during this time.
Ren et al. conducted a randomised controlled trial to validate the prophylactic efficacy between BSC with no creams and BSC with a 10% urea-based cream for sorafenib-induced HFRS in patients with advanced HCC. 17 The results suggested that the cream significantly prolonged the median time of incidence. The discrepancy with the results of the current study may relate to an inadequate treatment dose and the pharmaceutical factory from which the cream was obtained. One previous study recommended applying a 10% ureabased cream made in Germany three times a day after starting sorafenib on day 1. 17 The current study followed the recommendations of the pharmaceutical factory in Taiwan, which was to apply the 10% urea-based cream twice a day from 3 days prior to starting sorafenib. Thus, applying a nut-sized amount of 10% urea-based cream made by a Germany pharmaceutical factory three times a day may be more effective in preventing sorafenib-induced HFSK. Additionally, this study also supported that an inadequate treatment dose or non-Germany making of the 10% urea-based cream may have an efficacy similar to that of the moisturising cream.
A randomised controlled phase II trial revealed that a hydrocolloid dressing containing ceramide fixed on soles every 2-3 days was significantly more effective in treating sorafenib-induced grade 1 HFSR  not an effective substitute for the hydrocolloid dressing containing ceramide. Nevertheless, this study found that the 10% urea-based cream was more effective in maintaining cutaneous wetness during the first 6 weeks of starting sorafenib. Painful hyperkeratosis was a common consequence that resulted in the disruption of treatment whether in clinic quo or in adherence to the recommended guidelines. 14,15,23 Thus, the findings of this study indicate that a 10% ureabased cream may prevent hyperkeratosis within the first 6 weeks of starting sorafenib, after which a moisturising cream can be applied.
Although the prophylactic 10% urea-based cream eventually maintained cutaneous wetness in dry or mild dry skin it was confidence to prevent worse tendency of it during treatment. Besides, using it twice a day merely maintain patients' cutaneous wetness below 34%, which cannot significantly prevent dry skin. The dose also cannot reduce incidence of HFSR. The evidence possibly indicated an adequate dose to the 10% urea-based cream should be three times.

| LIMITATION
Questions still remain as to the most appropriate treatment dose and dermatological agent. Future research should validate the efficacy of an increased treatment dose of urea-based creams, such as comparing 10%, 12.5% or 20% as well as find a suitable substitute for hydrocolloid dressing containing ceramide, such as heparin cream that can be applied to the hands and feet.

| CONCLUSION
In conclusion, the application of nut-size amount of the 10% ureabased cream twice a day may preserve cutaneous wetness in the first 6 weeks after starting sorafenib, but it cannot reduce the occurrence of HFSR. Thus, the result supports nut-size dose of the 10% ureabased cream three times a day may be an appropriate dose to prevent HFSR.