Acute radiodermatitis in cancer patients: incidence and severity estimates*

Objective: To estimate the incidence and distribution of the degree of radiodermatitis in cancer patients submitted to radiotherapy in the regions of head and neck, breast, and pelvis. Method: Prospective, longitudinal, quantitative study conducted from March 15, 2016 to May 4, 2017 with patients observed for five weeks of treatment. The primary outcome was the proportion of grade 1 or higher radiodermatitis each week. The degree of radiodermatitis and mean time of occurrence for each patient were secondary outcomes. Results: This study included 112 patients. The incidence of radiodermatitis was 100% among those whose head and neck region was irradiated, followed by 98% for breast and 48% for the pelvic region. The degree of radiodermatitis varied with the irradiated site; the worst degrees were found in the head and neck region, followed by the breast. In the three groups, the mean time of the first occurrence of radiodermatitis was approximately eleven days. Conclusion: The incidence of radiodermatitis in the studied groups was high, which reinforces the need to perform the same evaluation nationwide. This would support creating and standardizing protocols and recommendations for a proper management of radiodermatitis, specially concerning its prevention. DESCRIPTORS Radiodermatitis; Head and Neck Neoplasms; Breast Neoplasms; Pelvic Neoplasms; Oncology Nursing. Corresponding author: Paula Elaine Diniz dos Reis Laboratório Interdisciplinar de Oncologia, Faculdade de Ciências da Saúde, Universidade de Brasília Campus Darcy Ribeiro, Asa Norte CEP 70910-900 – Brasília, DF, Brazil pauladiniz@unb.br Received: 08/06/2019 Approved: 08/03/2020 Priscila de Souza Maggi Bontempo1 Marcia Aparecida Ciol2 Amanda Gomes de Menêses1 Giovana Paula Rezende Simino3 Elaine Barros Ferreira1 Paula Elaine Diniz dos Reis1 * Extracted from the dissertation: “Ocorrência de radiodermatite em pacientes com câncer em um hospital de ensino de Brasília”, Programa de Pós-Graduação em Enfermagem, Universidade de Brasília, 2017. 1 Universidade de Brasília, Faculdade de Ciências da Saúde, Programa de Pós-Graduação em Enfermagem, Brasília, DF, Brazil. 2 University of Washington, School of Medicine, Department of Rehabilitation Medicine, Seattle, WA, USA. 3 Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil.


INTRODUCTION
Radiodermatitis is a cutaneous reaction to ionizing radiation exposure which varies from mild erythema to severer reactions, such as dry or moist desquamation. This reaction is commonly reported by the international literature to afflict cancer patients who undergo irradiation of the regions of head and neck, breast, and pelvis, whose skin is part of the target irradiation area (1)(2) . However, the incidence of this reaction in Brazil is unknown.
The pathogenesis of radiodermatitis results from cell damage due to ionizing radiation exposure followed by an inflammatory response which affects skin layers (epidermis, dermis, and hypodermis or subcutaneous tissue) (2) . Such cell damage takes place direct or indirectly by the formation of reactive oxygen species and impacts essentially the capacity of cell division and multiplication (3)(4) . Actively proliferating cells are thus, similarly to the basal cells constituting the epidermis, more radiosensitive (2) .
The skin presents a balanced system of cell production and destruction in the epidermis (basal layer). This balance is broken after the first radiotherapy session, when keratinocyte destruction starts. With subsequent radiotherapy sessions, there is an accumulation of dose, increasing unbalance, which predisposes to changes to epidermis integrity and skin healing processes. Such changes surface with the appearance of erythema, xerosis, desquamation, itching, and hyperpigmentation (1,5) .
Erythema develops from a response to damage to the skin basal layer and as a consequence of an inflammatory process which initiates with histamine release and hypervascularity. It occurs around 10 to 14 days after radiotherapy (1,6) in patients receiving doses between 6 and 20 Gy and may be associated to symptoms such as discomfort, itching, and xerosis (6) .
After three to four weeks of treatment, as a compensatory response, mitotic activity is increased and stimulates the migration of new cells to substitute the damaged ones. The new cells are produced faster than the old ones are eliminated, originating dry desquamation, common in doses higher than 20 Gy. Continuous exposure to radiotherapy decreases the capacity of the basal layer to efficiently produce new cells to substitute damaged cells, which results in epidermal detachment, release of serous fluid and formation of wet blisters, covering a partial or confluent area of the treated region. This reaction is called moist desquamation and may surface after doses higher than 30 Gy (6) .
The development of radiodermatitis and its severity are related to extrinsic and intrinsic factors. Extrinsic factors include the treatment (total dose, volume of irradiated area, fractionation, and planning technique) and intrinsic factors refer to the patient (age, genetic antecedents, molecular subtypes, size and shape of the irradiated area) (7)(8)(9) .
Acute radiodermatitis in severer degrees may limit the therapeutic dose of radiation (10) , contribute to delayed completion or even interruption of radiotherapy, causing a negative impact on therapy success (1,(11)(12) . Also, radiodermatitis is responsible for aesthetic alterations, discomfort, pain, and reduced patient quality of life (12) .
The available literature presents a gap concerning the incidence of radiodermatitis in Brazil and, although the physiopathology of radiodermatitis is known, studies presenting the distribution of the degree of radiodermatitis throughout treatment for each irradiated area are also required. Obtaining this data may contribute to establishing common grounds and recommendations for managing radiodermatitis. This study's objective was thus verifying the incidence of radiodermatitis in cancer patients submitted to radiotherapy in the regions of head and neck, breast, and pelvis.

Study deSign
Prospective, longitudinal, quantitative study.

Selection criteria
The sample comprised all patients with cancer in the regions of head and neck, breast, or pelvis (prostate, cervix, and endometrium) submitted to radiotherapy from March 15, 2016 to May 4, 2017 who agreed to participate in the study. The patients were treated in a radiotherapy device of the type Linear Accelerator manufactured by VARIAN, model CLINAC CX, employing a three-dimensional conformal (3D -CRT) planning.
The inclusion criteria were patients aged 18 or older with an indication to start radiotherapy for treatment of head and neck, breast, or pelvis (prostate, cervix, and endometrium) cancer in any clinical stage. The exclusion criteria were: patients with an indication of urgent radiotherapy due to tumor complications (superior vena cava syndrome, medullary compression syndrome, intracranial hypertension syndrome, and hemorrhage), patients with tumor wounds in the treatment region which could interfere with the evaluation of radiodermatitis, and patients with a record of radiotherapy of the evaluated region.

data collection
The participants were recruited during the nursing consultation carried out in the first day of radiotherapy in the outpatient unit of UNACON/HUB. After receiving information about the necessary cautions during radiotherapy according to the routine of the nursing outpatient unit, the patients were invited to participate in the study. After receiving the necessary information about the research during the nursing consultation and agreeing to participate, the patients signed the Informed Consent Form (ICF).
General recommendations of care with irradiated skin area include cleaning the skin with mild soap, hydrating the skin with water-based creams, photoprotection, and avoiding skin www.scielo.br/reeusp Rev Esc Enferm USP · 2021;55:e03676 friction (13) . Also, for each degree of radiodermatitis, specific topical products were indicated, namely: chamomile gel (used in a prophylactic way when patients have intact skin and in case of erythema), chamomile tea compress in case of dry desquamation, and silver sulfadiazine for moist desquamation (13) . In this first consultation, patients' sociodemographic and clinical data were collected, in addition to a skin evaluation abiding by the radiodermatitis classification criteria of the Oncology Nursing Society (ONS) (14) . Also, a photograph of their skin was taken with a Nikon -Coolpix P510 digital camera, excluding patients who had pelvic tumors, since this is an intimate area. For such patients, evaluation was performed through visual inspection of the treatment area by the nurse. The classification proposed by ONS (14) was employed to measure the degree of radiodermatitis: (0) no change, (1.0) faint or dull erythema, (1.5) bright erythema, (2.0) dry desquamation with or without erythema, (2.5) small to moderate amount of moist desquamation, (3.0) confluent moist desquamation, and (3.5) ulceration, hemorrhage, or necrosis.
The sociodemographic and clinical characteristics collected were: age (in years), gender (feminine or masculine), self-reported color (white, brown or black), educational level (illiterate, primary, secondary, or superior), smoking status (smoker, former smoker or never smoked), drinking status (drinker, former drinker or never drank), presence of diabetes, daily sun exposure, Performance Status, tumor location, current treatment (chemotherapy, radiotherapy or chemoradiotherapy), type of energy, and total dose (Gy). Also, for participants with breast cancer, Body Mass Index (BMI) was collected and transformed into the following classification: Obese (BMI ≥ 30 kg/m 2 ), Overweight (BMI > 25 and < 29.9 Kg/m 2 ), Normal Weight (BMI >18.5 and < 24.9 Kg/ m 2 ) and Underweight (BMI < 18.5 Kg/m 2 ), as well as bra size (Brazilian sizes: 42, 44, 46, 48, 50). All these variables were registered in a self-created instrument.
The nurse researcher evaluated the participants every week during the treatment, following the ONS criteria (14) for assessing skin integrity, occurrence of radiodermatitis, and photographic register (except pelvis). The patients were followed up during the phase 1 of the treatment (five weeks) since some of them had to wait for further therapeutic planning to proceed to subsequent treatment phases and radiodermatitis assessment would not be continuous.

data treatment and analySiS
The sociodemographic and clinical characteristics were presented in a descriptive manner.
The primary outcome was the presence of radiodermatitis, defined as the emergence of grade 1 or higher radiodermatitis during the treatment weeks. Each week, the proportion of patients presenting radiodermatitis was calculated and the estimate and 95% confidence interval were calculated by Bayesian method (using Jeffreys prior).
Concerning the secondary outcomes, the distribution of the degree of radiodermatitis by week was presented in a descriptive manner and the means, medians, standard deviation and extreme values of the time in days until the first observation of radiodermatitis within each group (head and neck, breast, and pelvis) were calculated.
The analyses of mean and frequency were carried out using SPSS version 25 for Mac, and the confidence interval for proportions and graphs was calculated with the software RStudio.

ethical aSpectS
The research project was approved by the Research Ethics Committee of the Health Sciences Faculty of UnB in Opinion 610.425, dated 2013, following the recommendations of Resolution 466/12 by the National Health Council concerning research with human subjects. All participants have signed the Informed Consent Form.

RESULTS
Out of 163 evaluated patients, 50 were non-eligible (30 presented tumor wounds in the irradiated region and 20 had received urgent radiotherapy). Among eligible patients, one refused participation and 112 were included (31 with head and neck cancer, 50 with breast cancer, and 31 with cancer in the pelvis region). There was no loss to follow-up during the assessment period.
Clinical and sociodemographic characterization of participants is presented in Table 1. The mean age of patients with head and neck cancer was 58.8 years (standard deviation [SD] =14.5 years); 68% were male and 68% were white. The predominant tumor site was the oral cavity, identified in 13 (42%) patients. Regarding the current treatment, 21 (68%) were undergoing chemoradiotherapy and 10 (32%) were undergoing exclusive radiotherapy. The treatment was conducted in a mean of 25 sessions, with a total dose of 50 Gy; the type of energy used was photons.
The mean age of participants with breast cancer was 53 years (SD = 13.4). The most frequent histological type in the sample was Invasive Ductal Carcinoma (94%). Most participants were above the recommended weight: 40% were obese and 36% were overweight (Table 1). Among participants who preserved their breast or underwent breast reconstruction, 16 (32%) wore bras of a size between 42 and 46, 5 (10%) used size 48, and 5 (10%) used a size higher than 50. Fifteen patients (30%) were receiving exclusive radiotherapy and 35 (70%), radiotherapy concomitant to another therapy (trastuzumab or hormone therapy). Hypofractionated radiation, in which there is a reduction in the total number of fractions and increase in the administered daily dose, was conducted in 9 patients (18%). The mean total radiotherapy dose was 48 Gy and the most employed type of energy was photons.
The mean age of patients who received pelvic radiotherapy was 62.4 years (SD = 11). Most (74%) had prostate cancer, followed by cervix cancer (16%) and endometrium cancer (10%). Seventeen patients (55%) were submitted to radiotherapy concomitantly to hormone therapy, 7 (23%) to chemoradiotherapy and seven (23%) to exclusive radiotherapy. The mean total dose was 49 Gy (cervix), 50 Gy (endometrium), and 54 Gy (prostate).  (14) 3 (10) radiodermatitiS in patientS receiving radiotherapy in the region of head and neck Table 2 presents the results for the occurrence and degree of radiodermatitis throughout time for patients whose region of head and neck was irradiated. All participants presented some degree of radiodermatitis during the treatment, with a mean time for first event occurrence of eleven days (SD = 5.4; median =10 [min = 5, max = 30]). In the first week of treatment (W1), 35% presented erythema, which increased to 68% in the second week (W2) and 87% in the third week (W3), out of which 7% evolved to dry desquamation. From the fourth week onwards, the proportion of individuals with erythema decreased, but there was an increase in dry desquamation (33%) and evolution to moist desquamation of a subject on week 4 and two on week 5.

occurrence of radiodermatitiS in the breaSt region
Among participants irradiated in the breast region, 98% acquired radiodermatitis during the treatment, with a mean time for first occurrence of radiodermatitis of eleven days (SD = 3.7). Erythema was the predominant sign among patients; it was present in all five weeks of follow-up, especially in weeks 3 and 4 (Table 3). Dry desquamation occurred in Weeks 4 (W4) and 5 (W5), with 6% and 10%, respectively. Moist desquamation started in Week 3 (W3), with an increased frequency in subsequent weeks. No patient presented ulcer, hemorrhage, or necrosis during the treatment.

occurrence of radiodermatitiS in the pelvic region
Among patients with cervix cancer, four presented some degree of radiodermatitis, with a mean occurrence time of 7 days (SD = 4.8). Among patients with endometrium cancer, only one presented radiodermatitis during the treatment, with an occurrence time of 11 days; among patients with prostate cancer, twelve presented radiodermatitis during the treatment and the mean time of occurrence was 18 days (SD=13), which was late when compared to the other regions.
Among patients whose pelvic region was irradiated, compared to the regions of head and neck and breast, there was a smaller occurrence of radiodermatitis (48.4%). However, the occurrence of moist desquamation during all the weeks of evaluation (Table 4) in one or two individuals per week is emphasized.  Figure 1 presents the proportion of patients who developed some degree of radiodermatitis by week of treatment according to the radiated region.

Head and Neck
Breast Pelvis Proportion and 95% Confidence Interval Proportion and 95% Confidence Interval Proportion and 95% Confidence Interval

DISCUSSION
In this study, the incidence of radiodermatitis among patients assisted in the Radiotherapy Outpatient Clinic of UNACON/HUB over a 14-month period was estimated. This is the first known Brazilian study to estimate the incidence of radiodermatitis in cancer patients submitted to radiotherapy. A total of 112 patients were followed-up, out of whom 31 had head and neck cancer, 50 had breast cancer, and 31 had a cancer in the pelvic region (23 prostate, 5 cervix, and 3 endometrium).
The incidence of radiodermatitis ranged from 48% to 100% and the time for development of radiodermatitis ranged from 7 to 18 days. These results are comparable to those found in the literature, according to which 90 to 95% of individuals submitted to radiotherapy develop some degree of radiodermatitis (15)(16) within a mean 10 to 14 days from the start of the treatment (17) . In patients whose breast region was irradiated, the incidence was 98%, a figure similar to that found in the international literature, which varied from 95 to 100% (18)(19) .
Among patients whose pelvis was irradiated, the proportion of patients who developed some degree of radiodermatitis varied from 8% to 26%; the literature does not present any data for the incidence of radiodermatitis in this population. The smaller incidence in this group may be hypothesized to be due to the fact that the areas for irradiation in the pelvic region are far from one another (anterior area, right side and left side) when compared to the regions of breast and head and neck, in which the areas for irradiation may overlap. The skin of the pelvic region may be thus more preserved regarding the accumulation of dose in the tissue.
Smoking may be an aggravating factor for the development of radiodermatitis since it hinders the process of skin healing (12,20) . This study's sample included a high number of smokers among patients whose region of head and neck was irradiated (81% had been smokers and 6% still were). Among patients with breast cancer, four (8%) reported being smokers, among whom three presented moist desquamation during the treatment.
Breast size is also related to the development of radiodermatitis (6,21) . Cutaneous folds may be causes of friction, resulting in lesions; this is common in obese people. This may also occur in patients with voluminous breasts, particularly in the inframammary region (22) . In this study, patients with voluminous breasts developed more erythema when compared to those with smaller breasts. Also, the two patients with the most voluminous breasts of the sample (bra size 54) had BMI higher than 30 Kg/m 2 (obese) and developed moist desquamation in the axillary and inframammary regions.
The time of occurrence and grade of radiodermatitis may be delayed if the care protocol recommended by nurses of the service is followed (23)(24) . The recommendations provided by the literature for skin care during radiotherapy are simple and involve skin hydration through topical care and water intake, wearing appropriate clothing, and sun protection (13,25) . This study's patients received information on care of the irradiated region and it is possible that some have thus delayed the time of development of radiodermatitis.
It is extremely important for Brazilian radiotherapy services to quantify and characterize radiodermatitis developed in cancer patients submitted to radiotherapy in a more systematic way. The obtention of national data may favor the creation of assistance and public policies, considering that this radiotoxicity constitutes a limiting factor in treatment continuation, possibly interfering with disease prognosis. Also, interrupting patient treatment due to radiotoxicity leads to postponing the start of radiotherapy for other patients, leading to a negative economic impact for the service.
This study's limitations include the fact that only one nurse performed skin assessment, even though the scale is well-defined concerning the objective criteria related to the signs. However, since the same person assessed the skin of all participants, the results were not biased by differences between evaluators.
The sample sizes of each group are not large, which could be considered a limitation. However, since this is an estimation study, sample size is reflected in the confidence interval size of each estimate. For a higher precision (i.e., smaller confidence intervals), bigger samples would be necessary, which could be calculated departing from this study. Despite this limitation, the results can be generalized to cancer patient populations in Brazil with the same inclusion and exclusion criteria, similar socioeconomic characteristics, and receiving radiation in circumstances which are like those of patients in this sample.

CONCLUSION
In this study, the incidence of radiodermatitis was 100% among patients whose region of head and neck was irradiated, 98% among those whose breast region was irradiated, and 48% among those whose pelvis (cervix, endometrium, and prostate) was irradiated.
The need for identifying the incidence of this radiotoxicity nationwide is emphasized, since it is necessary to create and standardize protocols and recommendations for proper management of radiodermatitis, specially concerning its prevention.