A technique to reduce skin toxicity in radiotherapy treatment planning for esophageal cancer

Abstract Purpose To demonstrate a specific skin dose limiting technique in radiotherapy treatment planning for esophageal cancer and carry out a comparative analysis combining with clinical cases. Material and methods Thirty patients with cervical and upper thoracic esophageal carcinoma previously treated in our institution were selected. A treatment plan had been finished previously according to the planning parameters directives from physician and delivered for each patient. In this study, we copied the previously delivered plans in radiotherapy treatment planning system and converted a low dose level (usually 5Gy) to a skin dose limiting structure (SDLS), then we set the objective functions of the SDLS in the Pinnacle Inverse Planning module and re‐optimize the plans to reduce the skin doses. Finally, we compared the dose distribution and other parameters of target volume and organs at risk (OARs) between the old plans and the new plans. Results There was no significant difference in most of OARs sparing. However, for all plans, the maximum dose to the SDLS decreased from 6145.90 ± 416.96 cGy to 5562.09 ± 616.69 cGy with maximum difference of 1361.30 cGy (P < 0.05), the percentage volume of 40Gy received by the SDLS decreased from (10.20 ± 6.36)% to (5.46 ± 4084)% with maximum difference of 9.89% (P < 0.05). For the target volume, there was no significant difference in the average dose and maximum dose, the approximate minimum dose to the target volume decreased from 5711.28 ± 164.61 cGy to 5584.93 ± 157.70 cGy (P < 0.05), the conformal index and homogeneity index of the target volume were hardly changed. Conclusion In radiotherapy treatment planning for esophageal cancer patients, the skin dose can be significantly reduced using the skin dose limiting technique, and the impact on the dose to target volume and OARs is little, this technique can be used in most radiotherapy treatment planning.


| INTRODUCTION
Esophageal cancer is one of the most common malignant tumors in the digestive tract. According to the latest global cancer report released by the World Health Organization in 2018, the incidence of esophageal cancer ranks seventh among all cancers (3.2 percent of new cancers in the world) and the mortality rate ranks sixth among all cancer deaths (5.3 percent of total cancer deaths). 1 The main treatments for esophageal cancer include surgery, radiotherapy, and chemotherapy, and most of the esophageal cancer patients need radiotherapy throughout the course of the disease. With the emergence and fast development of intensity modulated radiotherapy, the planning and delivery of radiation techniques have been greatly improved. We now can get higher prescription dose and better dose conformity to the target volume, and the 5-year survival rate of patients with esophageal cancer has been greatly improved. [2][3][4][5][6] However, because the anatomic position of the target volume is close to the skin (especially for the cervical and upper thoracic esophageal carcinoma), skin toxicity is inevitable during the process of radiotherapy, the skin injury can negatively affect the quality of life of the patients. [7][8][9] This skin reaction usually begins with the dose of 20-25 Gy, and radiation dermatitis occurs significantly after the cumulative dose to the skin reaches 40 Gy in the middle and late stages of radiotherapy. 10 The mild symptoms include local erythema, dryness, and desquamation, and the severe symptoms will be local skin pain, edema and exudates, moist desquamation and so on. [11][12][13][14][15] Radiationinduced skin reactions occur as a result of damage to the basal cell layer of the skin and resulting in an imbalance between the normal production of cells in this layer and the destruction of cells at the skin surface. [16][17][18] Although skin toxicity is inevitable in the process of radiotherapy, the dose to skin can be reduced as much as possible through ideal treatment planning, so as to reduce the degree of skin injury during radiotherapy. Within the last few years, multi-criteria optimization, knowledge-based planning approach (including model based planning, atlas-based planning, dose-volume histogram guidance planning and so on) have been used in Auto Planning, which is expected to improve the efficiency and quality of radiotherapy treatment planning. [19][20][21][22][23] Although significant progress has been made in this area, much work is still needed to explore practical issues

2.B | Treatment planning and utilization of skin dose limiting technique
After completing the contouring of target volume and OARs, a planning directive was completed. The planning directive outlined the physician's planning guidelines including target prescriptions (60Gy/ 2Gy/30Fx, V 60Gy ≥ 95%, D max < 66 Gy), normal structure goals (shown in Table 1), and other plan parameters. In clinical practice, we require that the dose should not be higher than 66 Gy. If it is inevitable, the volume of the dose above 66 Gy should not exceed 5% of the volume of PTV, and also should not be in the esophagus and trachea. Then the dosimetrists would complete a practicable treatment plan in accordance with the planning directives.
The patients were planned with five fixed fields (or with seven fixed fields for those whose planning directives were difficult to meet), the gantry angle of fields were set to be 200°, 330°, 0°, 40°, and 160°for five fields or 200°, 260°, 310°, 0°, 50°, 100°, and 160°for seven fields. The photon energy was 6 MV, the machine was Elekta Precise. In the previously finished and delivered plans, the optimization objectives related to the skin dose include some manually created rings around the PTV to compress the isodose curves, but it made little contribution to reducing the skin dose. In this study, the already finished plans of the 30 patients were copied, and an isodose level of 5 Gy was generated and converted into a structure for each plan. The structure "Outline" (external contour of OARs, Organs at risks. the patient) was then contracted 0.5cm to generate a structure "Outline-0.5," and then the structure generated by the 5 Gy isodose level subtracts the "Outline-0.5" and any overlapping parts with the structure PTV, a skin dose limiting structure (SDLS) with a thickness of 0.5cm just inside the external contour of the patient was created ( Fig. 1). Sometimes the SDLS can be manually modified in order to make it more practical.
After the SDLS had been created, it was added to the optimization objectives of the newly copied plan and the objective functions were set. In this study, the objective functions of SDLS were set as follows: D max < 50 Gy, weight 20, V 40Gy < 5%, weight 30. After the new objective functions were set, the newly copied plan was re-optimized and calculated to get a new dose distribution. In order to obtain a more ideal skin dose distribution, the optimization functions of the SDLS can be adjusted properly in different plans before re-optimization on the premise that the dose distribution of target volume is not adversely impacted and the dose to OARs is not increased significantly.

2.C | Plan evaluation
After the full optimization and calculation of the newly copied plan had been completed, we reviewed the two plans (the old plan and the new plan) side by side in the Pinnacle Plan Evaluation module, paying close attention to the changes of dose to target volumes, OARs and normal tissues (Fig. 2). In order to get a quantitative ana- and homogeneity index (HI ¼ ðD 2% À D 98% Þ = D 50% , where D 50% is the median absorbed dose of the PTV, D 2% , and D 98% represent the dose received by 2% and 98% of the volume of PTV) of the PTVs of the two plans ( Figure 3 shows the DVHs of both trials for the best and worst cases). 24

2.D | Statistical method
Statistical analyses were performed using software package SPSS paired sample t-Test was used to assess the differences between the two plans. P values < 0.05 were considered significant.   The results of this study were obtained by comparing two treatment plans of 30 patients, we can find a significant skin dose reduction through the use of the skin dose limiting technique demonstrated in this study, this can help us get an ideal skin protection for the patients in the process of radiotherapy, but there was a lack of clinical trial data. Previously, we did not use any techniques specifically aimed at reducing skin dose in treatment planning process. It was only after several times of treatment that the patient experienced severe skin reactions before we revised the treatment plan to obtain a lower skin dose. We will apply the skin dose limiting technique to future clinical work, and observe the symptoms of skin reaction of each patient during the process of radiotherapy.

| CONCLUSION
The purpose of this study was to demonstrate the process of a skin dose limiting technique in radiotherapy treatment planning for esophageal cancer patients. As of now, we have not been able to accurately predict the severity of radiation skin reactions a patient is

ACKNOWLEDGMENTS
Not applicable in this section.

CONF LICT OF I NTEREST
The authors declare no conflict of interests.

AUTHORS' CONTRI BUTIONS
All authors participated in patient treatment planning and plan evaluation, and all authors were involved in the preparation of the manuscript. All authors reviewed and approved the final manuscript.

ETHICS APPROVAL AN D CONSENT TO PARTICIPATE
Not applicable in this section.

CONS ENT F OR PU BLICATI ON
Not applicable in this section.

DATA AVAILABILITY STATEMENT
The datasets supporting the conclusions of this article are available from the corresponding author on reasonable request.