Dosimetric and Clinical Predictors of Acute Esophagitis in Lung Cancer Patients in Turkey Treated with Radiotherapy

Radiotherapy (RT) is used in the treatment of 64% of cases with lung cancer (with a curative intent in 46%) (Tyldesley et al., 2001). Tumors close to the midline and mediastinal lymphadenopathies increase the dose of exposure of the esophagus. For this reason, acute esophagitis (AE) is a frequent adverse effect seen in patients receiving RT for lung cancer. Grade 3-4 AE has been reported in 20-30% in simultaneous chemoradiotherapy (CRT) administrations (De Ruysscher et al., 2012). Parameters predicting the AE in advance would prevent interruptions of treatment and would increase the success of treatment (local control and overall survival). Administration of CRT, hyperfractioned RT, esophageal volume receiving a dose (%) more than 35 Gy (V35), V45, V50, V60, and mean esophageal dose (Dmean), and environmental and volumetric esophageal doses have been faulted for AE development in the literature (Maguire et al., 1999; Werner Wasik et al., 2000; Bradley et al., 2004; Patel et al., 2004; Rose et al., 2009). However, there is no


Introduction
Radiotherapy (RT) is used in the treatment of 64% of cases with lung cancer (with a curative intent in 46%) (Tyldesley et al., 2001). Tumors close to the midline and mediastinal lymphadenopathies increase the dose of exposure of the esophagus. For this reason, acute esophagitis (AE) is a frequent adverse effect seen in patients receiving RT for lung cancer. Grade 3-4 AE has been reported in 20-30% in simultaneous chemoradiotherapy (CRT) administrations (De Ruysscher et al., 2012). Parameters predicting the AE in advance would prevent interruptions of treatment and would increase the success of treatment (local control and overall survival).
Administration of CRT, hyperfractioned RT, esophageal volume receiving a dose (%) more than 35 Gy (V 35 ), V 45 , V 50 , V 60 , and mean esophageal dose (D mean ), and environmental and volumetric esophageal doses have been faulted for AE development in the literature (Maguire et al., 1999;Werner Wasik et al., 2000;Bradley et al., 2004; RESEARCH ARTICLE

Dosimetric and Clinical Predictors of Acute Esophagitis in Lung Cancer Patients in Turkey Treated with Radiotherapy
Durmus Etiz 1 *, Evrim Bayman 1 , Melek Akcay 1 , Bilgehan Sahin 1 , Cengiz Bal 2 generally accepted predictive factor for AE development. The aim of this study is to evaluate the association between the dose distribution, clinical factors, and AE development in the esophagus during the treatment of lung cancer.

Patient characteristics
One hundred and four patients who were planned to receive three-dimensional conformal RT in the Clinic of Radiation Oncology at Osmangazi University School of Medicine between July 2010 and January 2012 were evaluated. Cases with the histopathological diagnosis of non-small-cell lung carcinoma (NSCLC) or small-cell lung carcinoma (SCLC), with T3 or T4 tumors and/or mediastinal lymph node involvement, not operated on for medical reasons or according to the choice of the patient, and with a Karnofsky Performance Scale of 70 and above were included in the study. Cases with multiple primary diagnoses and/or distant metastasis were not included in the study. Cases that developed metastasis during the treatment were excluded from the study. Ninety-seven 4224 of the cases of this study group were males (93.3%) and 7 were females (6.7%). NSCLC/SCLC rate in the histopathological evaluation was 93/11. The median age was 61.5 years (39-79 years). The NSCLC group comprised of 50 cases with stage III B (48%), 27 cases with stage III A (26%), 18 cases with N0 (17.3%). The characteristics of the patients are presented in Table 1. Median tumor volume was 80 cm 3 (6-820 cm 3 ) and median lymph node volume was 2 cm 3 (0-46 cm 3 ).
Intravenous hydration was provided with 1-2 L of infusion fluid given over 8-12 hours. The balance of adequate hydration and urinary output were followed during the 24 hours following the treatment. Patients receiving simultaneous chemotherapy or patients with a decreased oral intake were hydrated, given enteral nutritional support and analgesics with hospitalization.

Radiotherapy
Three-dimensional conformal radiotherapy was planned for all patients. Tomographic planning sections were obtained in 5 mm intervals in the supine position with arms immobilized over the head with a T-bar. Gross tumor volume (primary tumor and lymph nodes with short diameter greater than 1 cm) was contoured according to ICRU-62 using CT and PET-CT (ICRU Report 62, 1999). Clinical target volume (CTV) was calculated by adding 6 mm in squamous cell carcinoma, 8 mm in adenocarcinoma, 3 mm in the presence of pathological lymph nodes, and 8 mm in SCLC. Elective mediastinal nodal irradiation was only used in SCLC. PTV was obtained by adding 1-1.5 cm to the CTV. PTV received a minimum of 93% and a maximum of 107% of the planned dose. Heterogeneity correction was performed. The esophagus was contoured from cricoid cartilage to the gastroesophageal intersection in order to include its external wall. Dose volume histograms were evaluated in all patients. Critical organ dose tolerances were identified as V 5 <50% (lung volume receiving more than 5 Gy was less than 50%), V 20 <30% (lung volume receiving more than 20 Gy was less than 30%), mean lung dose <15 Gy, and maximum spinal cord dose <50 Gy and cardiac V 30 <50% (cardiac volume receiving 30 Gy was less than 50%). Irradiation was performed with linear accelerator equipment (Precise-ELEKTA™) with 6 MV X-rays. Oblique fields following the anterior-posterior fields excluding the spinal cord outside the field were used in all patients. Fields were controlled by electronic portal imaging in the first treatment and followed at weekly intervals. The median dose of RT was 60 Gy (45-62 Gy). Fractions were applied as 1.8-2 Gy/day, 5 days a week. Mediastinal RT was administered in 86 cases (82.6%). Esophageal-tumor and lymph node volumes, and maximum-minimum-mean esophageal doses were recorded in dose volume histogram evaluations (XiO; Computerized Medical Systems™, St. Louis, MO).

Toxicity scoring and follow-up
Age and gender of the patients, tumor pathology, type of treatment (only RT/concurrent CRT), number of chemotherapy cycles before RT, localization of the tumor, T and N stage, severity of AE according to RTOG criteria. RTOG scoring system is presented in Table 2. (At the beginning, and the first, second, third, fourth, fifth, and sixth week, and following the first week, and first and second month). Days of interruption of RT, total RT dose, and tumor response results in the second month were recorded in the "RT reporting and oncological follow-up  All patients received sucralphate and nystatin suspension (455 ml/day) starting with the RT. Follow-up oncology visits were planned for every three months in the first two years, every six months between two and five years, and annually after five years.

Statistical analysis
Parameters that may be associated with AE development, such as age (Spearman correlation coefficient) and gender (Fisher's exact test) of the patients, simultaneous chemotherapy administration (Fisher's exact test), number of chemotherapy cycles before RT (Fisher's exact test), total RT dose (Fisher's exact test), tumor response to RT (Spearman correlation coefficient), pathological type of the tumor (Fisher's exact test), tumor localization (Fisher's exact test), T stage (Spearman correlation coefficient), N stage (Spearman correlation coefficient), tumor volume and esophageal volume (Spearman correlation coefficient), interruption of RT (Fisher's exact test), weight loss during RT (Spearman correlation coefficient) and administered doses and volumes (Spearman correlation coefficient, logistic regression analysis) were evaluated.
Statistical analysis was performed using the SPSS (Statistical Package for Social Science -Version 20.0-for Windows) program. A value of p<0.05 was accepted as statistically significant.

Results
Patients' visits accounted for total of 1040 for 104 patients. The mean number of days of interruption was two days (range 0-10 days), and a median weight of
AE was detected in 35 of 52 patients (67.3%) with ages older than 61.5 years (median) and in 20 out of 52 patients (38.5%) with ages less than 61.5 years in the fourth week, which is the most frequent time of occurrence of mucositis (r=-0.36, p=0.311). AE was seen in 28 of 55 cases (51%) with only RT administration, while 27 of 49 cases (55%) with CRT had AE (p=0.090).
Beginning in the 3 th week, 4 th week, 5 th week and 6 th week of RT, D max values greater than 5117 cGy, D mean . values greater than 1487, and V 10-60 was found to be correlated with the development of AE (Table 5). A logistic regression analysis demonstrated that D max . (p=0.002) and V 30 >31% (p=0.008) were significant risk factors effecting AE development (Table 6).

Discussion
The esophagus is exposed to high dose radiation in situations of centrally localized tumors and/or the presence of mediastinal lymph nodes. It is desirable that the esophagus receive lower doses because of its proximity to the spinal cord and heart. There is a relative lack of knowledge about clinical and dosimetric markers effecting the development of AE during RT in the literature. Increased rates of different concurrent chemotherapy protocols further complicate the evaluation of toxic doses for the esophagus. In addition, the efficiency of treatment and adverse effects may vary according to different societies (Lara et al., 2010;Soo et al., 2012). Emami et al. (1991) reported the TD 5/5 rates (5% stricture and perforation rate in 5 years) for the 1/3, 2/3, and whole esophagus as 60 Gy, 55 Gy, and 50 Gy, respectively. However, these findings are from the pre-three-dimensional RT planning and simultaneous chemotherapy applications period.
Various dosimetric and clinical factors have been reported to affect AE (Table 7) (Maguire et al., 1999;Werner Wasik et al., 2000;Singh et al., 2003;Bradley et al., 2004;Ahn et al., 2005;Court et al., 2012). Bradley et al. (2004) administered cisplatin, etoposide, gemcitabine, paclitaxel, and carboplatin-based CRT in their study performed in 166 cases with stages I-III NSCLC (25% patients). They identified that simultaneous chemotherapy doubled the risk of AE (p=0.001). Others reported this rate as a 12-fold increase (Werner Wasik et al., 1999;Qiao et al., 2005;Caglar et al., 2010). In a series of 105 lung cancers, grade 3 AE was not seen in patients who received only RT; while the rate of grade 3 AE was reported to be 18% in the CRT group (p=0.001) (Werner Wasik et al., 2000). Nevertheless, there are studies reporting no effects of CRT on the development of AE. In a series of 254 cases by Ahn et al. (2005) no differences were found in the rate of development of AE between the etoposide, gemcitabine, paclitaxel, carboplatin, and cisplatin-based CRT arm (32 cases) and no CRT arm (p=0.30). Takeda, confirmed the fact that CRT administration alone is not associated with the development of AE in his series of 35 cases (Takeda et al., 2005). In the present study, CRT was not a factor increasing the rate of AE, either. All patients receiving CRT were hospitalized and followed-up by daily patient visits, preventive drugs (sucralphate and nystatin) were administered under the control of the nurses, IV hydration was performed before chemotherapy, and enteral nutritional support was applied when necessary. These factors together may have prevented the progression of the symptoms of AE in this group. In addition, total RT dose (median 64 Gy) is higher than the present study (median 60 Gy) in the studies reporting an enhancing    DOI:http://dx.doi.org/10.7314/APJCP.2013.14.7.4223 Dosimetric and Clinical Predictors of Acute Esophagitis Lung Cancer Patients in Turkey Treated with Radiotherapy role of CRT on the rate of development of AE (Table 7) (Werner Wasik et al., 2000;Bradley et al., 2004;Caglar et al., 2010).
The clinical symptoms of AE in the present study were followed-up on. The answer to whether it is clear that there is no organ injury when the individual has no complaint of AE is found in a study of 82 lung cancer patients from Japan. Fields of grade 3 ulcers were seen in the endoscopic evaluation of cases with no AE symptoms or with mild symptoms (RTOG grade 0-1) who received CRT. On the contrary, grade 3 AE symptoms were reported in only 8.5% of the cases who had endoscopic grade 3 ulcers and thus, it was stressed that the esophagus, contradictory to the previous knowledge, was a "quiet" organ in terms of symptoms (Hirota et al., 2001). In our study, the presence of clinical grade 1-2 symptoms do not imply that the real damage is limited to this level. In addition, not only severe AE, but also intermediate and mild AE could be the reason of the late adverse effects according to Dorr et al. this feature (superficial barrier function against mechanical and/or chemical stresses) also is valid for the intestines, oral mucosa, and urinary bladder (Dorr et al., 2001). Treatment of mild mucositis would prevent possible late adverse effects (Rodriguez et al., 2009). For this reason, such cases should be followed-up on for late adverse effects even if they are asymptomatic in terms of AE.
A comparison of the parameters predicting the development of AE reported in the literature is difficult in many aspects. Variables such as different scoring systems for esophagitis (RTOG and Common Terminology Criteria for Adverse Events CTCAE), different techniques of RT (three-dimensional RT, intensity modulated RT, X-ray, proton treatment), different dose and fraction schemes (conventional and hyperfractionation), different dosevolume histogram parameters (esophageal volume or esophageal surface area), different esophageal contouring protocols (whole organ or external esophageal contour only), differences in the numbers of schemes and cycles of chemotherapy applied simultaneously or before the RT, radioprotector use (amifostine) and different ethnic origins of the patients [Western society (America and Europe) and Far Eastern society (China and Japan)] effect the results.
Another reason for different results originates from the anatomical location of the esophagus. A conversion to the oblique fields of the spinal cord prevention after the first field of lung RT is completed occurs around the fourth week in which AE is maximal. Due to the inclusion of a part of the esophagus immediately in front of the spinal cord in the field, and the exclusion of part of it results in dose-volume histograms with different dosimetric features in this period. In addition, the fractioned dose of the esophagus changes in the oblique boost period since it remains in the partial field, while it receives a 1.8-2 Gy daily dose during the first part of RT. This makes the unit of mean (D mean ) esophageal dose meaningless since it has the same continued fraction and weekly same dose accumulation (Huang et al., 2012). No association was identified in the present study between the D mean esophageal dose and AE in the regression analysis.
Some other factors that trigger AE are present, apart from RT and chemotherapy. These are esophageal infections (candida and herpes simplex esophagitis), the presence of gastroesophageal reflux disease (reflux complaints are confused with AE and are reported as a complaint of higher grade) and exposure of the stomach to RT (accompanying gastritis symptoms due to the inclusion of the stomach in the field in lower lobe tumors) (Werner Wasik et al., 2010). These additional factors render the identification of the association between RT and AE more difficult. Considering the serial structure of the spinal cord, and esophagus, it is expected to demonstrate findings of toxicity in exceeding doses of the maximal target value, such as the spinal cord.
In the present study, a maximal esophageal dose of greater than 51 Gy is found to increase the frequency of AE. This value is reported as 55 Gy by Singh et al. (2003) in their series of 207 cases from a Western society and 60 Gy by Qiao et al. (2005) in their series of 208 cases comprising a Far Eastern society.
In conclusions, maximal esophageal doses higher than 51 Gy and/or esophageal volumes approximately more than 1/3 receiving a 30 Gy dose are found to be risk factors for AE development in a series of 104 cases with lung cancer receiving RT administration.