AIM: To analyze the clinical and dosimetric predictive factors for radiation-induced esophageal injury in patients with non-small-cell lung cancer (NSCLC) during three-dimensional conformal radiotherapy (3D-CRT).

METHODS: We retrospectively analyzed 208 consecutive patients (146 men and 62 women) with NSCLC treated with 3D-CRT. The median age of the patients was 64 years (range 35-87 years). The clinical and treatment parameters including gender, age, performance status, sequential chemotherapy, concurrent chemotherapy, presence of carinal or subcarinal lymph nodes, pretreatment weight loss, mean dose to the entire esophagus, maximal point dose to the esophagus, and percentage of volume of esophagus receiving >55 Gy were studied. Clinical and dosimetric factors for radiation-induced acute and late grade 3-5 esophageal injury were analyzed according to Radiation Therapy Oncology Group (RTOG) criteria.

RESULTS: Twenty-five (12%) of the two hundred and eight patients developed acute or late grade 3-5 esophageal injury. Among them, nine patients had both acute and late grade 3-5 esophageal injury, two died of late esophageal perforation. Concurrent chemotherapy and maximal point dose to the esophagus ≥60 Gy were significantly associated with the risk of grade 3-5 esophageal injury. Fifty-four (26%) of the two hundred and eight patients received concurrent chemotherapy. Among them, 25 (46%) developed grade 3-5 esophageal injury (P = 0.0001<0.01). However, no grade 3-5 esophageal injury occurred in patients who received a maximal point dose to the esophagus <60 Gy (P = 0.0001<0.01).

CONCLUSION: Concurrent chemotherapy and the maximal esophageal point dose ≥60 Gy are significantly associated with the risk of grade 3-5 esophageal injury in patients with NSCLC treated with 3D-CRT.

Radiotherapy (RT) can achieve good results in the treatment of non-small-cell lung cancer (NSCLC). The key is how to monitor the dose to critical normal tissues such as the heart, normal lung, and esophagus. Several factors can minimize the dose of RT to the esophagus, including proximity of the esophagus to midline tumors or mediastinal adenopathy, proximity of the esophagus to higher priority organs such as spinal cord and heart and relative paucity of information on the clinical and dosimetric predictors of radiation-induced esophageal injury.

Clinicians familiar with RT planning are often forced to evaluate multiple plans that produce equivalent coverage of the target with variable sparing of critical structures. By our experience, the esophageal dose is not critically considered because the clinical and dosimetric variables predicting esophageal injury are not well characterized. The purpose of this investigation was to analyze the incidence of acute and late grade 3-5 esophageal injury in patients with NSCLC during three-dimensional conformal radiotherapy (3D-CRT) combined with or without chemotherapy (sequential or concurrent).

Twenty-five (12%) of the two hundred and eight patients developed acute (14) or late (11) grade 3-5 esophageal injury. Nine patients developed both acute and late grade 3-5 esophageal injury. Two patients died of perforation (grade 5 esophageal injury). The incidence of acute and late esophageal injury is summarized in Table 4.

Univariate analysis showed that concurrent chemotherapy, maximal point dose to the esophagus ≥60 Gy, subcarinal lymph nodes and mean esophageal dose ≥40 Gy were associated with the risk for grade 3-5 esophageal injury, whereas mean esophageal dose ≥40 Gy and subcarinal lymph nodes did not retain significance on multivariate analysis (Table 5).

Among the 208 patients, 54 (26%) received concurrent chemotherapy, 23 (92%) of the 25 patients who developed grade 3-5 esophageal injury received concurrent chemotherapy (P = 0.0001<0.01). Sequential chemotherapy was not associated with the grade 3-5 esophageal injury. Grade 3-5 esophageal injury did not occur in patients who received a maximal dose to the esophagus <60 Gy (P = 0.0001<0.01). Only two patients who did not receive concurrent chemotherapy developed grade 3-5 esophageal injury.

Damage to the esophagus is a major limiting factor in RT of intrathoracic tumors. Radiation-induced esophagitis presents various symptoms such as mild substernal burning sensation, severe pain, odynophagia and dysphagia. Acute change occurs, about 2 wk after initiation of therapy, and is largely limited to the mucosa. Most changes that occur later than 3 mo or more after irradiation are mortal disturbances possibly due to damage to nerves and smooth muscle, stricture may eventually develop[2]. It was reported that abnormalities include abnormal mortality with or without mucosal edema, stricture, ulceration, pseudodiverticulum and fistula, abnormal death occurs 4-12 wk following RT alone and as early as 1 wk after therapy when concurrent chemotherapy is given, strictures often develop 4-8 mo following completion of RT[3]. Radiation-induced esophageal injury occurs more frequently when RT and chemotherapy are combined. Frequent concurrent chemotherapy may complicate identification of the clinical and dosimetric predictors of esophageal injury. RT regimens that escalate the dose or add chemotherapy lay stress on defining the clinical and dosimetric predictors of acute and late esophageal injury. Intensificaiton of additional regimens is not feasible without better dosimetric methods to limit dose-induced esophageal injury. Additional characteristics of the predictors of esophagitis must define the dose limits for radiation and quantify the effect of chemotherapy.

The clinical and dosimetric predictors of acute and late esophageal injury are not typically characterized. Severe esophageal injury has been reported in 46% incidence of acute grade 3-4 esophageal injuries[4]. It was reported that concurrent chemotherapy increases nearly 12-fold risk for esophageal injury[5]. Emami et al[6], reported that RT may lead to late stricture and/or perforation of the esophagus. Dillman et al[7], reported a similar incidence of esophageal injury (<1%). It was reported that late esophageal injury following aggressive, high-dose conformal RT is common but rarely severe; dosimetric variables addressing the longitudinal and circumferential characteristics of the esophagus have biologic rationale and are predictive of late injury[8]. Patel et al[9], found that concurrent chemotherapy and hyperfractionated RT after surgery in patients with NSCLC are significantly associated with acute grade 2 or worse esophageal injury. Werner-Wasik et al[10], reported that concurrent chemotherapy and twice-daily RT are associated with higher grades and longer durations of acute esophagitis. Concurrent chemotherapy combined with RT appears to have lower esophageal radiation tolerance, resulting in markedly higher incidence of esophagitis[11].

We found a significant association between concurrent chemotherapy and the development of higher grades of esophagitis. Contrary to the findings of Maguire et al[8], we did not find any significant association between the incidence of esophagitis and the the longitudinal and circumferential characteristics of the esophagus. However, Kiura et al[12], reported that concurrent thoracic radiation therapy and chemotherapy are effective and tolerable in patients with advanced NSCLC. Singh et al[13], found that concurrent chemotherapy is significantly associated with the risk of grade 3-5 esophageal injury in patients with NSCLC. Advances in the delivery of RT and target delineation along with development of radioprotectors, intensity-modulated RT may allow the delivery of high doses to thoracic tumors while minimizing the dose to adjacent normal structures such as the esophagus. The introduction of fluoro-deoxy-glucose positron emission tomography scanning into traditional CT images for 3D treatment planning may improve target delineation, thus allowing reduced clinical target volumes with consequent reductions in the volume of irradiated esophagus. Marks et al[14], reported that, without a definitive threshold dose, a differential dose-volume-histogram (DVH) may be useful in predicting the outcome if the dose-response curve for regional organ dysfunction is known. Since the esophagus is a hollow tube, a dose-surface-histogram (DSH) might be a better descriptor than DVH. The relationship between the lumenal organ and the clinical/dosimetric factors is very difficult to investigate, thus both DVH and DSH are not ideal predictors of biologic outcome[8]. Yoshizawa et al[15], reported that carboplatin and 5-fluorouracil administered daily with concurrent thoracic radiation therapy in patients with NSCLC induce acceptable injury. Antonadou et al[16], found that amifostine, a radioprotector, reduces the incidence of pneumonitis and esophagitis without compromising the antitumor efficacy of radiation treatment in patients with lung cancer. In our study, concurrent chemotherapy and the maximal esophageal point dose were significantly associated with the risk of grade 3-5 esophageal injury in patients with NSCLC treated with high-dose 3D-CRT. In patients, who received concurrent chemotherapy, the maximal esophageal point dose for grade 3-5 esophageal injury was 60 Gy.

In conclusion, concurrent chemotherapy and the maximal esophageal point dose can predict grade 3-5 esophageal injury.