Abstract
Background/Aim: We assessed the efficacy and toxicity of concurrent chemoradiotherapy (CCRT) in patients with squamous cell carcinoma of the cervical esophagus. Patients and Methods: We retrospectively analyzed 37 patients treated with definitive CCRT. The patients received radiotherapy at a fraction dose of 2 Gy (total; 60 or 70 Gy) and concurrent chemotherapy. Adjuvant chemotherapy consisted of 1 to 2 cycles of 5-fluorouracil plus cisplatin or nedaplatin. Results: The median follow-up was 119.0 months, the 10-year overall survival, progression-free survival and laryngectomy-free survival rates were 35.6, 19.9 and 30.2% respectively. In the univariate analysis, T stage (T4 vs. T1-3) was the only prognostic factor for PFS. The most common acute toxicity was leukocytopenia (Grade 3; 27%). As for late toxicities, 4 patients (11%) developed Grade 2 or 3 esophageal strictures. Conclusion: The results of this study demonstrated that CCRT yielded satisfactory clinical outcomes with acceptable toxicities.
Radical surgical resection is the standard treatment for cervical esophageal carcinoma, similar to the case for esophageal carcinoma of other sites. However, in many cases of cervical esophageal carcinoma preservation of the larynx is difficult due to local extension of the disease, that is a potential cause of impairment of the quality of life (QOL) after the treatment. In regard to the radiotherapeutic approach, concurrent chemoradiotherapy (CCRT) has been shown to yield satisfactory clinical outcomes in cases of thoracic esophageal carcinoma, and has been well established as a curative treatment (1-4). In these settings, CCRT has also been applied for patients with cervical esophageal carcinoma, for the purpose of organ preservation. Although there were several publications on the CCRT for the disease, those reports had relatively short follow-up duration less than 5 years so far (5-12). In addition, neither the role of CCRT for the control of cervical esophageal carcinoma, nor the optimal fractionation schema for the RT dose or regimens for combined chemotherapy have been established yet. Here, we report the results of our retrospective review of the clinical results of definitive CCRT for squamous cell carcinoma of the cervical esophagus at our institution after a long follow-up for nearly 10 years, and analyze the prognostic factors potentially influencing the local control and survival rates of these patients after CCRT.
Patients and Methods
Patients. Data of a total of 37 consecutive patients with cervical esophageal carcinoma were reviewed in this study. The eligibility criteria for this study were as follows; patients with 1) pathologically proven squamous cell carcinoma, 2) who had received definitive CCRT without prior treatment between January 2002 and December 2009 at the National Cancer Center Hospital East (NCCHE), and 3) had loco-regionally confined disease without distant metastases. Patients with tumors showing invasion beyond the cervical esophagus were also included in this study, if the main tumor was located between the thoracic inlet and the cricopharyngeal muscle. Patients who were diagnosed as having clinical Stage IV disease, based on the presence of upper mediastinal lymph node metastases, were also included, because definitive RT was possible. Patients were staged according to the 2002 UICC-TNM Staging System. Initial staging was based on the results of history-taking and physical examination, blood tests, chest X-ray, upper gastrointestinal endoscopy, cervical computed tomography (CT) of the chest and upper abdomen, and barium contrast radiography performed where necessary. This study was approved by the institutional review board.
Radiation therapy. All patients received definitive RT at a fractional dose of 2 Gy daily 5 times a week, and the median overall treatment time for RT was 7 weeks (range, 6 to 9 weeks). From January 2002 to May 2004, the 2-dimensional (2D) simulator planning was performed with fluoroscopy, and initial RT was performed using opposed anterior and posterior beams. After that, the multiple beams technique was employed using a 3-dimensional (3D) treatment planning system (Xio; Elekta AB, Stockholm, Sweden). The prescription dose was 60 Gy for patents with resectable disease and 70 Gy for those with unresectable disease. In the former cases, elective nodal irradiation (ENI) at a dose of 40 Gy was delivered, which included the lymph node areas from the upper cervical region to the subcarinal region, while in the latter cases, ENI was not delivered. In 4 patients whose tumors could not be covered by 3D-conformal RT (3D RT), intensity-modulated radiation therapy (IMRT) was applied. The IMRT plan was optimized to achieve 95% coverage of the PTV by the prescribed dose. In the analysis, patients receiving IMRT were included in the 3D RT group.
Chemotherapy. All patients received CCRT. The chemotherapy consisted of 2 cycles of protracted infusions of 5-fluorouracil (5-FU) and cisplatin (CDDP), repeated once after an interval of 4 to 5 weeks. Figure 1 summarizes the chemotherapy regimens adopted in each of the treatment eras. After the CCRT, adjuvant chemotherapy was given to patients with adequate bone marrow function.
Evaluation and statistical analysis. The initial clinical response was evaluated within 3 months of completion of RT by upper gastrointestinal endoscopic examination and chest/abdominal CT. In this study, we used the complete response (CR) criteria proposed by Tahara (13), who defined CR of the primary tumor as (i) disappearance of the tumor lesion, (ii) absence of cancer cells in biopsy specimens, and (iii) disappearance of ulceration. If the clinical or endoscopic findings did not satisfy these criteria, the clinical response was classified as non-CR. In regard to the evaluation of lymph nodes metastases, CR was defined as complete disappearance of all assessable lesions. Furthermore, patients with persistence of metastatic lymph nodes measuring less than 1cm who showed no evidence of increase in the size of the nodes over a period of 3 months after the completion of RT were also classified as showing CR. Toxicities were graded using the Common Terminology Criteria (CTC), version 2.0. The overall survival (OS) and progression-free survival (PFS) rates were calculated from the start of RT. Laryngectomy-free survival (LFS) was defined as the interval from the start of RT to the date of laryngectomy or the date of death. The close-out date for the survival analysis was December 31, 2016. Data were analyzed using the SPSS version 22.0 (IBM Corporation, Armonk, NY, United States). Survival curves were estimated using the Kaplan-Meier product-limits method, and compared with the log-rank test. p-Values <0.05 were considered to indicate statistical significance.
Patient characteristics (n=37).
Results
Patient characteristics. The patient characteristics are shown in Table I. Five patients with M1 disease had upper mediastinal lymph node lesions. Table II summarize the RT methods. All patients, but one, received 2 cycles of concurrent chemotherapy during RT.
Clinical response and survival. Of the 37 patients, 12 patients (32.4%) showed CR at both the primary and lymph node sites. When the responses were stratified according to the T stage, 9 of 19 (47.3%) patients with T1-3 disease showed CR, while only 3 out of 18 (16.7%) patients with T4 disease showed CR; the difference was statistically significant (p=0.046). There were no significant differences in the CR rate depending on the duration of RT (≥56 days or <56 days), RT dose (>60 Gy or ≤60 Gy), RT method (2D-RT or 3D-RT), and status of lymph node metastasis.
Of the 12 patients who showed CR after the initial CCRT, the CR was maintained in 7 patients until the last follow-up. Regarding the failure or recurrence pattern after CCRT, 5 patients developed recurrent disease. The patterns of recurrence were divided into three groups: (i) local recurrence, (ii) lymph node recurrence, or (iii) distant metastases. Lymph node recurrence in the upper mediastinum, which represents spread outside the regional lymph node group (UICC staging system), was deemed as lymph node recurrence. The failure patterns were as follows: local recurrence in one patient, both local and lymph node recurrence in 3 patients, distant metastasis in one patient. None of the patients exhibited lymph node recurrence alone. Salvage or additional therapies after recurrence administered in the 5 patients were as follows: salvage surgery in one patient, endoscopic membrane resection (EMR)/endoscopic submucosal dissection (ESD) in one patient, and chemotherapy or best supportive care (BSC) in 3 patients.
The regimens of combined chemotherapy in each of the treatment eras included in the study period. The CRT basically consisted of 2 cycles of protracted infusions of 5-FU and CDDP or CDGP, and repeated once after an interval of 4 to 5 weeks during the period of the RT. (i) CDDP 40 mg/m2/day (days 1, 8 and 36, 43), 5-FU 400 mg/m2/day (days 1-5, 8-12, 36-40, and 43-47). (ii) CDDP 20 mg/m2/day (days 1-4 and 29-33), 5-FU 400 mg/m2/day (days 1-5 and 29-33). (iii) CDDP 70 mg/m2/day (days 1 and 29), 5-FU 700 mg/m2/day (days 1-4 and 29-32). (iv) CDGP 90 mg/m2/day (days 1 and 29), 5-FU 400 or 800 mg/m2/day (days 1-4 and 29-32). RT: Radiation treatment; 5-FU: 5-fluorouracil; CDDP: cisplatin; CDGP: nedaplatin.
Over a median follow-up period of 119.0 months, the 5-year OS, PFS and LFS rates of patients were 35.6%, 19.9% and 30.2% and the 10-year OS, PFS and LFS rates were 35.6%, 19.9% and 30.2%, respectively (Figure 2). In regard to the prognostic factors influencing the PFS, the univariate analysis identified T stage (T4 vs. T1-3) as the only significant prognostic factor (Hazard ratio 2.88, 95% confidence interval 1.31-6.37, p=0.0088) that might influence the PFS, while the RT duration (≥56 days or <56 days), RT dose (>60 Gy or ≤60 Gy), RT method (2D-RT or 3D-RT) and status of lymph node metastasis were not found to be significant (RT duration; p=0.72, RT dose; p=0.88, RT methods; p=0.43, status of lymph node metastasis; p=0.45).
Toxicity. Table III summarizes the incidence and severity of the acute and late toxicities. Grade 3 hematological toxicities were observed. In regard to the non-hematological toxicities, 7 patients developed Grade 3 mucositis or stomatitis. Four patients developed Grade 3 esophageal fistula during RT period or within 3 months after the completion of RT. One patient deceased within 30 days after the end of RT due to bleeding from the primary lesion. In regard to the late toxicities, esophageal stricture was observed after the CCRT, as shown in Table III. All patients who developed esophageal strictures underwent salvage EMR/ESD or photodynamic therapy (PDT) after the initial CCRT.
Patients' treatment (n=37).
Adverse events (n=37).
Salvage therapy. Of the 25 patients who were diagnosed as showing non-CR after the initial CCRT, 5 patients received salvage surgery (EMR/ESD in 3, PDT in one, and the remaining 16 patients salvage chemotherapy or BSC in 16 patients).
Discussion
Surgical resection has been thought to be the standard treatment for patients with cervical esophageal carcinoma. However, the treatment outcomes after surgery alone are not satisfactory, with the reported 5-year OS rate ranging from only 14-33% (14-18). In the era before the introduction of CCRT, RT alone or surgery after neoadjuvant RT was the treatment of choice for cervical esophageal carcinoma other than surgery alone, however, the clinical outcomes were not promising (19, 20). CCRT has been established as the non-surgical standard treatment for thoracic esophageal carcinoma, since several clinical trials have yielded excellent clinical outcomes (1-4). Based on the clinical outcomes of CCRT for thoracic esophageal carcinoma, CCRT has also been used widely for the treatment of cervical esophageal carcinoma. However, there have been only a few reports on clinical efficacy of CCRT for cervical esophageal carcinoma (5-8).
The 10-year OS, PFS and LFS rates of the patients. Over a median follow-up period of 119.0 months, the 10-year overall survival (OS) (a), progression-free survival (PFS) (b) and laryngectomy-free survival (LFS) (c) rates of the patients were 35.6%, 19.9%, 30.2%, respectively.
In this study, we reviewed the treatment outcomes of CCRT for squamous cell carcinoma of the cervical esophagus after a longer follow-up time of nearly 10 years and analyzed the prognostic factors according to the strategies for CCRT employed at our institution. As described above, the 10-year OS, PFS and LFS rates in this study were 35.6%, 19.9% and 30.2%, respectively. We considered that both the LFS rate and the OS rate in this study were satisfactory, and the results were quite consistent with the reported clinical results (7). However, there still appears to be room for improvement regarding the clinical outcomes: the CR rate and the PFS of the patients with T4 disease were significantly inferior to those of the patients with T1-3 disease. These findings indicated that the T stage might be an important prognostic factor influencing the responses to radiotherapy. In this study, about half the patients had T4 disease and most were inoperable, however, the outcomes in this study were not inferior to the reported clinical outcomes after surgical approaches (14-17). Taking the larynx preservation rate of the current study into consideration, we consider that CCRT could be offered as an alternative treatment option that provides satisfactory LFS without compromising the survival rate, even for patients with locally advanced disease. In regard to the late toxicities, only one patient developed grade 3 esophageal stricture. Development of esophageal stricture affects the swallowing function after treatment, leading to a decrease in the QOL of the patients, even if the disease is cured. Judging from the fact that all patients who developed esophageal strictures received EMR/ESD or PDT, definitive CCRT alone might not be responsible for the development of severe esophageal stricture.
A limitation of this report was the retrospective nature of the study, that caused variation of the regimens of combined chemotherapy adopted and the RT methods. Considering the low incidence of cervical esophageal carcinoma into account, some variations are inevitable.
In conclusion, the long-term clinical results of this study demonstrated the satisfactory clinical outcomes of CCRT in patients with squamous cell carcinoma of the cervical esophagus, with acceptable toxicity profiles. CCRT could be offered as a treatment option for the purpose of laryngeal preservation. The results also suggested that the T stage might be an important prognostic factor for the initial response and PFS after CCRT.
Acknowledgements
This study was supported, in part, by Grants-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan (JSPS KAKENHI grant numbers 15K19838 and 16K10412), Health Science Research grants from the Ministry of Health and Welfare, and the National Cancer Center Research and Development Fund (28-A-14).
- Received June 10, 2017.
- Revision received July 2, 2017.
- Accepted July 4, 2017.
- Copyright© 2017, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.