Efficacy of Preoperative Chemoradiation Therapy for cT 4 Esophageal Cancer : a Retrospective Study

*1)Departments of Radiation Oncology, Juntendo University Graduate School of Medicine, Tokyo, Japan, *2)Department of Radiology, Tokyo Rinkai Hospital, Tokyo, Japan, *3)Departments of Esophageal and Gastroenterological Surgery, Juntendo University Graduate School of Medicine, Tokyo, Japan, *4)Department of Data Science, Natonal Center for Global Health and Medicine Centr for Cininal Science, Tokyo, Japan, *5)Departments of Radiology, Nagoya City University Graduate School of Medical Sciences and Medical School, Aichi, Japan


Introduction
Esophageal cancer is a leading cause of cancerrelated death in Japan.The Cancer Statistics in Japan estimates that 21,965 individuals developed esophageal cancer in 2012, and 11, 576 died of the disease in 2014 (http://ganjoho.jp/reg_stat/statistics/stat/summary.html).Tumor resection is the best treatment option to achieve local control of the disease.However, many patients are diagnosed at advanced stages and have a poor prognosis.The 5-year overall survival (OS) rate of patients with this cancer is approximately 50% but is much worse in patients with locally advanced disease; the -survival rates for pathological T3 (pT3) and pT4 disease are 39% and 13%, respectively 1) .
Because local recurrence and distant metastases are the main issues after surgery 2) , preoperative chemotherapy with cisplatin (CDDP) plus 5-fluorouracil is the standard treatment for patients with stage II/III squamous cell carcinoma, based on the definitive results of the Japan Clinical Oncology Group (JCOG) 9907 trial in Japan 3) .In this trial, patients who received preoperative chemotherapy had a significantly higher 5-year OS rate (55%) than patients who received postoperative chemotherapy (43%).Recent meta-analyses also reported a benefit for patients who received preoperative chemoradiotherapy (CRT) compared to patients who did not 4)-6) .Subgroup analyses of the JCOG 9907 trial revealed that preoperative CRT was less effective for stage III or T3 lesions than limited lesions 7) .For patients with T1-3 disease, preoperative chemoradiotherapy improved survival among patients with potentially curable esophageal or esophagogastric-junction cancer 8) .Therefore, more intensive preoperative treatments, including CRT, have been introduced to treat these advanced cases.CRT is reportedly more effective than chemotherapy alone in terms of survival benefit 9) .
For patients with T4 disease, the standard treatment is definitive CRT or palliation 10) .Surgery is performed only for resectable tumors following preoperative or definitive CRT.In the present study, the effects of preoperative CRT on survival were evaluated in patients with advanced T3 or T4 disease from 1998 to 2008.We have reported the results of the treatment of patients with T3 disease elsewhere 11) .In that report, patients with lymph node metastasis exhibited a better prognosis in the DOC group than those in the FP group.Preoperative CRT for locally advanced esophageal cancer using DOC results in similar or better long-term outcomes compared with FP-based CRT.In this retrospective analysis, we assessed the efficacy of this treatment in patients with T4 stage disease.

Patients and Methods
This study was approved by the ethics board of our institution.Written informed consent from each individual patient was not required because of the retrospective nature of the study.

Patients
From November 1998 to November 2008, 57 patients with cT4 esophageal (any N) without distant metastases underwent preoperative CRT.All patients were reclassified according to the International Union Against Cancer (UICC) TNM classification, seventh edition 12) .The patient characteristics are presented in Table -  raphy (CT) scan. 18F-Fluorodeoxyglucose positron emission scanning and bronchoscopy were optional.Clinical stage T4 was defined according to Hironaka, et al. 13) as follows: 1) a tumor that extends into the lumen or causes deformity of the tracheobronchial tree or 2) a tumor that is attached to organs at a contact angle of ≥90°in the thoracic aorta, as observed on a CT scan.
The last follow-up was performed in November 2014.Complete follow-up was obtained for 44 patients.Among them, 6 were alive with no evidence of disease, 26 died of disease or treatmentrelated diseases.Twelve died of non-related disease (median: 44 months, range: 8-94 months).The median follow-up period for the six survivors was 73 months.Thirteen were lost for follow-up: 9 with diseases were treated as died of disease, and 4 without disease were censored at the last visit day in the follow-up analyses.

Chemotherapy regimen
In this study, 11 patients received 5-fluorouracil and cisplatin (FP group), and 46 were treated with docetaxel (DOC).The details of chemotherapy have been described elsewhere 10) .Briefly, the FP group received continuous infusion of 5-FU (500 mg/m 2 ) daily for five days and intravenous infusion of cisplatin (10 mg/m 2 ) on days 1 to 5, repeated every 4 weeks.The DOC group received DOC (10 mg/m 2 ) by intravenous infusion on day 1, repeated weekly.

Radiation methods
All but two patients received a total dose of 40 Gy administered in 20 fractions over 4 weeks.One patient received 41.4 Gy in 22 fractions.The other patient was treated with 30.6 Gy in 17 fractions.Radiation therapy was administered by a linear accelerator (Mevatrone KDX77; Siemens AG, Erlangen, Germany or Clinac 21EX; Varian Medical Systems, Palo Alto, CA, USA) using two opposing anterior-posterior and posterior-anterior 10-MV X-ray beams.One or two additional subfields were added if necessary to maintain the dose distribution from 95% to 107% of the prescribed dose within the target volume.
The clinical target volumes included the primary tumors and bilateral supraclavicular and mediastinal lymph nodal regions (T-field) in 46 patients; primary lesions and mediastinum (I-field) in 4 patients; and primary lesions, mediastinum, and pericardiac regions (L-field) in 7 patients.

Surgery
The response to CRT was assessed in all patients 1 month after completion of treatment using barium esophagography, esophagoscopy, endoscopic ultrasonography, cervical node ultrasonography, and cervical, chest, and abdominal CT scans.Surgery was performed within 4 to 6 weeks of completion of CRT for operable tumors.Additional irradiation of 20 to 26 Gy and/or combination chemotherapy was administered for inoperable tumors.
The pathological response to treatment was judged using the Rules for Classification of Esophageal Cancer in Japan of the Japan Esophageal Society 14) .Adverse effects were scored according to the National Cancer Institute Common Toxicity Criteria, v4.0 15) .

Statistical evaluation
Overall survival (OS) rates were calculated from the first day of treatment using the Kaplan-Meier method.Differences between curves were analyzed by the log-rank test appropriate when assessing the differences in OS, because the number of samples was not large, and it was difficult to assume a normal distribution.Differences in the incidence of factors between the two groups were compared using Fisherʼs exact test or the Mann-Whitney U test.The threshold of statistical significance was p < 0.05.These analyses were performed using HALBAU statistical software (http://halbau.jp/index.htm) and BellCurve for Execel ver.2.0.

Treatment sequelae
Treatment sequelae are presented in Table-2.All but one patient completed radiation therapy.One patient discontinued treatment at 30.6 Gy in 17 fractions due to deteriorating pulmonary symptoms.Four patients discontinued chemotherapy because of anaphylaxis in response to DOC (n = 1), skin reactions (n = 1), mucositis (n = 1), and bone marrow suppression (n = 1, WBC: 1,200/mm 3 ).Bone marrow suppression was a common acute adverse effect of CRT.One patient developed grade 4 thrombocytopenia, and six patients developed grade 3 leukocytopenia.One patient developed a grade 3 esophagobronchial fistula.

Responses to CRT 1 month after completion of treatment
Based on the imaging results, the clinical responses were partial response in 19 patients, stable disease in 34, and progressive disease in 4. Of the 57 patients, 36 (63%) were deemed to have operable tumors and underwent curative intent surgery.A total of 27 patients underwent right thoracotomy and total extirpation of the thoracoabdominal esophagus combined with lymph node dissection in the cervical, thoracic, and abdominal regions.Another 2 patients underwent the same resection only in the lymph node regions of the mediastinum and abdomen.A total of 7 patients underwent other surgeries.
Table-3 presents the pathological evaluation results of the resected lesions in 36 patients.A complete pathological response (grade 3) 13) was achieved in 3 patients.The other responses were grade 2 (The number of proliferating cells is 1/3 or less) in 16 patients, grade 1a (Proliferable cells are 2/3 or more) in 6 patients, grade 1b (Proliferable cells are 1/3 or more and less than 2/3) in 10 patients, and unknown in 1 patient.Thirty patients had a stage pT3 or lower tumor at the time of surgery, whereas six had pT4 tumors.Twenty-four patients underwent R0 (no residual tumor) 13) resection, and 12 underwent R1 (microscopic residual tumor) or R2 (macroscopic residual tumor) resection.
Of the 21 patients who were not designated to undergo curative intent surgery, 4 were treated with palliative surgery, 2 with additional CRT, 8 with radiation therapy only, 2 with chemotherapy only, and 1 with esophageal stent placement.Another 4 patients received no further treatment.

Survival
The OS rates for all patients were 39% and 21% 1 year and 2 years after the commencement of treatment, respectively.The median survival time was 9 months (Figure -1).Between curative intent surgery group and inoperable group, the 2-years survival rates were 31% vs 0%, median survival time were 10.5 months vs 6 months, respectively.The difference in OS between these two groups was significant.
In the surgery group, there was no significant  Data are presented as number of patients.＊ Among these 24 patients, tumor cells were observed at the surgical margins in 6 patients.(p =0.011).Difference in OS between patients who underwent R0 resection and those who underwent R1 or R2 resection (Figure -2).In the nonsurgery group, there was no significant difference in OS between patients who received additional radiation and patients who did not (p = 0.19).

Discussion
Preoperative CRT for cT4 esophageal cancer was performed relatively safely in 56 (98%) of 57 patients.The 2-year OS for the 36 patients who underwent resection with curative intent was 31%.Nishimura, et al. 16) reported a 2-year OS of 27% for 11 patients with cT4 stage III esophageal cancer treated with CRT; in that study, only one of the five patients who underwent resection after CRT (30-60 Gy) was still alive after 2 years.
Table-4 summarizes the treatment results for the patients with cT4 disease who received CRT 16)-20) .Overall, there was no difference in OS between patients who underwent CRT alone and patients who underwent resection after CRT.Fujita 10) has proposed that the standard treatment should be definitive CRT or palliation.Our results and those reported by others support this recommendation.In our study, survival rates were higher in the surgery group than in the nonsurgery group, but this result was likely influenced by a strong selection bias.
Among all patients with cT4 disease at diagnosis, 36 (63%) had stage T3 or lower disease after CRT as confirmed by imaging examination, and 30 (53%) had pathological T3 or lower disease at the time of surgery.These findings indicate either that CRT is effective in reducing tumor size or that patients were overdiagnosed during the pretreatment examinations.Moreover, 12 patients who underwent surgery did not achieve complete resection, indicating that one-third of tumors were misdiagnosed as operable by imaging examination after CRT.Uncertainties about the extent of tumor invasion based on imaging findings should be considered when selecting a treatment plan for individual patients.The extent of tumor infiltration into adjacent organs is usually diagnosed using the CT criteria described above 13) , but adequately judging tumor invasion to other structures can be difficult.More accurate diagnostic methods are needed to improve the outcomes of esophageal cancer treatment.A total of 46 patients were treated with DOC in this study.In our previous study of T3 disease, pathological evaluation of the resected lesions revealed a significant curative effect in the FP group but a higher survival rate in the DOC group than in the FP group 11) .The intensity of the FP treatment in this study was lower than the protocol used in standard induction chemotherapy or in definitive CRT for esophageal cancer.This is one possible reason for the insufficient effect of induction CRT in this study.Stahl,et al. 8) demonstrated that in patients who received neoadjuvant CRT, continuation of CRT resulted in OS equivalent to that of surgery.However, in the nonsurgery group in the present study, there was no difference in OS between the patients who did and did not receive additional radiation.This suggests that additional treatment of patients who did not respond to preoperative CRT had no effect, and palliative therapy should instead be considered for these patients.

Conclusions
Preoperative CRT for cT4 esophageal cancer was performed relatively safely but did not improve OS compared to the reported results of patients treated with CRT alone.
Data are presented as the number of patients.

Figure- 1
Figure-1Kaplan-Meier curve of overall survival for 57 patients with cT4 stage III esophageal cancer who received preoperative CRT according to operability.Median Survival Time: All patients 9 months, Surgery group 11 months, Non-surgery group 6 months

Table - 1
Patient characteristics

Table - 2
Acute adverse effects of preoperative chemoradiation therapy

Table - 4
Clinical results of CRT for T4 and/or M1 esophageal cancer