Diseases of the Esophagus (2015) 28, 460–467 DOI: 10.1111/dote.12217
Original article
Salvage chemoradiotherapy for locally advanced esophageal carcinomas Y. Nakajima,1 K. Kawada,1 Y. Tokairin,1 Y. Miyawaki,1 T. Okada,1 T. Ryotokuji,1 N. Fujiwara,1 K. Saito,1 H. Fujiwara,1 T. Ogo,1 M. Okuda,1 K. Nagai,1 S. Miyake,2 T. Kawano1 Departments of 1Esophageal and General Surgery and 2Clinical Oncology, Graduate School of Medicine, Tokyo Medical and Dental University, Tokyo, Japan
SUMMARY. ‘Salvage chemoradiotherapy (CRT)’ was introduced in 2005 to treat thoracic esophageal carcinomas deemed unresectable based on the intraoperative findings. The therapeutic concept is as follows: the surgical plan is changed to an operation that aims to achieve curability by the subsequent definitive CRT. For this purpose, the invading tumor is resected as much as possible, and systematic lymph node dissection is performed except for in the area around the bilateral recurrent nerves. The definitive CRT should be started as soon as possible and should be performed as planned. We hypothesized that this treatment would be feasible and provide good clinical effects. We herein verified this hypothesis. Twenty-seven patients who received salvage CRT were enrolled in the study, and their clinical course, therapeutic response, and prognosis were evaluated. The patients who had poor oral intake because of esophageal stenosis were able to eat solid food soon after the operation. The radiation field could be narrowed after surgery, and this might have contributed to the high rate of finishing the definitive CRT as planned. As a result, the overall response rate was 74.1%, and 48.1% of the patients had a complete response. No patient experienced fistula formation. The 1-, 3-, and 5-year overall survival rates were 66.5%, 35.2%, and 35.2%, respectively. Salvage CRT had clinical benefits, such as the fact that patients became able to have oral intake, that fistula formation could be prevented, that the adverse events associated with the definitive CRT could be reduced, and that prognosis of the patients was satisfactory. Although the rate of recurrent nerve paralysis was relatively high even after the suspension of aggressive bilateral recurrent nerve lymph node dissection, and the rate of the progressive disease after the definitive CRT was high, salvage CRT appears to provide some advantages for the patients who would otherwise not have other treatment options following a non-curative and residual operation. KEY WORDS: esophageal carcinoma, multimodality treatment, non-curative surgery, salvage therapy, T4.
INTRODUCTION Esophageal carcinoma is the sixth most frequently diagnosed cancer and the fifth most common cause of death among males worldwide.1 In addition, the prognosis of the disease remains unsatisfactory. This is partly because esophageal carcinoma easily invades into non-resectable structures, such as the aorta, trachea, and main bronchus, because the esophagus histologically lacks serosa and is surrounded by adventitia and because the esophagus is anatomically adjacent to several important Address correspondence to: Dr Yasuaki Nakajima, MD, PhD, Department of Esophageal and General Surgery, Graduate School of Medicine, Tokyo Medical and Dental University, 1–5-45 Yushima, Bunkyo-ku, Tokyo 113–8519, Japan. Email: [email protected] Conflicts of interest: The authors declare that they have no conflict of interest. 460
structures, with which fistula formation can directly lead to death. Definitive chemoradiotherapy (CRT) is performed worldwide for cases of T4b esophageal carcinoma2,3 according to the TNM Classification of Malignant Tumors, 7th edition.4 On the other hand, neoadjuvant CRT followed by surgery is the most common treatment for T3 carcinoma in Western countries,5 while neoadjuvant chemotherapy followed by surgery has become the standard treatment strategy in Japan.6 Because of the fact that precisely which neoadjuvant strategy is superior remains controversial, a multicenter, randomized trial is currently ongoing in Japan.7 Regardless of the type of treatment strategy, neoadjuvant therapy has been reported to have a strong survival advantage over surgery alone8 and is currently becoming a popular therapeutic strategy worldwide. However, in the actual clinical setting, surgery is often selected as the initial treatment in © 2014 International Society for Diseases of the Esophagus
Salvage CRT for esophageal carcinomas
patients with locally advanced tumors, as neoadjuvant chemotherapy using cisplatin and fluorouracil (5-FU) (FP) sometimes induces disease progression and bulky esophageal tumors are often unresectable under neoadjuvant chemotherapy. Moreover, affected patients generally suffer from poor food intake, severe weight loss, and failing physical strength; therefore, surgery is usually recommended in such cases if there is a possibility that complete tumor resection with systematic lymph node dissection can be performed. In particular, although such therapeutic strategies are established, it remains difficult to precisely diagnose the tumor depth preoperatively. As a result, the tumor sometimes turns out to be unresectable based on the intraoperative findings, and in such cases, the patient may experience a very poor clinical course. Since 2005, we have utilized so-called ‘salvage CRT’ in all cases diagnosed to be unresectable based on the intraoperative findings and applied this therapeutic strategy in 27 of 29 cases. The therapeutic concept of salvage CRT is as follows: the surgical plan is changed to an operation that aims to achieve curability by the subsequent administration of definitive CRT. For this purpose, the invading tumor is resected as much as possible, and systematic lymph node dissection is performed, excluding the area around the bilateral recurrent nerves. The definitive CRT should be started as soon as possible and should be performed as planned. Therefore, the operation should be performed to avoid morbidities and to allow the patient to quickly recover from the surgery so that the CRT can be started. We found that salvage CRT demonstrated clinical benefits such as the fact that patients became able to eat solid food, that the radiation field could be decreased and the adverse events associated with CRT could be reduced, and that fistula formation could be prevented. At the same time, feasible clinical effects were achieved. We herein describe our findings regarding the clinical validity of salvage CRT.
PATIENTS AND METHODS Therapeutic strategy for salvage CRT When the intraoperative findings showed that an advanced primary esophageal tumor and/or metastatic lymph nodes had invaded non-resectable structures, such as the aorta, trachea, or left main bronchus, and it was considered to be impossible to perform a curative operation, the surgical plan was changed. Although the invading tumor mass was resected as much as possible to reduce the tumor volume, systematic cervical and superior mediastinal lymph node dissection, so-called lymphadenectomy around the recurrent nerve, was suspended, and only sampling of the lymph nodes pointed out as being © 2014 International Society for Diseases of the Esophagus
461
metastatic by the preoperative diagnosis was performed in order to avoid the development of surgical morbidities, such as bilateral recurrent nerve paralysis. The middle and lower mediastinal lymph node dissection was performed systematically, and the abdominal lymph node dissection around the cardia, the lesser curvature of the stomach, and the left gastric artery was performed as usual when making the gastric tube because the lymphadenectomy around these areas did not increase the morbidity rate and allowed the radiation field to be narrowed. Gastric tube reconstruction via the retrosternal route and cervical esophageal-gastric end-to-side anastomosis using a circular stapler were the standard methods. The residual tumors were marked with surgical clips to set up the precise radiation field. During the procedure, a feeding tube was inserted via a gastric tube to the jejunum and used to compensate for oral feeding after surgery and during the subsequent treatment period with definitive CRT. CRT was planned to begin between 1 and 2 months after the operation. Radiotherapy was delivered with 4 or 10 MV photons. All patients underwent three-dimensional radiotherapy planning. Anterior-posterior opposing portal irradiation was initiated at approximately 40 Gy, and oblique portal irradiation was then performed to spare the spinal cord from the radiation. Patients received conventional fractionated radiation of 1.8–2.0 Gy per fraction, for a total dose of 59.4–60 Gy, five times a week over a period of 6 weeks. The radiation field was planned to include the residual tumor(s) and the areas where lymph node dissection had been avoided. Concurrent chemotherapy was performed mainly using cisplatin and 5-FU, and was administered every 4 weeks, for a total of two courses during the radiotherapy. The regimen consisted of 70 mg/m2 cisplatin on day 1 and 700 mg/m2 of 5-FU via a continuous drip infusion on days 1–4. However, the concurrent chemotherapeutic regimen was selected according to the patients’ performance status and tumor malignancy, such as the spread and the number of the metastatic lymph nodes demonstrated by the pathological findings. The therapeutic efficacy was primarily evaluated using computed tomography (CT). Prior to the administration of the definitive CRT, an enhanced CT study was carried out in order to confirm the current status of the residual tumor(s) and verify the absence of any new metastatic lesions. In cases in which CT was unable to detect the residual tumor(s) because the tumor volume was very small, positron emission tomography (PET) was also used to evaluate the tumor response. In cases in which neither enhanced CT nor PET detected the residual tumor, additional examinations were conducted every month using a combination of CT and PET. When the definitive CRT was effective, additional
462
Diseases of the Esophagus
Esophageal carcinoma patients treated in our institute between 2005 and 2012; n=928 Thoracic esophageal squamous cell carcinomas invading over the adventitia; n=347
Clinically resectable; n=253
Clinically unresectable; n=94
Underwent surgery with right thoracotomy or by thoracoscopic surgery without neoadjuvant therapy; n=184
pT3 and R0; n=123 (T3 group)
pT4a and R0; n=16 (T4a group)
Due to T4b tumor; n=73
pT4b and R2; n=29
Received definitive CRT for the residual tumor soon after surgery; n=27 (salvage CRT group)
Others; n=16
Due to distant organ metastasis; n=21
Received definitive CRT; n=61 (CRT group)
Others; n=12
Others; n=2 CRT with salvage operation; n=11 (salvage op group)
CRT without salvage operation; n=50
Fig. 1 A flow chart of the patients included in the study. T3 means that the tumor had invaded into the adventitia, T4a means that the tumor had invaded into resectable adjacent structures such as the pleura, pericardium, or diaphragm, and T4b means that the tumor had invaded into non-resectable adjacent structures such as the aorta, trachea, or main bronchus. R0 means no residual tumor, and R2 means a macroscopic residual tumor. CRT = chemoradiotherapy.
chemotherapy was performed. This therapeutic strategy was named ‘salvage CRT’ and was adopted when unresectable esophageal carcinoma was demonstrated by the intraoperative findings. Patients Between 2005 and 2012, 928 esophageal carcinoma patients were treated at our institute. Three hundred and seventy-three patients had thoracic esophageal carcinomas invading over the adventitia. Among the 263 patients diagnosed with resectable disease, 190 underwent esophagectomy with right thoracotomy or thoracoscopic surgery without preoperative CRT or chemotherapy. Among them, 29 patients were unable to receive curative resection and continued to have macroscopic residual tumor(s) because of the presence of advanced primary esophageal tumors and/or metastatic lymph nodes that had invaded nonresectable structures. Twenty-seven of these patients who could undergo salvage CRT were enrolled in this study (Fig. 1). The clinical course, therapeutic response, and prognosis of these patients were evaluated retrospectively, and the clinical validity of salvage CRT was verified. In order to compare the clinical course and therapeutic efficacy, 123 patients who were diagnosed with advanced thoracic esophageal squamous cell carcinoma invading adventitia without residual microscopic tumors pathologically (T3 group) and 16 patients whose tumors had invaded resectable adjacent structures and were able to undergo curative surgery without residual microscopic tumors pathologically (T4a group) were enrolled in this study.
Because all patients in the salvage CRT group had squamous cell carcinoma, only squamous cell carcinomas of the thoracic esophagus were assessed. Sixty-one patients who were diagnosed to have unresectable disease because of advanced primary esophageal tumors and/or metastatic lymph nodes that had invaded into non-resectable structures and who underwent definitive CRT as the initial treatment during the same period were also enrolled and evaluated (CRT group). The CRT group included 11 patients who underwent a salvage procedure after CRT (salvage op group). Clinical and statistical evaluation Surgical complications were graded according to the Dindo et al. classification,9 and the therapeutic efficacy was evaluated based on the Response Evaluation Criteria in Solid Tumors guidelines.10 Because in some patients, the residual tumors were very small and could not be detected on enhanced CT, PET was used to evaluate the tumor response. In cases in which neither CT nor PET detected any residual tumor and no new metastatic lesions were noted, the response was classified as a complete response (CR). A univariate analysis was performed using the χ2 test. The survival rates were calculated according to the Kaplan–Meier method and analyzed using the log-rank test. All statistical analyses were performed using the JMP ver. 10.0 software package for Macintosh (SAS Japan, Inc., Tokyo, Japan). A value of P < 0.05 was considered to be statistically significant. © 2014 International Society for Diseases of the Esophagus
Salvage CRT for esophageal carcinomas
RESULTS Characteristics of the salvage CRT patients Among the 29 patients whose tumors were diagnosed to be unresectable based on the intraoperative findings, 27 patients could undergo salvage CRT. The reasons why the other two patients could not undergo salvage CRT and their clinical courses were as follows. One rejected additional treatment and died due to the growth of the residual tumor and multiple liver metastases 9 months after the operation. The other patient could not start salvage CRT because of repeated aspiration pneumonia and a poor performance status. This patient had a residual tumor at the trachea and left main bronchus, and a CT study performed 2 months after the operation showed rapid growth of the tumor. The trachea and both sides of the main bronchus were surrounded by a tumor that caused deformation. The growing residual tumor invaded into the lumen of the trachea, and as a result, the patient died 3 months after the operation. The characteristics of the patients are listed in Table 1. With regard to the tumor location, both the salvage CRT and CRT groups had a significant tendency for the primary esophageal tumors to be located in the upper and middle thoracic esophagus (P < 0.0001), although there were no significant differences between the salvage CRT group and the CRT group. According to the intraoperative findings, 21 (77.8%) primary esophageal tumors and 8 (29.6%) metastatic lymph nodes had invaded non-resectable structures. Both the primary tumor and metastatic
463
nodes were unresectable in two patients. Invasion into the respiratory tract and non-resectable blood vessels was confirmed in 23 (85.2%) and 11 (40.7%) patients, respectively. Simultaneous respiratory tract and non-resectable blood vessel invasion was observed in seven patients. Five primary esophageal tumors had invaded both the left main bronchus and aorta, and the residual tumors were most frequently located around this area. There were no significant differences with respect to unresectable lesions and invaded structures between the salvage CRT and CRT groups. Intraoperative and postoperative courses The status of the lymph node dissection and postoperative morbidities are shown in Table 2. In the salvage CRT group, the rate of systematic lymph node dissection on both sides of the cervical and superior mediastinal areas was significantly lower than that in the T3 or T4a group (all P values were
Original article
Salvage chemoradiotherapy for locally advanced esophageal carcinomas Y. Nakajima,1 K. Kawada,1 Y. Tokairin,1 Y. Miyawaki,1 T. Okada,1 T. Ryotokuji,1 N. Fujiwara,1 K. Saito,1 H. Fujiwara,1 T. Ogo,1 M. Okuda,1 K. Nagai,1 S. Miyake,2 T. Kawano1 Departments of 1Esophageal and General Surgery and 2Clinical Oncology, Graduate School of Medicine, Tokyo Medical and Dental University, Tokyo, Japan
SUMMARY. ‘Salvage chemoradiotherapy (CRT)’ was introduced in 2005 to treat thoracic esophageal carcinomas deemed unresectable based on the intraoperative findings. The therapeutic concept is as follows: the surgical plan is changed to an operation that aims to achieve curability by the subsequent definitive CRT. For this purpose, the invading tumor is resected as much as possible, and systematic lymph node dissection is performed except for in the area around the bilateral recurrent nerves. The definitive CRT should be started as soon as possible and should be performed as planned. We hypothesized that this treatment would be feasible and provide good clinical effects. We herein verified this hypothesis. Twenty-seven patients who received salvage CRT were enrolled in the study, and their clinical course, therapeutic response, and prognosis were evaluated. The patients who had poor oral intake because of esophageal stenosis were able to eat solid food soon after the operation. The radiation field could be narrowed after surgery, and this might have contributed to the high rate of finishing the definitive CRT as planned. As a result, the overall response rate was 74.1%, and 48.1% of the patients had a complete response. No patient experienced fistula formation. The 1-, 3-, and 5-year overall survival rates were 66.5%, 35.2%, and 35.2%, respectively. Salvage CRT had clinical benefits, such as the fact that patients became able to have oral intake, that fistula formation could be prevented, that the adverse events associated with the definitive CRT could be reduced, and that prognosis of the patients was satisfactory. Although the rate of recurrent nerve paralysis was relatively high even after the suspension of aggressive bilateral recurrent nerve lymph node dissection, and the rate of the progressive disease after the definitive CRT was high, salvage CRT appears to provide some advantages for the patients who would otherwise not have other treatment options following a non-curative and residual operation. KEY WORDS: esophageal carcinoma, multimodality treatment, non-curative surgery, salvage therapy, T4.
INTRODUCTION Esophageal carcinoma is the sixth most frequently diagnosed cancer and the fifth most common cause of death among males worldwide.1 In addition, the prognosis of the disease remains unsatisfactory. This is partly because esophageal carcinoma easily invades into non-resectable structures, such as the aorta, trachea, and main bronchus, because the esophagus histologically lacks serosa and is surrounded by adventitia and because the esophagus is anatomically adjacent to several important Address correspondence to: Dr Yasuaki Nakajima, MD, PhD, Department of Esophageal and General Surgery, Graduate School of Medicine, Tokyo Medical and Dental University, 1–5-45 Yushima, Bunkyo-ku, Tokyo 113–8519, Japan. Email: [email protected] Conflicts of interest: The authors declare that they have no conflict of interest. 460
structures, with which fistula formation can directly lead to death. Definitive chemoradiotherapy (CRT) is performed worldwide for cases of T4b esophageal carcinoma2,3 according to the TNM Classification of Malignant Tumors, 7th edition.4 On the other hand, neoadjuvant CRT followed by surgery is the most common treatment for T3 carcinoma in Western countries,5 while neoadjuvant chemotherapy followed by surgery has become the standard treatment strategy in Japan.6 Because of the fact that precisely which neoadjuvant strategy is superior remains controversial, a multicenter, randomized trial is currently ongoing in Japan.7 Regardless of the type of treatment strategy, neoadjuvant therapy has been reported to have a strong survival advantage over surgery alone8 and is currently becoming a popular therapeutic strategy worldwide. However, in the actual clinical setting, surgery is often selected as the initial treatment in © 2014 International Society for Diseases of the Esophagus
Salvage CRT for esophageal carcinomas
patients with locally advanced tumors, as neoadjuvant chemotherapy using cisplatin and fluorouracil (5-FU) (FP) sometimes induces disease progression and bulky esophageal tumors are often unresectable under neoadjuvant chemotherapy. Moreover, affected patients generally suffer from poor food intake, severe weight loss, and failing physical strength; therefore, surgery is usually recommended in such cases if there is a possibility that complete tumor resection with systematic lymph node dissection can be performed. In particular, although such therapeutic strategies are established, it remains difficult to precisely diagnose the tumor depth preoperatively. As a result, the tumor sometimes turns out to be unresectable based on the intraoperative findings, and in such cases, the patient may experience a very poor clinical course. Since 2005, we have utilized so-called ‘salvage CRT’ in all cases diagnosed to be unresectable based on the intraoperative findings and applied this therapeutic strategy in 27 of 29 cases. The therapeutic concept of salvage CRT is as follows: the surgical plan is changed to an operation that aims to achieve curability by the subsequent administration of definitive CRT. For this purpose, the invading tumor is resected as much as possible, and systematic lymph node dissection is performed, excluding the area around the bilateral recurrent nerves. The definitive CRT should be started as soon as possible and should be performed as planned. Therefore, the operation should be performed to avoid morbidities and to allow the patient to quickly recover from the surgery so that the CRT can be started. We found that salvage CRT demonstrated clinical benefits such as the fact that patients became able to eat solid food, that the radiation field could be decreased and the adverse events associated with CRT could be reduced, and that fistula formation could be prevented. At the same time, feasible clinical effects were achieved. We herein describe our findings regarding the clinical validity of salvage CRT.
PATIENTS AND METHODS Therapeutic strategy for salvage CRT When the intraoperative findings showed that an advanced primary esophageal tumor and/or metastatic lymph nodes had invaded non-resectable structures, such as the aorta, trachea, or left main bronchus, and it was considered to be impossible to perform a curative operation, the surgical plan was changed. Although the invading tumor mass was resected as much as possible to reduce the tumor volume, systematic cervical and superior mediastinal lymph node dissection, so-called lymphadenectomy around the recurrent nerve, was suspended, and only sampling of the lymph nodes pointed out as being © 2014 International Society for Diseases of the Esophagus
461
metastatic by the preoperative diagnosis was performed in order to avoid the development of surgical morbidities, such as bilateral recurrent nerve paralysis. The middle and lower mediastinal lymph node dissection was performed systematically, and the abdominal lymph node dissection around the cardia, the lesser curvature of the stomach, and the left gastric artery was performed as usual when making the gastric tube because the lymphadenectomy around these areas did not increase the morbidity rate and allowed the radiation field to be narrowed. Gastric tube reconstruction via the retrosternal route and cervical esophageal-gastric end-to-side anastomosis using a circular stapler were the standard methods. The residual tumors were marked with surgical clips to set up the precise radiation field. During the procedure, a feeding tube was inserted via a gastric tube to the jejunum and used to compensate for oral feeding after surgery and during the subsequent treatment period with definitive CRT. CRT was planned to begin between 1 and 2 months after the operation. Radiotherapy was delivered with 4 or 10 MV photons. All patients underwent three-dimensional radiotherapy planning. Anterior-posterior opposing portal irradiation was initiated at approximately 40 Gy, and oblique portal irradiation was then performed to spare the spinal cord from the radiation. Patients received conventional fractionated radiation of 1.8–2.0 Gy per fraction, for a total dose of 59.4–60 Gy, five times a week over a period of 6 weeks. The radiation field was planned to include the residual tumor(s) and the areas where lymph node dissection had been avoided. Concurrent chemotherapy was performed mainly using cisplatin and 5-FU, and was administered every 4 weeks, for a total of two courses during the radiotherapy. The regimen consisted of 70 mg/m2 cisplatin on day 1 and 700 mg/m2 of 5-FU via a continuous drip infusion on days 1–4. However, the concurrent chemotherapeutic regimen was selected according to the patients’ performance status and tumor malignancy, such as the spread and the number of the metastatic lymph nodes demonstrated by the pathological findings. The therapeutic efficacy was primarily evaluated using computed tomography (CT). Prior to the administration of the definitive CRT, an enhanced CT study was carried out in order to confirm the current status of the residual tumor(s) and verify the absence of any new metastatic lesions. In cases in which CT was unable to detect the residual tumor(s) because the tumor volume was very small, positron emission tomography (PET) was also used to evaluate the tumor response. In cases in which neither enhanced CT nor PET detected the residual tumor, additional examinations were conducted every month using a combination of CT and PET. When the definitive CRT was effective, additional
462
Diseases of the Esophagus
Esophageal carcinoma patients treated in our institute between 2005 and 2012; n=928 Thoracic esophageal squamous cell carcinomas invading over the adventitia; n=347
Clinically resectable; n=253
Clinically unresectable; n=94
Underwent surgery with right thoracotomy or by thoracoscopic surgery without neoadjuvant therapy; n=184
pT3 and R0; n=123 (T3 group)
pT4a and R0; n=16 (T4a group)
Due to T4b tumor; n=73
pT4b and R2; n=29
Received definitive CRT for the residual tumor soon after surgery; n=27 (salvage CRT group)
Others; n=16
Due to distant organ metastasis; n=21
Received definitive CRT; n=61 (CRT group)
Others; n=12
Others; n=2 CRT with salvage operation; n=11 (salvage op group)
CRT without salvage operation; n=50
Fig. 1 A flow chart of the patients included in the study. T3 means that the tumor had invaded into the adventitia, T4a means that the tumor had invaded into resectable adjacent structures such as the pleura, pericardium, or diaphragm, and T4b means that the tumor had invaded into non-resectable adjacent structures such as the aorta, trachea, or main bronchus. R0 means no residual tumor, and R2 means a macroscopic residual tumor. CRT = chemoradiotherapy.
chemotherapy was performed. This therapeutic strategy was named ‘salvage CRT’ and was adopted when unresectable esophageal carcinoma was demonstrated by the intraoperative findings. Patients Between 2005 and 2012, 928 esophageal carcinoma patients were treated at our institute. Three hundred and seventy-three patients had thoracic esophageal carcinomas invading over the adventitia. Among the 263 patients diagnosed with resectable disease, 190 underwent esophagectomy with right thoracotomy or thoracoscopic surgery without preoperative CRT or chemotherapy. Among them, 29 patients were unable to receive curative resection and continued to have macroscopic residual tumor(s) because of the presence of advanced primary esophageal tumors and/or metastatic lymph nodes that had invaded nonresectable structures. Twenty-seven of these patients who could undergo salvage CRT were enrolled in this study (Fig. 1). The clinical course, therapeutic response, and prognosis of these patients were evaluated retrospectively, and the clinical validity of salvage CRT was verified. In order to compare the clinical course and therapeutic efficacy, 123 patients who were diagnosed with advanced thoracic esophageal squamous cell carcinoma invading adventitia without residual microscopic tumors pathologically (T3 group) and 16 patients whose tumors had invaded resectable adjacent structures and were able to undergo curative surgery without residual microscopic tumors pathologically (T4a group) were enrolled in this study.
Because all patients in the salvage CRT group had squamous cell carcinoma, only squamous cell carcinomas of the thoracic esophagus were assessed. Sixty-one patients who were diagnosed to have unresectable disease because of advanced primary esophageal tumors and/or metastatic lymph nodes that had invaded into non-resectable structures and who underwent definitive CRT as the initial treatment during the same period were also enrolled and evaluated (CRT group). The CRT group included 11 patients who underwent a salvage procedure after CRT (salvage op group). Clinical and statistical evaluation Surgical complications were graded according to the Dindo et al. classification,9 and the therapeutic efficacy was evaluated based on the Response Evaluation Criteria in Solid Tumors guidelines.10 Because in some patients, the residual tumors were very small and could not be detected on enhanced CT, PET was used to evaluate the tumor response. In cases in which neither CT nor PET detected any residual tumor and no new metastatic lesions were noted, the response was classified as a complete response (CR). A univariate analysis was performed using the χ2 test. The survival rates were calculated according to the Kaplan–Meier method and analyzed using the log-rank test. All statistical analyses were performed using the JMP ver. 10.0 software package for Macintosh (SAS Japan, Inc., Tokyo, Japan). A value of P < 0.05 was considered to be statistically significant. © 2014 International Society for Diseases of the Esophagus
Salvage CRT for esophageal carcinomas
RESULTS Characteristics of the salvage CRT patients Among the 29 patients whose tumors were diagnosed to be unresectable based on the intraoperative findings, 27 patients could undergo salvage CRT. The reasons why the other two patients could not undergo salvage CRT and their clinical courses were as follows. One rejected additional treatment and died due to the growth of the residual tumor and multiple liver metastases 9 months after the operation. The other patient could not start salvage CRT because of repeated aspiration pneumonia and a poor performance status. This patient had a residual tumor at the trachea and left main bronchus, and a CT study performed 2 months after the operation showed rapid growth of the tumor. The trachea and both sides of the main bronchus were surrounded by a tumor that caused deformation. The growing residual tumor invaded into the lumen of the trachea, and as a result, the patient died 3 months after the operation. The characteristics of the patients are listed in Table 1. With regard to the tumor location, both the salvage CRT and CRT groups had a significant tendency for the primary esophageal tumors to be located in the upper and middle thoracic esophagus (P < 0.0001), although there were no significant differences between the salvage CRT group and the CRT group. According to the intraoperative findings, 21 (77.8%) primary esophageal tumors and 8 (29.6%) metastatic lymph nodes had invaded non-resectable structures. Both the primary tumor and metastatic
463
nodes were unresectable in two patients. Invasion into the respiratory tract and non-resectable blood vessels was confirmed in 23 (85.2%) and 11 (40.7%) patients, respectively. Simultaneous respiratory tract and non-resectable blood vessel invasion was observed in seven patients. Five primary esophageal tumors had invaded both the left main bronchus and aorta, and the residual tumors were most frequently located around this area. There were no significant differences with respect to unresectable lesions and invaded structures between the salvage CRT and CRT groups. Intraoperative and postoperative courses The status of the lymph node dissection and postoperative morbidities are shown in Table 2. In the salvage CRT group, the rate of systematic lymph node dissection on both sides of the cervical and superior mediastinal areas was significantly lower than that in the T3 or T4a group (all P values were
