- Email: [email protected]
Salvage surgery after failed chemoradiotherapy in squamous cell carcinoma of the esophagus
Available online at www.sciencedirect.com
EJSO 35 (2009) 289e294
www.ejso.com
Salvage surgery after failed chemoradiotherapy in squamous cell carcinoma of the esophagus Y.K. Chao a,b, S.C. Chan c, H.K. Chang d, Y.H. Liu a, Y.C. Wu a, M.J. Hsieh a, C.K. Tseng e, H.P. Liu a,* a
Division of Thoracic Surgery, Department of Surgery, Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, Taiwan b Graduate Institute of Clinical Medical Sciences, College of Medicine, Chang Gung University, Taoyuan, Taiwan c Department of Nuclear Medicine, Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, Taiwan d Division of Haematology/Oncology, Department of Internal Medicine, Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, Taiwan e Department of Radiation Oncology, Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, Taiwan Accepted 27 February 2008 Available online 8 April 2008
Abstract Aims: To investigate the survival benefit and preoperative risk factors for hospital mortality of salvage surgery in esophageal cancer patients who had locoregional residual/recurrent tumor after definitive chemoradiotherapy. Methods: We retrospectively reviewed the esophageal cancer patients who presented at our hospital from 1997 to 2004. Forty-seven patients who had squamous cell cancer and developed locoregional recurrent/persistent disease after primary definitive chemoradiotherapy were elected. Twenty-seven of them received salvage esophagectomy (group 1) and the other 20 underwent non-operative treatment only (group 2). In order to assess the surgery-related mobility and mortality in group 1, 191 patients who received neoadjuvant chemoradiotherapy followed by operation during the same time period were also enrolled (group 3). Results: The 5-year overall survival of group 1 patients was 25.4%. In contrast, all of the patients in the group 2 died within 16.7 months. The difference was statistically significant ( p ¼ 0.0029). In comparison with group 3, group 1 patients had significantly more surgeryrelated complications and hospital mortality. In univariate analysis for preoperative risk factors, a low albumin or hemoglobulin level was associated with high hospital mortality in group 1 ( p ¼ 0.004 and 0.003, respectively). After multivariate analysis, only the low albumin level remained borderline significance. As for disease specific survival after salvage surgery, R0 resection was the only independent prognosticator ( p ¼ 0.049). Conclusion: Salvage surgery provides survival benefit in esophageal cancer patients with locoregional persistent or recurrent disease after primary definitive chemoradiotherapy. Preoperative albumin and hemoglobulin levels are associated with hospital mortality and may aid in selecting suitable patient for salvage surgery. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Salvage surgery; Definitive chemoradiotherapy; Squamous cell carcinoma; Esophageal cancer; Recurrence
Introduction Current strategies for treating primary locoregionaladvanced esophageal cancer consisted of radical surgery, definitive chemoradiotherapy, or neoadjuvant chemoradiotherapy followed by operation.1,2 Among them, definitive chemoradiotherapy had less treatment-related mortality * Corresponding author. Division of Thoracic & Cardiovascular Surgery, Chang Gung Memorial Hospital, 5 Fu-Hsing Street, Kweishan, Taoyuan, Taiwan. Tel.: þ886 3 328 1200x2118; fax: þ886 3 328 5818. E-mail address: [email protected] (H.P. Liu). 0748-7983/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.ejso.2008.02.014
and an similar overall survival rate, especially for patients with squamous cell carcinoma.2,3 But patients who had chemoradiotherapy also tend to develop residual or recurrent disease, which often poses difficulty in clinical management.2,3 Treatments for the residual/recurrent disease at the locoregional sites include salvage surgery or chemotherapy. However, many clinicians opposed the use of salvage surgery due to serious post-operation mobility and mortality. And only a few reports with regard to this issue are available in literature.4e6 Most importantly, although the reported overall survival for patients receiving salvage
290
Y.K. Chao et al. / EJSO 35 (2009) 289e294
surgery is better than those receiving non-surgery treatments, the improved survival in the salvage group may be due to bias in patient selection among different studies and needs further validation. Besides, no investigator has studied the preoperative prognostic factors for outcome after salvage surgery, which can enable better patient selection beforehand and thus avoid unnecessary surgery. In this context, we conducted this retrospective study with 2 aims. One is to compare the survivals of esophageal cancer patients who received salvage surgery or nonsurgery therapy due to locoregional residual/recurrent tumor. The other is to investigate the preoperative prognostic factor for salvage surgery. Materials and methods Patient population The clinical data of primary esophageal cancer patients with the histology of squamous cell carcinoma treated in our hospital from 1995 to 2004 were reviewed. A total of 84 patients received definitive chemoradiotherapy and 47 of them had locoregional residual/recurrence tumor. The detail of definitive chemoradiotherapy is showed in the following section. Histological biopsy to prove recurrent/residual tumor was done in all feasible cases. For the other patients, the diagnosis of recurrent or persistent tumor was made by concordant imaging and clinical findings. Before salvage treatment, complete work-up including chest-to-abdomen CT, panedocsopy, esophagogram, bone scan, and serum biochemical examinations had been arranged for all patients. Twenty-seven of the 47 patients received salvage esophagectomy (group 1). Criteria for operability including: (i) physiological fit for surgery: no liver cirrhosis (> Child B) and heart failure (NYHA class III, IV); (ii) no evidence of tracheoesophageal fistula; and (iii) no recurrent laryngeal nerve invasion. The other 20 underwent chemotherapy/ supportive treatment due to unwillingness to receive surgery (group 2). To assess the surgery-related mobility and mortality in group 1, another group of patients who receive neoadjuvant chemoradiotherapy and esophagectomy as the primary definitive treatment in the same period was also recruited (group 3). Primary definitive chemoradiotherapy The primary definitive chemoradiotherapy consisted of two continuous courses of sequential chemoradiation. The regimens of chemotherapy and radiotherapy in each course are as followed: 5-Fluorouracil administered as a continuous infusion over 96 h (1000 mg/m2 per day) on days 1 through 4 and days 29 through 33; cisplatin administered as an intravenous bolus (75 mg/m2) on day 1 and day 29. Radiation with a total dose of 30 Gy in 200 cGy daily fractions, 5 days a week, was given between days 4 and 29.
Neoadjuvant chemoradiotherapy with surgery The regimen in neoadjuvant chemoradiotherapy is the same with that of definitive treatment but was given for one course only. One month after completion of chemoradiation, these patients received radical surgery. Our standard surgical approach is through a right side limited thoracotomy and intrathoracic gastric tube reconstruction (Ivorelewis procedure) for middle/lower-third lesions and neck anastomosis (Mackeown procedure) for upper-third/ cervical lesions. Lymph node dissection was two-field. Statistical analysis Chi-square and Student’s t-test were used to compare categorical and quantitative variables, respectively. Overall survival was calculated from the day of salvage treatment to the time of death or last follow-up. Our database was linked to that of National Cancer Registry which is updated every 6 months. Survival curves were constructed using the KaplaneMeier method and compared by log-rank test. A Cox proportional hazard model was constructed for multivariate analysis (SPSS, version 12.0). Results Characteristics and survivals for patients receiving salvage surgery or non-surgery treatment Table 1 showed the demographic data of group 1 and group 2 patients. No significant difference exists between characteristics of these 2 groups. In group 1, the pathological result of salvage surgery was T1e2N0 (defined as early stage) in nine patients (37%) and T1e4N1 or T3e4N0 (advanced stage) in another 15 (52%). Three patients achieved pathologic complete remission (pCR). All of the patients in the group 2 died within 16.7 months due to cancer or cancer-related complication. Table 1 Characteristics of patients receiving salvage surgery (group 1) or nonsurgery treatment (group 2)
Gender (%) Male Female Median age [year] Tumor location (%) Cervical Upper 1/3 Middle 1/3 Lower 1/3 Initial clinical stage (%)* Stage II Stage III Stage IV
Group 1 (n ¼ 27)
Group 2 (n ¼ 20)
p value
26 1 62.44
19 1 57.55
0.675
1 7 14 5
0 5 10 5
0.806
11 14 2
14 6 0
0.1
*According to 2002 AJCC/UICC TNM staging system.
0.085
Y.K. Chao et al. / EJSO 35 (2009) 289e294
For group 1, with the median follow-up time of 70.7 months, fifteen patients died. Thirteen were due to cancer or cancer-related complications while the other two with intercurrent disease. The 5-year OS was 25.4%. The OSs between these two groups were statistically different (Fig. 1, p ¼ 0.0029). Surgery-related outcomes of patients receiving salvage surgery or neoadjuvant chemotherapy followed by operation Table 2 demonstrated the characteristics, and compared surgery details as well as complication or mortality due to surgery for group 1 and 3 patients. Group 1 patients had a significantly elder age ( p ¼ 0.04), more cervical or upper-third cancer ( p ¼ 0.046), and less R0 resection result ( p ¼ 0.001). As for surgery-related complications, group 1 showed a significantly more pulmonary complication ( p ¼ 0.006) and anastomosis leakage ( p ¼ 0.002). The hospital mortality of group 1 was also much higher than that of group 2 ( p ¼ 0.03). The operation time, blood loss during surgery, and days of hospital stay were similar in these two groups. Regarding the six mortality cases in group 1, three died of sepsis resulted from intrathoracic anastomosis leakage (at 21, 28, 48 days after surgery, respectively). The other 3 died due to acute respiratory distress syndrome and pneumonia (at 8, 14, and 23 days after surgery, respectively). The 5-year OS of group 1 was similar to that of group 3 (Fig. 2). Risk factors for hospital mortality in patients receiving salvage surgery Table 3 showed the separate results of univariate analysis for preoperative and perioperative risk factors for
291
Table 2 Characteristics and surgery-related outcomes of patients receiving salvage surgery (group 1) or neoadjuvant chemotherapy followed by operation (group 3)
Gender (%) Male Female Median age [year] Mean preoperative albumin level [g/dL] Tumor location (%) Cervical Upper 1/3 Middle 1/3 Lower 1/3 Initial clinical stage (%) Stage IIa Stage IIb Stage III Stage IV Conduit use (%) Stomach Colon Resection result (%) R0 Non-R0 Operation time [min] Blood loss [ml] Pulmonary complication (%) Yes No Anastomosis leakage (%) Yes No Hospital mortality (%) Yes No Hospital stay [day]
Group 1 (n ¼ 27)
Group 3 (n ¼ 191)
p value
26 1 62.44 3.881
188 3 54.52 3.982
0.4
1 7 14 5
4 36 104 47
0.046
4 7 14 2
25 49 92 25
0.58
20 7
167 24
0.077
17 10 370 335
161 30 350 330
0.001
9 18
22 169
0.006
4 23
2 191
0.002
6 21 22.4
15 176 20.1
0.03
0.04 0.368
0.35 0.38
0.3
hospital mortality. The levels of preoperative albumin ( p ¼ 0.004) and hemoglobin ( p ¼ 0.003), and anastomosis leakage ( p ¼ 0.025) were noted to be significantly associated with increased mortality following esophageal resection. After multivariate analysis, anastomosis leakage was the only independent significant perioperative risk factor ( p ¼ 0.021). As for preoperative prognosticators, only the albumin level showed borderline significance ( p ¼ 0.069) in multivariate analysis. Prognostic predictors of disease-specific survival in patients receiving salvage surgery
Figure 1. Overall survivals for group 1 and group 2 patients.
After excluding the six hospital mortality cases, 21 patients entered survival analysis. Univariate analysis revealed the following postoperative factors were significantly related to increased survival: superficial depth of tumor invasion (T1e2, p ¼ 0.0082), early pathologic stage (T1e2N0, p ¼ 0.0035) and R0 resection ( p ¼ 0.0002). While multivariate analysis revealed that R0 resection
Y.K. Chao et al. / EJSO 35 (2009) 289e294
292
Table 4 Cox proportional hazard model for overall survival of patients who received salvage surgery Risk factors
Resection Non-R0 R0 Depth of tumor invasion T1e2 T3e4 Pathologic stage Early Advanced
Overall survival OR (95% CI)
p value
Baseline 4.92 (1.08 w 22.4)
0.039
Baseline 2.83 (0.26 w 30.5)
0.392
Baseline 8.5 (0.75 w 95)
0.083
OR, odds ratio; 95% CI, 95% confidence interval; Early, T1e2N0; Advanced, T3e4 or N1.
Figure 2. Overall survivals for Group 1 and Group 3 patients.
was the most important prognosticator for overall survival ( p ¼ 0.039, HR: 4.919) (Table 4, Fig. 3). Discussion Survival benefit of salvage esophagectomy Definitive chemoradiotherapy is considered a standard treatment for locally advanced esophageal cancer. Despite of long-term survival of 15e20%1e3 after such treatment, locoregional recurrence is not uncommon and occurs in around 40e60% of patients.1e3 Prognosis after locoregional recurrence is dismal and all patients will die in one year without treatment.7 Treatment options for recurrent or persistent disease after definitive chemoradiotherapy are limited. Further radiotherapy is often contraindicated because the maximal tolerable dose has been achieved. As for salvage chemotherapy, the treatment outcome is
disappointing with a median survival of 7 months.8 Salvage surgery is another choice but its role still remains debatable. In this study, we compared two groups of patients who received salvage surgery and chemotherapy/supportive treatment, respectively. The baseline characteristics for these 2 groups were similar. We noted that patients receiving salvage surgery had a significantly higher OS than those underwent chemotherapy only ( p ¼ 0.007). A high 5-year OS of 35.71% can be achieved in group 1 patients. In contrast, all patients died within 1.5 year without surgical treatment. This finding highlights the importance of salvage surgery in the management of recurrent esophageal cancer after chemoradiotherapy. A potential bias in this analysis is that group 2 is a mixed cohort consisting of patients who received chemotherapy or underwent support care. We did not analyze them separately because the reported survivals for these 2 kinds of patients were similar and some patients did give up chemotherapy during treatment.
Table 3 Univariate analysis of preoperative and perioperative risk factors for hospital mortality Risk factors
p value
Preoperative risk factor Albumin level Hemoglobin level FEV1 Tumor location Persistent/recurrent disease
0.004 0.003 0.898 0.703 0.385
Perioperative risk factor Operation time Operation blood loss Conduit use (stomach/colon) Anastomosis site (neck/chest) R0 resection Pneumonia Anastomosis leakage
0.18 0.447 0.594 0.662 0.16 0.367 0.025
Figure 3. Disease specific survivals of patients achieving R0 versus NonR0 resection.
Y.K. Chao et al. / EJSO 35 (2009) 289e294
In this context, we recommend that esophageal cancer patients who had locoregional recurrence after chemoradiotherapy should be encouraged to receive surgical resection if feasible.
293
distal end of the graft using the ileocolic artery and vein and cervical vessels has also been suggested for reducing leakage.13 Patient selection for salvage esophagectomy
Risk of salvage esophagectomy and prognosticator for surgical mortality Although patients with recurrent/persistent esophageal cancer had a higher survival rate after salvage surgery, another concern is the surgical mobility and mortality. Similar to the previous studies,4e6 we also enrolled another group of patients who undergo neoadjuvant chemotherapy followed by surgery to assess the surgery-related complications. Some may argue for the comparability of these two groups because difference in pre-surgery conditions such as previous radiation dose, time interval after radiation, tumor location, or complication during surgery. But these data actually are important reference for consultation of patients because some of them do refuse further operation after neoadjuvant chemoradiotherapy. In this study, we found group 1 patients had a significantly more pulmonary complication, anastomostic leakage and subsequently higher hospital mortality than group 3, which can be explained by the aggressive nature of recurrent disease and complicated surgical approach. The ratios of these surgery-related complications in our study are similar to those of other series.4e6 Salvage surgery for esophageal cancer after chemoradiation has been reported to associate with a high hospital mortality rate of 8 w 15%.4e6 In our study, patients who received salvage surgery also showed a high hospital mortality rate of 22.2%. Thus, the importance of finding a reliable preoperative prognosticator to optimize the use of salvage surgery can not be overemphasized. In this study, we noted a lower preoperative blood albumin or hemoglobin level was associated with higher hospital mortality. Hypoalbuminemia is a well-known predictor for complication of standard esophagectomy.9 It has been reported that when serum albumin is below 3.2 g/dL, esophagectomy will result in more complication and hospital mortality than other types of surgery.10 This may be due to the fact that hypoalbuminemia reflects not only a poor nutrition status but also weak immune condition with high susceptibility of infection. In this context, the preoperative albumin and hemoglobulin levels should be seriously regarded before proceed to surgery. Intrathoracic anastomosis leakage-related mortality is an important cause of death in the current series. The three fetal leakages in our series occurred several weeks after surgery and were believed to be due to poor gastric perfusion as a result of high exposure of the proximal stomach to radiation.11 Using the radiation-sparing colon as the reconstruction substitute and changing into cervical anastomosis via the substernal route may be helpful for preventing lethal leakage.12 An additional microvascular anastomosis at the
R0 resection has been reported to be the most important good prognosticator in the postoperative setting for esophageal cancer patients who receive either radical surgery or neoadjuvant chemoradiotherapy followed by operation.4,6 R0 resection in our series also represented as the most significant risk factor for overall survival. Precise evaluation of respectability before operation is difficult. Conventional CT scan is incorrect for T factor determination after chemoradiation due to thickened esophageal wall and an unclear plane with adjacent mediastinal structure after high dose radiotherapy.14 This is also why we did not include the preoperative stage in the survival analysis. Piessen and colleagues had reported a high R0 resection rate of 81.4% if the salvage surgery was performed for tumor with: (i) height <5 cm on esophagogram; and (ii) aortic contact <90 degrees on computed tomography.15 The main patient group was focused on non-responder after induction chemoradiotherapy. The value of those indicators might need further investigation in the setting of local failure after definitive chemoradiotherapy. Conclusions Salvage surgery provides survival benefit for esophageal cancer patients with locoregional recurrent disease but may lead to serious surgery-related complication and hospital mortality. Preoperative albumin and hemoglobulin levels are important risk factors for hospital mortality and may aid in selecting suitable patient for salvage surgery. Conflict of interest We hereby declare that there is no potential or actual personal, financial or political interest related to this article.
References 1. Bedenne L, Michel P, Bouche O, et al. Chemoradiation followed by surgery compared with chemoradiation alone in squamous cancer of the esophagus: FFCD 9102. J Clin Oncol 2007;25:1160–8. 2. Stahl M, Stuschke M, Lehmann N, et al. Chemoradiation with and without surgery in patients with locally advanced squamous cell carcinoma of the esophagus. J Clin Oncol 2005;23:2310–7. 3. Minsky BD, Pajak TF, Ginsberg RJ, et al. INT 0123 (Radiation Therapy Oncology Group 94-05) phase III trial of combined-modality therapy for esophageal cancer: high-dose versus standard-dose radiation therapy. J Clin Oncol 2002;20:1167–74. 4. Swisher SG, Wynn P, Putnam JB, et al. Salvage esophagectomy for recurrent tumors after definitive chemotherapy and radiotherapy. J Thorac Cardiovasc Surg 2002;123:175–83.
294
Y.K. Chao et al. / EJSO 35 (2009) 289e294
5. Nakamura T, Hayashi K, Ota M, et al. Salvage esophagectomy after definitive chemotherapy and radiotherapy for advanced esophageal cancer. Am J Surg 2004;188:261–6. 6. Tomimaru Y, Yano M, Takachi K, et al. Factors affecting the prognosis of patients with esophageal cancer undergoing salvage surgery after definitive chemoradiotherapy. J Surg Oncol 2006; 93:422–8. 7. Shimada H, Kitabayashi H, Nabeya Y, et al. Treatment response and prognosis of patients after recurrence of esophageal cancer. Surgery 2003;133:4–31. 8. Govindan R, Read W, Faust J, et al. Phase II study of docetaxel and irinotecan in metastatic or recurrent esophageal cancer: a preliminary report. Oncology (Williston Park) 2003;17(9 Suppl. 8): 27–31. 9. Steyerberg EW, Neville BA, Koppert LB, et al. Surgical mortality in patients with esophageal cancer: development and validation of a simple risk score. J Clin Oncol 2006;24:4277–84.
10. Kudsk KA, Tolley EA, DeWitt RC, et al. Preoperative albumin and surgical site identify surgical risk for major postoperative complications. JPEN J Parenter Enteral Nutr 2003;27:1–9. 11. Urschel JD. Esophagogastrostomy anastomotic leaks complicating esophagectomy: a review. Am J Surg 1995;169:634–40. 12. Blewett CJ, Miller JD, Young JE, et al. Anastomotic leaks after esophagectomy for esophageal cancer: a comparison of thoracic and cervical anastomoses. Ann Thorac Cardiovasc Surg 2001;7:75–8. 13. Sakuraba M, Kimata Y, Hishinuma S, et al. Importance of additional microvascular anastomosis in esophageal reconstruction after salvage esophagectomy. Plast Reconstr Surg 2004;113:1934–9. 14. Jones DR, Parker Jr LA, Detterbeck FC, et al. Inadequacy of computed tomography in assessing patients with esophageal carcinoma after induction chemoradiotherapy. Cancer 1999;85:1026–32. 15. Piessen G, Briez N, Triboulet JP, et al. Patients with locally advanced esophageal carcinoma nonresponder to radiochemotherapy: who will benefit from surgery? Ann Surg Oncol 2007;14:2036–44.
EJSO 35 (2009) 289e294
www.ejso.com
Salvage surgery after failed chemoradiotherapy in squamous cell carcinoma of the esophagus Y.K. Chao a,b, S.C. Chan c, H.K. Chang d, Y.H. Liu a, Y.C. Wu a, M.J. Hsieh a, C.K. Tseng e, H.P. Liu a,* a
Division of Thoracic Surgery, Department of Surgery, Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, Taiwan b Graduate Institute of Clinical Medical Sciences, College of Medicine, Chang Gung University, Taoyuan, Taiwan c Department of Nuclear Medicine, Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, Taiwan d Division of Haematology/Oncology, Department of Internal Medicine, Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, Taiwan e Department of Radiation Oncology, Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, Taiwan Accepted 27 February 2008 Available online 8 April 2008
Abstract Aims: To investigate the survival benefit and preoperative risk factors for hospital mortality of salvage surgery in esophageal cancer patients who had locoregional residual/recurrent tumor after definitive chemoradiotherapy. Methods: We retrospectively reviewed the esophageal cancer patients who presented at our hospital from 1997 to 2004. Forty-seven patients who had squamous cell cancer and developed locoregional recurrent/persistent disease after primary definitive chemoradiotherapy were elected. Twenty-seven of them received salvage esophagectomy (group 1) and the other 20 underwent non-operative treatment only (group 2). In order to assess the surgery-related mobility and mortality in group 1, 191 patients who received neoadjuvant chemoradiotherapy followed by operation during the same time period were also enrolled (group 3). Results: The 5-year overall survival of group 1 patients was 25.4%. In contrast, all of the patients in the group 2 died within 16.7 months. The difference was statistically significant ( p ¼ 0.0029). In comparison with group 3, group 1 patients had significantly more surgeryrelated complications and hospital mortality. In univariate analysis for preoperative risk factors, a low albumin or hemoglobulin level was associated with high hospital mortality in group 1 ( p ¼ 0.004 and 0.003, respectively). After multivariate analysis, only the low albumin level remained borderline significance. As for disease specific survival after salvage surgery, R0 resection was the only independent prognosticator ( p ¼ 0.049). Conclusion: Salvage surgery provides survival benefit in esophageal cancer patients with locoregional persistent or recurrent disease after primary definitive chemoradiotherapy. Preoperative albumin and hemoglobulin levels are associated with hospital mortality and may aid in selecting suitable patient for salvage surgery. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Salvage surgery; Definitive chemoradiotherapy; Squamous cell carcinoma; Esophageal cancer; Recurrence
Introduction Current strategies for treating primary locoregionaladvanced esophageal cancer consisted of radical surgery, definitive chemoradiotherapy, or neoadjuvant chemoradiotherapy followed by operation.1,2 Among them, definitive chemoradiotherapy had less treatment-related mortality * Corresponding author. Division of Thoracic & Cardiovascular Surgery, Chang Gung Memorial Hospital, 5 Fu-Hsing Street, Kweishan, Taoyuan, Taiwan. Tel.: þ886 3 328 1200x2118; fax: þ886 3 328 5818. E-mail address: [email protected] (H.P. Liu). 0748-7983/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.ejso.2008.02.014
and an similar overall survival rate, especially for patients with squamous cell carcinoma.2,3 But patients who had chemoradiotherapy also tend to develop residual or recurrent disease, which often poses difficulty in clinical management.2,3 Treatments for the residual/recurrent disease at the locoregional sites include salvage surgery or chemotherapy. However, many clinicians opposed the use of salvage surgery due to serious post-operation mobility and mortality. And only a few reports with regard to this issue are available in literature.4e6 Most importantly, although the reported overall survival for patients receiving salvage
290
Y.K. Chao et al. / EJSO 35 (2009) 289e294
surgery is better than those receiving non-surgery treatments, the improved survival in the salvage group may be due to bias in patient selection among different studies and needs further validation. Besides, no investigator has studied the preoperative prognostic factors for outcome after salvage surgery, which can enable better patient selection beforehand and thus avoid unnecessary surgery. In this context, we conducted this retrospective study with 2 aims. One is to compare the survivals of esophageal cancer patients who received salvage surgery or nonsurgery therapy due to locoregional residual/recurrent tumor. The other is to investigate the preoperative prognostic factor for salvage surgery. Materials and methods Patient population The clinical data of primary esophageal cancer patients with the histology of squamous cell carcinoma treated in our hospital from 1995 to 2004 were reviewed. A total of 84 patients received definitive chemoradiotherapy and 47 of them had locoregional residual/recurrence tumor. The detail of definitive chemoradiotherapy is showed in the following section. Histological biopsy to prove recurrent/residual tumor was done in all feasible cases. For the other patients, the diagnosis of recurrent or persistent tumor was made by concordant imaging and clinical findings. Before salvage treatment, complete work-up including chest-to-abdomen CT, panedocsopy, esophagogram, bone scan, and serum biochemical examinations had been arranged for all patients. Twenty-seven of the 47 patients received salvage esophagectomy (group 1). Criteria for operability including: (i) physiological fit for surgery: no liver cirrhosis (> Child B) and heart failure (NYHA class III, IV); (ii) no evidence of tracheoesophageal fistula; and (iii) no recurrent laryngeal nerve invasion. The other 20 underwent chemotherapy/ supportive treatment due to unwillingness to receive surgery (group 2). To assess the surgery-related mobility and mortality in group 1, another group of patients who receive neoadjuvant chemoradiotherapy and esophagectomy as the primary definitive treatment in the same period was also recruited (group 3). Primary definitive chemoradiotherapy The primary definitive chemoradiotherapy consisted of two continuous courses of sequential chemoradiation. The regimens of chemotherapy and radiotherapy in each course are as followed: 5-Fluorouracil administered as a continuous infusion over 96 h (1000 mg/m2 per day) on days 1 through 4 and days 29 through 33; cisplatin administered as an intravenous bolus (75 mg/m2) on day 1 and day 29. Radiation with a total dose of 30 Gy in 200 cGy daily fractions, 5 days a week, was given between days 4 and 29.
Neoadjuvant chemoradiotherapy with surgery The regimen in neoadjuvant chemoradiotherapy is the same with that of definitive treatment but was given for one course only. One month after completion of chemoradiation, these patients received radical surgery. Our standard surgical approach is through a right side limited thoracotomy and intrathoracic gastric tube reconstruction (Ivorelewis procedure) for middle/lower-third lesions and neck anastomosis (Mackeown procedure) for upper-third/ cervical lesions. Lymph node dissection was two-field. Statistical analysis Chi-square and Student’s t-test were used to compare categorical and quantitative variables, respectively. Overall survival was calculated from the day of salvage treatment to the time of death or last follow-up. Our database was linked to that of National Cancer Registry which is updated every 6 months. Survival curves were constructed using the KaplaneMeier method and compared by log-rank test. A Cox proportional hazard model was constructed for multivariate analysis (SPSS, version 12.0). Results Characteristics and survivals for patients receiving salvage surgery or non-surgery treatment Table 1 showed the demographic data of group 1 and group 2 patients. No significant difference exists between characteristics of these 2 groups. In group 1, the pathological result of salvage surgery was T1e2N0 (defined as early stage) in nine patients (37%) and T1e4N1 or T3e4N0 (advanced stage) in another 15 (52%). Three patients achieved pathologic complete remission (pCR). All of the patients in the group 2 died within 16.7 months due to cancer or cancer-related complication. Table 1 Characteristics of patients receiving salvage surgery (group 1) or nonsurgery treatment (group 2)
Gender (%) Male Female Median age [year] Tumor location (%) Cervical Upper 1/3 Middle 1/3 Lower 1/3 Initial clinical stage (%)* Stage II Stage III Stage IV
Group 1 (n ¼ 27)
Group 2 (n ¼ 20)
p value
26 1 62.44
19 1 57.55
0.675
1 7 14 5
0 5 10 5
0.806
11 14 2
14 6 0
0.1
*According to 2002 AJCC/UICC TNM staging system.
0.085
Y.K. Chao et al. / EJSO 35 (2009) 289e294
For group 1, with the median follow-up time of 70.7 months, fifteen patients died. Thirteen were due to cancer or cancer-related complications while the other two with intercurrent disease. The 5-year OS was 25.4%. The OSs between these two groups were statistically different (Fig. 1, p ¼ 0.0029). Surgery-related outcomes of patients receiving salvage surgery or neoadjuvant chemotherapy followed by operation Table 2 demonstrated the characteristics, and compared surgery details as well as complication or mortality due to surgery for group 1 and 3 patients. Group 1 patients had a significantly elder age ( p ¼ 0.04), more cervical or upper-third cancer ( p ¼ 0.046), and less R0 resection result ( p ¼ 0.001). As for surgery-related complications, group 1 showed a significantly more pulmonary complication ( p ¼ 0.006) and anastomosis leakage ( p ¼ 0.002). The hospital mortality of group 1 was also much higher than that of group 2 ( p ¼ 0.03). The operation time, blood loss during surgery, and days of hospital stay were similar in these two groups. Regarding the six mortality cases in group 1, three died of sepsis resulted from intrathoracic anastomosis leakage (at 21, 28, 48 days after surgery, respectively). The other 3 died due to acute respiratory distress syndrome and pneumonia (at 8, 14, and 23 days after surgery, respectively). The 5-year OS of group 1 was similar to that of group 3 (Fig. 2). Risk factors for hospital mortality in patients receiving salvage surgery Table 3 showed the separate results of univariate analysis for preoperative and perioperative risk factors for
291
Table 2 Characteristics and surgery-related outcomes of patients receiving salvage surgery (group 1) or neoadjuvant chemotherapy followed by operation (group 3)
Gender (%) Male Female Median age [year] Mean preoperative albumin level [g/dL] Tumor location (%) Cervical Upper 1/3 Middle 1/3 Lower 1/3 Initial clinical stage (%) Stage IIa Stage IIb Stage III Stage IV Conduit use (%) Stomach Colon Resection result (%) R0 Non-R0 Operation time [min] Blood loss [ml] Pulmonary complication (%) Yes No Anastomosis leakage (%) Yes No Hospital mortality (%) Yes No Hospital stay [day]
Group 1 (n ¼ 27)
Group 3 (n ¼ 191)
p value
26 1 62.44 3.881
188 3 54.52 3.982
0.4
1 7 14 5
4 36 104 47
0.046
4 7 14 2
25 49 92 25
0.58
20 7
167 24
0.077
17 10 370 335
161 30 350 330
0.001
9 18
22 169
0.006
4 23
2 191
0.002
6 21 22.4
15 176 20.1
0.03
0.04 0.368
0.35 0.38
0.3
hospital mortality. The levels of preoperative albumin ( p ¼ 0.004) and hemoglobin ( p ¼ 0.003), and anastomosis leakage ( p ¼ 0.025) were noted to be significantly associated with increased mortality following esophageal resection. After multivariate analysis, anastomosis leakage was the only independent significant perioperative risk factor ( p ¼ 0.021). As for preoperative prognosticators, only the albumin level showed borderline significance ( p ¼ 0.069) in multivariate analysis. Prognostic predictors of disease-specific survival in patients receiving salvage surgery
Figure 1. Overall survivals for group 1 and group 2 patients.
After excluding the six hospital mortality cases, 21 patients entered survival analysis. Univariate analysis revealed the following postoperative factors were significantly related to increased survival: superficial depth of tumor invasion (T1e2, p ¼ 0.0082), early pathologic stage (T1e2N0, p ¼ 0.0035) and R0 resection ( p ¼ 0.0002). While multivariate analysis revealed that R0 resection
Y.K. Chao et al. / EJSO 35 (2009) 289e294
292
Table 4 Cox proportional hazard model for overall survival of patients who received salvage surgery Risk factors
Resection Non-R0 R0 Depth of tumor invasion T1e2 T3e4 Pathologic stage Early Advanced
Overall survival OR (95% CI)
p value
Baseline 4.92 (1.08 w 22.4)
0.039
Baseline 2.83 (0.26 w 30.5)
0.392
Baseline 8.5 (0.75 w 95)
0.083
OR, odds ratio; 95% CI, 95% confidence interval; Early, T1e2N0; Advanced, T3e4 or N1.
Figure 2. Overall survivals for Group 1 and Group 3 patients.
was the most important prognosticator for overall survival ( p ¼ 0.039, HR: 4.919) (Table 4, Fig. 3). Discussion Survival benefit of salvage esophagectomy Definitive chemoradiotherapy is considered a standard treatment for locally advanced esophageal cancer. Despite of long-term survival of 15e20%1e3 after such treatment, locoregional recurrence is not uncommon and occurs in around 40e60% of patients.1e3 Prognosis after locoregional recurrence is dismal and all patients will die in one year without treatment.7 Treatment options for recurrent or persistent disease after definitive chemoradiotherapy are limited. Further radiotherapy is often contraindicated because the maximal tolerable dose has been achieved. As for salvage chemotherapy, the treatment outcome is
disappointing with a median survival of 7 months.8 Salvage surgery is another choice but its role still remains debatable. In this study, we compared two groups of patients who received salvage surgery and chemotherapy/supportive treatment, respectively. The baseline characteristics for these 2 groups were similar. We noted that patients receiving salvage surgery had a significantly higher OS than those underwent chemotherapy only ( p ¼ 0.007). A high 5-year OS of 35.71% can be achieved in group 1 patients. In contrast, all patients died within 1.5 year without surgical treatment. This finding highlights the importance of salvage surgery in the management of recurrent esophageal cancer after chemoradiotherapy. A potential bias in this analysis is that group 2 is a mixed cohort consisting of patients who received chemotherapy or underwent support care. We did not analyze them separately because the reported survivals for these 2 kinds of patients were similar and some patients did give up chemotherapy during treatment.
Table 3 Univariate analysis of preoperative and perioperative risk factors for hospital mortality Risk factors
p value
Preoperative risk factor Albumin level Hemoglobin level FEV1 Tumor location Persistent/recurrent disease
0.004 0.003 0.898 0.703 0.385
Perioperative risk factor Operation time Operation blood loss Conduit use (stomach/colon) Anastomosis site (neck/chest) R0 resection Pneumonia Anastomosis leakage
0.18 0.447 0.594 0.662 0.16 0.367 0.025
Figure 3. Disease specific survivals of patients achieving R0 versus NonR0 resection.
Y.K. Chao et al. / EJSO 35 (2009) 289e294
In this context, we recommend that esophageal cancer patients who had locoregional recurrence after chemoradiotherapy should be encouraged to receive surgical resection if feasible.
293
distal end of the graft using the ileocolic artery and vein and cervical vessels has also been suggested for reducing leakage.13 Patient selection for salvage esophagectomy
Risk of salvage esophagectomy and prognosticator for surgical mortality Although patients with recurrent/persistent esophageal cancer had a higher survival rate after salvage surgery, another concern is the surgical mobility and mortality. Similar to the previous studies,4e6 we also enrolled another group of patients who undergo neoadjuvant chemotherapy followed by surgery to assess the surgery-related complications. Some may argue for the comparability of these two groups because difference in pre-surgery conditions such as previous radiation dose, time interval after radiation, tumor location, or complication during surgery. But these data actually are important reference for consultation of patients because some of them do refuse further operation after neoadjuvant chemoradiotherapy. In this study, we found group 1 patients had a significantly more pulmonary complication, anastomostic leakage and subsequently higher hospital mortality than group 3, which can be explained by the aggressive nature of recurrent disease and complicated surgical approach. The ratios of these surgery-related complications in our study are similar to those of other series.4e6 Salvage surgery for esophageal cancer after chemoradiation has been reported to associate with a high hospital mortality rate of 8 w 15%.4e6 In our study, patients who received salvage surgery also showed a high hospital mortality rate of 22.2%. Thus, the importance of finding a reliable preoperative prognosticator to optimize the use of salvage surgery can not be overemphasized. In this study, we noted a lower preoperative blood albumin or hemoglobin level was associated with higher hospital mortality. Hypoalbuminemia is a well-known predictor for complication of standard esophagectomy.9 It has been reported that when serum albumin is below 3.2 g/dL, esophagectomy will result in more complication and hospital mortality than other types of surgery.10 This may be due to the fact that hypoalbuminemia reflects not only a poor nutrition status but also weak immune condition with high susceptibility of infection. In this context, the preoperative albumin and hemoglobulin levels should be seriously regarded before proceed to surgery. Intrathoracic anastomosis leakage-related mortality is an important cause of death in the current series. The three fetal leakages in our series occurred several weeks after surgery and were believed to be due to poor gastric perfusion as a result of high exposure of the proximal stomach to radiation.11 Using the radiation-sparing colon as the reconstruction substitute and changing into cervical anastomosis via the substernal route may be helpful for preventing lethal leakage.12 An additional microvascular anastomosis at the
R0 resection has been reported to be the most important good prognosticator in the postoperative setting for esophageal cancer patients who receive either radical surgery or neoadjuvant chemoradiotherapy followed by operation.4,6 R0 resection in our series also represented as the most significant risk factor for overall survival. Precise evaluation of respectability before operation is difficult. Conventional CT scan is incorrect for T factor determination after chemoradiation due to thickened esophageal wall and an unclear plane with adjacent mediastinal structure after high dose radiotherapy.14 This is also why we did not include the preoperative stage in the survival analysis. Piessen and colleagues had reported a high R0 resection rate of 81.4% if the salvage surgery was performed for tumor with: (i) height <5 cm on esophagogram; and (ii) aortic contact <90 degrees on computed tomography.15 The main patient group was focused on non-responder after induction chemoradiotherapy. The value of those indicators might need further investigation in the setting of local failure after definitive chemoradiotherapy. Conclusions Salvage surgery provides survival benefit for esophageal cancer patients with locoregional recurrent disease but may lead to serious surgery-related complication and hospital mortality. Preoperative albumin and hemoglobulin levels are important risk factors for hospital mortality and may aid in selecting suitable patient for salvage surgery. Conflict of interest We hereby declare that there is no potential or actual personal, financial or political interest related to this article.
References 1. Bedenne L, Michel P, Bouche O, et al. Chemoradiation followed by surgery compared with chemoradiation alone in squamous cancer of the esophagus: FFCD 9102. J Clin Oncol 2007;25:1160–8. 2. Stahl M, Stuschke M, Lehmann N, et al. Chemoradiation with and without surgery in patients with locally advanced squamous cell carcinoma of the esophagus. J Clin Oncol 2005;23:2310–7. 3. Minsky BD, Pajak TF, Ginsberg RJ, et al. INT 0123 (Radiation Therapy Oncology Group 94-05) phase III trial of combined-modality therapy for esophageal cancer: high-dose versus standard-dose radiation therapy. J Clin Oncol 2002;20:1167–74. 4. Swisher SG, Wynn P, Putnam JB, et al. Salvage esophagectomy for recurrent tumors after definitive chemotherapy and radiotherapy. J Thorac Cardiovasc Surg 2002;123:175–83.
294
Y.K. Chao et al. / EJSO 35 (2009) 289e294
5. Nakamura T, Hayashi K, Ota M, et al. Salvage esophagectomy after definitive chemotherapy and radiotherapy for advanced esophageal cancer. Am J Surg 2004;188:261–6. 6. Tomimaru Y, Yano M, Takachi K, et al. Factors affecting the prognosis of patients with esophageal cancer undergoing salvage surgery after definitive chemoradiotherapy. J Surg Oncol 2006; 93:422–8. 7. Shimada H, Kitabayashi H, Nabeya Y, et al. Treatment response and prognosis of patients after recurrence of esophageal cancer. Surgery 2003;133:4–31. 8. Govindan R, Read W, Faust J, et al. Phase II study of docetaxel and irinotecan in metastatic or recurrent esophageal cancer: a preliminary report. Oncology (Williston Park) 2003;17(9 Suppl. 8): 27–31. 9. Steyerberg EW, Neville BA, Koppert LB, et al. Surgical mortality in patients with esophageal cancer: development and validation of a simple risk score. J Clin Oncol 2006;24:4277–84.
10. Kudsk KA, Tolley EA, DeWitt RC, et al. Preoperative albumin and surgical site identify surgical risk for major postoperative complications. JPEN J Parenter Enteral Nutr 2003;27:1–9. 11. Urschel JD. Esophagogastrostomy anastomotic leaks complicating esophagectomy: a review. Am J Surg 1995;169:634–40. 12. Blewett CJ, Miller JD, Young JE, et al. Anastomotic leaks after esophagectomy for esophageal cancer: a comparison of thoracic and cervical anastomoses. Ann Thorac Cardiovasc Surg 2001;7:75–8. 13. Sakuraba M, Kimata Y, Hishinuma S, et al. Importance of additional microvascular anastomosis in esophageal reconstruction after salvage esophagectomy. Plast Reconstr Surg 2004;113:1934–9. 14. Jones DR, Parker Jr LA, Detterbeck FC, et al. Inadequacy of computed tomography in assessing patients with esophageal carcinoma after induction chemoradiotherapy. Cancer 1999;85:1026–32. 15. Piessen G, Briez N, Triboulet JP, et al. Patients with locally advanced esophageal carcinoma nonresponder to radiochemotherapy: who will benefit from surgery? Ann Surg Oncol 2007;14:2036–44.