Complete Response to Neoadjuvant Chemoradiotherapy in Esophageal Carcinoma Is Associated With Significantly Improved Survival
http://www.100md.com
《临床肿瘤学》
the Departments of Surgical Oncology, Medical Oncology, and Radiation Oncology, and Division of Biostatistics, Fox Chase Cancer Center, Philadelphia, PA
ABSTRACT
PURPOSE: Attempts to improve survival of patients with esophageal cancer have been made using induction chemoradiotherapy (CRT) followed by surgery. A large single-center experience was reviewed to determine which treatment-related variables could predict survival and recurrence.
PATIENTS AND METHODS: All patients undergoing esophagectomy between January 1994 and December 2002 were reviewed. Univariate and multivariate analyses were performed using log-rank and Cox proportional hazards models, and survival curves were estimated using the Kaplan-Meier method.
RESULTS: Of 171 patients with invasive cancer, 131 (77%) underwent preoperative CRT. The average age was 60 years, and most patients were male (85%). Operations performed included Ivor-Lewis (60%), transhiatal (8%), three-hole (23%), or left thoracoabdominal (8%) esophagectomy. Perioperative mortality rate was 5%. Median overall survival (OS) of the entire group was 33 months, and the 5-year OS rate was 26%. Induction CRT was associated with a 33% 5-year survival rate compared with 11% for surgery alone (P = .43). Patients downstaged to pathologic stage 0 or I had an improved OS and disease-free survival (DFS) compared with those patients who were not downstaged (P = .022). Additionally, the ability to perform an R0 resection was a significant factor for OS and DFS (n = 130; P < .0001 and P <.0002, respectively).
CONCLUSION: Response to CRT and the ability to perform an R0 resection are associated with significantly improved survival in patients with esophageal carcinoma.
INTRODUCTION
Although relatively rare, esophageal carcinoma is increasing in frequency and is one of the most fatal malignancies in humans. It was estimated that in 2003 there were 13,900 new cases of esophageal cancer and 13,100 deaths in the United States.1 Although surgery remains the treatment of choice for patients with resectable and localized disease, systemic and local recurrences are still common, and 5-year survival rates range from 15% to 39%.2 Many physicians administer combinations of preoperative chemotherapy and radiation to downstage the primary tumor, thus increasing resectability rates as well as eliminating micrometastases and prolonging survival. However, despite a large number of phase II and III trials, the role of multimodality therapy remains unclear.
There has been only one phase III trial that has demonstrated a survival advantage to induction chemoradiotherapy (CRT) in patients with esophageal cancer.3 Conversely, several other randomized trials have failed to demonstrate an improvement for patients receiving induction CRT plus surgery versus surgery alone.4-7 However, many of these trials have identified several factors that may contribute to improved survival in patients undergoing esophagectomy; these include complete response to neoadjuvant CRT, completeness of resection, and size and location of tumor. We retrospectively reviewed our experience with 179 patients undergoing esophagogastrectomy to better understand the impact of induction therapy on survival after esophagectomy and attempt to determine which factors may contribute to improvements in overall survival (OS) and recurrence-free survival.
PATIENTS AND METHODS
All patients who underwent esophagectomy at Fox Chase Cancer Center (FCCC) between January 1994 and December 2002 were retrospectively reviewed. One hundred seventy-nine patients were identified; eight patients had high-grade dysplasia and were excluded from all survival analyses. Of the remaining patients, 140 (78%) had adenocarcinoma of the esophagus (middle or distal third), gastroesophageal junction, or cardia of the stomach. Thirty-one patients (17%) had squamous cell carcinoma of the proximal (n = 2), middle (n = 12), or distal third (n = 17) of the esophagus. Of the 171 patients with invasive cancer, 131 (77%) underwent preoperative CRT; these patients form the basis of this study.
Patient medical records were analyzed, and information was placed into an esophageal surgery database after approval by the FCCC Institutional Review Board with a waiver of consent. Data collected included preoperative factors (age, comorbidities, symptoms, and presence of Barrett's esophagus), treatment factors (type of chemotherapy, dose of radiation therapy, weight loss, and need for gastrostomy or jejunostomy feedings), tumor factors (histology, clinical stage, pathologic stage, and completeness of resection—R0, R1, or R2), operative and hospital course factors (type of operation performed, blood loss, type of anastomosis, and complications), and tumor recurrence and long-term survival.
The standard induction regimen consisted of external-beam radiotherapy (to a total dose of 45 Gy unless otherwise indicated) with two cycles of concurrent fluorouracil (FU) and cisplatin. Twenty-seven patients were treated on a phase I protocol at FCCC involving continuous-infusion FU and weekly cisplatin and paclitaxel administration in conjunction with 60 Gy of radiation.8 A recovery period of 4 to 6 weeks after induction therapy was mandated before esophageal resection. This period included restaging with computed tomography (CT) scan and, often, repeat esophagogastroduodenoscopy.
All patients were staged pre- and postoperatively according to the tumor-node-metastasis classification of the American Joint Committee for Cancer Staging 5th edition.9 Pretreatment clinical staging was performed using CT scan and, more recently, endoscopic ultrasound (EUS) and positron emission tomography (PET) scan. Patients with no residual, viable tumor cells in the surgical specimen (T0N0M0, stage 0-Rx) were classified as having a pathologic complete response (pCR). There were a large number of patients with minimal residual tumor; these patients (T1N0M0) were classified as having been downstaged (stage I-Rx). Follow-up consisted of examinations every 3 months for the first 2 years and then every 6 months subsequently. Most patients underwent only routine chest x-ray examination. Other imaging tests, such as CT scans, were prompted by patient symptomatology and were ordered at the discretion of the attending surgical, medical, or radiation oncologist. Follow-up data were obtained from patient medical records, referring physicians, and telephone interviews.
Operative mortality was defined as patient death within the first 30 days postoperatively or during the initial hospitalization. OS was estimated using the Kaplan-Meier method from the date of diagnosis to date of death or last follow-up. Disease-specific survival was not calculated. For survival curves, if no events (death or recurrence) happened after 5 years, we could not estimate the 5-year survival rate using the Kaplan-Meier method. Therefore, the survival used was the survival rate at the last time point before 5 years. The influence of variables on survival and recurrence was analyzed using the log-rank test for discrete variables and the Cox proportional hazards model for continuous variables. All statistical tests were two-sided, with significance defined as P < .05. All analyses were performed using SAS software (SAS Institute Inc, Cary, NC).
RESULTS
There were 157 men and 22 women. The average age was 60 years (range, 24 to 87 years). All patients underwent preoperative staging with CT scans of the chest and abdomen. Thirty-six patients underwent preoperative staging with EUS, which became prevalent in the latter years of this review. Preoperative staging is detailed in Table 1. The highest percentage of patients had stage IIA esophageal carcinoma. There were few patients with early disease (stage 0 or I). The majority of patients (n = 117) who underwent induction therapy had clinical stage IIA (T2-3N0M0), IIB (T2N1M0) or III disease (T3N1M0). The preoperative stage in patients not undergoing induction CRT included eight patients with carcinoma-in-situ, 15 patients with stage I disease, 15 patients with stage II disease, four patients with stage III disease, and one patient with recurrent stage IV squamous cell carcinoma of the upper esophagus who underwent palliative esophagectomy. Because the FCCC is a referral center for many community hospitals, several patients received their CRT at local centers for logistic reasons. Sixty-two patients (47%) underwent therapy at FCCC, whereas 69 (53%) had their preoperative therapy at outside locations. There was not a significant difference in survival between these two groups of patients. The most common regimen was FU and cisplatin with 45 Gy of external-beam irradiation (n = 65). All regimens contained FU; however, several other agents were combined with FU depending on where patients received their induction therapy. These agents included mitomycin, carboplatin, docetaxel, doxorubicin, and methotrexate.
Surgical therapy was dictated by location of tumor and surgeon preference. The most common operation performed was an Ivor-Lewis esophagectomy in 106 patients (60%). Other operations included three-hole esophagectomy (thoracic, abdominal, and cervical incisions) in 42 patients (23%), left thoracoabdominal esophagectomy in 15 patients (8%), and transhiatal esophagectomy in 14 patients (8%). All patients underwent an extensive regional lymphadenectomy whenever possible. Patients with middle and lower third esophageal carcinomas underwent mediastinal lymph node sampling (including levels 7, 8, and 9) as well as dissection of perigastric and celiac lymph nodes. Patients with gastroesophageal junction and gastric cardia tumors had periesophageal, perigastric, celiac, and hepatoduodenal lymph nodes sampled. Pyloroplasty or pyloromyotomy were not routinely used; the decision was based on preoperative gastroscopy and intraoperative assessment of the pylorus. Esophageal anastomoses were accomplished with either one layer of hand-sewn interrupted silk sutures or with an EEA stapler (US Surgical Corporation, Norwalk, CT). All patients underwent routine placement of a jejunostomy tube at the time of the definitive operation, if one was not already present.
Patients were maintained without oral intake and with nasogastric suction for the first 5 to 7 postoperative days. Gastrograffin (and/or thin barium) swallow examinations were routinely performed on postoperative days 5 to 7. If this study was negative, the nasogastric tube was removed, and an oral diet was commenced. Chest tubes and/or drains remained in place until patients were tolerating oral diet with no evidence of leakage. A leak was defined as a positive swallow study or clinical evidence of leak, including empyema, increasing pleural effusion, sepsis, or change in character of the drain effluent.
A major complication was recorded as any complication requiring additional intervention, such as medication or a procedure. In all, major complications occurred in 57% of patients; there was no difference in overall complication rate between induction and noninduction groups. Major complications are listed in Table 2. The perioperative mortality rate was essentially the same between both groups, and overall, it was 5%. The only complications that were more frequent in the induction CRT group versus the noninduction group were deep vein thrombosis or pulmonary embolism (8.4% v 0%, respectively) and vocal cord paresis (7.6% v 2.1%, respectively). Postoperative vocal cord dysfunction was present in 11 patients. Reasons for reoperation included tracheostomy in seven patients, mediastinitis in four patients, chylothorax in two patients, and tracheogastric fistula, bowel obstruction, gastric outlet obstruction, empyema, and bowel infarction in one patient each. The vast majority of leaks were managed conservatively with chest tube drainage, nasogastric tube decompression, and nutritional support using jejunostomy feedings or total parenteral nutrition. The median length of hospital stay was 13 days (range, 5 to 79 days; mean, 17 days). The median hospital stay was similar in patients who had complications compared with those who did not (12 days v 13 days, respectively) and in patients who underwent induction CRT versus those who did not (12 days v 13.5 days, respectively; P > .05).
On examination of the resected specimens (Table 3), a pCR was achieved in 42 patients (32% receiving CRT). Another 13 patients (10%) had a significant tumor response with only minimal tumor remaining and, thus, were downstaged to stage I-Rx. Seventy-six patients (58%) had either no change or an increase from their preoperative stage to the final, pathologic stage. The median time from the end of induction therapy to surgery was 43 days (range, 32 to 319 days); this data was not available for the approximately 50% of patients who underwent induction therapy outside of FCCC. There were six patients treated early in our experience with definitive CRT or who refused surgery after induction therapy who went on to develop local recurrence or persistent disease. These patients had a much longer interval between the end of CRT and surgery.
Follow-up was complete in all patients and ranged from 1 week to 99 months (median, 14 months). The median survival for the entire group was 33 months, with a 5-year OS rate of 26%. There was not a significant difference in OS between patients who did and did not receive induction CRT (P = .43), although the 5-year survival rate was 33% in the induction group compared with 11% in the noninduction group (Fig 1). However, those patients who achieved a pCR to induction therapy had a significantly improved survival compared with patients who did not. The 5-year survival rate and median OS time for patients with a pCR were 48% and 50 months, respectively, compared with 18% and 28 months, respectively, for those patients who had residual tumor (P = .015; Fig 2A). Additionally, the 13 patients who had minimal residual tumor had an improved survival compared with patients who did not have any pathologic response. These patients had a median survival time of 49 months, with a 5-year survival rate of 34% (P = .02; Table 4).
The median disease-free survival time (DFS) for all patients was 51 months, with a 5-year DFS rate of 49%. The recurrence-free survival rates were similar for induction and noninduction groups (50% and 46% at 5 years, respectively; P = .94). The response to induction CRT had a dramatic influence on recurrence. Patients who achieved a pCR had a 62% recurrence-free survival rate at 5 years compared with 31% for patients who had residual stage II to IV disease (P = .02; Fig 2B). However, patients with minimal residual tumor had an intermediate DFS rate of 42% at 5 years (Table 4).
Another factor that had a dramatic impact on OS and DFS was the ability to perform a complete (R0) resection. R0 resection was possible in 145 patients (85%) with invasive cancer. The median OS time and 5-year OS rate for patients undergoing R0 resection were 41 months and 35%, respectively, compared with 19 months and 10%, respectively, for patients undergoing less than R0 resection (P < .0001; Fig 3). DFS was also significantly improved in patients undergoing an R0 resection (P = .0002; Table 4).
Overall, 21 patients (12%) developed a locoregional recurrence, and 41 patients (21%) developed distant recurrence. Recurrence rates were similar between patients who did and did not undergo neoadjuvant therapy. In the induction group, 12% and 21% of patients had locoregional and distant recurrences, respectively. In patients who did not have preoperative therapy, the rates of locoregional and distant recurrences were 12% and 30%, respectively. There were no major differences in recurrence patterns between patients who achieved a pCR and those who did not; the rate of locoregional recurrence in the two groups was 12% and 8%, respectively, whereas the rate of distant recurrence was 17% and 22%.
DISCUSSION
Esophagectomy continues to be the standard treatment of patients with localized, resectable esophageal carcinoma. Despite recent improvements in technique and perioperative management of patients undergoing esophagectomy, systemic and local recurrences are common, and 5-year survival continues to be poor, with rates of 15% to 39%.2 To improve survival, many institutions have adopted induction CRT protocols. The goal is to increase the overall number of patients receiving CRT and, thus, improve DFS and OS after esophagectomy.
Despite encouraging results from many phase II trials, the results of phase III experiences have been disappointing. There has been only one randomized, controlled trial of induction CRT plus surgery versus surgery alone that has shown an advantage to induction CRT plus surgery. In this trial, Walsh et al3 randomized 113 patients with esophageal adenocarcinoma. Patients in the induction CRT arm received FU and cisplatin for two cycles concurrent with 40 Gy of radiotherapy. The median survival time for the multimodality group was 16 months compared with 11 months for the surgery alone group. Additionally, 3-year survival was significantly prolonged for the induction group versus the noninduction group (32% v 6%, respectively). This remains the only trial that has demonstrated a survival advantage to preoperative CRT; however, the survival rate for the surgery alone group is exceptionally low compared with historical controls. There have been other phase III trials reported that have not demonstrated a survival advantage to preoperative CRT.4-7,9,10 These trials are listed in Table 5.
Our data do not indicate a statistically significant survival benefit for patients undergoing neoadjuvant CRT, although 5-year survival rates are somewhat higher for the induction CRT patients versus the noninduction patients (33% v 11%, respectively). Several factors may have contributed to these results. This was not a prospective, randomized group of patients, and thus, selection bias is unavoidable. In fact, the majority of patients in the noninduction group had early-stage disease. Additionally, the groups were not equally matched; 131 patients underwent induction CRT compared with 48 patients who did not.
The survival of our patients who underwent induction CRT compares favorably with the survival reported in the literature. The median survival time of 33 months and 5-year survival rate of 33% is similar to that recently reported by the Johns Hopkins group.11 These authors reported a median survival time of 35 months and a 5-year survival rate of 40% for patients undergoing preoperative CRT with FU, cisplatin, and 44 Gy of radiotherapy. In our series, median DFS time had not yet been reached, and the 5-year DFS rate was 50%. This is also similar to the Hopkins group, which reported a median DFS time and 5-year DFS rate of 59 months and 49%, respectively.11
One of the more important factors in determining survival of patients undergoing esophagectomy is the completeness of resection. We performed a complete (R0) resection of all cancer in 85% of our patients; this number was not significantly different between patients who did and did not receive induction CRT. These patients had a significantly improved OS and DFS compared with patients who had less than an R0 resection. Our R0 resection rate compares favorably to those reported in the literature, which are typically above 80%.12,13 In a recent review of the M.D. Anderson experience, patients who had an R0 resection had a significantly improved OS compared to those who did not—5-year survivals of 30% versus less than 10%.12 In another report by Mariette et al,13 the 5-year survival of patients undergoing R0 resection was 47% compared to 4% in the R1/R2 group. These results compare favorably to our experience and substantiate the importance of a complete R0 resection. Probably the single most important prognostic factor in many reports of neoadjuvant CRT is the pathologic response to therapy. Most reports have demonstrated a significant improvement in survival for those patients who do not have evidence of residual tumor on pathologic examination (pCR).5,7,11 pCRs vary from 17% to 51% with most large randomized trials demonstrating rates of 10% to 30% (Table 5). Our pCR rate of 32% is consistent with these trials.
An important aspect of the current report is the fate of those patients who do not have significant downstaging of their tumor. The 5-year survival of this group of 76 patients was only 15% with a median survival of 25 months. In the RTOG 85-01 trial,14 in which nonsurgical patients were randomly assigned to receive 64 Gy of radiation alone or 50 Gy of radiation with concurrent cisplatin and 5-FU, patients who received combined therapy had a 26% 5-year survival rate. Because the 5-year survival of patients who do not respond to induction CRT is relatively poor (15% in our series), one could make the argument that patients who do not have a significant response to induction CRT may not benefit from surgery.15
However, a difficulty lies in determining which patients have had a significant response to therapy. The most accurate method of determining treatment effect has always been esophagectomy. CT scan and EUS are notoriously inaccurate in the restaging of patients who have undergone induction therapy. In a study by Jones et al,16 postinduction CT scanning accurately staged the T classification in 42% of patients, with overstaging in 36% and understaging in 20% of patients. On the other hand, EUS cannot distinguish tumor involvement of the esophageal wall and lymph nodes from the postinflammatory changes that characterize effective neoadjuvant treatment. In one study, EUS accurately predicted T stage in 27% and N stage in 58% of patients, and seven patients with a pCR were significantly overstaged.17
Several studies have examined PET scan in this capacity. In a recent study, Downey et al18 found that those patients with a greater than 60% decrease in the standardized uptake value had a significantly increased 2-year DFS and OS compared with patients who did not. Another group correlated major PET response to induction CRT with histopathology obtained during esophagectomy.19 In this series, the sensitivity of serial PET for a major CRT response was 71%, and the specificity was 82%. The median survival time of major PET responders after CRT was significantly increased at 16.3 months compared with 6.4 months for patients who did not respond. Therefore, fluorodeoxyglucose-PET shows some promise for restaging after induction CRT but needs to be further tested in larger groups of patients undergoing clinical trials. At some point in the future, this may prevent those patients who have not had a significant tumor response from undergoing surgery that may not be beneficial to them. However, at the present time, PET scans are not accurate enough at determining the presence and amount of residual disease after induction CRT. There are several patients who have a complete PET response with minimal residual standardized uptake value activity who have stage III disease in their esophagectomy specimen. Until such time that PET becomes more accurate or is replaced by another modality, surgery cannot be replaced to determine the pathologic staging of esophageal cancer.
In summary, induction CRT did not significantly prolong survival compared with surgery alone in this retrospective analysis. However, those patients who achieve a pCR or have an R0 resection have significantly prolonged survival. Future studies should address the role of PET and other modalities in restaging to determine which patients, after induction CRT, will benefit from surgery versus further chemotherapy. Additionally, more research is needed to determine the biologic differences between responders and nonresponders.
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Honoraria: Walter J. Scott, Aventis. For a detailed description of this category, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section of Information for Contributors found in the front of every issue.
NOTES
Presented in part in poster format at the 1st Annual American Society of Clinical Oncology/Society of Surgical Oncology/American Society for Therapeutic Radiology and Oncology/American Gastroenterological Association Gastrointestinal Symposium, San Francisco, CA, January 22-24, 2004.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Jemal A, Murray T, Samuels A, et al: Cancer statistics, 2003. CA Cancer J Clin 53:5-26, 2003
Refaely Y, Krasna MJ: Multimodality therapy for esophageal cancer. Surg Clin North Am 82:729-746, 2002
Walsh TN, Noonan N, Hollywood D, et al: A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. N Engl J Med 335:462-467, 1996
Bosset JF, Gignoux M, Triboulet JP, et al: Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. N Engl J Med 337:161-167, 1997
Le Prise E, Etienne PL, Mennier B, et al: A randomized study of chemotherapy, radiation therapy, and surgery versus surgery for localized squamous cell carcinoma of the esophagus. Cancer 73:1779-1784, 1994
Apinop C, Puttisak P, Preecha N: A prospective study of combined therapy in esophageal cancer. Hepatogastroenterology 41:391-393, 1994
Urba SG, Orringer MB, Turrisi A, et al: Randomized trial of preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. J Clin Oncol 19:305-313, 2001
Goldberg M, Farma J, Lampert C, et al:Survival following intensive preoperative combined modality therapy with paclitaxel, cisplatin, 5-fluorouracil, and radiation in resectable esophageal carcinoma: A phase I report. J Thorac Cardiovasc Surg 126:1168-1173, 2003
Nygaard K, Hagen S, Hansen HS, et al: Pre-operative radiotherapy prolongs survival in operable esophageal carcinoma: A randomized, multicenter study of pre-operative radiotherapy and chemotherapy—The second Scandinavian trial in esophageal cancer. World J Surg 16:1104-1110, 1992
Burmeister BH, Smithers, BM, Fitzgerald L, et al: A randomized phase III trial of preoperative chemoradiation followed by surgery (CR-S) versus surgery alone (S) for localized resectable cancer of the esophagus. Proc Am Soc Clin Oncol 21:130a, 2002 (abstr 518)
Kleinberg L, Knisely JP, Heitmiller R, et al: Mature survival results with preoperative cisplatin, protracted infusion 5-fluorouracil, and 44-Gy radiotherapy for esophageal cancer. Int J Radiat Oncol Biol Phys 56:328-334, 2003
Hofstetter W, Swisher SG, Correa AM, et al: Treatment outcomes of resected esophageal cancer. Ann Surg 236:376-385, 2002
Mariette C, Finzi L, Fabre S, et al: Factors predictive of complete resection of operable esophageal cancer: A prospective study. Ann Thorac Surg 75:1720-1726, 2003
Herskovic A, Martz K, al-Sarraf M, et al: Combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med 326:1593-1598, 1992
al-Sarraf M, Martz K, Herskovic A, et al: Progress report of combined chemoradiotherapy versus radiotherapy alone in patients with esophageal cancer: An intergroup study. J Clin Oncol 15:277-284, 1997
Jones DR, Parker LA Jr, Detterbeck FC, et al: Inadequacy of computed tomography in assessing patients with esophageal carcinoma after induction chemoradiotherapy. Cancer 85:1026-1032, 1999
Beseth BD, Bedford R, Isacoff WH, et al: Endoscopic ultrasound does not accurately assess pathologic stage of esophageal cancer after neoadjuvant chemoradiotherapy. Am Surg 66:827-831, 2000
Downey RJ, Akhurst T, Ilson D, et al: Whole body 18FDG-PET and the response of esophageal cancer to induction therapy: Results of a prospective trial. J Clin Oncol 21:428-432, 2003
Flamen P, Van Cutsem E, Lerut A, et al: Positron emission tomography for assessment of the response to induction radiochemotherapy in locally advanced oesophageal cancer. Ann Oncol 13:361-368, 2002(Adam C. Berger, Jeffrey F)
ABSTRACT
PURPOSE: Attempts to improve survival of patients with esophageal cancer have been made using induction chemoradiotherapy (CRT) followed by surgery. A large single-center experience was reviewed to determine which treatment-related variables could predict survival and recurrence.
PATIENTS AND METHODS: All patients undergoing esophagectomy between January 1994 and December 2002 were reviewed. Univariate and multivariate analyses were performed using log-rank and Cox proportional hazards models, and survival curves were estimated using the Kaplan-Meier method.
RESULTS: Of 171 patients with invasive cancer, 131 (77%) underwent preoperative CRT. The average age was 60 years, and most patients were male (85%). Operations performed included Ivor-Lewis (60%), transhiatal (8%), three-hole (23%), or left thoracoabdominal (8%) esophagectomy. Perioperative mortality rate was 5%. Median overall survival (OS) of the entire group was 33 months, and the 5-year OS rate was 26%. Induction CRT was associated with a 33% 5-year survival rate compared with 11% for surgery alone (P = .43). Patients downstaged to pathologic stage 0 or I had an improved OS and disease-free survival (DFS) compared with those patients who were not downstaged (P = .022). Additionally, the ability to perform an R0 resection was a significant factor for OS and DFS (n = 130; P < .0001 and P <.0002, respectively).
CONCLUSION: Response to CRT and the ability to perform an R0 resection are associated with significantly improved survival in patients with esophageal carcinoma.
INTRODUCTION
Although relatively rare, esophageal carcinoma is increasing in frequency and is one of the most fatal malignancies in humans. It was estimated that in 2003 there were 13,900 new cases of esophageal cancer and 13,100 deaths in the United States.1 Although surgery remains the treatment of choice for patients with resectable and localized disease, systemic and local recurrences are still common, and 5-year survival rates range from 15% to 39%.2 Many physicians administer combinations of preoperative chemotherapy and radiation to downstage the primary tumor, thus increasing resectability rates as well as eliminating micrometastases and prolonging survival. However, despite a large number of phase II and III trials, the role of multimodality therapy remains unclear.
There has been only one phase III trial that has demonstrated a survival advantage to induction chemoradiotherapy (CRT) in patients with esophageal cancer.3 Conversely, several other randomized trials have failed to demonstrate an improvement for patients receiving induction CRT plus surgery versus surgery alone.4-7 However, many of these trials have identified several factors that may contribute to improved survival in patients undergoing esophagectomy; these include complete response to neoadjuvant CRT, completeness of resection, and size and location of tumor. We retrospectively reviewed our experience with 179 patients undergoing esophagogastrectomy to better understand the impact of induction therapy on survival after esophagectomy and attempt to determine which factors may contribute to improvements in overall survival (OS) and recurrence-free survival.
PATIENTS AND METHODS
All patients who underwent esophagectomy at Fox Chase Cancer Center (FCCC) between January 1994 and December 2002 were retrospectively reviewed. One hundred seventy-nine patients were identified; eight patients had high-grade dysplasia and were excluded from all survival analyses. Of the remaining patients, 140 (78%) had adenocarcinoma of the esophagus (middle or distal third), gastroesophageal junction, or cardia of the stomach. Thirty-one patients (17%) had squamous cell carcinoma of the proximal (n = 2), middle (n = 12), or distal third (n = 17) of the esophagus. Of the 171 patients with invasive cancer, 131 (77%) underwent preoperative CRT; these patients form the basis of this study.
Patient medical records were analyzed, and information was placed into an esophageal surgery database after approval by the FCCC Institutional Review Board with a waiver of consent. Data collected included preoperative factors (age, comorbidities, symptoms, and presence of Barrett's esophagus), treatment factors (type of chemotherapy, dose of radiation therapy, weight loss, and need for gastrostomy or jejunostomy feedings), tumor factors (histology, clinical stage, pathologic stage, and completeness of resection—R0, R1, or R2), operative and hospital course factors (type of operation performed, blood loss, type of anastomosis, and complications), and tumor recurrence and long-term survival.
The standard induction regimen consisted of external-beam radiotherapy (to a total dose of 45 Gy unless otherwise indicated) with two cycles of concurrent fluorouracil (FU) and cisplatin. Twenty-seven patients were treated on a phase I protocol at FCCC involving continuous-infusion FU and weekly cisplatin and paclitaxel administration in conjunction with 60 Gy of radiation.8 A recovery period of 4 to 6 weeks after induction therapy was mandated before esophageal resection. This period included restaging with computed tomography (CT) scan and, often, repeat esophagogastroduodenoscopy.
All patients were staged pre- and postoperatively according to the tumor-node-metastasis classification of the American Joint Committee for Cancer Staging 5th edition.9 Pretreatment clinical staging was performed using CT scan and, more recently, endoscopic ultrasound (EUS) and positron emission tomography (PET) scan. Patients with no residual, viable tumor cells in the surgical specimen (T0N0M0, stage 0-Rx) were classified as having a pathologic complete response (pCR). There were a large number of patients with minimal residual tumor; these patients (T1N0M0) were classified as having been downstaged (stage I-Rx). Follow-up consisted of examinations every 3 months for the first 2 years and then every 6 months subsequently. Most patients underwent only routine chest x-ray examination. Other imaging tests, such as CT scans, were prompted by patient symptomatology and were ordered at the discretion of the attending surgical, medical, or radiation oncologist. Follow-up data were obtained from patient medical records, referring physicians, and telephone interviews.
Operative mortality was defined as patient death within the first 30 days postoperatively or during the initial hospitalization. OS was estimated using the Kaplan-Meier method from the date of diagnosis to date of death or last follow-up. Disease-specific survival was not calculated. For survival curves, if no events (death or recurrence) happened after 5 years, we could not estimate the 5-year survival rate using the Kaplan-Meier method. Therefore, the survival used was the survival rate at the last time point before 5 years. The influence of variables on survival and recurrence was analyzed using the log-rank test for discrete variables and the Cox proportional hazards model for continuous variables. All statistical tests were two-sided, with significance defined as P < .05. All analyses were performed using SAS software (SAS Institute Inc, Cary, NC).
RESULTS
There were 157 men and 22 women. The average age was 60 years (range, 24 to 87 years). All patients underwent preoperative staging with CT scans of the chest and abdomen. Thirty-six patients underwent preoperative staging with EUS, which became prevalent in the latter years of this review. Preoperative staging is detailed in Table 1. The highest percentage of patients had stage IIA esophageal carcinoma. There were few patients with early disease (stage 0 or I). The majority of patients (n = 117) who underwent induction therapy had clinical stage IIA (T2-3N0M0), IIB (T2N1M0) or III disease (T3N1M0). The preoperative stage in patients not undergoing induction CRT included eight patients with carcinoma-in-situ, 15 patients with stage I disease, 15 patients with stage II disease, four patients with stage III disease, and one patient with recurrent stage IV squamous cell carcinoma of the upper esophagus who underwent palliative esophagectomy. Because the FCCC is a referral center for many community hospitals, several patients received their CRT at local centers for logistic reasons. Sixty-two patients (47%) underwent therapy at FCCC, whereas 69 (53%) had their preoperative therapy at outside locations. There was not a significant difference in survival between these two groups of patients. The most common regimen was FU and cisplatin with 45 Gy of external-beam irradiation (n = 65). All regimens contained FU; however, several other agents were combined with FU depending on where patients received their induction therapy. These agents included mitomycin, carboplatin, docetaxel, doxorubicin, and methotrexate.
Surgical therapy was dictated by location of tumor and surgeon preference. The most common operation performed was an Ivor-Lewis esophagectomy in 106 patients (60%). Other operations included three-hole esophagectomy (thoracic, abdominal, and cervical incisions) in 42 patients (23%), left thoracoabdominal esophagectomy in 15 patients (8%), and transhiatal esophagectomy in 14 patients (8%). All patients underwent an extensive regional lymphadenectomy whenever possible. Patients with middle and lower third esophageal carcinomas underwent mediastinal lymph node sampling (including levels 7, 8, and 9) as well as dissection of perigastric and celiac lymph nodes. Patients with gastroesophageal junction and gastric cardia tumors had periesophageal, perigastric, celiac, and hepatoduodenal lymph nodes sampled. Pyloroplasty or pyloromyotomy were not routinely used; the decision was based on preoperative gastroscopy and intraoperative assessment of the pylorus. Esophageal anastomoses were accomplished with either one layer of hand-sewn interrupted silk sutures or with an EEA stapler (US Surgical Corporation, Norwalk, CT). All patients underwent routine placement of a jejunostomy tube at the time of the definitive operation, if one was not already present.
Patients were maintained without oral intake and with nasogastric suction for the first 5 to 7 postoperative days. Gastrograffin (and/or thin barium) swallow examinations were routinely performed on postoperative days 5 to 7. If this study was negative, the nasogastric tube was removed, and an oral diet was commenced. Chest tubes and/or drains remained in place until patients were tolerating oral diet with no evidence of leakage. A leak was defined as a positive swallow study or clinical evidence of leak, including empyema, increasing pleural effusion, sepsis, or change in character of the drain effluent.
A major complication was recorded as any complication requiring additional intervention, such as medication or a procedure. In all, major complications occurred in 57% of patients; there was no difference in overall complication rate between induction and noninduction groups. Major complications are listed in Table 2. The perioperative mortality rate was essentially the same between both groups, and overall, it was 5%. The only complications that were more frequent in the induction CRT group versus the noninduction group were deep vein thrombosis or pulmonary embolism (8.4% v 0%, respectively) and vocal cord paresis (7.6% v 2.1%, respectively). Postoperative vocal cord dysfunction was present in 11 patients. Reasons for reoperation included tracheostomy in seven patients, mediastinitis in four patients, chylothorax in two patients, and tracheogastric fistula, bowel obstruction, gastric outlet obstruction, empyema, and bowel infarction in one patient each. The vast majority of leaks were managed conservatively with chest tube drainage, nasogastric tube decompression, and nutritional support using jejunostomy feedings or total parenteral nutrition. The median length of hospital stay was 13 days (range, 5 to 79 days; mean, 17 days). The median hospital stay was similar in patients who had complications compared with those who did not (12 days v 13 days, respectively) and in patients who underwent induction CRT versus those who did not (12 days v 13.5 days, respectively; P > .05).
On examination of the resected specimens (Table 3), a pCR was achieved in 42 patients (32% receiving CRT). Another 13 patients (10%) had a significant tumor response with only minimal tumor remaining and, thus, were downstaged to stage I-Rx. Seventy-six patients (58%) had either no change or an increase from their preoperative stage to the final, pathologic stage. The median time from the end of induction therapy to surgery was 43 days (range, 32 to 319 days); this data was not available for the approximately 50% of patients who underwent induction therapy outside of FCCC. There were six patients treated early in our experience with definitive CRT or who refused surgery after induction therapy who went on to develop local recurrence or persistent disease. These patients had a much longer interval between the end of CRT and surgery.
Follow-up was complete in all patients and ranged from 1 week to 99 months (median, 14 months). The median survival for the entire group was 33 months, with a 5-year OS rate of 26%. There was not a significant difference in OS between patients who did and did not receive induction CRT (P = .43), although the 5-year survival rate was 33% in the induction group compared with 11% in the noninduction group (Fig 1). However, those patients who achieved a pCR to induction therapy had a significantly improved survival compared with patients who did not. The 5-year survival rate and median OS time for patients with a pCR were 48% and 50 months, respectively, compared with 18% and 28 months, respectively, for those patients who had residual tumor (P = .015; Fig 2A). Additionally, the 13 patients who had minimal residual tumor had an improved survival compared with patients who did not have any pathologic response. These patients had a median survival time of 49 months, with a 5-year survival rate of 34% (P = .02; Table 4).
The median disease-free survival time (DFS) for all patients was 51 months, with a 5-year DFS rate of 49%. The recurrence-free survival rates were similar for induction and noninduction groups (50% and 46% at 5 years, respectively; P = .94). The response to induction CRT had a dramatic influence on recurrence. Patients who achieved a pCR had a 62% recurrence-free survival rate at 5 years compared with 31% for patients who had residual stage II to IV disease (P = .02; Fig 2B). However, patients with minimal residual tumor had an intermediate DFS rate of 42% at 5 years (Table 4).
Another factor that had a dramatic impact on OS and DFS was the ability to perform a complete (R0) resection. R0 resection was possible in 145 patients (85%) with invasive cancer. The median OS time and 5-year OS rate for patients undergoing R0 resection were 41 months and 35%, respectively, compared with 19 months and 10%, respectively, for patients undergoing less than R0 resection (P < .0001; Fig 3). DFS was also significantly improved in patients undergoing an R0 resection (P = .0002; Table 4).
Overall, 21 patients (12%) developed a locoregional recurrence, and 41 patients (21%) developed distant recurrence. Recurrence rates were similar between patients who did and did not undergo neoadjuvant therapy. In the induction group, 12% and 21% of patients had locoregional and distant recurrences, respectively. In patients who did not have preoperative therapy, the rates of locoregional and distant recurrences were 12% and 30%, respectively. There were no major differences in recurrence patterns between patients who achieved a pCR and those who did not; the rate of locoregional recurrence in the two groups was 12% and 8%, respectively, whereas the rate of distant recurrence was 17% and 22%.
DISCUSSION
Esophagectomy continues to be the standard treatment of patients with localized, resectable esophageal carcinoma. Despite recent improvements in technique and perioperative management of patients undergoing esophagectomy, systemic and local recurrences are common, and 5-year survival continues to be poor, with rates of 15% to 39%.2 To improve survival, many institutions have adopted induction CRT protocols. The goal is to increase the overall number of patients receiving CRT and, thus, improve DFS and OS after esophagectomy.
Despite encouraging results from many phase II trials, the results of phase III experiences have been disappointing. There has been only one randomized, controlled trial of induction CRT plus surgery versus surgery alone that has shown an advantage to induction CRT plus surgery. In this trial, Walsh et al3 randomized 113 patients with esophageal adenocarcinoma. Patients in the induction CRT arm received FU and cisplatin for two cycles concurrent with 40 Gy of radiotherapy. The median survival time for the multimodality group was 16 months compared with 11 months for the surgery alone group. Additionally, 3-year survival was significantly prolonged for the induction group versus the noninduction group (32% v 6%, respectively). This remains the only trial that has demonstrated a survival advantage to preoperative CRT; however, the survival rate for the surgery alone group is exceptionally low compared with historical controls. There have been other phase III trials reported that have not demonstrated a survival advantage to preoperative CRT.4-7,9,10 These trials are listed in Table 5.
Our data do not indicate a statistically significant survival benefit for patients undergoing neoadjuvant CRT, although 5-year survival rates are somewhat higher for the induction CRT patients versus the noninduction patients (33% v 11%, respectively). Several factors may have contributed to these results. This was not a prospective, randomized group of patients, and thus, selection bias is unavoidable. In fact, the majority of patients in the noninduction group had early-stage disease. Additionally, the groups were not equally matched; 131 patients underwent induction CRT compared with 48 patients who did not.
The survival of our patients who underwent induction CRT compares favorably with the survival reported in the literature. The median survival time of 33 months and 5-year survival rate of 33% is similar to that recently reported by the Johns Hopkins group.11 These authors reported a median survival time of 35 months and a 5-year survival rate of 40% for patients undergoing preoperative CRT with FU, cisplatin, and 44 Gy of radiotherapy. In our series, median DFS time had not yet been reached, and the 5-year DFS rate was 50%. This is also similar to the Hopkins group, which reported a median DFS time and 5-year DFS rate of 59 months and 49%, respectively.11
One of the more important factors in determining survival of patients undergoing esophagectomy is the completeness of resection. We performed a complete (R0) resection of all cancer in 85% of our patients; this number was not significantly different between patients who did and did not receive induction CRT. These patients had a significantly improved OS and DFS compared with patients who had less than an R0 resection. Our R0 resection rate compares favorably to those reported in the literature, which are typically above 80%.12,13 In a recent review of the M.D. Anderson experience, patients who had an R0 resection had a significantly improved OS compared to those who did not—5-year survivals of 30% versus less than 10%.12 In another report by Mariette et al,13 the 5-year survival of patients undergoing R0 resection was 47% compared to 4% in the R1/R2 group. These results compare favorably to our experience and substantiate the importance of a complete R0 resection. Probably the single most important prognostic factor in many reports of neoadjuvant CRT is the pathologic response to therapy. Most reports have demonstrated a significant improvement in survival for those patients who do not have evidence of residual tumor on pathologic examination (pCR).5,7,11 pCRs vary from 17% to 51% with most large randomized trials demonstrating rates of 10% to 30% (Table 5). Our pCR rate of 32% is consistent with these trials.
An important aspect of the current report is the fate of those patients who do not have significant downstaging of their tumor. The 5-year survival of this group of 76 patients was only 15% with a median survival of 25 months. In the RTOG 85-01 trial,14 in which nonsurgical patients were randomly assigned to receive 64 Gy of radiation alone or 50 Gy of radiation with concurrent cisplatin and 5-FU, patients who received combined therapy had a 26% 5-year survival rate. Because the 5-year survival of patients who do not respond to induction CRT is relatively poor (15% in our series), one could make the argument that patients who do not have a significant response to induction CRT may not benefit from surgery.15
However, a difficulty lies in determining which patients have had a significant response to therapy. The most accurate method of determining treatment effect has always been esophagectomy. CT scan and EUS are notoriously inaccurate in the restaging of patients who have undergone induction therapy. In a study by Jones et al,16 postinduction CT scanning accurately staged the T classification in 42% of patients, with overstaging in 36% and understaging in 20% of patients. On the other hand, EUS cannot distinguish tumor involvement of the esophageal wall and lymph nodes from the postinflammatory changes that characterize effective neoadjuvant treatment. In one study, EUS accurately predicted T stage in 27% and N stage in 58% of patients, and seven patients with a pCR were significantly overstaged.17
Several studies have examined PET scan in this capacity. In a recent study, Downey et al18 found that those patients with a greater than 60% decrease in the standardized uptake value had a significantly increased 2-year DFS and OS compared with patients who did not. Another group correlated major PET response to induction CRT with histopathology obtained during esophagectomy.19 In this series, the sensitivity of serial PET for a major CRT response was 71%, and the specificity was 82%. The median survival time of major PET responders after CRT was significantly increased at 16.3 months compared with 6.4 months for patients who did not respond. Therefore, fluorodeoxyglucose-PET shows some promise for restaging after induction CRT but needs to be further tested in larger groups of patients undergoing clinical trials. At some point in the future, this may prevent those patients who have not had a significant tumor response from undergoing surgery that may not be beneficial to them. However, at the present time, PET scans are not accurate enough at determining the presence and amount of residual disease after induction CRT. There are several patients who have a complete PET response with minimal residual standardized uptake value activity who have stage III disease in their esophagectomy specimen. Until such time that PET becomes more accurate or is replaced by another modality, surgery cannot be replaced to determine the pathologic staging of esophageal cancer.
In summary, induction CRT did not significantly prolong survival compared with surgery alone in this retrospective analysis. However, those patients who achieve a pCR or have an R0 resection have significantly prolonged survival. Future studies should address the role of PET and other modalities in restaging to determine which patients, after induction CRT, will benefit from surgery versus further chemotherapy. Additionally, more research is needed to determine the biologic differences between responders and nonresponders.
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Honoraria: Walter J. Scott, Aventis. For a detailed description of this category, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section of Information for Contributors found in the front of every issue.
NOTES
Presented in part in poster format at the 1st Annual American Society of Clinical Oncology/Society of Surgical Oncology/American Society for Therapeutic Radiology and Oncology/American Gastroenterological Association Gastrointestinal Symposium, San Francisco, CA, January 22-24, 2004.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Jemal A, Murray T, Samuels A, et al: Cancer statistics, 2003. CA Cancer J Clin 53:5-26, 2003
Refaely Y, Krasna MJ: Multimodality therapy for esophageal cancer. Surg Clin North Am 82:729-746, 2002
Walsh TN, Noonan N, Hollywood D, et al: A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. N Engl J Med 335:462-467, 1996
Bosset JF, Gignoux M, Triboulet JP, et al: Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. N Engl J Med 337:161-167, 1997
Le Prise E, Etienne PL, Mennier B, et al: A randomized study of chemotherapy, radiation therapy, and surgery versus surgery for localized squamous cell carcinoma of the esophagus. Cancer 73:1779-1784, 1994
Apinop C, Puttisak P, Preecha N: A prospective study of combined therapy in esophageal cancer. Hepatogastroenterology 41:391-393, 1994
Urba SG, Orringer MB, Turrisi A, et al: Randomized trial of preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. J Clin Oncol 19:305-313, 2001
Goldberg M, Farma J, Lampert C, et al:Survival following intensive preoperative combined modality therapy with paclitaxel, cisplatin, 5-fluorouracil, and radiation in resectable esophageal carcinoma: A phase I report. J Thorac Cardiovasc Surg 126:1168-1173, 2003
Nygaard K, Hagen S, Hansen HS, et al: Pre-operative radiotherapy prolongs survival in operable esophageal carcinoma: A randomized, multicenter study of pre-operative radiotherapy and chemotherapy—The second Scandinavian trial in esophageal cancer. World J Surg 16:1104-1110, 1992
Burmeister BH, Smithers, BM, Fitzgerald L, et al: A randomized phase III trial of preoperative chemoradiation followed by surgery (CR-S) versus surgery alone (S) for localized resectable cancer of the esophagus. Proc Am Soc Clin Oncol 21:130a, 2002 (abstr 518)
Kleinberg L, Knisely JP, Heitmiller R, et al: Mature survival results with preoperative cisplatin, protracted infusion 5-fluorouracil, and 44-Gy radiotherapy for esophageal cancer. Int J Radiat Oncol Biol Phys 56:328-334, 2003
Hofstetter W, Swisher SG, Correa AM, et al: Treatment outcomes of resected esophageal cancer. Ann Surg 236:376-385, 2002
Mariette C, Finzi L, Fabre S, et al: Factors predictive of complete resection of operable esophageal cancer: A prospective study. Ann Thorac Surg 75:1720-1726, 2003
Herskovic A, Martz K, al-Sarraf M, et al: Combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med 326:1593-1598, 1992
al-Sarraf M, Martz K, Herskovic A, et al: Progress report of combined chemoradiotherapy versus radiotherapy alone in patients with esophageal cancer: An intergroup study. J Clin Oncol 15:277-284, 1997
Jones DR, Parker LA Jr, Detterbeck FC, et al: Inadequacy of computed tomography in assessing patients with esophageal carcinoma after induction chemoradiotherapy. Cancer 85:1026-1032, 1999
Beseth BD, Bedford R, Isacoff WH, et al: Endoscopic ultrasound does not accurately assess pathologic stage of esophageal cancer after neoadjuvant chemoradiotherapy. Am Surg 66:827-831, 2000
Downey RJ, Akhurst T, Ilson D, et al: Whole body 18FDG-PET and the response of esophageal cancer to induction therapy: Results of a prospective trial. J Clin Oncol 21:428-432, 2003
Flamen P, Van Cutsem E, Lerut A, et al: Positron emission tomography for assessment of the response to induction radiochemotherapy in locally advanced oesophageal cancer. Ann Oncol 13:361-368, 2002(Adam C. Berger, Jeffrey F)