1467 The Significance of Neuroendocrine Differentiation in Adenocarcinoma of the Esophagus and Esophagogastric Junction After Preoperative Chemoradiation Kim L. Wang, MD 1 Qinghua Yang, MD, PhD 1 Karen R. Cleary, MD 1 Stephen G. Swisher, MD 2 Arlene M. Correa, PhD 2 Ritsuko Komaki, MD 3 Jaffer A. Ajani, MD 4 Asif Rashid, MD, PhD 1 Stanley R. Hamilton, MD 1 Tsung-Teh Wu, MD, PhD 1 1 Departments of Pathology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas. 2 Department of Thoracic and Cardiovascular Surgery, The University of Texas M. D. Anderson Cancer Center, Houston, Texas. 3 Department of Radiation Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas. 4 Department of Gastrointestinal Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas. BACKGROUND. Esophageal and esophagogastric junction (EGJ) adenocarcinomas frequently have neuroendocrine (NE) differentiation, but the significance of NE differentiation in patients who have undergone preoperative chemoradiation and resection remains unclear. METHODS. The authors evaluated the presence of NE differentiation in esophageal and EGJ adenocarcinomas by immunohistochemistry for chromogranin A and synaptophysin and evaluated the clinical significance of NE differentiation in 83 patients (10 patients who had a complete tumor response and 73 patients who had residual tumor in resection specimens) who received preoperative chemoradiation. RESULTS. Of 73 patients who had residual tumor after preoperative treatment, 52% showed NE differentiation. The proportion of tumor cells with NE differentiation had increased from 6% 6 18% in pretreatment biopsy specimens to 47% 6 42% (P ¼.00003) in posttreatment resection specimens in 30 patients who had paired pretreatment biopsy and resection specimens available. Disease-free survival (P ¼.002) and overall survival (P ¼.006) were significantly better in patients who had a complete tumor response than in patients who had residual tumor. Among patients who had residual tumor after preoperative chemoradiation, disease-free survival (P ¼.03) and overall survival (P ¼.045) were significantly better in patients who had residual tumor without NE differentiation than in patients who had residual tumor with NE differentiation. In multivariate analysis, the presence of NE differentiation in residual tumor was a prognostic factor for worse disease-free survival (P ¼.02) independent of pathologic stage and extent of residual tumor. CONCLUSIONS. The results from this study suggested that tumor cells with NE differentiation were more resistant to neoadjuvant chemoradiation in patients with esophageal and EGJ adenocarcinomas. The presence of NE differentiation in residual tumor was associated with poor survival after preoperative neoadjuvant therapy. Cancer 2006;107:1467 74. Ó 2006 American Cancer Society. Address for reprints: Tsung-Teh Wu, MD, PhD, Department of Pathology, Unit 85, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030-4096; Fax: (713) 792-4049; E-mail: twu@mdanderson.org Received April 19, 2006; revision received June 19, 2006; accepted June 28, 2006. KEYWORDS: esophagus, adenocarcinoma, neuroendocrine differentiation, chemoradiation. The incidence of adenocarcinomas of the esophagus and esophagogastric junction (EGJ) has increased significantly in the past 2 decades. 1 Because most patients present with advanced-stage disease, the prognosis for patients with esophageal and EGJ adenocarcinomas remains poor. 2,3 Preoperative chemoradiation has improved overall survival and yields a complete response in approximately 25% to 30% of patients who have advanced localized disease. 4,5 Currently, there are no markers available that can predict chemoradiation response ª 2006 American Cancer Society DOI 10.1002/cncr.22179 Published online 5 September 2006 in Wiley InterScience (www.interscience.wiley.com).
1468 CANCER October 1, 2006 / Volume 107 / Number 7 and survival reliably in patients with adenocarcinomas of the esophagus and EGJ. Neuroendocrine (NE) tumors, including carcinoid tumors and high-grade NE carcinoma, are characterized by distinct histologic features and have different clinical behavior compared with adenocarcinomas of the gastrointestinal tract. Esophageal and EGJ adenocarcinomas with NE differentiation have been reported previously. 6,7 In rectal adenocarcinomas, it has been reported that residual tumors demonstrate increased NE differentiation after preoperative chemoradiation. 8 However, there are no detailed studies of the pathologic evaluation of NE differentiation after chemoradiation in patients with esophageal and EGJ adenocarcinomas. In the current study, we evaluated the prevalence and proportion of tumor cells with NE differentiation in patients with esophageal and EGJ adenocarcinomas before and after chemoradiation and associated those parameters with clinicopathologic features and patient survival. MATERIALS AND METHODS Patient Population This study included 83 patients with esophageal or EGJ adenocarcinomas who were treated with preoperative chemoradiation and surgery at The University of Texas M. D. Anderson Cancer Center from 1996 to 2002. Among them, 10 of 83 patients (12%) had no residual tumor in esophagectomy specimens, and 73 of 83 patients (88%) had residual tumor. Pretreatment biopsy specimens (multiple 2-mm mucosal tissue fragments obtained with standard endoscopic biopsy forceps) were available for immunohistochemistry on 10 patients with no residual tumor in esophagectomy specimens and on 30 patients with residual tumor. In addition, 10 patients with esophageal or EGJ adenocarcinoma who underwent surgery alone and who had both biopsy and resection specimens available were used as a control group in the evaluation of NE differentiation. All tumors were graded on the degree of gland formation into well differentiated, moderately differentiated, or poorly differentiated tumors based on the World Health Organization classification of esophageal adenocarcinoma. 9 This study was approved by the University of Texas M. D. Anderson Cancer Center Institutional Review Board. For clinical staging of potential candidates for preoperative chemoradiation, an upper gastrointestinal endoscopic biopsy, endoscopic ultrasound studies, and computed tomography scans of the chest, abdomen, and pelvis were obtained. Tumor was judged as resectable or localized without distant metastasis except for tumors that had celiac lymph node involvement (American Joint Committee on Cancer Stage II through Stage IVA). 10 The regimens for preoperative neoadjuvant therapy varied but, in general, consisted of a 30-day course of 5-fluorouracil, cisplatin, and paclitaxel in combination with radiotherapy in 25 fractions for a total dose of 40 grays (Gy) to 45 Gy. Resection was performed from 4 weeks to 6 weeks after the completion of neoadjuvant therapy. Immunohistochemical Staining NE differentiation was evaluated by immunohistochemical staining for chromogranin A (monoclonal antibody clone PHE5 at 1:4000 dilution; Chemicon, Temecula, CA) and synaptophysin (monoclonal antibody clone Snp88 at 1:50 dilution; BioGenex, San Ramon, CA) in formalin fixed, paraffin embedded tissues from esophagectomy specimens if residual tumor was present and in 40 pretreatment biopsies. For each tumor, NE differentiation was semiquantitated as the percentage of tumor cells with cytoplasmic staining for chromogranin A and/or synaptophysin. For statistical analysis, NE differentiation was defined as any immunoreactivity for chromogranin A and/or synaptophysin in tumor cells. Histologic Evaluation Routine hematoxylin and eosin-stained slides were used to evaluate for the presence of residual tumor in esophagectomy specimens. If no macroscopically identifiable tumor was present, then the entire ulcerated area was sectioned and submitted for histologic evaluation. Residual tumor was staged according to the American Joint Committee on Cancer Tumor, Lymph Node, Metastasis classification system. 10 The extent of residual tumor after chemotherapy was based on previously published grading systems for residual esophageal and gastric carcinomas status postchemotherapy as follows: The percentage of histologically identified, viable tumor was estimated in relation to the macroscopically identifiable, ulcerated tumor bed. 11,12 Four categories were used to evaluate the extent of residual carcinoma: no residual tumor, 1% to 10% residual tumor, 11% to 50% residual tumor, and >50% residual tumor. Statistical Analysis The NE differentiation status in each tumor was correlated with clinicopathologic features and patient survival. The Fisher exact test or Student test was used to determine the significance of the testing groups. Overall survival was calculated from the time of surgery to the time of death from any cause or to the time of last follow-up, at which point, the data were censored. Disease-free and overall survival curves were constructed
NE Differentiation in Esophageal Cancer/Wang et al. 1469 TABLE 1 Demographic and Clinical Characteristics of 83 Patients With Esophageal and EGJ Adenocarcinomas Treated With Pre-operative Chemoradiation and Resection Presence of residual tumor Absent (n 5 10) Present (n 5 73) neuroendocrine differentiation Absent (n 5 35) Present (n 5 38) Male/Female 9/1 34/1 36/2 Mean age 6 SD (years) 55.7 6 11.8 61.3 6 8.5 56.7 6 11.3 Tumor differentiation Moderate 7 15 16 Poor 3 20 22 Tumor location Mid esophagus 0 1 0 Distal esophagus 3 14 19 EGJ 7 20 19 SD indicates standard deviation; NE, neuroendocrine; EGJ, esophagogastric junction. by using the Kaplan Meier method, and the log-rank test was used to evaluate the statistical significance of differences. The prognostic significance of clinical and pathologic characteristics was determined by using univariate Cox regression analysis. Cox proportional hazards models were fitted for multivariate analysis. Statistical analysis was performed by using SPSS software (version 13 for Windows; SPSS Inc., Chicago, IL). Kaplan Meier survival curves were drawn with GraphPad Prism software (version 4 for Windows; GraphPad Software, San Diego, CA). We used a 2-sided significance level of.05 and a power of 0.90 for all statistical analyses. RESULTS Patient Characteristics Demographics of the 83 patients are summarized in Table 1. There was a male predominance (79 of 83 patients were male; 95%). The majority of tumors were located in the distal esophagus (n ¼ 36 tumors) or the EGJ (n ¼ 46 tumors). In 10 patients (12%), there was no residual adenocarcinoma in resection specimens. NE Differentiation in Posttreatment Residual Adenocarcinomas NE differentiation was present in 38 of 73 patients (52%) who had residual adenocarcinoma in posttreatment resection specimens. In addition, NE differentiation was identified by staining for synaptophysin alone in 14 of 38 tumors (37%), by chromogranin A staining alone in 6 of 38 tumors (16%), and by both chromogranin A and synaptophysin staining in 18 of FIGURE 1. Neuroendocrine (NE) differentiation is observed in residual esophageal and esophagogastric junction adenocarcinomas after preoperative neoadjuvant therapy. (A,B) A well to-moderately differentiated adenocarcinoma in resected esophagus (A) shows diffuse NE differentiation in residual tumor cells by synaptophysin immunostaining (B). (C) In another tumor, the residual tumor cells show prominent eosinophilic cytoplasm suggestive of NE differentiation within fibrotic stroma. (D) Chromogranin A immunostaining highlights these residual tumor cells with NE differentiation. 38 tumors (47%). Tumor differentiation was concordant between biopsy and resection specimens in 66 of 73 patients (29 with moderately differentiated tumors and 37 with poorly differentiated tumors) who had residual carcinoma. There was no difference between the presence of NE differentiation and tumor differentiation (16 of 31 of moderately differentiated tumors and 22 of 42 of poorly differentiated tumors; P ¼ 1.0) (Table 1). NE differentiation in >50% of residual tumor cells with was identified in 11 of 24 tumors (46%) by chromogranin A staining and in 17 of 32 tumors (53%) by synaptophysin staining (Fig. 1, Table 2). Pretreatment biopsies were available from 40 patients, and NE differentiation was present in 15 biopsy specimens (38%). NE differentiation was identified by synaptophysin alone in 5 of 40 tumors (13%), by chromogranin A alone in 1 tumor (3%), and by both chromogranin A and synaptophysin in 9 tumors (23%). Adenocarcinoma with from 1% to 5% NE-differentiated tumor cells were identified in 7 of 10 tumors (70%) by chromogranin A staining and in 6 of 14 tumors (43%) by synaptophysin staining (Table 2). The percent of tumor cells with NE differentiation in posttreatment resection specimens increased compared with NE differentiation of tumor in pretreatment biopsies (P ¼.11 for chromogranin A staining; P ¼.01 for synaptophysin staining) (Table 2). In 30 patients, the NE differentiation status was available for tumors both pretreatment and posttreatment.
1470 CANCER October 1, 2006 / Volume 107 / Number 7 TABLE 2 Extent of Neuroendocrine Differentiation in Biopsy and Resected Specimens for Esophageal and EGJ Adenocarcinomas by Chromogranin A and Synaptophysin Tumor cells with neuroendocrine differentiation (%) 0 1 5 6 20 21 50 >50 Pre-treatment tumor biopsy Chromogranin (n ¼ 40) 30 7 2 0 1* Synaptophysin (n ¼ 37) 23 6 5 2 1 y Resected residual tumor Chromogranin (n ¼ 73) 49 8 3 2 11* Synaptophysin (n ¼ 73) 41 8 2 5 17 y * P ¼ 0.11. y P ¼ 0.01. TABLE 3 Comparison of Neuroendocrine Differentiation in 30 Paired Tumor in Pre-treatment Biopsy and Resected Residual Tumor Tumor in pre-treatment biopsy NE differentiation Positive Negative P value Residual tumor in resection (n ¼ 30) Positive 11 10 0.44 Negative 3 6 The prevalence of NE differentiation in residual tumor after treatment in resection specimens was slightly higher (70%; 21 of 30 tumors) compared with tumor in pretreatment biopsies (46%; 14 of 30 tumors) (Table 3). In addition, the proportion of tumor cells with NE differentiation (6standard deviation) was significantly greater in posttreatment resection specimens (47% 6 42%) compared with paired pretreatment biopsy specimens (6% 6 18%; P ¼.00003) (Fig. 2). The presence of NE differentiation in tumor was concordant between pretreatment and posttreatment specimens in 17 of 30 patients (57% of patients) and was discordant in 13 of 30 patients (43%30) (Table 3). In contrast, NE differentiation status in tumor was concordant between biopsy and surgical specimens in 9 of 10 patients (4 patients with NE differentiation and 5 patients without NE differentiation) who underwent surgery alone. Clinical Significance of NE Differentiation after Preoperative Neoadjuvant Therapy There was no association between the presence of NE differentiation in tumors in pretreatment biopsies and the extent of residual tumor, residual tumor stage (ypt), or the presence of metastatic adenocarcinoma in regional lymph nodes in posttreatment resection FIGURE 2. These photomicrographs illustrate increased proportions of tumor cells with neuroendocrine (NE) differentiation in resection specimens after preoperative neoadjuvant therapy. (A) In a pretreatment biopsy specimen, adenocarcinoma shows focal NE differentiation with 2% of tumor cells stained for synaptophysin by immunohistochemistry. (B,C) In the resected esophagus, the residual tumor cells indicate a poorly differentiated adenocarcinoma (B), and synaptophysin immunohistochemistry shows that NE differentiation is present in approximately 100% of residual tumor cells (C).
NE Differentiation in Esophageal Cancer/Wang et al. 1471 TABLE 4 Correlation of Neuroendocrine Differentiation in Pre-treatment Tumor and Residual Tumor Status After Neoadjuvant Therapy Residual tumor Neuroendocrine differentiation Present (n 5 15) Absent (n 5 25) P value ypt Stage 0 3 7 0.61 1 3 7 2 1 3 3 8 8 ypn Stage 0.11 N1 9 8 N0 6 17 Extent of residual tumor 0.63 >50% 6 6 11 50% 1 4 1 10% 5 8 0% 3 7 specimens (Table 4). The presence of NE differentiation in posttreatment resection specimens was correlated with the extent of residual tumor but had no association with residual tumor stage (ypt), lymph node stage (ypn), distant metastasis (ypm), or the presence of Barrett mucosa (Table 5). TABLE 5 Correlation of Neuroendocrine Differentiation in Post-treatment Resection Specimens and Residual Tumor Status After Neoadjuvant Therapy Residual tumor Neuroendocrine differentiation Present (n 5 38) Absent (n 5 35) P value ypt Stage 0.12 0 0 3* 1 12 6 2 4 9 3 22 17 ypn Stage 0.16 N1 23 15 N0 15 20 ypm Stage 0.24 M0 35 35 M1 3 0 Extent of residual tumor 0.005 > 50% 14 13 11 50% 5 12 1 10% 19 7 0% 0 3* Barrett mucosa 0.47 Present 17 12 Absent 21 23 * In 3 cases, no residual tumor was identified in resected esophagectomy specimens and NE differentiation status was determined in residual metastatic tumor present in lymph nodes. Survival Analysis The median potential follow-up was 47.3 months. Patients who had no residual tumor in resected specimens had a significantly better overall survival and disease-free survival than patients who had residual tumor (Fig. 3). For patients who had residual tumor, disease-free survival and overall survival after preoperative chemoradiation were associated with pathologic stage of residual tumor and with the presence of NE differentiation in residual tumor in univariate analysis (Table 6). The presence of NE differentiation in residual tumor after preoperative chemoradiation therapy was associated with a worse outcome. The median disease-free survival and overall survival for patients who had NE differentiation present in residual tumor (11.7 months and 18.0 months, respectively) was significantly worse compared with those in patients without NE differentiation in residual tumor (22.5 months and 33.6 months, respectively; P ¼.03 and P ¼.045, respectively) (Fig. 3). In multivariate analysis, disease-free survival was predicted by NE differentiation status in residual tumor (P ¼.02), and overall survival was predicted by the extent of residual tumor (P ¼.02) (Table 7). There was no difference between pretreatment tumor with or without NE differentiation in predicting disease-free survival (P ¼.53) or overall survival (P ¼.19) after preoperative chemoradiation followed by resection on a limited set of patients. The median disease-free survival and overall survival of patients with NE differentiation (13.1 months and 22.5 months, respectively) and without NE differentiation (42.2 months and 42.3 months, respectively) in pretreatment biopsy specimens did not differ significantly (P ¼.27 and P ¼.14, respectively). DISCUSSION The clinical significance of NE differentiation in tumors is site-dependent. It has been reported that tumors with NE differentiation imply a good prognosis for patients pancreatic carcinomas, 13 a poor prognosis for patients with prostatic adenocarcinomas, 14,15 and have no effect on clinical outcome in patients with breast carcinomas. 16,17 The prognostic relevance of NE differentiation in colonic adenocarcinomas is controversial, 18 24 but most studies have observed that NE differentiation was associated with a worse prognosis in patients with colonic adenocarcinomas. 21,22,25 NE differentiation in esophageal and EGJ adenocarcinomas has been reported previously, 6,7 but its clinical significance remains unclear specifically in relation to preoperative chemoradiation and response to treatment.
1472 CANCER October 1, 2006 / Volume 107 / Number 7 FIGURE 3. Overall survival and disease-free survival is illustrated for patients with adenocarcinoma of the esophagus and esophagogastric junction who received preoperative chemoradiation. Patients who had no residual tumor (complete pathologic response [CR]) after chemoradiation had significantly better (A) overall survival and (B) disease-free survival compared with patients who had residual tumor. Patients who had residual tumor without neuroendocrine (NE) differentiation had significantly better (C) overall survival and (D) disease-free survival compared with patients who had residual tumor with NE differentiation. In the current study, NE differentiation was present in 38% of esophageal and EGJ adenocarcinomas in pretreatment biopsies and in 52% of residual adenocarcinomas after preoperative neoadjuvant therapy, similar to previously reported rates of from 21% to 49%. 6,7,26 In our study, the majority of tumors had focal NE differentiation (<5% of tumor cells). This finding was similar to previous studies, which found focal NE differentiation in tumors from multiple organ sites, including tumors of the pancreas, 13 colon, 21,22,25,27,28 and esophagus. 7,26 It is noteworthy that, in the current study, the percentage of tumor cells with NE differentiation was increased significantly in residual tumor in posttreatment resection specimens compared with paired tumor in pretreatment biopsy specimens. Increased ME cells also have been reported in colorectal and prostatic adenocarcinomas after neoadjuvant therapy, 8,29,30 and the extent of NE cells appears to be related to the degree of cytotoxic insult and treatment response. NE cell hyperplasia is observed in a variety of inflammatory conditions, such as chronic pancreatitis and inflammatory bowel disease, suggesting that NE cells may be more resistant to inflammatory injury. There are 2 main hypotheses to explain why NE differentiation increases in frequency and density in treated adenocarcinomas. First, tumor cells with NE differentiation may be more resistant to the cytotoxic insult caused by neoadjuvant therapy. 8,30 Second, cytotoxic injury may induce NE differentiation in tumor cells. 8 Recent studies showed that NE-secreted products, such as serotonin, somatostatin, and bombesin, may influence tumor growth by promoting proliferation, invasiveness, metastatic potential, and angiogenesis and by conferring antiapoptotic capabilities in prostatic and colorectal carcinoma. 14,31 Currently, there is no consensus regarding the prognostic significance of NE cells in esophageal and EGJ adenocarcinoma. Previous studies have shown that NE differentiation in esophageal and EGJ adenocarcinomas have no effect on survival 7,26 ; however, those studies included patients who had received preoperative chemotherapy as and patients who had undergone primary resection alone. In the current study, we evaluated only those patients who had received preoperative chemotherapy and demonstrated that, although the presence of NE differentiation in pretreatment tumor was not associated with patient s out-
NE Differentiation in Esophageal Cancer/Wang et al. 1473 TABLE 6 Univariate Analysis of Disease-Free and Overall Survivals Disease-free survival Overall survival Characteristic No. HR 95% CI P value HR 95% CI P value Age 73 1.00 0.97 1.03 0.97 1.01 0.97 1.04 0.76 Gender Male 70 0.99 0.24 4.09 0.99 1.15 0.28 4.77 0.84 Female (Reference) 3 1.00 1.00 Pathologic Stage 0.02 0.016 I (Reference) 12 1.00 1.00 II 34 0.99 0.42 2.36 0.99 1.07 0.42 2.71 0.89 III 22 2.06 0.87 4.91 0.10 2.39 0.95 6.01 0.06 IV 5 3.57 1.11 11.47 0.03 3.74 1.13 12.41 0.031 Extent of residual tumor 0.09 0.007 0% (Reference) 3 1.00 1.00 1 10% 26 1.14 0.25 5.08 0.87 2.48 0.31 20.25 0.40 11 50% 17 1.75 0.38 8.10 0.47 4.31 0.51 36.60 0.18 > 50% 27 2.64 0.60 11.64 0.20 7.84 0.96 64.13 0.055 NE differentiation Absent (Reference) 35 1.00 0.03 1.00 0.045 Present 38 1.89 1.07 3.32 1.83 1.01 3.31 HR indicates hazard ratio; CI, confidence interval; NE, neuroendocrine. TABLE 7 Multivariate Cox Regression Analysis of Disease-Free and Overall Survivals Disease-free survival Overall survival Characteristic No. HR 95% CI P value HR 95% CI P value Age 73 1.01 0.98 1.04 0.59 1.02 0.98 1.05 0.37 Gender Male 70 1.09 0.22 5.39 0.92 1.55 0.31 7.85 0.60 Female (Reference) 3 1.00 1.00 Pathologic Stage 0.18 0.21 I (Reference) 12 1.00 1.00 II 34 0.95 0.38 2.38 0.92 0.97 0.37 2.59 0.96 III 22 1.92 0.74 5.01 0.18 1.84 0.68 4.96 0.23 IV 5 2.93 0.56 6.65 0.30 2.35 0.64 8.70 0.20 Extent of Residual Tumor 0.09 0.02 0% (Reference) 3 1.00 1.00 0 10% 26 0.47 0.09 2.39 0.36 1.16 0.13 10.20 0.89 10 50% 17 1.06 0.22 5.17 0.94 2.66 0.31 22.83 0.37 > 50% 27 1.17 0.24 5.71 0.84 3.72 0.43 32.14 0.23 NE differentiation Absent (Reference) 35 1.00 1.00 Present 38 2.36 1.18 4.72 0.02 1.85 0.93 3.68 0.08 HR indicates hazard ratio; CI, confidence interval; NE, neuroendocrine. come, the presence of NE differentiation in residual tumor after preoperative chemotherapy was associated with worse disease-free survival and worse overall survival in patients with residual esophageal or EGJ adenocarcinoma. In addition, the presence of NE differentiation in residual tumor was an independent prognostic predictor of worse disease-free survival independent of the pathologic stage or the extent of residual tumor in multivariate analysis. In conclusion, the current results suggest that tumors with NE differentiation may be more resistant to preoperative neoadjuvant therapy than tumors with-
1474 CANCER October 1, 2006 / Volume 107 / Number 7 out NE differentiation. The presence of NE differentiation in residual esophageal or EGJ adenocarcinomas is associated with a worse patient outcome and may be used as an adjunct tool in predicting survival after preoperative neoadjuvant therapy. REFERENCES 1. Chen X, Yang CS. Esophageal adenocarcinoma: a review and perspectives on the mechanism of carcinogenesis and chemoprevention. Carcinogenesis. 2001;22:1119 1129. 2. Altorki N, Kent M, Ferrara C, Port J. Three-field lymph node dissection for squamous cell and adenocarcinoma of the esophagus. Ann Surg. 2002;236:177 183. 3. Orringer MB, Marshall B, Iannettoni MD. Transhiatal esophagectomy: clinical experience and refinements. Ann Surg. 1999;230:392 400; discussion, 403. 4. Heath EI, Burtness BA, Heitmiller RF, et al. Phase II evaluation of preoperative chemoradiation and postoperative adjuvant chemotherapy for squamous cell and adenocarcinoma of the esophagus. J Clin Oncol. 2000;18:868 876. 5. Kelsen DP, Ginsberg R, Pajak TF, et al. Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N Engl J Med. 1998;339:1979 1984. 6. Griffin M, Sweeney EC. The relationship of endocrine cells, dysplasia and carcinoembryonic antigen in Barrett s mucosa to adenocarcinoma of the oesophagus. Histopathology. 1987; 11:53 62. 7. Hamilton K, Chiappori A, Olson S, Sawyers J, Johnson D, Washington K. Prevalence and prognostic significance of neuroendocrine cells in esophageal adenocarcinoma. Mod Pathol. 2000;13:475 481. 8. Shia J, Tickoo SK, Guillem JG, et al. Increased endocrine cells in treated rectal adenocarcinomas: a possible reflection of endocrine differentiation in tumor cells induced by chemotherapy and radiotherapy. Am J Surg Pathol. 2002;26: 863 872. 9. Werner M, Flejou JF, Hainaut P, et. al. Adenocarcinoma of the esophagus. In: Hamilton SR, Aaltonen LA, editors. World Health Organization Classification of Tumors: Pathology and Genetics of Tumours of the Digestive System. Geneva: World Health Organization; 2000:20 25. 10. Greene FL, Page DL, Fleming ID, et al., editors. Esophaqus. American Joint Committee on Cancer Staging Manual. 6th ed. New York: Springer; 2002:91 103. 11. Chirieac LR, Swisher SG, Ajani JA, et al. Posttherapy pathologic stage predicts survival in patients with esophageal carcinoma receiving preoperative chemoradiation. Cancer. 2005; 103:1347 1355. 12. Mandard AM, Dalibard F, Mandard JC, et al. Pathologic assessment of tumor regression after preoperative chemoradiotherapy of esophageal carcinoma: clinicopathologic correlations. Cancer. 1994;73:2680 2686. 13. Tezel E, Nagasaka T, Nomoto S, Sugimoto H, Nakao A. Neuroendocrine-like differentiation in patients with pancreatic carcinoma. Cancer. 2000;89:2230 2236. 14. Abrahamsson PA. Neuroendocrine differentiation in prostatic carcinoma. Prostate. 1999;39:135 148. 15. Helpap B, Kollermann J, Oehler U. Neuroendocrine differentiation in prostatic carcinomas: histogenesis, biology, clinical relevance, and future therapeutical perspectives. Urol Int. 1999;62:133 138. 16. Miremadi A, Pinder SE, Lee AH, et al. Neuroendocrine differentiation and prognosis in breast adenocarcinoma. Histopathology. 2002;40:215 222. 17. Scopsi L, Andreola S, Pilotti S, et al. Mucinous carcinoma of the breast. A clinicopathologic, histochemical, and immunocytochemical study with special reference to neuroendocrine differentiation. Am J Surg Pathol. 1994;18:702 711. 18. Pagani A, Papotti M, Abbona GC, Bussolati G. Chromogranin gene expressions in colorectal adenocarcinomas. Mod Pathol. 1995;8:626 632. 19. Secco GB, Campora E, Fardelli R, et al. Chromogranin-A expression in neoplastic neuroendocrine cells and prognosis in colorectal cancer. Tumori. 1996;82:390 393. 20. Foley EF, Gaffey MJ, Frierson HF Jr. The frequency and clinical significance of neuroendocrine cells within Stage III adenocarcinomas of the colon. Arch Pathol Lab Med. 1998;122: 912 914. 21. Grabowski P, Schindler I, Anagnostopoulos I, et al. Neuroendocrine differentiation is a relevant prognostic factor in Stage III-IV colorectal cancer. Eur J Gastroenterol Hepatol. 2001;13: 405 411. 22. Hamada Y, Oishi A, Shoji T, et al. Endocrine cells and prognosis in patients with colorectal carcinoma. Cancer. 1992;69: 2641 2646. 23. Smith DM Jr., Haggitt RC. The prevalence and prognostic significance of argyrophil cells in colorectal carcinomas. Am J Surg Pathol. 1984;8:123 128. 24. Mori M, Mimori K, Kamakura T, Adachi Y, Ikeda Y, Sugimachi K. Chromogranin positive cells in colorectal carcinoma and transitional mucosa. JClinPathol. 1995;48:754 758. 25. de Bruine AP, Wiggers T, Beek C, et al. Endocrine cells in colorectal adenocarcinomas: incidence, hormone profile and prognostic relevance. Int J Cancer. 1993;54:765 771. 26. Koppert LB, Wijnhoven BP, Tilanus HW, Stijnen T, Van Dekken H, Dinjens WN. Neuroendocrine in Barrett s mucosa and adenocarcinomas of the gastroesophageal junction. Int J Surg Pathol. 2004;12:117 125. 27. Ferrero S, Buffa R, Pruneri G, et al. The prevalence and clinical significance of chromogranin A and secretogranin II immunoreactivity in colorectal adenocarcinomas. Virchows Arch. 1995;426:587 592. 28. Lapertosa G, Baracchini P, Delucchi F. Prevalence and prognostic significance of endocrine cells in colorectal adenocarcinomas. Pathologica. 1994;86:170 173. 29. Ahlgren G, Pedersen K, Lundberg S, Aus G, Hugosson J, Abrahamsson PA. Regressive changes and neuroendocrine differentiation in prostate cancer after neoadjuvant hormonal treatment. Prostate. 2000;42:274 279. 30. Bonkhoff H. Neuroendocrine cells in benign and malignant prostate tissue: morphogenesis, proliferation, and androgen receptor status. Prostate. 1998;8(Suppl):18 22. 31. Seretis E, Gavrill A, Agnantis N, Golematis V, Voloudakis- Baltatzis IE. Comparative study of serotonin and bombesin in adenocarcinomas and neuroendocrine tumors of the colon. Ultrastruct Pathol. 2001;25:445 454.