1638 Positron Emission Tomography of Esophageal Carcinoma Using 11C-Choline and 18 F-Fluorodeoxyglucose A Novel Method of Preoperative Lymph Node Staging Oichiro Kobori, M.D.1 Yujiro Kirihara, M.D.1 Noboru Kosaka, M.D.2 Toshihiko Hara, M.D.2 1 Department of Surgery, International Medical Center of Japan, Tokyo, Japan. 2 Department of Radiology, International Medical Center of Japan, Tokyo, Japan. BACKGROUND. Accurate preoperative staging is an important but difficult problem in determining therapy for patients with esophageal carcinoma. Positron emission tomography (PET) is used with [methyl-11C]choline (11C-choline) and 2-[18F]fluoro-2deoxy-D-glucose (18F-FDG) to detect a variety of malignancies. The authors used PET with both of these agents to detect lymph node metastases in patients with esophageal carcinoma. METHODS. Lymph node metastases in 33 patients with biopsy-proven esophageal carcinoma (16 patients with tumors classified as T1 and 17 patients with tumors classified as T2– 4) was examined by PET using 11C-choline and 18F-FDG, and the accuracy of the results was correlated with pathology findings after surgery. RESULTS. 11C-choline PET was more effective than 18F-FDG PET and computed tomography (CT) in detecting very small metastases localized in the mediastinum. It was ineffective, however, in detecting metastases localized in the upper abdomen, because of the normal uptake of 11C-choline in the liver. 18F-FDG PET was superior to CT in detecting metastases in the mediastinum and the upper abdomen, whereas 11C-choline PET was superior to 18F-FDG PET in detecting metastases in the mediastinum. When 11C-choline PET and 18F-FDG PET were used in combination, they were very effective in evaluating the lymph node status in both the mediastinum and the upper abdomen, and detected 85% of the metastatic lymph nodes (n 5 46). CONCLUSIONS. In this study, the combination of 11C-choline PET and 18F-FDG PET was very effective in evaluating the lymph node status of patients with esophageal carcinoma preoperatively. Cancer 1999;86:1638 – 48. © 1999 American Cancer Society. KEYWORDS: esophagus, carcinoma, diagnosis, staging, lymph node, positron emission tomography, 11C-choline, 18F-fluorodeoxyglucose. Supported in part by the Science and Technology Agency of Japan and the Japanese Smoking Research Foundation. Address for reprints: Toshihiko Hara, M.D., Department of Radiology, International Medical Center of Japan, 1-21-1 Toyama, Shinjuku-ku, Tokyo 162, Japan. Received November 23, 1998; revision received June 1, 1999; accepted June 1, 1999. © 1999 American Cancer Society T he chance of finding esophageal carcinoma at an early stage is increasing as the result of remarkable progress both in endoscopic technology and medical industries. Once it is found and treatment is considered, the next step is a survey of lymph node metastases. Recently, a cooperative study of 1740 patients with superficial esophageal carcinoma (T1 tumor) in which we are participating has elucidated that lymph node metastasis starts to increase as the tumor advances to the layer of lamina muscularis mucosae, a thin layer between the mucosa and the submucosa (unpublished data). This and other studies1– 4 are indicating that lymph node metastases are not rare events with the T1 tumor. The presence or absence of PET of Esophageal Carcinoma with Choline and FDG/Kobori et al. metastases determines the mode of surgical procedure. If there are no metastases, then endoscopic mucosal resection or esophagectomy without thoracotomy (i.e., transhiatal esophagectomy) are the treatments of choice. If there are any lymph node metastases, then esophagectomy with thoracotomy is the most justifiable treatment, but serious complications may ensue postoperatively. Endosonographic examination is imperfect in evaluating lymph node metastases of esophageal carcinoma,5–12 although it is useful in estimating the depth of invasion. Other noninvasive methods, including computed tomography (CT), are inaccurate in detecting lymph node metastases. Although thoracoscopic and laparoscopic examination seem to be the most reliable procedures for estimating lymph node metastases, they are invasive. Postoperative pathologic examination is still the gold standard for the staging of lymph node status. Positron emission tomography (PET) is a noninvasive imaging technique that can be used to identify focal areas of increased metabolism associated with malignancies. Primary tumor and metastases can be visualized by the increased focal uptake of a positronemitting tracer, 2-[18F]fluoro-2-deoxy-D-glucose (18FFDG), that is detected by PET scanning. The application of this method to esophageal carcinoma has been reported by Luketich et al.13,14 and Block et al.15 Recently, we introduced [methyl-11C]choline (11Ccholine) as another tumor-seeking tracer and successfully visualized brain tumor, prostate carcinoma, and other malignancies using PET.16,17 Choline is a natural blood constituent and penetrates cell membranes. 11 C-choline is incorporated in kidney and liver, converted to betaine, liberated into circulation again, then used for transmethylation reactions in various organs. In tumors, however, the only metabolic pathway of 11 C-choline is its integration into phospholipids. 11Ccholine is phosphorylated within tumor cells, and, after several biosynthetic processes, finally is integrated into phosphatidylcholine (lecithin), a major component of cell membrane phospholipids. Once 11 C-choline is phosphorylated within tumor cells, it remains there: This constitutes another “chemical trap.” Because tumor cells duplicate very fast, the biosynthesis of cell membranes also is very fast. It follows that the uptake rate of 11C-choline in tumors is proportional to the rate of tumor duplication. The principles of tumor imaging using 18F-FDG are as follows. Tumor cells frequently, but not always, consume a great deal of glucose, because they are inclined toward anaerobic glycolysis. 18F-FDG, a glucose analogue, is incorporated into tumors, but the immediate metabolic product, 18F-FDG-6-phosphate, 1639 is not metabolized further and is “chemically trapped” within tumor cells.18 Thus, 18F-FDG accumulates in tumors, thereby enabling their detection by PET. We performed PET studies in patients with esophageal carcinoma using 11C-choline and 18F-FDG, and the results were compared with those from CT scans. The radioactivity concentration in tumors and lymph nodes was measured in terms of the standardized uptake value (SUV). Finally, the effectiveness of these imaging techniques was evaluated compared with the pathologic findings obtained after surgery. MATERIALS AND METHODS Patients The PET studies using 11C-choline and 18F-FDG were performed in 33 patients with esophageal carcinoma (28 males and 5 females; ages, 50 – 81 years; mean age, 63.9 years; 16 patients with T1 tumors and 17 patients with T2–T4 tumors according to the TNM classification), of which the diagnosis was given in advance by esophageal endoscopy and biopsy. Informed consent was obtained from the patients. The PET study was performed before noon, when the patient was abstaining from breakfast. Radiopharmaceuticals 11 C-choline and 18F-FDG were prepared using a cyclotron and automated synthetic apparatuses that we constructed.19,20 Dosimetry of 11C-choline (decay halflife, 20 minutes) in 4 normal human subjects was performed by using the MIRD method21 (calculation by Dr. K. Fukushi of the National Institute of Radiological Sciences, Japan), and the results were as follows: 18.0 3 10212 Sv/Bq in kidney, 17.3 3 10212 Sv/Bq in liver, and 2.8 3 10212 Sv/Bq in the whole body, in which kidney and liver are the target organs. PET Imaging and Calculation PET images were obtained using a PET camera (6-mm spatial resolution; Headtome IV; Shimadzu, Kyoto, Japan) equipped with three rings producing five slices at 13-mm intervals. For 11C-choline PET scans, transmission scans and injection of 11C-choline (370 MBq) were followed immediately by emission (emission scan started 5 minutes after injection). For 18F-FDG PET scans, transmission scan was performed before injection of 18F-FDG (370 MBq), then, after 40 minutes, the patient was put back into the previous position, and the emission scan was performed. Both the transmission scans and the emission scans measured the whole range from liver to neck by shifting the bed position 6 times, with a scan time of 3 minutes each. By combining the emission and transmission data in a computer, attenuation-corrected 1640 CANCER November 1, 1999 / Volume 86 / Number 9 emission images (PET images) were obtained. The image of horizontal slices was presented on a computer screen by color display. The specific color represented the radioactivity concentration in each pixel (4 mm 3 4 mm 3 6 mm in real size) using the SUV, which is defined as SUV 5 Regional radioactivity concentration Total injected dose/bod y weight where the radioactivity concentration in a pixel (Bq/ mL) is to be determined from an apparent pixel count (cps/pixel volume) and a predetermined cofactor. The numeric value of the SUV in a region of interest (ROI) was provided by the computer. The reading of the PET image (interpretation and calculation) was performed by a group of radiologists who knew nothing more than the localization of the primary tumor. Surgery and Pathology After the PET study, all patients underwent surgery for esophageal carcinoma by esophagectomy and regional lymph node dissection. The esophagectomy was performed through a right thoracotomy followed by laparotomy and left neck incision. After labeling, the resected esophagus and the tissues adjacent to the esophagus that possibly contained metastatic lymph nodes were transferred to a pathology room. After surgery, the surgeons separated lymph nodes from the resected esophagus and the adjacent tissues and assigned specified numbers to them for indication of localization (according to the guidelines of the Japanese Society for Esophageal Diseases22) in the Pathology Department. All subsequent procedures were carried out in this department. The surgical specimens were fixed, embedded, sectioned, stained with hematoxylin and eosin, and finally examined microscopically. Small lymph nodes (,5 mm) were processed as a whole. Larger lymph nodes were cut into two pieces after being fixed, and one of them was submitted to further procedures. One section plane in each lymph node was examined microscopically. Dr. Michiyo Nasu, a pathologist who is experienced in esophageal carcinoma, reviewed all specimens. The total numbers of the examined lymph nodes were in the range of 20 –77 lymph nodes in each surgical case. The results of the lymph node metastases and localization were recorded according to the above-mentioned guideline. The status of the lymph node aggregation was not recorded, which is commonplace; however, this posed a problem later, when we wanted to compare pathologic findings with PET images. Their localizations finally were allotted to anatomic regions as described in the guidelines discussed above, i.e., cervical, upper thoracic, midthoracic, lower thoracic, and abdominal regions (equivalent to TNM classification). Superficial esophageal tumors in the T1 category were classified according to the depth of invasion. Esophageal carcinoma confined to the epithelium was called ep-cancer, tumors remaining within the muscularis mucosae were referred to as mm-cancer, and tumors involving the submucosa were called sm-cancer.22 Lymph node metastases of esophageal carcinoma were recorded as such, without applying the M category of the new (1987) TNM classification to the distant lymph node metastases. RESULTS Time Course of Carcinoma 11 C-Choline Uptake in Esophageal A stable tumor image was obtained by carrying out the PET scan when the tumor activity remained constant. With 11C-choline, the scan was started 5 minutes after injection, because the tumor activity reached a maximum in 5 minutes and stayed at a constant level afterward, whereas the blood activity (within the right ventricle) decreased rapidly then continued to be very low (Fig. 1). With 18F-FDG, however, the most stable image was obtained 40 – 60 minutes after injection.23 Case Presentation and Overall Results Typical cases from three patients are presented in Figures 2– 4. The overall results in detecting primary tumors are shown in Table 1. 11C-choline was superior to 18F-FDG in sensitivity for detecting small tumors. Concerning the T1 tumors (n 5 16), 15 tumors were visualized with 11C-choline (with 1 negative result for a tumor localized in the upper abdomen, which was indistinguishable from the normal liver), whereas only 6 tumors were visualized with 18F-FDG. Concerning the tumors with higher classifications (T2–T4; n 5 17), 16 tumors were visualized with 11C-choline (with 1 negative result for a tumor in the upper abdomen), and all 17 tumors were visualized with 18F-FDG. If 11 C-choline PET and 18F-FDG PET were used in combination, then all primary tumors were detected. The overall results in detecting lymph node metastases are shown in Table 2, in which the lymph nodes are assigned to pertaining anatomic regions. With regard to lymph node metastases in the mediastinum (n 5 32 lymph nodes), 11C-choline PET was positive in 28 lymph nodes and negative in 4 lymph nodes (sensitivity, 88%), whereas 18F-FDG PET was positive in 11 lymph nodes and negative in 21 lymph nodes (sensitivity, 34%). Concerning lymph node metastases in the upper abdomen (n 5 14 lymph nodes), 11 C-choline PET was completely ineffective because of the high uptake of 11C-choline in the liver (sensitivity, PET of Esophageal Carcinoma with Choline and FDG/Kobori et al. FIGURE 1. Time course of radioactivity concentration (standardized uptake value [SUV]) in the blood pool of the right ventricle and in a tumor after intravenous injection of [methyl-11C-]choline into a patient with esophageal carcinoma. 0%), whereas 18F-FDG PET was positive in 11 lymph nodes and negative in 3 lymph nodes (sensitivity, 79%). If 11C-choline PET and 18F-FDG PET were used in combination for the metastatic lymph nodes both in the mediastinum and in the upper abdomen (n 5 46 lymph nodes), then it was positive in 39 lymph nodes and negative in 7 lymph nodes (sensitivity, 85%). The minimal size of the metastatic lymph nodes detected by PET was 4 mm with 11C-choline and 8 mm with 18F-FDG. The sensitivity of 18F-FDG PET for lymph node metastases was high (79%) in the abdominal region and low (34%) in the mediastinum. This difference was accounted for by the difference in the size of the lymph nodes, which are larger in the abdomen and smaller in the mediastinum. DISCUSSION The survival rate for patients with esophageal carcinoma is dismal, with 6 –11% of all patients alive 5 years after diagnosis and treatment.24 Those esophageal carcinoma patients with a good prognosis, in general, have squamous cell carcinoma at an early stage. This 1641 FIGURE 2-1. Clinical and pathologic findings of the primary tumor. (A) Barium film. (B) Endoscopy (elevated lesion, 2 cm 3 1 cm). (C) Gross view of the esophagus (Lugol stain; unstained area, 6.5 cm 3 3.0 cm). (D) Histology (H & E, low magnification. (E) Histology (high magnification, 3 100). FIGURE 2-2. PET images with 11C-choline and 18F-FDG. (A) Primary tumor. (B) Lymph node metastasis. FIGURE. 2. Esophageal carcinoma of the midesophagus. There were no abnormal findings on computed tomography (CT) scan in this patient. [Methyl- 11 C]choline (11C-choline) positron emission tomography (PET) and [18F-fluoro2-deoxy-D-glucose (18F-FDG) PET showed high uptake in the primary tumor and lymph node metastasis (Patient 9 in Tables 1 and 2). 1642 CANCER November 1, 1999 / Volume 86 / Number 9 FIGURE 3-1. Clinical and pathologic findings of the primary tumor. (A) Lugol stain (unstained area, 3.7 cm 3 12 cm). (B) Endoscopy. (C) Histology (H & E, low magnification). (D) Histology (high magnification, 3 100). FIGURE 3-2. PET images with 11C-choline and 18F-FDG. (A) Primary tumor. (B) Lymph node metastasis. FIGURE. 3. Esophageal carcinoma of the midesophagus. There were no abnormal findings on CT in this patient. 11C-choline PET showed high uptake in the primary tumor and lymph node metastasis. 18F-FDG PET showed no high uptake (Patient 13 in Tables 1 and 2). FIGURE 4-1. Clinical and pathologic findings of the primary tumor. (A) Barium film. (B) Endoscopy (elevated lesion, 1.5 cm 3 1.5 cm). (C) Macroscopic view of the elevated lesion with ulceration. (D) Histology (H & E, low magnification). E. Histology (high magnification, 3 100). FIGURE 4-2. PET images with 11C-choline and 18F-FDG. (A) Primary tumor. (B) Lymph node metastasis. In addition, 11C-choline PET showed high uptake in bone marrow (resulting in a positive bone scan 4 months later), and 18F-FDG PET showed normal high uptake in myocardium. FIGURE. 4. Esophageal carcinoma of the midesophagus. CT scan showed a slightly thickened esophageal wall but no lymph node enlargement. Both 11 C-choline PET and 18F-FDG PET showed high uptake in the primary tumor and many lymph node metastases (Patient 10 in Tables 1 and 2). PET of Esophageal Carcinoma with Choline and FDG/Kobori et al. 1643 TABLE 1 Esophageal Carcinoma in Patients: Pathologic Location and Extent of the Primary Tumor and its Histologic Type Compared with the Findings of [Methyl-11C]Choline, Positron Emission Tomography, [18F]Fluoro-2Deoxy-D-Glucose, and Computed Tomography Primary tumor 11 18 F-FDG PET SUV CT wall thickness (mm) — — — 2.59 — — — — 3.45 3.13 3.46 1.44 — — — 0.97 5.49 3.54 2.02 3.09 2.67 2.18 6.78 3.60 2.78 6.18 3.31 3.83 6.24 5.61 4.70 5.82 9.88 — — — — — — — 13 — 6 — — — — — — — 32 13 16 10 17 44 64 33 45 30 17 18 19 13 61 23 Patient Region Pathology extent Histology C-choline PET SUV 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Lower thoracic Midthoracic Midthoracic Abdominal Midthoracic Midthoracic Lower thoracic Midthoracic Midthoracic Midthoracic Lower thoracic Midthoracic Midthoracic Midthoracic Upper thoracic Midthoracic Midthoracic Midthoracic Midthoracic Midthoracic Midthoracic Midthoracic Midthoracic Lower thoracic Lower thoracic Midthoracic Abdominal Midthoracic Lower thoracic Midthoracic Upper thoracic Lower thoracic Midthoracic T1 (sm) T1 (sm) T1 (sm) T1 (sm) T1 (sm) T1 (sm) T1 (sm) T1 (sm) T1 (sm) T1 (sm) T1 (sm) T1 (sm) T1 (mm) T1 (mm) T1 (mm) T1 (mm) T2 T2 T3 T3 T3 T3 T3 T3 T3 T3 T3 T3 T3 T3 T3 T3 T4 Well Mod Mod Mod Poor Poor Poor Poor Poor Poor Poor Poor Poor Poor Poor Poor Mod Mod Mod Mod Mod Mod Well Well Well Mod Mod Poor Poor Poor Poor Poor Mod 2.59 2.61 1.78 — 3.14 2.84 2.15 1.76 2.22 3.04 3.49 2.50 3.31 1.34 1.41 1.46 3.22 3.70 2.05 2.90 3.43 3.23 3.56 3.42 3.33 2.56 — 2.60 4.52 3.38 3.89 3.19 4.49 PET: positron emission tomography; 18F-FDG: [18F]fluoro-2-deoxy-D-glucose; CT: computed tomography; 11C-choline: [methyl-11C]choline; SUV: uptake rate in the PET image; T1, T2, T3, T4: tumor extent according to TNM classification; sm: submucosa; mm: muscularis mucosae (esophageal thickness at the tumor site); poor: poorly differentiated squamous cell carcinoma; well: well-differentiated squamous cell carcinoma; mod: moderately differentiated squamous cell carcinoma; —: negative. kind of esophageal carcinoma has been called either superficial esophageal carcinoma (not infiltrated beyond the submucosa) or early esophageal carcinoma (a subset of superficial esophageal carcinoma without regional lymph node metastases).22 Excellent 5-year survival rates in the range of 56.7–78% have been reported in patients undergoing surgical treatment of early-stage squamous cell carcinoma.1–3,25–30 It has been reported that the 5-year survival rate in these patients decreases progressively as invasion progresses from intraepithelial, to intramucosal, to sub- mucosal.26,29,31 It also has been reported that the likelihood of lymph node metastases increases in association with invasion of deeper esophageal wall, with the incidence of lymph node metastases in 5– 8% of patients with intramucosal squamous cell carcinoma.1– 4 Therefore, early identification and prompt surgical management of patients with superficial esophageal carcinoma before dissemination by way of lymphatics are crucial if survival rates from this lethal disease are to be improved. About one-half of the patients with esophageal 1644 CANCER November 1, 1999 / Volume 86 / Number 9 TABLE 2 Lymph Node Metastasis in Patients with Esophageal Carcinoma: Pathologic Location and Size of Metastasized Lymph Nodes Compared with the Findings of [Methyl-11C]Choline Positron Emission Tomography, [18F]Fluoro-2-Deoxy-D-Glucose, and Computed Tomography Lymph node metastasis Pathology Patient/region Diameter (mm) 11 1 2 3 4 5 6 — — — 10 — — 12 5 20 — 9, 5, 4 15, 13 10, 10 20, 14, 13, 10, 10, 8 17, 5, 4 — 20, 14, 10, 7, 7, 6 19 4 — 8, 7 6 — — 13 — 14, 10, 8 9 13 11, 6, 4 — 13, 11, 10, 10, 6 6, 5, 4 — 16, 5, 5 10 10, 8, 6, 6 — 14 5 10, 6 13, 8, 8, 6, 6, 5 8 7 4 16, 5, 5 10, 6, 6 9, 4 — 6, 6, 5, 4, 4 17, 13, 5 18, 7, 4 13 — 3.28, 3.14, 3.03 2.06 — — 2.99 3.11 3.22 — 2.59 — 2.20, 1.97, 1.50 1.82, 1.65 3.84 2.68, 2.16, 2.01 1.78 3.20, 3.02, 2.71, 2.53, 2.25 3.41 — 2.62 4.22 2.39 — 2.33 — — 2.71 2.37 — 2.24, 1.73 — 3.34, 2.72 2.56 3.23 — 3.06 — — 1.56 1.58 — — 2.11 1.72 — — 1.72 1.91, 1.83 2.27 — — 2.59 — 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 No Midthoracic Lower thoracic Abdominal No Cervical Upper thoracic Lower thoracic Abdominal Midthoracic Abdominal Upper thoracic Midthoracic Lower thoracic abdominal Upper thoracic Midthoracic Cervical Midthoracic Abdominal Lower thoracic Lower thoracic Midthoracic No Cervical Lower thoracic No Upper thoracic Lower thoracic Abdominal Upper thoracic No Cervical Upper thoracic Midthoracic Abdominal Midthoracic Abdominal No Upper thoracic Midthoracic Lower thoracic Abdominal Upper thoracic Midthoracic Lower thoracic Abdominal Upper thoracic Midthoracic Lower thoracic Abdominal Abdominal Upper thoracic Abdominal C-choline PET SUV 18 F-FDG PET SUV — — 2.41 2.43 — — — — 3.11 — 1.48 1.57, 1.38 1.37 3.98 3.26 — 3.71, 3.44 — — — 2.26 — — 2.67 — — — — 2.52 2.34 — 2.45 — 1.17 1.25 — 3.37, 3.10 — — — — 2.03 — — — 2.71 — — — — 2.52, 2.45 3.63 3.64 CT diameter (mm) — — — — — — — — 13 — — 14 11 19, 13, 10 — — 13 — — — 12 — — — — — 12 — 12 12 — 13 12 — — 10 10 — 15 — — 12, 11 — — — 25 — — — — 18 — — (continued) PET of Esophageal Carcinoma with Choline and FDG/Kobori et al. 1645 TABLE 2 (continued) Lymph node metastasis Pathology Patient/region Diameter (mm) 11 31 Cervical Upper thoracic Abdominal 32 Upper thoracic Midthoracic Lower thoracic Abdominal 33 No 10, 8 20, 5 13 8, 7, 6 6 9 5 — 3.01 3.17, 3.01 — 2.90, 2.77 — 3.04 — — C-choline PET SUV 18 F-FDG PET SUV — — 2.09 1.68, 1.66 — 5.06 — — CT diameter (mm) — — — — — — — — 11 C-choline: [methyl-11C]choline; PET: positron emission tomography; 18FDG: [18F]fluoro-2-deoxy-D-glucose; SUV: standardized uptake value (uptake rate in the PET image); CT: computerized tomography; Diameter: greatest dimension of the lymph node. carcinoma have a resectable tumor at first presentation.32,33 However, in one-half of all tumors, lymph node metastases are present already.34,35 If lymph node metastases are present, then the principle of oncologic radicality commands that one should aim not only at resecting the esophagus with primary tumor but also at dissecting the draining lymphatic system extensively. The main cause of death in patients with esophageal carcinoma, unlike other tumors of the gastrointestinal tract, is not hematogenous distant metastases but local recurrence, which is reported to be up to 80% in some series.36 About one-half of these recurrences seem to be attributable to lymph node metastases that were left behind.34 The consequence is that adequate lymphadenectomy is very important in surgery for patients with esophageal carcinoma. Since the first successful removal of a tumor of the thoracic esophagus by Turner37 in 1933, esophagectomy without thoracotomy (transhiatal approach), a procedure of bluntly stripping the esophagus from the mediastinum, has long been accepted as a standard procedure in esophagectomy. The disadvantage of this procedure is the inadequate removal of tumor-affected regional lymph nodes, if any. An alternative procedure is esophagectomy with thoracotomy (transthoracic approach), but the majority opinion holds that esophagectomy without thoracotomy obviously is safer than esophagectomy with thoracotomy if postoperative complications are taken into account.38 – 48 If there are no regional lymph node metastases, then esophagectomy without thoracotomy (and endoscopic resection of esophageal mucosa as well) seems to be the option of choice. In this case, however, confirmation of the absence of metastases is manda- tory. If there are lymph node metastases, however, then esophagectomy with thoracotomy is required. Surgical therapy yields the best results in esophageal carcinoma. For obtaining a good result, precise preoperative staging is essential: The patients should be evaluated carefully concerning the operability of their tumors, and the most appropriate surgical procedure should be applied. CT is not accurate enough for evaluating lymph node metastases in the esophageal carcinoma, with an accuracy rate of 51– 67%.6,7,10,49 –52 Endoscopic ultrasonography has a high accuracy rate for estimating the depth of primary esophageal carcinoma, but it is inaccurate in evaluating lymph node status.5–12 Recently, a combination of thoracoscopy and laparoscopy has been introduced for detecting locoregional and distant metastases11,52,53 and achieved high accuracy rates of 93% in thoracoscopy and 94% in laparoscopy,52 but this is an invasive methodology. Luketich et al.13,14 applied the 18F-FDG PET method for detecting regional lymph node metastases and distant metastases in esophageal carcinoma patients. They found that, for regional lymph node metastases, the sensitivity was 45%, the specificity was 100%, and the accuracy rate was 48%. Small regional lymph node metastases (mean greatest dimension, 5.2 mm; range, 2–10 mm) were not visualized by this method. Block et al.15 also applied this method to esophageal carcinoma staging. They reported that, in 58 patients with esophageal carcinoma who underwent surgery, 21 patients had lymph node metastases in resected specimens, and PET visualized the metastases in 17 patients, whereas CT was positive only in 5 patients. 1646 CANCER November 1, 1999 / Volume 86 / Number 9 We previously introduced 11C-choline PET for imaging brain tumor, lung carcinoma, esophageal carcinoma, colon carcinoma, prostate carcinoma, bladder carcinoma, and their metastases.17,18,54 11C-choline PET has a very high sensitivity compared with 18FFDG PET for the detection of small lymph node metastases. Conversely, 18F-FDG PET has a low sensitivity for the detection of small tumors. It seems that 18FFDG is incorporated actively in tumors only under the conditions that 1) the tumor is relatively large, 2) the blood supply (i.e., the oxygen supply) is short, and 3) the tumor is obtaining most of its energy from glycolysis.55– 60 If the blood supply is not deficient in the tumor, as in small tumors, then 18F-FDG may not be incorporated actively. In contrast, the rate of 11C-choline uptake is simply proportional to the rate of cell membrane synthesis, i.e., the rate of cell division, irrespective of the oxygen supply.17,18,61– 66 We have demonstrated again in the current study that 11Ccholine PET is more sensitive than 18F-FDG PET for the detection of small tumors. The current work was undertaken to evaluate the sensitivity of detecting the regional lymph node metastases involved with esophageal carcinoma using 11 C-choline PET and 18F-FDG PET. Pathologic findings were taken as the gold standard. However, it was possible that some of the small lymph nodes may have eluded pathologic detection, although surgeons elaborately carried out lymph node dissection (during and after surgery), and pathologists were meticulous in their work. An international congress has suggested that, for achieving accurate pathologic staging in patients with esophageal carcinoma, at least 15 lymph nodes, including both mediastinal and abdominal lymph nodes, should be examined.67 We examined 20 –77 lymph nodes in each case. Nevertheless, there were a number of high uptake areas with 11C-choline (sometimes together with 18F-FDG) for the areas that were possibly not dissected and not examined pathologically. This is a “sampling error”. In most of the cases, these areas were localized contralateral to the side of the surgical incision (Patients 2, 3, 6, 7, 9, 11, 15, 19, and 28 in Table 2). 11 C-choline PET was superior to 18F-FDG PET in its sensitivity for detecting small metastatic lymph nodes in the mediastinum. The minimal size of the metastatic lymph nodes detected with 11C-choline PET was 4 mm in greatest dimension. However, 11Ccholine PET was useless in imaging metastatic lymph nodes in the upper abdomen, because the uptake of 11 C-choline in the liver prevented the imaging of them. In contrast, 18F-FDG PET was effective in visualizing metastatic lymph nodes in the upper abdomen. However, it was not as effective as 11C-choline PET for detecting metastatic lymph nodes in the mediastinum, because 1) it could not visualize very small tumors in the mediastinum, and 2) sometimes, 18FFDG was incorporated into normal myocardium even in the fasting state, resulting in interference with the imaging of mediastinal lymph nodes. When 11C-choline and 18F-FDG were used in combination, the above-mentioned disadvantages were compensated, and lymph node metastases were detected in most cases regardless of their localization. We believe that the combined use of 11C-choline and 18F-FDG for PET will afford very reliable patient evaluation of regional lymph node metastases when the best surgical procedure is to be considered. REFERENCES 1. Bogomoletz WV, Molas G, Gayet B, Potet F. Superficial squamous cell carcinoma of the esophagus: a report of 76 cases and review of the literature. Am J Surg Pathol 1989;13:535– 46. 2. Goseki N, Koike M, Yoshida M. Histopathologic characteristics of early stage esophageal carcinoma: a comparative study with gastric carcinoma. Cancer 1992;69:1088 –93. 3. Sugimachi K, Ikebe M, Kitamura K, Toh Y, Matsuda H, Kuwano H. Long-term results of esophagectomy for early esophageal carcinoma. Hepatogastroenterology 1993;40: 203– 6. 4. Rice TW, Zuccaro G Jr., Adelstein DJ, Rybicki LA, Blackstone EH, Goldblum JR. Esophageal carcinoma: depth of tumor invasion is predictive of regional lymph node status. Ann Thorac Surg 1998;65:787–92. 5. Murata Y, Muroi M, Yoshida M, Ide H, Hanyu F. Endoscopic ultrasonography in the diagnosis of esophageal carcinoma. Surg Endosc 1987;1:11– 6. 6. Botet JF, Lightdale CJ, Zauber AG, Gerdes H, Urmacher C, Brennan MF. Preoperative staging of esophageal cancer: comparison of endoscopic US and dynamic CT. Radiology 1991;181:419 –25. 7. Tio TL, Cohen P, Coene PP, Udding J, den Hartog Jager FCA, Tytgat GNJ. Endosonography and computed tomography of esophageal carcinoma: preoperative classification compared to the new (1987) TNM system. Gastroenterology 1989; 96:1478 – 86. 8. Tio TL, Coene PPLO, den Hartog Jager FCA, Tytgat GNJ. Preoperative TNM classification of esophageal carcinoma by endosonography. Hepatogastroenterology 1990;37: 376 – 81. 9. Heinz A, Wahl W, Mildenberger P, Braunstein S, Georg M, Junginger T. Endosonographie beim Oesophaguscarcinom: Ergebnisse einer klinischen Untersuchung und einer Invitro-Analyse. Chirurg 1992;63:629 –33. 10. Goldschmid S, Nord J. Endoscopic diagnosis and treatment of esophageal cancer. Esophagus 1994;4:827–50. 11. Luketich JD, Kim R, Landreneau RJ, Urso K, Ferson P, Lembersky B, et al. A comparison of endoscopic ultrasound and minimally invasive surgical staging in esophageal cancer [abstract]. Am J Gastroenterol 1996;91:1892. 12. Natsugoe S, Yoshinaka H, Morinaga T, Shimada M, Baba M, Fukumoto T, et al. Ultrasonographic detection of lymphnode metastases in superficial carcinoma of the esophagus. Endoscopy 1996;28:674 –9. PET of Esophageal Carcinoma with Choline and FDG/Kobori et al. 13. Luketich JD, Schauer PR, Meltzer CC, Landreneau RJ, Urso K, Townsend DW, et al. Role of positron emission tomography in staging esophageal cancer. Ann Thorac Surg 1997;64: 765–9. 14. Luketich JD, Schauer P, Urso K, Kassis E, Ferson P, Keenan R, et al. Future directions in esophageal cancer. Chest 1998; 113:120S–2S. 15. Block MI, Patterson GA, Sundaresan RS, Bailey MS, Flanagan FL, Dehdashti F, et al. Improvement in staging of esophageal cancer with the addition of positron emission tomography. Ann Thorac Surg 1997;64:770 –7. 16. Hara T, Kosaka N, Shinoura N, Kondo T. PET imaging of brain tumor with [methyl-11C]choline. J Nucl Med 1997;38: 842–7. 17. Hara T, Kosaka N, Kishi H. PET imaging of prostate cancer using carbon-11-choline. J Nucl Med 1998;39:990 –5. 18. Gallagher BM, Fowler JS, Gutterson NI, MacGregor RR, Wan CN, Wolf AP. Metabolic trapping as a principle of radiopharmaceutical design: some factors responsible for the biodistribution of [18F]2-deoxy-2-fluoro-D-glucose. J Nucl Med 1978;19:1154 – 61. 19. Yuasa M, Yoshida H, Hara T. Computer-controlled synthesis of [18F]FDG by the tetrabutylammonium method: achievement of high yield, purity, reproducibility, reliability, and safety. Appl Radiat Isot 1997;48:201–5. 20. Hara T, Yuasa M. Automated synthesis of [11C]choline, a positron-emitting tracer for tumor imaging. Appl Radiat Isot 1999;50:531–3. 21. Zanzonico PB, Brill AB, Becker DV. Radiation dosimetry. In: Wagner HN Jr., Szabo Z, Buchanan JW, editors. Principles of nuclear medicine. 2nd ed. Philadelphia: Saunders, 1995: 106 –34. 22. Japanese Society for Esophageal Diseases. Guidelines for the clinical and pathological studies on carcinoma of the esophagus [in Japanese]. 8th ed. Tokyo: Kanehara Shuppan, 1992. 23. Lowe VJ, DeLong DM, Hoffman JM, Coleman RE. Optimum scanning protocol for FDG-PET evaluation of pulmonary malignancy. J Nucl Med 1995;36:883–7. 24. Wingo PA, Tong T, Bolden S. Cancer statistics, 1995. CA Cancer J Clin 1995;45:8 –30. 25. Yamada A, Kobayashi S. Diagnosis and treatment of early esophageal carcinoma. Zentralbl Chir 1985;110:1399 – 413. 26. Sugimachi K, Ohno S, Matsuda H, Mori M, Matsuoka H, Kuwano H. Clinicopathologic study of early stage esophageal carcinoma. Br J Surg 1989;76:759 – 63. 27. Sugimachi K, Kitamura K, Matsuda H, Mori M, Kawano H, Ide H. Proposed new criteria for early carcinoma of the esophagus. Surg Gynecol Obstet 1991;173:303– 8. 28. Haruma K, Tokutomi T, Tsuda T, Yoshihara M, Sumii K, Kajiyama G. Superficial esophageal carcinoma: a report of 27 cases in Japan. Am J Gastroenterol 1991;86:1723– 8. 29. Nabeya K, Hanaoka T, Li S, Nyumura T. What is the ideal treatment for early esophageal cancer? Endoscopy 1993;25: 670 –1. 30. Sabik JF, Rice TW, Goldblum JR, Koka A, Kirby TJ, Mendendorp SV, et al. Superficial esophageal carcinoma. Ann Thorac Surg 1995;60:896 –902. 31. Endo M, Kawano T. Detection and classification of early squamous cell esophageal cancer. Dis Esophag 1997; 10:155– 8. 32. Hennessy TP, O’Connel R. Carcinoma of the hypopharynx, esophagus and cardia. Surg Gynecol Obstet 1986;162:243–7. 33. Mannell A, Becker PJ. Evaluation of the results of oesophagectomy for oesophageal cancer. Br J Surg 1991;78:36 – 40. 1647 34. Akiyama H, Tsurumaru M, Kawamura T, Ono Y. Principles of surgical treatment for carcinoma of the esophagus. Ann Surg 1981;194:438 – 46. 35. Kato H, Watanabe H, Tachimori Y, Iizuka T. Evaluation of neck lymph node dissection for thoracic esophageal carcinoma. Ann Thorac Surg 1991;51:931–5. 36. Isono K, Onoda S, Ishikawa T, Seto H, Nakayama K. Studies on the causes of deaths from esophageal carcinoma. Cancer 1982;49:2173–9. 37. Turner GG. Excision of the thoracic oesophagus for carcinoma, with construction of an extra-thoracic gullet. Lancet 1933;2:1315– 6. 38. Stelzner F. Die abdomino-cervicale Oesophagektomie. Langenbecks Arch Chir 1981;355:63– 8. 39. Orringer MB. Transhiatal esophagectomy without thoracotomy for carcinoma of the thoracic esophagus. Ann Surg 1984;200:282– 8. 40. Rothmund M, Gamstätter G. Standardisierte Chirurgie des Ösophaguskarcinoms. Dtsch Med Wsch 1984;109:606 –12. 41. Ulrich B, Grabitz K, Kockel N. Ergebnisse der Magenhochzugsoperation ohne Thoracotomie bei der Behandelung des forgeschrittenen Ösophaguskarcinoms. Zbl Chir 1984;109: 1032– 43. 42. Goldfaden D, Orringer M, Appelmann HD, Kalish R. Adenocarcinoma of the distal esophagus and gastric cardia. J Thorac Cardiovasc Surg 1986;91:242–7. 43. Röher HD, Linn RM, Stahlknecht CD, Thon KP. Zur operativen Verfahrenswahl beim Ösophaguskarcinom. Chirurg 1988;59:582– 6. 44. Barbier PA, Becker CD, Wagner HE. Esophageal carcinoma: patient selection for transhiatal esophagectomy: a prospective analysis of 50 consecutive cases. World J Surg 1988;12: 263–9. 45. Finley RJ, Incluet RI. The results of esophagastrectomy without thoracotomy for adenocarcinoma of the esophagogastric junction. Ann Surg 1989;210:535– 43. 46. Hankins JR, Attar S, Coughlin TR, Miller JE, Hebel JR, Suter CM, et al. Carcinoma of the esophagus: a comparison of the results of transhiatal versus transthoracic resection. Ann Thorac Surg 1989;47:700 –5. 47. Tilanus HW, Hop WCJ, Langenhorst BLAM, van Lanschot JJB. Esophagectomy with or without thoracotomy: is there any difference? J Thorac Cardiovasc Surg 1993;105:898 –903. 48. Horstmann O, Verreet PR, Becker H, Ohmann C, Röher HD. Transhiatal oesophagectomy compared with transthoracic resection and systematic lymphadenectomy for the treatment of oesophageal cancer. Eur J Surg 1995;161:557– 67. 49. Quint LE, Glazer GM, Orringer MB, Gross BH. Esophageal carcinoma: CT findings. Radiology 1985;155:171–5. 50. Quint LE, Glazer GM, Orringer MB, Esophageal imaging by MR and CT: study of normal anatomy and neoplasms. Radiology 1985;156:727–31. 51. Becker CD, Barbier P, Porcellini B. CT evaluation of patients undergoing transhiatal esophagectomy for cancer. J Comput Assist Tomogr 1986;10:607–11. 52. Krasna MJ. Advances in staging of esophageal carcinoma. Chest 1998;113:107S–11S. 53. Krasna MJ, Flowers JL, Attar S, McLaughlin JS. Combined thoracoscopic/laparoscopic staging of esophageal cancer. J Thorac Cardiovasc Surg 1996;111:800 –7. 54. Hara T, Kosaka N, Kondo T, Kishi H, Kobori O. Imaging of brain tumor, lung cancer, esophagus cancer, colon cancer, prostate cancer, and bladder cancer with [C-11]choline [abstract]. J Nucl Med 1997;38:250. 1648 CANCER November 1, 1999 / Volume 86 / Number 9 55. Clavo AC, Brown RS, Wahl RL. Fluorodeoxyglucose uptake in human cancer cell lines is increases by hypoxia. J Nucl Med 1995;36:1625–32. 56. Minn H, Clavo AC, Wahl RL. Influence of hypoxia on tracer accumulation in squamous-cell carcinoma: in vitro evaluation for PET imaging. Nucl Med Biol 1996;23:941– 6. 57. Gulledge CJ, Dewhirst MW. Tumor oxygenation: a matter of supply and demand. Anticancer Res 1996;16:741–9. 58. Dang CV, Lewis BC, Dolde C, Dang G, Shim H. Oncogenes in tumor metabolism, tumorigenesis, and apoptosis. J Bioenerg Biomembr 1997;29:345–54. 59. Muthupala SP, Rempel A, Pedersen PL. Aberrant glycolytic metabolism of cancer cells: a remarkable coordination of genetic, transcriptional, post-translational, and mutational events that lead to a critical role for type II hexokinase. J Bioenerg Biomembr 1997;29:339 – 43. 60. Waki A, Fujibayashi Y, Yonekura Y, Sadato N, Ishii Y, Yokoyama A. Reassessment of FDG uptake in tumor cells: high FDG uptake as a reflection of oxygen-independent glycolysis dominant energy production. Nucl Med Biol 1997; 24:665–70. 61. Warden CH, Friedkin M. Regulation of choline kinase activity and phosphatidylcholine biosynthesis by mitogenic growth factors in 3T3 fibroblasts. J Biol Chem 1985;260:6006–11. 62. Macara IG. Elevated phosphocholine concentration in rastransformed NIH 3T3 cells arises from increased choline kinase activity, not from phosphatidylcholine breakdown. Mol Cell Biol 1989;9:325– 8. 63. Teegarden D, Taparowsky EJ, Kent C. Altered phosphatidylcholine metabolism in C3H10T[1/2] cells transfected with the Harvey-ras oncogene. J Biol Chem 1990;265: 6042–7. 64. Ronen SM, Rushkin E, Degani H. Lipid metabolism in T47D human breast cancer cells: 31P and 13C-NMR studies of choline and ethanolamine uptake. Biochim Biophys Acta 1991;1095:5–16. 65. Katz-Brull R, Degani H. Kinetics of choline transport and phosphorylation in human breast cancer cells: NMR application of the zero trans method. Anticancer Res 1996;16: 1375– 80. 66. Aiken NR, Gillies RJ. Phosphomonoester metabolism as a function of all proliferative status and exogenous precursors. Anticancer Res 1996;16:1393–7. 67. Funagalli U. Resective surgery for cancer of the thoracic esophagus: results of a consensus conference held at the Sixth World Congress of the International Society for Diseases of the Esophagus. Dis Esoph 1996;9(Suppl 1): 30 – 8.