|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2000 American Society for Clinical Oncology Utility of Positron Emission Tomography for the Staging of Patients With Potentially Operable Esophageal CarcinomaFrom the Departments of Nuclear Medicine, Internal Medicine, Radiology, Radiation Oncology, Pathology, and Thoracic Surgery, University Hospital Gasthuisberg, Katholieke Universiteit Leuven, Leuven, Belgium. Address reprint requests to Patrick Flamen, MD, Department of Nuclear Medicine, University Hospital Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium; email Patrick.Flamen{at}uz.kuleuven.ac.be
PURPOSE: A prospective study of preoperative tumor-node-metastasis staging of patients with esophageal cancer (EC) was designed to compare the accuracy of 18-F-fluoro-deoxy-D-glucose (FDG) positron emission tomography (PET) with conventional noninvasive modalities. PATIENTS AND METHODS: Seventy-four patients with carcinomas of the esophagus (n = 43) or gastroesophageal junction (n = 31) were studied. All patients underwent attenuation-corrected FDG-PET imaging, a spiral computed tomography (CT) scan, and an endoscopic ultrasound (EUS). RESULTS: FDG-PET demonstrated increased activity in the primary tumor in 70 of 74 patients (sensitivity: 95%). False-negative PET images were found in four patients with T1 lesions. Thirty-four patients (46%) had stage IV disease. FDG-PET had a higher accuracy for diagnosing stage IV disease compared with the combination of CT and EUS (82% v 64%, respectively; P = .004). FDG-PET had additional diagnostic value in 16 (22%) of 74 patients by upstaging 11 (15%) and downstaging five (7%) patients. Thirty-nine (53%) of the 74 patients underwent a 2- or 3-field lymphadenectomy in conjunction with primary curative esophagectomy. In these patients, tumoral involvement was found in 21 local and 35 regional or distant lymph nodes (LN). For local LN, the sensitivity of FDG-PET was lower than EUS (33% v 81%, respectively; P = .027), but the specificity may have been higher (89% v 67%, respectively; P = not significant [NS]). For the assessment of regional and distant LN involvement, compared with the combined use of CT and EUS, FDG-PET had a higher specificity (90% v 98%, respectively; P = .025) and a similar sensitivity (46% v 43%, respectively; P = NS). CONCLUSION: PET significantly improves the detection of stage IV disease in EC compared with the conventional staging modalities. PET improves diagnostic specificity for LN staging.
PRETREATMENT ASSESSMENT and classification of disease extent is essential in the management of esophageal cancer (EC).1,2 The tumor stage is the major determinant of prognosis and provides the basis for selection of the most appropriate therapeutic strategy. The therapeutic options in EC management include radical surgery in early disease, multimodal treatment schemes combining neoadjuvant chemotherapy and radiotherapy followed by surgery in locoregional advanced disease, and palliative schemes in case of distant metastatic disease. The current American Joint Committee on Cancer staging system for EC is entirely tumor-node-metastasis based.3,4 The pivotal variables of this stage system are the depth of wall penetration of the primary tumor (T stage), the presence of locoregional lymph node (LN) metastasis (stage IIB and stage III), and of distant LN or organ metastasis (stage IV). The standard noninvasive staging modalities are computed tomography (CT) of the chest and abdomen, for the evaluation of local tumor extent and for the detection of distant metastases, and endoscopic ultrasound (EUS), for the evaluation of tumor depth and locoregional LN staging in nonobstructing EC.2 However, these techniques entirely depend on structural characteristics for diagnosis. This inevitably causes limitations in diagnostic specificity (false-positive findings in enlarged inflammatory LN) and sensitivity (false-negative findings in nonenlarged invaded LN). Consequently, there is a clear need for more accurate preoperative staging.3-8 Some centers propagate the routine use of the minimally invasive surgical staging procedures, consisting of a thoracoscopy combined with a staging abdominal laparoscopy, thus aiming at a greater accuracy in the evaluation of regional and celiac LN and allowing the detection of unimaged pleural or peritoneal disease.9-11 However, the long surgery and hospitalization times and the high costs of these invasive procedures strongly limit their routine implementation. Recent reports suggested a role for positron emission tomography (PET) using the radiolabeled glucose analog 18-F-fluoro-deoxy-D-glucose (FDG) for preoperative staging of EC.12-15 The depiction of neoplastic foci by FDG-PET relies on the increased accumulation of the radiotracer in malignant tissues, which is believed to be the result of an increased expression of glucose transport enzymes in the tumoral cell membrane together with an increased activity of the enzymes of the first steps of the glycolytic pathway.16,17 Recent advances in technology allow the sensitive screening of the whole body in one examination session.18 The degree of FDG accumulation in a lesion can be assessed semiquantitatively and expressed as a standard uptake value. We hypothesized that FDG-PET, as a metabolism-based diagnostic modality, could improve the currently used structure-based noninvasive pretreatment staging, based on an increase of both sensitivity and specificity, thereby optimizing the stage-dependent therapeutic management in EC. Therefore, a prospective study was designed to compare the staging accuracy of FDG-PET with the standard combined use of CT and EUS.
Patients Patients with newly diagnosed biopsy-proven EC who were sent to the University Hospital of Leuven for evaluation of resectability were eligible. Exclusion criteria were prior EC treatment, diabetes mellitus, inflammatory lung disease, and inoperability for medical reasons. All patients underwent standard staging procedures, including history and physical examination, laboratory tests, an ultrasound examination of the neck, a barium esophagogram, bronchoscopy, spiral CT of the chest and abdomen, and a transesophageal EUS. Within the same week, an FDG-PET scanning was performed.
Spiral CT of the Chest and Abdomen LN measuring 10 mm or more at their maximum cross-sectional diameter were considered to be metastatic. All examinations were prospectively interpreted by a chest radiologist who was unaware of the results of the other imaging modalities.
EUS Endosonographic staging of tumor infiltration was relative to the five-layered structure of the gastrointestinal tract wall. Endosonographic criteria for LN metastasis were based on size, shape, margins, and echo-pattern. According to these characteristics, LN were classified as (probably) malignant or (probably) benign. Examinations were performed by one of three examiners with 4 to 12 years of experience. All patients were examined prospectively and blinded to the results of the other noninvasive diagnostic modalities.
FDG-PET For precise spatial localization of the PET lesions, an automated registration of the transaxial PET and CT slices was performed. For this, the reconstructed PET transmission images (and hence the PET emission images) and the CT images were registered using an algorithm based on information theory, maximizing the mutual information between the intensities of both images.21 An example of the automated image registration we used in the current study is shown in Fig 1.
Patient Management The PET, CT, and EUS findings, together with all other preoperative staging data, were reviewed and correlated at a multidisciplinary tumor conference involving thoracic surgeons, medical oncologists, a pathologist, radiation oncologists, a diagnostic radiologist, and a nuclear medicine physician. Discordant PET data that affected patient staging and/or management were taken into account only if confirmed by histology or by dedicated radiographic techniques. The new tumor-node-metastasis classification (fifth edition) proposed by the International Union Against Cancer was used for staging.22
Surgery If the primary tumor was located above the level of the carina, a posterior mediastinectomy was performed through a right thoracotomy. The operation is continued through a laparotomy performing a lymphadenectomy of the superior abdominal compartment, as described above. The abdominal lymphadenectomy and posterior mediastinectomy is called a two-field lymphadenectomy. A cervical location of the anastomosis allowed the performance of a cervical lymphadenectomy, the so-called third-field lymphadenectomy. This dissection includes, bilaterally, LNs lateral to the carotid vessels, the internal jugular, and supraclavicular nodes, the nodes along the recurrent nerves down to the point where the intrathoracic LN dissection ended. If the primary tumor invaded the gastric wall more than 5 cm, a total gastrectomy with a jejuno-esophagostomy on the thoracic esophagus was performed. In that case, the lymphadenectomy was limited to the abdominal and lower thoracic LN.
Data Analysis The gold standard for T stage was exclusively defined by histology and was, therefore, only available in the subgroup of patients in whom a tumorectomy was performed, whether curative or palliative. For the assessment of LN involvement, the gold standard was exclusively defined by histologic examination of the materials obtained in the patients in whom a two-field or three-field lymphadenectomy, in conjunction with esophagogastrectomy, was performed. For optimal correlation between imaging and histology results, all data were assigned to the following six anatomic regions per patient: local (LN located less than 3 cm from the primary tumor); supraclavicular (including supraclavicular and cervical LN); upper mediastinal (supracarinal); lower mediastinal (infracarinal); regional abdominal (including perigastric LN; excluding coeliac artery LN); and retroperitoneal (including coeliac artery and common hepatic and splenic artery LN). For the analysis of the accuracy of the combined use of CT and EUS, the positive results of both techniques were cumulated. Thus, a positive result with one technique overruled a negative result with the other.
Statistics
Patient Characteristics Seventy-four consecutive patients were included between October 1997 and December 1998. The primary carcinomas were located in the middle (n = 13) and distal esophagus (n = 30) or at the gastroesophageal junction (n = 31). Histology was squamous cell carcinoma (n = 21) and adenocarcinoma (n = 53). An overview of the tumor histology and localization in relation to therapeutic management is listed in Table 1. Eight of the 47 patients who underwent surgery with curative intent did not undergo an extensive two- or three-field lymphadenectomy. The motivation for this was the presence of a limited tumor load (T1, n = 1) or because a total gastrectomy was performed in case of major gastric wall invasion (n = 7). Reasons for not performing the cervical (ie, third-field) lymphadenectomy were diverse (n = 13) and included advanced biologic age, comorbidity, and/or peroperative instability.
Primary Tumor Assessment: T Stage FDG-PET demonstrated intense focal increased activity in the primary esophageal carcinoma in 70 of the 74 patients (sensitivity: 95%). The mean (± SD) SUV was 13.5 (± 7.3). False-negative PET images were found in four patients, all having small pT1 lesions (diameter less than 8 mm). The accuracy of FDG-PET and EUS for T-stage diagnosis was determined in the patient subset (n = 50) who underwent a tumorectomy. No correlation between intensity of the primary tumor FDG uptake, expressed as SUV, and the pT stage was found. High-grade malignant strictures at the primary tumor site precluded passage of the echoendoscope in 19 (26%) of 74 patients. The results of EUS are listed in Table 2. The accuracy of EUS for assessing the depth of tumor infiltration in the 42 patients with resectable nonobstructing EC was 27 (64%) of 42. EUS overstaged the T stage in eight (19%) of 42 patients and understaged the T stage in seven (17%) of 42 patients.
Detection of Stage IV Disease Thirty-four (46%) of the 74 patients had stage IV disease, based on the presence of malignancy in 40 lesions (18 organ metastases and 22 distant LN metastases). The proof of malignancy was based on the histology of 34 lesions (preoperative biopsy, nine lesions; diagnostic laparoscopy, four lesions; preoperative biopsy, three lesions; and surgical lymphadenectomy, 18 lesions) and by dedicated radiologic techniques or follow-up of six lesions. The sensitivity, specificity, and accuracy of PET, CT, and EUS for the diagnosis of stage IV disease on a patient base are listed in Table 3. FDG-PET had a higher accuracy compared with the combined use of CT and EUS (82% v 64%, respectively; P = .004), mainly by virtue of a superior sensitivity (74% v 47%, respectively). The analysis of the concordance between FDG-PET and the combination of CT and EUS for the diagnosis of stage IV disease is listed in Table 4. The patients were divided into two subgroups according to the concordance (n = 56) or discordance (n = 18) between the findings of FDG-PET and the combined use of CT and EUS. Through comparison with the gold standards, all diagnoses were then classified as true-positive, true-negative, false-positive, and false-negative. In the 18 patients with discordant results, FDG-PET was correct in 16 (89%), which resulted in the upstaging of 11 patients (15%) from M0 to M1 disease, and in the downstaging of five patients (7%) from M1 to M0 disease. In the 11 patients upstaged by FDG-PET, EUS and CT reported a T3N1 classification in 10 patients and a T2N1 stage in one patient. The FDG-PET lesions responsible for the upstaging were located in distant LNs of the supraclavicular and retroperitoneal region (n = 5), the liver (n = 2), the pleura (n = 1), the chest wall (n = 1), and in bone (n = 2). In one patient, PET detected a second primary carcinoma of the glottis. An illustrative example of the added value of FDG-PET is shown in Fig 2. In two other discordant patients, however, FDG-PET falsely understaged the M stage because of false-negative findings in two involved supradiaphragmatic LN, which were correctly reported by EUS.
In 20 (59%) of the 34 patients with stage IV disease, CT was negative for organ metastases. In this patient subgroup, EUS was positive for stage IV disease in three patients (sensitivity: 15%), negative in 14 patients (70%), and inconclusive because of incomplete scope passage in three patients (15%), whereas FDG-PET was true-positive in 12 patients (sensitivity: 60%). In the 17 stage-IV patients with negative or inconclusive EUS findings, FDG-PET correctly detected stage IV disease in 11 patients. On a lesion basis, FDG-PET was false-negative in five (28%) of 18 organ metastases (two peritoneal metastases, two small [< 10 mm] and superficially located liver metastases, and one lung metastasis). All these lesions were also missed by spiral CT and were only seen during surgery. PET was false-negative in seven (32%) of 22 distant LN metastases. Histology of these lesions indicated limited, micrometastatic disease in two normal-sized LN and gross invasion of two clinically enlarged LNs located in the supraclavicular area (n = 1) and around the splenic artery (n = 1). Three other false-negative LNs were found in the supradiaphragmatic region in the immediate vicinity of the primary tumor located at the gastroesophageal junction. The following false-positive PET results were found in four patients: two focal lung lesions (one granuloma and one without histologic diagnosis but with negative follow-up), one mediastinal LN with active inflammation on histology, and one in the hilus of the spleen, which was not verified during laparotomy.
Diagnosis of LN Involvement
For local LN, the sensitivity of FDG-PET was lower than EUS (33% v 81%, respectively; P = .027), but the specificity may have been higher (89% v 67%, respectively; P = not significant [NS]). CT scanning did not detect any local LN (sensitivity: 0%). For the assessment of regional and distant LN involvement, FDG-PET, compared with the combined use of CT and EUS, had a higher specificity (98% v 90%, respectively; P = .025) and a similar sensitivity (43% v 46%, respectively; P = NS). False-positive PET lesions were found in six LN regions in six patients. In two of these patients, histology indicated the presence of inflammation in enlarged mediastinal LN. Three other false-positive PET lesions were caused by a heterogenous tracer uptake in the primary tumor, which was incorrectly considered as reflecting local LN metastases. In another patient, PET indicated the presence of a focal lesion located in the hilus of the spleen, of which no evidence was found during surgery and which did not manifest itself during follow-up. False-negative PET lesions were found in 33 LN areas. Fourteen of these lesions (42%) were located in the immediate vicinity of the primary tumor. In 10 of the 19 nonlocal false-negative LN, histology reported the presence of micrometastases, with only partial invasion of the LN. In a patient-based analysis, the gold standard indicated the absence of LN involvement (N0 classification) in 11 patients, local LN involvement in eight patients, regional LN involvement in 10 patients, and distant LN involvement (M1 disease) in 10 patients. Table 6 lists the LN staging accuracy of FDG-PET, EUS, and CT. FDG-PET understaged the extent in 19 (49%) of 39 patients, whereas the combination of CT and EUS overstaged the LN status in 14 (36%) of 39 patients.
FDG-PET underestimated the extent of LN metastasis because of false-negative results in six local, seven regional, and six distant LN. PET overestimated the extent of LN metastasis in three (8%) of 39 patients, based on false-positive results in one local, one regional, and one distant LN.
Several pilot studies reported a high sensitivity of FDG-PET for primary squamous cell carcinoma and adenocarcinoma of the esophagus and of the gastroesophageal junction.12-15 False-negative FDG-PET findings were always related to small volume (Tis or T1) tumors. Our study confirmed the high sensitivity (95%) of FDG-PET for primary tumor visualization. False-negative PET imaging was found in four patients (5%) with small T1 tumors, suggesting that limitations in spatial resolution of the PET imaging device, currently around 8 mm, are the only cause for nonvisualization of primary EC. We found no significant relationship between the primary tumor SUV and the depth of tumor invasion (T classification) or between the primary tumor SUV and the extent of LN metastasis. These results are in conformity with an earlier report by Fukunaga et al14 who, moreover, demonstrated that the uptake was not related to clinicopathologic tumor grading. EUS has recently emerged as an important tool for the evaluation of the T stage. In our study population, the accuracy of EUS for the prediction of T stage was only 64%. The inaccuracy consisted of overstaging the T stage in 19% and understaging it in 17% of the patients. The accuracy of EUS for preoperative T-stage prediction was, in our experience, clearly lower than according to recent reports, where it ranged from 75% to 97%.2-8 This is probably related to a bias in patient selection in this study. Indeed, only those patients who underwent an esophagectomy were included for T-stage analysis, thus eliminating many T4 cases, which are relatively easy to distinguish from T1-3 cases. Our study in effect largely concerned T1-3 lesions in which differentiation by EUS is more challenging. In patients with stage IV EC (distant LN involvement or organ metastasis), surgery with a curative intent is contraindicated.2 This study found that FDG-PET had a superior accuracy for the diagnosis of stage IV disease compared with the combined use of CT and EUS (82% v 64%, respectively; P = .004). FDG-PET had an additional diagnostic value in 16 (22%) of 74 patients, by upstaging 11 (15%) and downstaging five patients (7%). These findings are in line with recent reports by Block et al12 and Luketich et al,24 who found metastases detected only by FDG-PET in 20% and 9%, respectively, of the studied patients. Interestingly, we found that 10 of the 11 patients in whom FDG-PET detected unsuspected distant metastases were staged by EUS as T3N1. It is known that in EC the frequency of lymphatic metastases, and therefore the probability of a positive PET scan, is related to increasing depth of tumor invasion.25 In this study design, CT, EUS, and FDG-PET were performed independently, and the results were blinded. In most staging algorithms, however, CT is performed first to assess operability and detect distant metastasis and is followed by EUS in the CT-negative patients. Lumping EUS and PET in the data with both CT-negative and -positive patients might have caused some bias. Therefore, a subanalysis was performed assessing the sensitivity for detection of stage IV disease of EUS and FDG-PET in the subset of patients in whom CT did not indicate organ metastatic disease. The sensitivity of EUS in this subgroup (including the three patients in whom endoscopy was inconclusive because of an incomplete passage of the scope) was only three (15%) of 20 patients compared with 12 (60%) of 20 patients with FDG-PET, confirming the superiority of the latter technique in this regard. Our study confirmed that EUS is the preferred method for the assessment of local LN involvement, with a superior sensitivity (81%) compared with CT (0%) or FDG-PET (33%). EUS, however, clearly suffers from a lack of specificity (67%), resulting in the overstaging of seven out of the 11 patients with pN0 disease. This is in line with a report by Botet et al8 who found a similar degree of overstaging by EUS in five (36%) of 14 of pN0 patients. Recent reports on the use of trans-EUS–guided LN biopsies promise new possibilities for more specific minimally invasive staging of EC.26 The inaccuracy of FDG-PET for the detection of local LN is not unexpected because this is a well known limitation also reported in other types of tumors such as primary colorectal or lung cancer. Thus, FDG-PET is not able to distinguish reliably between N0 and N1 disease. For the assessment of regional and distant LN involvement, FDG-PET, compared with the combined use of CT and EUS, had a higher specificity (98% v 90%, respectively; P = .025) and a similar sensitivity (43% v 46%, respectively; P = NS). However, on a patient base, these differences did not lead to an increase of LN staging accuracy using FDG-PET compared with the currently standard combined use of CT and EUS. This study also demonstrates that the underlying type of staging error depends on the modality used. PET understaged the extent in 19 (49%) of 39 patients, whereas the combination of CT and EUS overstaged the LN extent in 14 (36%) of 39 patients. This finding is in contrast with the high negative predictive value of FDG-PET for mediastinal staging of non–small-cell lung cancer recently published by our group.27 A plausible reason for this discrepancy would be a higher incidence of micrometastatic LN involvement in EC. Indeed, most false-negative PET lesions were found in LN with only partial or microscopic tumor invasion, indicating that limitation of spatial resolution of the PET apparatus is a major source of false-negative results. The accuracy of FDG-PET for LN staging found in the present study is clearly inferior to the available literature data. Flanagan et al,13 in a retrospective study, found an accuracy for LN staging of 76%; Block et al12 reported a sensitivity of 52%; and Kole et al28 reported a sensitivity of 92%. The major reason for the lower sensitivities found in our study compared with literature data is related to the quality of the gold standard. This is indeed the first study that assessed the accuracy of FDG-PET in reference to the histology of all relevant LN regions obtained through extensive two- or three-field lymphadenectomies. Moreover, in previous studies, esophagectomies were most often performed using transhiatal techniques, which presents greater difficulty for complete removal of LN and thus a potential for underestimation of the extent of LN involvement. This study proved that the diagnostic specificity of FDG-PET was significantly higher compared with the currently standard combined use of CT and EUS. PET specificity was 89% for local LN and 98% for regional and distant LN. Still, on a patient base, four patients were incorrectly overstaged by PET because of false-positive FDG accumulation in inflammatory LNs. Therefore, even if the positive predictive value of FDG-PE appears very high, we still recommend confirmation of the malignant nature of the PET lesions that could lead to a change in therapeutic management. The analysis of the accuracy of EUS for assessing LN involvement was performed in the subset of patients that underwent curative surgery for a nonmetastatic (M0) EC. Thus, all patients with a positive CT and/or a positive FDG-PET lesion in distant (M1) sites were excluded from the analysis, together with the patients with T4 lesions. This introduces a certain bias because the accuracy of these instruments tends to increase in case of nonobstructive advanced disease stages. This results in an underestimation of the real accuracy and explains the worse results found in this study compared with literature data. It is well known that EUS staging inaccuracies occur in cases where the entire tumor volume was not scanned because of partial obstruction and unwillingness to dilate. Therefore, in assessing the staging accuracy of EUS, the authors excluded these patients in their analysis. On the other hand, in assessing the staging accuracy of the combined use of EUS and CT, these patients were included, thereby considering only the CT results and rejecting the EUS information. The use of newer thinner EUS probes that are now available obviate the problem of nonpassage and could indeed lead to superior EUS accuracy, potentially decreasing the added value of FDG-PET.29 Based on these study results, the authors state that preoperative FDG-PET should be routinely used in patients in whom the standard staging algorithm (ie, CT scan followed by an EUS) suggests resectable locally advanced disease. In this patient subset, FDG-PET should improve the detection of stage IV disease and increase the specificity of LN staging. However, because FDG is not a tumor-specific tracer, histologic or radiographic confirmation is mandatory before a patient is considered as having unresectable disease based on PET imaging.30
We thank the whole nuclear medicine PET team, including all technologists of the imaging department and of the cyclotron facility, for their enthusiasm and support. We also thank S. Vleugels, for his dedicated assistance in PET-imaging acquisition and processing, and Dr Jan Bernheim, Department of Human Ecology, Vrije Universiteit Brussel, for the linguistic support.
1. American Joint Committee on Cancer: Manual for Staging of Cancer (ed 3). In, Beahrs OH, Henson DE, Hutter RVP, et al (eds): Philadelphia, PA, JB Lippincott, 1998, pp 63-67 2. Lightdale CJ: Practice guidelines: Esophageal cancer. Am J Gastroenterol 94: 20-29, 1999[Medline] 3. Rice TW, Boyce GA, Sivak MV: Esophageal ultrasound and the preoperative staging of carcinoma of the esophagus. J Thorac Cardiovasc Surg 101: 536-544, 1991[Abstract] 4. Zuccaro G Jr, Sivak MV, Rice TW: Endoscopic ultrasound and the staging of esophageal and gastric cancer. Gastrointest Endosc Clin N Am 2: 625-636, 1992 5. Tio Tl, Coene PP, den Hartog Jager FC, et al: Preoperative TNM classification of esophageal carcinoma by endosonography. Hepatogastroenterology 37: 376-381, 1990[Medline] 6. Maerz LL, Deveney CW, Lopez RR, et al: Role of computed tomographic scans in the staging of esophageal and proximal gastric malignancies. Am J Surg 165: 558-560, 1993[Medline] 7. Tio TL, Cohen P, Coene PP, et al: Endosonography and computed tomography of esophageal carcinoma. Gastroenterology 96: 1478-1486, 1989[Medline]
8.
Botet JF, Lightdale CJ, Zauber AG, et al: Preoperative staging of esophageal cancer: Comparison of endoscopic US and dynamic CT. Radiology 181: 419-425, 1991 9. Sugarbaker DJ, Jalitsch MT, Liptay MJ: Thoracoscopic staging and surgical therapy for esophageal cancer. Chest 107: 218S-223S, 1995[Medline] 10. Bonavina L, Incarbone R, Lattuada E, et al: Preoperative laparoscopy in management of patients with carcinoma of the esophagus and of the esophagogastric junction. J Surg Oncol 65: 171-174, 1997[Medline]
11.
Luketich JD, Schauer P, Landreneau R, et al: Minimally invasive surgical staging is superior to endoscopic ultrasound in detecting lymph node metastases in esophageal cancer. J Thorac Cardiovasc Surg 114: 817-823, 1997
12.
Block MI, Patterson GA, Sundaresan RS, et al: Improvement in staging of esophageal cancer with the addition of positron emission tomography. Ann Thorac Surg 64: 770-777, 1997
13.
Flanagan FL, Dehdashti F, Siegel BA, et al: Staging of esophageal cancer with 18F-fluorodeoxyglucose positron emission tomography. AJR 168: 417-424, 1997
14.
Fukunaga T, Okazumi S, Koide Y, et al: Evaluation of esophageal cancers using fluorine-18-fluorodeoxyglucose PET. J Nucl Med 39: 1002-1007, 1998 15. Luketich JD, Schauer P, Urso K, et al: Minimally invasive surgical biopsy confirms PET findings in esophageal cancer. Surg Endosc 11: 1213-1215, 1997[Medline]
16.
Warburg O: On the origin of cancer cells. Science 123: 309-314, 1956 17. Pauwels E, McCready VR, Stoot JH, et al: The mechanism of accumulation of tumour-localising radiopharmaceuticals. Eur J Nucl Med 25: 277-305, 1998[Medline]
18.
Dahlbom M, Hoffman EJ, Hoh CK: Whole-body positron emission tomography: I. Methods and performance characteristics. J Nucl Med 33: 1191-1199, 1992 19. Shepp LA, Vardi Y: Maximum likelihood reconstruction for emission tomography. IEEE Trans Med Imag 1: 113-122, 1982 20. Lange K, Carson R: EM reconstruction algorithm for emission and transmission tomography. J Comp Ass Tomography 8: 306-316, 1984 21. Maes F, Collignon A, Vandermeulen D, et al: Multi-modality image registration by maximization of mutual information. IEEE Trans Med Imag 16: 186-197, 1997 22. Sobin LH, Wittekind C: TNM Classification of Malignant Tumors (ed 5). New York, NY, Wiley-Liss, 1997 23. Beck J: Likelihood ratios: Another enhancement of sensitivity and specificity. Arch Pathol Lab Med 110: 685-686, 1986[Medline]
24.
Luketich JD, Friedman DM, Weigel TL, et al: Evaluation of distant metastases in esophageal cancer: 100 consecutive positron emission tomography scans. Ann Thorac Surg 68: 1133-1136, 1999
25.
Rice TW, Zuccaro G Jr, Adelstein DJ, et al: Esophageal carcinoma: Depth of tumor invasion is predictive of regional lymph node status. Ann Thorac Surg 65: 787-792, 1998 26. Chang KJ, Katz KD, Durbin TE, et al: Endoscopic ultrasound-guided fine-needle aspiration. Gastrointest Endosc 40: 694-699, 1994[Medline] 27. Vansteenkiste JF, Stroobants SG, De Leyn PR, et al: Lymph node staging in non-small-cell lung cancer with FDG-PET scan: A prospective study on 690 lymph node stations from 68 patients. J Clin Oncol 16: 2142-2149, 1998[Abstract] 28. Kole AC, Plukker JT, Nieweg OE, et al: Positron emission tomography for staging of oesophageal and gastroesophageal malignancy. Br J Cancer 9: 1863–1873, 1998 29. Mallery S, Van-Dam J: Increased rate of complete EUS staging of patients with esophageal cancer using the nonoptical, wire-guided echoendoscope. Gastrointest Endosc 50: 53-57, 1999[Medline] 30. Strauss LG: Fluorine-18 deoxyglucose and false-positive results: A major problem in the diagnostics of oncological patients. Eur J Nucl Med 23: 1409-1415, 1996[Medline] Submitted December 16, 1999; accepted May 9, 2000. This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
