Japanese Journal of Clinical Oncology 32:340-346 (2002)
© 2002 Foundation for Promotion of Cancer Research
Evaluation of Esophageal Cancer by Positron Emission Tomography
Department of Surgery, Tokai University School of Medicine, Isehara, Kanagawa, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
Background: A retrospective study was performed to determine the indications for positron emission tomography (PET) using [18F]fluorodeoxyglucose (FDG) in patients with esophageal cancer, including those with early cancer, and to investigate whether the tumor-to-normal ratio (T/N ratio) could be used as a substitute for the standardized uptake value (SUV).
Methods: Thirty-six patients were included in the study. Thirty-one patients who had 36 biopsy-proven lesions (35 squamous cell carcinomas and one small cell carcinoma) underwent PET study prior to treatment. PET images were evaluated visually and the relationship between the depth of invasion and the PET findings were examined in 22 lesions of 19 patients from whom specimens were obtained from the primary tumor by surgery or endoscopic mucosal resection. PET results were also compared with computed tomography (CT) and endoscopic ultrasonography (EUS) for detection of regional lymph node metastases in 18 patients who underwent extended lymph node dissection. Five patients underwent PET studies for the detection of recurrence and the PET findings were compared with their CT findings. The T/N ratio and the SUV were calculated for 20 primary tumors.
Results: Among the 15 tumors that were pT1b or greater, all 15 were positive on PET and all seven of the lesions confined to the mucosa (Tis or T1a) were negative. The sensitivity, specificity and accuracy of detecting nodal involvement were, respectively, 37.5, 96.1 and 88.3% by CT, 30.8, 88.5 and 81.0% by EUS and 41.7, 100 and 92.2% by PET. More sites of recurrence were detected by PET than by CT. There was no statistically significant correlation between the SUV and the T/N ratio.
Conclusions: PET imaging can detect primary esophageal cancer with a depth of invasion of T1b or greater, but Tis and T1a tumors are undetectable. PET seems to be more accurate than CT or EUS for diagnosing lymph node metastasis. The T/N ratio cannot be used as a substitute for the SUV.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
Positron emission tomography (PET) using [18F]fluorodeoxyglucose (FDG) has been applied to the diagnosis of cancer in various organs (1,2). In patients with esophageal cancer, FDG PET is reported to be useful for pretreatment staging (3–5), detection of lymph node metastasis (6) and distant metastasis (7), preoperative evaluation of prognosis (8) and assessment of the response to chemotherapy (9). The accuracy of this technique for detecting primary lesions has also been shown to be high (97–100%). However, the majority of the previous studies were conducted in patients with advanced cancer.
A large number of early cancers of the esophagus are now being detected following recent developments in endoscopic staining methods (10). Therefore, we performed a retrospective study to define better the indications for FDG PET imaging in patients with esophageal cancer, including those with early cancer.
To quantitate glucose metabolism during PET, the standardized uptake value (SUV) is calculated (11). To calculate the SUV, transmission scans for attenuation correction must be performed. With the latest whole-body PET scanners, attenuation correction by transmission scanning is sometimes omitted to shorten the examination time. In addition, it has been reported that the tumor-to-normal ratio (T/N ratio) evaluated using emission scans alone is correlated with the SUV (12). Therefore, we investigated whether the T/N ratio could be used as a substitute for the SUV in patients with esophageal cancer.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
We studied 36 patients with esophageal cancer who underwent FDG PET over a period of 6 years and 1 month between April 1994 and May 2000. They comprised 31 men and five women with a mean age of 59.2 years (range: 43–78 years). FDG PET was performed before the treatment of esophageal cancer in 31 patients and after surgery in the other five patients. Among the 31 patients in whom imaging was performed before treatment, 23 underwent surgery (five of them had surgery after receiving neoadjuvant chemotherapy), one underwent endoscopic mucosal resection (EMR) and seven had inoperable cancer that was treated with radiochemotherapy. The patients’ characteristics are shown in Tables 1 and 2.
|
|
Fluorine-18 was produced using an on-site medical cyclotron (Cypris HM-18, Sumitomo Heavy Industries, Tokyo, Japan). Scans were obtained using a whole-body PET scanner (ECAT EXACT 47, Siemens/CTI, Knoxville, TN, USA). Informed consent was obtained from all patients. The patients fasted for at least 6 h prior to PET. Transmission scans for attenuation correction were performed for 3 min in each bed position (in 31/36 patients). Forty-five minutes after administration of 260 MBq (7 mCi) of FDG, emission scanning was performed from the level of the pelvis to the maxilla for 7 min in each bed position. Gray-scale hard-copy images of the transaxial sections were printed and evaluated visually by two physicians (S.H. and S.Y.). Images of coronal and sagittal sections were also available and other test findings were referred to when evaluating the PET images. The PET results were compared with the final diagnosis obtained by surgery, and also with the clinical course and the results of computed tomography (CT), endoscopic ultrasonography (EUS) and other imaging techniques.
The following four parameters were investigated.
PET Features of the Primary Tumor
Thirty-six lesions were confirmed by endoscopic biopsy in the 31 patients who underwent PET before treatment and the PET findings of these 36 histologically proven lesions were investigated. The relationship between the depth of mural invasion and the PET findings was also examined in 22 lesions of 19 patients from whom specimens were obtained from the primary tumor by surgery or EMR.
PET Findings in the Regional Lymph Nodes
The findings obtained by PET, contrast-enhanced CT and EUS were compared for the detection of regional lymph node metastases in 18 patients who underwent extended lymph node dissection during surgery after PET. Contrast-enhanced CT was also performed in all 18 patients and EUS (EU-M20, Olympus Optical, Tokyo, Japan) was performed in 13 patients.
Diagnosis of Recurrence
This was investigated in five patients who underwent PET to identify recurrence of esophageal cancer.
Comparison of the SUV with the T/N Ratio
Among 25 patients (25 lesions) with a positive main tumor on PET, the SUV and the T/N ratio were compared in 20 patients (20 lesions) who underwent transmission scans. A circular region of interest (ROI) was drawn around the part of the main lesion that showed abnormal accumulation and the SUV was then calculated based on the patient’s total body weight. The T/N ratio was determined using ROIs placed over the main tumor and the mediastinum. The ROI for the main tumor virtually encircled the lesion, while the ROI for the mediastinum was of an arbitrary size and was located at a site that was not affected by the heart or the main tumor. Both the mean T/N ratio (mean T/N) and the maximum T/N ratio (max. T/N) were calculated. Correlations between the SUV and the T/N ratio were analyzed by calculating Spearman’s rank correlation coefficients.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
PET Features of the Primary Tumor
Of the 36 biopsy-proven malignant primary tumors in the 31 patients, 25 (69.4%) were positive on PET and the remaining 11 (30.6%) were negative.
Table 3 shows the PET features of the 22 tumors in the 19 patients in whom the depth of mural invasion was determined by surgery or EMR. The depth of invasion was classified according to the Guidelines for Clinical and Pathologic Studies of Carcinoma of the Esophagus released by the Japanese Society for Esophageal Disease (13). All 15 tumors (100%) with a depth of pT1b or more (involving the submucosa) were positive on PET. In contrast, all seven tumors (100%) that were confined to the mucosa (Tis or T1a) could not be detected by PET. An example of a positive tumor (case No. 30) is shown in Fig. 1.
|
|
PET Findings in the Regional Lymph Nodes
Regional lymph node dissection was performed for a total of 179 stations (node groups) in 18 of the 31 patients who underwent surgery with extended lymph node dissection. Classification of the regional lymph nodes was based on the Guidelines for Clinical and Pathologic Studies of Carcinoma of the Esophagus published by the Japanese Society for Esophageal Disease (13). The detection of lymph node metastasis by PET is shown in Table 4. On microscopy, regional lymph node metastasis was observed in 24 of the 179 stations. The PET and histopathology results were compared for each station because it was difficult to compare the results for individual lymph nodes. The sensitivity, specificity and accuracy of detecting nodal involvement by CT, respectively, were 37.5, 96.1 and 88.3% while the values for EUS were 30.8, 88.5 and 81.0% and those for PET were 41.7, 100 and 92.2%. PET was therefore superior to CT and EUS. Comparison of the sensitivity of PET, CT and US showed no significant difference among the three methods (Fisher’s exact probability test). In six patients, PET correctly identified extensive lymph node metastases and it was decided to use neoadjuvant radiochemotherapy instead of surgery. PET did not yield any false-positive results with regard to primary or regional lymph node lesions. Fig. 2 shows a patient (case No. 31) with lymph node metastases detected by PET.
|
|
Diagnosis of Recurrence
CT revealed a total of six lesions (five lymph node metastases and one pulmonary metastasis) in two of the five patients who underwent PET for the diagnosis of recurrence. In all six lesions, PET showed abnormal accumulation of FDG. PET also identified abnormal FDG accumulation at three other sites in two of the patients. Monitoring of the clinical course and follow-up diagnostic imaging after PET showed that these three sites were actually lymph node metastases.
PET revealed no abnormal findings in two of the three patients without abnormalities on conventional diagnostic imaging. No recurrence occurred during subsequent follow up for 12 months and nothing was detected by other imaging modalities, indicating that the PET results were true negative findings.
The other patient underwent PET 6 months after surgery, at which time abnormal FDG accumulation was noted at the thoracotomy rib resection site. This patient died of pulmonary and hepatic metastases at 1 year after surgery. It was not possible to determine whether there was a rib recurrence at the operation site. Fig. 3 shows PET imaging of recurrence in this patient (case No. 36).
|
Comparison of the SUV with the T/N Ratio
Table 1 shows the SUV and T/N ratio values for the 20 patients (20 lesions) in whom the main tumor was positive on PET and who underwent transmission scanning. Statistical analysis revealed no significant correlation between the SUV and the mean T/N (P = 0.246). Negative numerical values were sometimes obtained when calculating the mean T/N. The max. T/N of the ROI was therefore determined and the results were re-examined. There was a weak correlation between the SUV and the max. T/N (P = 0.08), but it was not statistically significant.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
The depth of mural invasion of the main tumor and the presence or absence of lymph node and distant metastases are important factors when deciding the management of patients with esophageal cancer. In patients who have early cancer without lymph node metastasis, complete cure can be expected after minimally invasive EMR (10). However, in patients with lymph node metastasis, three-field dissection involving thoracotomy and laparotomy is required to clear the cervical, thoracic and abdominal lymph nodes. This is far more invasive than EMR and also has a marked impact on the postoperative quality of life.
In patients with esophageal cancer, contrast radiography and endoscopy are excellent for identifying the main tumor. However, CT alone has limitations for detecting lymph node metastasis, with an accuracy of ~60% (14). EUS has a high sensitivity, but its specificity is inadequate, so diagnostic accuracy is improved when the EUS findings are combined with those of other tests (14–16).
FDG PET is an imaging method based on glucose metabolism. The metabolism of glucose is accelerated in cancer cells (17) and tumors with a high metabolic rate are reported to be highly aggressive (1). FDG is a glucose analogue that accumulates at sites of high glucose metabolism and is used for the metabolic imaging of tumors by PET (1,18). FDG PET allows quantitative evaluation of local glucose metabolism and can therefore be used to assess biological malignancy (1,19) and the efficacy of chemotherapy (20).
In 1995, Yasuda et al. used FDG PET for diagnostic imaging in a patient with esophageal cancer and were able to visualize clearly both the primary tumor and lymph node metastases (21). In 1997, Flanagan et al. used FDG PET for esophageal cancer staging in 36 patients (3). All of the patients showed abnormal FDG accumulation in the primary tumor. In five of them, PET identified new metastases, avoiding unnecessary surgery and primary lung cancer was also detected incidentally.
In 1997, Luketich et al. performed FDG PET preoperatively in patients with esophageal cancer and were able to identify the primary tumor with 97% accuracy (4). They also found that the technique was useful for detection of distant metastases and for staging to determine operability.
Although most of these previous studies involved patients with advanced cancer, the detection of superficial esophageal cancer has increased in recent years following advances in the endoscopic iodine staining method (10). We therefore included patients with superficial cancer diagnosed by endoscopic biopsy in the present study. Of the 22 lesions from which specimens were obtained during surgery or EMR, PET successfully diagnosed all 15 (100%) lesions with a depth of the submucosa (pT1b) or greater. However, PET yielded a false-negative result for all seven lesions (100%) that were confined to the mucosa (pTis and pT1a). There are therefore clear limitations to the use of PET imaging for the diagnosis of esophageal cancers confined to the mucosa. This is thought to be because detection by PET requires a certain minimum tumor volume.
EMR is indicated for esophageal cancer confined to the intraepithelial layer (m1) or the lamina propria (m2) and is relatively indicated for cancer that has invaded as far as the muscularis mucosae (m3) or the upper one-third of the submucosa (sm1) (10). Therefore, EMR is unlikely to be indicated if PET produces a positive finding.
Flanagan et al. reported that the accuracy of PET (76%) for determining the extent of nodal disease was superior to that of CT (45%) (3). Choi et al. compared preoperative CT and PET in 48 patients with esophageal cancer who had undergone lymph node dissection (6). The diagnostic sensitivity, specificity and accuracy of CT for detecting individual lymph node metastases were 18, 99 and 78%, respectively, while the values were 57, 97 and 86% for PET. Hence the results for PET were significantly better than those for CT.
As in previous studies, we also found that PET was superior to CT. However, PET was less sensitive than reported in previous studies. We thought that this was probably because lymph nodes with micrometastases were obtained by extended dissection in the present series.
In a 1999 study comparing the diagnostic accuracy of CT and PET for distant metastases, Luketich et al. reported a sensitivity of 46%, a specificity of 74% and an accuracy of 63% for CT, while PET had a sensitivity of 69%, a specificity of 93% and an accuracy of 84%, achieving significantly better results (7). A similar trend was observed for the detection of recurrence in the present study. However, our investigation of PET for the detection of recurrence involved only a small number of patients and was preliminary.
In a 1998 study of FDG PET performed in 48 patients with esophageal cancer, Fukunaga et al. found that measurement of tumor glucose metabolism was useful for predicting the prognosis, since there was a significantly worse prognosis in the patients with a high tumor SUV (8). The SUV of primary esophageal cancer may therefore serve as an indicator of biological malignancy. We did not examine the prognostic value of PET, so further investigation of this area is needed.
The disadvantage of omitting transmission scans during PET is that glucose metabolism cannot be quantitatively evaluated using the SUV. We found that there was no statistically significant correlation between the SUV and the T/N ratio. The T/N ratio varies depending on how the ROI is selected and the type of filter used. In addition, we determined the SUV using each patient’s total body weight. However, it has been reported that the SUV corrected for body surface area is less dependent on individual patient characteristics (12,22). Further studies are needed to identify parameters that can be used as indicators for quantitatively evaluating glucose metabolism instead of the SUV.
In conclusion, FDG PET can detect primary esophageal cancer with a depth of invasion of T1b or greater, but Tis and T1a tumors are undetectable by this modality. FDG PET is more accurate than CT or EUS for diagnosing lymph node metastasis, but is not highly sensitive. However, because it has a high specificity, it may be useful for deciding the management strategy. Although PET seems to be useful for identifying the site of recurrence, our results suggest that the T/N ratio cannot be used as a substitute for the SUV when PET images have not been corrected for attenuation.
|
Acknowledgments |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
The authors express their special thanks to Michiru Ide and Akira Shotsu of the Himedic Imaging Center at Lake Yamanaka for their helpful advice.
|
FOOTNOTES |
|---|
+ For reprints and all correspondence: Shinji Himeno, Department of Surgery, Tokai University School of Medicine,143 Shimokasuya, Isehara, Kanagawa 259-1193, Japan. E-mail: shinji-h@dc4.so-net.ne.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
1 Rigo P, Paulus P, Kaschten BJ, Hustinx R, Bury T, Jerusalem G, et al. Oncological application of positron emission tomography with fluorine-18 fluorodeoxyglucose. Eur J Nucl Med 1996;23:1641–74.[ISI][Medline]
2 Coleman RE. Clinical PET in oncology. Clin Positron Imaging 1998;1:15–30.[Medline]
3 Flanagan FL, Dehdashti F, Siegel BA, Trask DD, Sundaresan SR, Patterson GA, et al. Staging of esophageal cancer with 18F-fluorodeoxyglucose positron emission tomography. Am J Roentgenol 1997;168:417–24.
4 Luketich JD, Schauer PR, Meltzer CC, Landreneau RJ, Urso GK, Townsend DW, et al. Role of positron emission tomography in staging esophageal cancer. Ann Thorac Surg 1997;64:765–9.
5 Rankin SC, Taylor H, Cook GJR, Mason R. Computed tomography and positron emission tomography in the pre-operative staging of oesophageal carcinoma. Clin Radiol 1998;53:659–65.[ISI][Medline]
6 Choi JY, Lee KH, Shim YM, Lee KS, Kim JJ, Kim SE, et al. Improved detection of individual nodal involvement in squamous cell carcinoma of the esophagus by FDG PET. J Nucl Med 2000;41:808–15.
7 Luketich JD, Friedman DM, Weigel TL, Meehan MA, Keenan RJ, Townsend DW, et al. Evaluation of distant metastases in esophageal cancer: 100 consecutive positron emission tomography scans. Ann Thorac Surg 1999;68:1133–7.
8 Fukunaga T, Okazumi S, Koide S, Isono K, Imazeki K. Evaluation of esophageal cancers using fluorine-18-fluorodeoxyglucose PET. J Nucl Med 1998;39:1002–7.
9 Couper GW, Mcateer D, Wallis F, Norton M, Welch A, Nicolson M, et al. Detection of response to chemotherapy using positron emission tomography in patients with oesophageal and gastric cancer. Br J Surg 1998;85:1403–6.[ISI][Medline]
10 Makuuchi H. Endoscopic mucosal resection for early esophageal cancer-indication and techniques. Dig Endosc 1996;8:175–9.
11 Schomburg A, Bender H, Reichel C, Sommer T, Ruhlmann J, Kozak B, et al. Standardized uptake values of fluorine-18 fluorodeoxglucose: the value of different normalization procedures. Eur J Nucl Med 1996;23:571–4.[ISI][Medline]
12 Imran MB, Kubota K, Yamada S, Fukuda H, Yamada K, Fujiwara T, et al. Lesion-to-background ratio in nonattenuation-corrected whole-body FDG PET images. J Nucl Med 1998;39:1219–23.[ISI][Medline]
13 Japanese Society for Esophageal Disease. Guidelines for Clinical and Pathologic Studies on Carcinoma of the Esophagus, 9th ed. Tokyo: Kanehara 1999.
14 Kato H. Diagnosis and treatment of esophageal neoplasms. Jpn J Cancer Res 1995;86:993–1009.[ISI]
15 Fukuda M, Hirata K, Natori H. Endoscopic ultrasonography of the esophagus. World J Surg 2000;24:216–26.[ISI][Medline]
16 Menzel J, Hoepffner N, Nottberg H, Schulz C, Senninger N, Domschke W. Preoperative staging of esophageal carcinoma; Miniprobe sonography versus conventional endscopic ultrasound in a prospective histopathologically verified study. Endoscopy 1999;31:291–7.[Medline]
17 Warburg O. On the origin of cancer cells. Science 1956;123:309–14.
18 Delbeke D. Oncological application of FDG PET imaging: brain tumors, colorectal cancer, lymphoma and melanoma. J Nucl Med 1999;40:591–603.
19 Oshida M, Uno K, Suzuki M, Nagashima T, Hashimoto H, Yagata H, et al. Predicting the prognoses of breast carcinoma patients with positron emission tomography using 2-deoxy-2-fluoro[18F]-D-glucose. Cancer 1998;82:2227–34.[ISI][Medline]
20 Price P, Jones T. Can positron emission tomography (PET) be used to detect subclinical response to cancer therapy? Eur J Cancer 1995;31A:1924–7.[ISI][Medline]
21 Yasuda S, Raja S, Hubner KF. Application of whole-body positron emission tomography in the imaging of esophageal cancer: report of a case. Surg Today 1995;25:261–4.[ISI][Medline]
22 Kim CK, Gupta NC. Dependency of standardized uptake values of fluorine-18 fluorodeoxyglucose on body size: comparison of body surface area correction and lean body mass correction. Nucl Med Commun 1996;17:890–4.[ISI][Medline]
Received January 7, 2002; accepted May 31, 2002
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. G. Little, T. W. Rice, B. Bybel, D. P. Mason, S. C. Murthy, G. W. Falk, L. A. Rybicki, and E. H. Blackstone Is FDG-PET indicated for superficial esophageal cancer? Eur. J. Cardiothorac. Surg., May 1, 2007; 31(5): 791 - 796. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Rizk, R. J. Downey, T. Akhurst, M. Gonen, M. S. Bains, S. Larson, and V. Rusch Preoperative (18)[F]-Fluorodeoxyglucose Positron Emission Tomography Standardized Uptake Values Predict Survival After Esophageal Adenocarcinoma Resection. Ann. Thorac. Surg., March 1, 2006; 81(3): 1076 - 1081. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kan, Y. Shimada, F. Sato, T. Ito, K. Kondo, G. Watanabe, M. Maeda, S. Yamasaki, S. J. Meltzer, and M. Imamura Prediction of Lymph Node Metastasis with Use of Artificial Neural Networks Based on Gene Expression Profiles in Esophageal Squamous Cell Carcinoma Ann. Surg. Oncol., December 1, 2004; 11(12): 1070 - 1078. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.L. van Westreenen, M. Westerterp, P.M.M. Bossuyt, J. Pruim, G.W. Sloof, J.J.B. van Lanschot, H. Groen, and J.Th.M. Plukker Systematic Review of the Staging Performance of 18F-Fluorodeoxyglucose Positron Emission Tomography in Esophageal Cancer J. Clin. Oncol., September 15, 2004; 22(18): 3805 - 3812. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||