Eur J Cardiothorac Surg 2007;31:791-796. doi:10.1016/j.ejcts.2007.01.037
Copyright © 2007, European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved
Is FDG-PET indicated for superficial esophageal cancer?
Sherard G. Littlea,
Thomas W. Ricea,b,*,
Bohdan Bybelc,
David P. Masona,b,
Sudish C. Murthya,b,
Gary W. Falkb,d,
Lisa A. Rybickie,
Eugene H. Blackstonea,e
a Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, Cleveland, OH, USA
b The Center for Swallowing and Esophageal Disorders, Cleveland Clinic, Cleveland, OH, USA
c Department of Molecular and Functional Imaging, Cleveland Clinic, Cleveland, OH, USA
d Department of Gastroenterology and Hepatology, Cleveland Clinic, Cleveland, OH, USA
e Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH, USA
Received 11 September 2006;
accepted 15 January 2007.
* Corresponding author. Address: Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, 9500 Euclid Avenue/Desk F24, Cleveland, OH 44195, USA. Tel.: +1 216 444 1921; fax: +1 216 445 6876. (Email: ricet{at}ccf.org).
 |
Abstract
|
|---|
Objective: To ascertain whether fluorodeoxyglucose positron emission tomography is indicated for clinical staging of superficial cancer, we sought to determine if it accurately classifies tumor (T), regional nodal (N), and distant metastases (M), including distinguishing high-grade dysplasia (Tis) from invasive cancer (T1). Methods: Fifty-eight superficial esophageal cancer patients had preoperative positron emission tomography, 53 (91%) fused with computed tomography. Tumor characteristics, esophagoscopy findings, and pTNM were compared with positron emission tomography cTNM. pT1 was subdivided into intramucosal cancers with lamina propria or muscularis mucosa invasion and submucosal cancers with inner or outer invasion. Results: Fluorodeoxyglucose uptake increased with pT, from 5/11 (45%) for pTis to 11/16 (69%) for pT1 (outer submucosa), P
= 0.07, as it did for standardized uptake value, median 0 for pTis to 2.7 for pT1 (outer submucosa), P
= 0.06. Positron emission tomography could not differentiate Tis (5/11, 45%) from T1 (26/47, 55%; P
=
0.03). Regional nodal fluorodeoxyglucose uptake in three patients (standardized uptake value 2.8, 4.9, 11) was false positive; in six pN1 patients, it was false negative. Positron emission tomography had 0% sensitivity and positive predictive value for N1. There were no distant metastases; one patient developed a pulmonary metastasis 15 months postoperatively. Positron emission tomography detected three (5%) distant hypermetabolic sites, all synchronous tumors (papillary thyroid cancer, adrenal pheochromocytoma, rectal adenoma). Only increasing tumor length was related to greater fluorodeoxyglucose uptake (P
=
0.004) and higher standardized uptake value (P
=
0.001). Conclusions: Because positron emission tomography can neither differentiate pTis from T1 nor classify T, N, and M, it is not indicated in staging superficial esophageal cancer. Finding a synchronous primary tumor in approximately every 20th patient is its only benefit. Better, less expensive screening tools are available for common synchronous malignancies.
Key Words: TNM classifications • High-grade dysplasia • Synchronous tumors • Standardized uptake value • Diagnostic testing
|
1. Introduction
|
|---|
2[18F]fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET) has become a widely used clinical staging tool for esophageal cancer. However, its value in clinical staging of superficial carcinomas has been questioned [1–3]. The purposes of this study were to determine if FDG-PET accurately classifies tumor (T), regional nodal (N), and distant metastases (M), including distinguishing high-grade dysplasia (Tis) from invasive cancer (T1).
|
2. Patients and methods
|
|---|
2.1 Patients
From June 2003 to August 2005, 66 patients with superficial adenocarcinoma of the esophagus (pTis [high-grade dysplasia] or pT1) underwent esophagectomy at Cleveland Clinic. Of these, 58 (88%) had preoperative FDG-PET scanning and are the basis for this study. FDG-PET scanning was performed 50 ± 52 days before esophagectomy. Fifty-three (91%) had fused computed tomography PET scans (PET/CT), all but 1 performed at Cleveland Clinic; 5 (9%) had FDG-PET scans elsewhere without CT.
2.2 FDG-PET technique
The 52 patients who had PET/CT scans performed at Cleveland Clinic were scanned after typical preparation, including fasting after midnight before the test, hydration, and avoiding excessive muscle activity. Patients were injected with 15 mCi of FDG for weight up to 100 kg and 20 mCi for weight exceeding that. They were imaged 45–60 min later from skull base to mid-thigh with arms raised. A Siemens biograph 16 scanner was used with imaging time of 2–5 min per bed position, based on patient weight (1 min for weight <68 kg, 1.5 min for 68–100 kg, 2 min for >100 kg). Image reconstruction used the iterative OS-EM algorithm (two iterations and eight subsets) and 6-mm filter with the CT data used for attenuation correction. PET/CT studies were reviewed by one of three experienced nuclear medicine physicians using available co-registration software. Standardized uptake value (SUV) calculations were obtained using standard Siemens software for regions of interest drawn around findings of concern. Maximal SUV was recorded for analysis.
2.3 Tumor characteristics
At esophagoscopy, median length of Barrett mucosa was 3.0 cm (Table 1
). One patient each had endoscopically identified nodules, plaque, and ulcer. Endoscopic ultrasound (EUS) was performed in 53 patients.
At resection, in 20 patients in whom all 3 tumor measurements were available, median tumor dimensions were 1.8 cm x 1.4 cm x 0.2 cm (Table 1). Depth of tumor invasion (pT) was high-grade dysplasia (pTis) in 11 (19%), intramucosal cancer (pT1a) in 24 (41%), and submucosal cancer (pT1b) in 23 (40%) (Table 2
). Total number of nodes sampled was 994, median 15 per patient. Six patients had regional lymph node metastases (pN1); one cancer invaded the inner submucosa (T1b-inner) and five the outer submucosa (T1b-outer). All patients had adenocarcinoma.
2.4 Surgery
All patients proceeded to esophagectomy without induction chemoradiotherapy. Based on clinical stage, 38 (66%) had transhiatal esophagectomy with lymph node sampling, and 20 (34%) had thoracoabdominal esophagectomy with two-field lymph node sampling (Table 3
).
2.5 Data analysis
SUV was analyzed as a binary variable (any uptake) and as a continuous variable (actual uptake). The Cochran–Armitage test was used to determine if the percentage of patients with any uptake increased with greater tumor depth or worse tumor differentiation. The Jonckheere–Terpstra test was used to determine if actual uptake increased with greater tumor depth or worse tumor differentiation. Spearman rank correlation was calculated to assess the strength of the association of SUV with Barrett length, tumor length, and tumor size. Analyses were performed using SAS® software (SAS Institute Inc., Cary, NC). Data are presented as frequencies and percentages, medians, quartiles, and ranges, or means ± standard deviations, as appropriate.
|
3. Results
|
|---|
3.1 FDG-PET and pT
Of the 58 patients, 31 (53%) had FDG uptake at the primary tumor site. Patients with tumor uptake had SUV ranging from 2.1 to 16.6, with a mean of 3.9 and median of 3.5 (Fig. 1
). FDG uptake increased with increasing depth of tumor invasion (P
=
0.07), as did SUV (P
=
0.06; Table 4
). However, FDG-PET could not differentiate pTis from pT1, with 5 of 11 (45%) pTis tumors having FDG uptake and 26 of 47 (55%) pT1 tumors having it. Increasing tumor length at endoscopy was related to greater FDG uptake (P
=
0.004) and higher SUV (P
=
0.001; Fig. 2
).
3.2 FDG-PET and pN
Of the 58 patients, 3 (5.2%) had FDG uptake in regional lymph nodes, with SUVs of 2.8, 4.9, and 11 (Fig. 3
). FDG-positive nodes were found to be false positive after pathologic examination of 13, 25, and 26 nodes. (No primary tumor uptake was observed in these patients.) Six patients with regional lymph node metastases (pN1, Table 2) at esophagectomy had false-negative FDG-PET scans. Thus, sensitivity and positive predictive value of FDG uptake for pN1 were 0% and 0%, and specificity, negative predictive value, and accuracy were 94%, 89%, and 84%, respectively.
View larger version (96K):
[in this window]
[in a new window]
|
Fig. 3. FDG uptake in regional lymph nodes. Upper left panel is axial computed tomography (CT) section through tumor; lower left panel is corresponding PET/CT; middle panel is sagittal PET/CT; right panel is PET/CT coronal section. Arrows point to primary tumor FDG uptake. Subcarinal paraesophageal right hilar lymph node (arrows): SUV = 11.0, no primary tumor uptake.
|
|
3.3 FDG-PET and pM
Three patients (5.3%) demonstrated distant-site FDG uptake, with SUVs of 4.8, 4.9, and 7.4. Additional investigations and confirmatory biopsies revealed synchronous neoplasms in all three: papillary carcinoma of the thyroid, adrenal pheochromocytoma (Fig. 4
), and rectal adenoma. No distant metastases from esophageal cancers were found at esophagectomy. One patient developed a pulmonary metastasis 15 months postoperatively.
View larger version (105K):
[in this window]
[in a new window]
|
Fig. 4. FDG uptake at distant sites. Adrenal pheochromocytoma (arrows): SUV = 4.9, no primary esophageal tumor uptake. Format as in Fig. 3.
|
|
|
4. Discussion
|
|---|
4.1 Principal findings
Despite use of improved imaging technology (PET/CT), this study identifies three tumor and three imaging problems that compound to render FDG-PET useless for reliable detection of superficial carcinoma of the esophagus. Tumor problems include: (1) small size, (2) uncommon regional lymph node metastases for these small tumors, and (3) rare distant metastases. Imaging problems include: (1) using three-dimensional imaging for a unidimensional characterization of pT, (2) uncertainty in interpreting FDG uptake, and (3) small signal-to-noise ratio.
4.2 FDG-PET and pT
It is not surprising that many investigators have found FDG-PET results and pT to be positively, but poorly, correlated [2–5]. FDG-PET relates to volume of tumor and its metabolic activity, while pT is a unidimensional characterization of tumor depth. For superficial cancer, the problem is compounded by the fact that these tumors are small and may be below the resolution of FDG-PET [2,3,6]. Himeno and colleagues reported that 100% of 15 tumors confined to the mucosa (pTis and pT1a) were FDG-PET negative, and 100% of 15 tumors invading beyond the mucosa (T 
1b) were positive [2]. This led them to conclude that ‘Tis and T1a tumors are undetectable by this modality.’ Kato and colleagues reported that 18% of T1a tumors and 61% of T1b tumors were FDG-PET positive [3]. Cerfolio and colleagues reported that SUV increased with pT from 1.7 for pTis to 6.0 for T1–T2 and 9.8 for T3–T4 [5]. Our analysis found a higher uptake with Tis and T1a than previously reported, perhaps because of improved technology with PET/CT, but also because we had a larger patient population, some of whom had large superficial tumors. However, the power of PET/CT to discriminate T classification for superficial esophageal cancer is inadequate. Similarly, this test is unable to differentiate pTis from invasive cancer.
The problem is further compounded by the nonspecific nature of FDG uptake. Hypermetabolism is not unique to malignancy, which may lead to false-positive interpretation of FDG uptake. FDG-PET hypermetabolism has been reported secondary to esophagitis caused by chemotherapy [7], radiation therapy [8], candida infection [9], Barrett esophagus [10,11], gastroesophageal reflux disease-associated esophageal spasm, and bacterial esophagitis with peptic stricture [11]. Three percent of FDG-PET/CT studies have incidental lesions detected in the gastrointestinal tract [10]. These contribute also to 1 component of the ‘background noise’ of FDG-PET.
4.3 FDG-PET and pN
Similarly, it is not surprising that FDG-PET is highly specific and accurate in detecting pN1 disease, because few patients with superficial cancer have nodal metastases. However, it is insensitive, with no positive predictive value, because nodal metastases are likely microscopic (small volume), and any node detected likely reflects background hypermetabolism ‘noise’ for some reason other than tumor [12]. van Westreenen and colleagues performed a meta-analysis of 12 reports of FDG-PET for staging esophageal cancer [13]. Median prevalence of pN1 was only 0.55 (range, 0.33–0.84). Pooled sensitivity and specificity for pN1 were only 0.51 (95% CI, 0.34–0.69) and 0.84 (95% CI, 0.76–0.91), respectively. Positive predictive value and negative predictive value were 0.60 and 0.46, respectively. This rather poor performance of FDG-PET/CT is magnified in patients with superficial esophageal cancer, in which pN1 prevalence is low.
4.4 FDG-PET and pM
Exceedingly low prevalence of distant metastases compounds the problems of interpretation. Performance of PET/CT in M1 classification was undefined in patients with superficial esophageal cancer in our study because no pM1 was detected. In the same meta-analysis of van Westreenen and colleagues, median prevalence of pM1 was 0.36 (range, 0.09–0.50). Pooled sensitivity and specificity for pM1 was 0.67 (95% CI, 0.58–0.76) and 0.97 (95% CI, 0.90–1.00), respectively. Positive predictive value and negative predictive value were 0.92 and 0.83, respectively. False-positive examinations (signal-to-noise) result from incidental synchronous tumors, which have been reported to occur in 5.5% of 366 patients with esophageal cancer undergoing FDG-PET [14]. They were most common in the colon, kidney, and thyroid. The present study identified a similar number and locations of synchronous tumors, which simply reflect incidental tumors in these patients, not stage of the esophageal cancer.
4.5 Strengths and weaknesses
This is a small study from a single institution having a large experience with an uncommon tumor. Nevertheless, it highlights the limitation of current imaging technology, including PET and CT, to detect these small tumors. The study is also limited by our ability to accurately sample, detect, and classify tumors wherever they may be.
4.6 Recommendations
Because FDG-PET can neither differentiate pTis from T1 nor classify T, N, and M, it is not indicated in staging superficial esophageal cancer. Finding a synchronous primary tumor in approximately every 20th patient is its only benefit. Better, less expensive screening tools are available for common synchronous malignancies.
|
Appendix A
|
|---|
Conference discussion
Dr P. Thomas (Marseille, France): May I ask you the histology of your patients?
Dr Little: All these cancers were adenocarcinomas of the distal esophagus.
Dr Thomas: You said that the discovery of a synchronous tumor was infrequent. In fact, it was as high as 5%. It may significantly alter the management of those patients.
Dr Little: The patient who had the rectal adenoma underwent colonoscopy and polypectomy. The patient who had a thyroid carcinoma had thyroidectomy at the time of the esophagectomy, and the patient with the pheochromocytoma had resection of his tumor prior to esophagectomy.
Dr M. Lanuti (Boston, MA, USA): Of the six patients that had positive mediastinal lymph nodes on operation, were they detected by EUS?
Dr Little: One of these patients had lymph nodes detected on EUS and had fine-needle aspiration confirmation prior to surgery.
|
Acknowledgments
|
|---|
The authors thank Albert Chang, MD, and Ann Gamber for help with data collection, Ron Young for PET/CT consultation, Brian Kohlbacher for graphics assistance, and Tess Parry for editorial help.
|
Footnotes
|
|---|
\#9734; Presented at the joint 20th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 14th Annual Meeting of the European Society of Thoracic Surgeons, Stockholm, Sweden, 10–13 September 2006.
|
References
|
|---|
- Flamen P, Lerut A, Van Cutsem E, De Wever W, Peeters M, Stroobants S, Dupont P, Bormans G, Hiele M, De Leyn P, Van Raemdonck D, Coosemans W, Ectors N, Haustermans K, Mortelmans L. Utility of positron emission tomography for the staging of patients with potentially operable esophageal carcinoma. J Clin Oncol 2000;18:3202-3210.[Abstract/Free Full Text]
- Himeno S, Yasuda S, Shimada H, Tajima T, Makuuchi H. Evaluation of esophageal cancer by positron emission tomography. Jpn J Clin Oncol 2002;32:340-346.[Abstract/Free Full Text]
- Kato H, Miyazaki T, Nakajima M, Takita J, Kimura H, Faried A, Sohda M, Fukai Y, Masuda N, Fukuchi M, Manda R, Ojima H, Tsukada K, Kuwano H, Oriuchi N, Endo K. The incremental effect of positron emission tomography on diagnostic accuracy in the initial staging of esophageal carcinoma. Cancer 2005;103:148-156.[CrossRef][Medline]
- Rizk N, Downey RJ, Akhurst T, Gonen M, Bains MS, Larson S, Rusch V. Preoperative 18[F]-fluorodeoxyglucose positron emission tomography standardized uptake values predict survival after esophageal adenocarcinoma resection. Ann Thorac Surg 2006;81:1076-1081.[Abstract/Free Full Text]
- Cerfolio RJ, Bryant AS. Maximum standardized uptake values on positron emission tomography of esophageal cancer predicts stage, tumor biology, and survival. Ann Thorac Surg 2006;82:391-394discussion 4–5.[Abstract/Free Full Text]
- Flanagan FL, Dehdashti F, Siegel BA, Trask DD, Sundaresan SR, Patterson GA, Cooper JD. Staging of esophageal cancer with 18F-fluorodeoxyglucose positron emission tomography. AJR Am J Roentgenol 1997;168:417-424.[Abstract/Free Full Text]
- Bural GG, Kumar R, Mavi A, Alavi A. Reflux esophagitis secondary to chemotherapy detected by serial FDG-PET. Clin Nucl Med 2005;30:182-183.[CrossRef][Medline]
- Bhargava P, Reich P, Alavi A, Zhuang H. Radiation-induced esophagitis on FDG PET imaging. Clin Nucl Med 2003;28:849-850.[CrossRef][Medline]
- Shrikanthan S, Aydin A, Dhurairaj T, Alavi A, Zhuang H. Intense esophageal FDG activity caused by Candida infection obscured the concurrent primary esophageal cancer on PET imaging. Clin Nucl Med 2005;30:695-697.[CrossRef][Medline]
- Kamel EM, Thumshirn M, Truninger K, Schiesser M, Fried M, Padberg B, Schneiter D, Stoeckli SJ, von Schulthess GK, Stumpe KD. Significance of incidental 18F-FDG accumulations in the gastrointestinal tract in PET/CT: correlation with endoscopic and histopathologic results. J Nucl Med 2004;45:1804-1810.[Abstract/Free Full Text]
- Bakheet SM, Amin T, Alia AG, Kuzo R, Powe J. F-18 FDG uptake in benign esophageal disease. Clin Nucl Med 1999;24:995-997.[CrossRef][Medline]
- van Westreenen HL, Heeren PA, Jager PL, van Dullemen HM, Groen H, Plukker JT. Pitfalls of positive findings in staging esophageal cancer with F-18-fluorodeoxyglucose positron emission tomography. Ann Surg Oncol 2003;10:1100-1105.[CrossRef][Medline]
- van Westreenen HL, Westerterp M, Bossuyt PM, Pruim J, Sloof GW, van Lanschot JJ, Groen H, Plukker JT. Systematic review of the staging performance of 18F-fluorodeoxyglucose positron emission tomography in esophageal cancer. J Clin Oncol 2004;22:3805-3812.[Abstract/Free Full Text]
- van Westreenen HL, Westerterp M, Jager PL, van Dullemen HM, Sloof GW, Comans EF, van Lanschot JJ, Wiggers T, Plukker JT. Synchronous primary neoplasms detected on 18F-FDG PET in staging of patients with esophageal cancer. J Nucl Med 2005;46:1321-1325.[Abstract/Free Full Text]
Top
Abstract
1. Introduction
