Japanese Journal of Clinical Oncology Advance Access published online on March 1, 2007
Japanese Journal of Clinical Oncology, doi:10.1093/jjco/hyl150
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2007 Foundation for Promotion of Cancer Research
Unsuspected Bone and Soft Tissue Lesions Identified at Cancer Screening using Positron Emission Tomography
1 Division of Diagnostic Radiology
2 Divisions of Cancer Screening and Statistics,
3 Cancer Control Division, Research Center for Cancer Prevention and Screening, National Cancer Center, Tokyo, Japan
4 Division of Diagnostic Imaging, University of Texas MD Anderson Cancer Center, Houston, Texas, USA
5 Department of Radiology, Kobe University Graduate School of Medicine, Kobe, Japan
For reprints and all correspondence: Ukihide Tateishi, Division of Diagnostic Radiology, National Cancer Center Hospital, 5-1-1, Tsukiji, Chuo-ku, Tokyo, 104-0045, Japan. E-mail: utateish{at}ncc.go.jp
Received October 2, 2006; accepted October 30, 2006
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONConflict of interest statementReferences |
|---|
Background: [F-18]-fluorodeoxy-D-glucose (18FDG) positron emission tomography (PET) is a sensitive modality for detecting malignant lesions. The purpose of the present study was to describe unknown bone and soft tissue lesions in adults identified at cancer screening using PET.
Methods: A total of 4283 individuals of more than or equal to 40 years of age were enrolled. All individuals underwent scans from the base of the skull to proximal thigh. The images were reviewed and a consensus was reached by two board-certified radiologists and a nuclear medicine specialist for the diagnoses. Diagnoses of the lesions were confirmed by histological examination, typical radiologic findings, obvious progression in number and/or size of the lesion on follow-up examinations, and medical examination of interview.
Results: Unsuspected focal abnormality in the bone and soft tissue were found in 62 individuals (1.4%). The mean size of the lesion was 26 mm (range, 6–155 mm). There were 29 bone lesions (47%) and 33 soft tissue lesions (53%). A malignant lesion was found in one case (1.6%) and histologic diagnosis was primary non-Hodgkin lymphoma of the vertebra. Other major diagnoses were healing bone (n = 11, 18%) and benign cystic lesions of bone and soft tissue (n = 9, 15%), and brown fat of soft tissue (n = 4, 6%).
Conclusion: Unsuspected bone and soft tissue lesions of a wide variation of pathologic and clinical diagnoses were encountered at cancer screening using PET. Correlation with clinical history and other imaging findings is essential in the differential diagnosis.
Key Words: cancer • positron emission tomography • bone • soft tissue
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONConflict of interest statementReferences |
|---|
Cancer remains a large cause of death despite 50 years of anticancer efforts with advances in treatment. Patients with earlier stage disease have better survival rates but constitute a small proportion of presenting cases. The concept that earlier detection of cancer may improve the outcome of this disease has been a subject of controversy.
Some radiologic modalities are employed in cancer screening. In case of lung cancer, results of previous screening studies using chest X-ray contributed to the negative opinion. However, computed tomography (CT) detects lung cancer at an earlier stage with smaller size when compared with chest X-ray (1). The efficacy of lung cancer screening using CT is still undetermined, and whether it creates problems such as morbidity is unclear.
[F-18] fluorodeoxyglucose (18FDG) positron emission tomography (PET) cancer screening is not routinely performed in most countries, because of the lack of evidence that PET provides an advantage for the patient (2). The new-generation PET/CT improves the detectability of cancerous lesions because anatomic and molecular information can be precisely co-registered. Rapid technological advances in PET have ignited interest in cancer screening. Whether we can identify individuals at high risk to improve the detectability of cancer is still undetermined when we add PET study into the first screening method.
Although PET studies possibly detect a wide variety of cancers at potentially curable stages (3,4), we often encountered unsuspected bone and soft tissue lesions. The aim of the present study was to determine the rates of false positive findings and to report unsuspected bone and soft tissue lesions identified at cancer screening using PET.
|
MATERIAL AND METHODS |
|---|
|
TOPAbstractINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONConflict of interest statementReferences |
|---|
Patients
Participants over 40 years of age were enrolled at the research center for cancer prevention and screening from February 2004 to March 2006. There were 4283 individuals (male, n = 2491, 58%; female, n = 1792, 42%). Of these, 62 (1.4%) had unexpected bone or soft tissue lesion on PET scan, who were included for the retrospective analysis. Exclusion criteria for PET study were individuals with known malignancy, subjects of uncontrolled diabetes and pregnant or lactating women. This study was conducted in accordance with the amended Helsinki declaration and approved by the local ethics committees after all the patients had provided their informed consents for the review of their records and images.
Imaging Studies
Scans were acquired with dedicated PET (ECAT ACCEL; Siemens CTI, Knoxville, TN; n = 1483, 35%) and PET/CT device (Aquiduo; Toshiba Medical Systems, Tokyo, Japan; n = 2800, 65%) that consisted of a PET scanner (ECAT HR + ; CTI) and 16-section CT scanner (Aquilion V-detector; Toshiba Medical Systems) with a whole-body mode implemented as the standard software. Prior to the PET study, the patients were fasted for at least 6 h. CT was performed from the head to the proximal thigh according to a standardized protocol with the following setting: axial 3.0-mm collimation x 16 modes; 120 kVp; 80 mAs; and a 0.5-s tube rotation, a table speed of 11.0 mm/s. Most patients were scanned with arms up outside of the field of view. Only two patients with humerus and arm lesions were scanned with arms down. Patients maintained normal shallow respiration during the acquisition of CT scans. No iodinated contrast material was administered. Emission scans from the base of the skull to the proximal thigh were obtained starting 60 min after the intravenous administration of 185 MBq of 18FDG. The acquisition time for PET was 2 min per table position. CT data was resized from a 512 x 512 matrix to a 128 x 128 to match the PET data so that the images could be fused and CT transmission maps generated. Images were reconstructed with attenuation-weighted ordered-subset expectation maximization with two iterations and eight subsets using emission scans and CT data. Co-registered PET/CT images were displayed using dedicated software (Syngo; Siemens).
Radiograph and magnetic resonance imaging (MRI) of abnormal bone or soft tissue lesion sites was performed within 2 weeks of PET study. MRI was performed using 1.5 T systems (Signa, GE Medical Systems, Milwaukee, WI) with a phased array or body coil. Pulse sequence parameters and slice orientation varied with the examined anatomic site. T1- and T2-weighted images were obtained in the transverse plane and at least one longitudinal plane. T1-weighted acquisitions were obtained by using a 24–30 cm field of view, 4–8 mm section thickness, 600–700/8–15 (repetition time, ms/echo time, ms), 256 x 192–224 matrix, two signals acquired. T2-weighted acquisitions were performed by using a 15–35 cm field of view, 5–8 mm section thickness, a 2-mm intersection gap, 3000–5000/90–120 (repetition time, ms/effective echo time, ms) and 256 x 192–224 matrix, and two signals were acquired. After the intravenous administration of 0.1 mmol/kg of gadopentate dimeglumine (Magnevist, Schering, Berlin, Germany), fat-saturated T1-weighted images were obtained in the transverse plane and at least one longitudinal plane.
Image Interpretation
The images were reviewed and a consensus for the diagnosis was reached by two board-certified radiologists and a nuclear medicine specialist, who were unaware of any clinical or radiologic information using a multimodality computer platform. The initial review of the attenuation-corrected PET images was performed using transaxial, coronal and sagittal planes. Abnormal 18FDG uptake was defined as the focal increased activity greater than that in the adjacent or contralateral bone and soft tissue. Lesions with abnormal 18FDG uptake on PET and PET/CT images, and their likely anatomic location were recorded. A region of interest (ROI) was outlined in the peak activity within regions of increased 18FDG uptake. Qualitative evaluation for focally increased glucose metabolism as well as quantitative evaluation of standardized uptake values (SUV) was performed. When the tumor is extensively heterogeneous, the ROIs were set to cover all the components of the tumor.
Proof of Tumor Presence
Diagnoses were confirmed by histological examination after surgical resection (n = 12) and US- or CT-guided needle biopsy (n = 9). Diagnoses were also made by typical findings of radiograph or MRI and the presence of traumatic history, which was accomplished by means of medical examination of interview in which patients were questioned about their clinical signs or symptoms including induration or tenderness at the site where abnormal uptake was identified at PET study. The individuals whose diagnoses remained undetermined with abnormal PET findings were followed with CT, MRI and radiograph and checked whether there were an obvious progression in number or size of the lesions on follow-up examinations or not.
Statistical Analysis
The Student t-test was used for paired comparisons between SUV and size of the lesion. A P value of 0.05 was considered to indicate a statistically significant difference. Statistical analysis was performed with the SPSS version 11 software program (SPSS Inc., Chicago, IL).
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONConflict of interest statementReferences |
|---|
Of 4283 individuals who underwent cancer screening of PET study, we detected a total of 62 individuals (1.4%) who had unsuspected focal abnormalities in bone or soft tissue. There were 34 men (55%) and 28 (45%) women. The mean age of the subjects was 63 years. The mean size of the lesion was 26 mm (range, 6–155 mm). The majority of the lesions (75%) were less than or equal to 30 mm in the diameter. The mean SUV of the lesion was 3.4 (range, 0.9–10.6). There was no significant correlation between SUV and the mean size of the lesion (P = 0.55). All individuals underwent radiograph and MRI of the lesion. Diagnoses of the lesions were confirmed pathologically in 21 cases. In another 17 individuals, typical radiologic findings and the presence of traumatic history were revealed, leaving 24 cases for further assessment (12 bone lesions and 12 soft tissue lesions). These individuals whose diagnoses remained undetermined with abnormal PET findings were followed with CT (n = 16), MRI (n = 6), and radiograph (n = 12). The median follow-up duration was 5 months (range, 3–24 months).
Bone Lesion
In a total of 29 bone lesions, there were 23 (79%) in trunk and 6 (21%) in skull, mandible, and proximal portion of extremities (Table 1). The mean size of the lesion was 25 mm (range, 10–124 mm). The mean SUV was 3.2 (range, 1.0–7.3). Rib (n = 11, 38%) was the most frequent site of abnormal uptake and derived from healing bone after fracture which was verified by means of medical examination of interview and was long-standing and unchanged from previous radiograph or CT scans. The second frequent site was vertebra (n = 6, 21%): centrum of the vertebra (n = 3), acantha (n = 1), vertebral arch (n = 1), and inferior articular process (n = 1). Of these, diagnoses of four subjects remained undetermined. In one subject, osteoblastoma was found in the vertebral arch (C3) which was depicted as a osteolytic lesion on CT (SUV: 3.3). One subject was found to have a non-Hodgkin lymphoma of the bone. PET/CT showed that abnormal 18FDG uptake corresponded to osteolytic lesion of vertebral centrum (Fig. 1). CT-guided needle biopsy was performed and histologic diagnosis was diffuse large B-cell lymphoma. Whole-body PET/CT did not reveal other lesions. This patient had a normal physical examination at the time of PET/CT screening. This patient had a symptom of abrupt adynamia 17 days after biopsy. Progressive motor and sensory paralysis was found. On T2-weighted FSE MR images, the tumor showed hypersignal intense relative to adjacent vertebra. Pronounced enhancement was observed in the centrum as well as in posterior elements of the vertebra, and the surrounding soft tissue on post contrast SE fat-suppressed T1-weighted MR images. Vertebrectomy and subsequent chemotherapy were performed. The patient became ambulatory after therapy. Other histologic diagnoses of bone lesions (n = 5, 17%) were simple bone cyst, dentigerous cyst, fibrous dysplasia (Fig. 2), and hypertrophic change after enthesopathy. Those diagnoses were made by pathologic examination using biopsy specimens. Twelve bone lesions (41%) remained undetermined because the follow-up time for these lesions was too short to draw meaningful conclusions. The mean size of the lesion was 17 mm (10–38) and the mean SUV was 2.8 (range, 1.5–3.8). CT findings of these lesions were osteoblastic change in five, sclerotic rim with osteolytic change in two and normal in five. The median follow-up duration of these subjects was 3 months (range, 0–12 months). Most likely etiologies based on CT findings or other correlative imaging could not be speculated in these 12 lesions. These patients were scheduled for return for follow-up imaging.
|
|
|
Soft Tissue Lesion
In a total of 33 soft tissue lesions, there were 23 (70%) in the trunk and nine (27%) in the head and neck, and one (3%) in the upper arm (Table 2). The mean size of lesions was 26 mm (range, 6–60 mm). The mean SUV was 3.8 (range, 0.9–10.6). Buttock and inguinal region (n = 9) were the most common sites of abnormal uptake (Table 2). Most frequent pathologic diagnoses in these regions were cystic lesions: simple cyst (n = 3), epidermoid cyst (n = 2) and cystic neurinoma (n = 1). These cystic lesions arose from dermal or subcutaneous tissue of the trunk and showed homogeneous 18FDG uptakes. Although other diagnoses included reactive lymph node (n = 1) and dystrophic calcification with inflammation (n = 1), one case was undetermined. Truncal wall in chest and abdomen is a second frequent site of lesion (n = 6). These lesions showed localized abnormal uptake under the fascia. Of these, highest SUV (4.8) was detected in the chest wall of one subject (Fig. 3). Tumor showed heterogeneously hypersignal relative to muscle on T2-weighted FSE MR images. Post contrast SE fat-saturated T1-weighted MR images showed heterogeneous enhancement. Surgical biopsy was performed and pathologic diagnosis of neurinoma was made. Other diagnoses of truncal lesions determined by pathologic examinations were brown fat (n = 1), epidermoid (n = 1) and dystrophic calcification with inflammation (n = 1). However, two lesions were undetermined. The neck was the third most frequent region (n = 4). In one subject, PET/CT showed soft tissue mass with an abnormal accumulation adjacent to the left C1/2 facet joint (Fig. 4). SUV of this lesion was 10.6. On T2-weighted MR images, the tumor showed iso-signal intensity relative to adjacent muscle. Marked homogeneous enhancement was observed on post contrast SE fat-suppressed T1-weighted MR images. Pathologic examination revealed that the localized tumor was a tenosynovial giant cell tumor which extended to muscular membrane and synovial bursa. Other diagnoses of the neck determined by pathologic examinations were brown fat (n = 2). Two subjects with brown fat showed unilateral foci in the deep soft tissue of the neck. However, one lesion was undetermined. Other pathologic diagnoses of soft tissue lesions (n = 14) were non-specific inflammation (n = 2), keratosis (n = 1), dermal mucinosis (n = 1), foreign body granuloma (n = 1), reactive lymph node (n = 1), and undetermined lesions (n = 8). Eight undetermined lesions were situated superficially (n = 6) and deeply (n = 2). In the two deep-situated lesions, one lesion was associated with calcification in the iliopsoas and the other was adjacent to a post surgical scar. Twelve soft tissue lesions (36%) remained undetermined because the follow-up time for these lesions was too short to draw meaningful conclusions. The mean size of the lesions was 27 mm (6–60 mm) and the mean SUV was 1.9 (range, 1.1–3.2). The median follow-up duration of these subjects was 6 months (range, 0–18 months). Most likely etiologies based on CT or MR findings could not be speculated in these 12 lesions (Table 3).
|
|
|
|
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONConflict of interest statementReferences |
|---|
The results of the present study show that PET study can detect unexpected bone and soft tissue lesions at cancer screening setting. In our study, a total of 62 subjects (1.4%) had abnormal finding at PET study. Of these, one case was diagnosed as having a non-Hodgkin lymphoma of the bone which was localized in the vertebral centrum, and this patient received surgical resection and subsequent chemotherapy. The results might demonstrate potential in detecting bone and soft tissue malignancies by PET study. However, it is not proposed that PET study can be used to definitely rule out bone and soft tissue malignancies.
Normal physiological or benign pathologic 18FDG uptake can be confused with malignancy. Interpretive pitfalls are often encountered in PET/CT or PET scans where some parts of the image are mistaken for malignancy. Physiological uptakes of bone and soft tissue regions include skeletal muscle and bone marrow (5–7). Except for these physiological uptakes, non-specific uptakes which include healing bone, lymph nodes and joints can also mimic malignancy.
Healing bone associated with elevated 18FDG uptake was the most common unexpected bony lesion (8). Although pathologic confirmation was not obtained in all subjects, the rib was the most frequent site of abnormal uptake and derived from healing bone which could be verified by clinical history in our study. The granulation tissue in an earlier phase of healing will be attributable for enhanced 18FDG uptake. In the late healing phase, the procallus can be associated with elevated glucose metabolism. It is not possible to estimate how much of the effect on the healing process is due to significant 18FDG uptake in this retrospective study. Physiological uptake in single vertebral marrow can be misinterpreted as malignancy, but SUV is usually modest with less than 3.0 and uptake pattern is often uniform (9). In our study, vertebra was the second frequent site of 18FDG uptake. Uptake pattern of these lesions was non-uniform but SUV was modest around 3.0 except for two cases with non-Hodgkin lymphoma and osteoblastoma.
The generalized inflammatory response of lymph nodes is a common cause of 18FDG uptake, but reactive lymph nodes are found in two of our cases. Asymmetrical or isolated uptake in major muscles is a misleading finding, but none of our cases represent abnormal 18FDG uptake in major muscles. Glycolytic metabolism is elevated in the infiltration associated with inflammatory processes, resorption of necrotic debris, hematoma, or thrombus (5–7). In our study, five cystic soft tissue lesions except for one cystic neurinoma with homogeneous 18FDG uptake were the most common unexpected soft tissue lesions. The precise reasons for 18FDG uptake in cystic soft tissue lesions are unclear. However, such cystic lesions are often associated with inflammatory or hemorrhagic processes which can explain 18FDG uptake within the lesion.
In our study, one neurinoma arising from chest wall showed a high SUV of 4.8. Previous studies showed a significant 18FDG uptake in neurinoma with wide variations (10,11). Beaulieu et al. suggested that there was no correlation between the degree of 18FDG uptake and tumor size or proliferation rate (12). It is proposed that over-expression of glucose transporter protein may be attributable for high 18FDG accumulation in neurinoma. Over-expression of glucose transporter protein type 1 (GLUT-1) is one of the major causes of 18FDG uptake in malignant tumors. As for a neuronal histologic tumor type of benign perineurioma, over-expression of GLUT-1 is characteristic (13). Over-expression of glucose transporter protein type 3 (GLUT-3) which exists on the neuronal surface may be also attributable for 18FDG uptake in neurinoma.
Strauss and colleagues conducted two-compartment model to analyze 18FDG tracer uptake in 19 patients with giant cell tumor and gene chip analysis (U95A) (14). Tumor shows enhanced 18FDG uptake which correlated with bone cell differentiation and proliferative activity. 18FDG uptake also correlated with the expression of vascular endothelial growth factor. In our study, one tenosynovial giant cell tumor arising from soft tissue adjacent to the facet joint showed highest SUV of 10.6. Although giant cell tumors are classified as benign tumors, their angiogenetic activity is potent.
There were limitations in the present study. First, the of cancer screening PET study covered from head to thigh, which was not enough to include bone and soft tissue abnormalities in the lower extremities. However, our selective protocol with PET study was undertaken to screen for cancerous lesions mainly from the head to trunk. Second, a total of 24 lesions (38%) remained undetermined regardless of radiologic follow-up. Although CT, MRI and radiograph were used for follow-up, the follow-up duration was too short to determine the nature of these lesions. Most of these undetermined lesions demonstrate unclear findings on follow-up imaging studies and may not require immediate investigation.
In conclusion, the data presented do not suggest whether cancer screening using PET is effective or not. However, unsuspected bone and soft tissue lesions were encountered. There was a wide variation of pathologic and clinical diagnoses including one malignant lymphoma. The most common bone and soft tissue lesions were healing bone and cystic tumors, respectively. These could be interpreted as primary or metastatic lesions. Correlation with clinical history and other imaging findings is essential in differential diagnosis.
|
Conflict of interest statement |
|---|
|
TOPAbstractINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONConflict of interest statementReferences |
|---|
None declared.
|
Acknowledgments |
|---|
This work was supported in part by grants from Scientific Research Expenses for Health and Welfare Programs, No. 17-12, the promotion and standardization of diagnostic accuracy in PET/CT imaging, the Grant-in-Aid for Cancer Research from the Ministry of Health, Labour and Welfare, and Travel Grant of the Princess Takamatsu cancer Research Fund.
|
References |
|---|
|
TOPAbstractINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONConflict of interest statementReferences |
|---|
1 Sobue T, Moriyama N, Kaneko M, et al. (2002) Screening for lung cancer with low-dose helical computed tomography: anti-lung cancer association project. J Clin Oncol 20 911–20.
2 Yasuda S and Shohtsu A. (1997) Cancer screening with whole-body 18F-fluorodeoxyglucose positron-emission tomography. Lancet 350 1819.[CrossRef][Web of Science][Medline]
3 Yasuda S, Ide M, Fujii H, Nakahara T, Mochizuki Y, Takahashi W, et al. (2000) Application of positron emission tomography imaging to cancer screening. Br J Cancer 83 1607–11.[CrossRef][Web of Science][Medline]
4 Yasuda S and Ide M. (2005) PET and cancer screening. Ann Nucl Med 19 167–77.[Web of Science][Medline]
5 Gordon BA, Flanagan FL, Dehdashti F. (1997) Whole-body positron emission tomography: normal variations, pitfalls, and technical considerations. Am J Roentgenol 169 1675–80.
6 Cook GJR, Fogelman I, Maisey MN. (1996) Normal physiological and benign pathological variants of 18-fluoro-2-deoxyglucose positron-emission tomography scanning: potential for error in interpretation. Semin Nucl Med 26 308–14.[CrossRef][Web of Science][Medline]
7 Engel H, Steinert H, Buck A, Berthold T, Huch Boni RA, von Schulthess GK. (1996) Whole-body PET: physiological and artifactual fluorodeoxyglucose accumulations. J Nucl Med 37 441–6.
8 Meyer M, Gast T, Raja S, Hubner K. (1994) Increased F-18 FDG accumulation in an acute fracture. Clin Nucl Med 19 13–4.[CrossRef][Web of Science][Medline]
9 Sugawara Y, Fisher SJ, Zasadny KR, Kison PV, Baker LH, Wahl RL. (1998) Pre-clinical and clinical studies of bone marrow uptake of fluorine-18-deoxyglucose with or without granulocyte colony-stimulating factor during chemotherapy. J Clin Oncol 16 173–80.
10 Ahmed AR, Watanabe H, Aoki J, Shinozaki T, Takagishi K. (2001) Schwannoma of the extremities: the role of PET in preoperative planning. Eur J Nucl Med 10 1541–51.
11 Shah N, Sibtain A, Saunders MI, Townsend E, Wong WL. (2000) High FDG uptake in a schwannoma: a PET study. J Comput Assist Tomogr 24 55–6.[CrossRef][Web of Science][Medline]
12 Beaulieu S, Rubin B, Djang D, Conrad E, Turcotte E, Eary JF. (2004) Positron emission tomography of shwannomas: emphasizing its potential in preoperative planning. Am J Roentgenol 182 971–4.
13 Yamaguchi U, Hasegawa T, Hirose T, Fugo K, Mitsuhashi T, Shimizu M, et al. (2003) Screlosing perineurioma: a clinicopathological study of five cases and diagnostic utility of immunohistochemical staining for GLUT1. Virchows Arch 443 159–63.[CrossRef][Web of Science][Medline]
14 Strauss LG, Dimitrakopoulou-Strauss A, Koczan D, Bernd L, Haberkorn U, Ewerbeck V, et al. (2004) 18F-FDG kinetics and gene expression in giant cell tumors. J Nucl Med 45 1528–35.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||