| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Report |
1 Department of Radiology, Division of Nuclear Medicine, Box 356113, University
of Washington Medical Center, 1959 NE Pacific St., Seattle, WA 98195.
2 Department of Pathology, Box 356100, University of Washington Medical Center,
Seattle, WA 98195.
3 Department of Orthopedics, Box 356500, University of Washington Medical
Center, Seattle, WA 98195.
Received February 10, 2003; accepted after revision July 2, 2003.
Address correspondence to J. F. Eary.
OBJECTIVE. In this article, we describe FDG uptake in schwannoma as measured on positron emission tomography (PET). FDG uptake is compared with tumor cellularity, tumor size, and tumor proliferation rate (Ki-67 index).
CONCLUSION. Schwannomas generally have a high tumor-to-background ratio on FDG PET. Semiquantitative analysis with standardized uptake values (SUVs) reveals a wide variation in SUVs that can be explained by variations in the degree of cellularity. No correlation was found between FDG uptake and tumor size or tumor proliferation rate (Ki-67 index). Because these tumors often have a high level of FDG uptake, distinguishing schwannomas from malignant peripheral nerve sheath tumors before biopsy or surgery is not possible. Even in cases in which the maximum SUV or average SUV is greater than 6.0, schwannomas cannot be excluded. Therefore, schwannoma should be included in the differential diagnosis of peripheral nerve sheath tumors with low, intermediate, or high SUVs.
FDG positron emission tomography (PET) has been used extensively to quantify tumor metabolism in bone and soft-tissue sarcoma before surgery [1–3]. FDG uptake by malignant tumors correlates with tumor grade, is a good prognostic indicator, and is used to assess response to therapy. Compared with the information that is available about malignant tumors, little is known about FDG uptake in benign bone and soft-tissue tumors. More knowledge about FDG uptake in benign mesenchymal neoplasms is needed to determine whether FDG PET can be used to distinguish benign from malignant bone and soft-tissue tumors.
In this report, we describe our experience with FDG PET of schwannomas. Schwannomas are most commonly observed in men and women in the third to fifth decades of life and typically arise from the spinal nerve roots and the cervical, sympathetic, vagus, peroneal, and ulnar nerves in the head, neck, and flexor surfaces of the upper and lower extremities. Most of these tumors are solitary and unassociated with neurofibromatosis 1, although they frequently occur in patients with this disease. They commonly present as nonspecific masses that are associated with a peripheral nerve. Resection is performed only if they are symptomatic, are moderately large, or exhibit rapid growth.
Malignant peripheral nerve sheath tumor is an uncommon but aggressive sarcoma that most commonly arises in young and middle-aged adults as a rapidly growing mass, mainly in the trunk and proximal portions of the extremities. It has a propensity for systemic spread, especially to the lungs, which ultimately causes death in the majority of patients. Correct classification is necessary for optimal management in patients with malignant peripheral nerve sheath tumor.
CT and MRI are an effective means of delineating lesions and their relationship to surrounding structures, but CT and MRI are not considered sufficiently accurate to distinguish benign from malignant nerve sheath tumors [4, 5]. The use of needle biopsy should be restricted to highly selective situations because of the possibility of nerve injury. Open biopsy may not confirm the diagnosis. In light of these limitations, a noninvasive technique that enables distinction of schwannomas from malignant peripheral nerve sheath tumor or other malignant tumors would be of great value for treatment planning.
Subjects and Methods
Patients and Tumors
From 1996 to 2001, nine patients with schwannoma (four men and five women),
26 to 78 years old, were treated at the University of Washington Sarcoma
Clinic and underwent FDG PET before surgery. One patient with two synchronous
tumors that were proven to be schwannomas was known to have neurofibromatosis
1 and several spinal nerve root lesions confirmed to be schwannomas at
histologic examination. All the other patients had solitary soft-tissue
lesions without evidence of neurofibromatosis. The study was approved by the
human institutional review board, and informed consent was obtained from all
patients.
Histology and Immunohistochemistry
Tumor size was estimated by the product of the three axial measurements
obtained from the surgical specimen. Approximately one histologic section was
processed for each centimeter of greatest dimension of each tumor. Antibodies
to S-100 protein (polyclonal antibody at titer 1:4,000, Dako) and Ki-67
(MIB-1, monoclonal antibody at titer 1:400, Dako) antigens were used.
Cellularity was determined semiquantitatively by assessing the percentage of
each lesion that corresponded to a typical Antoni A area. The Ki-67 index was
determined by counting 1,000 cells in the most active area and dividing the
number of cells with unequivocal nuclear staining by the total number of cells
and expressing this number as a percentage.
PET Studies
All imaging studies were performed using the Advance PET scanner (General
Electric Medical Systems) in a 2D high-sensitivity mode, an axial field of
view of 15 cm (4.0 mm axial full width at half maximum at the center of the
tomograph), and an in-plane resolution of 4–5 mm
[6]. Before the PET study,
patients fasted for a minimum of 6 hr, and normal blood glucose levels were
confirmed before FDG injection. A mean dose of 385 MBq (10.4 mCi) FDG was
injected through an IV line placed in one of the patient's arms. Imaging was
initiated 47–73 min after injection in 1–5 fields of view with a
3- to 15-min emission scan covering the area of the tumor. This scan was
followed by a transmission scan that was obtained over the same regions for
attenuation correction. Images were reconstructed with a filtered
backprojection algorithm using a Hanning filter, resulting in a reconstructed
resolution of approximately 12 mm.
Image Analysis
Images were retrospectively reviewed on a workstation in three orthogonal
planes. Semiquantitative analysis was performed using standardized uptake
values (SUVs). Over all tumors, circular regions of interest composed of 12
pixels (0.7851 cm2) were drawn from the attenuation-corrected axial
slice that included the pixel showing the highest SUV. Both the maximum SUV
(SUVmax) and the average SUV (SUVav) were calculated
using the following formula:
 |
FDG uptake was compared with cellularity, tumor size, and Ki-67 index.
The mean SUVs for FDG (SUVmax and SUVav) in the hypocellular group and the hypercellular group were compared using a Mann-Whitney rank sum test for equality of medians. FDG SUV was also compared with tumor size and Ki-67 using a linear regression analysis.
Results
Gross inspection revealed the lesions were well circumscribed and had a firm and fibrous to myxoid cut surface. Histology results indicated that all the lesions were typical of conventional schwannoma without cellular or epithelioid variants. The lesions consisted of varying proportions of hypercellular (Antoni A) and hypocellular (Antoni B) areas. The global degree of cellularity varied widely among lesions. On the basis of the degree of cellularity, lesions were assigned to one of two groups. The hypocellular group consisted of lesions containing 10% or less Antoni A areas characterized by short fascicles of spindle cells, and the hypercellular group consisted of lesions with more than 10% Antoni A areas (Table 1). The stroma of each lesion ranged from hyalinized to myxoid, and the cells of the lesion were bland-appearing spindle cells with abundant palely eosinophilic cytoplasm and small nuclei with fine chromatin and inconspicuous nucleoli. Admixed with the spindle cells were variable numbers of hyalinized blood vessels and occasional lymphocytes (Fig. 1A, 1B). No mitotic figures were identified.
|
|
|
Immunohistochemical studies revealed each lesion to be diffusely and strongly positive for S-100 protein. For two of the larger lesions, one lesion in a 78-year-old woman had focal ischemic necrosis involving approximately 10% of the lesion, and one lesion in a 42-year-old man had a large number of cystic blood-filled spaces. The tumor proliferation rate was assessed by Ki-67 immunochemistry and showed a wide range (Table 1), all less than 50%.
Visual qualitative assessment of the images showed high tumor-to-background ratio for all the tumors proven to be schwannoma (Figs. 2,3,4), which allowed easy detection, with the exception of one patient for whom the tumor-to-background ratio was low. The three larger lesions had heterogeneous uptake, whereas the others had homogeneous uptake (Figs. 2 and 3). One patient underwent a whole-body PET study, and several foci of abnormally increased FDG uptake were found separate from the two proven to be schwannomas (Fig. 2). These foci of increased uptake, although suspected of being schwannomas, were not biopsied or surgically removed.
|
|
|
Table 1 lists the results of a semiquantitative analysis of FDG uptake, tumor size, cellularity, and Ki-67 index (expressed as a percentage). Tumor SUVs showed wide variations. The SUVmax varied from 1.9 to 7.2 (mean = 4.6) and the SUVav, from 1.6 to 6.3 (mean = 3.9). The mean SUV (SUVmax and SUVav) of the hypocellular tumors was significantly lower than the mean SUV (SUVmax and SUVav) of the hypercellular tumors (p = 0.010 for SUVmax and p = 0.010 for SUVav) (Fig. 5). No association was found between SUV and tumor size or Ki-67 index.
|
Discussion
Ahmed et al. [7] found FDG uptake in 22 lesions proven to be schwannomas. As in our series, the SUVav showed wide variation (SUVmax, not reported), but unlike our series, the SUVavs were lower (range, 0.3–3.7). Only 5% (1/22) of the lesions in their series showed an SUVav greater than 3.0, compared with 50% (5/10) in our series. Given that eight of 22 lesions had an SUVav in the malignant range (> 1.9 as defined in a previous study [8]), the researchers concluded that FDG PET was of limited value as a preoperative diagnostic imaging technique for the assessment of schwannoma versus sarcoma. Even with the use of a higher cutoff range of 2.5, all but one lesion in our study would have been considered malignant. We reached the same conclusion as Ahmed et al.: FDG PET has limited value for identifying benign versus malignant peripheral nerve sheath tumors. The FDG uptake of schwannomas can be variable. An SUVmax of 6.0 cannot exclude schwannoma from the differential diagnosis. A case report of high FDG uptake (SUV = 12) in a paravertebral schwannoma was recently described [9], which further supports our conclusions. A case of paravertebral schwannoma with high FDG uptake (SUV = 6.7) was described by Knight et al. [10] as a false-positive for malignant pulmonary lesion. Our data suggest that high FDG uptake is not an anomaly but rather a common occurrence in schwannoma.
One must exercise caution when using SUVav to quantify FDG uptake over a tumor because any change in the contour will change the average count. Therefore, SUVav is generally considered more technique-dependent than the SUVmax. The SUV increases as times progresses after injection, and the extent of this phenomenon is not well known for several tumor types. All tumors studied by us were imaged at a later time (> 40 min after injection) as opposed to the study by Ahmed et al. [7]. This difference might account for some of the increase in SUV observed in our series.
Our finding that FDG uptake was heterogeneous in the three larger tumors was not surprising because larger soft-tissue masses exhibit frequent cystic changes and necrosis [11] as found in two of the three largest tumors of our series. The wide range of SUVs for FDG appears to be explained by the different degrees of cellularity for each lesion (Fig. 5). However, the reason high FDG accumulation is found in benign tumors such as schwannoma remains unclear, and the higher FDG SUVs in our series did not correlate with increased tumor size or Ki-67 index. Perhaps overexpression of one of the glucose transporter proteins by tumor cells could provide an explanation, but this theory remains to be proven. Glucose transporter type 3 (GLUT-3) is found in all human tissues and is the major glucose transporter on the neuronal surface and could explain the high FDG accumulation in the constituent neoplastic cells in schwannomas.
Because FDG PET has limited value in distinguishing schwannomas from malignant peripheral nerve sheath tumors or other malignant soft-tissue tumors, PET radiotracers designed to study tumor characteristics other than glucose metabolism, such as tumor proliferation, should be assessed.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  F. A. Sawas, O. Lababede, M. A. Meziane, and A. V. Arrossi A 54-Year-Old Woman With Incidentally Discovered Mass on a Chest Radiograph Chest, June 1, 2009; 135(6): 1673 - 1678. [Full Text] [PDF]
|
|
|
|
 |
|
 J M Son, M I Ahn, K D Cho, J Yoo, and Y H Park Varying degrees of FDG uptake in multiple benign neurofibromas on PET/CT Br. J. Radiol., September 1, 2007; 80(957): e222 - e226. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Maeda, U. Tateishi, T. Terauchi, C. Hamashima, N. Moriyama, Y. Arai, E. E. Kim, and K. Sugimura Unsuspected Bone and Soft Tissue Lesions Identified at Cancer Screening using Positron Emission Tomography Jpn. J. Clin. Oncol., March 1, 2007; 37(3): 207 - 215. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-J. Kim, C.H. Baek, Y.H. Ko, and J.Y. Choi Neurothekeoma of the Tongue: CT, MR, and FDG PET Imaging Findings. AJNR Am. J. Neuroradiol., October 1, 2006; 27(9): 1823 - 1825. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
= mean SUVmax.