© 2004 Foundation for Promotion of Cancer Research
Diagnosis of Peripheral Nerve Sheath Tumors around the Pelvis
1 Division of Orthopedic Surgery, Graduate School of Medical and Dental Sciences, Niigata University, 2 Department of Orthopedic Surgery, Niigata Cancer Center Hospital, 3 Department of Radiology, Niigata Citizen Hospital, 4 Section of Pathology, Niigata University Hospital, Niigata, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: To distinguish between benign and malignant peripheral nerve sheath tumors around the pelvis.
Methods: A retrospective study of 30 patients with benign and malignant peripheral nerve sheath tumors located around the pelvis was performed. Clinical, imaging and histological features of 19 benign and 11 malignant peripheral nerve sheath tumors around the pelvis were reviewed retrospectively.
Results: Nearly all patients exhibited pain at presentation in cases involving both benign and malignant tumors. Although tumor size, duration of symptoms and presence of sensory disturbance possessed little value in differential diagnosis, severe motor weakness was observed exclusively in patients presenting with malignant tumors. On CT or MRI, central enhancement was apparent in 11 of the19 benign tumors; in contrast, central enhancement was evident in one of the 11 malignant tumors. Fine needle aspiration cytology was performed in 11 tumors; correct diagnosis was achieved in four tumors. Core needle biopsy was performed in five tumors, all of which were correctly diagnosed with no neurological deficits. Immunohistochemically, all benign tumors were diffusely positive for S-100 protein, whereas malignant tumors were negative or focally positive for S-100 protein. Ki-67 index was less than 4% in all benign tumors; additionally, this index was 7–36% in malignant tumors.
Conclusion: Central enhancement pattern on imaging studies strongly suggests a benign tumor; in contrast, severe motor weakness suggests malignant lesions. Core needle biopsy was reliable with respect to preoperative diagnosis.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Peripheral nerve sheath tumors (PNSTs) around the pelvis are relatively uncommon (1–11). Benign peripheral nerve sheath tumors (BPNSTs) in this region may develop large tumoral mass and may display marked nuclear atypia on a degenerative basis; as a result, differential diagnosis between benign and malignant peripheral nerve sheath tumors (MPNSTs) remains a challenge. Preoperative pathological diagnosis based on limited samples is particularly difficult when evaluating the degenerative features of tumors (1,2,6). Although a few series of studies exist on clinical features of pelvic benign schwannomas (1,5), details of clinical, imaging and histological features of MPNSTs around the pelvis have been scarcely reported. The purpose of this study was to determine the value of clinical signs, imaging features, biopsy and immunohistochemical evaluation in terms of differentiation of BPNST from MPNST around the pelvis.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Clinical Data
All patients at our institution diagnosed with and treated for BPNST or MPNST around the pelvis between 1980 and 2002 were reviewed. The tumors arose from the retroperitoneal, sacral, presacral, buttock and inguinal regions. Thirty patients presented with PNSTs, including 16 benign schwannomas, three neurofibromas and 11 MPNSTs. Clinical charts, preoperative images and histologic slides of all the 30 cases were reviewed. Strict criteria were used for the diagnosis of MPNST. The malignant neoplasms appeared to arise in a nerve or in patients displaying typical neurofibromatosis.
Muscle strength was graded from 0 (no activity) to 5 (normal) on manual muscle testing (MMT) (12). The worst grade of the affected muscle was recorded.
Preoperative computed tomography (CT) scans were available for review in all but one patient (15 schwannomas, 3 neurofibromas and 11 MPNSTs). Axial sections (5 or 10 mm in thickness) were obtained from all patients. Magnetic resonance imaging (MRI) of 24 patients (13 schwannomas, 2 neurofibromas and 9 MPNSTs) was performed by employing a variable system with varying field strength (0.5 to 1.5 T). Conventional scanning sequences included T1-weighted (400-700/15-32) and T2-weighted (1700-3300/80-110) spin echo sequences.
Histologic Data
Fine needle aspiration was performed with an 18-gauge needle with or without image guidance. Core biopsy was conducted with an 18-gauge needle (ASAP Detachable, Boston Scientific Corporation, Watertown, MA, USA) with or without image guidance. Each specimen (resected specimens and biopsy materials) was stained with routine hematoxylin and eosin. Immunoperoxidase methodology in terms of a Simple Stain MAX PRO system (Nichirei, Tokyo, Japan) was employed. Immunohistochemical reagents utilized in this study included primary monoclonal antibodies against Ki-67 (MIB-1) and a polyclonal antibody against S-100 protein (Dako, Tokyo, Japan).
Possible statistical correlation was examined by using Fisher’s exact method and Student’s t-test. P values <0.05 were considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The results are summarized in Tables 1, 2 and 3.
|
|
|
Clinical Findings
Benign Schwannoma and Neurofibroma
A total of 19 patients underwent surgery for BPNSTs (16 schwannomas and 3 neurofibromas). Twelve patients displayed solitary schwannomas, four patients demonstrated multiple schwannomas (schwannomatosis), two patients exhibited neurofibromatosis (NF1) and a single patient presented with segmental NF. Age at surgery ranged from 10 to 76 years (median, 54 years). In patients presenting with multiple tumors, the largest tumor occurring in each patient was examined in this series. The maximum dimension of individual tumors ranged from 4 to 20 cm; median size was 9 cm. Six, nine and four patients displayed sacral and presacral tumors, retroperitoneal tumors and buttock tumors, respectively. Six tumors arose from the sacral nerve root, five from the sciatic nerve, four from the femoral nerve, three from the lumbo-sacral plexus and a single tumor arose from the genitofemoral nerve. Direct connection with a major nerve was not identified in one patient with NF1.
Fifteen patients complained of pain at rest. Ten patients demonstrated sensory disturbance and six patients exhibited faint to mild motor weakness (MMT 4, five patients; MMT 3, one patient). The duration of symptoms ranged from 2 to 120 months (median, 12 months).
MPNST
Of the 11 cases of MPNST, seven patients displayed NF1. The maximum tumor size ranged from 9 to 17 cm (median, 12 cm). Age at surgery ranged from 16 to 61 years (median, 34 years). Retroperitoneal tumors were present in seven patients; furthermore, sacral and presacral tumors were observed in a single patient, whereas three patients exhibited buttock tumors. Three tumors arose from the lumbo-sacral plexus, two from the sciatic nerve, two from the femoral nerve, and a single tumor arose from the S2 root. Direct connection with a major nerve was not identified in three patients diagnosed with NF1.
All patients complained of pain at rest; additionally, eight and seven patients displayed sensory disturbance and faint to severe motor weakness (MMT 4, one patient; MMT 3, two patients; MMT 2, two patients; and MMT 0, two patients), respectively. The median duration of symptoms was 12 months (range, 3 to 96 months).
Imaging Findings
BPNST
Eighteen patients underwent examination by CT. All tumors, which were slightly hypodense relative to muscle, possessed smooth margins. Six of these tumors arose from sacral nerve roots, or the eroded sacral foramen of the individual nerve root. The margins of the tumors in the sacrum were sharp and marginal sclerosis was observed in these six cases. Eight tumors were homogenous, whereas 10 tumors were heterogeneous. Six tumors exhibited small, mineralized areas. Contrast-enhanced studies were available in all 18 cases; furthermore, nine cases displayed a round to irregular enhancement at the center of the tumoral mass or center of the nodules in the tumor (central enhancement) (Fig. 1A). Five tumors showed peripheral enhancement (Fig. 1B); irregular and homogenous enhancement were evident in three and one tumor, respectively.
|
On T1-weighted MR images, 14 tumors were iso- to hypo-intense with muscle. On T2-weighted images, all tumors revealed inhomogeneous hyperintensity; moreover, peripheral hyperintense rim and central low intensity (target sign) (Figs 2A, C) were detected in four of the 14 tumors. Central enhancement was observed in six of the 14 cases involving Gd-DTPA administration (Figs 2B, D). Massive cystic, hemorrhagic or necrotic degeneration (degenerative schwannoma) was apparent in five cases.
|
MPNST
CT features of 11 MPNSTs were reviewed. Eight tumors possessed smooth margins, and three were invasive. No MPNST showed central enhancement, whereas seven displayed irregular enhancement and peripheral enhancement was evident in four (Figs 1C, D).
MRI features of eight MPNSTs were reviewed. On T1-weighted images, all tumors were isointense with muscle; furthermore, all but one tumor possessed smooth margins. On T2-weighted images, all tumors were hyperintense. Enhanced T1-weighted imaging revealed heterogeneous enhancement in all but one case. Target sign and central enhancement were observed in a single case. Enhancement occurred predominantly in the peripheral area in two cases of MPNST; four cases exhibited irregular enhancement (Figs 2E, F). Only one case revealed central enhancement in the nodules of the tumor.
Preoperative Biopsy and Cytology
A histological diagnosis was made prior to or during operation in all cases. Open biopsy was performed in 15 cases (10 BPNSTs, five MPNSTs). Six of these cases were diagnosed by intra-operative frozen section. All permanent sections of biopsy were accurately diagnosed; however, one MPNST was diagnosed as benign schwannoma by frozen section. Fine needle aspiration cytology (FNA) was performed in five BPNSTs and six MPNSTs. Of these 11 cases, CT guided procedure was conducted in four patients. Only four tumors were accurately diagnosed. Five samples did not yield adequate information, and additional open biopsy was performed. One schwannoma was diagnosed as spindle cell sarcoma; furthermore, one MPNST was diagnosed as neurofibroma by FNA. Four schwannomas and a single MPNST were accurately diagnosed by core needle biopsy. All needle biopsy procedures resulted in temporary radiating pain; however, neurological deficits were not observed.
Pathologic Findings and Imaging Correlation
Of the 16 schwannomas, 14 tumors were classic schwannomas with Antoni A and B areas, and two neoplasms were cellular variants. Mitotic structures were scarcely observed in all schwannomas. The central enhancement in the BPNSTs, which was apparent in classic schwannoma or neurofibroma, corresponded to the central location of the hypercellular Antoni A area (Fig. 3A) or compact, packed tumor cells and collagen in neurofibroma. Peripheral enhancement in BPNSTs usually corresponded to cystic or hemorrhagic degeneration of the degenerative schwannomas. Peripheral enhancement in MPNSTs corresponded to peripheral location of vivid tumor tissues and central necrosis (Fig. 3B). One MPNST with central enhancement displayed a relatively hypercellular area at the center of the tumor. One MPNST showed rhabdomyoblastic differentiation (Malignant Triton Tumor, Case M5).
|
Immunohistochemical Results
Immunohistochemical staining results for S-100 protein and Ki-67 are illustrated in Table 2. All 19 BPNSTs displayed diffuse immonoreactivity for the S-100 protein. No BPNST exhibited nuclear staining of Ki-67 exceeding 4% (median, 1.2%). In six of the 11 MPNSTs, S-100 protein was positive; however, the positive cells were focal in all six instances. 7–38% of the tumor cell nuclei revealed immonoreactivity for Ki-67 in MPNST (median, 23%) (Figs 4A–G).
|
Age, association of NF1, central enhancement on CT or MRI and immunostaining of S-100 protein and Ki-67 were significantly different between BPNST and MPNST (Table 3).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
While optimal treatment of MPNST involves en-bloc surgical resection so as to include surrounding nerves, with or without radiotherapy, BPNST can usually be resected without neurological deficit. BPNSTs around the pelvis often develop large lesions. Moreover, imaging studies reveal an inhomogeneous, hypervascular, bone-destructive tumoral mass; consequently, differentiation from MPNST is often difficult (1–11).
This study demonstrated that MPNST around the pelvis is very rare in patients without neurofibromatosis. Twelve and four cases of BPNST and MPNST, respectively, in the absence of NF were observed in the present study. On the other hand, seven cases each of BPNST and MPNST with NF were observed. The true ratio of BPNST to MPNST in NF patients is unclear due to the rarity of treatment of non-symptomatic tumors in patients with NF. However, the clinician should consider the possibility of MPNST when pelvic tumors in NF patients produce these symptoms.
It has been reported that in the extremities, careful clinical and imaging examinations could distinguish between most benign and malignant peripheral nerve sheath tumors (8). Pain at rest or neurological deficit is often observed in MPNST; in contrast, BPNST seldom produces these clinical signs in the extremities (11). However, our data indicate that pain at rest, tumor size and duration of symptoms possess little value in terms of differentiation of BPNST from MPNST around the pelvis. Pelvic PNST tend to be larger in size than PNST in the extremities; in addition, pelvic PNST often arise from the sacral foramen. As a result, both benign and malignant PNST may produce pain. Although faint motor weakness (MMT 4) was occasionally present in patients with BPNST, severe motor weakness (MMT 0–2) was observed exclusively in patients with MPNST. Therefore, evaluation of motor weakness is important with respect to differentiation of benign and malignant tumors.
Imaging and Histological Findings
Most previous reports have stated that imaging studies offer some benefit in terms of differentiation of BPNST from MPNST (8–10). It has been documented that 40–50% of BPNSTs occurring in the extremities display central enhancement pattern (8). Histologically, the pattern comprises a central zone of tightly packed cellular components (Antoni A area in schwannoma) surrounded by hypocellular myxoid material (Antoni B area in schwannoma). This zoning also corresponds to the target sign in T2-weighted MR images, which consists of a central low signal intensity with a peripheral ring of high signal intensity. The central enhancement and target sign are rarely observed in MPNST (12–14). In this series, central enhancement was apparent in half of the BPNST cases; however, it was present in only one of the 11 cases of MPNST. In general, large mesenchymal tumors often show degenerative changes at the tumoral center; enhancement is usually evident at the periphery of the tumor. Therefore, central enhancement and target sign have diagnostic value in BPNST (12). Although the target sign was scarcely observed in CT, central enhancement was often present in CT as well as in MRI.
Degenerative schwannoma, which often arises in this lesion, exhibits cyst formation, hemorrhage, calcification and hyalinization. Imaging findings of degenerative schwannoma also revealed such degenerative changes; moreover, enhancement was observed at the periphery of the tumor. Invasive margin was apparent in three of the 11 MPNSTs; in contrast, no BPNST displayed invasive margin in this series. Although the incidence of invasive margin in MPNST is low, margin of the tumor also possesses diagnostic value.
Preoperative pathological diagnosis based on a limited number of samples is particularly difficult. FNA is often not beneficial with respect to the investigation of degenerative tumors, falsely suggesting malignancy due to cellular atypism and poor sampling from degenerative tissue (15,16). Among the 10 cases presenting with FNA in this series, one degenerative schwannoma was overdiagnosed as sarcoma; additionally, five samples failed to provide useful information. These disappointing results may be due to the degenerative nature of pelvic PNST and deep tumoral location. We recently employed core needle biopsy with ASAP needle system in five cases. These five samples exhibited suitable quality and were diagnosed accurately. In instances where the sample showed cytological atypism, additional immunohistochemical analysis was useful for arriving at an accurate diagnosis. However, it is not possible to perform needle biopsy in all patients with pelvic PNST. We performed the needle biopsy under CT guidance, when the needle could be inserted through the posterior approach. Careful biopsy planning is needed to prevent complications such as bleeding and intestinal damage.
The S-100 protein is widely used as an immunohistochemical marker for PNST. Although S-100 protein can be identified in 50–90% of MPNSTs, staining is typically focal and limited to a small number of cells. Strong, diffuse immunoreactivity for S-100 protein always suggests BPNSTs, including degenerative or cellular schwannomas (17). The Ki-67 protein is expressed in the proliferative phase (G1, S and G2) of the cell cycle. Ki-67 indices were reported in 5–65% of MPNSTs; in contrast, indices for BPNST were typically less than 5% (18). In this series, strong, diffuse immunoreactivity for S-100 protein was observed exclusively in BPNST, and over 5% of Ki-67 index was observed solely in MPNST. Therefore, these immunohistochemical analyses offer a powerful tool with respect to differentiation of BPNST from MPNST. Core biopsy specimens provide suitable quality for immunohistochemical analysis. Although all patients included in the present study, who underwent core needle biopsy, complained of radiating pain along the nerves, no neurological deficits were observed following the biopsy. Clinicians should be aware that the needle tract of a biopsy is often contaminated by tumor cells, and it should be resected or irradiated when the tumor is malignant.
Positron emission tomography employing the glucose analogue 18F-fluorodeoxyglucose (18FDG PET), which is a dynamic imaging technique, has been reported to be a useful tool for the detection of MPNST in patients presenting with NF (19). Although we have no experience with 18FDG PET, a dynamic study of this caliber may improve diagnostic accuracy.
Differentiation of BPNST from MPNST remains problematic. However, careful evaluation of motor weakness, recognition of central enhancement in BPNST, core needle biopsy and immunohistochemical examination possess diagnostic value in terms of differentiation of benign and malignant PNST.
|
FOOTNOTES |
|---|
+ For reprints and all correspondence: Akira Ogose, Division of Orthopedic Surgery, Graduate School of Medical and Dental Sciences, Niigata University, Asahimachi 1–751, Niigata 951-8510, Japan. E-mail: aogose{at}med.niigata-u.ac.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 Abernathey CD, Onoerio BM, Scheithauer B, Pairolero PC, Shives TC. Surgical management of giant sacral schwannoma. J Neurosurg 1986;65:286–95.[Web of Science][Medline]
2 Turk PS, Peters N, Libbey NP, Wanebo HJ. Diagnosis and management of giant intrasacral schwannoma. Cancer 1992;70:2650–7.[CrossRef][Web of Science][Medline]
3 Ortolan EG, Sola CA, Gruenberg MF, Vazquez FC. Giant sacral schwannoma. A case report. Spine 1996;21:522–6.[CrossRef][Web of Science][Medline]
4 Acciarri N, Staffa G, Poppi M. Giant sacral schwannoma: removed by anterior, transabdominal approach. Br J Neurosurg 1996;10:489–92.[CrossRef][Web of Science][Medline]
5 Dominguez J, Lobato RD, Romas A, Rivas JJ, Gomez PA, Castro S. Giant intrasacral schwannomas: report of six cases. Acta Neurochir 1997;139:954–60.
6 Hide LG, Bansouin CJ, Murray SA, Malcolm AJ. Giant ancient schwannoma of the pelvis. Skeletal Radiol 2000;29:538–42.[CrossRef][Web of Science][Medline]
7 Ducatman BS, Scheithauer BW, Piepgras DG, Reiman HM, Ilstrup DM. Malignant peripheral nerve sheath tumors. A clinicopathologic study of 120 cases. Cancer 1986;57:2006–21.[CrossRef][Web of Science][Medline]
8 White W, Shiu MH, Rosenblum MK, Erlandson RA, Woodruff JM. Cellular schwannoma. A clinicopathologic study of 57 patients and 58 tumors. Cancer 1990;66:1266–75.[CrossRef][Web of Science][Medline]
9 Takeyama M, Koshino T, Nakazawa A, Nitto H, Nakamura J, Saito T. Giant intrasacral cellular schwannoma treated with high sacral amputation. Spine 2001;25:E216–9.[CrossRef]
10 Nakano M, Miwa K, Kanimoto Y. A case of retroperitoneal ancient schwannoma in the pelvis. Hinyokika Kiyo 2001;47:473–6.[Medline]
11 Ogose A, Hotta T, Morita T, Yamamura S, Hosaka N, Kobayashi H, et al. Tumors of peripheral nerves: correlation of symptoms, clinical signs, imaging features, and histologic diagnosis. Skeletal Radiol 1999;28:183–8.[CrossRef][Web of Science][Medline]
12 Hislop HJ, Montgomery J. Danniels and Worthingham’s muscle testing, 6th edn. Philadelphia: WB Sauders 1995;1–9.
13 Bhargava R, Parham DM, Lasater OE, Chari RS, Chen G, Fletcher BD. MR imaging differentiation of benign and malignant peripheral nerve sheath tumors: use of the target sign. Pediatr Radiol 1997;27:124–9.[CrossRef][Web of Science][Medline]
14 Murphey MD, Smith WS, Smith SE, Kransdorf MJ, Temple HT. Imaging of musculoskeletal neurogenic tumors: radiologic-pathologic correlation. Radiographics 1999;19:1253–80.
15 Lopez-Rios F, Perez-Barrios A, Alberti N, Vargas J, Lozano F, de Agustin P. Fine-needle aspiration biopsy of the retroperitoneum: a series of 111 cases not including specific organs. Diagn Cytopathol 2002;27:85–9.[CrossRef][Web of Science][Medline]
16 Henke AC, Salomano DR, Hughes JH. Cellular schwannoma mimics a sarcoma: an example of a potential pitfall in aspiration cytodiagnosis. Diagn Cytopathol 1999;20:312–6.[CrossRef][Web of Science][Medline]
17 Enzinger FM, Weiss SW. Soft tissue tumors, 3rd edn. St. Louis: Mosby, 1995:821–928.
18 Kindblom LG, Ahlden M, Meis-Kindblom JM, Stenman G. Immunohistochemical and molecular analysis of p53, MDM2, proliferating cell nuclear antigen and Ki67 in benign and malignant peripheral nerve sheath tumors. Virchows Arch 1995;427:19–26.[Web of Science][Medline]
19 Ferner RE, Lucas JD, O’Doherty MJ, Huges RAC, Smith MA, Cronin BF, et al. Evaluation of 18fluorodeoxyglucose positron emission tomography (18FDG PET) in the detection of malignant peripheral nerve sheath tumours arising from within plexiform neurofibromas in neurofibromatosis 1. J Neurol Neurosurg Psychiatry 2000;68:353–7.
Received April 8, 2004; accepted May 7, 2004
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. R. Bhosale, M. Patnana, C. Viswanathan, and J. Szklaruk The Inguinal Canal: Anatomy and Imaging Features of Common and Uncommon Masses RadioGraphics, May 1, 2008; 28(3): 819 - 835. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||