Discussion
SFT, a rare neoplasm, was first described in 1931 by Klemperer and Rabin as a spindle cell tumour that originated from the pleura [
1]. Although SFTs are not frequently observed in extrapleural locations, they have been found to arise in any location, including the spine [
2‐
4]. Spinal SFTs usually arise within intramedullary (58%), intradural extramedullary (24%), or extradural (18%) components [
6]; to the best of our knowledge, no previous reports in the scientific literature have described a malignant SFT arising from the vertebral arch.
On MRI scans, most SFTs appear hypointense or isointense relative to the muscle on T1-weighted images and have a heterogeneously hyperintense tumour component on T2-weighted images [
7]. Additionally, SFTs are usually well delineated and tend to exhibit a lobulated shape with displaced adjacent structures. Local invasion is rare [
8]. However, in the current case, the findings of mild tumour hyperintensity on T1-weighted images, an unclear margin, and local invasion were suggestive of malignancy.
Histologically, SFTs comprise spindle cells in a “patternless pattern” characterised by a combination of alternating hypocellular and hypercellular areas interspersed with thick bands of hyalinised collagen [
9]. However, it may be difficult to distinguish collagen from other eosinophilic extracellular materials, especially osteoid, as this process is subjective [
7]. Accordingly, it is extremely difficult to histologically distinguish SFTs arising in the bone from other neoplasms, especially osteosarcoma. Initially, in this case, we considered the extracellular matrix to be osteoid, leading to an initial diagnosis of osteosarcoma.
Immunohistochemical examination is useful for distinguishing SFTs from other neoplasms. Most SFTs exhibit positive staining for CD34, bcl-2, and CD99, but negative staining for αSMA, desmin, and S-100 protein. However, these immunohistochemical markers are nonspecific and can be detected in a variety of other neoplasms [
10]. Recently, whole-genome sequencing was used to detect a fusion gene of
NAB2 and
STAT6 on chromosome 12 in the majority of SFTs [
11]. Moreover, immunohistochemical analyses of STAT6 revealed a diffuse nuclear expression in almost all SFTs, but not in other tumours [
12,
13]. Therefore, STAT6 could be considered a highly sensitive and nearly ideal marker for the differentiation of SFT from other neoplasms [
13]. In this case, although the initial diagnosis was osteosarcoma, a re-examination that included immunohistochemical detection of STAT6 and RT-PCR detection of the
NAB2 exon 6-STAT6 exon 16 fusion led us to confirm a diagnosis of malignant SFT.
Although most SFTs have benign features, these tumours may recur or metastasise in the absence of any predictive morphological features [
14]. As a result, SFTs are now classified as an intermediate category within the 2013 World Health Organization criteria of tumours of the soft tissue and bone [
7]. Vallet-Decouvelare et al. reported that 80% of extrathoracic SFTs with atypical histological features—increased cellularity, nuclear pleomorphism, necrosis, and four or more mitoses per ten high-power fields—developed local recurrences or distant metastases [
14]. In addition, Gold et al. reported that positive surgical margins and a tumour size of 10 cm or larger are prognostic factors for poor metastasis-free survival [
3]. From a genetic point of view, Barthelmess et al. reported that in comparison to other types of SFTs, those harbouring
NAB2-exon 6-STAT6 exon 16/17 fusions more frequently exhibited cellular histologic features associated with more aggressive behaviour [
15]. The present case, which featured infiltration, nuclear atypia, and mitosis with clinically aggressive behaviour, was ultimately diagnosed as malignant SFT.
In this case, we performed an interbody fusion using anterior, rather than posterior, instrumentation because posterior reconstruction after extensive soft tissue removal could increase the risk of postoperative infection. Fortunately, the patient’s postoperative course was uneventful, and she was immediately able to receive adjuvant chemotherapy after surgery.
We found the T-saw to be useful for cutting pedicles. The exact locations of these pedicles were difficult to identify because the posterior elements of the vertebrae were almost completely covered with extensive soft tissue, which had to be resected en bloc. It might have been dangerous to cut the pedicles from the lateral side to the medial side using an osteotome in this case, given the risk of neural damage. Although additional time was needed to set the T-saw around the pedicles, the process of cutting the pedicle from the medial wall to the lateral wall was safer.
Some authors have reported the surgical resection of SFTs arising from the spine [
2,
6,
8]. Bouyer et al. reported the case of a SFT of the thoracic spine with two recurrences, and another author reported a case in which a malignant SFT was resected surgically but a local recurrence developed rapidly [
8]. These cases demonstrate the difficulty associated with complete resection of a SFT arising from the spine. In our case, a total resection was achieved, and the patient had no evidence of local recurrence. However, she developed a lung metastasis 5 years after the primary resection and died of disease progression 8 years after surgery. Accordingly, a long follow-up is required because a late recurrence or metastasis can occur after more than 10 years [
14].