Primary Intraosseous Solitary Fibrous Tumor of the Mandible: Report of a Diagnostically Challenging Case with NAB2::STAT6 Fusion and Review of the Literature

Abstract

Introduction

Solitary fibrous tumor (SFT) represents an uncommon mesenchymal neoplasm affecting primarily the extremities and deep soft tissues with, overall, benign but locally aggressive biologic behavior and an underlying pathognomonic NAB2::STAT6 fusion. Intraosseous SFTs are infrequent, and involvement of the jawbones is exceedingly rare.

Case presentation

A 54-year-old woman presented with an asymptomatic, well-demarcated, multilocular radiolucency of the left posterior mandible featuring focally irregular borders, root resorption and lingual cortex perforation. The lesion had shown progressive growth over a 6-year period. Microscopically, a proliferation of predominantly ovoid and spindle-shaped cells with indistinct cell membrane borders, elongated, plump or tapered, hyperchromatic nuclei, and lightly eosinophilic cytoplasm was noted. Marked cytologic atypia, pleomorphism and mitoses were absent. A secondary population of epithelioid cells exhibiting ovoid or elongated vesicular nuclei, and abundant, pale eosinophilic or vacuolated cytoplasm was also present. The supporting stroma was densely fibrous with areas of marked hyalinization and variably-sized, ramifying, thin-walled vessels. By immunohistochemistry, lesional cells were strongly and diffusely positive for STAT6 and CD99, and focally immunoreactive for MDM2 and SATB2. Ki-67 was expressed in less than 5% of lesional cells, while most interspersed epithelioid cells were positive for the histiocyte marker, CD163. Molecular analysis disclosed a NAB2::STAT6 fusion confirming the diagnosis of SFT. The patient underwent segmental mandibulectomy.

Conclusions

Herein, we report the first case of primary intraosseous SFT of the mandible with complete documentation of its characteristic immunohistochemical and molecular features. Diagnosis of such unusual presentations may be further complicated by the challenging histomorphologic diversity of SFT.

Introduction

According to the 5th edition of the W.H.O. classification of bone and soft tissue tumors (WHO), solitary fibrous tumor (SFT) is defined as a fibroblastic neoplasm characterized by a haphazardly-arranged population of spindled-to-ovoid cells surrounding a prominent, branching and hyalinized vasculature within a variably collagenous stroma [1]. Paracentric inversion at chromosomal region 12q13 resulting in a NAB2::STAT6 gene fusion is pathognomonic for SFT [1–3]. Although most SFTs behave in a benign fashion, locally aggressive biologic behavior with distal or local recurrences are reported in 10–30% of cases [1, 4–6].

Clinically, the majority of SFTs affect adults of a broad age range with a peak incidence between 40 and 70 years without strong sex predilection [1]. SFTs are anatomically ubiquitous presenting as slow-growing, asymptomatic masses most commonly in the extremities or deep soft tissues, 30–40% of cases each, the abdominal cavity, pelvis, or retroperitoneum [1, 7]. Furthermore, the trunk and head and neck region account for 10–15% of SFTs each. In the head and neck, the sinonasal tract and orbit are favored although involvement of the oral cavity and salivary glands may also infrequently occur [1, 8].

Intraosseous SFTs are uncommon and may represent primary tumors or metastases. Bona fide examples of SFT affecting the jawbones are exceedingly rare with only sparse, previously reported, cases in the English literature [9–14]. However, in most of these cases, diagnosis of SFT was rendered solely upon histomorphologic criteria and, occasionally, in conjunction with a limited panel of ancillary immunostains. Notably, no previous case of SFT involving the jawbones has been shown to harbor an underlying NAB2::STAT6 fusion.

Herein, we present the clinico-radiologic, histopathologic, immunophenotypic and molecular characteristics of a diagnostically challenging case of primary intraosseous SFT of the mandible, together with a comprehensive review of the pertinent literature.

Materials and Methods

Surgical/Pathology Workup and Immunohistochemistry (IHC)

Multiple, tannish-brown soft tissue fragments measuring 19 × 8 × 4 mm in aggregate were submitted for histopathologic examination. Four µm FFPE tissue sections were immunohistochemically stained using the following antibodies: STAT6, CD99, CD163, SMA, CD34, SATB2, pancytokeratin AE1/AE3, S100, MDM2 and Ki-67. Information regarding the clone, source and provider of the above antibodies, as well as primary antibody dilution, incubation period and antigen retrieval method were tabulated (Table 1). Tissue staining was performed using the Bond-Max fully-automated IHC staining platform (Leica Biosystems), along with appropriate positive and negative controls.

Table 1.

The list of primary antibodies utilized for immunohistochemical analysis of the current case of primary intraosseous SFT of the mandible

Antibody	Clone	Source	Company	Antigen Retrieval	Primary Ab Dilution	Incubation period
SATB2	EP281	Rabbit, monoclonal	Cell Marque	ER1, Bond’s Low pH retrieval	1:500	15 min
STAT6	YE361	Rabbit monoclonal	Abcam	ER2, Bond’s High pH retrieval	1:200	15 min.
CD99	O13	Mouse monoclonal	Roche	CC1, Ultra’s High pH retrieval	RTU	40 min.
CD34	QBEnd/10	Mouse monoclonal	Cell Marque	ER1, Bond’s Low pH retrieval	1:400	30 min.
CD163	10D6	Mouse monoclonal	Leica/Novocastra	ER1, Bond’s Low pH retrieval	1:900	30 min.
SMA	1A4	Mouse monoclonal	Agilent Dako	None	1:600	30 min.
S100	4C4.9	Mouse monoclonal	Cell Marque	ER1, Bond’s Low pH retrieval	1:400	15 min.
AE1/AE3	AE1 & AE3	Cocktail	Sakura	High	RTU	21 min.
MDM2	IF2	Mouse monoclonal	Cell Marque	ER1, Bond’s Low pH retrieval	RTU	15 min.
Ki-67	MIB-1	Mouse monoclonal	Agilent Dako	ER2, Bond’s High pH retrieval	1:400	15 min.

SFT, solitary fibrous tumor; RTU, ready to use

Genomic Evaluation

RNA was extracted from macrodissected FFPE sections of tumor using the PureLink FFPE Total RNA Isolation Kit (Thermo Fisher, Waltham, MA). A190-gene fusion RNA panel was performed by a custom probe-bait next-generation sequencing (NGS) assay, with library preparation using KAPA Stranded RNA-Seq Kit with RiboErase and KAPA unique dual-indexed adapter kits (Roche, Indianapolis, IN) and sequencing in the NextSeq platform (Illumina, San Diego, CA), as previously described [15]. Analysis was performed using Arriba software. Solid tumor DNA NGS panel was performed on FFPE-extracted genomic DNA (EZ1 platform, Qiagen, Chatsworth, CA) using a 92-gene custom AmpliSeq panel on the Ion Torrent S5 platform (Life Technologies, Carlsbad, CA) with analysis by Gene Studio/Ion tools and GenomOncology Workbench (Cleveland, OH). Fluorescence in situ hybridization (FISH), using a chr12q15/MDM2 locus-specific probe (Vysis LSI MDM2 SpectrumOrange probe and a chromosome-specific alpha-satellite CEP12 SpectrumGreen probe (D12Z3; Abbott Molecular) was negative for MDM2 copy number amplification.

Literature Review

Publicly available electronic databases, including PubMed, Medline and Google Scholar, were searched during the period 2000–2024 for previously reported cases of primary or metastatic, intraosseous SFT involving the jawbones using the following combination of keywords: “hemangiopericytoma” or “solitary fibrous tumor” and “jaws”, “jawbone”, “mandible” or “maxilla”. Inclusion criteria comprised case reports and case series published in the English-written literature with adequate documentation of the histopathologic features of the lesions. SFT cases originally reported as intraosseous but, upon careful review, determined to derive from the soft tissues were excluded from the literature review. Five case reports [9–11, 13, 14] and one case series [12] satisfying the above criteria were identified to a total of N = 14 jawbone SFTs. Patient age and sex, size, location and clinico-radiographic characteristics of the lesion, as well as treatment and follow-up information was retrieved and tabulated. When available, positive IHC stains were also recorded.

Case Presentation

Clinical and Radiographic Features

A 54-year-old woman presented for evaluation of an asymptomatic, partially demarcated, multilocular radiolucency of the left posterior body of the mandible causing resorption of the root of the 2nd left mandibular premolar (tooth #20) together with loss of cortices of the left inferior alveolar canal (Fig. 1C). Radiographically, the lesion lacked marked peripheral cortication and demonstrated irregular medial and inferior borders (Fig. 1C). No paresthesia, facial asymmetry, swelling, or temporomandibular joint dysfunction was noted. Intraorally, slight expansion of the left buccal mandibular cortex was observed along with loss of the lingual cortex. Upon review of previous radiographs, the lesion was detectable approximately 6 years prior as a small, circumscribed, unilocular radiolucency of the left posterior mandible (Fig. 1A) with notable enlargement over the subsequent 6-year period (Fig. 1B and C). A computed tomography scan with contrast revealed an expansile, bone destructive, intraosseous lesion causing thinning and perforation of the left lingual cortex of the mandible (Fig. 2A and B). Due to spatial approximation of the lesion to the inferior alveolar nerve canal, the overall gradual progression, and its partially circumscribed borders, a provisional diagnosis of benign peripheral nerve sheath tumor or vascular malformation was favored. An incisional biopsy was performed 6 months after initial evaluation.

Fig. 1.

Radiographic characteristics and clinical progression of primary intraosseous solitary fibrous tumor of the mandible. (A) Panoramic radiograph at initial presentation showing a small in-size unilocular radiolucency of the left posterior mandible; (B) Progression of the lesion to a larger, well-defined, bone-destructive, multilocular radiolucency; (C) Partially demarcated, multilocular radiolucency of the left posterior body of the mandible exhibiting rugged, irregular, medial and inferior borders, and resorption of the root of the 2nd left mandibular premolar

Fig. 2.

(A) Axial view of CT scan with contrast revealing an expansile, bone destructive central lesion involving the left posterior mandible, causing thinning of the cortices; (B) Coronal view of CT scan showing perforation of the left lingual mandibular cortex

Histopathologic, Immunophenotypic and Molecular Findings

Histopathologic sections revealed a well-vascularized (Fig. 3A) proliferation of predominantly ovoid and spindle-shaped cells with relatively indistinct cell membrane borders and lightly eosinophilic cytoplasm (Fig. 3B and C). The ovoid-to-spindled lesional cells featured elongated, plump or tapered, hyperchromatic nuclei with finely granular or coarse chromatin (Fig. 3C and E). A secondary population of epithelioid cells exhibiting rich, pale eosinophilic and often vacuolated or optically clear cytoplasm with distinct cytoplasmic borders was also identified (Fig. 3D and F). Nuclei of the epithelioid cells varied from rounded to ovoid to vaguely elongated (“banana-shaped”), and exhibited vesicular or coarse chromatin and smooth nuclear contours (Fig. 3F). Prominent cytologic atypia, nuclear pleomorphism, mitotic figures and geographic necrosis were absent. The surrounding stroma comprised dense fibrous connective tissue with areas of marked collagen hyalinization (Fig. 3B, C and E) and supported variably-sized, irregularly-shaped, thin-walled vascular channels with frequent perivascular hyalinization (Fig. 3A). Mild chronic inflammatory cell infiltration including rare eosinophils was also seen.

Fig. 3.

Histopathologic characteristics of primary intraosseous solitary fibrous tumor of the mandible. (A) Low-power photomicrograph showing a proliferation of ovoid and spindle cells surrounding numerous, thin-walled, arborizing vascular channels with perivascular hyalinization; (B) Low-power photomicrograph highlighting marked stromal hyalinization; (C) and (E) Medium- and high-power photomicrographs depicting ovoid and spindle-shaped, lesional cells with relatively indistinct cell membrane borders and lightly eosinophilic cytoplasm featuring elongated, plump or tapered, hyperchromatic nuclei; (D) and (F) Medium- and high-power photomicrographs showing a secondary population of epithelioid cells with rich, pale eosinophilic and often vacuolated or optically clear cytoplasm, and rounded to ovoid to vaguely elongated nuclei with vesicular or coarse chromatin

Immunohistochemically, lesional cells showed strong, uniform, nuclear expression of STAT6 (Fig. 4A) and membranous CD99. Focal, weak-to-moderate, nuclear MDM2 immunostaining was also noted (Fig. 4B), while SATB2 staining was overall patchy and weak in tumor cells (Fig. 4C). Interestingly, strong nuclear SATB2 expression was focally observed at the tumor periphery, often in the vicinity of staghorn-like vessels (Fig. 4C inset). CD34 highlighted the prominent stromal vasculature but was negative in tumor cells (Fig. 4D), whereas CD163 decorated the cytoplasm of most interspersed, vacuolated, epithelioid cells confirming their histiocytic lineage (Fig. 4E). Proliferation marker Ki-67 was expressed in less than 5% of the cell population (Fig. 4F). IHC probes against pancytokeratin AE1/AE3, S100 and SMA were invariably negative in lesional cells.

Fig. 4.

Immunophenotypic characteristics of primary mandibular solitary fibrous tumor. (A) Lesional cells show strong and diffuse, nuclear staining for STAT6; (B) Focal, weak-to-moderate, nuclear MDM2 immunostaining; (C) Patchy and weak SATB2 immunoreactivity with focally strong SATB2 expression at the tumor periphery (inset); (D) CD34 highlighted only the prominent stromal vasculature but not the lesional cells; (E) CD163 decorated the cytoplasm of most epithelioid, vacuolated cells confirming their histiocytic lineage; (F) Ki-67 proliferation marker is expressed in < 5% of lesional cells

RNA sequencing detected the most common NAB2::STAT6 fusion seen in SFT, with exon 6 of NAB2 fused to exon 16 of STAT6 [16], with abundant support (798 fusion reads) (Fig. 5). There were no DDIT3,FGFR1/2/3,FUS,MDM2 or NCOA2 fusions seen, as relevant negatives. DNA NGS panel detected no pathogenic variants, including no CTNNB1,DICER1,H3-1 A,H3C2,NF1,NF2,TERT or TP53 mutations as the most relevant negatives. Additionally, no MDM2 amplification was detected by FISH.

Fig. 5.

NAB2::STAT6 fusion identified by RNA sequencing. The predominant gene transcript identified shows exon 6 of NAB2 (ENST00000300131.3/NM_005967.4) fused to exon 16 of STAT6 (ENST00000556155.1/NM_003153.5). Visualization performed using Arriba software

Treatment and Follow-Up

As the patient had no previous history of SFT or hemangiopericytoma and no additional lesions were identified by CT imaging of the head and neck, chest, abdomen and pelvis, a diagnosis of primary mandibular SFT was rendered. Using the modified Demicco risk assessment model [4] and given the clinical, i.e., patient’s age (< 55 years) and tumor size (2.4 cm), and histopathologic findings, including absence of mitotic activity and necrosis, the current SFT was classified as low-risk. The patient underwent left segmental mandibulectomy with limited level IB lymph node dissection, followed by reconstruction of the mandible. The examined lymph nodes showed unremarkable histopathologic findings. No recurrence has been reported within a brief 2-month follow-up period.

Discussion

We report the first fully characterized example of primary intraosseous SFT of the mandible with NAB2::STAT6 fusion. Although SFT may develop in any anatomic site, intraosseous occurrence is considered infrequent [17] with involvement of the jawbones being exceptionally rare. The exact prevalence of bona fide intraosseous jawbone SFT, however, is unclear since the vast majority of cases have been diagnosed under the obsolete term “hemangiopericytoma”, which has been loosely utilized in the literature to include other spindle cell lesions of presumed pericytic differentiation with a prominent arborizing vasculature. Even with this uncertainty, only 14 previously-reported examples of intraosseous jawbone SFT/hemangiopericytoma were identified in the English literature as summarized in Table 2 [9–14]; 9 (64.3%) affected men and 5 (35.7%) women (M: F ratio = 1.8:1) with a mean age of 39.6 years (age range = 23–54 years). A strong predilection for the posterior portion of the mandible/mandibular ramus was noted (11 of 14; 78.6%) with only 3 cases (21.4%) occurring in the maxilla. Nine cases (64.3%) were primary and the remaining (5 of 14; 35.7%) recurrent or metastatic. Reported clinical findings included paresthesia, pain, tooth mobility, dysphagia, and even progressive ipsilateral hearing loss. Radiographically, all intraosseous SFTs/hemangiopericytomas presented as expansile, multilocular, bone destructive radiolucencies of markedly variable duration (mean = 9 months; range = 2–36 months) and size (mean = 6.5 cm; range = 3.0–15.0 cm), causing cortical bone perforation and occasional tooth resorption. All cases were treated surgically with hemimandibulectomy, with 4 individuals also receiving adjuvant chemo- or radiation therapy [12]. When outcome information was available, 10 individuals remained with no evidence of disease during a mean follow-up period of 44 months (range = 10–101 months), whereas 2 died of disease after developing locoregional and/or distant metastases 42 and 101 months after diagnosis, respectively [12].

Table 2.

Clinico-epidemiologic, histopathologic and immunophenotypic characteristics of previously reported cases of hemangiopericytoma/SFT involving the jawbones

Author (Year)/ Number of cases	Age (years)/ Sex	Location/ Size (cm)	Clinico-radiographic presentation	Positive IHC markers	Diagnosis/ Grading or Risk Stratification	Treatment	Follow-up
Guerrissi et al. [6] N = 1	37/M	R posterior mandible, ramus 15.0 × 10.0	Expansile, bone destructive mass of 3-years duration; progressive ipsilateral deafness	CD34	Hemangiopericytoma, primary	Surgery; hemimandibulectomy	24 months; NED
Bhutia and Roychoudhury [10] N = 1	26/F	R posterior mandible, ramus 3.0 × 2.0 × 1.5	Expansile, multilocular RL of 1-year duration with cortical perforation; tooth mobility	NA	Hemangiopericytoma, primary	Surgery; hemimandibulectomy	48 months; NED
Thiele et al. [11] N = 1	41/F	R mandibular ramus 2.8 × 3.4 × 3.6	Progressive, painless, expansile, bone destructive mass of 3 months duration; dysphagia	CD34, vimentin, CD99, MIB-1 (low)	Hemangiopericytoma, primary; potentially malignant	Surgery; hemimandibulectomy	10 months; NED
Wushou et al. [12] N = 9	Median = 38 (range: 23–51) M: F = 8:1	Mandible (n = 6); maxilla (n = 3) Median = 6.5 (range = 3.7–12.0)	Expansile, bone destructive lesions with cortical perforation and soft tissue involvement (n = 5); pain (n = 3); paresthesia (n = 4) Mean duration = 7 months (range = 2–12)	NA	Hemangiopericytoma, primary (n = 5); recurrent (n = 4) High-grade (n = 7); Intermediate-grade (n = 1); Low-grade (n = 1)	Surgery (n = 9); adjuvant radiotherapy (n = 2) or chemotherapy (n = 2)	Median = 49 months (range = 10–101); NED (n = 7); DOD (n = 2)
Mishra et al. [13] N = 1	54/F	L posterior mandible, ramus 4.3 × 2.3	Well-defined, expansile, multilocular RL of 2 months duration with cortical perforation; tooth mobility, root resorption	CD34*, SMA, MIB-1 (low)	Hemangiopericytoma/ SFT, primary	Surgery; hemimandibulectomy	NA
Dudde et al. [14] N = 1	53/F	L posterior mandible NA	Poorly-defined RL with displacement of the IAN and cortical thinning; paresthesia	NA	SFT, metastatic (FOM primary); high-grade ƒ	Surgery; mandibular continuity resection	NA

*In this case report, CD34 immunostain was reported as positive in tumor cells. However, the provided photomicrograph shows convincing CD34 staining only in vascular endothelial cells.

^ƒIn this case, IHC and molecular studies performed in the primary tumor of the floor of mouth showed nuclear expression of STAT6 and an underlying NAB2::STAT6 fusion. However, IHC or molecular confirmation of the diagnosis of metastatic SFT was not provided for the intraosseous mandibular lesion.

SFT, solitary fibrous tumor; IHC, immunohistochemical; RL, radiolucency; IAN, inferior alveolar nerve; FOM, floor of mouth; NED, no evidence of disease; DOD, dead of disease; NA, not available

Histopathologic grading or biologic risk assessment information was provided in 11 previous cases of intraosseous jawbone SFT/hemangiopericytoma with 8 of them classified as high-grade and 1 each as intermediate- or low-grade, and “potentially malignant”. In the vast majority, diagnosis was based solely on light microscopic findings with ancillary immunostains performed in 6 of 14 cases and detailed description of the respective IHC results in just 3 cases [9, 11, 13]. Diffuse CD34 positivity was reported in all 3 cases studied [9, 11, 13]. However, in one of the studies careful evaluation of the provided photomicrographs shows unequivocally positive CD34 staining in vascular endothelial cells, while tumor cells appear uniformly negative [13]. Of note, confirmation of the SFT diagnosis either by means of STAT6 IHC or identification of the NAB2::STAT6 fusion is not documented in any previous intraosseous SFT of the jaws. Recently, a case of metastatic mandibular SFT presumed to originate from a primary lesion involving the floor of mouth was reported [14]. Although molecular confirmation of the diagnosis of SFT was provided for the primary floor-of-mouth tumor, no IHC or cytogenetic work-up was performed for the mandibular lesion and no illustrations of its histopathologic features were provided [14]. Furthermore, jawbone metastasis from a SFT of the floor of mouth, although not impossible, would be considered extremely unlikely.

Depending on the clinical setting, when the soft tissues of the head and neck are affected, the histopathologic differential diagnosis of SFT is remarkably broad and may encompass various benign, e.g., myofibroma, cellular schwannoma, fibrous histiocytoma, cellular angiofibroma, nodular fasciitis, and spindle cell lipoma, as well as malignant neoplasms including low-grade fibromyxoid sarcoma, monophasic synovial sarcoma, and even sarcomatoid squamous cell carcinoma and spindle cell melanoma [1, 18]. However, in the event of mandibular involvement, as seen in the current case, additional diagnostic considerations such as mesenchymal chondrosarcoma, low-grade central osteosarcoma, Ewing sarcoma, phosphaturic mesenchymal tumor and undifferentiated small round cell sarcomas, for example, CIC-rearranged sarcoma and sarcoma with BCOR aberrations, should be ruled out.

Mesenchymal chondrosarcoma is characterized by a primitive-appearing, cellular, round-to-ovoid cell proliferation in association with branching vascular channels exhibiting a hemangiopericytoma-like architecture and variable amounts of hyaline cartilage. Of note, the cartilaginous component may be entirely absent within a submitted specimen [1]. The latter may pose significant diagnostic challenges especially in small sample biopsies, similar to the current case. Although, both mesenchymal chondrosarcoma and intraosseous SFT may show strong membranous CD99 positivity, the majority of mesenchymal chondrosarcomas also demonstrate nuclear NKX2.2, or alternatively NKX3.1, and SOX9 staining, together with HEY1::NCOA2 rearrangements [18–20]. The foci of stromal collagen hyalinization in this mandibular SFT were highly reminiscent of early osteoid and in conjunction with the relatively bland appearance of the spindle cell population, raised suspicion for a low-grade central osteosarcoma. Low-grade central osteosarcoma accounts for merely 1–2% of all osteosarcomas and is characterized by variably cellular fascicles of spindle cells with subtle nuclear atypia and scarce mitotic activity, embedded in a fibrosclerotic stroma admixed with interlacing, woven bone trabeculae [1]. Most cases of low-grade central osteosarcoma of the long bones demonstrate diffuse, moderate-to-strong, nuclear immunostaining for MDM2 and CDK4, mirroring the genetic amplification of 12q13-q15 involving MDM2 and CDK4 [21–25]. Notably, MDM2 and CDK4 IHC staining pattern and intensity may vary substantially in jawbone osteosarcomas with only focal and weak-to-moderate expression, akin to this mandibular SFT, in up to 75% and 62.5% of cases, respectively [26]. Furthermore, SATB2, a commonly utilized IHC marker in daily practice for confirmation of osteogenic lineage, is strongly and diffusely expressed in the vast majority (> 90%) of osteosarcomas [1, 27, 28], irrespective of the amount of malignant osteoid and histopathologic subtype [27–29]. Aberrant SATB2 expression in head and neck SFT has been previously reported [30]. Interestingly, the current mandibular SFT showed predominantly patchy and weak SATB2 staining together with areas of strong positivity at the periphery. Underscoring its lack of specificity and need for cautious interpretation, variable SATB2 staining has been documented in a broad spectrum of other non-osteosarcomatous spindle cell neoplasms [31–33].

Albeit exceptionally infrequent, jawbone occurrence of phosphaturic mesenchymal tumors has been reported with such lesions clinically inducing systemic osteomalacia via secretion of FGF23 and severe hypophosphatemia due to phosphate waste [34–36]. Histologically, phosphaturic mesenchymal tumors are characterized by a variably cellular population of cytologically bland, spindle to stellate cells with occasional osteoclast-type multinucleated giant cells, in association with a hyalinized or lightly basophilic, calcified matrix showing granular, i.e., grungy, or flocculent appearance [1, 34–36]. A prominent hemangiopericytomatous vascular network is also commonly encountered, thus bearing overt morphologic similarities to the current mandibular SFT. Although in our SFT case only areas of stromal hyalinization were observed without any evidence of dystrophic calcifications, such a feature may be entirely absent from a fraction of phosphaturic mesenchymal tumors or not readily identifiable, particularly in incisional biopsies, due to sampling error. In addition to clinical findings, IHC studies may be diagnostically useful since, in contrast to SFT, phosphaturic mesenchymal tumors are invariably negative for STAT6 and show consistent expression of CD56, ERG, SATB2 and SSTR2A [33, 37]. FGF23 mRNA and protein expression levels are also upregulated, while genetic analysis reveals FN1::FGFR1 and FN1::FGF1 fusions in ~ 60–70% of cases [37–39]. Although interspersed inflammatory cells may be present in SFT, a prominent histiocytic reaction such as seen in the current intraosseous lesion is not typically anticipated and may imitate fibrous histiocytoma. Both SFT and fibrous histiocytoma exhibit overlapping histopathologic characteristics that include cell morphology, i.e., uniformly monomorphic spindled cells with plump, ovoid to elongated vesicular nuclei and indistinct, palely eosinophilic cytoplasm, presence of thin-walled, branching, hemangiopericytoma-like vasculature, and stromal hyalinization [1, 40]. A storiform architecture together with the presence of foamy histiocytes and/or multinucleated giant cells, features commonly utilized to distinguish fibrous histiocytoma may be subtle or sparse. Adding to the level of diagnostic confusion, CD34 positivity may be observed in up to 40% of fibrous histiocytomas [41], while weak, nuclear STAT6 staining has also been rarely reported [7].

Nuclear STAT6 expression is a highly sensitive and almost perfectly specific IHC marker for the diagnosis of SFT and can be helpful in discerning SFT from its histologic mimics in daily pathology practice [7, 42]. Indeed, strong and diffuse STAT6 immunoreactivity is observed in almost the entirety of SFTs irrespective of their histopathologic attributes, anatomic site, including jawbone lesions as shown here, and CD34 expressivity status [7, 42]. In SFT, aberrant STAT6 expression reflects the pathognomonic for this tumor NAB2::STAT6 fusion [7]. Positive STAT6 immunostaining, however, is not limited to SFT and has also been documented to a lesser extent in other mesenchymal neoplasms, e.g., dedifferentiated liposarcoma [43, 44] and GLI1-amplified soft tissue tumors [45]. In these lesions, unlike SFT, STAT6 protein overexpression is not the result of a gene fusion product but rather the outcome of co-amplification of the STAT6 encoding gene located at the 12q13.3-q14.1 locus when amplification of the 12q13-15 chromosomal region occurs [7, 43, 44]. Together with STAT6, MDM2, CDK4, and GLI1 located at the chromosomal loci 12q15, 12q14.1 and 12q13.3, respectively, are also amplified, thus explaining why concomitant IHC positivity for MDM2, CDK4 and STAT6 is a frequent phenomenon in this group of tumors [7, 43–45]. Notably, STAT6 staining pattern in dedifferentiated liposarcoma and GLI1-amplified soft tissue tumors may vary from patchy and weak, to strong and diffuse, with both nuclear and cytoplasmic localization [7, 44, 45]. The latter markedly differs from the uniformly strong and exclusively nuclear STAT6 immunophenotype characterizing SFTs. Recently, aberrant nuclear STAT6 expression was reported in a single case of mesenchymal chondrosarcoma of the spine shown to harbor the HEY1::NCOA2 fusion [46]. In addition to STAT6, CD34 and CD99 positivity is encountered in 90% and 70% of SFTs, respectively, while BCL2 staining is also seen in virtually all cases, despite lack of specificity [1, 7, 8, 47, 48].

In conclusion, we present a fully-documented case and literature review of central jawbone SFTs. There tumors are rare and such a diagnosis may prove particularly challenging. Strong and diffuse nuclear STAT6 IHC staining is diagnostically useful, particularly in the event of CD34 negativity, as shown here. Molecular confirmation of a typical NAB2::STAT6 fusion is recommended for definitive diagnosis of central jawbone presentations, since other histologic mimics may also demonstrate variable STAT6 reactivity. Intraosseous SFTs of the jawbones may show locally aggressive biologic behavior with cortical bone perforation and tooth root resorption. Complete surgical excision remains standard of care with a guarded overall prognosis.

Author Contributions

Material preparation and data collection was performed by all authors. The first draft of the manuscript was written by PPA and KLW, while accompanying illustrations were prepared by PPA and DMJ. All authors have reviewed, edited and approved the final version of the submitted manuscript.

Funding

Not applicable.

Data Availability

No datasets were generated or analysed during the current study.

Code availability:

Not applicable.

Declarations

Ethical Approval

This case report is exempt from IRB approval.

Consent to Participate

Not applicable.

Consent for Publication

Not applicable.

Competing Interests

The authors declare no competing interests.