Solitary Fibrous Tumour

Risk stratification models are preferred over anatomical staging.

The genetic hallmark of SFT is a paracentric inversion involving chromosome 12q, resulting in the fusion of the NAB2 and STAT6 genes. Some studies have shown a correlation between fusion types and histological features. The NAB2-STAT6 fusion is thought to convert wildtype NAB2 from a transcriptional repressor of EGR1-mediated signaling into a transcriptional activator via replacement of the C-terminal repression domain by the transcriptional activation domain of STAT6, thereby resulting in a feedforward loop of constitutive EGR1-mediated transactivation of proliferation and survival-associated growth factors, including IGF2 and FGFR1. Overexpression of ALDH1A1 (ALDH1), EGFR, JAK2, histone deacetylases, and retinoic acid receptor may also contribute to tumorigenesis. Other alterations associated with aggressive behavior and dedifferentiation include TERT promoter mutations and deletions or mutations of TP53.

SFTs are well-circumscribed masses that typically measure 5-10 cm, although some lesions may exceed 25 cm in greatest dimension. The cut surface is nodular and tan to reddish-brown, and it occasionally shows hemorrhage, myxoid change, or cystic degeneration.

SFTs are composed of haphazardly arranged spindled to ovoid cells with indistinct, pale eosinophilic cytoplasm within a variably collagenous stroma, admixed with branching and hyalinized staghorn-shaped (haemangiopericytomatous) blood vessels. There is a wide histological spectrum, ranging from paucicellular lesions with abundant stromal keloidal-type collagen to highly cellular tumors consisting of closely spaced cells with little or no intervening stroma. Myxoid change may be present. SFTs most often have low mitotic counts, without substantial nuclear pleomorphism or necrosis. tumors demonstrating a high mitotic count with or without increased cellularity, atypia, necrosis, and infiltrative growth have traditionally been termed malignant, but new risk stratification models more accurately predict prognosis. 

Fat-forming (lipomatous) SFT harbors a component of mature adipose tissue. Giant cell–rich SFT, formerly known as giant cell angiofibroma, shows features of conventional SFT with an admixed population of multinucleated giant cells within the stroma and lining pseudovascular spaces. Dedifferentiated (anaplastic) SFTs show transition to high-grade sarcoma with or without heterologous elements such as rhabdomyosarcoma or osteosarcoma. By immunohistochemistry, SFT typically shows strong and diffuse expression of CD34 and nuclear STAT6, but expression may be lost in dedifferentiated SFT.

Cytological examination (studying cells under a microscope) reveals oval, elongated, or rounded cells with wispy cytoplasm and eosinophilic (pink-staining) collagenous stroma.

Recurrence (distant or local) occurs in 10-30% of SFTs, with 10-40% of recurrences reported after 5 years and rare recurrences seen after 15 years. TERT promoter mutations are more common in tumors with aggressive features, but they may also be seen in low-risk SFT. Various single clinical or histological features have been reported to correlate with metastatic or local recurrence potential in large series, including high mitotic count (> 2 mitoses/mm2, equating to > 4 mitoses per 10 high-power fields of 0.5 mm in diameter and 0.2 mm2 in area), tumor size, necrosis, patient age, tumor cellularity and nuclear pleomorphism, and tumor site. However, individual studies contradict one another regarding which specific features are important.

The development of multivariate risk models has resulted in improved prognostication over the traditional benign/malignant distinction. Of these models, the one most similar to the traditional definition of malignant SFT stratifies tumors into four risk tiers based on mitotic count, pleomorphism, and tumor cellularity. A set of risk calculators proposed by the French Sarcoma Group (FSG) incorporates clinical data (patient age, tumor site), pathological features (mitotic count), and history of radiotherapy to variously predict overall survival, local recurrence, and distant metastatic risk. The most widely used model for metastatic risk incorporates mitotic count (≥ 2 mitoses/mm2), patient age (≥ 55 years), and tumor size stratified by 5 cm tiers to classify tumors into low, intermediate, and high-risk groups. This model has been validated for both thoracic and extrathoracic SFTs; a subsequent refinement includes necrosis as a fourth variable, resulting in higher numbers of cases being classified as low-risk.

NAB2-STAT6 gene fusions are pathognomonic (specific indicators) for SFT. However, because NAB2 and STAT6 are in close proximity on chromosome 12q, detection of their fusion is difficult by conventional cytogenetic methods, and the diversity of breakpoints occurring in both exons and introns makes PCR-based detection of fusion variants difficult without multiplexed sequencing assays. STAT6 immunohistochemistry is a sensitive and specific surrogate for all fusions.

Essential: spindled to ovoid cells arranged around a branching and hyalinized vasculature; variable stromal collagen deposition; CD34 and/or STAT6 expression by immunohistochemistry.

Desirable (in selected cases): demonstration of NAB2-STAT6 gene fusion.