Solitary fibrous tumors arising from the bone are an extremely rare event and only few cases have been previously described in the literature.
1. Introduction
Solitary fibrous tumor (SFT) is a rare mesenchymal tumor of fibroblastic origin that can occur at any anatomic site and typically affects middle-aged adults
[1][2][3]. It is characterized by a strong morphologic heterogeneity with a wide spectrum of biologic features. The histological and molecular diagnostic criteria used in soft tissue SFT (S-SFT) have been recently applied on “non otherwise classified” primary bone tumors, drawing out a new category of SFT of the bone (B-SFT)
[4][5][6][7]. Nevertheless, B-SFT is exceedingly rare, with only few cases are described in the literature
[8][9][10][11], and its biological behavior has not yet been assessed. From a histopathological and molecular point of view, primary B-SFT shares the same features of S-SFT. It is characterized by a prominent, branched vascularization, with a thin and dilated vascular texture defined as “staghorn” and by the presence of the
NAB2-STAT6 gene rearrangement (NGFI-A binding protein 2—Signal Transducer and Activator of Transcription 6), present in about 90% of cases and considered a pathognomonic feature
[1][12]. Positivity to CD34 stain is distinctive in 90–95% of the cases. S-SFT has an intermediate malignant potential with a low risk of metastasis. Some studies have investigated the prognostic role of previously described molecular markers, without, however, obtaining conclusive results; the aforementioned prognostic criteria have never been explored in B-SFT
[13][14][15][16]. Most S-SFTs are clinically indolent, with an intermediate malignant potential and a low risk of metastasis, showing an overall 5- and 10-year distant metastasis (DM)-free rates of 74% and 55%, respectively. In recent times, different stratification risk models have been proposed
[17][18][19][20][21][22]. The current most utilized scoring system to discriminate different risk groups for S-SFT—also related to the development of distal metastasis—is the one proposed by Demicco et al.
[22], which considers patient age, mitotic activity, tumor necrosis, and size. To date, few prognostic molecular markers have been analyzed.
NAB2–STAT6 chimeric transcripts, with a frequency ranging from 55 to 100%
[23][24], and characterized by different breakpoints in fusion genes, might contribute to the morphologic diversity of SFT; some studies evidenced associations between specific fusion variants and different clinical features
[21][25]. In addition, specific point mutations within the promoter region of telomerase reverse transcriptase (
TERT)—C228Tand C250T—have been recently reported in S-SFT subsets and other tumors
[15][25][26][27]. These mutations confer enhanced
TERT promoter activity and have been suggested as predictive factors to identify high-risk patients. Finally, TP53 has also been proposed as an SFTs risk factor. In particular, tumors with TP53 mutations were almost always classified as high risk
[21][28]. Due to the rarity of B-SFT and taking advantage of the availability of a large and homogeneous cohort of patients, the goal of this study was to better characterize the biological behavior of this specific SFT subset located in the bone considering both the clinical, histological, and molecular features, as well as the applicability of the risk stratification model used for S-SFT.
Histopathological and Immunohistochemical Features
2. Histopathological and Immunohistochemical Features
From a histopathological point of view, 15 cases showed more than 4 mitotic figures per 10 HPF and were associated with high cellularity, cytologic atypia, and >10% of necrosis, defining high-grade tumors (, ). CD34 and STAT6 immunopositivity was observed in 95% (23/24) and in 100% (24/24) of cases, respectively ().
Figure 1. Solitary fibrous tumor: A spindle cell proliferation showing hemangiopericytoma-like blood vessels is seen (Hematoxylin &Eosin, original magnification, ×100).
Figure 2. The nuclei of neoplastic cells express STAT6 (original magnification, ×200).
According to Demicco score
[22], 8 patients (33%) were classified in the low-risk group, 11 (46%) in the intermediate-risk group, and 5 (21%) in the high-risk group ().
Two of the nine patients who developed distant metastasis belonged to the low-risk groups, five to the intermediate-risk group, while two patients belonged to the high-risk group. The three patients with metastasis at presentation were equally distributed in the three risk groups.
3.3. NAB2–STAT6 Fusion Variants
3. NAB2–STAT6 Fusion Variants
The analysis of fusion transcripts identified
NAB2–STAT6 fusion variants in 10 out of 12 (83.3%) samples (). In two cases, no variant was found. Considering the 24 types of fusion variants evaluated, 2 breakpoints were detected with a higher frequency:
NAB2exon6—
STAT6exon17 (4 cases) and
NAB2exon4-STAT6exon2 (3 cases), followed by the breakpoint
NAB2exon6—
STAT6exon16,
NAB2exon2—
STAT6exon2 and
NAB2exon6-STAT6exon16/NAB2exon6—
STAT6exon17 in one case (). Regarding the Demicco score risk, the
NABex6-STAT6ex17 fusion variant was present only in high- and intermediate-risk patients, even if
NAB2-STAT6 fusion variants and Demicco score risk were not significantly correlated (
p = 0.25).
3.4. TERT Promoter Mutations: C228T and C250T
4. TERT Promoter Mutations: C228T and C250T
The wild-type C250C genotype was shown in all 16 samples while no C250T mutations were detected. In only one DNA sample, a heterozygous C228T substitution was detected.
The only patient presenting this variant died one day after surgery due to complications; therefore, it was not possible to evaluate its prognostic role ().
3.5. p53 Mutations
5. p53 Mutations
Overall, we detected p53 genetic alterations in 11 samples (). Three samples presented point mutations: a nonsense heterozygous variant (p.Gln165*) was detected in patient 1; a missense heterozygous variants (p.Ala63Val), already described as a variant of uncertain significance (VUS), was detected in patient 16; and a homozygous splice site alteration (c.375 + 1G > A) was observed in patient 11. All samples except two (1 and 14) showed the presence of a copy number variation (CNV) involving at least exon 4 of
p53. In detail, CNV deletions were detected in patient 2, 7, 13, 15, 18, 19, and 23 whereas CNV amplifications were detected in patient 16 and 21.
3.6. Correlations between Clinicopathological, Immunohistochemical, and Molecular Data
6. Correlations between Clinicopathological, Immunohistochemical, and Molecular Data
Regarding the entire population of study (24 cases), 5- and 10-year DSS were respectively 64% and 42%, whereas on the localized tumor, 5- and 10-year disease-related-specific DSS were respectively 80% and 60%. As expected, localized and surgically treated patients (16 out of 24, 66%) showed a better 5-year DSS than metastatic ones (74% vs. 33%) ().
Figure 3. Cumulative survival of 24 patients affected by primary bone solitary fibrous tumor; stratification by metastases at presentation.
summarizes the results of the Kaplan–Meier survival analysis of the clinicopathological variables (histological grade, tumor size, age, mitosis, necrosis, Demicco score risk). Stratification by tumor size did not correlate with DSS either for localized patients (
p = 0.54) or for the whole series (
p = 0.44). However, the only patient with tumor size <5 cm was alive at follow up (). Stratification based on mitotic count was carried out (A ≤ 1 mitosis, B = 1–3 mitosis, and C ≥ 4); no correlation was found in terms of DSS at the 5- and 10-year follow up either for the whole series (
p = 0.54) or for patients with localized disease (
p = 0.33) ().
Table 3.
Disease-specific survival (DSS) analysis related to clinicopathological parameters.
Variables |
Disease Specific Survival (24 pts) |
Localized Disease * (16 pts) |
5 Years-DSS |
10 Years-DSS |
p-Value |
5 Years-DSS |
10 Years-DSS |
p-Value |
Histological Grade |
|
|
|
|
|
|
|
Low |
62% |
31% |
0.52 |
100% |
67% |
0.84 |
|
High |
65% |
58% |
|
82% |
71% |
|
Size |
|
|
|
|
|
|
|
(A) 0–4.99 cm |
100% |
100% |
0.44 |
100% |
100% |
0.54 |
|
(B) 5–9.99 cm |
70% |
36% |
|
87% |
45% |
|
|
(C)10–14.99 cm |
62% |
62% |
|
80% |
80% |
|
|
(D) >15 |
50% |
50% |
|
50% |
50% |
|
Age |
|
|
|
|
|
|
|
<55 years |
86% |
27% |
0.06 |
100% |
77% |
0.15 |
|
≥55 years |
61% |
27% |
|
100% |
60% |
|
Mitosis |
|
|
|
|
|
|
|
(A) <1 |
60% |
30% |
0.54 |
60% |
30% |
0.33 |
|
(B) 1–3 |
66% |
33% |
|
100% |
50% |
|
|
(C) ≥4 |
65% |
58% |
|
76% |
68% |
|
Necrosis |
|
|
|
|
|
|
|
<10% |
80% |
47% |
0.66 |
100% |
62% |
0.95 |
|
≥10% |
51% |
51% |
|
78% |
78% |
|
Gene Fusion |
|
|
|
|
|
|
|
Exon6 |
|
|
|
80% |
40% |
0.68 |
|
Other |
|
|
|
100% |
67% |
|
Demicco Score Risk |
|
|
|
|
|
|
|
High |
|
|
|
54% |
54% |
0.43 |
|
Intermediate |
|
|
|
72% |
46% |
|
|
Low |
|
|
|
64% |
28% |
|