PEComas of the bone and soft tissues are extremely rare. In the current study, we examined the clinical and pathological characteristics of PEComas of bone and soft tissues (Tables 1 and 2) and compared them with the characteristics of PEComas of various organs, including the skin, uterus, gastrointestinal tract, pancreas, kidney, liver, bone, and soft tissue (Tables 3 and 4) (Mahera et al. 1997; Kuroda et al. 2000; Diment and Colecchia 2003; Fukunaga 2004; Harris et al. 2004; Folpe et al. 2005; Mai and Belanger 2006; Blechet et al. 2007; Pikoulis et al. 2007; Osei et al. 2007; Boussouga et al. 2008; Argani et al. 2010; Varshney et al. 2011; Charli-Joseph et al. 2014; Conlon et al. 2015; Chen et al. 2016; Zhang et al. 2017; Alnajar et al. 2018; Bao et al. 2019; Harvey et al. 2019; Zhong et al. 2020; Rehman et al. 2021; Zhang et al. 2021).
Table 3
Previous reports of soft-tissue PEComa
No. | Authors | Publication year | Age | Gender | Location | Size (cm) | Melan-A | HMB45 | SMA | Desmin | CK | CD34 | S-100 | TFE3 |
1 | Mahera et al. | 1997 | 79 | F | Leg | 21 | N.A | - | + | + | NA | + | NA | NA |
2 | Kuroda et al. | 2000 | 41 | M | Thigh | 10 | N.A | + | N.A | NA | NA | NA | NA | NA |
3 | Diment & Colecchia | 2003 | 59 | F | Thigh | 10 | N.A | + | + | - | - | NA | - | NA |
4 | Fukunaga | 2004 | 44 | F | Abdominal wall | 3.5 | + | + | ++ | - | + | - | - | NA |
5 | Harris et al. | 2004 | 87 | M | Knee | 5.5 | N.A | + | + | - | - | - | - | NA |
6 | Folpe et al. | 2005 | 43 | F | Thigh | 3.5 | - | + | + | - | NA | NA | - | NA |
7 | | | 71 | M | Forearm | 9 | N.A | N.A | N.A | N.A | N.A | N.A | N.A | NA |
8 | | | 77 | F | Neck | 2.6 | N.A | N.A | N.A | N.A | N.A | N.A | N.A | NA |
9 | | | 49 | F | Shoulder | Large | N.A | N.A | N.A | N.A | N.A | N.A | N.A | NA |
10 | | | 22 | F | Abdominal wall | 8.5 | N.A | N.A | N.A | N.A | N.A | N.A | N.A | NA |
11 | | | 24 | F | Abdominal wall | 10.5 | N.A | N.A | N.A | N.A | N.A | N.A | N.A | NA |
12 | | | 19 | F | Retroperitoneum | 2.1 | N.A | N.A | N.A | N.A | N.A | N.A | N.A | NA |
13 | | | 18 | F | Retroperitoneum | 6 | N.A | N.A | N.A | N.A | N.A | N.A | N.A | NA |
14 | | | 72 | F | Retroperitoneum | 24 | N.A | N.A | N.A | N.A | N.A | N.A | N.A | NA |
15 | Mai & Belanger | 2006 | 56 | M | Thigh | 11 | N.A | ++ | N.A | ++ | + | NA | - | NA |
16 | | | 46 | M | Groin | 8 | N.A | ++ | N.A | - | - | NA | - | NA |
17 | | | 60 | F | Thigh | 12 | N.A | + | N.A | + | + | NA | - | NA |
18 | Pikoulis et al. | 2007 | 23 | F | Buttock | 8.5 | N.A | - | ++ | NA | NA | NA | NA | NA |
19 | Osei et al. | 2007 | 49 | F | Shoulder | 5.3 | N.A | + | + | NA | NA | N.A | - | NA |
20 | Blechet et al. | 2007 | 13 | M | Knee Joint | 4 | N.A | + | + | NA | NA | NA | - | NA |
21 | Boussouga et al. | 2008 | 26 | F | Shoulder | 6 | N.A | + | + | + | NA | NA | NA | NA |
22 | Argani et al. | 2010 | 33 | F | Buttock | N.A | - | + | - | - | - | - | - | ++ |
23 | | | 46 | F | Thigh | 6.5 | + | + | - | - | - | - | - | ++ |
24 | | | 71 | M | Arm | 4 | - | + | - | + | - | N.A | - | ++ |
25 | | | 26 | F | Thigh | 3 | + | N.A | - | N.A | N.A | N.A | N.A | + |
26 | | | 48 | F | Foot | N.A | + | + | + | + | N.A | N.A | N.A | + |
27 | Varshney et al. | 2011 | 12 | F | Knee joint, foot | 15 | N.A | - | + | + | NA | NA | NA | NA |
28 | Alnajar et al. | 2017 | 44 | M | Knee | 23 | - | + | + | NA | - | NA | - | - |
29 | Harvey et al. | 2019 | 44 | M | Knee | 6.5 | + | - | + | - | N.A | N.A | - | - |
30 | Rehman et al. | 2021 | 49 | F | Buttock | 10.5 | N.A | ་ | ་ | - | N.A | - | - | NA |
Summary of previous reports | 1997–2021 | 45 | F/M: 21/9 | Upper extremity: 5 Lower extremity: 15 Trunk: 10 | 8.9 | 56% (5/9) | 81% (17/21) | 78% (14/18) | 44% (7/16) | 30% (3/10) | 17% (1/6) | 0% (0/15) | 71% (5/7) |
Present study | 2023 | 49 | F/M: 6/4 | Upper extremity: 3 Lower extremity: 3 Trunk: 4 | 9.3 | 30% (3/10) | 90% (9/10) | 50% (5/10) | 30% (3/10) | 10% (1/10) | 0% (0/10) | 0% (0/10) | 89% (8/9) |
Table 4
Comparison of immunohistochemical findings among various PEComa origins.
Authors | Publication year | Primary origin (no. of patients) | Age (median) | Gender F/M | Melan-A | HMB45 | SMA | Desmin | CK | CD34 | S-100 | TFE3 | High MIB-1 index (> 10%) |
Charli-Joseph et al. | 2014 | Skin (8) | 46 | 5/3 | 0% (0/8) | 88% (7/8) | 88% (7/8) | 13% (1/8) | NA | NA | 38% (3/8) | NA | 13% (1/8) |
Conlon et al. | 2015 | Uterine corpus (78) | 47.5 | 78/0 | 46% (21/46) | 99% (71/72) | 80% (53/68) | 63% (39/62) | 5% (2/43) | 0% | NA | NA | NA |
Chen et al. | 2016 | GI-tract (26) | 38.9 | 32/18 | 65% (22/34) | 96% (44/46) | 64% (31/33) | 44% (18/44) | 0% (0/21) | 0% (0/32) | 11% (4/37) | 60% (6/10) | NA |
Zhang et al. | 2017 | Pancreas (21) | 47.9 | 17/3 | NA | 100% (20/20) | 88% (15/17) | NA | 6% (1/18) | NA | NA | NA | 0% (0/12) |
Bao et al. | 2019 | Kidney (24) and others (2) | 51 | 22/4 | 100% (26/26) | 96% (25/26) | 100% (26/26) | 85% (22/26) | 0% (0/26) | 0% (0/26) | 27% (7/26) | 0% (0/26) | NA |
Zhang et al. | 2021 | Liver (26) | 50 | 17/9 | 88% (23/26) | 96% (25/26) | 84% (22/26) | 4% (1/26) | NA | 69% (18/26) | 54% (14/26) | 0% (0/26) | 4% (1/26) |
Zhong et al. | 2020 | Bone (20) | 37 | 10/10 | 50% (10/20) | 80% (16/20) | 45% (9/20) | 15% (3/20) | NA | NA | 5% (1/20) | 4cases | NA |
Summary of previous reports* | 1997–2021 | Soft tissue (30) | 45 | 21/9 | 56% (5/9) | 81% (17/21) | 78% (14/18) | 44% (7/16) | 30% (3/10) | 17% (1/6) | 0% (0/15) | 71% (5/7) | NA |
Present study | 2023 | Bone (1) and soft tissue (9) | 49 | 6/4 | 30% (3/10) | 90% (9/10) | 50% (5/10) | 30% (3/10) | 10% (1/10) | 0% (0/10) | 0% (0/10) | 89% (8/9) | 50% (5/10) |
Abbreviations: GI-tract: gastrointestinal tract, F: female, M: male, *Summary of previous reports was obtained from Table 3. |
The present study included four males and six females with a median age of 49 years (31–83 years). Bone and soft tissue PEComas are more frequent in females than in males and usually occur in middle or old age. In terms of age and sex, PEComas of the bone and soft tissue showed a similar trend to that reported in previous studies of PEComas arising in other organs (Table 4). Bone and soft tissue PEComas occurred almost equally in the upper extremities, lower extremities, and trunk, and there was no specific site of predilection. The mean tumor diameter was 9.3 cm, and tumors were larger than 5 cm in six cases. The PEComas of the bone and soft tissue were usually large at the time of detection. These findings are comparable to those of previously reported soft tissue PEComas (Table 3).
There have only been a few detailed reports on the clinical prognosis of PEComas. Conlon et al. reviewed 78 cases of uterine corpus PEComas and reported that ten of the 63 patients (16%) died of the disease; the median survival of these ten patients was 20 months. They also reported that nine patients (14%) were alive with disease, while 44 (70%) had no evidence of disease (Conlon et al. 2015). In the present study, two patients had lung metastasis at the time of initial treatment (cases 3 and 6). One patient had postoperative distant metastasis to the lung and common iliac lymph nodes (case 4). One patient experienced local recurrence 8 months after surgery and received radiation therapy. Thereafter, the patient developed postoperative distant metastasis to the lung 4 years after the initial therapy and underwent pulmonary metastasectomy (case 2). The 1-year survival rate was 77% (Fig. 1). The prognosis of bone and soft tissue PEComas is comparable to that of the uterine corpus (Conlon et al. 2015). Although it was difficult to conclude the clinical risk of developing bone and soft tissue PEComas due to the short follow-up period of the current study, bone and soft tissue PEComas were found to have a poor prognosis, similar to that of high-grade soft tissue sarcomas.
PEComas typically show a nested architecture, comprising epithelioid cells with abundant granular eosinophilic or clear cytoplasm, round nuclei, and small nucleoli. Nests or trabeculae are typically surrounded by thin-walled capillary vessels. In contrast, a small subset of PEComas has predominantly spindle cell morphology (Bonetti et al. 2001). In the bone and soft tissue PEComas, the morphological findings included six tumors with predominantly epithelioid cells, three tumors with predominantly spindle cells, and one tumor with equal proportions of epithelioid and spindle cells. Malignant PEComas are typically characterized by large tumor size, mitoses, necrosis, and nuclear atypia. Folpe et al. developed a prognostic system based on the retrospective analysis of 26 PEComas at multiple sites and divided them into benign, uncertain malignancy potential, and malignancy groups based on histological criteria (Folpe et al. 2005). Based on this classification, malignancy was reported in 57% (12/21 cases) of PEComas of the bone (Zhong et al. 2020), 54% (42/78 cases) of uterine PEComas (Conlon et al. 2015), and 52% (26/50 cases) of gastrointestinal PEComas (Chen et al. 2016). In the study of soft tissues by Folpe et al., excluding those of uterine, intra-abdominal, and genital origin, seven cases were classified as malignant, one case had uncertain malignancy potential, and one case was benign (cases 6–14) (Folpe et al. 2005). In the present study, 80% of the bone and soft tissue PEComas were classified as malignant, while 20% were classified as having uncertain malignancy potential. None of the patients were categorized as benign. Half of the cases (5/10) showed high MIB-1 index values of > 10% (Table 2-B). In contrast, the rates of occurrence of high MIB-1 index values (> 10%) were 13% (1/8) in skin PEComa, 0% (0/12) in pancreatic PEComa, and 4% (1/26) in liver PEComa (Charli-Joseph et al. 2014; Zhang et al. 2017; Zhang et al. 2021). These results support the hypothesis that bone and soft tissue PEComas have a particularly higher cell proliferation and malignancy potential than other visceral PEComas (Hasegawa et al. 2002).
PEComa is characterized by the immunohistochemical expression of both melanocytic and muscle markers (Utpatel et al. 2020). In this study, the melanocyte marker HMB45, which is considered the most sensitive immunostaining marker, showed a high positivity rate of 90%. Melanoma also showed a high positivity rate of 89%. Meanwhile, the muscle markers SMA, MSA and desmin were positive only in 50%, 33%, and 30% cases, respectively. Muscle markers were less abundant than melanocytic markers and tended to be less abundant than those in PEComas of other primary sites (Table 4) (Charli-Joseph et al. 2014; Conlon et al. 2015; Chen et al. 2016; Zhang et al. 2017; Bao et al. 2019; Zhong et al. 2020; Rehman et al. 2021; Zhang et al. 2021). In general, epithelioid PEComas tend to show higher expression of melanocytic markers than that of myogenic markers, and spindle cell PEComas show an opposite expression profile (Conlon et al. 2015). In this study, 80% of the cases showed epithelioid or mixed morphology and only 20% showed spindle cell morphology. This may have resulted in the high expression of melanocytic markers and low expression of myogenic markers in the bone and soft tissue PEComa.
PEComas are classified into two subtypes. Conventional PEComas harbor mutations and loss of heterozygosity (LOH) in the TSC2 gene and, more rarely, the TSC1 gene, which may be associated with angiomyolipomas and PEComas. The significance of LOH in TSC1/2 is the subsequent upregulation of mTOR signaling, which is the basis of the action of mTOR inhibitors that are often utilized in PEComa treatment. In contrast, a distinct small subset of PEComas harboring rearrangements of the TFE3(Xp11) gene locus has been identified (Utpatel et al. 2020). The TFE3-rearranged PEComas harbor TFE3 gene fusions, which correlate with a strong nuclear immunoreactivity for TFE3 (Argani et al. 2010); approximately 15% cases show strong nuclear staining for TFE3 (Bonetti et al. 2001). These tumors tend to have a prominent alveolar pattern and epithelioid morphology and lack the expression of smooth muscle markers (Utpatel et al. 2020). In our study, TFE3 expression was positive in eight of the nine cases that could be observed; three of these cases showed strong (3+) nuclear staining for TFE3 (Table 2-B; cases 4, 5, and 7). In addition, tumors with high expression of TFE3 (2 or higher) (cases 2–7 and 9) were classified as PEComas with malignant potential according to Folpe’s classification.These characteristics suggest that PEComas of the bone and soft tissue are prone to TFE3 rearrangement with high malignant potential.
Radical resection is the primary treatment modality for PEComas, as PEComas are characterized by resistance to radiation and chemotherapy (Bleeker et al. 2012; Jia et al. 2020). Radical resection is associated with an increased disease-free survival (Sobiborowicz et al 2021). Recently, mTOR inhibitors such as sirolimus have been shown to be effective for inoperable, recurrent, or advanced PEComas (Switaj et al. 2021). In contrast to conventional PEComas, TFE3-rearranged PEComas were shown to lack TSC2 inactivating mutations (Switaj et al. 2021). These findings have theoretically critical treatment implications, particularly for the efficacy of targeted mTOR inhibitors, as the hypothetical benefit of this therapy is likely minimized. Therefore, recognition of the rearranged variant of PEComa may assist in making important decisions regarding clinical management (Schoolmeester et al. 2015). Our study shows that PEComas of the bone and soft tissues are more likely to develop into TFE3-rearranged PEComas, which may be useful for developing future treatment strategies.
The present study has several limitations. The first serious limitation was that this study included only a small number of cases. Because PEComas of bone and soft tissue tumors are extremely rare, an international study is necessary to collect more information. Second, we could not genetically identify TSC1/TSC2 alterations or TFE3 gene rearrangements. Third, we excluded primary PEComa of the visceral organs, such as the gastrointestinal tract, uterus, and kidney, from our analysis. Thus, we could not compare the chronological characteristics or histological details of PEComas of the bone and soft tissue with those of the other visceral organs. Further genetic investigations involving a larger number of patients are necessary to establish an appropriate treatment strategy for PEComas.