Primary Culture of Undifferentiated Pleomorphic Sarcoma: Molecular Characterization and Response to Anticancer Agents

Undifferentiated pleomorphic sarcoma (UPS) is an aggressive mesenchymal neoplasm with no specific line of differentiation. Eribulin, a novel synthetic microtubule inhibitor, has shown anticancer activity in several tumors, including soft tissue sarcomas (STS). We investigated the molecular biology of UPS, and the mechanisms of action of this innovative microtubule-depolymerizing drug. A primary culture from a patient with UPS was established and characterized in terms of gene expression. The activity of eribulin was also compared with that of other drugs currently used for STS treatment, including trabectedin. Finally, Western blot analysis was performed to better elucidate the activity of eribulin. Our results showed an upregulation of epithelial mesenchymal transition-related genes, and a downregulation of epithelial markers. Furthermore, genes involved in chemoresistance were upregulated. Pharmacological analysis confirmed limited sensitivity to chemotherapy. Interestingly, eribulin exhibited a similar activity to that of standard treatments. Molecular analysis revealed the expression of cell cycle arrest-related and pro-apoptotic-related proteins. These findings are suggestive of aggressive behavior in UPS. Furthermore, the identification of chemoresistance-related genes could facilitate the development of innovative drugs to improve patient outcome. Overall, the results from the present study furnish a rationale for elucidating the role of eribulin for the treatment of UPS.


Introduction
Undifferentiated pleomorphic sarcoma (UPS), previously known as malignant fibrous histiocytoma (MFH), is a mesenchymal malignancy of soft tissue or bone that shows no definable line of differentiation. It is characterized by highly variable morphologic features, for the most part consisting of transition from storiform to pleomorphic areas. UPS is the fourth most common soft tissue sarcoma (STS) usually affecting elderly patients, with a slight male predominance [1]. Its incidence in children is very rare [2]. Although the majority of UPSs are high-grade lesions with a predilection for the lower extremities, the upper extremities can also be affected. The incidence of local recurrence varies between 19% and 31%, the rate of metastases is 31-35%, and the 5-year overall survival is 65-70% [3]. At present, there are no specific immunohistochemical markers available for the standard differential diagnosis, which is based on the exclusion of other entities i.e., dedifferentiated liposarcoma, pleomorphic leiomyosarcoma, pleomorphic liposarcoma, pleomorphic rhabdomyosarcoma, myxofibrosarcoma, poorly differentiated carcinoma, and melanoma [4]. The mainstay of therapy for localized disease is surgical resection with negative margins (R0 resection) combined with (neo)adjuvant treatments in selected cases. En bloc resection with 2-cm healthy tissue margins has been widely supplanted by limb-or function-sparing surgery, and amputation is only considered in selected cases [5]. Surgical resection with R0 margins remains the primary goal for retroperitoneal localized disease. However, large dimensions and the specific anatomy of lesions may make it difficult to obtain wide surgical margins, in which case, adjuvant radiotherapy may be useful [6,7]. The gold standard for metastatic disease is chemotherapy, but outcome is generally very poor. The most common sites of metastasis are the lungs (90% incidence), followed by bone (85%) and liver (1%) [8,9]. Anthracyclines are the most widely used drugs for the treatment of advanced and metastatic STS, including UPS, with a 16-27% response rate when used as single agents [10,11]. The association of both anthracyclines and ifosfamide has led to an improvement in response rates, but not in overall survival (OS) [12], and is used to obtain tumor shrinkage for clinical benefit. In other cases, a monotherapy regimen with doxorubicin is preferred, especially in a palliative setting [13]. The recent advent of novel agents with encouraging preliminary results may further improve treatment options and clinical outcomes.
Eribulin mesylate ( Figure 1) is a novel marine-derived synthetic macrocyclic ketone analog of halichondrin B, with an innovative mechanism of action involving the inhibition of microtubule dynamics. In 2010, eribulin was approved by the U.S. Food and Drug Administration (FDA) as monotherapy for patients with advanced/metastatic breast cancer previously treated with an anthracycline or taxane [14]. Recently, eribulin was also FDA-and European Medicines Agency (EMA)-approved for the treatment of patients with inoperable liposarcoma (LPS) after the failure of anthracycline-based therapy [15]. This microtubule inhibitor drug has an antitumor activity that differs from any other microtubule-depolymerizing drug. In particular, its unique mechanism of action would seem to involve the disruption of microtubule polymerization through site-specific binding of β-tubulin, with no effect on depolymerization [16]. The consequent sequestration of β-tubulin into non-functional aggregates leads to the inhibition of microtubule dynamics, cancer cell growth-and cell cycle-arrest, and finally, apoptosis [17][18][19]. However, although several studies have demonstrated the antitumor activity of eribulin, its mechanism of action is still not clearly understood. Recent preclinical studies have shown other complex effects mediated by eribulin, including vasculature alteration, inhibition of Wnt/β-catenin signaling, suppression of migration and invasion, and reversal of epithelial-mesenchymal transition in breast cancer, adenocarcinoma, and liposarcoma [20][21][22][23]. In a non-randomized multicenter phase II study, the European Organization for Research and Treatment of Cancer (EORTC) Soft Tissue and Bone Sarcoma Group (STBSG) assessed the activity and safety of eribulin in patients with different STS histotypes, but did not report specific results on UPS [24]. The following phase III study by the same group compared the efficacy of eribulin and dacarbazine in advanced LPS and leiomyosarcoma patients [25]. However, few data are available on eribulin activity in other aggressive sarcoma subtypes known for their refractoriness to chemotherapy [26,27].
Our understanding of the molecular biology of UPS is limited, as other STS, by the small number of cell lines available [28,29]. In this regard, primary cultures offer the ideal material to study the heterogeneous biology and behavior of UPS. We established a UPS primary culture from a surgically resected specimen of a 68 year old male patient. The primary culture was characterized in terms of gene expression, and, as the disease is aggressive, we also investigated epithelial mesenchymal transition (EMT) and drug resistance-related gene expression. We compared the anticancer activity of eribulin with that of standard drugs, to elucidate the role of eribulin in UPS treatment. We also investigated the efficacy of trabectedin, a marine-derived anticancer agent that has shown to improve the clinical outcome of patients with metastatic STS, especially liposarcomas and leiomyosarcomas (referred to as L-sarcomas in the literature) [30,31]. Trabectedin was approved in 2007 by the EMA, and in 2015 by the FDA, for the treatment of advanced STS after failure of anthracyclines or for disease not amenable to treatment with these agents [32]. In addition, we explored the efficacy of dacarbazine, a treatment option for metastatic STS (including UPS), which was used as control in a phase III trial evaluating eribulin efficacy in patients with advanced liposarcoma or leiomyosarcoma [25]. Finally, we studied the molecular mechanism by which eribulin exerted its cytotoxic effect on our UPS primary culture to improve current understanding of the antitumor effect of this tubulin inhibitor.
The present work provides a valuable insight into the natural history of UPS, and sheds light on the potential role of eribulin and other drugs, including trabectedin, for the treatment of this poorly understood cancer. the clinical outcome of patients with metastatic STS, especially liposarcomas and leiomyosarcomas (referred to as L-sarcomas in the literature) [30,31]. Trabectedin was approved in 2007 by the EMA, and in 2015 by the FDA, for the treatment of advanced STS after failure of anthracyclines or for disease not amenable to treatment with these agents [32]. In addition, we explored the efficacy of dacarbazine, a treatment option for metastatic STS (including UPS), which was used as control in a phase III trial evaluating eribulin efficacy in patients with advanced liposarcoma or leiomyosarcoma [25]. Finally, we studied the molecular mechanism by which eribulin exerted its cytotoxic effect on our UPS primary culture to improve current understanding of the antitumor effect of this tubulin inhibitor.
The present work provides a valuable insight into the natural history of UPS, and sheds light on the potential role of eribulin and other drugs, including trabectedin, for the treatment of this poorly understood cancer.

Tumor Diagnosis
The patient underwent an MRI scan of the right thigh, revealing the presence of a large, solid mass in the vastus lateralis of the quadriceps femoris muscle (maximum diameter 20 cm and transverse diameter around 5 cm) (Figure 2a,b). Macroscopic examination of the resected tumor mass, which weighed 630 g, showed a muscle segment of 21 cm × 10 cm × 6 cm covered by a lozenge of skin (18 cm × 8 cm). When incised, the mass had the appearance of a gelatinous, polylobulated nodule (9 cm × 10 cm) with healthy tissue margins. Microscopically, the hematoxylin and eosin (H&E) stained sections, reviewed by an experienced sarcoma pathologist, showed a muscle mass diffusely infiltrated by a tumor with spindle, markedly atypical, and pleomorphic cells with necrotic areas and several atypical mitoses (Figure 2e). Immunohistochemical analysis showed positivity for CD34 ( Figure 2g) and negativity for muscular markers and S100. There was no MDM2 amplification (MDM2/Cep12 ratio of 1.0), thus, the diagnosis of dedifferentiated liposarcoma was excluded [33,34]. The diagnosis was UPS with R0 resection, and the negative margins were confirmed by H&E staining (Figure 2c).

Tumor Diagnosis
The patient underwent an MRI scan of the right thigh, revealing the presence of a large, solid mass in the vastus lateralis of the quadriceps femoris muscle (maximum diameter 20 cm and transverse diameter around 5 cm) (Figure 2a,b). Macroscopic examination of the resected tumor mass, which weighed 630 g, showed a muscle segment of 21 cm × 10 cm × 6 cm covered by a lozenge of skin (18 cm × 8 cm). When incised, the mass had the appearance of a gelatinous, polylobulated nodule (9 cm × 10 cm) with healthy tissue margins. Microscopically, the hematoxylin and eosin (H&E) stained sections, reviewed by an experienced sarcoma pathologist, showed a muscle mass diffusely infiltrated by a tumor with spindle, markedly atypical, and pleomorphic cells with necrotic areas and several atypical mitoses (Figure 2e). Immunohistochemical analysis showed positivity for CD34 ( Figure 2g) and negativity for muscular markers and S100. There was no MDM2 amplification (MDM2/Cep12 ratio of 1.0), thus, the diagnosis of dedifferentiated liposarcoma was excluded [33,34]. The diagnosis was UPS with R0 resection, and the negative margins were confirmed by H&E staining (Figure 2c).

Establishment of Patient-Derived Undifferentiated Pleomorphic Sarcoma (UPS) Culture
Patient-derived cells continued to grow after culture passages. Morphological analysis of the cultured cells performed by an experienced sarcoma pathologist confirmed the establishment of a UPS primary culture (Figure 2f). The proportion of UPS cells was 50%. Furthermore, the morphological evaluation of matched patient-derived cells from healthy tissue confirmed the establishment of the UPS primary culture (Figure 2d). Given that the immunohistochemical analysis of the UPS lesion revealed positivity for CD34, we evaluated the expression of the antigen in the primary culture, confirming the positivity of the cultured cells for that marker (Figure 2h).

Establishment of Patient-Derived Undifferentiated Pleomorphic Sarcoma (UPS) Culture
Patient-derived cells continued to grow after culture passages. Morphological analysis of the cultured cells performed by an experienced sarcoma pathologist confirmed the establishment of a UPS primary culture ( Figure 2f). The proportion of UPS cells was 50%. Furthermore, the morphological evaluation of matched patient-derived cells from healthy tissue confirmed the establishment of the UPS primary culture (Figure 2d). Given that the immunohistochemical analysis of the UPS lesion revealed positivity for CD34, we evaluated the expression of the antigen in the primary culture, confirming the positivity of the cultured cells for that marker (Figure 2h).

Gene Expression Profile of Patient-Derived UPS Culture
Several markers involved in EMT and chemoresistance were evaluated to characterize the aggressiveness of the tumor (Figure 3a,b). The expression of the mesenchymal marker vimentin was 2.6-fold higher than that of the control tissue, whereas e-cadherin, an epithelial marker, was downregulated with respect to control. Tgf-β, a marker involved in several tumor-associated pathways, and EMT regulators, such as snail, were downregulated. Expression of the matrix modifying enzyme mmp2 was 2-fold higher that of control, while mmp9 and the antiapoptotic, EMT-related gene slugwere downregulated. Higher levels of some chemoresistance-related genes

Gene Expression Profile of Patient-Derived UPS Culture
Several markers involved in EMT and chemoresistance were evaluated to characterize the aggressiveness of the tumor (Figure 3a,b). The expression of the mesenchymal marker vimentin was 2.6-fold higher than that of the control tissue, whereas e-cadherin, an epithelial marker, was downregulated with respect to control. Tgf-β, a marker involved in several tumor-associated pathways, and EMT regulators, such as snail, were downregulated. Expression of the matrix modifying enzyme mmp2 was 2-fold higher that of control, while mmp9 and the antiapoptotic, EMT-related gene slug were downregulated. Higher levels of some chemoresistance-related genes were also observed. In particular, the expression of laptm4a and laptm4b genes, both involved in the transport of small molecules across endosomal and lysosomal membranes [35,36], were 3.2-and 1.05-fold higher that of control, respectively. transport of small molecules across endosomal and lysosomal membranes [35,36], were 3.2-and 1.05-fold higher that of control, respectively.

Chemotherapy Assessment in UPS Primary Culture
The antiproliferative activity of eribulin in the UPS primary culture was assessed by the mitochondrial reduction assay MTT (Figure 4a,b). The efficacy of eribulin (ERI) treatment was compared with that of a combination of an anthracycline (epirubicin, EPI) and ifosfamide (IFO), and an anthracycline alone (doxorubicin, DOXO), both of which represent the standard of care for unresectable or metastatic sarcomas, including UPS [12,13]. The primary culture was also exposed to the promising drug, trabectedin (TRABE), and to one of the second-line treatment options for metastatic STS, dacarbazine (DACA).

The Activity of Eribulin in the UPS Primary Culture
Cell morphology was examined after treatment to gain a further insight into the mechanism through which eribulin exerts its anticancer activity. Morphological changes, such as rounding up and cell shrinkage, were observed after exposure to eribulin, while untreated cells continued to proliferate with a storiform pattern, and did not show these specific features (Figure 4c). We thus analyzed the expression level of some apoptosis-related proteins to determine how this microtubuletargeted drug induces its cytotoxic effect (Figure 4d). Cell cycle arrest mediated by eribulin was confirmed by the upregulation of p21, whose expression was 2.75-fold higher than that of control. Furthermore, the expression levels of pro-apoptotic protein Bax and anti-apoptotic protein Bcl-xL were evaluated to investigate the mechanisms involved in inducing apoptosis. The results showed an upregulation of Bax and downregulation of Bcl-xL, confirming the drug-mediated induction of apoptosis. Finally, since the expression of Bax was promoted by eribulin, its downstream proteins were further assessed. Caspase-3 and caspase-9 were upregulated, indicating that eribulin exerts its antitumor activity through the activation of a caspase-dependent apoptotic pathway.

Chemotherapy Assessment in UPS Primary Culture
The antiproliferative activity of eribulin in the UPS primary culture was assessed by the mitochondrial reduction assay MTT (Figure 4a,b). The efficacy of eribulin (ERI) treatment was compared with that of a combination of an anthracycline (epirubicin, EPI) and ifosfamide (IFO), and an anthracycline alone (doxorubicin, DOXO), both of which represent the standard of care for unresectable or metastatic sarcomas, including UPS [12,13]. The primary culture was also exposed to the promising drug, trabectedin (TRABE), and to one of the second-line treatment options for metastatic STS, dacarbazine (DACA).
Patient  (Figure 4b) acquired after drug exposure were consistent with data obtained from the cytotoxicity assay. In particular, a similar cell confluence of EPI/IFO, DOXO and ERI was seen, while TRABE and DACA showed a higher confluence compared to the previous treatments.

The Activity of Eribulin in the UPS Primary Culture
Cell morphology was examined after treatment to gain a further insight into the mechanism through which eribulin exerts its anticancer activity. Morphological changes, such as rounding up and cell shrinkage, were observed after exposure to eribulin, while untreated cells continued to proliferate with a storiform pattern, and did not show these specific features (Figure 4c). We thus analyzed the expression level of some apoptosis-related proteins to determine how this microtubule-targeted drug induces its cytotoxic effect (Figure 4d). Cell cycle arrest mediated by eribulin was confirmed by the upregulation of p21, whose expression was 2.75-fold higher than that of control. Furthermore, the expression levels of pro-apoptotic protein Bax and anti-apoptotic protein Bcl-xL were evaluated to investigate the mechanisms involved in inducing apoptosis. The results showed an upregulation of Bax and downregulation of Bcl-xL, confirming the drug-mediated induction of apoptosis. Finally, since the expression of Bax was promoted by eribulin, its downstream proteins were further assessed. Caspase-3 and caspase-9 were upregulated, indicating that eribulin exerts its antitumor activity through the activation of a caspase-dependent apoptotic pathway.

Discussion
UPS, formerly known as malignant fibrous histiocytoma, was recognized for the first time as a distinct histologic STS subtype in the early 1960s [37,38]. It represents a spectrum of tumors with fibroblastic/myofibroblastic features, and no definable line of differentiation [39]. The disease manifests a wide range of histologic appearances, such as marked pleomorphism with spindle cells, multinucleated giant cells, and storiform areas [3]. UPS occurs mainly in adults, with the highest incidence rates between 50 and 70 years of age [40]. Diagnosis is based on the evaluation of hematoxylin-eosin stained sections from biopsied lesions, as specific tumor biomarkers have yet to be identified. Thus, the role of immunohistochemistry is mainly to exclude other diseases [3]. From a clinical point of view, the majority of UPS are high-grade lesions, and the incidence of local recurrence ranges between 19% and 31%. Metastases occur in 31-35% of patients, and the 5-year overall survival is 65-70% [41]. The standard therapeutic strategy for localized disease is surgery with (neo)adjuvant treatments in selected cases [5]. Chemotherapy is the standard of care in a metastatic setting, but its role is limited, and outcomes are generally poor [42]. A better understanding of the molecular background and treatment sensitivity of UPS is needed to improve the management and outcome of patients. In this regard, well-characterized cell lines could be fundamental in elucidating tumor pathophysiology, mechanisms of resistance, and the role of chemotherapy in UPS. However, given that UPS was only recently acknowledged as a distinct pathological entity, there are still no

Discussion
UPS, formerly known as malignant fibrous histiocytoma, was recognized for the first time as a distinct histologic STS subtype in the early 1960s [37,38]. It represents a spectrum of tumors with fibroblastic/myofibroblastic features, and no definable line of differentiation [39]. The disease manifests a wide range of histologic appearances, such as marked pleomorphism with spindle cells, multinucleated giant cells, and storiform areas [3]. UPS occurs mainly in adults, with the highest incidence rates between 50 and 70 years of age [40]. Diagnosis is based on the evaluation of hematoxylin-eosin stained sections from biopsied lesions, as specific tumor biomarkers have yet to be identified. Thus, the role of immunohistochemistry is mainly to exclude other diseases [3]. From a clinical point of view, the majority of UPS are high-grade lesions, and the incidence of local recurrence ranges between 19% and 31%. Metastases occur in 31-35% of patients, and the 5-year overall survival is 65-70% [41]. The standard therapeutic strategy for localized disease is surgery with (neo)adjuvant treatments in selected cases [5]. Chemotherapy is the standard of care in a metastatic setting, but its role is limited, and outcomes are generally poor [42]. A better understanding of the molecular background and treatment sensitivity of UPS is needed to improve the management and outcome of patients. In this regard, well-characterized cell lines could be fundamental in elucidating tumor pathophysiology, mechanisms of resistance, and the role of chemotherapy in UPS. However, given that UPS was only recently acknowledged as a distinct pathological entity, there are still no commercial UPS cell lines available [29]. Thus, patient-derived primary cultures represent the ideal experimental material to study the heterogeneous biology of UPS [43][44][45][46][47].
In the present work, cells were derived from a UPS lesion of a 68-year old male patient who underwent tumor resection. Morphological and immunohistochemical analysis confirmed the establishment of the UPS primary culture (Figure 2f,h). Gene expression analysis of the tumor specimen confirmed the mesenchymal origin of this disease. In particular, vimentin and mmp2, both mesenchymal-related genes, were upregulated, whereas the expression of e-cadherin, an epithelial gene, was lower than that of matched healthy tissue (Figure 3a). Furthermore, upregulation of genes involved in drug resistance, especially to anthracyclines, was observed. In this regard, the amplification of laptm4a and laptm4b genes has been reported in numerous chemoresistant tumors, including breast and gallbladder cancer [36,48]. These genes would seem to be involved in the membrane trafficking of drugs, especially anthracyclines. Thus, their overexpression could partially explain the similar antiproliferative activity of the anthracycline-based regimens (EPI/IFO and DOXO) used in our study. Further analyses are needed to elucidate the role of these genes in UPS. We also evaluated the diagnostic impact of CD109 expression, previously analyzed by our group in a series of patient-derived myxofibroscarcoma primary cultures [49]. Results showed a downregulation of cd109 gene in this STS histotype with respect to healthy donors, indicating the potential usefulness of cd109 in the standard differential diagnosis of MFS [49,50].
Finally, in order to elucidate the role of chemotherapy in UPS, we exposed the primary culture to different drugs currently used for the treatment of advanced STS. The activity of novel drugs, such as trabectedin and eribulin, the latter under investigation for use in STS, was also assessed. Results revealed a similar activity of doxorubicin alone, the epirubicin-ifosfamide combination, and eribulin alone. Although trabectedin showed a lower antiproliferative effect than the above drugs, it nevertheless induced a cytotoxic effect, confirming previous reports of its antitumor activity in UPS [51,52]. Dacarbazine did not affect cell survival. The significant anticancer activity of eribulin in the primary culture with respect to the first-line ifosfamide-epirubicin combination, prompted us to analyze the mechanism of action through which eribulin exerts its antitumor activity in UPS. Morphological changes in the cell culture, such as rounding up and cell shrinkage, were detected after eribulin treatment with respect to control cells, which were characterized by a storiform pattern (Figure 4b,c). We thus analyzed apoptosis-related proteins to better understand the cytotoxic activity mediated by this new interfering microtubule. Results indicated that eribulin exerted its antiproliferative activity through the activation of p21 and Bax, and the suppression of Bcl-xL. The upregulation of the 2 pro-apoptotic proteins led to the collapse of mitochondrial transmembrane potential. Finally, investigation of the downstream proteins caspase-3 and caspase-9 showed that eribulin enhanced their expression. These data suggest that eribulin induces cell cycle arrest and cell death programming through a caspase-dependent mitochondrial pathway ( Figure 5). commercial UPS cell lines available [29]. Thus, patient-derived primary cultures represent the ideal experimental material to study the heterogeneous biology of UPS [43][44][45][46][47].
In the present work, cells were derived from a UPS lesion of a 68-year old male patient who underwent tumor resection. Morphological and immunohistochemical analysis confirmed the establishment of the UPS primary culture (Figure 2f,h). Gene expression analysis of the tumor specimen confirmed the mesenchymal origin of this disease. In particular, vimentin and mmp2, both mesenchymal-related genes, were upregulated, whereas the expression of e-cadherin, an epithelial gene, was lower than that of matched healthy tissue (Figure 3a). Furthermore, upregulation of genes involved in drug resistance, especially to anthracyclines, was observed. In this regard, the amplification of laptm4a and laptm4b genes has been reported in numerous chemoresistant tumors, including breast and gallbladder cancer [36,48]. These genes would seem to be involved in the membrane trafficking of drugs, especially anthracyclines. Thus, their overexpression could partially explain the similar antiproliferative activity of the anthracycline-based regimens (EPI/IFO and DOXO) used in our study. Further analyses are needed to elucidate the role of these genes in UPS. We also evaluated the diagnostic impact of CD109 expression, previously analyzed by our group in a series of patient-derived myxofibroscarcoma primary cultures [49]. Results showed a downregulation of cd109 gene in this STS histotype with respect to healthy donors, indicating the potential usefulness of cd109 in the standard differential diagnosis of MFS [49,50].
Finally, in order to elucidate the role of chemotherapy in UPS, we exposed the primary culture to different drugs currently used for the treatment of advanced STS. The activity of novel drugs, such as trabectedin and eribulin, the latter under investigation for use in STS, was also assessed. Results revealed a similar activity of doxorubicin alone, the epirubicin-ifosfamide combination, and eribulin alone. Although trabectedin showed a lower antiproliferative effect than the above drugs, it nevertheless induced a cytotoxic effect, confirming previous reports of its antitumor activity in UPS [51,52]. Dacarbazine did not affect cell survival. The significant anticancer activity of eribulin in the primary culture with respect to the first-line ifosfamide-epirubicin combination, prompted us to analyze the mechanism of action through which eribulin exerts its antitumor activity in UPS. Morphological changes in the cell culture, such as rounding up and cell shrinkage, were detected after eribulin treatment with respect to control cells, which were characterized by a storiform pattern (Figure 4b,c). We thus analyzed apoptosis-related proteins to better understand the cytotoxic activity mediated by this new interfering microtubule. Results indicated that eribulin exerted its antiproliferative activity through the activation of p21 and Bax, and the suppression of Bcl-xL. The upregulation of the 2 pro-apoptotic proteins led to the collapse of mitochondrial transmembrane potential. Finally, investigation of the downstream proteins caspase-3 and caspase-9 showed that eribulin enhanced their expression. These data suggest that eribulin induces cell cycle arrest and cell death programming through a caspase-dependent mitochondrial pathway ( Figure 5).

Ethics Statement
Briefly, the tumor mass was harvested intraoperatively from a patient undergoing surgical resection, as reported in the Results section. The study was approved by IRST-Area Vasta Romagna Ethics Committee, approval no. 4751, 31 July 2015, and all the procedures were performed following Good Clinical Practice and in accordance with the principles laid down in the Helsinki declaration. The patient gave written informed consent to take part in the study.

Patient History
A 68 year old male presented with a solitary, rapidly growing lesion in his right thigh. His past medical history was unremarkable. An MRI scan showed a large solid lesion in the quadriceps femoris muscle of the limb (longitudinal diameter 20 cm and transverse diameters 53 mm × 88 mm). A biopsy revealed the presence of a malignant mesenchymal tumor with poorly differentiated myxoid aspects compatible with undifferentiated pleomorphic sarcoma. A full-body CT scan was negative. The patient underwent surgical resection of the lesion, the pathology report describing a nodular lesion (9 cm × 10 cm) from high-grade undifferentiated pleomorphic sarcoma with wide disease-free margins. As there was no evidence of metastatic disease post-surgery, adjuvant radiotherapy was administered. Chest CT re-staging after radiation treatment revealed the presence of a single lung lesion suggestive of metastasis, and subsequently confirmed as such by an FDG PET/CT scan (Siemens Healthcare Ltd., Milan, Italy). The patient recently underwent metastasectomy.

Histological and Immunohistochemical Diagnosis
The surgical specimen was selected and analyzed by an experienced sarcoma pathologist, and processed within 3 h of surgical resection. Normal tissue comprising thigh muscle and adipose tissue and blood vessels obtained by tumor R0 resection was used as study control (Figure 2c). Differential diagnosis was performed by hematoxylin and eosin (H&E, Sigma Aldrich, St. Louis, MO, USA) staining. Briefly, the surgical material was washed twice with sterile phosphate buffered saline (PBS), paraffin-embedded in a cryomold and then frozen at −80 • C. The frozen tissue blocks were then sectioned into 5 µm-thick slices using a microtome, after which the sections were hydrated and stained with hematoxylin (Sigma Aldrich, St. Louis, MO, USA) and eosin (Sigma-Aldrich), according to the manufacturer's instructions. Finally, the stained sections were washed three times with PBS 1×, mounted with Cytoseal™ XYL (Thermo Scientific™ Richard-Allan Scientific™, San Diego, CA, USA) mounting media, covered with a coverslip, and analyzed. CD34 determination was performed by immunohistochemical staining. Briefly, 5 µm-thick sections cut from paraffin-embedded tissue were deparaffinized with xylene for 1 h, after which they were rehydrated and incubated with antigen retrieval solution in a water bath at 98.5 • C for 30 min. The sections were cooled for 20 min, incubated for 10 min with 3% hydrogen peroxide solution, and washed twice with demineralized water. They were then incubated with 3% bovine serum albumin in PBS for 20 min, and incubated with CD34 antibody (Abcam ® , Cambridge, UK) diluted 1:200 at room temperature for one hour. CD34 antibody was detected by immunoperoxidase techniques using the streptavidin-biotin-peroxidase complex (ABC) method [53]. Cell nuclei were counterstained with hematoxylin (Sigma Aldrich). The slides were mounted with Cytoseal™ XYL (Thermo Scientific™ Richard-Allan Scientific™) mounting media, covered with a coverslip, and analyzed. Finally, MDM2 gene amplification was assessed by FISH analysis (Vysis MDM2/CEP 12 FISH Probe Kit, Abbott Park, IL, USA) to exclude a diagnosis of dedifferentiated liposarcoma.

Establishment of UPS Primary Cell Culture
A patient-derived UPS primary cell culture was established from surgical tissue. The tumor specimens were washed twice in PBS, and shredded into 1-2 mm 3 pieces with surgical scalpels.
The obtained pieces were then incubated with a PBS solution of 2 mg/mL collagenase type I (Millipore Corporation, Billerica, MA, USA) at 37 • C in stirring conditions for 15 min, after which the sample was stored overnight at room temperature. Collagenase digestion was then blocked by adding DMEM supplemented with 10% fetal bovine serum, 1% glutamine, and 10% penicillin/streptomycin, after which cells were isolated from the aggregates using a 100 µm sterile filter (CellTrics, Partec, Münster, Germany). Cells were counted and seeded in standard monolayer cultures at a density of 80,000 cells/cm 2 , and maintained in complete DMEM medium at 37 • C in a 5% CO 2 atmosphere. All the experiments were performed with the use of low-passage primary cultures, and analyzed by an experienced sarcoma pathologist.

Histological and Immunohistochemical Analysis of UPS Primary Culture
Hematoxylin and eosin (H&E) staining was performed to evaluate the morphological features and distribution of cells of the patient-derived UPS culture. Cells (100,000) were cytospun to glass slides, fixed in acetone for 10 min, and then in chloroform for 5 min, and stored at −20 • C. Finally, the slides were thawed, hydrated, and stained with H&E following the manufacturer's instructions. The stained slides were analyzed with an optical Zeiss Axioskop microscope (Carl Zeiss, Gottïngen, Germany) equipped with a Polaroid camera. CD34 immunohistochemical analysis was performed on the cytological slides following the protocol reported in the Histological and Immunohistochemical Diagnosis section. The stained slides were analyzed with an optical Zeiss Axioskop microscope equipped with a Polaroid camera.

Gene Expression Analysis
RNA extraction was carried out using the cells isolated from tissue specimens without being cultured in vitro, in order to avoid molecular changes. Briefly, mRNA isolation was performed using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer's instructions. iScript cDNA Synthesis Kit (BioRad, Hercules, CA, USA) was used to reverse transcribe 500 ng of extracted RNA. Gene expression analysis was then carried out by Real-Time PCR using a 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). Amplification was performed in a total volume of 20 µL containing 2× Taqman Universal PCR Master Mix (Applied Biosystems) and 2 µL of cDNA. The following markers were analyzed: vimentin, e-cadherin, tgf-β, snail, mmp2, mmp9, and slug. Actb and hprt were used as reference genes. The resulting amount of the transcripts was normalized to the reference genes with the 2(-delta delta C(T)) method. We used SYBR Select Master Mix (Applied Biosystems) with 2 µL of cDNA for tp53i3, laptm4a, rab22a, s100p, cd109, and laptm4b analysis. Actb and gapdh were used as reference genes.

Protein Expression Analysis
Patient-derived cell cultures were exposed to eribulin at the human plasma peak concentration, as reported in the Drug Testing section. After 72 h, cells were trypsinized, immediately frozen with liquid nitrogen, and stored at −80 • C. Proteins were then extracted using a RIPA buffer with 10% PMSF, 1% HALT phosphatase inhibitor cocktail, and 0.5% protease inhibitor cocktail. The cell suspension was centrifuged at 15,000 rpm for 20 min at 4 • C and protein contents were determined using a BCA protein assay kit (Pierce™ BCA Protein Assay Kit, Thermo Scientific, Waltham, MA, USA). An equal amount of protein from each sample was separated on Criterion™ Precast Gel Tris-HCl (Biorad, Hercules, CA, USA) and transferred to polyvinylidene fluoride membranes (Millipore Corporation). The membranes were blocked for 2 h with a solution containing 5% fat-free milk PBS with 0.1% Tween 20 (Sigma Aldrich, St. Louis, MO, USA) at room temperature, and incubated overnight at 4 • C with each of the following antibodies: anti-p21 (1:2000 Cell Signaling, Danvers, MA, USA), anti-Bax (1:1000 Cell Signaling), anti-Bcl-xL (1:1000 Cell Signaling), anti-caspase-3 (1:1000 Cell Signaling), anti-caspase-9 (1:1000 Cell Signaling), and anti-vinculin (1:1000 Thermo Scientific). After washing, the membranes were incubated for 1 h at room temperature with horseradish peroxidase-conjugated secondary antibody. Densitometric analysis of proteins was performed on Western blot with Quantity One software version 4.6.9 (BioRad, Hercules, CA, USA).

Statistical Analysis
Three independent replicates were performed for each experiment. Data are presented as mean ± standard deviation (SD), or mean ± standard error (SE), as stated, with n indicating the number of replicates. Differences between groups were assessed by a two-tailed Student's t-test, and accepted as significant at p < 0.05.

Conclusions
In conclusion, the results from the present study highlight the potential importance of using patient-derived primary cultures to study the tumor heterogeneity and clinical diversity of UPS. They also increase our understanding of the molecular background of this STS histotype, and could contribute to the development of new therapeutic strategies. Finally, our investigation of the mechanism of action of the promising microtubule-depolymerizing drug, eribulin, provides a rationale for a more in-depth evaluation of its role in the treatment of this mesenchymal malignancy.