The VEGF- and PDGF-family of angiogenic markers have prognostic impact in soft tissue sarcomas arising in the extremities and trunk

Background Soft-tissue sarcomas are rare malignant tumors of mesenchymal lineage that can arise in any part of the body. Prognosis, and hence also treatment may vary according to histologic subtype and localization. Angiogenesis is the process of forming new blood vessels from pre-existing ones. The deregulation of this process is thought to be an important step in malignant transformation. This study investigates the prognostic impact of platelet derived growth factor- (PDGF), vascular endothelial growth factor- (VEGF) and fibroblast growth factor (FGF) families in soft-tissue sarcomas of the extremities & trunk (ET) and visceral & retroperitoneal (VR) locations. Methods Tumor samples from 181 patients (115 ET and 66 VR) with resected soft tissue sarcomas were collected and tissue microarrays were constructed. Immunohistochemistry was used to evaluate angiogenic marker expression. Recurrence-free survival (RFS), metastasis-free survival (MFS) and disease-specific survival (DSS) were used as endpoints in prognostic impact assessment. Results In univariate analyses, almost all investigated angiogenic markers had prognostic impact in the ET group. In contrast, only FGFR-1 showed any significant prognostic impact in the VR group. In the multivariate analyses, PDGF-D (HR = 1.863, 95% CI = 1.057-3.283, P = 0.031), VEGFR-1 (HR = 2.106, 95% CI = 1.038-4.272, P = 0.039) and VEGF-A (HR 2.095, 95% CI 1.028-4.271, P = 0.042) were independent negative prognosticators for DSS, MFS and RFS, respectively, in the ET group. FGFR-1 was an independent positive prognosticator for DSS (HR = 0.243, 95% CI = 0.095-0.618, P = 0.003) in the VR group. Conclusions Angiogenic molecules from the PDGF and VEGF families have prognostic impact in soft-tissue sarcomas arising in the ET, but not in VR locations. In the latter histological grade and resection margins are the most important prognostic factors.


Background
Soft tissue sarcomas (STS) constitute a highly heterogeneous collection of tumors comprising over 50 histological subtypes, arising from mesenchymal tissue and capable of forming tumors in all parts of the human body [1]. This group amounts to 0.5-1% of the annual tumor burden with a mortality of about 40-60%, resulting in an estimated 11 280 cases and 3 900 deaths in the US in 2012 [2].
It is good practice to distinguish between STSs arising in the extremity & trunk (ET), head & neck (HN) and visceral & retroperitoneal (VR) localizations as treatment and prognosis vary widely according to localization [3]. Further subdivision, according to histological type, malignancy grade, stage and vascular invasion among others, can be conducted [3]. Definitive treatment is radical surgery followed by radiotherapy in case of non-radical surgical margins [4]. Adjuvant chemotherapy for adult STS is still under investigation, and hence the routine use of such treatment is today limited to the palliative setting [5].
Angiogenesis is the process of forming new blood vessels from pre-existing ones. Folkman and coworkers proved this to be a pivotal step in carcinogenesis by showing that tumors would not grow beyond > 2 mm in diameter without forming vasculature [6,7]. In 2001, Hanahan and Weinberg, suggested angiogenesis as one of the hallmarks of cancer [8] and in the 2011 updated version angiogenesis was still considered one of the most important aspects of cancer progression [9].
Vascular endothelial growth factors (VEGF) and receptors (VEGFR) are pivotal in endothelial cell proliferation and sprouting during angio-and lymphangiogenesis [10]. Platelet-derived growth factors (PDGF) and receptors (PDGFR) play an important part in the regulation of tumor stroma through the recruitment of pericytes and vascular smooth muscle cells helping to stabilize newly formed vessels and through stimulation of stromal cells to produce VEGF-A and thus drive angiogenesis [11,12]. Fibroblast growth factors (FGF) and receptors (FGFR) drives endothelial cell proliferation and sprouting and activate several molecules involved in extracellular matrix remodelling including matrix metallo-proteinases and urokinase-like plasminogen activator [13].
Our group has previously reported on the expression of VEGF, PDGF and FGF families of growth factors in STSs of all sites [14][15][16]. This report investigates the differential impact of these growth factors in STSs arising in ET versus VR localizations.

Patients and clinical samples
Primary tumor tissue from anonymized patients diagnosed with STS at the University Hospital of North-Norway and the Hospitals of Arkhangelsk County, Russia, from 1973 through 2006, were collected. In total 496 patients were registered from the hospital databases. Of these, 388 patients were excluded from the study because of: missing clinical data (n = 86), inadequate formalin-fixed paraffin-embedded (FFPE) tissue blocks (n = 161), no surgery performed and/or metastasis present at the time of diagnosis (n = 55) or head and neck sarcomas (n =13). Thus 115 patients with STSs of the extremities and trunk wall and 66 patients with STSs of visceral or retroperitoneal origin, with complete medical records and FFPE tissue blocks were eligible.
This report includes follow-up data as of September 2009. The median follow-up was 53.9 (range 0.5-391.7) months for extremity and trunk patients and 59.4 (range 0.10-366.7) months for visceral and retroperitoneal patients. Complete demographic and clinical data were collected retrospectively. Formalin-fixed and paraffin-embedded tumor specimens were obtained from the archives of the Departments of Pathology at the University Hospital of North-Norway and the Hospitals of Arkhangelsk County, Russia. The tumors were graded according to the French Fédération Nationale des centres de Lutte Contre le Cancer (FNCLCC) system and histologically subtyped according to the World Health Organization guidelines [1,17]. Wide resection margins were defined as wide local resection with free microscopic margins or amputation of the affected limb or organ.

Microarray construction
All sarcomas were histologically reviewed by two trained pathologists (S. Sorbye and A. Valkov) and the most representative areas of tumor cells (neoplastic mesenchymal cells) were carefully selected and marked on the hematoxylin and eosin (H/E) slide and sampled for the tissue microarray (TMA) blocks. The TMAs were assembled using a tissue-arraying instrument (Beecher Instruments, Silver Springs, MD). The Detailed methodology has been previously reported [18]. Briefly, we used a 0.6 mm diameter stylet, and the study specimens were routinely sampled with four replicate core samples from different areas of neoplastic tissue. Normal tissue from the patients was used as staining control.
To include all core samples, 12 TMA blocks were constructed. Multiple 5-μm sections were cut with a Micron microtome (HM355S) and stained by specific antibodies for immunohistochemistry (IHC) analysis.

Immunohistochemistry
The applied antibodies were subjected to in-house validation by the manufacturer for IHC analysis on paraffin-embedded material. The detailed methodology has previously been reported [14][15][16].

Statistical methods
All statistical analyses were done using the statistical package SPSS (Chicago, IL), version 16. The IHC scores from each observer were compared for interobserver reliability by use of a two-way random effect model with absolute agreement definition. The intraclass correlation coefficient (reliability coefficient) was obtained from these results. The Chi-square test and Fishers Exact test were

Ethical clearance
The Norwegian National Data Inspection Board and The Regional Committee for Research Ethics (Northern Norway) approved the study.
The impact of angiogenic markers on DSS, MFS and RFS in the VR group is summarized in Table 3.
FGRF-1 (P = 0.023) was the only prognostic indicator for DSS and PDGF-C (P = 0.045) for RFS. Table 4 presents multivariate analyses of clinicopathological and angiogenic marker variables with respect to DSS, MFS and RFS in the ET and VR groups, respectively.
In the VR group, high malignancy grade (P = 0.003) and non-wide resection margins (P = 0.014) were significant independent adverse prognostic indicators of DSS whereas high FGFR-1 expression (HR = 0.243, 95% CI = 0.095-0.618, P = 0.003) was an independent positive prognostic indicator of DSS. Female gender (P = 0.038) was an independent negative prognostic indicator of MFS while non-wide resection margins (P < 0.001) was an independent negative prognostic indicator of RFS. Abbreviations: PDGF Platelet-derived growth factor, PDGFR Platelet-derived growth factor receptor, VEGF Vascular endothelial growth factor, VEGFR Vascular endothelial growth factor receptor, FGF Fibroblast growth factor, FGFR Fibroblast growth factor receptor.

Discussion and conclusions
In our univariate analyses high expression of most examined angiogenic markers were prognosticators of DSS and/or MFS and/or RFS in the ET group. Further, PDGF-D was an independent negative prognostic indicator of DSS, VEGFR-1 an independent negative prognostic indicator of MFS and VEGF-A an independent negative prognostic indicator of RFS. In contrast, only FGFR-1 was a prognosticator of DSS in both the univariate and multivariate analyses of the VR group. To our knowledge, this is the first comparison of the expression of angiogenic molecules in ET versus VR STSs. Current knowledge of the importance of tumor localization (ET versusVR tumors) when it comes to the prognostic impact of angiogenic markers in STSs is limited. Yudoh et. al. investigated the level of VEGF-A in tissue from ET patients and found high levels to predict survival, local recurrence and metastasis [18]. We have previously reported on the expression of PDGFs, VEGFs and FGFs in a larger cohort of STS of mixed sites and histology and found high expression of VEGFR-3, PDGF-B and FGF2 to have independent negative prognostic impact on DSS [14][15][16]. When comparing the expression of angiogenic markers based on tumor location, it becomes apparent that these variables almost exclusively have prognostic impact in STS arising in the ET group (Tables 2, 3 and 4). This difference could to some extent be due to a smaller number of patients in the VR group, with a resulting increased risk of false negative results. However, near all angiogenic markers showed significant prognostic impact in the univariate analyses of the ET group, whereas only FGFR-1 showed prognostic impact in the VR group. Table 1 summarizes the clinopathological values in the ET and VR groups and it is apparent that the VR group contains a higher percentage of leiomysarcomas and liposarcomas. The different distribution of histologies between the ET and VR groups might suggest that angiogenic markers have higher impact in STSs arising in ET locations. Another explanation may be that ET tumors, even the slow growing ones, will produce symptoms when they reach a certain size due to limits created by connective and muscle tissue and blood and lymph vessels. VR tumors could in contrast grow to significant size before producing symptoms. This may explain our results as VR tumors in many cases only are found after the angiogenic switch have occurred, thus the impact of angiogenic markers have been negated in these tumors.
In the PDGF-axis, all markers were prognosticators of DSS, all but PDGF-C were prognosticators of MFS and all but PDGF-C and PDGFR-β were prognosticators of RFS in the ET group (Table 2), while none of the PDGFs were prognosticators in the VR group. Further, PDGF-D was found to be an independent negative prognostic factor for DSS in the ET group. In our previous study, PDGF-B was an independent prognosticator of DSS [15], and in this study PDGF-D is an independent prognosticator of DSS. PDGF-B binds all PDGFRs while PDGF-D binds PDGFR-αβ and-ββ [11]. Both PDGF-B and PDGF-D has been shown to exhibit similar and extensive angiogenic and transforming abilities [19,20]. Although our results cannot distinguish whether PDGF signalling drives tumor development through angiogenesis or other pathways, they strongly suggest PDGF signalling to be an important part of STS growth and progression.
In the VEGF-axis, VEGF-A, and VEGFR-1 were prognosticators of DSS, MFS and RFS in the ET group, while none of the VEGFs were prognosticators in the VR group (Table 2). Further, VEGFR-1 was an independent prognostic indicator of MFS and VEGF-A was an independent prognostic indicator of RFS in the ET group. VEGF-A signalling is the major angiogenic pathway, and high tumor expression and availability in serum has previously been associated with malignancy grade, metastasis, local recurrence and worse overall survival in STS patients [18,[21][22][23][24][25][26]. VEGFR-1 is thought to modulate VEGF-A signalling through VEGFR-2, has anti-angiogenic properties in its soluble form, and has been linked to metastasis in experimental studies suggesting a feasible biological link for our finding in these STS patients [27,28]. This latter finding is quite interesting as antibodies and small-molecules targeting VEGFR-1 are being developed [29,30].
In the FGF-axis, FGF-2 was an unfavorable prognostic indicator of DSS in ET group. FGF2 is thought to drive cellcycling, activate extracellular matrix remodelling and to rescue PDGF-B and VEGF-A driven angiogenesis in the presence of their respective inhibitors [13,31,32]. Surprisingly, FGFR-1 was an independent positive indicator of DSS in the VR group. To our knowledge these are new data, but these results have to be validated before a firm conclusion may be drawn due to the low number of patients.
This study enhances our current knowledge on angiogenic prognosticators in STSs, strongly indicates the involvement of the PDGF and VEGF pathways in ET STS development and adds to the growing body of evidence suggesting that STSs of different sites and histology should be analyzed independently in future studies. Further emphasis should also be put on validating VEGFR-1 as a predictor of MFS in ET STS patients, as these patients may benefit from adjuvant therapy targeting VEGFR-1.