Despite improvements in diagnostic techniques, surgical techniques and cancer therapies, the prognosis of patients with PDAC remains extremely poor and further studies are needed to higher knowledge of tumor pathogenesis and to lead to new preventive and therapeutic options. In recent years, special attention is being paid to the role of tumor-associated stroma as it plays an important part in tumor development and progression, as well as, resistance to therapies(22). The role of the CXCR4 receptor is key to the interaction of cancer cells and their microenvironment and it is considered a promising prognostic biomarker for cancer. It is expressed in numerous cell types, but little is known about the relationship between its expression in the stroma and its prognostic implication. Its value in pancreatic cancer is yet to be determined, and it is known that PDCA contains a very important tumor-bound stromal component, so elucidating the relationship of the CXCR4 expression in the stroma and its involvement in PDAC is of particular interest.
Importance of stroma in PDAC
In the present study, we first compared the percentage of tumor-associated stroma with non-tumor stroma tissue in 23 patients diagnosed with PDAC. It was observed that the percentage of stroma in the total volume was significantly higher (P-value = 0.001) in tumor tissue than in non-tumor tissue samples (67.4 ± 4.8 and 32.4 ± 5.2, respectively), which may indicate that the stroma might be involved in the development of pancreatic cancer (Table 1). It has been proposed that neoplastic cell-matrix interactions stimulate the large desmoplastic reaction and the stroma production is promoted by the activation of multiple neoplastic cell-derived signaling pathways; furthermore, numerous studies have provided evidence that the microenvironment co-evolves with transformed epithelial cells in different carcinomas; hence we find a significant stromal component(24). This is consistent with our findings. Therefore, the percentage of stroma in the tumor samples is quite high as has been found in other research projects (2,22,24,25).
Expression of CXCR4 as a biomarker of PDCA
Once the importance of the stroma in the tumor samples was proven, we must take into account that the chemokine CXCR4 receptor has been related to tumor-stroma interactions, and that its expression has also been detected in important cells in tumor microenvironment(3), in CSCs(17,26), tumor cells(27), or in cells of the immune system such as naive T cells, some memory T cells, B cells, and mature dendritic cells. These play a central role in lymphocyte trafficking and homing to lymph nodes(28). The expression of CXCR4 in the different cellular components was analysed by the pathologist (Table 1). Like Wehler et al proved, cytoplasmatic staining of CXCR4 was observed in all cases, although some have reported cases with an additional membranous localization of CXCR4. This is consistent with the previously described inducible translocation of CXCR4 from the membrane to the cytoplasm(27,29).
As depicted in Table 1, on the one hand, as in the data series, the overall expression of CXCR4 was studied; on the other hand, the percentage of CXCR4 expression was only analysed in the stroma of the samples, and the CXCR4 expression found in the cells of the immune system was also analysed. The CXCR4 expression was significantly higher (P-value = 0.022) in the stroma of tumor tissues than in the stroma of non-tumor tissues (23.5 ± 6.1 and 8.7 ± 4.6, respectively), whereas no statistical association was possible in the evaluation of expression in tissues globally or in immune cells. The data set described here constitutes a relatively small set of samples to be analysed compared to other larger data sets of CXCR4 expression studies in PDCA. These studies analyse the whole tissue(28), so we propose to focus the analysis on the CXCR4 expression in the stroma and eliminate possible artefacts that may be generating CXCR4 expression in other cellular subtypes.
Association of the CXCR4 expression in the stroma with patient characteristics
In order to determine the role of the CXCR4 stroma expression in the pathology of PDAC, the present study also analysed the association between overexpression or underexpression of CXCR4 in tumor tissues versus their matched non-tumor tissue samples with clinicopathological factors, but we found no statistical significance (Table 2), this happened in contrast to other datasets such as the studies by Darash-Yahana et al, which positively correlate the CXCR4 expression with tumor growth, vascularization and metastasis in prostate cancer(4). We probably also found no association with mortality or recurrence like Kure et al, due to the small number of samples (Figure 2)(30).
Relevance of the CXCR4 expression in tumor stroma in PDAC differentiation
Focusing exclusively on tumor tissues of PDAC patients, we evaluated the role of the CXCR4 expression in tumor stroma with respect to prognostic factors, finding a relationship between poor differentiation and high CXCR4 expression. Studies in glioblastoma (GBM) suggest an important role of CXCR4 in cell proliferation and in maintaining the neoplastic phenotype of GBM cells(12). This could justify that poor differentiation promoted by proliferative activity that results in loss of original conformation is mediated by CXCR4 expression. In addition, abnormalities affecting signal transduction pathways involved in the control of cell growth and other malignant properties, which may be affected by CXCR4, might compromise tumor staging and the degree of differentiation(28). Indeed Heinrich et al, observe increased CXCR4-mediated proliferation in pancreatic cell lines(31).
There is controversy about the prognostic value of CXCR4 overexpression in different tumors and although there are numerous related studies, Zhao et al, provided a negligible association between CXCR4 and clinical outcome(21). There are numerous series of data correlating CXCR4 overexpression with poor prognosis in pancreatic cancer, Zhan et al, found no association between poor differentiation and CXCR4 overexpression, but did find an association with pathological type, lymph node stage, and TNM stage(32). In experiments carried out in mice, Malik et al. linked the CXCR4/CXCL12 axis to PDAC growth, spread, chemoresistance, and immune evasion(2). On the other hand, Kure et al, found high level of correlation between the CXCR4 expression with well-differentiated PDAC(30). As Gebauer et al, no association with mortality or recurrence was found(33).