Discussion
It is well known that inflammation is one of the major hallmarks of cancer (21), and an inflammatory microenvironment abundant in various signaling molecules, particularly cytokines, is a characteristic feature of RCC. It has been suggested that the inflammatory profile may be useful for predicting prognosis, survival and/or treatment response in patients with this type of tumor (22). Although there are some previous data on serum cytokines in RCC (15), the cytokine profile of urine from patients with ccRCC is not fully clear. Therefore, the present study aimed to investigate the role of urinary and circulating cytokines in the pathogenesis of ccRCC, as well as their potential capacity to be used for noninvasive diagnosis of ccRCC. Furthermore, the present study examined the effect of surgery and tumor grade on the urinary cytokine profile.
Regarding the potential diagnostic and predictive value of urinary cytokines in ccRCC, the present study found that IL-1Ra, IL-6, IL-15, CCL2, CCL3, CCL4, CXCL10, GM-CSF and PDGF-BB reached significantly higher levels in the urine of patients prior to operation compared with those in control urine. Since changes in urinary immune profile reflect ongoing pathological processes in the kidney (23–25), increased levels of these cytokines may be indicative of their tumorigenic character. On the other hand, the cytokine profile of urine may not always exactly match the immune changes occurring in ccRCC tissue, as demonstrated in the case of plasma cytokines (15). Therefore, several factors need to be considered when interpreting the current results. Firstly, cytokines have a short half-life and their resulting concentration measured in urine samples may be influenced by the presence of cytokine-binding proteins, inhibitors, soluble cytokine receptors and/or the method of sample processing (26). In addition, some cytokines, after binding to their target receptor in tissue, may be internalized and subsequently degraded (27), implying that the levels of urinary cytokines may represent only a fraction of their originally secreted levels. Finally, individual cytokines may have different kidney excretion rates (28). Even if the cytokine profiles of urine and tissue were correlated, another important fact that should be considered is that numerous cytokines do not have a clearly defined character in the TME and can have either tumor-promoting or tumor-suppressing effects, depending on the tissue context. Therefore, further investigation is needed to determine the effect of each cytokine on the development of ccRCC in order to uncover the underlying mechanism of each of them in the pathogenesis of ccRCC. Nevertheless, it could be hypothesized that IL-1Ra, IL-6, IL-15, CCL2, CCL3, CCL4, CXCL10, GM-CSF and PDGF-BB may be promising candidates for urinary markers of ccRCC.
Notably, presurgical urine, as well as urine from patients with HG cancer, contained significantly lower levels of G-CSF than urine from control subjects. It has been shown that the hematopoietic factor G-CSF can be secreted by tumor cells, and that it contributes to tumor aggression and is negatively associated with prognosis (29). The tumor-promoting character of G-CSF has also been proposed in the context of ccRCC. Patients with nonmetastatic ccRCC and high intratumoral expression levels of G-CSF have been reported to be more likely to have higher grade tumors and a higher probability of experiencing recurrence (30). However, it remains unclear why healthy subjects exhibited higher levels of G-CSF in urine than patients with ccRCC in the present study, and whether these values are outside the physiological range or not, since the cut-off values of individual cytokines have not yet been established (31).
To determine the possible predictive ability of the examined molecules, the current study further evaluated the changes in their levels regarding tumor grade. The analysis further revealed that both patients with LG and HG cancer had significantly higher levels of IL-1Ra, IL-6, IL-15, CCL2 and CCL4 in their urine compared with controls. However, the levels of these cytokines (except IL-6) did not change significantly with the increase of tumor grade. These findings suggest that these cytokines probably would not have much clinical value in terms of tumor stage prediction in patients with ccRCC. By contrast, the results of PDGF-BB, CCL5, CXCL10, GM-CSF and G-CSF with regard to tumor stage appeared to be more clinically relevant and thus may deserve more attention. These cytokines were detected in significantly different concentrations only in one group of patients, either LG (GM-CSF) or HG (the remaining cytokines), compared with those in the control group. This may indicate that they represent a promising novel tool for distinguishing an advanced form of ccRCC from a localized one, and subsequently may enable us to estimate the prognosis of patients. Additionally, CXCL10 reached a significantly higher concentration in HG urine than in LG urine, whereas IL-6 levels were higher in both stages compared with the controls and were also significantly increased with tumor dedifferentiation, which suggests that these two could be potential markers of tumor progression. Previous reports seem to support our assumption, as the majority of the aforementioned cytokines have been suggested to have tumorigenic effects. PDGF-BB, a major mitogenic factor, serves a crucial role in both physiological and pathological blood vessel development, including tumor angiogenesis (32). Aberrant PDGF signaling contributes to several cancer hallmarks and has been detected in several human tumors (33,34). Its tumor-promoting effects in RCC underscore the common use of various PDGF receptor inhibitors in the treatment of patients with an advanced stage of this disease (35). CCL5 is an inflammatory chemokine, and pathologically increased activity of the CCL5/C-C chemokine receptor type 5 axis is considered to facilitate tumor progression through various mechanisms (36). The pro-tumorigenic role of CCL5, and its association with tumor pathological stage and grade, have also been demonstrated in ccRCC tissue (37). CXCL10 signaling increases cancer cell proliferation and angiogenesis (38), and its high expression in RCC tissue and association with poor survival have also been demonstrated (39,40). IL-6 is a potent proinflammatory mediator that is involved in chronic inflammation and carcinogenesis, and is responsible for cancer progression and maintenance of immune reactions (41,42). IL-6 enhances tumor cell proliferation, promotes tumor survival and the formation of metastasis (43), and has been reported to act as an important regulator of RCC pathogenesis (44).
Regarding GM-CSF, the situation appears to be more complicated compared with that of the other investigated cytokines. Significantly higher levels of this hematopoietic factor were detected in LG urine compared with in control urine, but this trend disappeared in HG urine. This may indicate that GM-CSF, in contrast to the aforementioned cytokines, could be a potential marker of localized ccRCC, and that its role in the development of ccRCC may change during tumor progression. Notably, previous studies have demonstrated that GM-CSF has a dual role. The inhibitory effects of GM-CSF on tumor growth and metastasis have been described; however, there is also evidence to suggest that, in multiple types of cancer, it acts as an immune-independent tumor promoter (45,46). Nevertheless, the present results concerning the predictive role of urinary cytokines in ccRCC are only preliminary and further investigation is needed before any definitive conclusions can be drawn.
When comparing preoperative and postoperative urine samples with control samples, the present study revealed that the urinary cytokine profile of patients changed only minimally after tumor removal. IL-1Ra, IL-6, IL-15, CCL2, CCL3, CCL4, CXCL10, GM-CSF and PDGF-BB levels were significantly higher in urine both before and after surgery compared with those in control urine. This may imply that a longer time is needed to detect any significant changes in the urinary cytokine profile of patients with ccRCC after tumor removal. Furthermore, it may be assumed that these cytokines are not only involved in the pathogenesis of ccRCC, but also in the local immune defensive and reparative processes occurring in the kidney tissue after surgical injury. For example, PDGF is known to serve a major role in almost all stages of wound healing (47) and the majority of the other aforementioned cytokines have been reported to participate in the acute phase of the inflammatory response, which peaks between the second and third day after injury (48).
There were three cytokines that reached markedly different levels in only one type of urine sample with respect to the controls, namely IL-5, VEGF and G-CSF. As the urinary levels of IL-5 and VEGF were higher on the 3rd day after surgery, it may be hypothesized that these molecules could be involved in the inflammatory and wound-healing processes occurring in the kidney after tumor resection, rather than in the pathogenesis of ccRCC. This notion seems plausible considering that IL-5 is the primary regulator of differentiation, maturation, expansion, survival and activation of eosinophils (49), which have critical roles in tissue healing and can promote angiogenesis (50,51). Thus, an increase in IL-5 shortly after surgery may be associated with the recruitment of eosinophils to the site of injury, where they participate in various reparatory processes. Regarding VEGF, this factor is considered to be the most important regulator of angiogenesis (52). Compared with other angiogenic growth factors, such as bFGF and TGF-β, VEGF has been reported to affect not only the formation of blood vessels, but also other important processes of the wound-healing cascade, such as epithelization and collagen deposition (53). The maximal VEGF activity in the wound is estimated to occur between days 3 and 7 after injury (54). As for G-CSF, its levels were lower in urine samples from patients before surgery compared with those in samples from the controls, but this trend disappeared after removal of the tumor. Based on this, it can be assumed that this hematopoietic factor may be involved in the antitumor immunity of the host. The decreased levels of G-CSF in presurgical urine samples could possibly reflect tumor-induced suppression of the immune system of the host, which, after tumor removal, was restored. However, this does not appear to be in line with its character proposed by previous studies (29,30) as aforementioned.
Regarding the diagnostic value of plasma cytokines in ccRCC, the results of the present study showed that the plasma and urine cytokine profiles of patients with ccRCC were markedly different. While in urine changes in the levels of 10 cytokines were observed, in plasma only two cytokines exhibited significant changes. This may be explained by the fact that two different control groups were used as a reference for urine and plasma analysis. Additionally, when correlation analysis was performed, no statistically significant correlation was detected between plasma and presurgical urine samples from patients. However, due to the low number of samples in the present dataset, only 11 matched plasma-urine samples could be investigated and, since the detection rate of cytokines in patient plasma was relatively low, the test could not be performed on all of the examined molecules. There are limited reports on the immune profiles of plasma and urine in ccRCC. However, Nobles et al (55), who studied a population with low levels of inflammation, did not find any correlation or association between plasma and urinary cytokine profiles. Thus, further research is warranted to accurately determine whether in patients with ccRCC the changes in the immune profile of plasma are somehow related to the changes in urine.
Regarding the potential plasma markers of ccRCC, the present study detected significantly higher levels of IL-6 in patients compared with those in the controls. This is in line with the results of previous reports. An increase in the systemic levels of IL-6 has also been observed in other patients with different types of cancer, and this feature has been revealed to be positively associated with tumor stage and negatively associated with prognosis (56). High levels of IL-6 in serum and its prognostic value have also been reported in patients with RCC (16,57–60). The current study observed lower levels of CCL2 in patient plasma than in the controls. This observation is notable, since CCL2 is mostly considered to be a tumor-promoting chemokine since it serves a critical role in the progression of various cancer types by supporting several major tumor-promoting processes (61,62). High levels of CCL2 expression and its correlation with advanced stage and recurrence have also been reported in ccRCC (63–65). Additionally, circulating CCL2 has been suggested as a marker of breast (66) and prostate (67) cancer. On the other hand, there is evidence that CCL2 is able to enhance the antitumor capability of monocytes and neutrophils (68), which could explain the present findings. It may be hypothesized that, in the case of ccRCC, CCL2 could serve a role in the immune defensive mechanisms of the host, which are suppressed by the tumor.
With regard to the analyses presented in Tables SII and SIII, the sample size used was relatively low. Therefore, none of these results were considered to be statistically and/or clinically relevant. Further research is needed to elucidate more deeply the dynamics of urinary cytokines in patients with ccRCC after surgery, particularly to determine whether the trend observed in CCL2 is also preserved even if the number of samples increases, and to identify potential markers of recurrence.
The present study has several limitations, including: i) The number of participants, and the control and patients' datasets were limited, and it would be advisable to verify the present results using a larger dataset; ii) due to a low number of samples from patients with different tumor grades, it was not possible to perform more detailed analysis of cytokine levels with respect to this clinical feature; iii) for monitoring the urinary cytokine dynamics after surgery, the time interval of sample collection appears to be insufficient and it would be advisable to increase it; iv) the control group for urine analysis consisted of different subjects compared with the control group for plasma analysis, which may have affected the present observations; and v) monitoring plasma cytokines after surgery would further increase the quality of the current study.
In conclusion, the results of the present study suggested that G-CSF, IL-1Ra, IL-6, IL-15, CCL2, CCL3, CCL4, CXCL10, GM-CSF and PDGF-BB may represent promising non-invasive biomarkers of ccRCC. G-CSF, GM-CSF, IL-6, CXCL10, CCL5 and PDGF-BB could potentially be used to determine the grade of ccRCC and consequently estimate the prognosis of patients with this cancer type. Furthermore, the levels of circulating CCL2 and IL-6 may have potential diagnostic value for ccRCC. The decrease in the urinary G-CSF and circulating CCL2 levels in patients with this tumor type requires further research.