Original Paper
Urologia Internationalis
Received: December 6, 2012 Accepted after revision: May 30, 2013 Published online: December 19, 2013
Urol Int 2014;92:270–275 DOI: 10.1159/000353401
Prognostic Impact of Intratumoral C-Reactive Protein Expression in Patients with Clear Cell Renal Cell Carcinoma Cavit Can a Mustafa Fuat Acikalin b Ata Ozen a Emine Dundar b Departments of a Urology and b Pathology, Eskişehir Osmangazi University Medical Faculty, Eskişehir, Turkey
Key Words Renal cell carcinoma · C-reactive protein · Tumoral expression · Prognosis
than patients with a negative CRP score. Conclusion: Our results suggest that the assessment of intratumoral CRP expression may be a useful tool for predicting the prognosis in patients with clear cell renal cell carcinoma. © 2013 S. Karger AG, Basel
Abstract Background: Elevated serum C-reactive protein (CRP) level has been demonstrated to predict poorer survival of both localized and metastatic renal cell carcinoma. However, the prognostic value of intratumoral CRP expression has not been sufficiently studied. Patients and Methods: In the present study, the expression of CRP was evaluated with immunohistochemical analysis in 127 patients who had undergone curative surgery for clear cell renal cell carcinoma. CRP staining was scored using the immunoreactivity score. An immunoreactivity score of 4 (median value) or lower was considered negative, a score higher than 4 was considered positive CRP expression. Results: Univariate analysis revealed that Fuhrman grade, necrosis, vascular invasion, TNM stage and CRP expression were associated with tumor-specific survival. Multivariate analysis using the Cox proportional hazards regression method revealed only TNM stage as an independent predictor of tumor-specific survival (p = 0.001). A trend towards significance (p = 0.066) was observed with CRP expression, but it did not reach significance. Patients with a positive CRP score were 3.46 times more likely to die
© 2013 S. Karger AG, Basel 0042–1138/13/0923–0270$38.00/0 E-Mail [email protected] www.karger.com/uin
Introduction
About 2% of all new cancers worldwide originate in the kidney, and clear cell renal cell carcinoma (ccRCC) is the most common subtype of adult kidney cancer [1]. In 2010, it is estimated that 58240 new RCC cases were diagnosed in the United States and 13040 deaths occurred [2]. The clinical course of ccRCC is related to some extent to a number of clinicopathological factors, including histologic tumor necrosis, sarcomatoid differentiation, Fuhrman nuclear grade and TNM staging parameters (tumor size, local tumor extension, lymph node involvement, large vessel invasion, adrenal invasion, distant metastases). Unfortunately, these factors are not entirely reliable in predicting the clinical course in patients with ccRCC. For example, there are reports that have documented the metastatic potential of even low-grade, earlystage ccRCC [3]. Therefore, it is clinically important to Dr. Mustafa Fuat Acikalin Gültepe mah. Üniversite evleri C5 Blok Daire 4 Eskişehir (Turkey) E-Mail acikalin @ ogu.edu.tr
define factors that have a significant impact on prognosis in ccRCC patients. In recent years, new molecular markers have been evaluated as potentially important prognostic markers, including hypoxia-inducible factor, carbonic anhydrase IX, vascular endothelial growth factor, p53, Bcl-2, p27, PTEN and EpCam. However, there is a strong demand to identify new factors because a better prognostic stratification of patients should lead to a more effective treatment. Recent studies showed that the evaluation of serum levels of systemic inflammatory markers, particularly C-reactive protein (CRP), may be valuable in predicting the biologic behavior of ccRCC. CRP is a constituent of normal human serum. It is an acutephase protein and a marker of systemic inflammatory response. CRP is produced mainly by hepatocytes, and its production is regulated by pro-inflammatory cytokines such as IL-1, tumor necrosis factor and IL-6 [4]. Serum levels of CRP have been found to be elevated in various disorders, including infections, inflammatory conditions and malignancies. CRP is frequently increased in patients with cancer and was found to be an indicator of a poor prognosis in various cancer types, such as esophageal, colorectal and ovarian cancer [5–7]. An elevated serum CRP level was also demonstrated to predict poorer survival of both localized and metastatic RCC [8–11]. Despite the existence of several studies investigating the relationship between serum levels of CRP and prognosis in patients with RCC, the prognostic value of intratumoral CRP expression has not been sufficiently studied. The aim of the present study was to determine the relationship of intratumoral CRP expression with clinicopathological parameters and patient prognosis in ccRCC. Patients and Methods
Immunohistochemistry All specimens had been fixed in 10% neutral buffered formalin. Sections from representative paraffin-embedded tissue blocks were cut at 4 μm. Immunohistochemical staining with monoclonal antibodies specific for CRP (Abcam, Cambridge, Mass., Ab32412, rabbit monoclonal Y284, dilution 1:500) was performed on a Benchmark staining platform (Ventana Medical Systems Inc.) with the Ventana Universal DAB detection kit (Ventana Medical Systems Inc.). The slides were then dehydrated and mounted. All sections were assessed independently by two pathologists (E.D. and M.F.A.) without knowledge of the clinicopathological features. CRP staining was scored using the immunoreactivity score (IRS) [14]. The IRS was calculated as the product of staining intensity and percentage values, leading to a scale from 0 to 12. Staining intensity was scored 0 for negative staining, 1 for barely detectable staining, 2 if positivity could be easily detected using highpowered magnification (×200), and 3 if low-powered magnification allowed clear signal detection. The percentage score was defined as 0 if 80% of cells stained positively. We used the median IRS of our cohort to separate CRP-negative and -positive cases (‘dichotomy at the median’). Therefore, an IRS ≤4 was considered negative CRP expression and an IRS >4 was considered positive CRP expression. Statistical Analysis The association between the clinicopathological features and CRP expression was estimated using χ2 tests (Fisher’s exact and Pearson). Cumulative patient survival was estimated using the Kaplan-Meier method, and for univariate analysis the log-rank test was used. Variables included in the univariate analysis were patient age, nuclear grade, tumor necrosis, vascular invasion, TNM stage and CRP score. The Cox proportional hazards regression model was used for multivariate analysis. Variables that were statistically significant in univariate analysis (nuclear grade, tumor necrosis, vascular invasion, TNM stage and CRP score) were included in the multivariate analysis. A p value 4
0.4
0.2
0 0
50
100
150
200
Survival (months)
Fig. 2. Kaplan-Meier plot for negative versus positive CRP score
(log-rank test, p < 0.001).
Table 2. Relationship between CRP expression and clinicopatho-
logical variables Variable
Age, years ≤50 >50 Fuhrman grade G1+G2 G3+G4 Necrosis Negative Positive Vascular invasion Negative Positive TNM stage I II III IV
CRP expression
p
negative
positive
25 55
17 30
59 21
13 34
68 12
17 30
70 10
34 13
40 19 11 10
18 7 14 8
0.696
Urologia Internationalis
Received: December 6, 2012 Accepted after revision: May 30, 2013 Published online: December 19, 2013
Urol Int 2014;92:270–275 DOI: 10.1159/000353401
Prognostic Impact of Intratumoral C-Reactive Protein Expression in Patients with Clear Cell Renal Cell Carcinoma Cavit Can a Mustafa Fuat Acikalin b Ata Ozen a Emine Dundar b Departments of a Urology and b Pathology, Eskişehir Osmangazi University Medical Faculty, Eskişehir, Turkey
Key Words Renal cell carcinoma · C-reactive protein · Tumoral expression · Prognosis
than patients with a negative CRP score. Conclusion: Our results suggest that the assessment of intratumoral CRP expression may be a useful tool for predicting the prognosis in patients with clear cell renal cell carcinoma. © 2013 S. Karger AG, Basel
Abstract Background: Elevated serum C-reactive protein (CRP) level has been demonstrated to predict poorer survival of both localized and metastatic renal cell carcinoma. However, the prognostic value of intratumoral CRP expression has not been sufficiently studied. Patients and Methods: In the present study, the expression of CRP was evaluated with immunohistochemical analysis in 127 patients who had undergone curative surgery for clear cell renal cell carcinoma. CRP staining was scored using the immunoreactivity score. An immunoreactivity score of 4 (median value) or lower was considered negative, a score higher than 4 was considered positive CRP expression. Results: Univariate analysis revealed that Fuhrman grade, necrosis, vascular invasion, TNM stage and CRP expression were associated with tumor-specific survival. Multivariate analysis using the Cox proportional hazards regression method revealed only TNM stage as an independent predictor of tumor-specific survival (p = 0.001). A trend towards significance (p = 0.066) was observed with CRP expression, but it did not reach significance. Patients with a positive CRP score were 3.46 times more likely to die
© 2013 S. Karger AG, Basel 0042–1138/13/0923–0270$38.00/0 E-Mail [email protected] www.karger.com/uin
Introduction
About 2% of all new cancers worldwide originate in the kidney, and clear cell renal cell carcinoma (ccRCC) is the most common subtype of adult kidney cancer [1]. In 2010, it is estimated that 58240 new RCC cases were diagnosed in the United States and 13040 deaths occurred [2]. The clinical course of ccRCC is related to some extent to a number of clinicopathological factors, including histologic tumor necrosis, sarcomatoid differentiation, Fuhrman nuclear grade and TNM staging parameters (tumor size, local tumor extension, lymph node involvement, large vessel invasion, adrenal invasion, distant metastases). Unfortunately, these factors are not entirely reliable in predicting the clinical course in patients with ccRCC. For example, there are reports that have documented the metastatic potential of even low-grade, earlystage ccRCC [3]. Therefore, it is clinically important to Dr. Mustafa Fuat Acikalin Gültepe mah. Üniversite evleri C5 Blok Daire 4 Eskişehir (Turkey) E-Mail acikalin @ ogu.edu.tr
define factors that have a significant impact on prognosis in ccRCC patients. In recent years, new molecular markers have been evaluated as potentially important prognostic markers, including hypoxia-inducible factor, carbonic anhydrase IX, vascular endothelial growth factor, p53, Bcl-2, p27, PTEN and EpCam. However, there is a strong demand to identify new factors because a better prognostic stratification of patients should lead to a more effective treatment. Recent studies showed that the evaluation of serum levels of systemic inflammatory markers, particularly C-reactive protein (CRP), may be valuable in predicting the biologic behavior of ccRCC. CRP is a constituent of normal human serum. It is an acutephase protein and a marker of systemic inflammatory response. CRP is produced mainly by hepatocytes, and its production is regulated by pro-inflammatory cytokines such as IL-1, tumor necrosis factor and IL-6 [4]. Serum levels of CRP have been found to be elevated in various disorders, including infections, inflammatory conditions and malignancies. CRP is frequently increased in patients with cancer and was found to be an indicator of a poor prognosis in various cancer types, such as esophageal, colorectal and ovarian cancer [5–7]. An elevated serum CRP level was also demonstrated to predict poorer survival of both localized and metastatic RCC [8–11]. Despite the existence of several studies investigating the relationship between serum levels of CRP and prognosis in patients with RCC, the prognostic value of intratumoral CRP expression has not been sufficiently studied. The aim of the present study was to determine the relationship of intratumoral CRP expression with clinicopathological parameters and patient prognosis in ccRCC. Patients and Methods
Immunohistochemistry All specimens had been fixed in 10% neutral buffered formalin. Sections from representative paraffin-embedded tissue blocks were cut at 4 μm. Immunohistochemical staining with monoclonal antibodies specific for CRP (Abcam, Cambridge, Mass., Ab32412, rabbit monoclonal Y284, dilution 1:500) was performed on a Benchmark staining platform (Ventana Medical Systems Inc.) with the Ventana Universal DAB detection kit (Ventana Medical Systems Inc.). The slides were then dehydrated and mounted. All sections were assessed independently by two pathologists (E.D. and M.F.A.) without knowledge of the clinicopathological features. CRP staining was scored using the immunoreactivity score (IRS) [14]. The IRS was calculated as the product of staining intensity and percentage values, leading to a scale from 0 to 12. Staining intensity was scored 0 for negative staining, 1 for barely detectable staining, 2 if positivity could be easily detected using highpowered magnification (×200), and 3 if low-powered magnification allowed clear signal detection. The percentage score was defined as 0 if 80% of cells stained positively. We used the median IRS of our cohort to separate CRP-negative and -positive cases (‘dichotomy at the median’). Therefore, an IRS ≤4 was considered negative CRP expression and an IRS >4 was considered positive CRP expression. Statistical Analysis The association between the clinicopathological features and CRP expression was estimated using χ2 tests (Fisher’s exact and Pearson). Cumulative patient survival was estimated using the Kaplan-Meier method, and for univariate analysis the log-rank test was used. Variables included in the univariate analysis were patient age, nuclear grade, tumor necrosis, vascular invasion, TNM stage and CRP score. The Cox proportional hazards regression model was used for multivariate analysis. Variables that were statistically significant in univariate analysis (nuclear grade, tumor necrosis, vascular invasion, TNM stage and CRP score) were included in the multivariate analysis. A p value 4
0.4
0.2
0 0
50
100
150
200
Survival (months)
Fig. 2. Kaplan-Meier plot for negative versus positive CRP score
(log-rank test, p < 0.001).
Table 2. Relationship between CRP expression and clinicopatho-
logical variables Variable
Age, years ≤50 >50 Fuhrman grade G1+G2 G3+G4 Necrosis Negative Positive Vascular invasion Negative Positive TNM stage I II III IV
CRP expression
p
negative
positive
25 55
17 30
59 21
13 34
68 12
17 30
70 10
34 13
40 19 11 10
18 7 14 8
0.696
