What's known on the subject? and What does the study add?
The exposed proliferation antigen 210 (XPA-210) of the proliferation marker thymidine kinase 1 (TK1) showed higher expression levels in metastatic renal cell carcinoma.
The current study used a new XPA-210 antibody to clarify the role of TK1 tissue expression in the largest reported cohort of different renal cell carcinoma types and oncocytomas.
• To determine the clinical role of the exposed proliferation antigen 210 (XPA-210) of the proliferation marker thymidine kinase 1 (TK1) in a large cohort of different renal cell carcinoma (RCC) types, oncocytomas and normal renal tissues samples, as TK1 is reported to be of clinical significance in several cancer entities and is suggested as a prognostic serum biomarker for RCC.
PATIENTS AND METHODS
• Expressions of XPA-210 were determined immunohistochemically in 40 clear cell RCCs (ccRCC), 25 papillary RCCs (papRCC), 17 chromophobe RCC (chRCC), 27 oncocytomas and 64 normal renal parenchyma paraffin-embedded specimens.
• Immunohistochemistry was performed with a monoclonal anti-XPA-210 antibody. Staining was measured by the percentage of positive cells.
• Expression was compared between subgroups and correlated with respective clinical data using one-way analysis of variance with post hoc Tukey-Kramer analyses.
• XPA-210 staining in the RCC subgroup was significantly different from the oncocytomas (mean [sem] 4.1 [0.4] vs 2.2 [0.4]; P = 0.004) and from normal renal tissue (1.0 [0.1]; P < 0.001], whereas oncocytomas did not differ from normal renal parenchyma staining (P = 0.18).
• Subdivided into RCC groups, only ccRCC (mean [sem] 5.1 [0.6]; P < 0.001) and papRCC (4.4 [0.6]; P < 0.001) varied from normal renal parenchyma, whereas chRCC (1.4 [0.3]; P = 0.99) did not.
• RCC XPA-210 staining was significantly associated with higher tumour stage (T = 3, P = 0.002) and grade (G = 3, P = 0.001).
• The malignant character of RCC is reflected by higher XPA-210 expression as compared with oncocytomas and normal kidney.
• The ccRCC and papRCC subgroups had higher XPA-210 levels.
• XPA-210 could be considered a potential marker for the assessment of the proliferative activity in primary RCC.
exposed proliferation antigen centred around amino acid 210
(clear cell) (papillary) (chromophobe) RCC.
RCC is the most frequent kidney tumour lesion with ≈30 000 newly diagnosed patients within the European Union and over 38 000 new cases in the USA. Annually ≈15 000 (European Union) and ≈12 000 (USA) disease-related deaths are reported . More recently, new targeted therapies have increased in popularity showing improved response rates and prolonged progression-free survival times when compared with cytokine-based therapeutic approaches . However, distinct molecular markers in metastatic RCC are needed to identify patients at risk of progression and to monitor response to antineoplastic therapies targeting complex signalling pathways. One candidate marker thymidine kinase 1 (TK1) is expressed in normal proliferating cells during the S-phase and has been investigated in different malignant tissues as a measure for cancer cell proliferation. Serum TK1 expression has shown clinical significance in breast [3–5], lung [4,6], colorectal , and bladder cancer . Consistent with these findings Mizutani et al.  reported positive correlations between TK1 tissue activity, tumour stage, grade and sensitivity to 5-fluorouracil treatment in patients with RCC. Contrarily, there are other reports of higher expression levels of TK1 in normal non-proliferating kidney tubules compared with RCC cells . The exposed proliferation antigen centred around amino acid 210 (XPA-210) is contained only in intact TK1  and has already been proven to be useful in the determination of TK1 in tissue and serum analysis [3,9,11].
To further clarify the role of TK1 tissue expression in a large cohort of different RCC subtypes, oncocytomas and normal renal tissues the present study used a monoclonal anti-XPA-210 antibody with enhanced epitope specificity in conjunction with high throughput tissue microarray (TMA) analysis.
PATIENTS AND METHODS
In all, 173 tumour and normal paraffin-embedded tissue specimens from 109 patients (40 clear cell RCC [ccRCC], 25 papillary RCC [papRCC], 17 chromophobe RCC [chRCC], 27 oncocytomas) who underwent nephrectomy or nephron-sparing surgery for RCC at the University of Tuebingen, Germany between May 2002 and October 2007 were included in this analysis. By histology, 36.7% of tumours were ccRCC, 22.9% papRCC, 15.6% chRCC, and the remaining 24.8% were oncocytomas. Informed consent was obtained according to procedures approved by the Institutional Review Board of the University of Tuebingen. The mean age of the 109 cases was 62.9 years and 60% were male. Most of the RCC cases (including all histological subgroups) had pT1 (69.5%) or pT3 (29.3%). There was lymph node involvement in 10 patients with RCC (12.2%), whereas 72 (87.8%) had no evidence of distant metastatic disease. Nuclear grading according to the Fuhrman classification was G1 in 11 (13.4%) patients, G2 in 60 (73.2%), and G3 in the rest (13.4%). Samples from pathologically representative tumour regions and benign renal tissues were obtained. Specimens were classified according to the seventh edition of the American Joint Cancer Committee/Union Internationale Contre le Cancer (AJCC/UICC) TNM system. The patient and tumour characteristics are given in Table 1. After an independent second histological evaluation of haematoxylin and eosin-stained slides, the TMA slides (Fig. 1) were prepared in the laboratory (U.V.) as described previously [12,13].
Table 1. Characteristics of 109 patients with different renal tumours analysed for XPA-210 protein tissue expression
ccRCC (n = 40)
papRCC (n = 25)
chRCC (n = 17)
Oncocytoma (n = 27)
Age at surgery, years
Tumour diameter, cm
Lymph node involvement
Immunostaining was manually performed using a commercially available monoclonal anti-XPA-210 mouse antibody (Clone 5, AroCell, Uppsala, Sweden). Tissues were deparaffinized by passing the specimens through xylene and rehydrated though serial dilutions of ethanol (100%, 96% and 70%). Endogenous peroxidase was blocked with 3% H2O2 in Aqua dest. for 30 min. Antigen retrieval was accomplished by incubating the sections for 3 × 5 min in 10 mm citrate buffer at 95 °C. To block non-specific binding sites, sections were incubated in antibody diluent with background reducing components (Dako S3022, Dako Cytomation, Glostrup, Sweden). The primary antibody was used in a dilution of 1 : 250 in Dako S3022 and sections were incubated for 60 min at room temperature. The EnVision G/2 System/AP, Rabbit/Mouse (Permanent Red, Dako Cytomation) was used for detection before sections were counterstained with haematoxylin and mounted.
The anti-XPA-210 antibody was tested for specific staining in competition with the free synthetic peptide antigen (corresponding to the amino acids 195–225 of the human TK1 protein sequence; AroCell). Tonsillar tissue was used for positive control and for negative control the primary antibody was omitted in the experiments. TMA slides were evaluated in a ‘blinded’ manner by two independent investigators. Staining was quantitatively given as the number of positive cells per 100 cells. Divergent results were re-evaluated. Expression was compared between subgroups and renal tissue and correlated with respective clinical data. Statistical analysis was carried out using one-way anova with post hoc Tukey-Kramer analyses; P < 0.05 was considered to indicate statistical significance.
There were no significant differences in the patients' characteristics (age and gender). Significant differences in tumour characteristics were only found for the frequency of systemic spread between eight distant metastasis in ccRCC, two in papRCC and no metastatic disease in chRCC group (P = 0.05, Table 1). In normal kidney parenchyma (proximal tubule, distal tubule and glomerulus) XPA-210 expression was low. XPA immunoreactivity was infrequently found in the cytoplasm of proximal and distal tubule cells. Oncocytoma and tumour tissue predominantly showed cytoplasmic XPA-210 expression. Representative results of XPA-210 immunostaining in different renal tissue samples are shown in Fig. 1. Overall staining results (mean number of positive cells/100cells [sem]) differed significantly between the RCC (4.1 [0.4]) vs the oncocytoma (2.2 [0.4]; P = 0.004) and the normal renal tissue groups (1.1 [0.1]; P < 0.001, Fig. 2). There were no differences between oncocytomas and normal renal tissue expression (P = 0.18). Further histological RCC subgroup analysis (Fig. 3) showed significant staining variations between ccRCCs (mean [sem] 5.1 [0.6]; P < 0.001), papRCCs (4.4 [0.6]; P < 0.001) and normal renal parenchyma. However, chRCC (1.4 [0.3]; P = 0.99) showed no significant difference. Overall RCC staining results were significantly associated with advanced T3 tumour stages (5.9 [0.9]) vs limited stages (3.4 [0.4]; P = 0.002) and high- (7.9 [1.5]) vs low-grade tumours (3.5 [0.3]; P = 0.001). Tumour stage and grade correlations are shown in Fig. 4.
TK1 is part of the salvage pathway of deoxynucleotide metabolism, and thus closely linked to cell proliferation. TK1 activity is low in resting cells and starts to rise between the late G1 and S-phase. One promising cell proliferation marker, named XPA-210, is based on a specific epitope derived from TK1. Serum TK1 activity and concentration have been used to monitor the relative extent of tumour metastasis and prognosis in various human cancers. Chen et al.  reported a correlation between serum TK1 expression in 1247 patients with cancer (lung, oesophagus, gastric, head and neck, and thyroid cancer) to pathological stage and grade. Serum TK1 activity was already used for risk estimation of tumour spread and prognostic implications in patients with leukaemia . Similar findings have been reported for serum TK1 concentration in patients with breast cancer with recurrent disease . In bladder cancer, serum concentrations of TK1 are also strongly associated with tumour stage and grade at initial diagnosis . In addition, the use of serum TK1 protein level was evaluated in a health screening programme including 11 880 patients undergoing parallel medical examination. In the serum TK1-positive screening population significantly more persons with non-malignant proliferation of breast and prostate tissue were detected. However, only one malignant case was found in the serum TK1-positive group, whereas there were no malignant cases in the TK1-negative candidates . A recent study confirmed higher preoperative serum TK1 levels in 116 patients with RCC compared with 20 healthy control subjects. Furthermore, serum TK1 activity correlated with T stage and lower serum TK1 was associated with extensive tumour necrosis . In conclusion, Luo et al.  showed increased serum TK1 activity with higher grade, stage and RCC tumour size. However, there was relationship between serum TK1 activity, serum TK1 protein concentration, and increased TK1 expression in RCC tumour sections. Similarly, He et al.  reported only slight correlations between serum TK1 concentration and activity levels in nine different cancer types. The authors concluded that serum TK1 concentration is a more sensitive tumour marker in solid malignancies than is serum TK1 activity. In fresh-frozen tumour tissue from 66 patients with RCC, TK1 activity was correlated with increased tumour stage, grade and sensitivity to 5-fluorouracil treatment . Another group reported higher TK1 protein levels in proximal tubule cells compared with RCC specimens using immunohistochemistry and western blot analysis with two different antibodies directed against XPA-161 and XPA-210. The reported higher TK1 expression of kidney tubule cells was partially explained by an enhanced proliferative capacity .
Recent findings in our group showed higher XPA-210 levels in metastatic than in localized ccRCC disease and were able to distinguish patients with metastatic RCC without bone involvement from those with solitary osseous lesions. However, XPA-210 immunostaining showed abundant nuclear expression in ccRCC tissues eventually caused by a lower epitope specificity of the used polyclonal XPA-210 antibody variant . Furthermore, the aforementioned conflicting results for TK1 tissue and serum analysis can be partially explained by the huge variety of measurement methods. Whereas radioactively labelled substrates are mainly used for measuring serum TK1 activity [18,19], anti-TK1 antibodies have been developed to determine serum and tissue TK1 concentrations [20–23]. The most specific and sensitive TK1 antibodies rely on the XPA-210 peptide in the C-terminal part of TK1 [10,22,24]. The present study used a new monoclonal XPA-210 antibody to detect TK1 expression in the largest reported cohort of different RCC types, oncocytomas compared with normal renal parenchyma. In addition to that, we used high-volume TMA analysis to investigate the clinical compatibility and the use of XPA-210 as a potential marker to evaluate tumour proliferation. There were higher XPA-210 expressions in ccRCC and papRCC tissue with higher clinical stage and grade. In agreement with previously reported data, the immunoreactivity was predominantly found in the cytoplasm . XPA-210 expression was significantly increased in malignant renal tumours compared with the oncocytoma (P = 0.004) and the normal renal tissue group (P < 0.001). Statistical analysis showed no differences between oncocytomas and normal renal tissue staining. The examination of different histological subgroups showed significant staining variations between ccRCCs as well as papRCCs compared with normal renal parenchyma. Only chRCCs showed no significant differences for XPA-210 staining from normal samples as well as from renal oncocytomas. The histopathological differentiation of oncocytomas and chRCCs is sometimes difficult and standardized molecular markers are still missing. However, the differential diagnosis between benign oncocytoma, the most common benign epithelial neoplasm of the kidney and chRCC with the potential risk of systemic spread is fundamental for clinical and surgical management. Based on these results the use of XPA-210 in patients with chRCC has to be critically discussed. Interestingly, there was a trend, albeit not statistically significant, towards an increasing XPA-210 expression in RCCs up to a diameter of 7 cm and a subsequent XPA-210 decrease >7 cm. Variations can be explained by higher proliferation rates during tumour growth followed by lower proliferative activity due to insufficient blood supply or the beginning of necrosis.
XPA-210 showed the potential to quantify proliferative activity and might therefore contribute to a better understanding of the signalling pathways in RCC. To date, the individual biological aggressiveness of RCC is almost unpredictable. Considering this, XPA-210 may provide additional information for characterization of patient-risk profiles and determination of individual therapy. In this context, measuring XPA-210 within serum diagnostics especially compared with TK1 tissue results in RCC might be of further interest. The present RCC tissue protein expression data could serve as a basis to propose serum-based analysis as the main application of XPA-210 in future clinical routine.
We thank Anne-Charlotte Aronsson (AroCell, Uppsala, Sweden) for providing the synthetic XPA-210 peptide antigen.