not logged in

European Urology

Journal Article Page

Jump to

European Urology

Volume 61, issue 5, pages e41-e52, May 2012

Kidney Cancer

Clinical, Molecular, and Genetic Correlates of Lymphatic Spread in Clear Cell Renal Cell Carcinoma

Nils Kroeger a b, David B. Seligson c, Tobias Klatte a, Edward N. Rampersaud a, Frédéric D. Birkhäuser a, P. Nagesh Rao c, John T. Leppert a, Nazy Zomorodian a, Fairooz F. Kabbinavar a d, Arie S. Belldegrun a and Allan J. Pantuck a lowast

Accepted 9 January 2012, Published online 18 January 2012, pages 888 - 895

Full Text Full-Text PDF (0,9 MB) Create Platinum Slide Series

Abstract

Background

While it is well known that clear cell renal cell carcinoma (ccRCC) that presents with lymphatic spread is associated with an extremely poor prognosis, its molecular and genetic biology is poorly understood.

Objective

Define the clinicopathologic, molecular, and genetic biological characteristics of these tumors in comparison to nonmetastatic (N0M0) renal cell carcinomas.

Design, setting, and participants

A retrospective study defined clinicopathologic features, expression of 28 molecular markers, and occurrence of chromosomal aberrations for their correlation with lymphatic spread in three cohorts of 502, 196, and 272 patients, respectively.

Measurements

Fisher exact test or the χ2 test were used to compare categorical variables; continuous variables were compared with the Mann-Whitney U test or student t test. Cut-off values were calculated based on receiver operating characteristic curves and the Youden Index. Uni- and multivariate regression analyses were used to investigate the correlation with lymphatic spread.

Results and limitations

In clinical analyses, a predictive model consisting of smoking history (p = 0.040), T stage (p < 0.0001), Fuhrman grade (p < 0.0001), Eastern Cooperative Oncology Group performance status (p < 0.0001), and microvascular invasion (p < 0.0001) was independently associated with lymphatic spread. After adjustment with these clinical variables, low carbonic anhydrase IX (CAIX) (p = 0.043) and high epithelial vascular endothelial growth factor receptor 2 (p = 0.033) protein expression were associated with a higher risk of lymphatic spread, and loss of chromosome 3p (p < 0.0001) with a lower risk. The current study is limited by its retrospective design, small sample size, and single-center experience.

Conclusions

The low rates of CAIX expression and loss of chromosome 3p suggest that lymphatic spread in ccRCC occurs independently of von Hippel-Lindau tumor suppressor inactivation.

Take Home Message

Lymph node–positive clear cell renal cell carcinomas (RCCs) are more likely to be wild-type von Hippel-Lindau tumors, to express low carbonic anhydrase IX, and to have the worst prognosis and are poor immunotherapy responders, like non–clear cell RCCs.

Keywords: Renal cell carcinoma, Lymph nodes, CAIX, VEGF, VEGF receptor, Loss 3p, VHL gene.

Article Outline

1. Introduction

According to the World Health Organization, renal cell carcinoma (RCC) accounted for 116 500 deaths in 2008 [1]. Although the incidence of localized (N0M0) RCCs has risen [2], 20–30% of patients still present with metastatic disease [3], and despite the implementation of effective targeted drug treatments, metastatic disease is still associated with a grim prognosis [4]. This is particularly true for RCCs having lymph node involvement, which is an entity reported to have particularly poor survival [5] and [6]. For example, the US National Institutes of Health reported a median overall survival of 14.7 mo for patients without lymph nodes compared to 8.5 mo for patients with lymph nodes [7]. Likewise, in data from the University of California, Los Angeles (UCLA) [8], the presence of positive lymph nodes in M1 patients was associated with larger, higher-grade primary tumors that were more commonly associated with sarcomatoid features. Moreover, N + M1 patients had a median survival of 10.5 mo compared with 20.4 mo for N0M1 patients, respectively. The median survival of N0M1 patients was improved to 28 mo in those who received adjunctive immunotherapy, while the median survival of patients with N + M1 disease was the same in those treated with and those treated without adjunctive immunotherapy. While these are well known clinical observations, the biological background that underlies the poor outcomes of lymphatic spread in RCC is still poorly understood. We hypothesized that clear cell RCCs (ccRCCs) with lymphatic spread represent a unique subtype of ccRCC, having their own definable clinical, genetic, and molecular features. To investigate this hypothesis, we analyzed the clinicopathologic features, protein expression profiles, and chromosomal aberrations that were independently associated with node-positive ccRCC.

2. Materials and methods

2.1. Clinical analysis

We first aimed to examine a broad spectrum of clinicopathologic features to identify those associated with lymphatic spread. Therefore, 502 patients treated for RCC between 1989 and 2007 were extracted from the institutional review board-approved UCLA kidney cancer database. Inclusion criteria were a full dataset in terms of age, gender, Eastern Cooperative Oncology Group performance status (ECOG PS) [9], smoking history, and comorbidities such as hypertension, diabetes mellitus, chronic obstructive pulmonary disease (COPD), and coronary artery disease. Pathologic data included TNM stages [10], tumor size, Fuhrman grade [11], and the presence of microvascular invasion (MVI) or macrovascular invasion (MAVI), collecting duct invasion (CDI), multifocality, and sarcomatoid features.

N-stage status was defined according to pathologic findings or by clinical data when applicable. Clinically, lymphadenopathy was defined as enlarged (≥1 cm3) hilar or retroperitoneal lymph nodes on preoperative computerized tomography or magnetic resonance imaging. All patients underwent a minimum of an ipsilateral hilar lymph node dissection (LND) in the absence of clinically suspicious nodes. The decision to perform an extended LND was based on the finding of abnormal lymph nodes detected on preoperative imaging or at the time of surgery, a practice that historically has resulted in only an 8% incidental positive lymph node rate [12]. The extent of LND was determined by individual surgeon practice without a standardized institutional template. All positive lymph nodes were confirmed by histologic examination. Data on number of lymph nodes removed and their rate of positivity are not available in our database.

2.2. Molecular analysis

Secondary investigation using a tissue microarray (TMA) was performed to evaluate molecular characteristics. The study subcohort consisted of 196 randomly selected ccRCC patients who underwent nephrectomy at UCLA between 1989 and 2000. For molecular analysis, we chose markers relevant to RCC biology as part of the hypoxia-inducible factor (HIF) pathway (HIF-1α, carbonic anhydrase IX [CAIX], vascular endothelial growth factor [VEGF]-A, VEGF-C, VEGF-D, and VEGF-R1-3) and the mammalian target of rapamycin (mTOR) pathway (p27, pAkt, pS6, phosphatase and tensin homolog [PTEN]) or have otherwise been demonstrated to be important prognostic markers (CAXII, Ki67, p53, p21, Gelsolin, EpCam, Vimentin, and chemokine (C-X-C motif) receptor 3 [CXCR3]) in RCC.

Three core-tissue biopsies, 0.6 mm in diameter, were taken from selected morphologically representative regions of formalin-fixed, paraffin-embedded, primary tumor specimens and arrayed using methods described previously [13]. Immunohistochemical staining was performed as described previously [14] using appropriate positive and negative controls. A single pathologist (D.B.S.) blinded to clinicopathologic variables and clinical outcome performed quantitative assessment for each antibody. The extent of expression (staining frequency) was recorded as a percentage of the entire tumor sample that stained positive as a pooled mean of three spots. Staining intensity was not considered because in prior experience using our TMA, staining intensity appears to be a subjective variable influenced by fixation protocols, tissue types, and storage time of tissue specimens [15].

2.3. Cytogenetic analysis

Tumor samples were collected immediately after nephrectomy from 272 ccRCC tumors treated between 1989 and 2007 and aseptically minced into 2- to 3-mm pieces. The detailed tissue preparation protocol was described previously [16]. From each tumor, 20 metaphases were analyzed in accordance with the International System for Human Cytogenetic Nomenclature [17] by one clinical cytogeneticist (P.N.R.).

2.4. Statistical analysis

The clinicopathologic features and chromosomal aberrations in a single group of lymph node–positive ccRCCs (N + M0) (26, 8, and 6 in the clinical, molecular, and cytogenetic analyses respectively) or N + M1 (73, 30, and 22 in the clinical, molecular, and cytogenetic analyses, respectively) were compared to another group of lymph node–negative (N0M0) tumors using the χ2 test and Fisher exact test for categorical variables and student t test for continuous variables. In TMA analysis, protein expressions of lymphatic RCCs were compared to those for N0M0 tumors using the Mann-Whitney U test. Cut-off values were calculated based on receiver operating characteristic (ROC) curves and the Youden Index.

Univariate and multiple logistic regression analyses were used to test the independent association of clinical, molecular, and cytogenetic variables with lymphatic spread. For univariate analysis, only those variables that occurred in N+ patients with significant differences or had statistically different expression compared to N0M0 patients were investigated. Clinicopathologic features that demonstrated an independent correlation in univariate logistic regression analysis were further evaluated by adjustment in multiple logistic regression analysis including molecular and cytogenetic variables. For calculation of multiple logistic regression models, a backward stepwise selection with the likelihood ratio criterion (including and exclusion criteria were p≤ 0.05 and p≥ 0.10, respectively) was used. The rank of elimination was given when a variable was removed from the equation, and the odds ratio (OR), 95% confidence interval (CI), and the p value for the removed variable were obtained during the removal step. All variables were investigated as continuous variables to avoid overfitting of the models. The statistical tests were two-sided and performed at a significance level of 0.05 using the PASW (SPSS) v.18.0 software package (IBM Corp., Armonk, NY, USA).

3. Results

3.1. Clinical analysis

The comparisons of patient and tumor characteristics are summarized in Table 1. Tumor size (OR: 1.176; 95% CI, 1.113–1.243; p < 0.0001); sarcomatoid features (OR: 6.618; 95% CI, 2.741–15.976; p < 0.0001); multifocality (OR: 2.039; 95% CI, 1.153–3.605; p = 0.014); hypertension (OR: 0.588; 95% CI, 0.368–0.939; p = 0.026); smoking history (OR: 2.012; 95% CI, 1.290–3.196; p = 0.002); T stage (OR: 4.427; 95% CI, 3.169–6.138; p < 0.0001); Fuhrman grade (OR: 3.103; 95% CI, 2.267–4.248; p < 0.0001); ECOG PS (OR: 2.521; 95% CI, 1.722–3.689; p < 0.0001); CDI (OR: 1.858; 95% CI, 1.057–3.267; p = 0.031); MAVI (OR: 2.515; 95% CI, 1.502–4.211; p < 0.0001); and MVI (OR: 5.795; 95% CI, 3.032–11.077; p < 0.0001) all showed statistically significantly associations with lymphatic spread in univariate logistic regression analysis. The results of multiple logistic regression analysis are summarized in Table 2.

Table 1 Patients’ clinical and pathologic characteristics

Localized Patients, no. (%) Lymphatic Patients, no. (%) p value Test
Age, yr
 mean ± SD 59.79 ± 12.79 58.15 ± 11.90 0.248 Student t
Gender
 Female 176 (43.7) 38 (38.4) 0.366 Fisher exact
 Male 227 (56.3) 61 (61.6)
CAD
 No 352 (87.3) 89 (89.9) 0.607 Fisher exact
 Yes 51 (12.7) 10 (10.1)
Hypertension
 No 227 (56.3) 68 (68.7) 0.030 Fisher exact
 Yes 176 (43.7) 31 (31.3)
Diabetes mellitus
 No 352 (87.3) 89 (89.9) 0.607 Fisher exact
 Yes 51 (12.7) 10 (10.1)
COPD
 No 393 (97.5) 95 (96.0) 0.492 Fisher exact
 Yes 10 (2.5) 4 (4.0)
Smoking history
 No 260 (64.5) 47 (47.5) 0.003 Fisher exact
 Yes 143 (35.5) 52 (52.5)
Tumor size, cm, mean ± SD 6.22 ± 3.82 8.93 ± 3.64 <0.0001 Student t
T stage
 T1 230 (57.1) 6 (6.1)
 T2 63 (15.6) 11 (11.1) Chi-square
 T3 104 (25.8) 72 (72.7) <0.0001
 T4 6 (1.5) 10 (10.1)
Organ-confined (T1 + 2) tumor 293 (72.7) 17 (17.2) <0.0001 Fisher exact
vs advanced (T3 + 4) tumor 110 (27.3) 82 (82.8)
Fuhrman grade
 I 78 (19.3) 3 (3.0)
 II 211 (52.4) 34 (34.3) Chi-square
 III 102 (25.3) 42 (42.4) <0.0001
 IV 12 (3.0) 20 (20.2)
ECOG PS
 0 239 (59.3) 35 (35.4)
 1 156 (38.7) 56 (56.6) Chi-square
 2 8 (2.0) 6 (6.1) <0.0001
 3 0 (0.0) 2 (2.0)
Sarcomatoid features
 No 394 (97.8) 86 (86.9) <0.0001 Fisher exact
 Yes 9 (2.2) 13 (13.1)
CDI
 No 352 (87.3) 78 (78.8) 0.037 Fisher exact
 Yes 51 (12.7) 21 (21.2)
MAVI
 No 346 (85.9) 70 (70.7) 0.001 Fisher exact
 Yes 57 (14.1) 29 (29.3)
MVI
 No 383 (95.0) 76 (76.8) <0.0001 Fisher exact
 Yes 20 (5.0) 23 (23.2)
Multifocality
 No 356 (88.3) 78 (78.8) 0.021 Fisher exact
 Yes 47 (11.7) 21 (21.2)

SD = standard deviation; CAD = coronary artery disease; COPD = chronic obstructive pulmonary disease; ECOG PS = Eastern Cooperative Oncology Group performance status; CDI = collecting duct invasion; MAVI = macrovascular invasion; MVI = microvascular invasion.

Table 2 Clinical analysis multiple logistic regression: association of clinicopathologic features with lymphatic spread

Factor Rank of elimination from analysis OR 95% CI p value
Smoking history – 1.791 1.028 3.122 0.040
T stage – 3.732 2.591 5.376 <0.001
Fuhrman grade – 2.012 1.383 2.927 <0.001
ECOG PS – 1.828 1.143 2.924 <0.001
MVI – 6.359 2.739 14.763 <0.001
Hypertension 6 0.633 0.350 1.146 0.131
Tumor size 2 1.018 0.944 1.098 0.991
CDI 1 1.031 0.512 2.075 0.973
Multifocality 4 1.351 0.640 2.854 0.430
Sarcomatoid features 5 2.028 0.643 6.395 0.227
MAVI 3 0.793 0.408 1.542 0.494

OR = odds ratio; CI = confidence interval; ECOG PS = Eastern Cooperative Oncology Group performance status; MVI = microvascular invasion; CDI = collecting duct invasion; MAVI = macrovascular invasion.

3.2. Molecular analysis

Compared to N0M0 tumors, an expression pattern characterized by low expression of CAIX, CAXII, and endothelial VEGFR-3, and high expression of Ki67, epithelial VEGF-A, epithelial VEGFR-1, epithelial VEGFR-2, and pS6 was associated with lymphatic tumors (Fig. 1), as follows (percentages represent the mean positive staining within each group plus or minus the standard deviation):

  • • CAIX: 84.93% ± 32.81% versus 68.59% ± 41.34%; p = 0.005
  • • CAXII: 93.80% ± 21.60% versus 81.58% ± 34.02%; p = 0.001
  • • Endothelial VEGFR-3: 34.03% ± 27.49% versus 18.05% ± 17.51; p = 0.001
  • • Ki67: 6.39% ± 8.07% versus 12.99% ± 14.13%; p = 0.002
  • • Epithelial VEGF-A: 33.55% ± 27.72% versus 47.64% ± 30.74%; p = 0.013
  • • Epithelial VEGFR-1: 48.98% ± 29.99% versus 67.75% ± 27.78%; p = 0.001
  • • Epithelial VEGFR-2: 32.78% ± 29.89% versus 40.38% ± 26.09%; p = 0.041
  • • pS6: 35.23% ± 35.24% versus 58.50% ± 32.39%; p = 0.0001.
gr1

Fig. 1 Comparison of protein expression of localized clear cell renal cell carcinoma versus lymphatic spread. Error bars represent the standard deviation.* p < 0.05.** p < 0.001.

ROC analyses were conducted to calculate cut-off values that best discriminated between lymphatic and N0M0 tumors. Appropriate cut-off values were CAIX <92%, CAXII <96%, endothelial VEGFR-3 <28%, Ki67 >9%, epithelial VEGF-A >56%, epithelial VEGFR-1 >57%, epithelial VEGFR-2 >17%, and pS6 >28%.

Univariate logistic regression demonstrated an association of low CAIX (OR: 3.656; 95% CI, 1.708–7.829; p = 001); low CAXII (OR: 4.354; 95% CI, 1.908–9.936; p < 0.0001); low endothelial VEGFR-3 (OR: 4.100; 95% CI, 1.769–9.501; p = 0.001); high epithelial VEGF A (OR: 2.841; 95% CI, 1.339–6.026; p = 0.007); high epithelial VEGFR-1 (OR: 3.567; 95% CI, 1.652–7.703; p = 0.001); high epithelial VEGFR-2 (OR: 3.670; 95% CI, 1.451–9.286; p = 0.006); high Ki67 (OR: 3.291; 95% CI, 1.546–7.007; p = 0.002); and high pS6 (OR: 5.117; 95% CI, 2.113–12.392; p < 0.0001) with lymphatic spread. The final logistic regression model is summarized in Table 3.

Table 3 Molecular analysis multiple logistic regression: association of molecular features with lymphatic spread

Factor Rank of elimination from analysis OR 95% CI p value
T stage – 4.228 1.837 9.732 0.001
Fuhrman grade – 5.895 2.286 15.202 <0.0001
ECOG PS – 5.616 1.620 19.476 0.007
MVI – 20.413 2.663 156.486 0.004
Low vs high CAIX – 3.362 1.037 10.906 0.043
High vs low VEGFR-2 epithelial – 4.821 1.137 20.450 0.033
Smoking history 6 1.881 0.571 6.193 0.162
Ki 67 1 0.874 0.259 2.943 0.828
VEGFR-3 endothelial 4 1.662 0.465 5.940 0.434
VEGF-A epithelial 3 0.616 0.157 2.414 0.487
VEGFR-1 epithelial 2 1.185 0.325 4.313 0.797
pS 6 7 2.299 0.606 8.720 0.221
CAXII 5 2.157 0.475 9.786 0.319

OR = odds ratio; CI = confidence interval; ECOG PS = Eastern Cooperative Oncology Group performance status; MVI = microvascular invasion; CAIX = carbonic anhydrase IX; VEGFR = vascular endothelial growth factor receptor; CAXII = carbonic anhydrase XII.

3.3. Cytogenetic analysis

An abnormal karyotype was demonstrated by 247 of 272 (90.8%) tumors, (223 of 243 [91.8%] N0M0 tumors; 24 of 28 [85.7%] lymphatic tumors). The comparison of cytogenetic alterations between nonmetastatic and metastatic tumors is summarized in Table 4. Lymphatic metastasized tumors presented significantly less often with chromosome 3p deletion (p < 0.0001) when compared using Fisher exact test. In univariate logistic regression, loss of 3p (OR: 0.123; 95% CI, 0.045–0.334; p < 0.0001) was associated with lymphatic spread. After adjustment for clinical variables, 3p loss remained significantly associated with a lower risk for metastatic lymph node involvement. Multiple logistic regression analysis results of the cytogenetic analyses are summarized in Table 5.

Table 4 Comparison of chromosomal aberrations

N0M0 Lymphatic spread p value Test
Loss 3p
 Yes 156 (63.9) 5 (17.9) <0.0001 Fisher exact
 No 88 (36.1) 23 (88.9)
Loss 9p
 Yes 27 (11.1) 6 (21.4) 0.126 Fisher exact
 No 217 (88.9) 22 (78.6)
Loss 4q
 Yes 25 (10.2) 4 (14.3) 0.517 Fisher exact
 No 219 (89.8) 24 (85.7)
Loss 4p
 Yes 19 (7.8) 4 (14.3) 0.273 Fisher exact
 No 225 (92.2) 24 (85.7)
Loss 14q
 Yes 56 (23.0) 9 (32.1) 0.348 Fisher exact
 No 188 (77.0) 19 (67.9)
Gain 5q
 Yes 75 (30.7) 5 (17.9) 0.192 Fisher exact
 No 169 (69.3) 23 (82.1)
Loss 6q
 Yes 43 (17.6) 2 (7.1) 0.189 Fisher exact
 No 201 (82.7) 26 (92.9)
TRIS 7
 Yes 58 (23.8) 8 (28.6) 0.642 Fisher exact
 No 186 (76.2) 20 (71.4)
Loss 8p
 Yes 45 (18.4) 7 (25.0) 0.446 Fisher exact
 No 199 (81.6) 21 (75.0)
Gain 17p
 Yes 7 (2.9) 3 (10.7) 0.072 Fisher exact
 No 237 (97.1) 25 (89.3)
Loss 17p
 Yes 12 (4.9) 1 (3.6) 1.00 Fisher exact
 No 232 (95.1) 27 (96.4)

Table 5 Cytogenetic analysis multiple logistic regression: association of cytogenetic alterations with lymphatic spread

Rank of elimination from analysis OR 95% CI p value
Fuhrman grade – 3.559 1.801 7.032 <0.0001
T stage – 3.087 1.605 5.934 0.001
Loss 3p – 0.105 0.032 0.344 <0.0001
Smoking history 3 2.311 0.802 6.659 0.121
ECOG PS 1 1.514 0.569 4.029 0.406
MVI 2 1.581 0.503 4.965 0.433

OR = odds ratio; CI = confidence interval; ECOG = Eastern Cooperative Oncology Group performance status; MVI = microvascular invasion.

4. Discussion

The main findings of our current investigation are that RCCs that present with lymphatic spread can be characterized by a unique profile of clinicopathologic, molecular, and genetic alterations. In clinical analysis, a model consisting of smoking history, T stage, Fuhrman grade, ECOG PS, and MVI was correlated with lymphatic spread. Low CAIX and high epithelial VEGFR-2 protein expression on the molecular level, and a low rate of chromosome 3p loss on the genetic level, were further characteristic of lymphatic spread.

Recent reports have demonstrated a link between smoking and advanced tumor stages [18] and [19]. Our present results similarly demonstrate that smoking is an independent risk factor for lymphatic spread in ccRCC. We have therefore identified a modifiable lifestyle factor that appears to be capable of influencing the biological behavior of ccRCC. Interestingly, advanced primary T stage, but not tumor size, was an independent predictor for lymph nodes. This finding is particularly interesting with regard to active surveillance as a treatment strategy for localized tumors [20]. Our findings suggest that tumor size alone may be insufficient as a selection criterion to exclude subjects for active surveillance. Hypertension was less often observed in patients with lymph node–positive tumors, though hypertension was the weakest covariable in multivariate analysis. We hypothesize that patients with hypertension more often have contact with the health care system and with preventive medical examinations. Thus, it is possible that tumors are diagnosed earlier. Finally, the presence of MVI has been previously reported to be an important predictor of disease recurrence [21]. In the current study, MVI was the most important factor to predict lymphatic spread and, therefore, may serve as a pathologic criterion of paramount importance for postnephrectomy surveillance protocols.

Low CAIX and high epithelial VEGFR-2 expression independently predicted the presence of lymphatic spread. High expression of CAIX in RCC is mainly confined to clear cell tumors, and low CAIX expression has been reported to be associated with worse cancer-specific survival (CSS) [22]. In addition, several reports have shown it to be a marker for poor response to immunotherapy [23]. Clinical analyses have demonstrated that node-positive RCC is likewise associated with worse CSS and diminished response to immunotherapy [7] and [12]. Therefore, the current data provide a link between these molecular and clinical findings: low CAIX expression is a significant, independent marker for lymphatic spread, and both low CAIX and lymphatic spread are related to poor treatment response and unfavorable survival outcome. In having a low rate of chromosome 3p loss, low expression of CAIX, and poor response to immunotherapy, N+ ccRCC appears to share a number of features with non–clear cell RCC subtypes.

High expression of epithelial VEGFR-2 was associated with lymphatic spread. As it is well recognized, the VEGF pathway, with its various ligands and associated receptor tyrosine kinases, is of paramount importance for the regulation of angiogenesis and lymphangiogenesis in cancer. VEGF-A is a potent direct-acting angiogenic signal working through the tyrosine kinase receptors VEGFR-1 and VEGFR-2 [24]. While VEGF-mediated lymphangiogenesis is thought to be signaled predominantly through the binding of VEGF-C and VEGF-D to VEGFR-3 [25] and [26], the lymphangiogenic VEGF ligands C and D can also signal through the VEGFR-2 receptor to promote lymphangiogenesis [27]. In the current study, patients with lymphatic spread had significantly lower VEGFR-3 expression, but higher VEGFR-2 expression; moreover, high epithelial VEGFR-2 was correlated with metastatic lymph nodes. While low VEGFR-3 expression in N+ tumors is an unexpected finding, it must be acknowledged that much is still unknown about the mechanisms of lymphangiogenesis, and the predicted relationships may oversimplify the reality that VEGF receptors can be regulated in both a positive and negative manner by a wide variety of ligands and by an even wider variety of other receptors, including integrins, cadherins, and neuropilins [28]. Clearly there remains much to understand regarding VEGF/VEGFR biochemical and cellular signaling.

In univariate analysis, the current data reveal an association between low incidence of chromosome 3p loss and lymph node involvement that remained consistent after adjustment for clinicopathologic variables. Deletions of chromosome 3p were observed in 18% and 64% of N+ and N0 tumors, respectively, and its loss was correlated with an almost 90% lower risk for lymphatic spread. The chromosome 3p25 contains the von Hippel-Lindau (VHL) tumor suppressor gene [29], whose loss of function is responsible for the oncogenesis of the majority of clear cell RCCs. Beroukhim et al. [30] recently compared the different gene expression patterns of wild-type and biallelic VHL-inactivated RCCs, demonstrating that CAIX is the most differently expressed gene measured, with low CAIX expression being confined to a subset of VHL wild-type clear cell tumors. Since low expression of CAIX and a low rate of chromosome 3p loss are characteristic for VHL wild-type tumors, we suggest that ccRCCs that are capable of development of lymphatic spread may develop via a VHL-independent mechanism.

The present study has several limitations, including the retrospective study design, the small sample size (particularly in some subgroups that disallowed certain desired analyses), missing variables in the molecular analyses that did not permit all comparisons to be performed for the entire cohort, and concomitant distant metastases that may have influenced our results. It will be necessary to assess these clinical, molecular, and genetic features in a prospective validation cohort before they should be used in clinical practice for risk group assessment.

5. Conclusions

Low CAIX and high epithelial VEGFR-2 protein expression, as well as low rate of chromosome 3p loss, are associated with metastatic lymph node spread, suggesting that ccRCCs that spread via the lymphatics develop independently of VHL inactivation. These results may provide a genetic and molecular basis for the poor prognosis and lack of response to interleukin-2 in lymph node–positive ccRCC. These results warrant further investigation into non-VHL-related molecular pathways contributing to the node-positive phenotype in ccRCC.


Author contributions: Allan J. Pantuck had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Kroeger, Seligson, Pantuck, Belldegrun.

Acquisition of data: Kroeger, Seligson, Klatte, Rao, Zomorodian, Kabbinavar, Pantuck.

Analysis and interpretation of data: Kroeger, Klatte, Seligson, Birkhäuser, Belldegrun, Pantuck.

Drafting of the manuscript: Kroeger, Klatte, Rampersaud, Birkhäuser, Pantuck.

Critical revision of the manuscript for important intellectual content: Seligson, Rao, Belldegrun.

Statistical analysis: Kroeger, Klatte.

Obtaining funding: None.

Administrative, technical, or material support: Belldegrun, Zomorodian, Kabbinavar, Pantuck.

Supervision: Belldegrun, Pantuck.

Other (specify): None.

Financial disclosures: I certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: This study was supported in part by the Wissenschaftliche Urologische Gesellschaft e.V. (Germany) (N.K.) and the Swiss National Science Foundation individual fellowship PBBSP3-133403 (F.B.).

References

  • [1] Ferlay J, Shin HR, Bray F, Forman D, Mathers CD, Parkin D. Globocan 2008. Cancer incidence and mortality worldwide: IARC CancerBase No. 10. Lyon, France: International Agency for Research on Cancer (IARC); 2010. IARC Web site. http://globocan.iarc.fr/.
  • [2] W.H. Chow, S.S. Devesa. Contemporary epidemiology of renal cell cancer. Cancer J. 2008;14:288-301 Crossref.
  • [3] W.H. Chow, L.M. Dong, S.S. Devesa. Epidemiology and risk factors for kidney cancer. Nat Rev Urol. 2010;7:245-257 Crossref.
  • [4] M. Sun, R. Thuret, F. Abdollah, et al. Age-adjusted incidence, mortality, and survival rates of stage-specific renal cell carcinoma in North America: a trend analysis. Eur Urol. 2011;59:135-141 Abstract, Full-text, PDF, Crossref.
  • [5] F. Rodriguez-Covarrubias, R. Castillejos-Molina, M. Sotomayor, et al. Impact of lymph node invasion and sarcomatoid differentiation on the survival of patients with locally advanced renal cell carcinoma. Urol Int. 2010;85:23-29 Crossref.
  • [6] D.P. Zubac, L. Bostad, T. Seidal, T. Wentzel-Larsen, S.A. Haukaas. The prognostic relevance of interactions between venous invasion, lymph node involvement and distant metastases in renal cell carcinoma after radical nephrectomy. BMC Urol. 2008;8:19 Crossref.
  • [7] J.R. Vasselli, J.C. Yang, W.M. Linehan, D.E. White, S.A. Rosenberg, M.M. Walther. Lack of retroperitoneal lymphadenopathy predicts survival of patients with metastatic renal cell carcinoma. J Urol. 2001;166:68-72
  • [8] U. Capitanio, F. Becker, M.L. Blute, et al. Lymph node dissection in renal cell carcinoma. Eur Urol. 2011;60:1212-1220 Abstract, Full-text, PDF, Crossref.
  • [9] M.M. Oken, R.H. Creech, D.C. Tormey, et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol. 1982;5:649-655
  • [10] American Joint Comitte on Cancer. AJCC Staging Manual. 6th ed. New York, NY: Springer-Verlag; 2002. p. 323–5.
  • [11] S. Fuhrman, L. Lasky, C. Limas. Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol. 1982;6:655-663
  • [12] A.J. Pantuck, A. Zisman, F. Dorey, et al. Renal cell carcinoma with retroperitoneal lymph nodes. Impact on survival and benefits of immunotherapy. Cancer. 2003;97:2995-3002 Crossref.
  • [13] J. Kononen, L. Bubendorf, A. Kallioniemi, et al. Tissue microarrays for high-throughput molecular profiling of tumor specimens. Nat Med. 1998;4:844-847 Crossref.
  • [14] T. Klatte, D.B. Seligson, J. LaRochelle, et al. Molecular signatures of localized clear cell renal cell carcinoma to predict disease-free survival after nephrectomy. Cancer Epidemiol Biomarkers Prev. 2009;18:894-900 Crossref.
  • [15] D. Atkins, K.A. Reiffen, C.L. Tegtmeier, H. Winther, M.S. Bonato, S. Storkel. Immunohistochemical detection of EGFR in paraffin-embedded tumor tissues: variation in staining intensity due to choice of fixative and storage time of tissue sections. J Histochem Cytochem. 2004;52:893-901 Crossref.
  • [16] J. La Rochelle, T. Klatte, A. Dastane, et al. Chromosome 9p deletions identify an aggressive phenotype of clear cell renal cell carcinoma. Cancer. 2010;116:4696-4702 Crossref.
  • [17] L.G. Schaffer. International system of human cytogenetic nomenclature. (S. Karger AG, Basel, Switzerland, 2009)
  • [18] A. Parker, C. Lohse, J. Cheville, B. Leibovich, T. Igel, M. Blute. Evaluation of the association of current cigarette smoking and outcome for patients with clear cell renal cell carcinoma. Int J Urol. 2008;15:304-308 Crossref.
  • [19] M. Tsivian, D.M. Moreira, J.R. Caso, V. Mouraviev, T.J. Polascik. Cigarette smoking is associated with advanced renal cell carcinoma. J Clin Oncol. 2011;29:2027-2031 Crossref.
  • [20] H. Van Poppel, F. Becker, J.A. Cadeddu, et al. Treatment of localised renal cell carcinoma. Eur Urol. 2011;60:662-672 Abstract, Full-text, PDF, Crossref.
  • [21] H. Lang, V. Lindner, H. Letourneux, M. Martin, C. Saussine, D. Jacqmin. Prognostic value of microscopic venous invasion in renal cell carcinoma: long-term follow-up. Eur Urol. 2004;46:331-335 Abstract, Full-text, PDF, Crossref.
  • [22] M.H. Bui, D. Seligson, K.R. Han, et al. Carbonic anhydrase IX is an independent predictor of survival in advanced renal clear cell carcinoma: implications for prognosis and therapy. Clin Cancer Res. 2003;9:802-811
  • [23] M. Atkins, M. Regan, D. McDermott, et al. Carbonic anhydrase IX expression predicts outcome of interleukin 2 therapy for renal cancer. Clin Cancer Res. 2005;11:3714-3721 Crossref.
  • [24] N. Ferrara, H.P. Gerber, J. LeCouter. The biology of VEGF and its receptors. Nat Med. 2003;9:669-676 Crossref.
  • [25] M.J. Karkkainen, T.V. Petrova. Vascular endothelial growth factor receptors in the regulation of angiogenesis and lymphangiogenesis. Oncogene. 2000;19:5598-5605 Crossref.
  • [26] S.A. Stacker, M.G. Achen, L. Jussila, M.E. Baldwin, K. Alitalo. Lymphangiogenesis and cancer metastasis. Nat Rev Cancer. 2002;2:573-583 Crossref.
  • [27] H. Takahashi, M. Shibuya. The vascular endothelial growth factor (VEGF)/VEGF receptor system and its role under physiological and pathological conditions. Clin Sci (Lond). 2005;109:227-241
  • [28] N. Rahimi. VEGFR-1 and VEGFR-2: two non-identical twins with a unique physiognomy. Front Biosci. 2006;11:818-829
  • [29] B.R. Seizinger, G.A. Rouleau, L.J. Ozelius, et al. Von Hippel-Lindau disease maps to the region of chromosome 3 associated with renal cell carcinoma. Nature. 1988;332:268-269 Crossref.
  • [30] R. Beroukhim, J.P. Brunet, A. Di Napoli, et al. Patterns of gene expression and copy-number alterations in von-Hippel Lindau disease-associated and sporadic clear cell carcinoma of the kidney. Cancer Res. 2009;69:4674-4681 Crossref.
return to article outline

Footnotes

a Institute of Urologic Oncology, David Geffen School of Medicine, University of California, Los Angeles, CA, USA

b Department of Urology, University Hospital Greifswald, Greifswald, Germany

c Department of Pathology and Laboratory Medicine, David Geffen School of Medicine, University of California, Los Angeles, CA, USA

d Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, CA, USA

lowast Corresponding author. Institute of Urologic Oncology, 924 Westwood Blvd., Ste. 1050, Los Angeles, CA 90095-7207, USA. Tel. +1 310 206 2436; Fax: +1 310 794 3513.

Recommend this article

Currently this article has a rating of 0. Please log in to recommend it.

Article tools

Elsevier Copyright © Elsevier B.V. All rights reserved