| | Prognostic significance of vessel architecture and vascular stability in non-small cell lung cancerReceived 14 March 2006; received in revised form 12 July 2006; accepted 20 September 2006. 1. Introduction  Angiogenesis, the formation of a neovascular blood supply derived from preexisting blood vessels, plays a central role in tumour growth, maintenance and metastasis [1] in several solid tumour systems including non-small cell lung cancer (NSCLC). However, the prognostic value of the vessel number remains controversial for NSCLC. While some studies have reported microvessel density to be a prognostic factor atleast in a subset of patients [2], several previous studies have failed to detect the vessel number as a prognostic factor independent from already established prognostic factors [3], [4]. Besides inducing angiogenesis, some authors have proposed that, in particular circumstances, tumours could acquire blood supply in alternative ways such as vascular mimicry [5] and by exploration of the already established vasculature of the respective organ [5], [6]. Particularly the lung provides a dense vasculature system which derives from the pulmonary arteries and – potentially even more relevant – from the aorta via bronchial arteries. However, besides structural differences a deeper analysis of preexisting and newly formed vessels has not been undertaken. In general, newly formed vessels have been characterized as structurally distinct from the preexisting vasculature [7], [8]. After the formation of new endothelial cell sprouts, stabilization factors such as the vascular endothelial growth factor (VEGF) or attachment by perivascular cells such as pericytes and vascular smooth muscle cells (VSMCs) are needed for endothelial cell maintenance [9]. Besides VEGF, one of the most potent endothelial cell survival factors [1], [10], the angiopoietins (Ang-1 to Ang-4) have been shown to be important regulators of neovascularization and endothelial cell survival in malignant and non-malignant tissues [11]. Ang-1 has been recognized as the major activating ligand to the tyrosine kinase receptor Tie2, thereby promoting endothelial cell survival and vessel stabilization by recruiting and sustaining peri-endothelial supporting cells [5], [12]. Ang-2 is a naturally occurring antagonist to Ang-1 and prevents Tie2 activation. However, the effects of the angiopoietins on angiogenesis remain controversial [13], [14]. Some studies suggest that Ang-1 may be pro-angiogenic [13], whereas others have shown that Ang-1 inhibits angiogenesis, tumour growth and vascular permeability [14], [15], [16]. In this regard, it has been hypothesized that the presence of VEGF may determine the effect of the angiopoietins. For example, in the presence of VEGF, vessel destabilization by Ang-2 has been hypothesized to induce an angiogenic response; however, in the absence of VEGF, Ang-2 leads to vessel regression [17]. To evaluate characteristics and prognostic impact of different vessel structure types as possible indication of preexisting versus newly formed (i.e. angiogenic) blood vessels, we investigated the vascular pattern and microvessel density on 72 NSCLC patients with stages I and II disease. Besides a structural analysis of tumour vessels, we questioned whether the different vessel types may be further characterized by differential vascular stability as exerted by pericyte coverage and distinct expression patterns of known stabilization factors such as VEGF and angiopoietins. 2. Patients and methods 2.1. Patients Vessel density and expression levels of Ang-1, Ang-2 and VEGF proteins were determined by immunohistochemistry in 72 patients with primary NSCLC surgically treated with curative intention from February 1993 to June 1994. After written informed consent, patients underwent thoracotomy with the objective of achieving complete resection of the tumour (resection margin microscopically free of tumour cells) and extensive mediastinal lymph node dissection. Diagnosis of malignant disease was confirmed pathologically and classified according to World Health Organization criteria. The postsurgical stage of each tumour was determined according to the revised International System for Staging of Lung Cancer [18]. Inclusion criteria for this study were surgical stages I and II, complete resection of the tumour (resection margin microscopically free of tumour cells) and perioperative survival (within 30 days after surgery). Patient characteristics are shown in Table 1. The median age was 62 years (range 34–79 years). 2.2. Immunohistochemical analyses Antibodies for immunohistochemical analyses were obtained as follows: rabbit anti-human angiopoietin-1 (Alpha Diagnostic International, San Antonio, TX, USA), rabbit anti-human angiopoietin-2, rabbit anti-human VEGF (Santa Cruz Biotechnology, Santa Cruz, CA), mouse anti-human IgG1k CD31 (Dako, Glostrup, Denmark), mouse anti-human IgG2ak α-smooth muscle actin (αSMA, Dako), goat anti-mouse IgG1 Fab2-fragment Biotin labelled (Southern Biotechnology, Birmingham, AL, USA), goat anti-mouse IgG2a AP labelled (Southern Biotechnology), mouse anti-rabbit IgG M0737 (Dako), rabbit anti-mouse Z0259 (Dako). One paraffin block from each patient containing tumour tissue was chosen blindly and sliced in 4 mm sections. To study the overall tissue morphology, slides were stained with H&E [19]. For assessment of the expression of VEGF, Ang-1 and Ang-2, immunohistochemical staining was performed by the alkaline phosphatase/antialkaline phosphatase double bridge technique (APAAP kit; Dako) as previously described [20], [21]. In short, after deparaffinization and rehydration, antigen retrieval was performed in a microwave oven at 450W twice for 7min in 10mM sodium citrate (pH 6.0; Dako). After application of the primary antibodies (dilutions for VEGF, Ang-1 and Ang-2 were 1:1000, 1:100, 1:500, respectively) overnight at 4°C, subsequent steps were performed according to the manufacturer's instructions. The fast red substrate (Dako) supplemented with 0.1% (w/v) levamisole was used for revelation of phosphatase activity. For CD31 staining, endogenous peroxidases were blocked with 3% H2O2 in methanol, and slides were incubated with an anti-CD31 directed antibody (1:100) overnight at 4°C. Subsequently, a biotin-conjugated secondary antibody and peroxidase-conjugated streptavidin were applied for 1h and 30min at room temperature, respectively. Slides were stained with AEC substrate (Dako), and the reaction was stopped by washing with distilled water. For determination of pericyte coverage, double-staining for CD31 and αSMA was performed using a modified protocol of a previously described procedure [22]. Briefly, slides were stained with an anti-αSMA directed antibody applying the alkaline phosphatase/antialkaline phosphatase double bridge technique as described above using nitro blue tetrazolium (NBT; Dako) as substrate. Subsequently, vessels were stained with an anti-CD31 antibody as described and visualized with a diaminobenzidine substrate (Dako). Staining was monitored under a bright-field microscope, and the reaction was stopped by washing with distilled water. 2.4. Definition of the vascular pattern Two vascular patterns were distinguished: the “alveolar” pattern was characterized by a vasculature, which looked similar to the normal alveolar structure of the lung. Tumour cells were filling up the alveoli and seemed to exploit the existing alveolar vessels. In the centre of the tumour cell-filled alveoli, a small necrosis often was seen, and the surrounding cell layer typically was seen up to 10 cells in thickness. In contrast, the “angiogenic” pattern showed a quite disorganized microvessel pattern, which is typically seen in angiogenic solid tumours. The normal lung architecture was no longer apparent, and associated–associated stroma and vessels were present. 2.5. Quantification of protein expression In each tissue sample, four distinct areas with high VEGF, Ang-1, and Ang-2 expression were evaluated at 100× magnification. Expression was semiquantitatively assessed as the product of positive stained cell quantity and staining intensity as described [25] and reported as AU (arbitrary units). The percentage of positive cells stained with each specific antibody within cellular areas was estimated according to a three-grade scale (1, ≤10% positive cells; 2, 10–50% positive cells; 3, ≥50% positive cells). Subsequently, the intensity of positive staining of the majority of cells was evaluated in four representative 500× fields (0.126mm2 field area) selected in each section after the initial screening at 100–250× magnification (1, faint or negative; 2, moderate staining; 3, intense staining). The cut-off of all factors was set at 3AU since the median expression status of all factors was approximately 3AU. To ensure the reliability of the quantification assay, additional slides from the same tissue blocks were stained in independent immunostainings and reanalysed with excellent agreement. 2.6. Statistics Statistical analysis was performed using SPSS (SPSS GmbH Software, Munich, Germany). The influence of variables on survival was analysed using the log-rank test. The disease-free interval (DFI) was defined as the interval between lung resection and local recurrence and/or occurrence of distant metastases. The overall survival time was calculated as the interval between surgery and death or last clinical evaluation. Patients who died from other causes than lung cancer were censored. 3. Results 3.1. Analysis of the microvascular pattern A total of 72 patients with stages I and II NSCLC disease were included into the study. Seventy-one patients were eligible for survival analysis. With an overall median follow-up of 70.5 months and of 139 months for surviving patients, 2-, 5- and 10-year survival rates were 73%, 55% and 51%, respectively. The median DFI was 68 months. A total of 56 patients was eligible for analysis of the vessel pattern with sufficient staining quality of the slides and a size of the tissue section large enough for analysis of blood vessels in atleast four different spots. Forty-five of these 56 tumours (80%) showed evidence of an alveolar pattern in atleast parts of the sample (Fig. 1A). Among these, the vascular pattern was entirely “alveolar” in 27 tumours. In contrast, 11 tumours (20%) showed an entirely quite disorganized microvascular pattern characteristic of angiogenic solid tumours (Fig. 1B). However, a specific vascular pattern was not linked to either histological subtype, gender or associated stage. In addition, out of the collective of 56 samples assessed for vascular patterns, 31 random tissue samples were double stained using anti-CD31 and anti-αSMA antibodies. Interestingly, the alveolar vascular pattern was characterized by a nearly complete coverage of αSMA positive perivascular cells (Fig. 1C and E). In contrast, slides with presumably newly formed angiogenic vessels showed only 26–74% coverage by perivascular cells (Fig. 1D and F). Taking these characteristics, the samples were re-assessed for their respective vascular pattern with differentiation of nearly complete pericytes coverage versus less pericytes coverage with no significant variations in results. 3.3. Evaluation of VEGF, Ang-1 and Ang-2 expression To further analyse differences in vascular stabilization factors, the expression pattern of VEGF, Ang-1, and Ang-2 was evaluated by performing immunohistochemistry. Representative examples of protein expression are displayed in Fig. 3. VEGF expression was increased in 31 patients (46% of total 68 evaluated patients). In 24 tumours with determined “angiogenic” vascular pattern, VEGF expression was not significantly associated with the vascular density (p=0.37). Ang-1 and Ang-2 protein expressions were detected in a variety of cells. Since both proteins were predominantly expressed by associated cells, the following analyses were based on associated cell-derived expression. Forty-four (63% of total 70 evaluated patients) and 22 (34% of total 65 patients) samples displayed increased Ang-1 and Ang-2 expression, respectively. High Ang-1 expression was often detected in proximity of areas with necrosis. Moreover, increased Ang-1 expression was significantly linked with elevated VEGF expression (p=0.005) and tended to be associated with increased Ang-2 expression (p=0.054). Interestingly, there was no significant association of the expression of either factor with a specific vessel pattern. However, tumours with “angiogenic” vascular pattern tended to have lower Ang-1 expression levels (p=0.2) and higher Ang-2 expression levels (p=0.34). 4. Discussion Targeting the formation of new blood vessels has become an attractive approach in solid tumours including NSCLC. In the present study, we analysed the characteristics and prognostic significance of the tumour vasculature in 72 patients with early-stage NSCLC. Interestingly, besides tumours displaying an “angiogenic” disorganized microvascular pattern as evidence of angiogenesis induction we further found a distinct “alveolar” vascular pattern in a major subset of the investigated samples which looked similar to the normal alveolar structure of the lung. A similar pattern has been previously addressed as evidence of coopting the preexisting vasculature [6], [26]. In fact, the normal lung provides two vasculature systems deriving from the pulmonary and the bronchial arteries that both may be adopted by a growing tumour without the development of hypoxic areas and the subsequent need for angiogenesis induction [27]. Similarly, Holash et al. described in a rat glioma model that even smallest gliomas were vascularized with vessels similar to normal brain vessels and without evidence of angiogenesis such as judged by the lack of vascular sprouts and hyperplastic vessels [5]. However, while previous authors defined distinct vascular patterns based on the structural appearance [6], [26], there is no definition of a coopted preexisting vessel based on histopathological criteria. Since colocalized endothelial and perivascular cells have repeatedly been used to define mature blood vessels [9], [24], we defined the distinct vascular patterns by characterization of the vascular stability. Performing double staining for endothelial cells and αSMA-positive cells, detection of αSMA-positive cells lined along blood vessels was interpreted as coverage with perivascular cells such as pericytes and VSMCs [22], [23]. While newly formed “angiogenic” vessels were covered by perivascular cells only to some extent, the “alveolar” vascular pattern was characterized by a complete coverage with αSMA-positive cells in our study similarly to the normal vasculature. Collectively, these data indicate that NSCLC tumours are characterized by distinct vascular structures with differential coverage with perivascular cells which may indicate cooption of the preexisting pulmonary vasculature atleast in some parts of the tumour. In the present study, we found an “alveolar” pattern in the majority of tumours atleast in some sections. With a median follow-up of 139 months for surviving patients, evidence of an alveolar vascular pattern was significantly associated with improved survival while an entirely “angiogenic” pattern indicated a worse outcome. These data are in agreement with a study by Offersen et al. on 143 patients with stages I–IIIA NSCLC which used a definition of vessel types similar ours [26]. In contrast, Pezzella and associates found no association between vascularity and outcome in 515 patients with stage I NSCLC [4], [6]. In this study, the non-angiogenic pattern accounted for only 16% of the tumours and was defined as such only if the “alveolar” pattern was present throughout the tumour [6]. In addition, the same authors acknowledged that the majority of tumours showed evidence of an “alveolar” pattern at the associated invasion front. Presumably, the “alveolar” pattern may indicate that the tumour has not yet upregulated the induction of angiogenesis but has been able to establish itself as an invasive carcinoma without a selection pressure of aggressiveness. In case of formation of new blood vessels, high vascular density may serve as an adverse prognostic marker as suggested in our study. To further characterize vascular stability, we investigated the expressions of various angiogenic factors including VEGF and angiopoietins. Similar to Tanaka et al. [28], we analysed the expression of angiopoietins based on the expression in associated cells because of the heterogeneity of angiopoietin expression in endothelial cells. In our study, VEGF, Ang-1 and Ang-2 were intensely expressed by associated cells in 46%, 63% and 34% of the patients, respectively. There was no significant association of expression of either factor with a specific vascular type. Interestingly, the tumours with “angiogenic” vascular pattern tended to have lower Ang-1 but higher Ang-2 expression levels possibly indicating increased destabilization. In the entire collective, high expression of Ang-1 was significantly linked to poor outcome while high expression of VEGF or Ang-2 showed no association to the clinical outcome. These data may look somewhat contradictory to the study by Tanaka et al., who found high Ang-2 expression in 17% of 236 patients with stages I–IIIA NSCLC that was correlated to poor survival [28]. However, angiopoietins have been implicated in vessel cooption and regulation of angiogenesis in several ways [5], [17]. To date, most authors have linked the biologic effect of Ang-2 in the context of VEGF expression. However, in the present study we provide evidence that the effect of Ang-1 may be modulated by the expression of VEGF, as well. Increased expression of VEGF was associated with improved outcome in tumours with high expression of Ang-1, but with shortened survival in tumours with low Ang-1 status. These results suggest that: (1) VEGF may somewhat counterbalance the adverse prognostic effects of high Ang-1 expression and (2) the importance of each factor for the regulation of angiogenesis has to be interpreted in the context of the expression of all VEGF, Ang-1 and Ang-2. In summary, an alveolar vessel type as indication of the adoption of preexisting vessels appears to be a common phenomenon in NSCLC while induction of angiogenesis may reflect an increased aggressive behaviour. However, heterogeneity of vascular patterns within the same tumour may complicate a correlation to prognosis. With upcoming clinical studies addressing anti-angiogenesis as a potential therapeutic target it remains to be analysed whether anti-angiogenic agents act differently on presumably coopted lung vessels compared to newly formed vessels. As a hypothesis, alveolar type vessels may be more resistant to anti-angiogenic therapy which may be atleast partly mediated by the coverage with VSMCs, soluble factors, or both [1]. Hence, targeting these vessels may require other strategies than preventing angiogenesis. Possibly, targets that not only inhibit the induction of new blood vessels but also destabilize the already established vasculature are more likely to improve the current treatment options in patients with advanced NSCLC. Consequently, to better evaluate the potential benefit and to improve anti-angiogenic agents in NSCLC, future clinical studies should incorporate detailed molecular and histopathological analyses of the associated tissue including vessel cooption and expression of angiogenic and anti-angiogenic factors. Acknowledgements Supported, in part, by the Dr. Mildred Scheel Stiftung für Krebsforschung, Deutsche Krebshilfe (Grant 10-2207-Re 1; to N.R.), the Deutsche Forschungsgemeinschaft Grant Me 950-3-2 (to R.M.) and the Interdisciplinary Centre of Clinical Research Muenster (IZKF Project no. Kess 2/023/04 to R.M.), Germany. References [1]. [1]Reinmuth N, Liu W, Jung YD, et al. Induction of VEGF in perivascular cells defines a potential paracrine mechanism for endothelial cell survival. FASEB J. 2001;15:1239–1241. MEDLINE [2]. [2]Kojima H, Shijubo N, Abe S. Thymidine phosphorylase and vascular endothelial growth factor in patients with stage I lung adenocarcinoma. Cancer. 2002;94:1083–1093. [3]. [3]Giatromanolaki A, Koukourakis M, O’Byrne K, et al. Prognostic value of angiogenesis in operable non-small cell lung cancer. J Pathol. 1996;179:80–88. MEDLINE |
CrossRef
[4]. [4]Pastorino U, Andreola S, Tagliabue E, et al. Immunocytochemical markers in stage I lung cancer: relevance to prognosis. J Clin Oncol. 1997;15:2858–2865. [5]. [5]Holash J, Maisonpierre PC, Compton D, et al. Vessel cooption, regression, and growth in tumors mediated by angiopoietins and VEGF. Science. 1999;284:1994–1998. MEDLINE |
CrossRef
[6]. [6]Pezzella F, Pastorino U, Tagliabue E, et al. Non-small cell lung carcinoma tumor growth without morphological evidence of neo-angiogenesis. Am J Pathol. 1997;151:1417–1423. MEDLINE [7]. [7]Folkman J. The role of angiogenesis in tumor growth. Semin Cancer Biol. 1992;3:65–71. [8]. [8]Jain RK. Normalizing tumor vasculature with anti-angiogenic therapy: a new paradigm for combination therapy. Nat Med. 2001;7:987–989. MEDLINE |
CrossRef
[9]. [9]Korff T, Kimmina S, Martiny-Baron G, et al. Blood vessel maturation in a three-dimensional spheroidal coculture model: direct contact with smooth muscle cells regulates endothelial cell quiescence and abrogates VEGF responsiveness. FASEB J. 2001;15:447–457. MEDLINE |
CrossRef
[10]. [10]Yang AD, Bauer TW, Camp ER, et al. Improving delivery of antineoplastic agents with anti-vascular endothelial growth factor therapy. Cancer. 2005;103:1561–1570. [11]. [11]Tait CR, Jones PF. Angiopoietins in tumors: the angiogenic swith. J Pathol. 2004;204:1–10. MEDLINE |
CrossRef
[12]. [12]Hayes AJ, Huang WQ, Mallah J, et al. Angiopoietin-1 and its receptor Tie-2 participate in the regulation of capillary-like tubule formation and survival of endothelial cells. Microvasc Res. 1999;58:224–237. MEDLINE |
CrossRef
[13]. [13]Suri C, McClain J, Thurston G, et al. Increased vascularization in mice overexpressing angiopoietin-1. Science. 1998;282:468–471. MEDLINE |
CrossRef
[14]. [14]Ahmad SA, Liu W, Jung YD, et al. Differential expression of angiopoietin-1 and angiopoietin-2 in colon carcinoma. A possible mechanism for the initiation of angiogenesis. Cancer. 2001;92:1138–1143. [15]. [15]Chae JK, Kim I, Lim ST, et al. Coadministration of angiopoietin-1 and vascular endothelial growth factor enhances collateral vascularization. Arterioscler Thromb Vasc Biol. 2000;20:2573–2578. [16]. [16]Thurston G, Rudge JS, Ioffe E, et al. Angiopoietin-1 protects the adult vasculature against plasma leakage. Nat Med. 2000;6:460–463. MEDLINE |
CrossRef
[17]. [17]Maisonpierre PC, Suri C, Jones PF, et al. Angiopoietin-2, a natural antagonist for Tie2 that disrupts in vivo angiogenesis. Science. 1997;277:55–60. MEDLINE |
CrossRef
[18]. [18]Mountain CF. Revisions in the international system for staging lung cancer. Chest. 1997;111:1710–1717. MEDLINE |
CrossRef
[19]. [19]Shaheen RM, Davis DW, Liu W, et al. Antiangiogenic therapy targeting the tyrosine kinase receptor for vascular endothelial growth factor receptor inhibits the growth of colon cancer liver metastasis and induces tumor and endothelial cell apoptosis. Cancer Res. 1999;59:5412–5416. MEDLINE [20]. [20]Bieker R, Padro T, Kramer J, et al. Overexpression of basic fibroblast growth factor and autocrine stimulation in acute myeloid leukemia. Cancer Res. 2003;63:7241–7246. MEDLINE [21]. [21]Kreuter M, Bieker R, Bielack SS, et al. Prognostic relevance of increased angiogenesis in osteosarcoma. Clin Cancer Res. 2004;10:8531–8537. MEDLINE |
CrossRef
[22]. [22]Reinmuth N, Fan F, Liu W, et al. Impact of insulin-like growth factor receptor-I function on angiogenesis, growth, and metastasis of colon cancer. Lab Invest. 2002;82:1377–1389. MEDLINE [23]. [23]Reinmuth N, Liu W, Ahmad SA, et al. Alphavbeta3 integrin antagonist S247 decreases colon cancer metastasis and angiogenesis and improves survival in mice. Cancer Res. 2003;63:2079–2087. MEDLINE [24]. [24]Eberhard A, Kahlert S, Goede V, et al. Heterogeneity of angiogenesis and blood vessel maturation in human tumors: implications for antiangiogenic tumor therapies. Cancer Res. 2000;60:1388–1393. MEDLINE [25]. [25]Padro T, Bieker R, Ruiz S, et al. Overexpression of vascular endothelial growth factor (VEGF) and its cellular receptor KDR (VEGFR-2) in the bone marrow of patients with acute myeloid leukemia. Leukemia. 2002;16:1302–1310. MEDLINE |
CrossRef
[26]. [26]Offersen BV, Pfeiffer P, Hamilton-Dutoit S, et al. Patterns of angiogenesis in non-small cell lung carcinoma. Cancer. 2001;91:1500–1509. [27]. [27]Milne EN, Noonan CD, Margulis AR, et al. Vascular supply of pulmonary metastases. Experimental study in rats. Invest Radiol. 1969;4:215–229. MEDLINE |
CrossRef
[28]. [28]Tanaka F, Ishikawa S, Yanagihara K, et al. Expression of angiopoietins and its clinical significance in non-small cell lung cancer. Cancer Res. 2002;62:7124–7129. MEDLINE a Department of Medicine/Hematology and Oncology, University of Muenster, Muenster, Germany b IZKF Muenster, University of Muenster, Muenster, Germany c Clinic for Thoracic Diseases, University of Heidelberg, Amalienstr. 5, D-69126 Heidelberg, Germany d Clinic for Pulmonary Diseases, Hemer, Germany  Corresponding author at: Clinic for Thoracic Diseases, University of Heidelberg, Amalienstr. 5, D-69126 Heidelberg, Germany. Tel.: +49 6221 396 1301; fax: +49 6221 396 1302.
PII: S0169-5002(06)00516-2 doi:10.1016/j.lungcan.2006.09.025 © 2006 Elsevier Ireland Ltd. All rights reserved. | |
|
FULL TEXT
