ATS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Ann Thorac Surg 2008;85:395-405. doi:10.1016/j.athoracsur.2007.08.054
© 2008 The Society of Thoracic Surgeons
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Paul Van Schil
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sardari Nia, P.
Right arrow Articles by Van Marck, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sardari Nia, P.
Right arrow Articles by Van Marck, E.
Related Collections
Right arrow Lung - cancer

Original Articles: General Thoracic

Different Growth Patterns of Non-Small Cell Lung Cancer Represent Distinct Biologic Subtypes

Peyman Sardari Nia, MDa,*, Cecile Colpaert, MD, PhDb, Peter Vermeulen, MD, PhDb, Joost Weyler, MD, PhDc, Francesco Pezzella, MD, PhDd, Paul Van Schil, MD, PhDa, Eric Van Marck, MD, PhDb

a Department of Thoracic and Vascular Surgery, University Hospital Antwerp, Antwerp, Belgium
b Department of Pathology, University Hospital Antwerp, Antwerp, Belgium
c Department of Epidemiology and Social Medicine, University Hospital Antwerp, Antwerp, Belgium
d Cancer Research Tumor Pathology Group, John Radcliffe Hospital, Oxford, United Kingdom

Accepted for publication August 23, 2007.

* Address correspondence to Dr Sardari Nia, Department of Thoracic and Vascular Surgery, University Hospital Antwerp, Wilrijkstraat 10, Antwerp, Edegem B2650, Belgium (Email: peymansardarinia{at}yahoo.com).


   Abstract
Top
 Abstract
Introduction
 Material and Methods
 Results
 Comment
 References
 
Background: We have recently shown the prognostic value of growth pattern classification in non-small cell lung cancer. The aim of this study is to validate the hypothesis that these growth patterns have a distinct angiogenic and proliferative profile.

Methods: Hematoxylin-eosin stained tissue sections of 239 patients with non-small cell lung cancer were classified into growth patterns. One representative tissue section per patient was double immunostained with CD34 and Ki-67 antibodies. Endothelial cell proliferation fraction, tumor cell proliferation fraction, microvessel density, and Chalkley count were assessed at the invading front and the center of the selected tumor section.

Results: According to the growth pattern classification, 161 patients (67.4%) had a destructive, 33 (13.8%) a papillary, and 45 (18.8%) an alveolar growth pattern. There were significant differences in endothelial cell proliferation fraction (p < 0.001), tumor cell proliferation fraction (p < 0.001), microvessel density (p < 0.001), and Chalkley count (p < 0.001) between the growth patterns. Multiple Cox regression analysis showed that a low endothelial cell proliferation fraction was consistently an independent prognostic factor for overall poor (hazard ratio = 0.93; confidence interval: 0.88 to 0.97, p = 0.002) and disease-free survival (hazard ratio = 0.94; confidence interval: 0.89 to 0.98, p = 0.007).

Conclusions: Growth patterns have a distinct angiogenic and proliferative profile. In non-small cell lung cancer, a low degree of angiogenesis (a low endothelial cell proliferation fraction) is associated with poor prognosis.


   Introduction
Top
 Abstract
 Introduction
Material and Methods
 Results
 Comment
 References
 
Lung cancer is the main cancer worldwide in terms of incidence and mortality [1]. Our recent search on Pubmed for the keywords "lung cancer" revealed more than 100,000 publications since 1950 [2]. Despite this enormous wealth of scientific literature, lung cancer remains the leading cause of cancer deaths worldwide, and tumor-node-metastasis (TNM) staging is still the most important tool used to estimate the prognosis for lung cancer and select the best possible combination of treatment modalities [3].

Although TNM staging yields an accurate estimate of the progression of disease at the time of diagnosis, it does not always account for survival differences. Moreover, TNM staging does not provide information about the biologic profile of the tumor. We have recently studied a classification of non-small cell lung cancer (NSCLC) according to the growth patterns (Fig 1) [4]. The growth pattern is the biologic behavior of the "tumor as a whole" with respect to normal lung parenchyma at the invading front of the tumor. Growth pattern classification is based on the observation of nonangiogenic growth in lung, liver, and lymph nodes [4–8]. The nonangiogenic growth pattern in NSCLC was first described by Pezzella and colleagues [5]. An essential prerequisite for nonangiogenic growth appears to be the ability of the tumor to preserve the stromal architecture of the host tissue where the preexisting blood vessels lie. We have shown that the classification of NSCLC into these growth patterns has a significant prognostic value [4]. We hypothesize that growth patterns are the synthesis of different biologic characteristics and represent distinct biologic subtypes. The attractiveness of the growth pattern hypothesis lies in the fact that it probably integrates several distinct molecular mechanisms into a single feature. The validation of this hypothesis could be important for clinical practice as different growth patterns could respond differently to different treatment modalities [2].


Figure 1
View larger version (14K):
[in this window]
[in a new window]

  		Fig 1. Classification of non-small cell lung cancer into the growth patterns based on morphologic characteristics of the tumor tissue at the invading front.

 
The growth pattern classification is based on morphologic characteristics of the interface (the invading front) of a tumor only. We have deliberately chosen the interface as it seems to be the most homogeneous region of the tumor. We argued that the interface is the most active region in the tumor representing the actual tumor growth activity (at the time of sampling) whereas the center of the tumor probably represents the past as well as the present tumor growth activity providing a more heterogeneous region.

Therefore, the aims of the current study are to validate that the interface of the tumor in NSCLC has the highest tumor growth activity and that the morphologically different growth patterns have distinct angiogenic and proliferative profiles.


   Material and Methods
Top
 Abstract
 Introduction
 Material and Methods
Results
 Comment
 References
 
Study Population
All relevant clinical information was gathered retrospectively from 239 consecutive patients undergoing curative surgical resection for primary NSCLC at the University Hospital of Antwerp between January 1991 and January 2001. The Institutional Review Board approval and permission was granted before data collection and conduction of this study, and individual patient consent was waived. All patients were operated on by the same surgeon (P.V.S.) and had survived 30 days beyond the operation. None of the patients had received a systemic therapy or radiotherapy before the curative resection. The patients were followed up every 4 months in the first 2 years, every 6 months from the third year, and annually from the sixth year. In every follow-up contact, an interview, physical examination, and chest radiography were performed. The pathologic TNM staging was used based on histopathology reports and on the perioperative findings [3]. The tumors were classified according to the current World Health Organization histologic classification by a pathologist (C.C.) [9]. The disease-free interval was defined as the interval between surgery and recurrence. Recurrence was defined as first radiologic sign of local recurrence, metastasis, or both, provided that it was substantiated with further clinical progression of the disease and administration of antitumoral or palliative therapy, or clear radiologic progression or confirmation on biopsy. The endpoints were death and recurrence of the disease based on information gathered from the hospital registry and the medical records.

Growth Pattern Analysis
Hematoxylin and eosin–stained tissue sections from all tumor blocks from each patient were examined. Tumors were classified according to growth patterns based on two considerations as described before [4]: firstly, preservation or destruction of lung parenchyma at the lung-tumor interface, and, secondly, presence or absence of a tumor-associated stroma at the lung-tumor interface (Fig 1). The interface was defined as a field of x100 magnification at the edge of the tumor, containing only tumor tissue, next to normal lung tissue. Based on these criteria, tumors were classified into three growth patterns as follows: (1) Destructive growth pattern: destruction of lung parenchyma with the presence of tumor-associated stroma at the interface. (2) Papillary growth pattern: preservation of the alveolar structure of the lung parenchyma at the interface with formation of stromal stalks containing capillary vessels originating from the alveolar septa, suggesting cooption (the tumor uses the preexisting vascularization of the host tissue for the oxygen supply) of alveolar blood vessels with subsequent angiogenesis. (3) Alveolar growth pattern: preservation of the alveolar structure of the lung parenchyma with cooption of septal blood vessels and without evidence of new stroma formation at the interface. In this growth pattern, solid tumor cell nests fill the alveolar spaces, often with the presence of necrosis in the center of these nests. This group does not include bronchioloalveolar carcinoma, which is characterized by orderly replacement of pneumocytes by tumor cells along the alveolar septa without infiltration, necrosis, or fibrovascular proliferation.

Although all tumor blocks were examined, the growth patterns could only be assessed on the tumor blocks on which there was a clear lung-tumor interface. On average, two to three tumor blocks were available with clear lung-tumor interface based on which the growth pattern was determined. For a tumor to be classified as alveolar or papillary, this tumor growth had to be predominantly (90%) present throughout the whole interface on the different tumor blocks.

In only two cases was a mixed growth pattern observed at the interface. In these two cases, a major destructive component was always observed with a minor alveolar or papillary component. So, these tumors were included in the destructive group.

The growth pattern classification must not be confused with the histologic classification of NSCLC. The focus in histology is predominantly on the morphologic characteristics of individual tumor cells, whereby enormous morphologic heterogeneity exists even in different fields of the same tumor section. The focus in growth pattern classification is the following question: what does the tumor do as a whole with regard to the normal lung parenchyma at the interface? There are only three possible ways in which a tumor can behave according to growth pattern classification, namely preservation, destruction, or preservation with modulation (stroma formation). Therefore, this classification is less bound to the heterogeneity.

Immunohistochemistry
One formalin-fixed, paraffin-embedded tissue block containing a representative tumor fragment with a clear interface between lung and tumor tissue and also containing a substantial central part of the tumor was selected per patient. We chose the tumor block with the largest lung-tumor interface for immunohistochemistry analysis.

Five-micrometer sections were cut. After deparaffinization, a double immunohistochemical technique was used to stain simultaneously, on the one hand, proliferating cells using monoclonal antibodies against Ki-67 antigen (Clone MIB-1, Clone No. M7240, Dako) and, on the other hand, endothelial cells using monoclonal antibodies against CD34 antigen (Clone QBEnd 10, Clone No. M7165, Dako), on a Dako Autostainer [10].

Endothelial Cell Proliferation Fraction
An area with highest vascular density (vascular hotspot) at the interface of the tumor was identified at low magnification (x10 ocular and x10 objective). At higher magnification (x10 ocular and x40 objective), a total number of 200 endothelial cells (CD34 positive) were evaluated in consecutive fields; and endothelial cell proliferation fraction (ECPF) was the number of endothelial cells with Ki-67–stained nuclei per 200 endothelial cells, divided by 2 to yield percentage values.

Consequently, on the same tissue section an area with highest vascular density in the center of the tumor was identified at low magnification, and ECPF was assessed in the same manner as done at the interface based on evaluation of 200 endothelial cells.

Tumor Cell Proliferation Fraction
An area with the highest immunoreactivity with Ki-67 at the interface of the tumor was identified at low magnification (x10 ocular and x10 objective). At higher magnification (x10 ocular and x40 objective), a total number of 500 tumor cells were evaluated; and tumor cell proliferation fraction (TCPF) was the number of tumor cells with Ki-67–stained nuclei per 500 tumor cells, divided by 5.

Consequently, on the same tissue section an area with highest immunoreactivity with Ki-67 in the center of the tumor was identified at low magnification, and TCPF was assessed in the same manner as done at interface based on evaluation of 500 tumor cells.

The ratio of TCPF to ECPF at the interface, roughly reflecting the degree of angiogenesis-independent tumor growth at the interface, was determined.

Tumor Vascularization
Chalkley method
Microvessel density (MVD) was determined according to the Chalkley method [11]. This is a morphometric point counting system using a microscope eyepiece graticule. Four areas of highest vascular density ("hot spots," two at the interface and two in the center) were identified at low magnification (x10 ocular and x10 objective). On a higher magnification (x10 ocular and x25 objective, Chalkley grid area 0.22 mm2), a 25-point Chalkley eyepiece graticule was applied to each hot spot and oriented to permit the maximum number of points to hit on or in an immunostained microvessel.

Weidner method
Microvessel density was also determined according to the Weidner method [12]. Four areas of highest vascular density ("hot spots," two at the interface and two in the center of the same tissue section) were identified at low magnification (x10 ocular and x10 objective). On a higher magnification (x10 ocular and x25), the number of CD34 immunostained blood vessels were counted in the microscopic field.

Statistics
Distribution of variables between groups was analyzed by the {chi}2 test for categorical variables. Continuous variables—age, size of the tumor, ECPF, TCPF, MVD, and Chalkley count—were tested for normality with the Kolmogorov-Smirnov test. The distribution of these factors between groups was assessed by the Kruskal-Wallis test and Mann-Whitney U test, because these factors were not normally distributed. Comparison between different variables determined at the interface and in the center of the tumor, was assessed by two-related-samples test of Wilcoxon. Survival effects were evaluated using Kaplan-Meier plots, and the differences were assessed by the log-rank test. Potential confounders were entered into a Cox multiple regression model for overall survival as well as for disease-free survival. The model building was guided by the influence of the inclusion of these factors on the regression coefficient. A p value of less than 0.05 was considered to be significant. The estimates were presented with a 95% confidence interval or with standard error. The analyses were performed with SPSS for Windows (Release 12.0; SPSS, Chicago, Illinois).


   Results
Top
 Abstract
 Introduction
 Material and Methods
 Results
Comment
 References
 
Population Characteristics and Patient Status
The median patient age at the time of the operation was 66 years (range, 39 to 87). According to pathologic TNM staging, 66 patients were in stage IA, 79 patients in stage IB, 11 patients in stage IIA, 52 patients in stage IIB, 11 patients in stage IIIA, and 20 patients in stage IIIB. Forty-three patients received postoperative radiotherapy. None of the patients received any postoperative chemotherapy. The median follow-up of the whole group was 31 months (range, 2 to 135). Kaplan-Meier survival probability at 5 years for all patients was 51.9% (standard error = 3.9%).

At the time of final evaluation for this study, 139 patients were alive, of whom 114 were disease free, 103 had relapsed, 13 had a second primary, and 100 had died. Thirty-one percent of the patients with a relapse had a local recurrence, 67% had metastases, and 2% had both.

Growth Pattern Analysis
Based upon growth pattern classification, 161 patients (67.4%) had a tumor with a destructive growth pattern (Fig 2A and B), 33 (13.8%) with a papillary growth pattern (Fig 2C and D), and 45 (18.8%) with an alveolar growth pattern (Fig 2E and F).


Figure 2
View larger version (101K):
[in this window]
[in a new window]

  		Fig 2. Growth patterns of non-small cell lung cancer on a routine hematoxylin and eosin–stained and on CD34-Ki-67 double immunostained tissue sections. Destructive growth pattern (angiogenic): parenchymal structures of the lung are not preserved, but replaced by carcinoma cells and tumor-associated stroma (A) with formation of new blood vessels (B [arrow]). Papillary growth pattern (intermediate): the lung parenchyma has been preserved with formation of fibrovascular stalks originating from the alveolar septa (C) with formation of new blood vessels (D [arrow]). Alveolar growth pattern (nonangiogenic): the tumor cell nests fill the alveolar spaces with preservation of alveolar septa (E) and with cooption of alveolar blood vessels (F [arrow]) without formation of a tumor-associated stroma.

 
Comparison of Interface and Center of the Tumor
The ECPF, TCPF, MVD, and Chalkley count were significantly higher at the interface than in the center of the tumor (Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1 Proliferative and Angiogenic Activity of the Tumor at the Interface and the Center
 
Clinical, Biologic, and Pathologic Profile of Growth Patterns
There were no significant differences in distribution of clinical factors between the growth patterns (Table 2). The association of histologic types with growth patterns is presented in Table 3. The alveolar and destructive growth patterns were present in all the histologic types. The papillary growth pattern was exclusively seen in adenocarcinomas. However, adenocarcinomas expressed all three patterns of growth. A high number of large cell carcinoma was observed in tumors with an alveolar growth pattern.


View this table:
[in this window]
[in a new window]

  		Table 2 Association of Different Clinical Factors With Growth Patterns
 

View this table:
[in this window]
[in a new window]

  		Table 3 Association of Histologic Classification With Growth Patterns
 
With regard to the adenocarcinomas, the majority of the tumors with an alveolar growth pattern were of the solid histologic subtype, whereas the tumors with a papillary growth pattern were of the papillary or mixed (predominantly, more than 50%, with bronchioloalveolar features) histologic subtype. Even adenocarcinomas with mixed histologic subtypes showed a single growth pattern at their interface. For example, the adenocarcinomas with mixed histologic types with bronchioloalveolar carcinoma features showed a bronchioloalveolar carcinoma subtype at interface (papillary growth pattern) and another subtype in the center. All the tumors with bronchioloalveolar carcinoma features described are tumors with invasive components.

The association of biologic factors with growth patterns is presented in Table 4. Tumors with a papillary and an alveolar growth pattern tended to be smaller. The ECPF at the interface was significantly different between growth patterns, with each growth pattern representing a distinct ECPF profile. Alveolar growth pattern had the lowest ECPF and destructive growth pattern had the highest ECPF, whereas both growth patterns expressed a similar high TCPF. The papillary growth pattern had the lowest TCPF.


View this table:
[in this window]
[in a new window]

  		Table 4 Association of Different Biologic Factors With Growth Patterns
 
The MVD and Chalkley count were significantly higher in tumors with a papillary and an alveolar growth pattern.

The ratio of TCPF to ECPF was compared between the growth patterns. The alveolar growth pattern had a 10-fold higher ratio than the papillary and the destructive growth pattern.

Survival Analyses
Univariate survival analyses
The univariate survival analyses of different clinicopathologic factors for overall and disease-free survival are presented in Table 5. Significant factors were type of the operation, size of the tumor, T status, N status, tumor stage, growth pattern, and ECPF at the interface. Pneumonectomy, larger tumors, advanced T status, lymph node metastasis, advanced stage, alveolar growth pattern, and a low ECPF were associated with early death and recurrence of the disease in univariate analysis. Microvessel density and Chalkley count had no significant prognostic value. The Kaplan-Meier survival curves of the different growth patterns in stage IA and IB are presented in Figure 3A and B.


View this table:
[in this window]
[in a new window]

  		Table 5 Univariate Survival Analysis for Overall and Disease-Free Survival
 

Figure 3
View larger version (11K):
[in this window]
[in a new window]

  		Fig 3. (A) Growth pattern and stage IA. Overall survival according to growth pattern for patients in stage IA (destructive versus papillary growth pattern, p = 0.038, and destructive versus alveolar growth pattern, p = 0.788). (B) Growth pattern and stage IB. Overall survival according to growth pattern for patients in stage IB (destructive versus papillary growth pattern, p = 0.250, and destructive versus alveolar growth pattern, p = 0.008).

 
Multiple cox regression analyses
Different models were applied. Microvessel density and Chalkley count were not entered together in the model as they are both measurements of vascularization. The ECPF at the interface and the growth pattern were entered together as well as separately in the model. In model A, in which growth pattern was entered without ECPF (Table 6), older age, pneumonectomy, alveolar growth pattern, papillary growth pattern, and lymph node metastasis were independent predictors of poor overall survival. For disease-free survival, lymph node metastasis and alveolar growth pattern were independent predictors of early local recurrence or metastasis. Further analyses showed that while MVD and Chalkley count had no significant prognostic value, a low ECPF was consistently an independent predictor of poor prognosis and early local recurrence or metastasis. When ECPF and growth pattern were entered together in model B, growth pattern was outpowered by ECPF (Table 7): ECPF was one of the most significant independent prognostic factors.


View this table:
[in this window]
[in a new window]

  		Table 6 Adjusted a Hazard Ratios of Mortality and Recurrence in 239 Patients With NSCLC Undergoing Curative Resection by a Single Surgeon (Model A b )
 

View this table:
[in this window]
[in a new window]

  		Table 7 Adjusted a Hazard Ratios of Mortality and Recurrence in 239 Patients With NSCLC Undergoing Curative Resection by a Single Surgeon (Model B b )
 
Cox multiple regression analyses for stage IA and IB
For clinical relevance, a subanalysis was performed for patients in stage IA and IB. For overall survival, alveolar growth pattern and papillary growth pattern were independent predictors of poor prognosis (Table 8). For disease-free survival, none of the variables reached statistical significance (Table 8).


View this table:
[in this window]
[in a new window]

  		Table 8 Adjusted a Hazard Ratios of Mortality and Recurrence in 145 Patients in Stage IA and IB With NSCLC Undergoing Curative Resection by a Single Surgeon (Model C b )
 

   Comment
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Comment
References
 
We have provided a model in which NSCLC can be classified into three distinct biologic subtypes based on easily assessable morphologic characteristics on a routine hematoxylin and eosin–stained tissue section independent of histologic subtypes. We had previously shown that these growth patterns have a distinct prognostic profile. In this study, we have validated the hypothesis that morphologically different growth patterns have distinct angiogenic and proliferative profiles.

The growth pattern classification is based on morphologic characteristics of the interface of a tumor only. We have deliberately chosen the interface as it seems to be the most homogeneous region of the tumor. We have shown that the interface has a higher TCPF, ECPF, MVD, and Chalkley count compared with the center of the tumor, providing evidence that it has the highest tumor growth activity. The observation that the interface has the highest expression of growth-associated factors has been observed before in NSCLC. The highest mean MVD, the highest vascular endothelial growth factor bound to its receptor on endothelial cells (VEGF/KDR) activated microvessel density, and the highest lymph vessel density have been shown to be at the invading front of the tumors in NSCLC [13, 14]. Furthermore, the expression of other angiogenic factors, such as angiopoietin I and angiopoietin II have been mainly observed at the edge of the tumor in NSCLC [15].

The alveolar growth pattern has been described by several investigators before and its nonangiogenic character hypothesized based on its morphologic characteristics [4, 5, 16, 17]. However, no study has shown that the alveolar growth pattern is indeed nonangiogenic by measuring the fraction of proliferating endothelial cells. Moreover, no study to date has measured the rate of ongoing angiogenesis by determining the ECPF in a large group of patients with NSCLC.

We have shown that the alveolar growth pattern is nonangiogenic, as this growth pattern is characterized by a low ECPF and a high TCPF. The alveolar growth pattern has a 10-fold higher ratio of TCPF to ECPF than papillary and destructive growth pattern, suggesting an angiogenesis-independent growth. These data validate the hypothesis that tumors with an alveolar growth pattern proliferate in the alveolar spaces with preservation of the interalveolar septa and derive their oxygen supply from the septal blood vessels (cooption), without new stroma formation at the interface (Fig 2).

The papillary growth pattern is quite different from the alveolar and destructive growth pattern, as it is characterized by an intermediate ECPF, the lowest TCPF, highest MVD, and highest Chalkley count. The ratio of TCPF to ECPF was the lowest in papillary growth pattern, suggesting an angiogenic growth. Together with the high MVD and Chalkley count, these data corroborate the hypothesis that in tumors with a papillary growth pattern, there is extensive cooption of microvessels due to preservation of the normal densely vascularized lung parenchyma. Also, there is formation of new stromal stalks containing capillary vessels originating from the alveolar septa (Fig 2). Therefore, we consider this growth pattern to be intermediate between the angiogenic and nonangiogenic growth pattern.

The destructive growth pattern is characterized by the lowest MVD and Chalkley count but the highest ECPF at the interface. The TCPF is high but the TCPF/ECPF is low, suggesting an angiogenic growth. These data validate the hypothesis that in this growth pattern, there is destruction of the preexisting lung parenchyma and formation of a new stroma with brisk angiogenesis (Fig 2).

Microvessel density and Chalkley count are traditionally considered to be measurements of angiogenesis and numerous studies have shown an association between the MVD/Chalkley and poor prognosis for many cancers [18]. We have shown that the papillary and the alveolar growth pattern, which are characterized by a low ECPF, have the highest MVD and Chalkley count. This finding can be explained by preservation of the normal densely vascularized lung parenchyma and strongly suggests that MVD and Chalkley count, in contrast to ECPF, are no good measurements of angiogenesis in NSCLC.

The prognostic value of MVD has been a controversial subject in NSCLC [16, 19–25]. The largest study of 515 patients with the longest follow-up time found no prognostic value for MVD [26]. We could also not demonstrate any prognostic value of MVD or Chalkley count in our patient population.

The notion that nonangiogenic growth pattern is associated with poor prognosis seemed initially paradoxical, in view of numerous studies describing an association between a high microvessel density and poor prognosis. However, we have shown that growth patterns with the lowest ECPF (ongoing angiogenesis) in NSCLC have the highest microvessel density. Moreover, a low ECPF was an independent predictor of poor prognosis and recurrence of disease. This finding strongly suggests that in NSCLC a low degree of ongoing angiogenesis is predictive of poor prognosis.

There are several possible explanations for this finding. Angiogenesis induces a chaotic and inefficient vascularization. Angiogenesis leads to architecturally different blood vessels from those in physiologic conditions: vessels are irregularly shaped, dilated, and tortuous and can have dead endings. They are not organized into definitive venules, arterioles, and capillaries, but rather share features of all of these types haphazardly. The vascularization of the nonangiogenic growth pattern is therefore more efficient and denser, because it consists of coopted preexisting pulmonary vasculature. It contributes probably to a better growth and progression for nonangiogenic tumor tissue. A recent study by Hu and associates [17] has also shown that angiogenic and nonangiogenic growth patterns of NSCLC can be clustered apart by gene expression profiles. In this study, the global gene expression profile of tumors with an alveolar growth pattern was compared to tumors without an alveolar growth pattern. Nonangiogenic tumors had higher expression levels of genes concerned with mitochondrial metabolism. That implies that in nonangiogenic tumors oxygen consumption is more efficient by the high mitochondrial metabolism.

The clinical implications of the results presented here might be profound. Firstly, tumors with a nonangiogenic growth pattern will probably respond less to treatment with angiogenesis-inhibitors. This resistance to antiangiogenic therapy will probably occur irrespective of whether a tumor is entirely or partially (only at the interface) nonangiogenic: nonangiogenic clones might be selected by antiangiogenic treatment. The growth pattern can be used as a surrogate marker of angiogenesis, and pretreatment patient selection based on the growth pattern could enlarge the fraction of patients responding to angiogenesis inhibitors.

Secondly, the proposed classification has a strong independent prognostic value for overall and disease-free survival. The growth patterns are a possible explanation for differences in survival of patients in the same stage. In stage I resected NSCLC, many patients are long-term survivors, but some relapse early and die. The patients with early relapse and reduced overall survival are those with an alveolar and papillary growth pattern. These "at risk" patients may be identified based on the growth pattern and can be intensively followed or treated with high-dose adjuvant therapy.

Thirdly, we have clearly shown that each growth pattern has its own proliferative and angiogenic profile. The TCPF to ECPF ratio, roughly reflecting the degree of angiogenesis-independent tumor growth, was distinct for each growth pattern. These observations together with observed prognostic differences corroborate the hypothesis that these growth patterns represent distinct biologic subtypes.

In conclusion, the different growth patterns of NSCLC have distinct morphologic, angiogenic, and proliferative characteristics with distinct prognostic value. Further research has to be conducted to test whether the predictions and premises of this holistic simplification of NSCLC are valid.


   References
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Comment
 References
 

  1. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002 CA Cancer J Clin 2005;55:74-108.[Abstract/Free Full Text]
  2. Sardari Nia P, Van Marck E, Van Schil P. The prospect of biologic staging of non-small-cell lung cancer Clin Lung Cancer 2005;6:217-224.[Medline]
  3. Mountain C. Revisions in the international system for staging lung cancer Chest 1997;111:1710-1717.[Medline]
  4. Sardari Nia P, Colpaert C, Blyweert B, et al. Prognostic value of nonangiogenic and angiogenic growth patterns in non-small cell lung cancer Br J Cancer 2004;91:1293-1300.[Medline]
  5. 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.[Abstract]
  6. Pezzella F. Evidence of novel non-angiogenic pathway in breast-cancer metastasis Lancet 2000;355:1787-1788.[Medline]
  7. Vermeulen PB, Colpaert C, Salgado R, et al. Liver metastases from colorectal adenocarcinomas grow in three patterns with different angiogenesis and desmoplasia J Pathol 2001;195:336-342.[Medline]
  8. Vermeulen PB, Sardari Nia P, Colpaert C, et al. Lack of angiogenesis in lymph node metastases of carcinomas is growth pattern dependent Histopathology 2002;40:105-107.[Medline]
  9. Brambilla E, Travis WD, Colby TV, Corrin B, Shimosato Y. The new World Health Organization classification of lung tumors Eur Respir J 2001;18:1059-1068.[Abstract/Free Full Text]
  10. Van den Eynden G, Van der Auwera I, Van Laer S, et al. Angiogenesis and hypoxia in lymph node metastases is predicted by the angiogenesis and hypoxia in the primary tumor in patients with breast cancer Br J Cancer 2005;93:1128-1136.[Medline]
  11. Chalkley HW. Method for quantitative morphologic analysis of tissues J Natl Cancer Inst 1943;4:47-53.
  12. Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis and metastasis—correlation in invasive breast carcinoma N Engl J Med 1991;324:1-8.[Abstract]
  13. Giatromanolaki A, Koukourakis M, Sivridis E, O'Byrne K, Gatter KC, Harris AL. "Invading edge vs. inner" (edvin) patterns of vascularisation: an interplay between angiogenic and vascular survival factors defines the clinical behaviour of non-small cell lung cancer J Pathol 2000;192:140-149.[Medline]
  14. Koukourakis MI, Giatromanolaki A, Thorpe PE, et al. Vascular endothelial growth factor/KDR activated microvessel density versus CD31 standard microvessel density in non-small cell lung cancer Cancer Res 2000;60:3088-3095.[Abstract/Free Full Text]
  15. 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.[Abstract/Free Full Text]
  16. Offersen B, Pfeiffer P, Hamilton-Dutoit S, Overgaard J. Patterns of angiogenesis in nonsmall-cell lung carcinoma Cancer 2001;91:1500-1509.[Medline]
  17. Hu J, Bianchi F, Ferguson M, et al. Gene expression signature for angiogenic and nonangiogenic non-small-cell lung cancer Oncogene 2005;24:1212-1219.[Medline]
  18. Vermeulen PB, Gasparini G, Fox SB, et al. Second international consensus on the methodology and criteria of evaluation of angiogenesis quantification in solid human tumors Eur J Cancer 2002;38:1564-1579.[Medline]
  19. 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]
  20. Chandrachud LM, Pendleton N, Chisholm DM, Horan MA, Schor AM. Relationship between vascularity, age and survival in non-small-cell lung cancer Br J Cancer 1997;76:1367-1375.[Medline]
  21. Fontanini G, Lucchi M, Vignati S, et al. Angiogenesis as a prognostic indicator of survival in non-small-cell lung carcinoma: a prospective study J Natl Cancer Inst 1997;89:881-886.[Abstract/Free Full Text]
  22. 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.[Abstract]
  23. Duarte IG, Bufkin BL, Pennington MF, et al. Angiogenesis as a predictor of survival after resection for stage I non-small-cell lung cancer J Thorac Cardiovasc Surg 1998;115:652-658.[Abstract/Free Full Text]
  24. Matsuyama K, Chiba Y, Sasaki M, Tanaka H, Muraoka R, Tanigawa N. Tumor angiogenesis as a prognostic marker in operable non-small-cell lung cancer Ann Thorac Surg 1998;65:1405-1409.[Abstract/Free Full Text]
  25. Ohta Y, Tomita Y, Oda M, Watanabe S, Murakami S, Watanabe Y. Tumor angiogenesis and recurrence in stage I non-small cell lung cancer Ann Thorac Surg 1999;68:1034-1038.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Paul Van Schil
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sardari Nia, P.
Right arrow Articles by Van Marck, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sardari Nia, P.
Right arrow Articles by Van Marck, E.
Related Collections
Right arrow Lung - cancer

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
ANN THORAC SURG ASIAN CARDIOVASC THORAC ANN EUR J CARDIOTHORAC SURG
J THORAC CARDIOVASC SURG ICVTS ALL CTSNet JOURNALS