Heparanase is over-expressed in lung cancer and inversely correlates with patient's survival

Abstract

Condensed abstract

We report that heparanase is over-expressed in 75% of lung cancer patients, correlating with N-, and M-stages (p=0.04 and 0.01, respectively), and inversely correlating with patients’ survival (p=0.007). These findings suggest that heparanase expression is decisive for lung cancer patients’ prognosis, likely due to heparanase mediated tumor cell dissemination by blood and lymph vessels.

Background

Heparanase is an endo-β-D-glucuronidase capable of cleaving heparan sulfate (HS) side chains at a limited number of sites, yielding HS fragments of still appreciable size (~5–7 kDa). Heparanase activity has long been detected in a number of cell types and tissues. Importantly, heparanase activity correlates with the metastatic potential of tumor-derived cells, attributed to enhanced cell dissemination as a consequence of HS cleavage and remodeling of the extracellular matrix barrier.

Methods

We examined heparanase expression in 114 lung cancer patients by means of immunohistochemistry, and correlated clinical-pathological data with heparanase immunostaining and cellular localization.

Results

We report that heparanase is over-expressed in 75% of the patients. Heparanase expression correlated with lung cancer N-, and M-stages (p=0.04 and 0.01, respectively), and inversely correlated with patients’ survival (p=0.007). Notably, this adverse effect is largely dependent on the cellular localization of heparanase. Thus, while cytoplasmic staining of heparanase is associated with poor prognosis, nuclear heparanase predicts a favorable outcome of lung cancer patients.

Conclusions

These findings suggest that heparanase expression and cellular localization are decisive for lung cancer patients' prognosis, likely due to heparanase mediated tumor cell dissemination by blood and lymph vessels.

Introduction

Heparan sulfate (HS) glycosaminoglycans bind to and assemble extracellular matrix (ECM) proteins (i.e., laminin, fibronectin, collagen type IV) and thereby contribute significantly to the ECM self assembly and integrity. HS also tether a multitude of growth factors, chemokines, cytokines and enzymes to the ECM and cell surface, providing a low affinity storage depot ^1–3. Cleavage of HS side chains is therefore expected not only to alter the integrity of the ECM, but also to release HS-bound biological mediators. The ECM provides an essential physical barrier between cells and tissues, as well as a scaffold for cell growth, migration, differentiation and survival, and undergoes continuous remodeling during development and in certain pathological conditions such as wound healing and cancer ⁴. ECM-remodeling enzymes are thus expected to profoundly affect cell and tissue function.

Heparanase is an endo-β-D-glucuronidase capable of cleaving HS side chains at a limited number of sites, yielding HS fragments of still appreciable size (~5–7 kDa) ^{5, 6}. Heparanase activity has long been detected in a number of cell types and tissues. Importantly, heparanase activity correlates with the metastatic potential of tumor-derived cells, attributed to enhanced cell dissemination as a consequence of HS cleavage and remodeling of the ECM barrier ^{7, 8}. More recently, heparanase up-regulation was documented in an increasing number of human carcinomas and hematological malignancies ^9–11. In many cases, heparanase expression correlated with increased tumor metastasis, vascular density, and shorter post operative survival rate, thus providing a strong clinical support for the pro-metastatic function of the enzyme ^{9, 10}, making it an attractive target for the development of anti-cancer drugs ^{9, 12–16}.

Lung cancer is the leading cause of cancer death worldwide. Despite the introduction of new agents and schedules, chemotherapy still obtains unsatisfactory overall response rates, rare complete remissions and responses of relatively short duration ¹⁷. Therefore, better understanding of basic biology aspects of the disease is required for the development of new therapeutic approaches ¹⁸. The expression of heparanase in lung cancer has not been so far reported. Here, we examined heparanase expression in 114 lung cancer patients, and correlated clinical-pathological data with heparanase immunostaining and cellular localization. We report that heparanase expression is induced in 75% of the patients and inversely correlates with patients' survival. This adverse effect is largely dependent on the cellular localization of heparanase.

Materials and methods

Experimental design

The study included 114 patients with lung cancer that were diagnosed in the Department of Thoracic Surgery, Rambam Health Care Campus, Haifa, Israel, whose archival paraffin-embedded pathological material was available for immunohistochmical analysis. The study protocol was approved by the Institutional Review Board. The clinical data of all patients was reviewed and patients were re-staged according to the AJCC 2006 staging system. Clinical data included demographics, site of tumor, tumor-node-metastasis (TNM) staging, treatment modality, status at the end of the study (dead or alive), failure (local, regional or distant), time to failure, follow-up and survival.

Immunostaining

Staining of formalin-fixed, paraffin-embedded 5 micron sections was performed essentially as described ^19–21. Briefly, slides were deparaffinized, rehydrated and endogenous peroxidase activity was quenched (30 min) by 3% hydrogen peroxide in methanol. Slides were then subjected to antigen retrieval by boiling (20 min) in 10 mM citrate buffer, pH 6.0. Slides were incubated with 10% normal goat serum (NGS) in phosphate buffered saline (PBS) for 60 min to block non specific binding and incubated (20 h, 4°C) with anti heparanase 733 polyclonal antibody ²² diluted 1:100, or the D2–40 monoclonal antibody (1:100; Dako, Glostrup, Denmark), which specifically decorates lymphatic endothelial cells ²³. Slides were extensively washed with PBS containing 0.01% Triton X-100 and incubated with a secondary reagent (Envision kit) according to the manufacturer’s (Dako) instructions. Following additional washes, color was developed with the AEC reagent (Dako), sections were counterstained with hematoxylin and mounted, as described ^19–21.

Immunostained specimens were examined by a senior pathologist (IN) who was blind to clinical data of the patients, and heparanase staining was scored according to the intensity (0: none, 1: weak; 2: moderate-strong) and percentage (extent staining) of tumor cells that were stained (0: <10%; 1: 10–50%; 2>50%). Lymph vessel density (LVD) was evaluated by counting D2–40-positive lymph vessels in the tumor section (0: < 1 lymph vessel per section; +1: 2–5 lymph vessels per section; +2: >5 lymph vessels per section). Specimens that were similarly stained with pre-immune serum, or subjected to the above procedure but lacking the primary antibody, yielded no detectable staining. For statistical analysis, we combined the cases diagnosed as heparanase negative with cases diagnosed as weak heparanase staining compared to cases with strong heparanase intensity and/or extent (subgroups 0+1 vs. subgroup 2).

Statistical analysis

Univariate association between heparanase staining (intensity and extent of staining) and localization (nuclear vs. cytoplasmic), LVD, and clinical and pathological parameters as well as patients’ outcome, were analyzed using Chi Square tests (Pearson, Fisher exact test). Multivariable logistic regression was performed to detect independent parameters that affect patients’ status and to estimate relevant Odd's ratio (OR) with 95% confidence interval (CI). Univariate association with survival was evaluated by Kaplan Meier curves, and tested using Log-Rank test. A multivariable Cox’s proportional hazard model was perform with stepwise selection, to identify independent predictors of survival (P for enter and p to stay were set as 0.1). The model included all parameters with p<0.2 by the univariate analysis.

Results

Heparanase is overexpressed in lung cancer and predicts poor outcome

114 patients (62 males and 52 females) were included in this study; median age at diagnosis was 66.8±10 years. Mean follow-up was 22±16 months for the entire group and 64.8±11.05 months for surviving patients. Clinical description of the patients is presented in table 1. Adenocarcinoma was the predominant type (87%; 99/114), whereas bronchoalveolar carcinoma consisted 13% of the patients (15/114). Metastatic nodes were found in 31 patients, 12 with one node (N1) and 19 with two or more metastatic nodes (N2–3, Table 1). Eight patients exhibited metastatic disease at presentation (M1, Table 1).

Table 1.

Clinical description of patients

Parameter	Number of patients (%)
Type
Adenocarcinoma	99 (87)
Bronchoalveolar carcinoma	15 (13)
T stage
T1	52 (46)
T2	48 (42)
T3–4	14 (12)
N stage
N0	83 (73)
N1	12 (10)
N2–3	19 (17)
M stage
M0	106
M1	8

Normal looking lung tissue far from the tumor lesion did not stain for heparanase (Fig. 1A, B). In contrast, positive heparanase staining was found in 75% (85/114) of the tumor specimens, while 25% (29/114) of the specimens were negative. There was no difference among genders in heparanase staining: 16 males (25.8% of males) and 13 females (25%) were stained negative for heparanase. The heparanase-positive group was further categorized according to the intensity and extent of staining. Thus, weak staining (+1) was found in 55% (47/85) of the positive specimens (Fig. 1C, D), while 45% (38/85) exhibited strong staining (+2) for heparanase (Fig. 1E, F). According to the extent criteria (see "Materials and Methods"), 66% (56/85) of the heparanase positive specimens had a high extent score (+2), whereas 34% (29/85) had low extent score (+1). Strong staining intensity (+2) of heparanase was far more prevalent in adenocarcinoma than in bronchoaleolar carcinoma, a type of lung cancer associated with a markedly better prognosis compared with invasive adenocarcinoma. Here, only 13% of the bronchoalveolar carcinomas exhibited strong staining intensity of heparanase (+2) compared with 36% of the adenocarcinomas (Table 2). Although statistical significance is borderline (p=0.07), likely due to the relatively small number of bronchoaleolar carcinomas, the results suggest that heparanase expression is associated with a more advanced and aggressive type of the disease. Indeed, heparanase staining in terms of extent and even more so of intensity, correlated with lymph nodes (N-stage, p=0.037) and distant (M-stage, p=0.009) metastases (Table 2). For example, while 70% of the patients without distant metastasis (M0) exhibited no or weak staining of heparanase (0+1), 75% of patients with distant metastasis (M1) stained strongly for heparanase (Table 2), differences that are highly statistically significant (p=0.009). In agreement with this finding is the inverse correlation between heparanase staining and patients' overall survival (Table 3; Fig. 2).

Figure 1.

Immunohistochemical staining of heparanase in lung tumor specimens. Formalin-fixed, paraffin-embedded 5 micron sections of lung tumors were subjected to immunostaining of heparanase, applying anti-heparanase 733 polyclonal antibody, as described under 'Materials and Methods'. Shown are representative photomicrographs of normal looking lung tissue adjacent to the tumor lesion stained negative for heparanase (A, B), and tumor specimens that were stained weakly (C, D) or strongly (E, F) for heparanase. Tumor specimens exhibiting nuclear localization of heparanase (G, arrows) or stained negative for heparanase (H) are also shown. Original magnification: A, C, E × 100, B, D, F-H × 200.

Table 2.

Association of heparanase staining (intensity and extent) with clinical and histological parameters (univariate analysis)

Parameter	Heparanase Intensity		Heparanase Extent		Total
	0–1	2	0–1	2
Histology	P=0.07		P=0.19
Bronchoalveolar carcinoma	13 (87%)	2 (13%)	10 (67%)	5 (33%)	15
Adenocarcinoma	63 (64%)	36 (36%)	48 (48%)	51 (52%)	99
N Stage	p=0.03		P=0.04
N0	60 (72%)	23 (28%)	47 (57%)	36 (43%)	83
N1–3	16 (52%)	15 (48%)	11 (35%)	20 (65%)	31
M stage	P=0.009		P=0.01
M0	74 (70%)	32 (30%)	57 (54%)	49 (46%)	106
M1	2 (25%)	6 (75%)	1 (13%)	7 (87%)	8

Table 3.

Univariate 5 years overall survival of lung cancer patients

Parameter	Patients at risk	5 years overall survival	p
T stage			0.002
T 1	52	76%
T 2	48	69%
T 3–4	14	32%
N stage
N0	83	79%	0.003
N1–3	31	39%
Heparanase intensity
0 –1	84	75%	0.01
2	38	52%
Heparanase extent
0 –1	58	86%	0.007
2	56	50%
Heparanase localization
0	29	92%	0.02
Nuclear	32	72%
Cytoplasmic	53	55%

Figure 2.

Survival analysis. Overall survival stratified by heparanase intensity (A; 0+1 vs. 2), extent (B; 0+1 vs. 2), and localization (C; 0 vs. nuclear vs. cytoplasmic staining of heparanase) (Kaplan-Meier survival plots).

Univariate analysis revealed that the T-stage is the most significant parameter for the survival of lung cancer patients (p=0.002), followed by the N-stage (p=0.003), heparanase extent (p=0.007; Fig. 2A), and heparanase intensity (p=0.01; Fig. 2B) (Table 3), suggesting that heparanase plays an important role in lung cancer progression. Gender was not found to be a significant parameter for overall survival (Log-Rank Test, p = 0.14).

Close examination of the heparanase positive group revealed a distinct cellular localization pattern. Thus, in 62% of the specimens (53/85) heparanase staining appeared cytoplasmic (Fig. 1C–F), while in the rest 38% (32/85) heparanase was found also in the cell nucleus (Fig. 1G). Interestingly, nuclear localization of heparanase was associated with a favorable outcome of lung cancer patients. Clearly, patients stained negatively for heparanase, or exhibiting nuclear localization of the enzyme had a significantly higher overall survival than patients endowed with only cytoplasmic staining (p=0.02; Table 3; Fig. 2C). Cox’s proportional hazard model was subsequently performed for overall survival. The most significant and independent parameters that influenced overall survival were T-stage (0.0004) followed by heparanase localization (p=0.002), further supporting a critical role of heparanase in lung cancer.

Heparanase staining is associated with lymph vessel invasion

The close association between heparanase staining and tumor metastasis (Table 2) suggests that heparanase facilitates the dissemination and mobilization of tumor cells by lymph (N-stage) and blood vessels (M-stage). We utilized monoclonal antibody D2–40, which specifically decorates the lymphatic endothelium ²³, to evaluate lymph vessels density (LVD) and lymphatic invasion by tumor cells, and analyzed its association with clinical parameters and heparanase staining. Seven biopsies (6%) were found negative for tumor lymph vessels; 71 exhibited a low number of lymph vessels (+1; Fig. 3A), and 36 percents showed a high number of lymphatics (+2; Fig. 3B). The functionality of the lymphatic system in lung carcinoma is clearly evident by the appearance of metastatic tumor cells, often in the form of tumor emboli, inside lymph vessels (Fig. 3, C, D), noted in 31 patients. LVD was found to be the most significant parameter for lymphatic invasion by tumor cells (p=0.002; Table 4), as intuitively anticipated. Notably, heparanase staining intensity correlated with lymphatic invasion. Thus, 21% of the specimens that stained negatively (0) or weakly (+1) for heparanase exhibited lymph vessels invaded by tumor cells, compared with 42% of the specimens that stained strongly for heparanase (+2; Table 4), differences that are statistically significant (p=0.01). These findings suggest that heparanase expression and cellular localization are decisive for lung cancer patients' prognosis, likely due to tumor cell dissemination by blood and lymph vessels.

Figure 3.

Lymph angiogenesis and lymphatic invasion by lung tumor cells. Sections of lung tumors were stained with monoclonal antibody D2–40, which specifically decorate lymphatic endothelial cells. Shown are representative photomicrographs of lung tumors exhibiting low (A) or high (B) lymph vessels density (LVD), and lymph vessels invasion by tumor cells (C, D). Note, disruption of the lymphatic endothelium by invading cancer cells. Original magnifications: A, B × 100, C × 200, D × 400.

Table 4.

Number of lymph vessels harboring metastatic tumor cells in correlation with LVD, heparanase intensity, and N stage

Parameter	Tumor cells in LV		Total	p
	No	Yes
Lymph vessel density				0.002
0	7 (100)	0	7
1	57 (80)	14 (20)	71
2	19 (53)	17 (47)	36
Heparanase intensity				0.01
0–1	62 (79)	16 (21)	78
2	21 (58)	15 (42)	36

Discussion

Lung cancer incidence and mortality have slowly but steadily declined in recent years, due to a decrease in cigarette smoking, particularly among men. However, the disease remains the leading cause of cancer deaths in the United States, and is expected to kill more than 160,000 people in 2007 ²⁴. Although some targeted therapies are proving effective against lung cancer, the specific characteristics of an individual's tumor is expected to better predict prognosis and response to treatment, an approach known as personalized medicine ²⁴. Our results suggest that heparanase expression may be considered among the molecular parameters that define lung tumors, yet clinical research is required to elucidate the association between heparanase levels and response to treatment. This is largely supported by the association between heparanase levels and tumor metastasis (Table 2), and the inverse correlation with patients' survival (Fig. 2). Heparanase activity has long been implicated in tumor metastasis, a notion that is now well supported experimentally ^{25, 26} and clinically ^{10, 13–16}. This function of heparanase was largely attributed to enhanced cell invasion through the sub-epithelial and sub-endothelial basement membranes as a consequence of HS cleavage and remodeling of the ECM ²⁷, and the formation of new blood vessels that mobilize metastatic tumor cells to distant organs ⁹. In lung carcinoma, as well as in gastric ^{28, 29}, esophageal ³⁰, endometrial ^{31, 32}, and bladder³³ carcinomas, heparanase induction correlated with increased lymph nodes metastases, indicating dissemination of tumor cells primarily through the lymphatic system. Indeed, invasion of lymph vessels by tumor cells was associated with heparanase staining intensity (Table 4). Although anticipated, this type of correlation is demonstrated here for the first time. This is significant finding since lymph vessels, in contrast with blood vessels, are not surrounded by a supporting basement membrane ³⁴. Thus, heparanase may facilitate lymphatic invasion by means other than its enzymatic activity required to break through ECM barriers ²⁷. For example, locally secreted heparanase may activate signal transduction, resulting in loosen endothelial cell-cell contacts. This effect may be mediated by VEGF induction and Src activation ³⁵, shown to modulate endothelial cell permeability and integrity ^{36, 37}, possibly via clustering and internalization of syndecan family members ³⁸, or by syndecan shedding ³⁹. Noticeably, penetrating tumor cells appear to markedly disrupt the lymphatic endothelium (Fig. 3, C, D) in a manner significantly different from the elegant crawling of transmigrating immune cells between neighboring endothelial cells. Studies examining the effect of heparanase on lymphatic endothelial cell-cell interaction and monolayer integrity are currently in progress.

Not only heparanase levels, but also its cellular localization appear critical for lung cancer patients' survival and accordingly for the classification of an individual's tumor. Thus, while cytoplasmic heparanase predicts poor survival, nuclear heparanase was associated with a favorable outcome of lung cancer patients (Table 3; Fig. 2). Cox’s Propotional Hazard Model for overall survival showed that heparanase localization is a highly significant parameter that influenced overall survival, independent of the T or N stage. Moreover, N stage became insignificant upon the inclusion of heparanase localization in the Cox’s model.

This is in agreement with a favorable outcome of head and neck ²⁰, gastric ⁴⁰, and esophageal ³⁰ cancer patients exhibiting nuclear heparanase. Furthermore, nuclear heparanase was correlated with sustained differentiation of gastric and esophageal carcinomas ^{30, 40}, which was recapitulated, to some extent, in vitro ⁴¹. It is tempting to suggest that heparanase mediated cleavage of HS, shown previously to be localized in the cell nucleus ^{42, 43}, affects gene transcription. Alternatively, the interaction between heparanase and HS may modulate the formation of protein complexes on the chromatin and thereby regulate the expression of selected genes. Studies examining the role of nuclear heparanase in cell differentiation are currently in progress.

Taken together, our results describe, for the first time a role for heparanase in lung cancer, thus contributing to the basic understanding of this deadly disease. It is hoped that heparanase inhibitors ^{16, 44, 45} will add to the limited arsenal currently available against cancer in general and lung carcinoma in particular.