Ann Thorac Surg 1998;66:1915-1918
© 1998 The Society of Thoracic Surgeons
Original Articles
The clinical significance of hepatocyte growth factor for non–small cell lung cancer
Jill M. Siegfried, PhDa,
Lisa A. Weissfeld, PhDb,
James D. Luketich, MDd,
Robert J. Weyant, PhDc,
Christopher T. Gubish, MSa,
Rodney J. Landreneau, MDa,b,c,d
a Department of Pharmacology and University of Pittsburgh Cancer Institute, Lung Cancer Basic Science Program, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
b Department of Biostatistics, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
c Department of Dental Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
d Department of Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
Address reprint requests to Dr Siegfried, Department of Pharmacology, University of Pittsburgh, E1340 Biomedical Science Tower, Pittsburgh, PA 15261
e-mail: (siegfrie{at}server.pharm.pitt.edu)
Presented at the Thirty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Feb 3–5, 1997.
 |
Abstract
|
|---|
Background. Hepatocyte growth factor (HGF) is a cytokine that is released after injury. It is a paracrine factor that is produced by mesenchymal cells; epithelial and endothelial cells respond to HGF through its receptor, the c-met protein. Hepatocyte growth factor induces cell growth and cell movement and is also highly angiogenic. Evidence from breast cancer patients suggests that HGF is a negative prognostic indicator for breast cancer and is associated with invasive disease.
Methods. We measured the HGF content in tumor tissue from 56 non–small cell lung cancer patients using the Western blot technique. The amount of HGF in tumor extracts was quantitated by densitometry after transfer of proteins to nitrocellulose and exposure to antibodies. Survival curves were generated based on clinical information obtained for each patient.
Results. Our data indicate that HGF is also a negative prognostic indicator in lung cancer. As in the study of breast cancer patients, HGF was associated with recurrence and poor survival; the relative risk was seen to increase with increasing HGF tumor content. At levels of HGF greater than 100 units, the relative risk was 10, compared with that in patients with an HGF level of 1 unit. Node-negative patients with an elevated HGF tumor content had a significantly poorer outcome than node-positive patients with a low HGF tumor content. The same relationship was observed if the patients were stratified by stage: elevated HGF was associated with stage I patients whose disease recurred and who died of their disease, and stage I patients with elevated HGF had a worse survival than higher stage patients with a low level of HGF.
Conclusions. These results suggest that elevated HGF may predict a more aggressive biology in non–small cell lung cancer patients. The level of HGF may be useful as an indicator of high risk in early stage lung cancer patients.
|
Introduction
|
|---|
Hepatocyte growth factor (HGF) was originally discovered as a blood-borne factor that was released during liver regeneration [1]. It is now recognized that HGF is found in many organs, including the mammary gland, lung, and kidney, as well as the liver, and that it is released after injury both locally and in distant organs [2, 3]. Hepatocyte growth factor is produced by mesenchymal cells, but acts on cells of epithelial and endothelial origin [4]. It has multiple effects, including mitogenic, motogenic, and angiogenic effects [5–7]. The receptor for HGF, the c-met protooncogene product, is expressed on most epithelial cells, including those of the lung [8]. The receptor is a tyrosine kinase, and the signaling pathway involves both the ras gene product and the rho gene product.
High levels of HGF in ductal carcinomas of the breast are associated with poor survival [9]. For breast cancer, the relative risk of elevated HGF (the one-third of patients with the highest levels) was found to be 3.5, and node-negative patients with a high HGF tumor content had a worse outcome than node-positive patients with low levels of HGF [9]. Hepatocyte growth factor was also found to be higher in invasive ductal carcinoma of the breast than in ductal carcinoma in situ [10]. The presence of von Willebrand factor, a marker of blood vessels, was correlated with HGF content [10], implying that well-vascularized tumors have a higher HGF content. The combined effect of HGF to stimulate angiogenesis, cell movement, and cell division could contribute to invasion and metastasis.
|
Material and methods
|
|---|
Tumor tissue was collected sequentially from 56 patients undergoing curative lung resection. The final diagnoses of the carcinomas examined were 47 adenocarcinomas, three adenosquamous carcinomas, three bronchiolo-alveolar carcinomas, and three squamous cell carcinomas. Pathologic staging was carried out according to standard criteria [11]. During follow-up, 42 patients were still living and 14 deaths had occurred. Mean follow-up period for censored patients was 29 months. Mean follow-up period for deceased patients was 12.2 months.
Tissue was frozen at the time of resection and stored at -70°C. Frozen tissue was homogenized and extracted with 50 mmol/L Tris HCl, pH 7.4, in the presence of 0.25% Triton X-100, as described by Yamashita et al. [9]. Quantification of HGF was carried out by Western blot analysis. Equal amounts of tumor protein were subjected to electrophoresis, as described by Singh-Kaw et al. [8]. The proteins were transferred to nitrocellulose membranes and standard Western procedure was performed using a goat polyclonal antiHGF antibody (R & D Systems, Minneapolis, MN). Horseradish-peroxidase–conjugated goat-antiHGF antibody was used as the secondary antibody. Densitometric scans were done comparing lanes with known recombinant HGF standards, and the results were expressed as ng HGF/40 µg tumor protein, which was the amount of HGF applied per lane during electrophoresis.
The relationship of HGF content to clinical outcome was analyzed using one-way ANOVA. Kaplan-Meier survival curves [12] were also generated for overall and disease-free survival. The prognostic significance of HGF content, assessed using the log-rank and Wilcoxon tests, as well as a Cox proportional hazards model [13], were fit to the data to assess the effect of HGF in the presence of other factors.
|
Results
|
|---|
The amount of HGF detected in tumor extracts from different patients ranged from 1 to 195 ng/40 µg protein. The median HGF content was 15.3 ng/40 µg protein and the mean was 27.2 ng/40 µg protein. Both the median and the mean were used as cutoff points to analyze survival parameters in the 56 patients. The results were statistically significant whether the analysis was done for the 53 patients with some type of adenocarcinoma or for those patients and the 3 patients with squamous cell carcinoma. When patients were divided by stage, there was no statistical difference between either the mean or the median HGF content, demonstrating that HGF level was independent of stage, and is not simply a measure of stage. There was an association between recurrence during the follow-up period and HGF level (Table 1). For all patients, those with no evidence of disease had a mean HGF level of 15.5 ng/40 µg protein. For those whose disease recurred within the follow-up period, the mean HGF level was 46.7 ng/40 µg protein (p = 0.009). The patients analyzed included 34 stage I patients, 9 stage II patients, and 13 stage IIIa patients. We also analyzed the stage I patients separately (Table 1). For stage I patients only, the mean HGF was 12.8 ng/40 µg protein in patients with no evidence of disease and 46.1 ng/40 µg protein in those whose disease recurred during the follow-up period (p = 0.001). The same association was found for median HGF (Table 1). This demonstrates that higher HGF levels are associated with aggressive disease at all stages of lung cancer.
In analysis of survival, elevated HGF was found to be associated with poor disease-free and overall survival. When using either the mean, the median, or the upper third of HGF level as the cutoff, elevated HGF was always associated with poor outcome. Figure 1 shows the stratification for disease-free survival using the median HGF level as the cutoff point between high and low HGF (p = 0.01, log-rank test and Wilcoxon test); the same relationship was found with the mean (not shown). Overall survival was also stratified by nodal status and stage (Figs 2, 3). In Figure 2, survival is shown for node-negative and node-positive patients, divided by HGF level. The cutoff used was the median HGF. Curve 1 is the survival of node-negative patients who had low HGF. Curve 2 is the survival of node-positive patients who had low HGF. Curves 3 and 4 are node-negative and node-positive patients, respectively, who had elevated HGF. These data illustrate that patients who are node-negative but have elevated HGF have poorer outcomes than node-positive patients with low HGF (p = 0.05 by the log-rank test and 0.06 by the Wilcoxon test). This again shows that elevated HGF is associated with a more aggressive biology that is unrelated to tumor burden. Thus it could be a useful predictor of high risk for node-negative patients. The same relationship was found in Figure 3 when the patients were stratified by stage. Again, those with low HGF (curves 1 and 2) had much better survival than those with elevated HGF (curves 3 and 4), irrespective of stage (p = 0.01 by the log-rank test and the Wilcoxon test).
View larger version (20K):
[in this window]
[in a new window]
|
Fig 1. Disease-free survival for non–small cell lung cancer patients stratified by hepatocyte growth factor (HGF) content. Median HGF was used as the cutoff between low and high HGF. There are 28 patients in each group.
|
|
View larger version (21K):
[in this window]
[in a new window]
|
Fig 2. Overall survival of non–small cell lung cancer patients stratified by HGF content and nodal status. Median HGF was used as the cutoff between low and high HGF. Curve 1 = node-negative and low HGF (n = 20); curve 2 = node-positive and low HGF (n = 8); curve 3 = node-negative and high HGF (n = 16); curve 4 = node-positive and high HGF (n = 12). (HGF = hepatocyte growth factor).
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Fig 3. Overall survival of non–small cell lung cancer patients stratified by HGF content and stage. Median HGF was used as the cutoff between low and high HGF. Curve 1 = stage I and low HGF (n = 18); curve 2 = stage II and IIIa and low HGF (n = 10); curve 3 = stage I and high HGF (n = 16); curve 4 = stage II and IIIa and high HGF (n = 12). (HGF = hepatocyte growth factor).
|
|
In a multivariate Cox analysis, HGF emerged as the strongest component in survival, compared with nodal status, T status, stage, age, sex, or histologic type. The significance of HGF was consistently p = 0.0001, regardless of the number of other parameters examined or the order in which the parameters were entered into the model. The best model was one of continuous risk, in which the risk was expressed per unit of HGF. In this model the ß coefficient, an estimate of the risk, was 0.021/ng HGF. Thus the relative risk continuously increased as the HGF level increased, and can be calculated for any level of HGF relative to another using the equation: relative risk = eß(HGF1 - HGF2). Applying this equation, the relative risk of an HGF level of 15 ng/40 µg protein was 2.7, compared with an HGF level of 1 ng/40 µg protein. At 50 ng HGF/40 µg protein, the relative risk was 2.9, compared with 1 ng of HGF/40 µg protein, and at 100 ng HGF/40 µg protein, the relative risk was approximately 10, compared with 1 ng/40 µg protein. In contrast, the relative risk of T status greater than 1 was 3.7 in this patient population, and the relative risk of age greater than 65 years was 3.8. The relative risk of stage was 1.8. Stage did not emerge as a strong prognostic indicator in this study, most likely because the majority of the patients were stage I and half the deaths occurred in stage I patients. This observation underscores the need for more accurate predictors of the survival of stage I patients.
|
Comment
|
|---|
These results confirm the findings in breast cancer patients described earlier and suggest that HGF level can be useful as an independent predictor of aggressive tumor biology. Because HGF is a known stimulator of invasion, and is both mitogenic and angiogenic, the presence of high HGF levels in the primary tumor may be an indicator that occult metastases have already occurred. Preliminary evidence from our laboratory suggests that cytokines released by the tumor may stimulate lung fibroblasts to produce HGF. If tumor cells can induce local HGF release as they migrate out of the primary tumor, this could also contribute to progression of disease.
Our findings suggest that the HGF level in the primary tumor could identify early stage patients who are at risk for recurrence and death. These individuals may be candidates for adjuvant therapy, and may be as much at risk as stage II or IIIa patients.
|
Acknowledgments
|
|---|
This research was supported by a grant from the American Cancer Society (CN-163), awarded to J.M.S.
|
References
|
|---|
-
Zarnegar R., Michalopoulos G. Purification and biological characterization of human hepatopoietin A, a polypeptide growth factor for hepatocytes. Cancer Res 1989;49:3314-3320.[Abstract/Free Full Text]
-
Igawa T., Kanda S., Kanetake H., et al. Hepatocyte growth factor is a potent mitogen for cultured rabbit renal tubular epithelial cells. Biochem Biophys Res Commun 1991;174:831-838.[Medline]
-
Yamashita J., Ogawa M., Beppu T. Immunoreactive hepatocyte growth factor is present in tissue extracts from human breast cancer but not in conditioned medium of human breast cancer cell lines. Res Commun Chem Pathol Pharmacol 1993;82:249-252.[Medline]
-
Rosen E., Meromsky L., Setter E., Vinter D.W., Goldberg I.D. Smooth muscle-derived factor stimulates mobility of human tumor cells. Invasion Metastasis 1990;10:49-64.[Medline]
-
Gherardi E., Stoker M. Hepatocytes and scatter factor. Nature (Lond) 1990;346:228.[Medline]
-
Morimoto A., Okamura K., Hamanaka R., et al. Hepatocyte growth factor modulates migration and proliferation of human microvascular endothelial cells in culture. Biochem Biophys Res Commun 1991;179:1042-1049.[Medline]
-
Weidner K.M., Behrens J., Vandekerckhove J., Birchmeier W. Scatter factor: molecular characteristics and effects on the invasiveness of epithelial cells. J Cell Biol 1990;111:2097-2108.[Abstract/Free Full Text]
-
Singh-Kaw P., Zarnegar R., Siegfried J.M. Stimulatory effects of hepatocyte growth factor on normal and neoplastic human bronchial epithelial cells. Am J Physiol 1995;268(Lung Cell Mol Physiol 12):L1012-L1020.[Abstract/Free Full Text]
-
Yamashita J-i, Ogawa M., Yamashita S-i, et al. Immunoreactive hepatocyte growth factor is a strong and independent predictor of recurrence and survival in human breast cancer. Cancer Res 1994;54:1630-1633.[Abstract/Free Full Text]
-
Yao Y., Jin L., Fuchs A., et al. Scatter factor protein levels in human breast cancers: clinicopathological and biological correlations. Am J Pathol 1996;149:1707-1717.[Abstract]
-
Mountain C.F. The new international system for staging lung cancer. Chest 1986;89:225S-233S.[Free Full Text]
-
Kaplan E.L., Meier P. Nonparametric estimation for incomplete observations. J Am Stat Assoc 1958;53:457-481.
-
Cox D.R. Regression models and life tables (with discussion). JR Stat Soc Ser B 1972;34:187-220.
This article has been cited by other articles:
|
|
|
|
 
A. Coxon, K. Rex, S. Meyer, J. Sun, J. Sun, Q. Chen, R. Radinsky, R. Kendall, and T. L. Burgess
Soluble c-Met receptors inhibit phosphorylation of c-Met and growth of hepatocyte growth factor: c-Met-dependent tumors in animal models
Mol. Cancer Ther.,
May 1, 2009;
8(5):
1119 - 1125.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. A. Zucali, M. G. Ruiz, E. Giovannetti, A. Destro, M. Varella-Garcia, K. Floor, G. L. Ceresoli, J. A. Rodriguez, I. Garassino, P. Comoglio, et al.
Role of cMET expression in non-small-cell lung cancer patients treated with EGFR tyrosine kinase inhibitors
Ann. Onc.,
September 1, 2008;
19(9):
1605 - 1612.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Kong-Beltran, S. Seshagiri, J. Zha, W. Zhu, K. Bhawe, N. Mendoza, T. Holcomb, K. Pujara, J. Stinson, L. Fu, et al.
Somatic Mutations Lead to an Oncogenic Deletion of Met in Lung Cancer
Cancer Res.,
January 1, 2006;
66(1):
283 - 289.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
Z. Zheng, G. Bepler, A. Cantor, and E. B. Haura
Small Tumor Size and Limited Smoking History Predicts Activated Epidermal Growth Factor Receptor in Early-Stage Non-small Cell Lung Cancer
Chest,
July 1, 2005;
128(1):
308 - 316.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. Singhal, A. Vachani, D. Antin-Ozerkis, L. R. Kaiser, and S. M. Albelda
Prognostic Implications of Cell Cycle, Apoptosis, and Angiogenesis Biomarkers in Non-Small Cell Lung Cancer: A Review
Clin. Cancer Res.,
June 1, 2005;
11(11):
3974 - 3986.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S.-M. Sheen-Chen, Y.-W. Liu, H.-L. Eng, and F.-F. Chou
Serum Levels of Hepatocyte Growth Factor in Patients with Breast Cancer
Cancer Epidemiol. Biomarkers Prev.,
March 1, 2005;
14(3):
715 - 717.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. C. Ma, R. Jagadeeswaran, S. Jagadeesh, M. S. Tretiakova, V. Nallasura, E. A. Fox, M. Hansen, E. Schaefer, K. Naoki, A. Lader, et al.
Functional Expression and Mutations of c-Met and Its Therapeutic Inhibition with SU11274 and Small Interfering RNA in Non-Small Cell Lung Cancer
Cancer Res.,
February 15, 2005;
65(4):
1479 - 1488.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
T C Mineo, V Ambrogi, A Baldi, C Rabitti, P Bollero, B Vincenzi, and G Tonini
Prognostic impact of VEGF, CD31, CD34, and CD105 expression and tumour vessel invasion after radical surgery for IB-IIA non-small cell lung cancer
J. Clin. Pathol.,
June 1, 2004;
57(6):
591 - 597.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
X. Wang, P. Le, C. Liang, J. Chan, D. Kiewlich, T. Miller, D. Harris, L. Sun, A. Rice, S. Vasile, et al.
Potent and selective inhibitors of the Met [hepatocyte growth factor/scatter factor (HGF/SF) receptor] tyrosine kinase block HGF/SF-induced tumor cell growth and invasion
Mol. Cancer Ther.,
November 1, 2003;
2(11):
1085 - 1092.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. Maulik, T. Kijima, P. C. Ma, S. K. Ghosh, J. Lin, G. I. Shapiro, E. Schaefer, E. Tibaldi, B. E. Johnson, and R. Salgia
Modulation of the c-Met/Hepatocyte Growth Factor Pathway in Small Cell Lung Cancer
Clin. Cancer Res.,
February 1, 2002;
8(2):
620 - 627.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Bechard, T. Gentina, M. Delehedde, A. Scherpereel, M. Lyon, M. Aumercier, R. Vazeux, C. Richet, P. Degand, B. Jude, et al.
Endocan Is a Novel Chondroitin Sulfate/Dermatan Sulfate Proteoglycan That Promotes Hepatocyte Growth Factor/Scatter Factor Mitogenic Activity
J. Biol. Chem.,
December 14, 2001;
276(51):
48341 - 48349.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
