|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Loss of MDA-7 Expression With Progression of MelanomaByFrom the Departments of Molecular and Cellular Oncology, Pathology, and Biostatistics, The University of Texas M.D. Anderson Cancer Center, and Introgen Therapeutics, Houston, TX. Address reprint requests to Julie A. Ellerhorst, MD, PhD, Department of Molecular and Cellular Oncology, Box 79, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX, 77030-4095; email: jaellerh{at}mail.mdanderson.org
PURPOSE: Ectopic transfer of the melanoma differentiation-associated gene-7 (mda-7) has been shown in vitro to suppress growth and induce apoptosis in a variety of human tumor cell lines; similar effects are not elicited in normal cells. Thus, the mda-7 gene seems to function as a novel tumor suppressor, and there is interest in the potential of mda-7 gene transfer as cancer therapy. The objective of this study was to determine if MDA-7 protein is lost during primary melanoma progression from superficial to invasive stages and from localized to metastatic tumor. As a secondary objective, we analyzed MDA-7 protein expression in primary melanomas for correlation with predictors of outcome and with survival. MATERIALS AND METHODS: MDA-7 protein expression was evaluated by immunohistochemistry in 41 primary melanomas and 41 metastases, including 24 paired samples. Each sample was scored for the percentage of positive cells and the overall intensity of immunolabeling. RESULTS: Significant decreases in MDA-7 immunostaining, reflected in both number and intensity scores, were observed when comparing the intraepidermal and superficially invasive portions with the deeply invasive portions of primary tumors. Significant differences were also observed when comparing primary tumors to paired metastases. CONCLUSION: Downregulation of MDA-7 expression in primary melanomas facilitates progression to invasive and metastatic stages. These data support the development of Ad-mda7 as gene therapy for advanced melanoma.
THE MELANOMA differentiation-associated gene-7 (mda-7) encodes a 24-kd protein, the cDNA for which was originally isolated from a human melanoma cell line induced to differentiate.1 In vitro, mda-7 gene transfer has resulted in phenotypic alterations generally associated with tumor suppressor gene activity. Introduction of mda-7 cDNA into cultured human tumor cells by plasmid transfection or by infection using an adenoviral vector (Ad-mda7) results in growth arrest and apoptotic cell death, findings observed not only in melanoma cells but also in cells from a variety of other solid tumor types.2-5 Importantly, normal cells do not undergo such changes on overexpression of MDA-7.2-5 These features form the basis of support for the concept of Ad-mda7 as cancer gene therapy. To date, there has been relatively little published regarding MDA-7 expression in human melanoma samples. Jiang et al1 used comparative RT-PCR to analyze mda-7 message in five samples of cultured human melanocytes, seven primary melanomas, and seven metastases. They reported a significant decrease in message as the samples proceeded from benign to malignant to metastatic. As part of a previous study performed in our laboratory, immunohistochemistry (IHC) was used to survey MDA-7 protein expression in a spectrum of tumor samples, including nevi, primary melanomas, and melanoma metastases.4 Although the sample numbers were again small, we observed a trend for high MDA-7 expression in nevi and declining levels proceeding from thin primaries to thick primaries to metastatic foci. Together, these two studies provide evidence that loss of MDA-7 may be one of the requisite steps for melanoma disease progression, culminating in invasion and metastasis. The data also suggest that MDA-7 levels in primary tumors might be of prognostic importance. To confirm and expand these preliminary findings, we have performed a large IHC study of MDA-7 protein expression in 82 melanoma samples, evenly divided between primary tumors and metastases, using paired samples when available. The primary objective of this study was to conclusively establish a decline in MDA-7 protein levels with melanoma progression, data that would support research into the development of Ad-mda7 as therapy for advanced disease. A secondary objective was to examine the potential correlation of MDA-7 levels in primary tumors with patient survival or with factors known to influence survival.
Patient Samples The tumor samples used in this study consisted of primary cutaneous melanomas and melanoma metastases from various sites. Formalin-fixed, paraffin-embedded sections of these tumors were obtained from the Melanoma and Skin Cancer Core Laboratory of The University of Texas M.D. Anderson Cancer Center. When available, a sample pair of primary tumor and metastasis was analyzed from the same patient.
Reagents
Antibody Specificity
Immunohistochemistry For each sample, vimentin and rabbit immunoglobulin G served as positive and negative primary antibody controls, respectively. All tissue samples from a given patient were immunolabeled in the same experiment.
Immunohistochemistry Scoring
Statistical Analysis Differences in scores for paired samples (different areas within the same tumor; primary tumors and metastases from the same patient) were analyzed for significance by the Wilcoxin signed rank test. Correlation of immunolabeling with pathologic stage, clinical stage, or site of metastasis was tested by  2 analysis. The log-rank test was used to analyze the correlation of MDA-7 expression with survival.
Tissue Samples Eighty-two samples from 48 patients were analyzed for expression of MDA-7 protein. The tumor specimens consisted of 41 primary cutaneous melanomas and 41 metastases from diverse sites (Table 2). Included in this group were pairs of matched primary tumors and metastases from 24 individual patients.
MDA-7 Antibody Specificity The binding specificity of the anti–MDA-7 polyclonal antibody was evaluated in a competition assay using bacterially derived recombinant human MDA-7 protein. Specificity of the MDA-7 polyclonal antibody is shown in Fig 1. Incubation of the antibody with recombinant MDA-7 before use for IHC yielded results identical to those of the negative control.
MDA-7 Expression by Primary Melanomas To determine if a pattern of MDA-7 loss occurred with tumor progression from superficial to invasive stages, we compared, within the same sample, MDA-7 expression by tumor cells in the epidermis or superficial dermis and cells at the tumor front invading into the deep dermis. Among the 25 Clark level III and IV tumors examined, both the percentage of MDA-7–positive cells and the intensity of immunolabeling were found to be decreased significantly (P = .003 and P = .008, respectively) in the invasive portion of the tumor (Fig 2).
Despite the differences in MDA-7 expression in superficial and deep portions of the tumors, we were unable to establish a significant correlation of MDA-7 number or intensity scores with Breslow thickness or Clark level among 41 primary tumors ranging from melanoma in situ to a thickness of 10 mm (Clark level IV). Similarly, expression of MDA-7 by the primary tumor did not correlate with clinical stage or with survival from the date of diagnosis (Table 3).
MDA-7 Expression by Melanoma Metastases On the basis of the differences in MDA-7 expression detected between the superficial and invasive portions of the primary tumors, we predicted that a similar pattern would be found when comparing the overall number and intensity scores of a primary tumor with the scores of a metastasis derived from it. To address this hypothesis, we examined MDA-7 expression in 24 pairs of primary tumors and metastases (Table 4). Significant reductions in the number (P = .001) and intensity (P = .001) scores of the metastases, relative to the primary tumors, were found (Fig 3).
We next examined a total of 41 metastases to determine if a pattern of MDA-7 loss existed along American Joint Committee on Cancer (AJCC) stages of metastatic progression. For analysis, the metastases were categorized as regional nodal (AJCC stage N), subcutaneous (AJCC stage M1a), and visceral (AJCC stage M1b) (Table 2). No differences were detected in the number or intensity scores of metastases in the three categories (Table 3).
MDA-7 Expression in Germinal Centers
In vitro studies using human tumor cell lines of diverse origin have provided sound evidence for the role of MDA-7 as a tumor suppressor. We have previously shown that infection of cultured human melanoma cells with Ad-mda7 results in growth arrest and apoptotic cell death.4 We now present immunohistochemical data in human melanoma that clearly demonstrate a decline of MDA-7 protein levels with progression of disease. We are able to show a significant decrease in MDA-7 expression when comparing the intraepidermal and superficially invasive to the deeply invasive portions of primary melanomas. Furthermore, using paired samples, we have found a significant decrease in MDA-7 expression between primary melanomas and the metastases derived from them. In contrast, no difference was detected among various stages of metastases, including regional lymph nodes, distant cutaneous sites, and distant visceral sites, suggesting that downregulation of MDA-7 occurs during earlier stages of progression. Taken together, these data provide evidence that loss of MDA-7 is a critical step in the progression of melanoma from a noninvasive primary tumor to an invasive malignancy with metastatic potential. Our findings thus lend additional support to the proposed role of mda-7 as a tumor suppressor gene. Previous data from our laboratory, analyzing a smaller number of primary tumors, suggested an inverse correlation between MDA-7 expression and pathologic stage, raising the possibility that MDA-7 might have prognostic utility. However, in the present study with a larger number of samples, we were unable to confirm a correlation of MDA-7 immunolabeling in the primary tumor with overall survival or with factors associated with survival, including primary tumor depth and clinical stage of the patient at presentation. Thus, it would seem that MDA-7 expression in a given primary tumor is not an independent predictor of tumor behavior or patient outcome. We observed several in situ melanomas and nonmetastatic thin primary tumors in which MDA-7 was nearly or totally absent, contrary to our expectation of intense immunoreactivity in the earlier stage tumors. Although we cannot rule out the possibility that other factors, such as differences in specimen handling or formalin fixation times, may have resulted in loss of MDA-7 antigenicity, it is more likely that MDA-7 is not uniformly expressed by every developing primary melanoma. It would follow that MDA-7 downregulation, loss, or absence may be necessary but not sufficient for melanoma progression to occur. The finding of MDA-7 immunoreactivity in germinal centers of some lymph nodes, in a morphologic pattern suggestive of dendritic cells, was unexpected and has not been previously described. The immunopositive germinal centers were frequently in nodal regions uninvolved with tumor, making it unlikely that metastatic melanoma cells accounted for the immunoreactivity. The extent and intensity of the immunolabeling argues against MDA-7 peptide, presented on the surface of dendritic cells, as the source of the positive findings. The favored explanation is that the dendritic cells themselves, in response to an unknown stimulus, produced the MDA-7 observed in these sections. Although it is not intuitive that MDA-7 expression should be induced in dendritic cells, it has been proposed, on the basis of structural and preliminary functional data, that MDA-7 may possess cytokine-like activity.5-7 Finally, it is possible that, even though the polyclonal antiserum used in this study was completely blocked by recombinant MDA-7, the antibodies might still cross-react with other molecules, as of yet undefined, that are homologous to MDA-7 and specific to a subset of dendritic cells residing in germinal centers. Despite the apparent lack of utility of MDA-7 as a diagnostic or predictive tool for melanoma, our data add to the present literature supporting the development of mda-7–based therapy as a novel approach for this disease. Accordingly, we would propose that Ad-mda7, presently in phase I clinical evaluation, be developed for the treatment of specific manifestations of advanced melanoma. One potential application would be gene therapy by direct inoculation for in transit disease. Although in transit disease can be treated effectively in 50% to 70% of patients by means of isolated hyperthermic limb perfusion, this form of treatment is not available to patients with head and neck or truncal primary tumors or patients with significant arterial insufficiency.8,9 Furthermore, patients who fail treatment with limb perfusion and those who suffer disease recurrence after initial response have few options for long-term disease control; these patients may be good candidates for Ad-mda7 gene therapy, as well. Other manifestations of metastatic melanoma that might similarly be considered include unresectable regional nodal disease and leptomeningeal disease, arguably the most refractory and morbid metastatic site. In conclusion, we now have convincing evidence that downregulation or loss of MDA-7 protein expression plays an important role in the progression of human melanoma. Our data support the concept that introduction of the mda-7 gene into advanced melanoma tumors might provide clinical benefit in terms of tumor remission or control. Future studies will focus on the feasibility of testing Ad-mda7 in clinical trials for melanoma.
Supported by grant no. P30 CA16672 from the Cancer Center Support to M.D. Anderson Cancer Center Melanoma Research Program Core Laboratory, training grant no. T32 CA72371 to J.A.E., and grant no. R41 CA89778 to E.A.G. We thank Golfers Against Cancer for their philanthropic support, Sandra Ortega for procurement and banking of tumor tissue samples, and Gary Fanger and Thomas Vedvick (Corixa Inc, Seattle, WA) for the preparation of reagents.
Both V.G.P. and S.E. contributed equally to this work.
1. Jiang H, Lin JJ, Su Z-Z, et al: Selective hybridization identifies a novel melanoma differentiation associated gene, mda-7, modulated during human melanoma differentiation, growth and progression. Oncogene 11: 2477-2486, 1995[Medline]
2.
Jiang H, Su Z-Z, Lin JJ, et al: The melanoma differentiation associated gene mda-7 suppresses cancer cell growth. Proc Natl Acad Sci U S A 93: 9160-9165, 1996 3. Saeki T, Mhashilkar A, Chada S, et al: Tumor-suppressive effects by adenovirus-mediated mda-7 gene transfer in non-small cell lung cancer in vitro. Gene Ther 7: 2051-2057, 2000[CrossRef][Medline] 4. Ekmekcioglu S, Ellerhorst J, Mhashilkar AM, et al: Down-regulated melanoma differentiation associated gene (mda-7) expression in human melanomas. Int J Cancer 94: 54-59, 2001[CrossRef][Medline] 5. Mhashilkar AM, Schrock RD, Hindi M, et al: Melanoma-differentiation associated gene-7 (mda-7): A novel anti-tumor gene for cancer gene therapy. Mol Med 7: 271-282, 2001[Medline] 6. Caudell EG, Chada S, Grimm EA: Expression and activity of the putative cytokine MDA-7. Third International Workshop on IL-10 and Related Molecules, Berlin, Germany, April 4-7, 2001 (abstr)
7.
Schaefer G, Venkataraman C, Schindler U: FISP (Il-4-induced secreted protein), a novel cytokine-like molecule secreted by Th2 cells. J Immunol 166: 5859-5863, 2001
8.
Thompson JF, Hunt JA, Shannon KF, et al: Frequency and duration of remission after isolated limb perfusion for melanoma. Arch Surg 132: 903-907, 1997 9. Lienard D, Eggermont AM, Kroon BBR, et al: Isolated limb perfusion in primary and recurrent melanoma: Indications and results. Semin Surg Oncol 14: 202-209, 1998[CrossRef][Medline] Submitted June 20, 2001; accepted October 4, 2001.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
