|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 1999 American Society for Clinical Oncology Polymerase Chain Reaction-Based Detection of Circulating Melanoma Cells as an Effective Marker of Tumor ProgressionFrom the Divisions of Clinical Immunology and Medical Oncology A, Department of Medicine, National Tumor Institute, Naples, Italy. Address reprint requests to Dr. Giuseppe Castello, Division of Clinical Immunology, National Tumor Institute "G. Pascale," Via M. Semmola, 80131 Naples, Italy
PURPOSE: Reverse transcriptase (RT) polymerase chain reaction (PCR) with multiple markers has been demonstrated to be highly sensitive in detecting circulating cells from patients with malignant melanoma (MM). We evaluated the clinical significance of the presence in peripheral blood of specific PCR-positive mRNA markers as an expression of circulating melanoma cells. PATIENTS AND METHODS: Total cellular RNA was obtained from the peripheral blood of 235 patients with either localized (n = 154) or metastatic (n = 81) melanoma. We performed RT-PCR using tyrosinase, p97, MUC18, and MelanA/MART1 as gene markers. The PCR products were analyzed by gel electrophoresis and Southern blot hybridization. In addition, 20 healthy subjects and 21 patients with nonmelanoma cancer were used as negative controls. RESULTS: Although detected at various levels among assessable patients, each mRNA marker was significantly correlated with disease stage. A significant correlation with disease stage was demonstrated for patients who were positive to all four markers (P < .0001) or to at least three markers (P < .001). Univariate analysis showed a significant correlation between risk of recurrence (evaluated in stage I, II, and III patients) and increasing number of PCR-positive markers (P = .0002). Logistic regression multivariate analysis indicated that each single marker (except tyrosinase) and, more especially, the presence of four PCR-positive markers remained statistically independent prognostic factors for tumor progression. CONCLUSION: Our data establish the existence of a significant correlation among clinical stages, tumor progression, and presence of circulating melanoma-associated antigens in peripheral blood of MM patients. Preliminary assessment of a subset of patients with a higher risk of recurrence needs longer follow-up and further studies to define the role of RT-PCR in monitoring MM patients.
THE INCIDENCE AND MORTALITY RATE of malignant melanoma (MM) have increased more than those of any other malignancy among all white populations in the past 15 years.1,2 Ideally, early detection would decrease both the incidence and mortality rate of MM through the identification and excision of promelanoma lesions. Malignant melanoma is associated with a poor prognosis because of its high frequency of metastasis. Considering the small size of most primary melanoma lesions, the metastatic potential of MM is considerably greater than that of most other solid tumors. In fact, for patients with lesions thinner than 1.5 mm, the 5-year disease-free survival prospects are over 90%; for those with lesions thicker than 3 mm, the 5-year disease-free survival rate falls to below 50%.2 The relationship between circulating tumor cells and the development of secondary disease is not fully understood. Metastatic melanoma cells are not found in the circulation of healthy individuals. Consequently, the presence of such cells in peripheral blood samples from patients with early-stage MM could indicate a dissemination of the tumor cells and, thus, a high risk of metastasis. If confirmed, circulating melanoma cells could represent a new prognostic marker with which to predict clinical behavior, assess tumor progression, and improve patient management. Reverse transcriptase (RT) polymerase chain reaction (PCR) using tyrosinase mRNA is reported to be a useful tool for detection of circulating tumor cells in the peripheral blood of patients with melanoma.3 The tyrosinase gene (TYR)4 encodes a key enzyme in melanin biosynthesis and its expression is not found in normal peripheral blood. Therefore, tyrosinase (TYR) mRNA could indicate the presence of circulating melanoma cells. Brossart et al5-7 reported that the number of circulating cells in MM patients, evaluated by a semiquantitative tyrosinase RT-PCR assay, correlates with tumor burden. However, different research groups reported conflicting results on the sensitivity and clinical value of tyrosinase RT-PCR.8-17 In patients with localized disease (stages I and II), the percentage of blood samples that test positive for TYR varies greatly, ranging from 010 to 45%.14 Similarly, blood samples from untreated patients with advanced disease show remarkable differences in TYR-positive rates, which range from 016 to 44%9 for stage III patients and from 13%12 to 100%5-7 for stage IV patients. Taken altogether, published data indicate that only 44 (14%) of 305 patients with localized disease and 253 (49%) of 511 patients with advanced disease are TYR-positive.5-17 Because of such inconsistent results, PCR-based analysis of TYR mRNA seems unreliable as a marker for detection of melanoma cells in patients' blood. In addition, low sensitivity in patients with disseminated and progressive disease suggests that TYR mRNA may be of limited value in the management of MM patients. However, the heterogeneity of TYR expression in melanomas and methodologic differences in blood sample preparation, RNA extraction, and cDNA synthesis may also account for these discrepancies. In an attempt to resolve the latter problems, the European Organization for Research and Treatment of Cancer's Melanoma Cooperative Group (EORTC-MCG) has provided standardized methods and standard quality control measures to be used to avoid any PCR-based artifacts and facilitate comparison of results from different laboratories.18 However, the heterogeneity of TYR expression could be related to the peculiar biologic properties of melanomas, as the expression level of specific melanoma-associated antigens can vary considerably, becoming more heterogeneous in the advanced stages of tumor progression. Therefore, a single mRNA marker assay could be less inclusive than a multimarker mRNA assay for detecting the heterogeneous population of occult metastatic melanoma cells. To monitor MM patients with either localized or advanced disease, maximal sensitivity and reliability of the diagnostic RT-PCR approach could be achieved by using additional markers expressed at detectable levels and with high frequencies in melanoma. In recent years, two major studies demonstrated that RT-PCR assays with multiple mRNA markers are more effective than assays with TYR alone for the detection of micrometastasis in the peripheral blood of MM patients.19,20 However, most of the blood samples analyzed by these investigators were from patients with advanced disease, and thus the role of each single marker and/or combination of PCR-positive markers on tumor progression was not well defined. The aims of this study were to assess the sensitivity and specificity of a multimarker RT-PCR assay (combined with Southern blot analysis) in detecting circulating melanoma cells and to evaluate the statistical correlation between the presence of PCR-positive markers and the clinical status of MM patients. We investigated a large number of control subjects and 235 MM patients at various stages of disease, strictly following the indications of the EORTC-MCG for blood sample preparation, RNA extraction, and cDNA synthesis. All four gene markers used in our RT-PCR assay (TYR and the other three melanoma-associated antigens, ie, p97,21 MUC18,22 and MelanA/MART123) were previously found expressed in most human melanomas but only in trace amounts in normal adult tissues.24 Among normal cells, only melanocytes seem to express these genes at detectable levels.24
Patient Selection Patients with a histologically documented diagnosis of cutaneous malignant melanoma, with either localized or metastatic disease, were included in the study. Disease stage was recorded as IA, IB, IIA, IIB, III, or IV, according to the American Joint Committee on Cancer (AJCC) guidelines. Blood samples were taken from nonmetastatic melanoma patients within the 2 weeks after surgical treatment, from stage III patients within 2 weeks after lymph node dissection or local recurrence (in transit metastasis) excision, and from stage IV patients before systemic treatment. Clinical stage and current or past therapy (if any) were documented at the time of enrollment and followed up prospectively. Clinical staging was determined by medical history, physical examination, cell blood count, and blood biochemistry. Other complementary examinations were performed if indicated. Disease status was defined depending on the absence or presence of clinical melanoma at the moment of blood extraction, and disease progression was determined by a worsening in disease status. No clinical decisions were made on the basis of the RT-PCR assay results. All stage I, II, and III patients were examined every 4 months after the diagnosis. At each follow-up visit, a clinical history, physical examination, cell blood count, and blood biochemistry were performed. Other tests were performed if clinically indicated. Negative controls were healthy subjects and patients with other malignancies. The human melanoma-derived cell lines SK-MEL-28 and SK-MEL-29 (American Type Culture Collection, Rockville, MD) were used as a positive control.
Sample Preparation
Oligonucleotide Primers The integrity of RNA for RT-PCR assays was determined using parallel RT-PCR assays with primers specific for the housekeeping gene GAPDH (sense, 5'-TCT GAC TTC AAC AGC GAC AC-3'; antisense, 5'-TCT TCC TCT TGT GCT CTT GG-3'). A 160-bp fragment was produced. Samples that failed to amplify products for GAPDH RNA were considered noninformative and discarded.
Reverse Transcriptase Polymerase Chain Reaction Assay In each RT-PCR assay, respective controls included a melanoma cell line RNA as reaction-positive control, total human DNA (to detect illegitimate gene marker amplification at the genomic level), PCR reagents and primers without RNA as reaction-negative control (to reveal abnormal PCR-mixture contamination), and amplification control for the housekeeping gene GAPDH (to facilitate quantitative and qualitative assessment of both RNA extraction and cDNA synthesis). Final products were separated by electrophoresis on a 2% agarose gel and analyzed by direct visualization after ethidium bromide staining. To assess the specificity of RT-PCR products, agarose gels were transferred onto a nylon membrane for Southern blot analysis.26 All cDNA probes were prepared from PCR products inside the outermost PCR primers to ensure their specificity. Hybridization of membranes was performed using labeled probes, as previously described.27 The number of positive patients was determined by adding positive results from blot analysis to those from direct visualization analysis.
Statistical Analysis
Specificity and Sensitivity of the RT-PCR Assay Peripheral blood nucleated cells (PBNCs), obtained by dextran separation of 5 mL of blood from 20 healthy donors and 21 nonmelanoma cancer patients (eight colorectal carcinomas, five non–small-cell lung cancers, three head and neck squamous cell carcinomas, three non-Hodgkin's lymphomas, and two multiple myelomas; all patients presented advanced disease at the time of blood sampling), were screened with all four markers to test the specificity of the RT-PCR assay (Table  1). Polymerase chain reaction markers were not detected in PBNCs from healthy donors, with the exception of three cases (15%) that were positive to MUC18 (Table 1). Higher percentages of positive results were obtained in PBNCs from nonmelanoma cancer patients, ranging from 10% for MelanA/MART1 to 43% for MUC18 (Table 1). Among all 41 control subjects, 13 (32%) were found to be positive to a single mRNA marker, and only two (5%) were found to be positive to two markers (Table 1). Calculation of specificity (true negative/true negative + false positive) indicated that maximal value was reached by TYR (1.00) and similar good results were obtained using MelanA/MART1 and p97 (0.95 and 0.93, respectively), but a low specificity was shown by MUC18. However, we had included MUC18 in the assay because of its sensitivity in detecting metastatic tumor cells in peripheral blood, as previously reported.19
To assess the sensitivity of the RT-PCR assay, serial dilutions of melanoma cell lines (105 to 100 cells) in 107 PBNCs were obtained and a PCR assay was performed using 1 µg of total RNA from each dilution. This in vitro model system mimicked circulating melanoma cells in blood, allowing the detection of melanoma cells mixed with PBNCs. After separation by gel electrophoresis, simple visual examination of PCR cDNA products stained with ethidium bromide could still detect, although weakly, one melanoma cell in 107 PBNCs (Fig
Patient Analysis
The PBNC from melanoma patients were screened using the RT-PCR assay previously described to detect transcripts of TYR, p97, MUC18, and MelanA/MART1 genes. The RT-PCR products were separated by electrophoresis on a 2% agarose gel and analyzed by direct visualization after ethidium bromide staining. Figure 2 shows the test results of a subset of patients tested with primers specific for each individual marker as well as for GAPDH gene expression. For p97 and GAPDH, only one-round amplification was needed to detect the specific PCR fragments (286 and 160 bp, respectively; Fig 2). For TYR, MUC18, and MelanA/MART1, cDNA bands (207, 262, and 292 bp, respectively) shown in Fig 2 represent nested PCR products. In these cases, only samples that showed specific bands after a second round of amplification with nested primers were considered positive. Samples that showed no specific amplification with nested primers were considered negative. Samples that showed no specific amplification for individual gene marker RNA and positivity for GAPDH RNA were considered negative.
All PCR products, including PBNC samples with negative test results after gel fractionation, were transferred onto nylon membranes, and hybridization was performed with respective cDNA probes. Southern blot analysis increased the number of PCR-positive patients only for the TYR, p97, and MelanA/MART1 markers. Specifically, 18, 11, and four samples that were negative after direct visualization analysis were found to be positive after blot analysis with the TYR, p97, and MelanA/MART1 probes, respectively. All samples that were still negative after specific hybridization were considered truly negative.
RT-PCR Results and Statistical Analysis
When all of the PCR positive markers were considered (Table
No statistical correlation was observed between the RT-PCR results and Breslow thickness, Clark level, histology, primary site, number of regional lymph nodes involved, sites of distal metastases, growth pattern, sex, or age. Logistic regression multivariate analysis was performed on the patient group as a whole to estimate the relative risk of clinical stage, presence of individual mRNA markers, and total number of PCR-positive markers, as well as to adjust potential confounding effects and to assess possible multiplicative interactions. With AJCC stage as the dependent variable, the relative risk was estimated to increase 1.8 times (P = .06) for TYR-positive patients, 5.6 times (P = .0001; 95% confidence interval [CI], 2.4 to 12.8) for p97-positive patients, 8.48 times (P < .005; 95% CI, 1.9 to 37.6) for MUC18-positive patients, 2.1 times (P = .025; 95% CI, 1.1 to 4.0) for MelanA/MART1-positive patients, and 19.5 times (P = .008; 95% CI, 2.1 to 179.4) for patients positive to all four PCR markers. With the exception of TYR, each individual mRNA and the presence of four PCR-positive markers (although the confidence interval was too wide in this case) remained statistically independent prognostic factors for tumor progression.
Previous studies have demonstrated the usefulness of RT-PCR assays that use only TYR as marker to detect circulating melanoma cells in peripheral blood.3,5 However, rates of TYR-positive patients vary considerably among the published reports.5-17 These studies did not consider the heterogeneity of individual gene marker expression in tumor lesions within individual patients or among patients. The continuous production of genetic variant cells by a progressing cancer can contribute to tumor heterogeneity and variation in the level of individual gene marker expression.24 Melanoma-associated antigen expression can differ among metastatic melanomas from the same patient and between a metastatic lesion and its primary tumor.24 The use of single-gene-marker assays may be unreliable because of the heterogeneity and level of single-marker expression in tumor cells. Therefore, a multiple-marker RT-PCR assay combined with Southern blot analysis should be more inclusive and avoid possible false-negative and false-positive results. Hoon et al19 and Sarantou et al20 demonstrated the efficacy of using a combination of mRNA markers to detect micrometastasis in the blood circulation of MM patients. However, the majority (134 of 154, 87%) of blood samples analyzed in these two studies were from patients with advanced disease (AJCC stages III and IV) who were positive to most of the markers used.19,20 A significant correlation with disease stage was shown only for TYR19; no significance as tumor progression marker was demonstrated for other single mRNA or for their combinations in the RT-PCR assays analyzed.19,20 When all existing data were taken into account, TYR seemed to be the only marker useful as a prognostic factor in monitoring MM patients. Unfortunately, this outcome was not unanimously confirmed. In our report, we demonstrate that mRNA expression of two or more markers identified by a multiplex RT-PCR assay can be more specific in validation of a positive result, thereby improving sensitivity in the detection of metastatic tumor cells. Among the gene markers used, we did not observe any TYR mRNA transcript in nonmelanoma controls, confirming the highest specificity of this marker. Nonetheless, p97 and MelanA/MART1 showed a good specificity, with only 5% to 7% false-positive results in nonmelanoma controls. Although expression of MUC18 mRNA has been found at detectable levels only in melanocytes,22,24 specificity of this marker was unexpectedly low in our nonmelanoma controls (29% false-positive results). All genes described above specify for melanoma-associated antigens that are involved in cell-mediated immune response and exert cytolytic activity by T lymphocytes. Two additional antigens, MAGE-330 and gp100/pMel-17,31 which are commonly recognized by T cells in melanoma, were not used as markers in our study because of their low specificity. In fact, both MAGE-3 and gp100/pMel-17 were demonstrated to be expressed at the mRNA level not only in cells of the melanocytic lineage but also in many tumor samples and cell lines of nonmelanocytic origin.32,33 False-positive results using RT-PCR assays have been reported by several authors, reflecting the possibility of illegitimate transcription processes or the detection of specific transcripts in nonspecific cells.13-17 However, false-positive results were usually generated by PCR conditions stronger than those used in our protocols (ie, after 40 cycles of a second round of amplification with nested primers). Southern blot analysis with probes that recognize internal sequences can be considered as an additional control to confirm the specificity of the PCR products and avoid false-positive results. The limited number of cases in most of the previous studies and the different methodologic approaches in the sample preparation might be responsible for the contrasting results on the reliability of the RT-PCR assay to test tumor progression in MM patients. In our study, we screened a large group of MM patients, according to the reproducible conditions indicated by the EORTC-MCG.18 We found a significant correlation between clinical stage of disease and RT-PCR status, in terms of both individual marker expression and number of PCR-positive markers. An increasing number of markers were detected in a majority of patients with clinical metastases, whereas the percentage of detection progressively decreased in patients with nodal involvement and localized disease. Our analysis confirms that multiple-marker assays are more sensitive detectors of neoplastic circulating cells than single-marker assays.
In addition, a significant preliminary observation in our series was the association between the increasing number of PCR-positive markers and the risk of relapse in patients with no evidence of clinical disease. Presence of relapse was evaluated in the 203 patients in the series with stage I, II, and III disease. Although the median follow-up period was too short (13 months; range, 8 to 18 months) from the time of blood sampling to the date of this writing (July 1998), there were no relapses among the 87 stage I patients, 9 (13%) among the 67 stage II patients, and 15 (31%) among the 49 stage III patients ( A longer follow-up period for all of these patients is required to better define the short- and long-term clinical significance of these findings. However, one could speculate that the presence of a higher number of PCR-positive markers in peripheral blood could be related to circulating multiple metastatic clones, the first step in the tumor progression cascade toward establishment of distant metastases. If confirmed, this information may be used to predict tumor progression in patients with circulating tumor cells and no evidence of disease and to identify a subset of patients who may need a more detailed follow-up. However, it is important to keep in mind that establishment of distant metastases depends on many parameters, including the host's antitumor response, the viability of metastatic cells in blood circulation, the capacity of tumor cells to colonize the target tissue, and the growth potential at the distant site. In conclusion, our results establish the existence of statistically significant associations between clinical stage of malignant melanoma and detection by RT-PCR of tumor-associated antigens in peripheral blood as an expression of circulating neoplastic cells. A preliminary outcome analysis of our series seems to define a subset of patients with a higher risk of recurrence. Further studies are needed to better define the significance of the presence of such circulating antigens for tumor prognosis, early detection of relapse, and monitoring of the effectiveness of systemic therapy in patients with melanoma.
The members of the Melanoma Cooperative Group are P. Aprea, P.A. Ascierto, G. Botti, C. Caraco', G. Castello, E. Celentano, C. Comella, A. Daponte, F. Graziano, N. Mozzillo, G. Palmieri, R. Parasole, A. Picone (all from the National Tumor Institute, Naples, Italy), L. Bosco, R. Satriano, (both from the 2nd University of Naples, Naples Italy), and M. D'Urso (International Institute of Genetics and Biophysics, C.N.R., Naples, Italy).
We thank Drs. Giovanna Romano, Maria Napolitano, and Enrico Leonardi for their helpful technical assistance. Supported by Italian Ministry of Health, Health Department of Campania Region (Italy), and C.N.R. "Target Project on Biotechnology."
Drs. Palmieri and Strazzulo contributed equally to this work. Present address of Dr. Palmieri is Institute of Molecular Genetics, Alghero (SS), Italy.
1. Schuchter LM: Melanoma and other skin neoplasms. Curr Opin Oncol 9:175-177, 1997[Medline] 2. Berwick M, Halpern AH: Melanoma epidemiology. Curr Opin Oncol 9:178-182, 1997[Medline] 3. Smith B, Selby P, Southgate J, et al: Detection of melanoma cells in peripheral blood by means of reverse transcriptase and polymerase chain reaction. Lancet 338:1227-1229, 1991[Medline]
4.
Brichard V, Van Pel A, Wolfel T, et al: The tyrosinase gene codes for an antigen recognized by autologous cytolytic T lymphocytes on HLA-A2 melanomas. J Exp Med 178:489-495, 1993 5. Brossart P, Keilholz U, Willhauck M, et al: Hematogenous spread of malignant melanoma cells in different stages of disease. J Invest Dermatol 101:887-889, 1993[Medline] 6. Brossart P, Keilholz U, Scheibenbogen C, et al: Detection of residual tumor cells in patients with malignant melanoma responding to immunotherapy. J Immunother 15:38-41, 1994
7.
Brossart P, Schmier J, Kruger S, et al: A PCR-based semiquantitative assessment of malignant melanoma cells in peripheral blood. Cancer Res 55:4065-4068, 1995 8. Foss AJ, Guille MJ, Occleston NL, et al: The detection of melanoma cells in peripheral blood by reverse transcription-polymerase chain reaction. Br J Cancer 72:155-159, 1995[Medline]
9.
Battayani Z, Grob J, Xerri R, et al: PCR detection of circulating melanocytes as a prognostic marker in patients with melanoma. Arch Dermatol 131:443-447, 1995
10.
Kunter U, Buer J, Probst M, et al: Peripheral blood tyrosinase messenger RNA detection and survival in malignant melanoma. J Natl Cancer Inst 88:590-594, 1996 11. Stevens GL, Scheer WD, Levine EA: Detection of tyrosinase mRNA from the blood of melanoma patients. Cancer Epidemiol Biomarkers Prev 5:293-296, 1996[Abstract]
12.
Pittman K, Burchill S, Smith B, et al: Reverse transcriptase-polymerase chain reaction for expression of tyrosinase to identify malignant melanoma cells in peripheral blood. Ann Oncol 7:297-301, 1996 13. Reinhold U, Ludtke Handjery HC, Schnautz S, et al: The analysis of tyrosinase-specific mRNA in blood samples of melanoma patients by RT-PCR is not a useful test for metastatic tumor progression. J Invest Dermatol 108:166-169, 1997[Medline]
14.
Mellado B, Colomer D, Castel T, et al: Detection of circulating neoplastic cells by reverse-transcriptase polymerase chain reaction in malignant melanoma: Association with clinical stage and prognosis. J Clin Oncol 14:2091-2097, 1996 15. Proebstle T, Voit C, Weber W, et al: Association of positive RT-PCR for tyrosinase in peripheral blood of malignant melanoma patients with clinical stage, disease-free survival and other risk factors. Symposium, Diagnostic PCR in Oncology. Heidelberg, Germany, 1997 (abstract 22) 16. Glaser R, Rass K, Seiter S, et al: Detection of circulating melanoma cells by specific amplification of tyrosinase complementary DNA is not a reliable tumor marker in melanoma patients: A clinical two-center study. J Clin Oncol 15:2818-2825, 1997[Abstract] 17. Jung FA, Buzaid AC, Ross MI, et al: Evaluation of tyrosinase mRNA as a tumor marker in the blood of melanoma patients. J Clin Oncol 15:2826-2831, 1997[Abstract] 18. Keilholz U: Diagnostic PCR in melanoma, methods and quality assurance. Epalinges, Switzerland, 26/27 January 1996. Eur J Cancer 32A:1661-1663, 1996
19.
Hoon DSB, Wang Y, Dale PS, et al: Detection of occult melanoma cells in blood with a multiple-marker polymerase chain reaction assay. J Clin Oncol 13:2109-2116, 1995
20.
Sarantou T, Chi DDJ, Garrison DA, et al: Melanoma-associated antigens as messenger RNA detection markers for melanoma. Cancer Res 57:1371-1376, 1997
21.
Rose TM, Plowman GD, Teplow DB, et al: Primary structure of the human melanoma-associated antigen p97 (melanotransferrin) deduced from the mRNA sequence. Proc Natl Acad Sci USA 83:1261-1265, 1986
22.
Lehmann JM, Riethmuller G, Johnson JP: MUC18, a marker of tumor progression in human melanoma, shows sequence similarity to the neural cell adhesion molecules of the immunoglobulin superfamily. Proc Natl Acad Sci USA 86:9891-9895, 1989
23.
Coulie PG, Brichard V, Van Pel A, et al: A new gene coding for a differentiation antigen recognized by autologous cytolytic T lymphocytes on HLA-A2 melanomas. J Exp Med 180:35-42, 1994 24. Slingluff CL Jr, Hunt DF, Engelhard VH: Direct analysis of tumor-associated peptide antigens. Curr Opin Immunol 6:733-740, 1994[Medline] 25. Chomczynski P, Sacchi N: Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156-159, 1987[Medline] 26. Maniatis T, Fritsch EF, Sambrook J (eds): Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory, 1989 27. Palmieri G, Romano G, Ciccodicola A, et al: YAC contig organization and CpG island analysis in Xq28. Genomics 24:149-158, 1994[Medline] 28. Breslow A: Thickness, cross-sectional area and depth of invasion in the prognosis of cutaneous melanoma. Ann Surg 172:902-908, 1970[Medline] 29. Clark WH Jr, Elder DE, Guerry D IV,et al: A study of tumor progression: The precursor lesions of superficial spreading and nodular melanoma. Hum Pathol 15:1147-1165, 1984[Medline]
30.
Gaugler B, Van den Eynde B, van der Bruggen P, et al: Human gene MAGE-3 codes for an antigen recognized on a melanoma by autologous cytolytic T lymphocytes. J Exp Med 179:921-930, 1994
31.
Bakker AB, Schreurs MWJ, de Boer AJ, et al: Melanocyte lineage-specific antigen gp100 is recognized by melanoma-derived tumor-infiltrating lymphocytes. J Exp Med 179:1005-1009, 1994 32. Brouwenstijn N, Slager EH, Bakker AB, et al: Transcription of the gene encoding melanoma-associated antigen gp100 in tissues and cell lines other than those of the melanocytic lineage. Br J Cancer 76:1562-1566, 1997[Medline]
33.
de Vries TJ, Fourkour A, Wobbes T, et al: Heterogeneous expression of immunotherapy candidate proteins gp100, MART1, and tyrosinase in human melanoma cell lines and in human melanocytic lesions. Cancer Res 57:3223-3229, 1997 Submitted July 29, 1998; accepted September 30, 1998.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 1999 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 test. Analysis of the exact coefficient for sample proportion was performed to determine whether there was any significant difference between using one specific marker alone and using multiple markers. All significant parameters (below .05 level) in the univariate analysis were included in a logistic regression for multivariate analysis. 
