Plasma DNA microsatellites as tumor-specific markers and indicators of tumor progression in melanoma patients.

Multiple DNA microsatellites with frequent loss of heterozygosity (LOH) in melanomas have been demonstrated. The finding that free DNA is enriched in blood of melanoma patients prompted studies to determine whether tumor-specific DNA, such as DNA microsatellites exhibiting LOH, can be detected in blood and have clinical use. In this study, 57 advanced and 19 early clinically staged melanoma patients were assessed using 10 microsatellite markers on six chromosomes. Matched plasma and melanoma tissues from 40 patients showed significant concordance of LOH (P < 0.0001). The frequency of LOH microsatellite markers detected in plasma significantly increased in more advanced-staged patients. At locus D3S1293, LOH detection showed significant correlation to clinical disease progression (P = 0.02). Additionally, the combination of LOH microsatellite markers D9S157 and D3S1293 (P = 0.01), D9S157 and D1S228 (P = 0.05), and D11S925 and D3S1293 (P = 0.01) were significantly correlated to progression of different clinical stages of disease. These studies indicate that tumor-specific LOH markers in plasma have a potential clinical use as diagnostic and prognostic markers in melanoma patients.

Melanoma-associated antigens as messenger RNA detection markers for melanoma.

Melanoma is heterogeneous for its biological properties and melanoma-associated antigens (MAAs). This diversity is partially observed in the expression of the MAAs involved with the melanin synthesis pathway. We therefore developed a sensitive multimarker reverse transcription-PCR plus Southern blot assay using five MAAs as molecular markers to detect primary and metastatic melanoma cells. Melanoma cell lines, melanocytes (cultured), primary and metastatic malignant melanoma tissues, and blood from patients with American Joint Committee on Cancer stage I-IV melanoma were assessed for tyrosinase, tyrosinase-related proteins 1 and 2, Pmel 17, and MART-1/Melan-A. All of the MAA mRNA markers were expressed in 100% of melanoma cell lines and cultured melanocytes, 74% of primary and metastatic tumors (excluding tumor-draining lymph nodes), 43% of tumor-involved lymph nodes, and 43% of patients' bloods. Hypomelanotic melanoma tissues expressed a lower frequency of individual mRNA markers. Overall, at least one mRNA marker was expressed in more than 86% of specimens assayed. Normal tissue specimens from patients and blood from normal volunteer donors were negative for MAA mRNA expression. The multimarker MAA reverse transcription-PCR plus Southern blot analysis was more reliable and sensitive than a single-molecular marker assay for the detection of melanoma cells. This molecular assay can also provide information on MAA mRNA expression of metastatic melanoma cells that may assist in monitoring the therapeutic efficacy of active specific immunotherapy toward specific MAA-bearing melanomas.

Limitations of specific reverse-transcriptase polymerase chain reaction markers in the detection of metastases in the lymph nodes and blood of breast cancer patients.

PURPOSE: This study was performed to evaluate the potential of specific mRNA markers to detect micrometastases by reverse-transcriptase polymerase chain reaction (RT-PCR) and Southern blot analysis of sentinel lymph nodes (SNs) and blood from patients with breast cancer. PATIENTS AND METHODS: We assessed the specificity of carcinoembryonic antigen (CEA), cytokeratin-19 (CK-19), CK-20, gastrointestinal tumor-associated antigen-733.2 (GA733.2), and mucin-1 (MUC-1) in the blood of healthy donors (n = 13) and lymph nodes from patients without cancer (n = 3) by RT-PCR assay. The sensitivity of the RT-PCR assay for the target mRNA markers was assessed in breast cancer cell lines (n = 4), primary breast tumors (n = 8), and the frozen sections of SNs (n = 22) from 22 patients with American Joint Committee on Cancer (AJCC) stages I to IIIA breast cancer. RESULTS: CK-20 was the only mRNA marker not detected in lymph nodes or blood from patients without cancer. Both the blood and lymph nodes from patients without cancer expressed CEA, CK-19, GA733.2, and MUC-1 mRNA. All four breast cancer cell lines and six of eight primary breast tumors expressed all five mRNA markers. Expression of mRNA by the RT-PCR assay in the frozen-section SNs (n = 12) without metastases by conventional histopathology ranged from 8% (CK-20) to 92% (GA733.2). Detection of RT-PCR cDNA products in frozen-section SNs was increased with Southern blot analysis compared with ethidium bromide gel electrophoresis (EtBr) for all mRNA markers except CK-19. CONCLUSION: CEA, CK-19, GA733.2, and MUC-1 show no diagnostic value as mRNA markers for the detection of micrometastases by the RT-PCR assay because they are expressed in the blood and lymph nodes of patients without cancer. Further studies are needed to assess the sensitivity of CK-20 to detect micrometastases by the RT-PCR assay in the blood and frozen-section SNs of patients with breast cancer.

Predictive testing for multiple endocrine neoplasia type 2A (MEN 2A) based on the detection of mutations in the RET protooncogene.

BACKGROUND: The identification of inherited mutations in the RET protooncogene (RET) associated with multiple endocrine neoplasia type 2A (MEN 2A) has enabled the development of a genetic test to identify asymptomatic carriers of disease. METHODS: Genomic DNA was extracted from 96 members of an MEN 2A kindred. The polymerase chain reaction was used to amplify the RET exon known to contain the associated mutation. The mutation results in a new restriction endonuclease site and is detected by digestion with the appropriate enzyme. Inheritance of the mutation was verified with a previously developed genetic linkage test. RESULTS: We found that (1) mutations vary among kindreds but are consistently inherited within kindreds, (2) invariable correlation exists between mutation and disease (43 mutations in 43 affected individuals), (3) determination of the genetic status by linkage-based testing was precluded by recombination events and the informativeness of genetic markers, and (4) mutation analysis presymptomatically identified two genetically affected individuals. CONCLUSIONS: Direct genetic analysis for mutations in RET circumvents the limitations of linkage-based genetic testing and current biochemical screening assays. This method will be the diagnostic test of choice for the identification of asymptomatic individuals at risk for MEN 2A.

Medullary thyroid carcinoma: genetic advances, treatment recommendations, and the approach to the patient with persistent hypercalcitoninemia.

Medullary thyroid cancer is a tumor of the thyroid C cells that occurs in sporadic and hereditary clinical settings. Genetic testing of at-risk individuals is available and has been applied to patient management. Plasma calcitonin levels are a sensitive marker for the presence of disease. Surgery offers the best hope for cure and also is an effective modality for managing metastatic and recurrent disease.

Cellular immuno-PCR. Detection of a carbohydrate tumor marker.

Carbohydrate tumor-antigens are important tumor markers for diagnosis and functional characteristics of human cancer cells. Detection of these carbohydrate tumor antigens on metastatic cancer cells in blood is a difficult task. We developed a highly sensitive method to detect a cell surface carbohydrate antigen using a hybrid technology referred to as cellular immuno-PCR. This technique uses the human monoclonal antibody (HumAb) L612, specific to a tumor-related antigen (ganglioside) GM3 that is expressed on the cell surface of human tumor cells and not normal cells. L612 coupled to a DNA oligonucleotide for exponential amplification by DNA polymerase chain reaction (PCR) can be used to enhance the detection signal. The DNA-HumAb conjugate was assessed for detection of a small number of human cancer cells after PCR amplification and Southern blot analysis. To assess the assay specificity human melanoma and other cancer cell lines, as well as healthy donor and melanoma patients, bloods were assessed. Cellular immuno-PCR requires < 1 ng/ml DNA-HumAb complex and was shown to have a detection level of < 10 cells in titration studies in which melanoma cells were diluted in 2 million healthy donor peripheral blood lymphocytes. The assay was shown to be very sensitive and could detect low levels of GM3 antigen expression by tumor cells. This novel approach for detecting a carbohydrate tumor antigen on tumor cells in blood provides a potential useful clinicopathological assay.

Induction of antibody response to human tumor antigens by gene therapy using a fusigenic viral liposome vaccine.

Development of effective cancer vaccines would help prevent and control tumor progression. A novel approach of immunizing against tumor antigens is in vivo gene vaccination. We have developed a fusigenic viral liposome vector using HVJ (hemagglutinating virus of Japan) and liposome to deliver human tumor antigen genes effectively to cells in vivo. Plasmids containing the human tumor antigen genes MAGE-1 and MAGE-3 were encapsulated in fusigenic viral liposomes and injected into mice intramuscularly. MAGE-1 and -3 recombinant proteins were used in Western blotting and affinity ELISA for assessment of antibody responses. Mice immunized with MAGE-1 and -3 gene vaccine individually were shown to produce anti-MAGE-1 and -3 IgG antibody responses respectively. Animals immunized with plasmid alone did not induce anti-MAGE-1 or -3 IgG responses. Antibody responses could be enhanced on reimmunization with the gene vaccines. Muscle biopsies taken after vaccine injection were verified to express gene-specific mRNA transcripts. Mice immunized with MAGE-1 or -3 gene vaccines were shown to induce antibodies that could cross-react with the respective recombinant proteins. This study demonstrates that in vivo immunization using HVJ-liposome containing human tumor antigen genes can effectively deliver and induce immune responses to the respective whole proteins.

Molecular detection of tumor-associated antigens shared by human cutaneous melanomas and gliomas.

Both melanocytes and glial cells are derived embryologically from the neural ectoderm. Their malignant transformed counterparts, melanoma and glioma cells, respectively, may share common antigens. Numerous tumor-associated antigens have been identified in melanomas but only a few a gliomas. Using an established reverse transcriptase polymerase chain reaction plus Southern blot assay, we compared the mRNA expression of melanoma-associated antigens (MAAs) of melanomas to brain tumors primarily derived from glial cells. The MAAs studied included tyrosinase (Tyr), tyrosinase-related protein-1 and -2 (TRP-1 and TRP-2), gp100, human melanoma antigen-encoding genes 1 and 3 (MAGE-1 and MAGE-3), and melanotransferrin (p97). Glioblastoma multiforme (n = 21), anaplastic astrocytoma (n = 3), ependymoma (n = 2), meningioma (n = 3), oligodendroglioma (n = 1), and melanoma (n = 12) tumor specimens were assayed for MAA mRNA expression. Glioblastoma multiforme, astrocytoma, and melanoma cell lines were also assayed. We observed that individual MAA mRNAs were expressed in these brain tumors and cell lines at varying frequencies. The melanogenesis-pathway-related MAAs Tyr, TRP-1, TRP-2, and gp100 mRNAs were also expressed at different levels in normal brain tissues but at a much lower frequency than in glioblastoma multiforme and melanoma. MAGE-1 and MAGE-3 mRNA were expressed in different types of tumor specimens and cell lines but never in normal brain tissue. Tumor antigen p97 was expressed in all types of tumors and also in normal brain tissues. These studies demonstrate that melanomas and primary brain tumors express common MAAs and could be exploited in patients with malignant glioma by active specific immunotherapy against these common MAAs.

Detection of metastatic breast cancer by beta-hCG polymerase chain reaction.

Reverse transcriptase-polymerase chain reaction (RT-PCR) for detection of occult malignancies in breast cancer patients is evolving as a useful diagnostic tool. However, no reliable molecular mRNA markers are available. We developed an RT-PCR plus Southern blot assay using beta-hCG (beta-subunit of human chorionic gonadotropin) gene expression as a tumor marker for detection of breast malignancies metastatic to tumor-draining lymph nodes and blood. Breast carcinoma cell lines, primary breast malignancies and human placenta were used as positive controls for establishing the beta-hCG RT-PCR assay. Peripheral blood leukocytes (PBL) from normal volunteer donors, normal breast tissue and lymph nodes from cancer-free patients were used as negative controls. beta-hCG RT-PCR was used to assess tumor cell presence in PBL and tumor-draining axillary nodes from patients with AJCC stage I-IV breast cancer. The assay sensitivity and specificity were enhanced by restriction endonuclease digestion of an Sty I site of the RT-PCR cDNA product followed by Southern blot analysis. beta-hCG mRNA was expressed in all breast cancer cell lines and 80% of primary breast cancers; it was not expressed in negative controls. The assay reliably detected one cancer cell in > 10(7) PBL, with a sensitivity of 10(-5) microgram RNA. Eighty percent of PBL and 61% of tumor-draining axillary nodes from breast cancer patients expressed beta-hCG mRNA. The assay is a sensitive and specific method of identifying breast cancer cells in breast tissues, lymph nodes and blood.

Detection of beta-human chorionic gonadotropin mRNA as a marker for cutaneous malignant melanoma.

The beta chain of human chorionic gonadotropin (hCG) hormone is produced by fetal cells, gonadal cell tumors and several types of non-gonadal carcinoma. hCG is composed of an alpha and a beta chain, the latter of which can be used to distinguish the molecule from other related gonadotropin hormones. Detection of beta-hCG mRNA transcripts can be potentially useful as a marker to identify tumor cells. We devised a highly specific and sensitive assay to detect the atavistic expression of beta-hCG in cutaneous melanoma by RT-PCR. Twenty-four melanoma cell lines and 43 melanoma biopsies were evaluated for beta-hCG mRNA expression. An RT-PCR assay was developed to specifically distinguish beta-hCG poly-A mRNA from other related gonadotropin beta chains. This was performed by endonuclease digestion of a unique Sty 1 site in the beta chain, followed by Southern blot analysis with a beta-hCG cDNA probe. Of the 24 melanoma cell lines analyzed, 18 expressed beta-hCG mRNA. Analysis of melanoma biopsy specimens revealed beta-hCG mRNA expression in 17/25 melanoma-positive TDLN, and in only 5/15 non-lymphoid melanoma metastases. Beta-hCG mRNA expression had a 53% correlation to tyrosinase mRNA, a predominant melanoma marker. Beta-hCG mRNA was not detected in normal donor PBL and normal lymph nodes. Detection of beta-hCG mRNA expression may be a useful molecular marker to define a subset of malignant melanoma.

Predictive DNA testing and prophylactic thyroidectomy in patients at risk for multiple endocrine neoplasia type 2A.

BACKGROUND: Missense germ-line mutations in the RET protooncogene are associated with multiple endocrine neoplasia type 2A (MEN 2A). Detection of these mutant alleles in kindred members predicts disease inheritance and provides the basis for preventative thyroidectomy. METHODS: A polymerase chain reaction (PCR)-based genetic test for the 19 known RET mutations was designed to study 132 members of 7 kindreds with MEN 2A. Haplotypes also were constructed using genetic markers flanking the MEN 2A locus. Plasma calcitonin (CT) concentrations were determined before and after provocative testing. RESULTS: Direct DNA testing and haplotype analysis showed that 21 of 58 kindred members at risk for disease had inherited a mutation in the RET protooncogene associated with MEN 2A. Plasma CT concentrations were elevated in 9 of the 21 family members, but were normal in 12. After genetic counseling, 13 of the 21 kindred members (6 with normal and seven with elevated plasma CT levels), consented to immediate thyroidectomy. In each patient, the resected thyroid gland showed C-cell hyperplasia with or without medullary thyroid carcinoma. There were no metastases to regional lymph nodes, and postoperative stimulated plasma CT levels were normal. CONCLUSION: The PCR-based direct DNA test for RET mutations is accurate, rapid, and reproducible. For all 132 individuals evaluated, the results of direct DNA analysis were consistent with haplotype studies. The direct test for mutations in the RET protooncogene is the preferred method for screening MEN 2A kindreds. In family members who have inherited a RET mutation, total thyroidectomy is indicated, regardless of the plasma CT values.