Biomarkers and intermediate-high risk non-muscle invasive bladder cancer: a multivariate analysis of three different cellular pathways with pronostic implications

Purpose To determine the presence of a group of mutations, and establish the prognostic value for recurrence and progression. Materials and methods Prospective observational study. Intermediate-to-high-risk non-muscle invasive bladder cancer (NMIBC) was evaluated. Data from genetic analyses were included in a database along with clinicopathological variables of interest. Results Seventy-four patients. Twenty-five (33.8%) recurred and 3 (4.1%) progressed. Median time to recurrence: 8 months (5.7–12.7). Median time to progression: 14 months (P75: 12). Mutation distribution: KRAS codon 12: one patient (1.4%), BAT25: five patients (6.8%), BAT-26: four patients (5.4%), and D2S123: 6 patients (8.1%). Arg72Pro polymorphism: 50 patients (67.6%) exhibited homozygous mutations, 23 (31.1%) were heterozygous, and 1 patient (1.4%) did not present the mutation. We found an association between presence of MSI at BAT26 and female sex (p < 0.05) and tumor stage and the TP53 Arg72Pro polymorphism. Recurrence-free survival (RFS) was significantly associated with presence of MSI at D2S123, with a HR of 5.44 for patients presenting the mutation (95% CI 1.83–16.16). On multivariate analysis, we found a statistically significant increase in risk of recurrence among patients with MSI at D2S123 (HR 5.15; p < 0.05) and more than 2 previous transurethral bladder resections (TURBs) (HR 5.07; p < 0.05) adjusted for tumor stage and grade. Harrell’s concordance index revealed an accuracy of 0.74 (p < 0.05). Conclusion An association was found between presence BAT26 MSI and female sex, Arg72Pro polymorphism with tumor stage and D2S123 and more than 2 TUR procedures were associated with RFS adjusted to tumor stage and grade (AU)

Biomarkers and intermediate-high risk non-muscle invasive bladder cancer: a multivariate analysis of three different cellular pathways with pronostic implications.

PURPOSE: To determine the presence of a group of mutations, and establish the prognostic value for recurrence and progression. MATERIALS AND METHODS: Prospective observational study. Intermediate-to-high-risk non-muscle invasive bladder cancer (NMIBC) was evaluated. Data from genetic analyses were included in a database along with clinicopathological variables of interest. RESULTS: Seventy-four patients. Twenty-five (33.8%) recurred and 3 (4.1%) progressed. Median time to recurrence: 8 months (5.7-12.7). Median time to progression: 14 months (P75: 12). Mutation distribution: KRAS codon 12: one patient (1.4%), BAT25: five patients (6.8%), BAT-26: four patients (5.4%), and D2S123: 6 patients (8.1%). Arg72Pro polymorphism: 50 patients (67.6%) exhibited homozygous mutations, 23 (31.1%) were heterozygous, and 1 patient (1.4%) did not present the mutation. We found an association between presence of MSI at BAT26 and female sex (p < 0.05) and tumor stage and the TP53 Arg72Pro polymorphism. Recurrence-free survival (RFS) was significantly associated with presence of MSI at D2S123, with a HR of 5.44 for patients presenting the mutation (95% CI 1.83-16.16). On multivariate analysis, we found a statistically significant increase in risk of recurrence among patients with MSI at D2S123 (HR 5.15; p < 0.05) and more than 2 previous transurethral bladder resections (TURBs) (HR 5.07; p < 0.05) adjusted for tumor stage and grade. Harrell's concordance index revealed an accuracy of 0.74 (p < 0.05). CONCLUSION: An association was found between presence BAT26 MSI and female sex, Arg72Pro polymorphism with tumor stage and D2S123 and more than 2 TUR procedures were associated with RFS adjusted to tumor stage and grade.

Immunomagnetic quantification of circulating tumoral cells in patients with prostate cancer: clinical and pathological correlation.

OBJECTIVES: To detect and enumerate circulating prostatic tumor cells (CTC) in the peripheral blood of patients with prostate cancer (PC) and study the relationship between CTCs and clinical-pathological parameters. METHODS: Prospective three-arm study: 26 patients (p) with localised PC (LPC); 24 P with metastatic PC (MPC) and 30 healthy volunteer controls. A single 7.5 ml sample of peripheral blood was retrieved; CTCs were isolated using an immunomagnetic method based on the CellSearch system (Veridex). CTCs were identified as nucleated cells negative for CD45 (leukocytes) and positive for cytokeratins. (8, 18 y 19) The relationship between CTC numbers and PSA levels, Gleason score and TNM classification was studied. RESULTS: Only 10% of the healthy controls had 1 CTC/7.5 mL, none of the patients with localised PC had more than 3 CTCs (88% < or = 2 CTCs), and patients with MPC had significantly higher CTC levels [m: 29 (1-178)] compared with the other two groups (P: 0.000). A positive correlation was demonstrated between the CTC count and PSA levels, tumor size, and presence or absence of enlarged lymph nodes. Gleason score was the only parameter that did not show any correlation with CTC levels, and although the number of CTCs was higher in patients with visceral metastases [m: 297 (0-416)] compared with bone metastases patients [m: 68 (9.5-168)] , these differences were not significant. CONCLUSIONS: Immunomagnetic analysis permits CTCs to be enumerated in peripheral blood and could be a possible way to correctly stage and make a reasonable prognosis of metastatic disease.

Papel del laboratorio en la práctica clínica del cáncer de mama / Laboratory's function in clinic practice of breast cancer

No disponible

[Loss of heterozygosity in the 9p21 region as an inactivation mechanism of the p16 suppressor gene]. / Pérdida de heterocigosidad en la región 9p21 como mecanismo de inactivación del gen supresor p16.

OBJECTIVE: To determine the loss of heterozygosity (LOH) on 9p21 (locus D9S1747) in patients with renal carcinoma by analysis of microsatellite polymorphisms. METHODS: 40 patients with sporadic renal cancer were studied. LOH on 9p21 was performed by analysis of microsatellite polymorphisms. RESULTS: 23.7% showed LOH on 9p21. No correlation was found between this genetic alteration and tumor features. CONCLUSIONS: LOH on 9p21 was found in 23.7% of the patients in this series. LOH was found in 26.9% of renal cell carcinomas, 25% of papillary carcinomas and 25% of Bellini duct carcinomas. LOH was not found in the other histological types.

[Epidermal growth factor receptor (EGFR) in the prognosis of bladder carcinoma. Experience of 5 years]. / Receptor del factor de crecimiento epidérmico (EGFR) en el pronóstico del carcinoma vesical. Experiencia a 5 años.

METHODS: From November 1992 to November 1993, a prospective study was conducted on 20 controls and 61 patients with bladder carcinoma. EGFR expression was determined by radioimmunoassay and the correlation of the results of histological analysis and the clinical course was analyzed. The follow-up period was from November 1992 to July 1998. The association between qualitative variables and the x2 or Fisher exact test was compared using the hypothesis of the proportional ordinal trend for the ordinal variables, and the quantitative variables were analyzed using Student's t test and/or variance analysis (ANOVA). Survival was analyzed by the Kaplan-Meier method and comparison was performed using the Breslow exact test. The Cox proportional hazards regression model was utilized. The SPSS software for Windows 7.0 was used for the analysis. RESULTS: The EGFR values were higher for patients with bladder carcinoma than in controls (14.48 vs 2.54 fmol/mg of protein). EGFR values were higher in patients with superficial bladder tumor than in those with infiltrating tumors (27.03 fmol/mg vs. 10.05 fmol/mg of protein; p = 0.000). Poorly differentiated tumors showed higher values of EGFR (6.73, 14.48 and 17.07 fmol/mg of protein for grades I, II and III, respectively; p < 0.05). The EGFR values were higher in patients that died from cancer during follow-up (64.8) than in those who died from other causes (47.5) and those who are alive and on follow-up (42). An increase in EGFR values did not carry a risk of death from cancer (p = 0.1269; ns). Analysis of the grade of tumor differentiation showed that for the more aggressive tumor grade, a positive EGFR was a sign of reduced survival. Survival in patients with superficial and infiltrating tumor did not appear to change significantly according to the EGFR value. EGFR determination was not useful in predicting recurrence and increased EGFR values did not correlate with a higher risk of recurrence. CONCLUSIONS: 1) The normal pattern of EGFR could not be established. 2) EGFR was not useful in identifying subgroups at risk of death. 3) Knowledge about these proteins synthesized by oncogenes offers new possibilities in the treatment of cancer.

[P185 (Neu) oncoprotein in the prognosis of bladder carcinoma. Experience of 5 years]. / Oncoproteína P185 (Neu) en el pronóstico del carcinoma vesical. Experiencia a 5 años.

OBJECTIVE: To determine the utility of p185 oncogene in the biological characterization of transitional cell carcinoma and in the prediction of recurrence, and to analyze survival at 5 years mean follow-up. METHODS: A prospective clinical cohort study was conducted on 81 patients. Tissue specimens were obtained between November 1992 and November 1993. The study comprised two groups: nontumoral bladder tissue specimens from 20 patients (group I) and tissue specimens from 61 patients with bladder carcinoma (group II). p185 expression was determined by enzyme immunoanalysis (EIA). A statistical analysis of the results was performed. RESULTS: p185 oncoprotein levels were higher in patients with recurrence (1098.97 HNU/mg protein vs. 924.54 HNU/mg). Although higher levels of p185 were found in the patients that had died vs those who are alive, the differences were not statistically significant for overall survival or stratification by tumor grade or infiltration (p = 0.556; ns). CONCLUSIONS: Determination of p185 oncoprotein was found to be useful in the prediction of tumor recurrence at 5 years mean follow-up.

Oncoproteína P185 (Neu) en el pronóstico del carcinoma vesical. Experiencia a 5 años / P185 (neu) oncoprotein in the prognosis of bladder carcinoma: 5-years' experience

OBJETIVOS: El objetivo de este trabajo es conseguir establecer la utilidad de la determinación de la oncoproteína p185 en la caracterización biológica del carcinoma de células transicionales y evaluar su influencia en la predicción de recidiva y el análisis de la supervivencia tras un periodo medio de 5 años de seguimiento. MÉTODO: Hemos realizado un estudio de cohortes prospectivo que analiza un grupo de 81 pacientes cuyas muestras fueron recogidas entre Noviembre de 1992 y Noviembre de 1993 y que fueron divididos en 2 grupos: el primero formado por 20 muestras de tejido vesical no tumoral y el segundo integrado por 61 pacientes diagnos ticados de carcinoma vesical. Determinamos la expresión de la oncoproteina p185 mediante enzimoinmunoanálisis (EIA), realizando posteriormente un análisis estadístico detallado de los resultados. RESULTADOS: Los niveles detectados de oncoproteína p185 fueron mayores en pacientes con recidiva (1098,97 HNU/mg de proteína vs 924,54 HNU/mg). Pese a que los niveles de p185 fueron mayores en los pacientes fallecidos que en los pacientes vivos, no encontramos diferencias significativas en cuanto a la supervivencia global o estadificada por grado o infiltración (P=0,556, NS).CONCLUSIONES: La determinación de oncoproteína p185 demostró ser de utilidad para la predicción de recidiva tumoral en un seguimiento medio de 5 años (AU)

[Usefulness of p53 oncoprotein immunohistochemistry in the follow-up of bladder carcinoma: a 5-year study]. / Utilidad de la oncoproteína p53-IHQ en el seguimiento del carcinoma vesical: estudio a 5 años.

OBJECTIVE: Mutations in the TP53 gene are frequently detected in some types of malignant tumors (bladder, prostate, kidney, lungs, breast, colon and rectum). This study analyzed the utility of semi-quantitative determination of p53 in transitional cell carcinoma of the bladder by an immunohistochemical technique and evaluated the results at 5 years. METHODS/RESULTS: A prospective clinical cohort study was conducted on 81 patients. The study comprised two groups: nontumoral bladder tissue specimens from 20 patients (group I) and tissue specimens from 61 patients with bladder carcinoma (group II). In both groups the tissue specimens were obtained between November 1992 and November 1993, and during the follow-up period until July 1998. p53 expression was determined by a semi-quantitative method based on an immunohistochemical technique (NCL-p53-DO7, Novocastra). CONCLUSIONS: p53 oncoprotein was not found to be useful in the characterization of carcinoma of the urinary bladder.

[Quantification of p53 oncoprotein in bladder carcinoma: 5-year experience]. / Cuantificación de la oncoproteína p53 en el carcinoma vesical: experiencia a 5 años.

OBJECTIVE: Mutations in the p53 gene are frequently detected in some types of malignant tumors (bladder, prostate, kidney, lung, breast, colon and rectum). This study analyzed the utility of p53 quantitation in transitional cell carcinoma of the bladder and evaluated the results at 5 years. METHODS: A prospective clinical cohort study was conducted on 81 patients. The study comprised two groups: nontumoral bladder tissue specimens from 20 patients (group I) and tissue specimens from 61 patients with bladder carcinoma (group II). In both groups the tissue specimens were obtained between November 1992 and November 1993, including the follow-up period until July 1998. p53 expression was determined by a quantitative method based on immunoluminiscence (LIA-MAT p53). RESULTS: p53 expression was higher in bladder carcinoma than in healthy urothelial tissue; higher values of p53 were found for infiltrating and undifferentiated tumors. The p53 values were higher in patients with tumor recurrence than in those without (NS). The Bonferroni multiple comparisons test showed a higher mortality in patients with p53 > 0.9 than in patients who are alive or have died from other causes (p = 0.000). CONCLUSION: The results show that p53 LIA-MAT is an independent prognostic factor at a cutoff of 0.9 and permits identification of a subgroup of patients at high risk.

[The loss of heterozygosity on the short arm of chromosome 3 in renal carcinoma]. / Estudio de la pérdida de heterocigosidad en el brazo corto del cromosoma 3 en carcinoma renal.

OBJECTIVE: Renal cancer accounts for 2% of tumors. The most common chromosome abnormality found in renal cancer is the loss of heterozygosity (LOH) on the short arm of chromosome 3 (3p), which suggests that near the gene responsible for von Hippel-Lindau disease, there may be one or more tumor suppressor genes between 3p14 and 3p21 with a relevant role in the development of renal cancer. METHODS: 41 patients with sporadic renal cancer were tested for three microsatellites mapped to the short arm of chromosome 3 (3p14.1-3p14.3, 3p21.2-3p21.3 and 3p25) by polymerase chain reaction. The results were compared with patient habits and tumor features. RESULTS/CONCLUSIONS: 43.9% of the patients showed LOH on at least one locus. Thirty-four percent showed LOH only on one locus, 4.9% on two loci and 7.3% on the three loci tested. All the patients who showed LOH on 3p21 had a tumor size greater than 25 mm. There is a risk 1.76 times higher of no loss in tumors less than 25 mm in size than in tumors greater than 25 mm (Cl 95% 1.33-2.33).

Serum and tissue CEA in colorectal cancer: clinical relevance.

PURPOSE: This article is an analysis of the information derived from the determination of tumor-tissue concentration of CEA in patients with colorectal cancer. To ascertain the relationship between tumor marker content with the histologic aspects and serologic levels of CEA of this neoplam. MATERIALS AND METHODS: 136 patients with colorectal adenocarcinoma and 41 with colorectal benign processes are analyzed and followed during an average time of 27 months. The CEA of the serum were obtained preoperatively and postoperatively and measured by radioimmunoassay (RIA). Tissular CEA levels were determined with RIA. The histological characteristics are analyzed (Dukes classification, grade of differentiation, index of atypia, microscopic vascular and lymphatic involvement. RESULTS: 1) The cut off point of the tissular CEA with the best sensitivity and specificity for the diagnosis of normal mucosa is 386 ng/mg and for tumoral tissue is 1160 ng/mg. 2) There is no correlation between tissue and serologic CEA value. 3) The tissular level of CEA have a significant statistical correlation with Dukes stage (p < 0.003); other histological characteristics were no significative. 4) There are significant statistical correlations between serologic CEA and relapse but no with survival rates. CONCLUSIONS: 1) Serologic CEA levels depend on numerous factors. 2) There aren't correlations between preoperative serologic levels and tissular CEA levels. 3) Tissular CEA do not predict what patients will have an elevated serologic CEA level in relapse.

[New quantitative method for determining p53 protein in bladder carcinomas]. / Nuevo método cuantitativo para determinar la proteina P53 en carcinomas vesicales.

OBJECTIVE: To continue a study protocol on the molecular biology of bladder tumors, analyze protein p53 expression using a new quantitative analytical method and the biological implications of the changes in p53 expression. METHOD: From January, 1993 to January, 1995, 74 patients were studied. These patients were divided into two groups: the first group comprised 14 controls of urothelial tissue and the second comprised 60 cases of transitional cell carcinoma of the bladder. A quantitative method of immunoluminescence (LIA-mat p53 method) was utilized to analyze p53 expression. RESULTS: Tissue oncoprotein p53 was higher in patients with bladder carcinoma than in healthy urothelial tissue. Higher values of protein p53 was found in infiltrating and undifferentiated tumors and in those patients who died than in those who are alive. CONCLUSION: Protein p53 determination using this new quantitative method permits identification of subgroups of patients with tumors that have a more aggressive biological behaviour.

[Advances regarding tumor markers in bladder cancer]. / Avances en marcadores tumorales en cáncer vesical.

INTRODUCTION: Currently, there is not known tumoral marker for vesical carcinoma that would allow to distinguish when a surface tumour may become invasive. OBJECTIVE: To analyze the functionality of a series of biological substances (CEA, CA 50, CA 19.9 and TPS) in vesical carcinoma. MATERIAL AND METHODS: Between September 1992 and June 1994, a total of 385 biological specimens divided into two groups were analyzed. The first group comprised 271 serum samples from 81 control subjects and 190 patients with vesical carcinoma. The second group included 114 urothelial tissue samples (56 controls and 58 vesical carcinoma). Serum and tissue levels of CA, CA 50, CA 19.1 and TPS were determined in both groups by fluoroimmunoassay, RIA and IRMA, respectively. An statistical evaluation was done using Student's 't' and/or Mann-Whitney tests depending on whether data distribution adjusted to normal or not. RESULTS: Patients with vesical carcinoma, ana within this group those with infiltrant tumours, showed higher CEA serum levels. Also CEA tissue levels found in neoplastic vesical urothelium were higher than those in the control group (p < 0.05). Tissue levels were higher in infiltrant tumours. Higher TPS serum and tissue levels were found in the vesical tumours group. Same as with CEA, CA 50 also exhibited higher serum levels in the group with vesical Ca than in the controls (p < 0.01). Likewise, CA 50 tissue values were higher in the group with vesical Ca, more specifically in the infiltrant tumours group (p < 0.001). Statistically significant differences become apparent when the above values were compared to tissue samples from the control group (p < 0.001). On the other hand, serum CA 19.9 levels were lower in the vesical carcinoma group although tissue levels were higher in the vesical Ca group (p < 0.001). CONCLUSIONS: Transitional cell vesical carcinoma is a tumour that produces and secretes CEA, CA 50, CA 19.1 and TPS. CEA and CA 50 levels could be used as prognostic factors.

[The molecular biology of bladder carcinoma]. / Biología molecular del carcinoma vesical.

OBJECTIVE: The clinical course of transitional cell carcinoma of the bladder can be difficult to predict due to its potential to invade the muscle layer and/or develop to a high grade lesion. Bladder carcinoma can arise from genetic changes that may activate the oncogenes (-c-erbB2, c-erbB1, c-myc, ras, etc.) and/or inactivate the suppressor genes (p53, Rb). The aim of the present study is to continue a study protocol on the molecular biology of bladder tumors. METHODS/RESULTS: From January, 1993 to January, 1995, 85 patients were studied. These patients were divided into two groups: the first group comprised 14 controls of urothelial tissue and the second comprised 65 cases of transitional cell carcinoma of the bladder. p53 expression was determined by an immunohistochemical method (NCL-p53-DO7 monoclonal antibody). Quantification of the p8 oncoprotein in cytosol and EGFR (epidermal growth factor receptor) in membrane was performed by ELISA (Oncogene Science) and RIA (Vienna Lab), respectively. A statistically significant relationship between the expression of p53 and EGFR with tumor stage and grade was found. Quantification of p185 and EGFR showed higher values in the tumor tissue than in the control samples, but a worse survival could not be determined. CONCLUSIONS: The present study shows that p53 expression can be considered to be a prognostic factor. It provides useful information on the aggressive behaviour of the tumor and has a direct relation with the survival rates.

[Serum and tissue quantification of tissue polypeptide antigen (TPA) in transitional cell carcinoma of the bladder]. / Cuantificación sérica y tisular de antígeno polipeptídico tisular (TPS) en el carcinoma transicional de vejiga.

Tissue polypeptide antigen (TPA) is a marker of proliferative cellular activity. The aim of this paper is to demonstrate the production of this marker in bladder carcinomas and to study the biological behaviour in this type of patients. From September, 1992 to June, 1993, we studied 50 patients divided into two groups. The first group comprised healthy subjects and the second one comprised 30 patients with bladder carcinoma. In both groups, we determined the TPS in blood and tumoral tissue by RIA (TPS-IRMA Beki-Diagnostic AB). Our results demonstrated higher levels of TPS in tumoral tissue and blood than in healthy subjects (1887.83 and 197.33 vs 231.5 and 58.23 IU/ml) and higher levels of tissular and blood TPS for the undifferentiated tumors (989.66 and 231.5, 1748.2 and 210, 1842.6 and 219, 2010.7 and 220 IU/ml for Broders' classification 1, 2, 3 and 4).

[Involvement of epidermal growth factor receptor (EGFR) in the etiopathogenesis of prostatic proliferative processes]. / Implicación del receptor del factor de crecimiento epidérmico (EGFR) en la etiopatogenia de los procesos proliferativos prostáticos.

The prostatic growth factors require a membrane specific receptor to which they must bind in order to carry out their biological activities correctly. The aim of this study was to isolate and quantify the epidermal growth factor receptor in prostatic tissue and indirectly determine the growth factors acting on it (EGF, TGF alpha, PDGF, NGF, IGF). From September, 1992 to June, 1993, we studied 55 patients. These were divided into two groups: the first group comprised 49 patients with benign prostatic hyperplasia (BPH) and 6 patients with prostatic carcinoma comprised the second group. Samples of the prostate were obtained following suprapubic (12 cases), TUR (38 cases), radical prostatectomy (1 case) and transrectal biopsy (4 cases). The EGFR was determined by radioimmunoassay (EGFR-RIA, Vienna Lab, Labordiagnostica GmbH). For the overall group of patients, we obtained mean EGFR values of 6.36 +/- 0.59 fmol/mg of protein and a positivity of 96.36% and 100% for BPH and malignant proliferative processes, respectively. The foregoing data show that EGFR was isolated from the tissue we analyzed and has an evident role in the regulation of prostate growth.

[Quantitative determination of epidermal growth factor urothelial receptor (EGFR) in superficial and invasive bladder carcinoma]. / Dosificación cuantitativa del receptor urotelial del factor de crecimiento epidérmico (EGFR) en carcinomas vesicales superficiales e invasivos.

The epidermic growth factor (EGF) is a polypeptide which stimulates tissue proliferation. The mechanism of action takes place through an specific membrane receptor, known as the EGFR. Recent scientific contributions have allowed to know its implication in various tumoral processes (breast, ovary, bladder, etc), so that its expression may be used as a major prognostic factor. The objective of this work is to quantify and analyze the epidermic growth factor receptor (EGFR) in surface and invasive vesical carcinoma. To this end, 43 tissular samples divided in two groups were studied. The groups were: 1) Group 1 or control group, comprising 14 samples of "control" vesico-urothelial tissue, and 2) Group 2, assembling 29 patients with vesical carcinoma (17 surface and 12 infiltrant). All tissue samples underwent a process of homogenization and subsequent determination of membrane EGFR by means of radioimmunoassay (EGFR-Receptor Assay, Vienna Lab, Labordiagnostika GmbH). Our results demonstrate the expression of EGFR in both control and tumoral vesico-urothelial tissue in 100% of cases, with detection of significantly higher concentrations (p) in samples from vesical carcinoma than in those from control specimens (15.24 vs. 5.02 fmol/mg protein) and higher levels in infiltrant rather than in surface vesical carcinomas (18.92 vs. 11.4 fmol/mg protein).

[Usefulness of carcinoembryonic antigen in transitional cell carcinoma of the bladder]. / Utilidad del antígeno carcinoembrionario en los carcinomas vesicales de células transicionales.

Carcinoembryonic antigen (CEA), described by Gold and Freedman, was the first fetal tumor antigen isolated in tumor homogenates. The present study investigated the biological behaviour of CEA in superficial transitional cell carcinoma of the bladder, deep localized and disseminated. The study comprised 100 subjects; 30 carefully-selected healthy subjects comprised the first group and 70 patients with a diagnosed transitional cell carcinoma of the bladder comprised the second group. Serum CEA was determined by enzyme immunoassay (CEA, EIA, Roche). Our results suggest that serum CEA determination affords no diagnostic benefit in this type of malignant tumor.