Serum CEA and CA 15-3 as prognostic factors in primary breast cancer.

In the present study, we investigated the association of the serum levels of the tumour markers carcinoembryonic antigen and cancer antigen 15-3 with disease free survival and death from disease in 1046 women with breast cancer without metastases at the time of primary diagnosis in relation to age and the established prognostic factors tumour size, lymph node status, histological grading and hormone receptor status. We found that elevated pre-operative serum marker values were correlated with early relapse (cancer antigen 15-3; P=0.0003) and death from disease (carcinoembryonic antigen, cancer antigen 15-3; P=0.0001 both) in univariate analyses. By comparing pre- and post-operative values we found a decline in values post-surgery. In those patients where marker levels of carcinoembryonic antigen decreased more than 33%, a significantly higher risk for relapse and death from disease (both P=0.0001) in univariate analyses was observed. In multivariate analysis this decrease of carcinoembryonic antigen proved to be an independent prognostic factor. The results for cancer antigen 15-3 were comparable to carcinoembryonic antigen in univariate analyses but showed no significance in multivariate analysis. In this study the post-operative decrease of the serum tumour marker carcinoembryonic antigen was a strong independent prognostic factor for disease free survival and death from disease in breast cancer patients.

The significance of Tenascin-C serum level as tumor marker in squamous cell carcinoma of the head and neck.

BACKGROUND: Tenascin, an extracellular matrix glycoprotein, is transiently present in embryonic tissue, in benign granulation tissue, but also in several highly anaplastic tumors like fibrosarcoma, melanoma and squamous cell carcinoma of the skin. This study was performed to validate elevated Tenascin serum levels as a possible marker for head and neck squamous cell carcinomas (HNSCC). PATIENTS AND METHODS: Tenascin serum levels were evaluated in patients with primary (n = 92) and with recurrent (n = 28) HNSCC. Patients with benign, non inflammatory ear, nose and throat diseases (n = 16) served as the control. The Tenascin serum levels were measured by ELISA (Aventis). RESULTS: Serum Tenascin concentrations of patients with benign ENT diseases ranged between 0.37 and 2.19 micrograms/ml (n = 16, mean +/- SD: 1.23 +/- 0.59 micrograms/ml), of patients with HNSCC (primary diagnosis) between 0.05 and 8.75 micrograms/ml (n = 92, mean +/- SD: 1.81 (1.36 micrograms/ml) and of patients with recurrent HNSCC between 0.53 and 10.0 micrograms/ml (n = 28, mean +/- SD: 2.78 +/- 2.2 micrograms/ml). CONCLUSION: We found a significant elevation of Tenascin serum levels only in patients with higher tumor stages (T4/UICC4) (p < 0.01/p < 0.1) or recurrent disease compared to Tenascin serum levels in healthy controls. Thereby Tenascin serum levels cannot be used clinically as a routine serum marker for the control of head and neck cancer. Further investigations are necessary to evaluate whether the measurement of Tenascin levels as tumor markers could offer additional information to the clinical outcome of patients with HNSCC.

Nuclear mitotic apparatus protein (NuMA) in benign and malignant diseases.

Nuclear mitotic apparatus protein (NuMA) is a 239 kDa internal nuclear matrix protein described to be elevated in cancer patients, especially in colorectal carcinoma and early colorectal cancers. We tested the significance of NuMA as tumour marker in colorectal cancer and also the sensitivity/specificity profile in general. Therefore, we investigated in a retrospective clinical study 507 sera from patients suffering from solid tumours, with the main emphasis on colorectal carcinoma, and 418 sera from patients with benign diseases and healthy individuals. Testing was done with a double monoclonal enzyme immunoassay detecting head and rod domain of NuMA and results were compared to the established tumour associated antigens. Based on a specificity of 95% versus the benign reference group of gastrointestinal diseases, we found--at the time of primary diagnosis--a sensitivity for colorectal cancer of 8% for NuMA, 36% for CEA and 17% for CA 19-9. Regarding T-stages of colorectal cancer no marker detected T1 when regarding 95% specificity-cut-off value but NuMA showed little more sensitivity when based on a 95% specificity cut off value versus healthy. This could not be shown in Dukes' stages. Regarding all other solid tumours tested--all based on a specificity of 95% for the corresponding benign reference groups--no advantage of NuMA in sensitivity for all other solid tumours investigated was found. No additional sensitivity could be observed. Based on our results, the NuMA-assay in its present form has no clinical relevance.

Tumour markers CEA and CA 15-3 as Prognostic factors in breast cancer--univariate and multivariate analysis.

Tumour markers are putative prognostic indicators for patients with breast cancer, but have not been elevated independently by multivariate analysis in a large patient number. In 550 patients with breast cancer without known metastases the levels of the serum tumour markers CEA und CA 15-3 were determined preoperatively and during follow-up. The prognostic relevance of these markers for recurrence (n = 128/487) and death of disease (n = 55/550) was evaluated in relation to established prognostic factors. In univariate analysis tumour size, lymph nodes, histological grading, age, hormone receptors, preoperative value of CEA (cut-off 2 ng/mL) and CA 15-3 (cut-off 25 U/mL) and their decrease of more than 33% within seven months after operation were significant for relapse. The results for death of disease were similar except for age. In multivariate analysis tumour size, lymph nodes and decrease of CEA > 33% (p < 0.001) were independent prognostic factors for recurrence. For overall survival tumour size, lymph nodes, histological grading and preoperative levels of CEA > or = 2 ng/mL (p = 0.038) and of CA 15-3 > or = 25 U/mL (p = 0.007) were independent prognostic factors. Pre- and postoperative values of the tumour markers CEA und CA 15-3 are strong independent prognostic factors for relapse and survival in breast cancer patients.

BTA-TRAK--a useful diagnostic tool in urinary bladder cancer?

During recent years the BTA-TRAK-assay (Bard Diagnostics, Redmont, USA) has been described in several investigations to be of clinical utility for patients suffering from bladder cancer. In a prospective study we investigated over four months the voided urine samples of all consecutive patients undergoing cystoscopy independent of their clinical background (n = 244) with the BTA-TRAK-assay. With a specificity of 95% for benign urological diseases (cut off: 1300 U/mL) we found a sensitivity of 13% for active bladder tumours. Using healthy individuals as a reference group (cut off: 40 U/mL) we found a sensitivity of 56% (specificity 67%). Using the cut off value recommended by the manufacturer (14 U/mL) a specificity of 54% and a sensitivity of 62% was found. For patients without relapse (NED) versus patients with active bladder tumours we got a specificity of 55% and a sensitivity of 62%. Due to an insufficient specificity and sensitivity the BTA-TRAK-test is not able to replace cystoscopy nor to improve existing diagnostic strategies in bladder cancer.

Pro-gastrin-releasing peptide (ProGRP)--a useful marker in small cell lung carcinomas.

Gastrin-releasing-peptide (GRP), the mammalian counterpart of amphibian bombesin, has been reported to be produced by cells of SCLC. Using recombinant ProGRP Yamaguchi et al developed an enzyme immunoassay for the measurement of this more stable precursor of GRP. We focused our interest on the comparability of ProGRP to neuron specific enolase (NSE), CYFRA 21-1 and CEA. For this purpose we investigated the sera of 272 patients with histologically proven carcinomas of the lung (87 SCLC, 185 NSCLC). The sera of 74 patients with benign diseases of the lung and smokers served as a reference group. At a specificity of 95% ProGRP and NSE possessed comparable sensitivities (47% versus 45%) in small cell lung carcinomas. ProGRP showed only a few more positive test results than NSE, but reached much higher value levels than NSE. ProGRP and NSE showed a clear additive sensitivity of about 20%. In NSCLC CYFRA 21-1 was the leading marker with 63% sensitivity, whereas ProGRP seldom showed a "false positive" test result. ProGRP proved a very high specificity and good sensitivity for small cell lung carcinomas and therefore enables diagnosis of small cell lung carcinoma in patients with lung tumours of unknown origin as well as good control of efficiency of therapy.

Carbohydrate-deficient transferrin is not a useful marker for the detection of chronic alcohol abuse.

BACKGROUND: The role of carbohydrate-deficient transferrin (CDT) as a reliable marker for the detection of chronic alcohol abuse has been discussed controversially. METHODS: Therefore, we investigated CDT in the sera from 405 subjects with different alcohol intake. Besides healthy control subjects (n = 42), inpatients and outpatients in a department of gastroenterology (n = 325) and patients admitted to a department of otorhinolaryngology (n = 38) were studied. A total of 213 patients suffered from various forms of liver diseases, and 89 patients had liver transplantation. CDT values were determined by a double-antibody radioimmunoassay. RESULTS: In the 241 alcohol-abstinent subjects, CDT levels ranged from 3 to 90 units L-1 (median = 12); the 92 moderate drinkers (20-60 g of alcohol per day) showed values from 3 to 40 units L-1 (median = 12), and the 72 subjects with chronic alcohol abuse (> 60 g per day) revealed CDT levels from 3 to 100 units L-1 (median = 16). The diagnostic specificity for alcohol abuse was 86.8% for men (sensitivity 36.9%) and 95% for women (sensitivity 0%). CONCLUSION: Our data indicate that measurement of CDT does not reach clinical use in the detection of chronic alcohol abuse in an unselected population because of its insufficient specificity and sensitivity.

Serum levels of intercellular adhesion molecule-1 in squamous cell carcinoma of the head and neck.

Soluble intercellular adhesion molecule-1 (s-ICAM-1) was measured in the sera of 131 patients with primary and 50 patients with recurrent squamous cell cancer of the head and neck (HNSCC). 30 patients with benign ear, nose and throat diseases served as controls. s-ICAM-1 levels in serum are high in patients with HNSCC, particularly in the advanced tumor stages (UICC IV). Highest levels can be measured at the time of tumor recurrence and locoregional lymph node metastases. The sensitivity (95% specificity) of s-ICAM-1 (cutoff-level: 473 ng/ml) is 4% at primary diagnosis and 12% for recurrent disease. A coefficient of correlation for s-ICAM-1 in combination with SCC, carcinoembryonic antigen and CYFRA 21-1 indicates that no correlation can be found of s-ICAM-1 compared with traditional tumor markers. Due to overlapping values in control and patient groups s-ICAM-1 is not suitable for a specific clinical use in HNSCC.

Methodological and clinical evaluation of two automated enzymatic immunoassays as compared with a radioimmunoassay for neuron-specific enolase.

We evaluated the clinical and methodological features of the neuron-specific enolase radioimmunoassay (NSE RIA) (Pharmacia = Ph) with the neuron-specific enolase enzyme immunoassay (NSE EIA) on the ES 700 (Boehringer Mannheim = BM) and the NSE EIA on the Cobas Core System (Roche = Ro). A total of 253 serum samples obtained from 37 healthy persons, 45 patients with benign lung diseases, 124 patients with lung cancer (42 with small cell lung cancer, 23 with adenocarcinoma, 21 with squamous cell carcinoma, 11 with large cell carcinoma, and 27 with unknown histology), 34 with lung metastases, 7 patients with sarcoma and 6 patients with malign lymphatic diseases were stored at -80 degrees C and assayed retrospectively. The intra- and inter-assay imprecisions were lower for the automatized test systems than for the RIA. Correlation between the EIA's and the RIA was better for NSE (Ro) than for NSE (BM) (BM/Ph: r = 0.93 and slope = 0.54; Ro/Ph: r = 0.95, slope = 0.79), but weaker than the correlation between the two EIA's: over the whole range r = 0.96, neuron-specific enolase < 50 micrograms/l: r = 0.97, neuron-specific enolase < 20 micrograms/l: r = 0.92. Fixing the specificity at 95% versus benign lung diseases we found a cut off value of 11.9 micrograms/l for NSE RIA (Ph), 15.9 micrograms/l for NSE EIA (BM) and 13.5 micrograms/l for NSE EIA (Ro). Based on this specificity of 95% versus benign lung diseases as the clinically relevant reference group, the sensitivity for NSE RIA was 32% for all lung cancer and 45% for small cell lung cancer, for NSE EIA (BM) 35% for all lung cancer and 43% for small cell lung cancer, the NSE EIA (Ro) had a sensitivity of 42% for all lung cancer and 57% for small cell lung cancer. In a follow-up study of two patients with small cell lung cancer a good comparability for all three assays in the kinetics, but a marked difference in the neuron-specific enolase value levels was found. The results show that the NSE EIA (Ro) on Cobas Core system is the most sensitive assay for the detection of small cell lung cancer.