Improved performance of a double antibody radioimmunoassay for carcinoembryonic antigen

Evaluation of gastric carcinoembryonic antigen analysis as an aid during screening for gastric neoplasia in atrophic gastritis

The value of gastric juice and tissue carcinoembryonic antigen (CEA) analysis as an adjunct to endoscopic screening for gastric neoplasia was investigated in 61 patients with atrophic gastritis of whom 41 had other (superimposed) gastric lesions: six adenocarcinoma, four carcinoid, 23 regenerative polyps with or without dysplasia and eight fundic, or antral mucosal dysplasia. The gastric concentration of CEA did not differ between patient groups with different superimposed lesions. In these patients the gastric juice CEA concentrations were significantly increased in comparison with those in patients without superimposed lesions (p = 0.002). Gastric juice CEA concentrations above the upper range (+2SD) of those observed in normal controls were found in 40 (98%) of 41 patients with superimposed lesions and in 13 (65%) of 20 patients without such lesions (p = 0.001). At re-examination of 26 patients without neoplasia initially, after a mean interval of 32 months two (without polyps initially), had developed regenerative polyps, one an adenoma, and one an adenocarcinoma. These four had raised gastric juice CEA concentrations at the initial examination.

Urinary modified nucleosides as tumor markers in cancer of the urinary organs or female genital tract

Using a sensitive and specific method involving high-performance liquid chromatography, urinary levels of four modified nucleosides--pseudouridine (psi), 1-methylinosine (m1I), 1-methyladenosine (m1A), and 1-methylguanosine (m1G)--were investigated before and after treatment in 31 patients with cancer of the urinary organs or the female genital tract. Before treatment m1I was the most frequently elevated nucleoside (77%). Pretreatment urinary levels of psi, m1I, and m1A in patients with stage 2-4 cancer of the female genital tract were significantly elevated compared to human healthy volunteers (p less than 0.005). Compared with the other nucleosides, psi appeared to correlate more closely with the clinical outcome (progression or regression) of patients with cancer of the female genital tract. In the case of patients with cancer of the urinary organs, m1I followed the clinical outcome better than the other nucleosides measured. Therefore psi and m1I seem to be useful for monitoring genito-urinary cancers.

Tumour and cellular localization by use of monoclonal and polyclonal antibodies to placental alkaline phosphatase

Monoclonal and polyclonal antibodies against placental alkaline phosphatase (PLAP) were evaluated for tumour immunolocalization of human PLAP-producing Hep 2 tumours in nude mice. The antibodies were labelled with 125I and injected i.p. in mice with developing Hep 2 tumours. The distribution of 125I-anti PLAP in various tissues showed that the labelled antibody was enriched in the tumour, the mean concentration ratio being 7.1 and 6.8 for polyclonal and monoclonal antibodies, respectively. A PLAP negative tumour (RD) showed a mean ratio of 1.2. There was a positive correlation between PLAP content and uptake of labelled antibody in the tumours. Hep 2 tumour cells in tissue culture showed 100% positivity for PLAP, while imprints of the tumour after passage in nude mice showed 40-50% positivity. PLAP offers potential as a useful marker for localizing tumours in humans.

NCA: a differentiation antigen of myelopoietic cells in humans and hominoid monkeys

NCA, a normal colonic and granulocytic antigen, could be demonstrated in serum and in myelopoietic, but not lymphopoietic or erythropoietic, cells of Homo sapiens and other Primates. The levels of NCA in both serum and myelopoietic cells of Homo and hominoids were higher than those of more distant relatives of the same order. Thus, the classic phylogenetic differences are reflected also by the distribution of NCA. Hyperimmunization of Macaca irus, in which the NCA content of serum and cells is low, led to occurrence of anti-NCA IgG in all animals. The phylogenetic differences may accordingly have to do with slight antigenic NCA differences between Homo and other Primates rather than differences in amount only. Purified NCA did not affect growth and maturation of myelopoietic stem cells in vitro, whereas anti-NCA inhibited development of the majority of myelopoietic clusters and colonies.

Antigenic reactivities of purified preparations of carcinoembryonic antigen (CEA) and related normal antigens using four different radioimmunoassay systems for CEA

Antigenic reactivities of 4 purified carcinoembryonic antigen (CEA) preparations including the WHO standard CEA and 4 CEA-related antigens--nonspecific cross-reacting antigen (NCA) from normal lungs, nonspecific cross-reacting antigen-2 (NCA-2) from meconium, normal fecal antigen-1 (NFA-1) and normal fecal antigen-2 (NFA-2) from normal adult feces--were investigated using 4 different radioimmunoassay (RIA) systems (Hoffman-La Roche, CEA-IRE-SORIN, Dainabot and our system). In general, all CEA preparations including the respective standard CEAs gave similar reactions with these assay systems. Differences in reaction intensity were, however, observed among the CEA preparations, and the order of reaction intensity among the preparations varied due to assay systems, e.g., a given CEA preparation which reacted most strongly with an assay system reacted very weakly with another assay system, and vice versa. Both NCA-2 and NFA-2, which possessed physiochemical properties very similar to those of CEA, revealed reactivities indistinguishable from those of the CEA preparations in all systems. NCA, which showed a partial cross-reactivity with CEA in the double immunodiffusion test, revealed a remarkably weak reactivity in all systems. NFA-1, however, which was also partially cross-reactive with CEA but antigenically unrelated to NCA, showed a fairly strong reactivity in all systems, but its dose-dependent slope was apparently different from that of the other antigen preparations.

Decrease of a normal colon and granulocyte antigen (NCA) in whole lysed blood of myelosuppressed patients

A normal colon and granulocyte antigen (NCA) is found in large amounts in neutrophil granulocytes. A method combining freeze-thawing and ultrasonication of whole blood proved to be the most efficient in recovery of NCA from separated granulocytes or whole blood. NCA in whole blood was studied in healthy persons and in 8 patients with embryonal carcinoma undergoing chemotherapy, with a myelosuppressive effect. A strong correlation (r = 0.96) between the total number of peripheral granulocytes and the amount of NCA in lysed whole blood was found in both healthy persons and myelosuppressed patients. The NCA content per 10(9) granulocytes was equal in granulocytopenic and healthy persons. NCA measurements therefore give information about the total number of granulocytes in peripheral blood within the limits measured (0.5-6.5 x 10(9) granulocytes/l).

A new direct solid-phase radioimmunoassay for carcinoembryonic antigen without pretreatment of serum samples

We examined some fundamental conditions for establishing a new simplified radioimmunoassay system for CEA. The system consists of antibody-coated plastic beads and radiolabeled tracer antibody. The reactivity of this system with several purified CEA preparations can vary due to differences in the nature of antibody preparations used either for coating beads or for tracer antibody. Among purified CEA preparations, we also found some differences in reactivity with a given immobilized antibody preparation. The assay system that resulted, referred to as bead-direct-RIA, needs neither pretreatment of serum samples nor centrifugation in any of its steps, and has a wide assay range, measuring CEA concentrations of 1-300 ng/ml using only 50 microliters of a serum sample. Its accuracy, reproducibility, dilution effect, and analytical recovery proved the reliability of this method. The mean value +/- S.D. for 48 normal adults was 2.5 +/- 1.0 ng/ml. It was found by comparative studies that the Z-gel method of Hoffmann-La Roche generally gave the highest values, the antibody-disc method of Dainabot gave the lowest, and the values by bead-direct-RIA fell between those of the 2 other methods and correlated well with either value. Follow-up studies on patients with malignancy indicated that the values by bead-direct-RIA fluctuated in parallel with those by Roche and Dainabot kits.

Tumor localization of CEA-containing human tumors in nude mice by means of monoclonal anti-CEA antibodies

Three different monoclonal antibodies (mabs) directed against conformation-dependent protein determinants in carcinoembryonic antigen (CEA) were evaluated for tumorimmunolocalization of CEA-containing human tumors grown in nude mice. The antibodies were of high affinity and of the IgG1 subclass. They did not react with non-specific cross-reacting antigen (NCA), biliary glycoprotein I (BGP I) or orosomucoid. 10-20 muCi (1-2 micrograms) of 131I-labelled mab was injected intraperitoneally into mice bearing heterografts of the CEA-containing human tumor cell lines LS 174T, Detroit 562 or HT-29 or into mice bearing grafts of the non-CEA-containing human tumor cell line RD. The concentration of radiolabelled mab in the tumor compared with the mean concentration in the mouse increased up to day 4 after injection. In contrast, the concentration ratios of radioactivity in the other tissues decreased with time. Excellent tumor localization was obtained with mab 38S1 in mice bearing LS 174T tumors. The concentration of radioactivity (cpm/g) tissue wet weight) was 24-390 times higher in the tumor than in the muscle. The difference between tumor and liver and between tumor and blood was 4-89 times and 2-25 times, respectively. A mean value of 40% of the total radioactivity in the mouse at day 4 was localized to the tumor. Mab 38S1 also showed positive tumor localization in mice carrying Detroit and HT-29 tumors. Means of 22% and 18% of the total radioactivity in the mouse at day 4 were found in these tumors, respectively. Mab 38S1 was, however, not enriched in the tumor of mice bearing the non-CEA-containing sarcoma RD. Furthermore, when LS 174T tumor-bearing mice were injected with normal mouse IgG, no enrichment of radioactivity in the tumor was seen. Positive tumor localization with mab 48 was obtained while mab 9 was not enriched in the tumor. Scanning pictures of LS 174T-grafted mice on day 4 after injection of 131I-mab 38S1 showed a definite radioactive spot corresponding to the tumor and in addition a spot corresponding to the liver or the heart.

Urinary CEA for prediction of survival time and recurrence in bladder cancer

The prognostic information from carcinoembryonic antigen was evaluated in bacteria-free urines of patients with bladder carcinoma. Patients with elevated (greater than or equal to 30 ng/ml) U-CEA had a poorer relative and symptom-free survival than patients with initial U-CEA below 30 ng/ml. Patients in whom U-CEA decreased from before to after radiation treatment had a better survival rate than patients with increasing U-CEA. These findings were most significant in cases with large (T3 + T4) tumors or with tumors of a lower differentiation (G3). U-CEA appeared to be an independent variable for prognostic evaluation of survival, since these differences were also true within the subgroups of known variables such as classes or grades. All but one of the patients, in whom short-term local control was obtained, had a posttreatment U-CEA below 50 ng/ml. In oncology units, where the more malignant bladder tumors are treated, U-CEA determinations may indicate which patients require intensified monitoring or treatment.

Assessment of serial CEA determinations in urine of patients with bladder carcinoma

Carcinoembryonic antigen (CEA) levels in urine and serum from 294 patients with bladder cancer in varying stages have been clinically evaluated. All urine samples were obtained from patients with intact renal function and without bacterial infection in the bladder. The samples were collected before, during, and at follow-up examination after radiotherapy. They were perchloric acid extracted before being assayed in a double-antibody radioimmunoassay. The geometric mean of urine CEA levels for patients with primary tumors of Stage T1 or T2 was significantly lower than that for those with Stage T3 or T4 disease. The urine CEA levels for patients with tumors of various histologic grades did not differ. The urine CEA levels decreased from before to after radiation treatment of the primary tumor. Patients with recurrence within six months after undergoing primary treatment had higher initial mean urine CEA levels than did those without evidence of recurrence. The prognostic information for recurrence was limited to the more advanced tumors. Differences were also found between the means of samples taken before recurrence and after treatment of recurrent tumors; with regression of the tumor, a lower mean urine CEA level was found; with progression, a higher value. Urine CEA levels before any treatment were higher when the patients had a short survival time. Serum CEA levels were not related to stage or grade of the primary bladder tumor but levels were slightly elevated with metastases. The determination of urine CEA levels seems to be useful in the follow-up of patients with bladder carcinoma because when initially high, it adds to the information of the T classification and predicts early recurrence, and the monitoring of individual patients after primary treatment is useful for detecting recurrence.

Placental alkaline phosphatase, alphafetoprotein and carcinoembryonic antigen in testicular tumors. Tissue typing by means of cytologic smears

Indirect immunofluorescence and radioimmunoassay with specific rabbit antisera demonstrated the occurrence of alphafetoprotein (AFP), carcinoembryonic antigen (CEA) and placental alkaline phosphatase (PLAP) in primary testicular tumor cells. Embryonal carcinomas had AFP- and CEA-containing cells, sometimes PLAP. PLAP and sometimes CEA were found in seminoma cells. Sera from patients with advanced non-seminomatous tumors could contain any of these antigens or any combination of them. Sera from patients with seminomas had raised PLAP or CEA. PLAP appears to be a new marker for seminoma.