Since 1972 plasma CEA levels of 25 cancer patients have been assayed to evaluate the reliability of CEA as an early indicator of recurrent gastrointestinal cancer. Identification of significant elevations in CEA levels required definition of exactly what a given value meant. Intraassay and interassay accuracy was determined and graphed as a CEA NOMOGRAM, which measures the observed CEA level against the 95% confidence limits for that observation and thus can be used to identify statistically significant increases. A statistically significant rise above a baseline value established by the NOMOGRAM proved to be a correct indicator of tumor recurrence in 22 (88%) of 25 patients who underwent second-look intraabdominal operations (22 colorectal, 2 gastric, and 1 pancreatic). In each case, other accepted procedures, such as liver enzymes, scans, and x-rays, were nondiagnostic. Of the 22 patients with proved tumor recurrence, 16 (73%) had distant metastases and 6 (27%) had localized tumors. One patient remains tumor-free three years after second-look operation and has had no significant change in CEA levels. More frequent serial CEA determinations combined with sound clinical judgment should facilitate earlier detection of recurrent gastrointestinal cancer.

A commercially available enzyme immunoassay system for detecting autoantibodies to double-stranded DNA, deoxyribonucleoprotein, Smith, ribonuclearprotein, Sjögren's syndrome-associated antigens A and B, and scleroderma-associated antigen 70 was compared to the conventional immunofluorescence assay for double-stranded DNA and double diffusion assays for extractable nuclear antigens. There was excellent correlation between methods, but it appears that the enzyme immunoassays are more sensitive. Based on the results of this study, the authors recommend performing anti-nuclear antibody screening at two dilutions, with enzyme immunoassay follow-up of appropriate patient sera that are positive on anti-nuclear antibody testing. Nucleolar and centromere pattern anti-nuclear antibodies are diagnostic for variants of scleroderma and need no further evaluation. Negative anti-nuclear antibody tests performed using HEp-2 tissue culture cells require no further evaluation.

C-reactive protein (CRP) 1 is an acute-phase protein in man that is of considerable interest, in part because serum concentrations increase from 20-to several hundredfold during inflammation or acute tissue injury (1, 2). The association of elevated levels of CRP with activation of certain immune responses led to the postulation that CRP is involved in initiation or regulation of lymphocyte function (3). Recent evidence of similarities between CRP-binding cells and the effector cell responsible for natural killer (NK) cell-mediated lysis prompted our investigation of the role of CRP in the NK response. NK and CRP-binding cells are large granular lymphocytes (LGL), bear IgG Fc receptors, represent only a small percentage of the total lymphocyte population, and rosette similarly with sheep erythrocytes (4-8).These experiments were initiated to determine whether NK cells bear CRP, whether cell-bound CRP is involved in the effector phase of the NK response, and whether CRP or CRP-complex binding would alter NK function (9-12), that is, they were designed to examine whether CRP or CRP complexes are involved in NK cellmediated killing. The results demonstrate that CRP is present on at least certain NK effectors, that CRP or a molecule that co-caps with CRP is required for optimal NK function, and that cell-surface CRP and not fluid-phase CRP is involved in the mediation of NK activity. Materials and MethodsPurification and Characterization of CRP and CPS. CRP was prepared from pooled human serous fluids by phosphocholine-affinity, DE-52, and Sephacryl S-200 (Pharmacia Fine Chemicals, Div. of Pharmacia Inc., Piscataway, N J) column chromatography by methods previously described (13). CRP preparations were concentrated by ultrafiltration, sterile filtered, and stored at 4°C. When examined by reduced sodium dodecyl sulfate-polyacrylamide gel electrophoresis (14), purified CRP produced a single band corresponding to CRP subunits with a molecular weight of 21,500 + 500. A single peak of 1251 was obtained after sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis of labeled CRP (15). Purified CRP was concentrated to 10 mg/ml by ultrafiltration and no IgG, IgA, IgM, C3, or serum amyloid P component were detected using radial immunodiffusion plates with limits of sensitivity of 5/zg/ ml. Ouchterlony analysis of the concentrated material revealed a single precipitation line with

No abstract

Changes in immunoglobulin class and subclass levels and the development of antitumor antibodies were assessed in normal and tumor-bearing mice challenged with Corynebacterium parvum. C. parvum administration resulted in a marked increase in certain immunoglobulin levels, especially Ig G2b, and in the development of antibodies reacting with syngeneic and allogeneic tumor cells. The serologic changes induced by C. parvum were dependent on the dose and route of administration; preliminary studies suggested that they may have been largly independent of T-cell function. These changes were suppressed by the administration of gold salts, which also inhibited the antitumor effect of C. parvum.