Clinical correlates of circulating immune complex levels in advanced lung cancer. A discrimination analysis.

Sera from 53 patients with unresectable lung cancer were tested for the presence of immune complexes by 12 assays. 5 assays (EA rosette inhibition, ADCC inhibition, platelet aggregation, IgG and C3 concentrations in PEG precipitates) could discriminate cancer patients from healthy subjects with over 80% reliability. On the basis of 3 assays (EA-I, ADCC-I and PEG-C3) a function allowing a 100% correct classification could be formulated:--(EA-I)--0.5 (ADCC-I) + 2.4 (PEG-C3) greater than 69.3, i.e., results higher than 69.3 are characteristic for cancer patients and lower than 69.3 for normal subjects. The relationship between the immune complex levels and the average survival time was not altered by sex, age, histology and treatment. None of the immune complex assays or their combination were useful for the estimation of individual life expectation.

Evaluation of different methods for detecting circulating immune complexes. Studies in patients with lung cancer.

In a collaborative study involving 7 laboratories, sera from 53 patients with lung cancer, 37 primary and 16 secondary tumours, and sera of 40 healthy blood donors were tested by 19 different assays or assay modifications used for detecting immune complexes. In 12 out of 19 assays, significantly higher immune complex levels were found in the cancer patients than in the healthy subjects. Assays based on interactions between immune complexes and Fc receptors of different cells (lymphocytes, macrophages of platelets) discriminated between cancer patients and health subjects and a high percentage (47-87%) of positivity was observed in such assays in patients with lung cancer. In contrast, none of the tests based on immune complex-complement interactions discriminated between cancer patients and health subjects. Immunochemical analyses of the PEG precipitates obtained from the sera tested revealed that the concentrations of IgG, IgA and C3 were significantly higher in the precipitates obtained from patients sera than from control sera, but no significant differences were seen in IgM and C1q concentrations. A 100% correct classification of individuals tested was obtained on discriminant analysis of results with 3 assays: EA rosette inhibition, ADCC inhibition and C3 concentration in PEG precipitates. Correlation between results obtained with individual sera by the different assays was very poor: significant correlation coefficients were found in only 13% of all possible paired comparisons. Our results suggest that Fc receptor-dependent assays are more suitable for detection and measurement of circulating immune complexes in lung cancer than tests based on interactions with complement.

Possible mechanisms of the first step of the classical complement activation pathway: binding and activation of C1.

Different immunoglobulin preparations of human monoclonal IgM, normal human and rat IgG, as well as purified rabbit antibodies were treated by various methods, fragmentation, aggregation and complexing with antigen. The ability of the treated and untreated preparations to fix isolated human C1, to activate the classical complement pathway (to consume C4 in normal human serum) were compared. It was found that the different methods affected the conformation of the immunoglobulin molecules in different ways and induced changes to a greater or lesser extent in the two capacities of the preparations tested. In the case of the monoclonal IgM preparation a strong C1-fixation was observed without measurable complement activation. Other preparations, interfacially aggregated human IgG, BSA-anti-BSA and OA-anti-OA immune complexes had a very weak C1-fixing but a marked complement activating capacity. Some preparations, e.g. heat-aggregated IgG, both fixed and activated C1 effectively, aggregates with a complement-activating capacity without C1-fixing effect were separated by gel-filtration. It was demonstrated further, that at a given time only a part of the activated C1 molecules could be found fixed to the immunoglobulins, the other part was released into the fluid phase after activation. On the basis of the results of this and previous studies a hypothesis is proposed suggesting three possible results of the interaction between C1 and the different preparations: (1) firm fixation and activation; (2) binding not followed by activation and (3) a transient binding leading to activation. The possible application of this hypothesis for the interpretation of the results of the different methods for detecting immune complexes is discussed.