The natural history of antitumour immunity in human breast cancer assayed by tube leucocyte adherence inhibition

Tumor size, leukocyte adherence inhibition and serum levels of tumor antigen in dogs with the canine transmissible venereal sarcoma

Tumor antigen (TA) associated with the canine transmissible venereal sarcoma (CTVS) was detected in the sera of dogs bearing the tumor. Rabbit antisera specific for tumor antigen and 3 M KCl extracts of CTVS cells were used in both a competitive enzyme-linked immunosorbent assay (ELISA) and antigen-capture ELISA to quantify levels of circulating TA. In a study of 29 dogs bearing the transplanted CTVS, levels of circulating TA correlated positively with tumor volume. In a longitudinal study of four dogs receiving a transplant of 10(8) viable CTVS cells, circulating CTVS antigen was detected transiently 2 days after transplantation, while persistent levels of TA associated with increasing tumor volume were demonstrable 19-34 days after transplantation. In three of four tumor-bearing dogs, levels of serum TA correlated inversely with values obtained with peripheral blood leukocytes in the leukocyte adherence inhibition (LAI) assay; elevated levels of circulating TA found in dogs with large (greater than 7 cm3) tumors were associated with decreased LAI reactivity of peripheral blood leukocytes. TA could not be detected in sera 48-72 h after surgical removal of CTVS whereas LAI reactivity of peripheral blood leukocytes to CTVS antigen rebounded 1-3 weeks following tumor excision. Results of this study support the use of the competitive ELISA and LAI techniques in assessing levels of circulating tumor antigen, tumor burden and tumor-specific immunity.

Circulating anti-tumor and autoantibodies in breast carcinoma: relationship to stage and prognosis

Serum antibodies to breast tumor antigen(s) and circulating autoantibodies were tested in 175 patients with various stages of carcinoma of the breast, followed for a mean period of 51 months. Antibodies to surface membrane and to cytoplasmic antigens of autologous and allogeneic tumor cells were measured. Peripheral lymphocyte count and skin reaction to six recall antigens were also tested. Patients with metastatic disease had significantly lower prevalence of antibodies to autologous tumor cells and lower total lymphocyte count than patients with early breast cancer. Patients with locally advanced disease (greater than or equal to 4 positive axillary nodes) had the highest frequency of anti-tumor antibodies, the second highest lymphocyte count, but with the lowest prevalance of autoantibodies. Presence or absence of anti-tumor or autoantibody did not correlate with results of skin tests or other standard blood tests. Among patients with locally advanced or metastatic breast cancer, those who had a positive skin test or whose lymphocyte count was 1500 to 2500 per cu mm had significantly better 5-year absolute survival rates (p = 0.04, p = 0.002, respectively). This study suggests that in patients with locally advanced or metastatic breast cancers, skin test reactivity and optimal peripheral lymphocyte count may be useful prognostic indicators. In contrast, neither the presence of anti-tumor antibodies to membrane or cytoplasmic antigens, nor the presence of autoantibodies, correlates with prognosis in patients with early or late breast cancers.

A study of the immune response to the organ-specific neoantigen of human bladder cancer

The antitumor immune response of monocytes armed with cytophilic antitumor antibody to an organ-specific neoantigen of human bladder cancer was measured by the tube leukocyte adherence inhibition assay. Of 29 patients with Stage A bladder cancer, 27 had positive tests, whereas of 11 patients with Stage B, C and D bladder cancer, 2 had positive tests. If the leukocytes from advanced bladder cancer patients were preincubated briefly with prostaglandin E2, the negative response was converted to positive. Of 189 patients with diseases of other organs, 2 had positive tests. Patients with acute or chronic cystitis had negative tests. The sensitivity of the assay was 88 per cent; specificity was 95 per cent. The bladder cancer extracts prepared from metastatic tissue or the tissue-cultured cell line, T24, had similar activity in the assay. Normal bladder tissue did not share the bladder cancer organ-specific neoantigen nor did cancers of other organs. Bladder epithelial cells expressed the OSN before they acquired invasive properties since patients with dysplastic bladder epithelium or in-situ cancer had positive tests. Antitumor immunity was often detectable before a recurrence became obvious by cystoscopy or cytology of urine and became undetectable quickly after removal of the bladder cancer. Of 100 patients with prior bladder cancer without evidence of exophytic tumor, 18 had positive tests; the possibility must be considered seriously that some had precancerous changes which formed no visible gross abnormalities. The results suggested that antitumor immunity to bladder cancer was a sensitive indicator of precancerous or cancerous changes existing in epithelial cells.

Predictive value of tube leukocyte adherence inhibition (LAI) assay for breast, colorectal, stomach and pancreatic cancer

The tube LAI assay measures accurately antitumor immunity in patients with early cancer but fails to detect up to 75% of patients with advanced cancer due to excess circulating organ-specific neoantigen (OSN). Substances such as prostaglandin E2 (PGE2) or aminophylline, which increase intracellular nucleotides in leukocytes of patients with advanced cancer reversed this nonreactivity and greatly increased the sensitivity of the assay without any loss of specificity. Antitumor immunity can now be detected in advanced cancer, and a combination of the two assays gives prognostic potential to the assay: a positive test with PGE2 and negative test without indicates the patient has a large tumor burden. The specificity of the assay for each cancer was high and in most instances was greater than or equal to 95%. The PGE2 stimulated assay retained the high specificity. The sensitivity of the regular tube assay was often low, 33-56% because of the many advanced cancer patients tested, whereas the PGE2 stimulated assay showed almost a two-fold increase in sensitivity, 67-93%. The diagnostic value of the assay was estimated by calculating the predictive value for different prevalences of cancer. It was found that at low prevalences of cancer as found in the general population, the assay had a low diagnostic value since few patients with a positive test would have the cancer tested for. With prevalences of cancer of 5% or greater as might be found in a tertiary care clinical setting, the assay would seem to have diagnostic value since one half or more patients with a positive test would have the cancer tested for. Most false positives, but not all, are found in patients who have lesions that are often considered to increase their risk for cancer: severe dysplasia of the breast, colon adenomas, chronic atrophic gastritis and chronic pancreatitis, suggesting that the assay predicts oncogenesis.

Prospective evaluation of the leukocyte adherence inhibition (LAI) test in breast cancer using a panel of extracts from known and unknown primary tumors

The antigen-induced leukocyte adherence inhibition (LAI) test is an assay of cellular immune reactivity which was expected to offer a promising adjunct to currently available diagnostic and monitoring procedures in cancer patients. We have applied this assay to (1) 83 inpatients clinically suspected of having breast cancer, (2) 50 out-patients before mammography, and (3) 37 healthy women. In order to account for the known heterogenicity of breast cancer, we performed LAI assays with extracts from 15 primary tumors of known and 14 primary tumors of initially unknown histology. Thus, the problem of tumor specificity in the LAI assay was tackled in a double-blind fashion. The results obtained clearly show that breast-tissue-specific rather than tumor-specific responses were detected by LAI testing. The LAI assay failed to discriminate between high-risk patients with and without cancer and between extracts from breast tumors with and without carcinoma. The results are discussed with regard to the many claims for the detection of specific tumor immunity in the literature.

Sequential changes in peripheral blood leukocyte adherence inhibition (LAI) reactivity during progressive growth and spontaneous regression of canine transmissible venereal sarcoma

Sequential changes in canine transmissible venereal sarcoma (CTVS) extract-induced leukocyte adherence inhibition (LAI) reactivity of peripheral blood leukocytes (PBL) were monitored in dogs from the time of tumor inoculation through progressive growth and spontaneous regression or surgical excision of CTVS. The time-course profiles of LAI reactivity of tumor dog PBL correlated with the clinical stage of tumor growth, substantiating our previous findings in a cross-sectional study. The kinetics of the time-course appearance of LAI reactivity observed and the demonstration of a rapid anamnestic response following tumor challenge indicate further that LAI is a reliable in vitro measure of in vivo tumor immunity. A significant increase in CTVS antigen extract-induced LAI reactivity observed in the PBL during tumor regression suggest that LAI reflects a function effector cell mechanism associated with spontaneous regression of CTVS. By contrast, absence of significant LAI reactivity of PBL during progressive tumor growth suggests the presence of serum blocking factors. In addition a substantial rebound in LAI reactivity observed in the PBL of dogs 3 to 7 days following surgical excision of progressively growing CTVS provides further evidence that tumor-cell components are associated with progressive tumor growth and diminished LAI reactivity.

Inhibition of leukocyte adherence by a 3 M potassium chloride extracts of human malignant melanoma

The tube leukocyte adherence inhibition (LAI) assay was performed using 3 M KCl extracts of malignant melanoma materials and leukocytes taken from 24 malignant melanoma patients. Of these patients, 21 participated in sequential studies which employed a minimum of 3 time points. An increasing non-adherence index (NAI) was found in 5 of 6 patients who exhibited a favorable disease course. In 5 of 6 patients with an unfavorable disease course, NAI values decreased. In 8 of 9 patients with stable disease, the NAI also was stable. These results indicate that the trend of sequential LAI studies reflects the clinical course of malignant melanoma patients.

A coded study of antitumor immunity to human lung cancer assayed by tube leukocyte adherence inhibition

The present study was undertaken to evaluate the specificity of antitumor immunity to human lung cancer, measured by an in vitro assay--tube leukocyte adherence inhibition (LAI). We standardized and monitored the putative tumor antigen activity of the extracts by testing leukocytes from controls and patients with lung cancer in the Montreal General Hospital. A specific antitumor response to a lung cancer antigen was detected with coded leukocytes from 56% (20 out of 36) of patients with epidermoid lung cancer. By contrast, 4% (2 out of 53) of patients with inflammatory lung disease and none of 46 other patients with cancer metastatic to the lung or with other diagnoses had an LAI-positive result. The LAI response was inversely related to the extent of cancer: 80% (8 of 10) with Stage I, 66% (2 of 3) with Stage II, 54% (6 of 11) with localized Stage III, and 33% (4 of 12) with widespread Stage III were LAI positive. Leukocytes from patients with epidermoid, adenocarcinoma, or small cell lung cancer reacted to a common tumor antigen shared by extracts of epidermoid and small cell lung cancer. This study with coded samples from a remote hospital confirms the results of other investigators that the LAI measures an antitumor immune response to human organ-specific neoantigens.

Canine transmissible venereal sarcoma: leukocyte adherence inhibition (LAI) reactivity of various lymphoid tissues of dogs with tumors at different stages of growth

Leukocyte adherence inhibition (LAI) reactivity of various lymphoid tissues of dogs with canine transmissible venereal sarcoma (CTVS) at different stages of growth was determined by the tube LAI test. Tumors were classified at the time of excision into progressor, steady state, and regressor stages of growth. The LAI reactivity to CTVS antigen extract of spleen, draining and non-draining lymph-node cells, and peripheral blood leukocytes of regressors (non-adherence index--NAI of 172.8 +/- 46.8, 148.1 +/- 64.7, 138.7 +/- 47.3, and 172.2 +/- 60.7, respectively) was significantly greater than that of progressors (46.1 +/- 20.0, 38.5 +/- 21.5, 50.2 +/- 30.0, 24,6 +/- 37.2, respectively, p less than 0.001) and normal dogs (47.5 +/- 22.8, 54.6 +/- 24.6, 26.7 +/- 14.0, 50.9 +/- 22.4, respectively, p less than 0.001). In contrast, LAI reactivity of progressor lymphoid tissues to CTVS antigen extract did not differ significantly from that of normal dogs. LAI reactivity of lymphoid tissues from steady state tumor bearers (97.9 +/- 39.2, 80.7 +/- 47.3, 87.1 +/- 40.0, 85.1 +/- 53,9, respectively) was intermediate between and significantly different from LAI reactivities of regressor (p less than 0.05) and progressor (p less than 0.01) lymphoid tissues. Significant LAI reactivity observed in regressors suggests the presence of a functional effector cell mechanism associated with spontaneous regression of CTVS. The three distinct patterns of LAI reactivity observed in tumor-bearing dogs appear to correlate with the clinical course of tumor growth.

Partial purification of organ-specific neoantigens from human colon and breast cancer by affinity chromatography with human tumour-specific gamma-globulin

Organ-specific neoantigens (TA) shed from the tumours of patients with metastatic breast or colon cancer and which had filtered into the urine were partially purified by a combination of physicochemical methods and affinity chromatography. TA activity of the isolated materials was monitored by the blocking Tube LAI assay. Urinary protein was precipitated by 80% saturated ammonium sulphate. Albumin was removed by affinity chromatography with blue Sepharose CL-6B. Affinity columns of human IgG were prepared from sera of patients whose leucocytes were LAI+ to the breast- or colon-cancer extracts. The anti-breast-TA affinity column bound the TA in the urine of patients with metastatic breast cancer but not that of patients with metastatic colon cancer. The TA in urine of patients with metastatic colon cancer was bound by the anti-colon-TA affinity column. Analysis by SDS PAGE revealed that the isolates with and without TA activity were composed mostly of urinary protein which had bound nonspecifically to the human IgG affinity columns. With an affinity column of anti-NHS and Protein A, some of the contaminants were removed, to reveal SDD PAGE unique bands at about 38,000 and 12,000 mol. wt in the isolate with breast-TA activity. Rabbit antisera, raised to the material that had bound nonspecically to the anti-breast-TA affinity column, were used as an anti-nonspecific affinity column to remove the contaminants in the isolates from the affinity columns of anti-breast TA and anti-colon TA. After passage through the anti-nonspecific affinity column, the material that contained the putative breast or colon cancer TA revealed a unique band at about 38,000-40,000 mol. wt and residual fine bands at about 25,000-30,000 mol. wt. Both the control material and material with TA activity had similar bands at about 25,000 and 50,000 mol. wt. The specific activity of the putative colon or breast TAs, as measured by the blocking Tube LAI assay, was increased from about 30 to 5000-10,000 u/mg, a 125-400-fold enrichment.

Tube leukocyte adherence inhibition (LAI) assay in gastrointestinal (GIT) cancer

Tumor-specific immunity to carcinoma of the colon, pancreas and stomach was assayed by tube LAI. Cancers of the colon, pancreas and stomach, were shown to possess organ-type specific neoantigens. In 115 patients with colon cancer, 100%, 75%, 61% with Dukes' A, B and C cancer were LAI positive, respectively. Even a microfocus of in situ cancer in a colon adenoma was sufficient to stimulate measurable tumor-specific immunity in the host. In Dukes' D cancer, 25% of patients with widespread metastasis were positive, whereas 100% with solitary lesions were positive. Reactive leukocytes from patients with colon cancer did not react to extracts of normal bowel mucosa or villous adenoma from LAI-negative patients. Leukocytes from 19% (3 of 16) of patients with colon adenomas reacted to the extract of colon cancer but not normal colon mucosa. Moreover, the LAI-positive response of the patients with colon adenomas or colon cancer is directed to a colon cancer TSA which is linked to beta2-microglobulin. These studies suggest that some colon adenomas express TSA before morphological evidence of cancer. It is not known if the acquisition of a cell surface TSA is an irreversible step toward unrestrained growth and metastasis. In pancreatic cancer, 100% of patients with cancers less than 5 cm and without metastasis were LAI positive, whereas 29% were positive when the cancer was greater than 5 cm or had metastasized. In Patients with stomach cancer, 100% with Stage II and 46% with Stage III and IV cancer were LAI-positive. Leukocytes from patients with other GIT cancers and from patients with inflammatory bowel disease or pancreatitis did not react with extracts of colon, stomach or pancreatic cancer. Leukocytes from patients with metastatic cancer, usually did not react in the tube LAI assay because their surfaces were coated in vivo with TSA. LAI reactivity was present when CEA was not detectable and when CEA levels were elevated LAI activity was often absent. The present study suggests that the automated tube LAI shows sufficient promise to warrant studies to determine its efficacy for the diagnosis of GIT cancers.

A study of false positive and negative responses in the tube leucocyte adherence inhibition (tube LAI) assay

A panel of 5 different breast-cancer and 2 other cancer extracts was used to clarify the false-negative responses in patients with Stage I and II breast cancer and the false-positive responses in control subjects. Most patients with Stage I and II breast cancer who had an initially negative LAI response were positive when tested against the panel. The false negatives occurred because of (1) the experimental errors of the assay; (2) changes in the antigenic strength of the extracts; (3) antigenic heterogeneity of a few tumours and (4) lack of tumour-specific reactivity of the host. 3% of control subjects had a false-positive LAI response. The leucocytes from most of these positive patients did not react to the panel of antigens, and hence the false positives appeared to result from experimental error. In-hospital patients with benign breast disease had a 12% positivity rate when initially assayed, and 63% of these patients reacted to the panel of breast-cancer antigens. Those patients with benign breast disease who reacted to the panel of breast-cancer antigens had cytophilic anti-breast-cancer antibody in their serum; their leucocyte LAI reactivity was blocked in an immunologically specific manner by serum from advanced Stage IV breast-cancer patients; their leucocytes reacted to extracts of breast cancer and not fibrocystic breast tissue; their leucocyte reactivity was blocked by isolated breast-cancer TSA that was linked to beta 2 microglobulin, but not by normal breast-tissue proteins; and the kinetics of the LAI response after excision of the breast mass was identical to that observed with breast-cancer patients after mastectomy. In these patients, the breast tissue within the breast lump expressed breast TSA similar to unequivocal breast cancer.