G. R. Giles

Lymphokine activated killer (LAK) cells in patients with gastrointestinal cancer.

Lymphokine activated killer (LAK) cells are a recently described cellular immune phenomenon with exciting potential for the treatment of tumours arising from solid organs. A comparison of some aspects of LAK cell precursors and LAK cell function was undertaken in 44 control subjects and 44 preoperative patients suffering from gastrointestinal cancer (20 localised and 24 advanced). Lymphokine activated killer cell precursor (natural killer (NK) cell) activity was significantly diminished in patients with advanced tumours (p less than 0.02) as was fully mature LAK cell activity against an NK resistant target cell (p less than 0.012). T-lymphocyte responses were not significantly different between the three groups. The reduced LAK cell generation was associated with a significantly diminished proliferative response of LAK precursors to stimulation with high dose IL-2 in vitro (p less than 0.012). Impaired LAK cell generation may explain the failure of adoptive cellular immunotherapy with LAK cells in some patients with advanced gastrointestinal cancer and prompts the search for means of augmenting this activity in such patients.

Free radical detoxifying systems in human colorectal cancer

In aerobically metabolizing cells most oxygen undergoes tetravalent reduction to water by efficient intracellular mechanisms, principally the cytochrome system. A small amount, however, is metabolized by univalent reduction and associated alternative pathways which produce highly reactive species such as superoxide (02-j) that may undergo secondary reactions leading to H202 hydrogen peroxide and hydroxyl (OH-) free radicals. Accumulation of these metabolites may result in damage to intra and extracellular structures and they have been implicated in a number of disease processes (Buckley, 1983). All normal cells are protected from such damage by antioxidant systems which detoxify the reactive substances. They include the enzymes superoxide dismutase, catalase and glutathione peroxidase, and scavenging agents, such as the tocopherols, ascorbic acid and reduced glutathione. A number of studies of malignant cell lines and tumour biopsies have demonstrated abnormalities in the detoxifying pathways in particular a reduction of the activity of the superoxide dismutases which would result in increased amounts of potentially toxic, oxygen-derived free radicals (Marklund et al., 1982; Oberley & Beuttner, 1979). It has been proposed that alterations to intracellular structures and metabolic pathways caused by such a build-up of these substances may account for some of the properties of malignant cells and may even be involved in the process of malignant transformation (Oberley & Beuttner, 1979). Studies of animal tumours, both in vivo and from cell culture, from virally transformed, chemically induced and spontaneous tumours, have shown abnormalities in the levels of both the copper/zinc(Cu/Zn-SOD) and manganese (MnSOD) containing superoxide dismutases (Oberley & Beuttner, 1979). Human solid tumours have not been investigated to the same extent and reported levels of these enzymes are not consistently abnormal (Oberley & Beuttner, 1979; Westman & Marklund, 1981). There is little information about these enzymes in groups of histologically similar human tumours or about the other antioxidant enzyme systems. The aim of this study was to determine whether abnormalities in Cu/Zn-SOD, Mn-SOD and other detoxifying enzymes are present in human colorectal adenocarcinomas. We measured activities of superoxide dismutase and its two components Cu/Zn-SOD and Mn-SOD as well as two other antioxidant enzymes, catalase (CAT) and glutathione peroxidase (GPX), in tumours and ostensibly normal mucosa from 23 patients with colonic or rectal cancer. In addition, from the same specimens, levels of thiobarbituric (TBA) reactive compounds were measured as a possible indicator of lipid peroxidation and thus of the extent of free radical damage to cells. Operative specimens from patients undergoing colonic resection for large bowel tumours were opened in theatre and tissue samples, -1 g in weight, were taken from the viable growing edge of the tumour and from normal colonic mucosa at least 10 cm away from the tumour. These were washed and placed in 5 ml PBS at pH 7.4 and frozen at 20°C for later assay. After thawing they were homogenised on ice by ultrasound. Protein content of the homogenate was estimated by the method of Lowry et al. (1951) using human albumin as a standard. Enzyme assays were made at 37°C using a Pye Unicam SP8000 recording spectrophotometer. Superoxide dismutase activity was estimated by the method described by Crapo et al. (1978) at pH 7.8. This method uses xanthine and xanthine oxidase to generate superoxide anion at a constant rate, and cytochrome c as an indicator with which superoxide dismutase can compete. Cytochrome oxidase and peroxidase can cause problems with this assay which can readily be overcome by adding potassium cyanide (10um) and catalase. However, this was found not to be necessary in our samples.

Cisplatin‐resistant metastatic hepatoblastoma: Complete response to carboplatin, etoposide, and liver transplantation

A child with stage 4 hepatoblastoma failed to respond to treatment with cisplatin and adriamycin. She then showed a response to carboplatin with complete clearing of pulmonary metastases. Bilobar liver disease persisted, although significantly reduced in size. A liver transplant was subsequently performed and she remains in complete remission 36 months later. After the first course of carboplatin, there was a dramatic rise in alpha-feto protein which then fell exponentially. Carboplatin warrants further study in hepatoblastoma.

Pyrimidine nucleoside phosphorylase activity in tumour and matched normal gastrointestinal mucosa.

5-Fluorouracil (5-FU) requires activation to the metabolites FdUMP and FUTP for its cytotoxic effect. This activation involves the intracellular enzymes uridine and thymidine phosphorylase. We have assayed the levels of these enzymes in colorectal and gastric cancers and have shown that, in the majority of cases, the enzyme levels were higher than in the adjacent normal mucosa. It is suggested that, while the ratio tumour/normal mucosa enzyme activity may give some indication of the relative toxicity of 5-FU in these tissues, the absolute activities of the phosphorylases in tumour tissue could give a better indication of tumour responsiveness.

Evaluation of an enzyme-linked immunosorbent Raji cell assay (ELISA) in the investigation of gastrointestinal cancer.

SummaryThe levels of circulating immune complexes (CICs) have been estimated in a group of patients with colorectal cancer and gastric cancer, in addition to which a normal range has been established in a group of patients with benign gastrointestinal disease. A newly developed enzyme-linked immunosorbent Raji cell assay has been used in this study. Overall only 30% of patients with gastrointestinal cancer showed elevation of CIC levels outside the normal range. Elevated levels correlated with tumour differentiation bud did not correlate with site of disease or with the presence of metastases. In an attempt to define the specificity of CIC estimation, soluble tumour extract was added to sera from tumour-bearing patients. Specific IC elevations were produced by addition of allogeneic tumour extract of colon cancer in patients with colorectal cancer; this phenomenon was not seen when the same extract was added to the sera of patients with gastric cancer.

Incorporation of intermediary products of 5-FU anabolism into colorectal cancer.

In 18 patients intermediary anabolic metabolites of 5-FU were measured in normal colonic mucosa and colorectal cancer tissue of intravenous bolus injection or continuous infusion. Higher total concentrations of 5-FU products were found in the cancers when compared with normal colonic tissue. There appeared to be no evidence, however, that some patients had a selectively increased ability to incorporate 5-FU into their tumours. Overall higher concentrations of tumour incorporation of 5-FU were found after bolus injection rather than the infusion method, although this difference is statistically not significant.