Phase I and pharmacokinetic study of the orally administered farnesyl transferase inhibitor R115777 in patients with advanced solid tumors.

R115777 is a novel selective inhibitor of farnesyl transferase, an enzyme that is involved in the proliferation of the malignant cell type. This study was designed to determine the toxicity, maximal tolerated dose and pharmacokinetics of R115777 when given orally b.i.d. for 28 days followed by 1-2 weeks of rest. Patients with advanced solid tumors for whom no standard therapy was available could enter the study. The starting dose of R115777 was 200 mg/dose and inter- as well as intra-patient dose escalations were performed with increments of 100 mg/dose. Nine patients entered the study and received in total 23 treatment cycles. A dose of 300 mg b.i.d. proved feasible with grade 4 neutropenia occurring in one of six patients who completed the first treatment cycle. Other toxicities were infrequent. Pharmacokinetic analysis demonstrated that peak plasma concentrations of 881+/-393 ng/ml were reached within 1-5 h. No accumulation of R115777 was observed over a 28-day period. The study was terminated based on these results together with the observation from a related phase I study in which higher doses of R115777 were associated with the frequent occurrence of grade 3-4 myelosuppression. We conclude that the recommended dose of R115777 given for 28 days followed by 1-2 weeks of rest is 300 mg b.i.d. Myelosuppression is the dose-limiting toxicity.

Itraconazole oral solution for primary prophylaxis of fungal infections in patients with hematological malignancy and profound neutropenia: a randomized, double-blind, double-placebo, multicenter trial comparing itraconazole and amphotericin B.

Systemic and superficial fungal infections are a major problem among immunocompromised patients with hematological malignancy. A double-blind, double-placebo, randomized, multicenter trial was performed to compare the efficacy and safety of itraconazole oral solution (2.5 mg/kg of body weight twice a day) with amphotericin B capsules (500 mg orally four times a day) for prophylaxis of systemic and superficial fungal infection. Prophylactic treatment was initiated on the first day of chemotherapy and was continued until the end of the neutropenic period (>0.5 x 10(9) neutrophils/liter) or up to a maximum of 3 days following the end of neutropenia, unless a systemic fungal infection was documented or suspected. The maximum treatment duration was 56 days. In the intent-to-treat population, invasive aspergillosis was noted in 5 (1.8%) of the 281 patients assigned to itraconazole oral solution and in 9 (3.3%) of the 276 patients assigned to oral amphotericin B; of these, 1 and 4 patients died, respectively. Proven systemic fungal infection (including invasive aspergillosis) occurred in 8 patients (2.8%) who received itraconazole, compared with 13 (4.7%) who received oral amphotericin B. Itraconazole significantly reduced the incidence of superficial fungal infections as compared to oral amphotericin B (2 [1%] versus 13 [5%]; P = 0.004). Although the incidences of suspected fungal infection (including fever of unknown origin) were not different between the groups, fewer patients were administered intravenous systemic antifungals (mainly intravenous amphotericin B) in the group receiving itraconazole than in the group receiving oral amphotericin B (114 [41%] versus 132 [48%]; P = 0.066). Adequate plasma itraconazole levels were achieved in about 80% of the patients from 1 week after the start of treatment. In both groups, the trial medication was safe and well tolerated. Prophylactic administration of itraconazole oral solution significantly reduces superficial fungal infection in patients with hematological malignancies and neutropenia. The incidence of proven systemic fungal infections, the number of deaths due to deep fungal infections, and the use of systemic antifungals tended to be lower in the itraconazole-treated group than in the amphotericin B-treated group, without statistical significance. Itraconazole oral solution is a broad-spectrum systemic antifungal agent with prophylactic activity in neutropenic patients, especially for those at high risk of prolonged neutropenia.

Late effects of intraoperative radiation therapy in anastomotic rat colon.

PURPOSE: to determine whether intraoperative radiotherapy causes long-term negative effects on the healing of colonic anastomoses in the rat. METHODS AND MATERIALS: 175 rats were divided into seven equal groups. One group served as sham-irradiated control group. In the others, following a colonic resection, 1 or 2 cm of the distal bowel limb was irradiated with a single dose of 10, 15, or 20 Gy (groups 10/1, 15/1, 20/1, 10/2, 15/2, and 20/2, respectively). Subsequently, an anastomosis was constructed. The animals were killed after 6 (n = 10 in each group) or 12 (n = 15) months. The abdomen was inspected for abnormalities and the colonic diameter was measured. The anastomotic segment was analyzed biochemically (hydroxyproline) and histologically. RESULTS: During the experimental period, 1 rat (group 15/1) died because of anastomotic leakage and 3 others died from unknown causes. There was no difference in colonic diameter between groups. Altogether 17 rats developed an adenocarcinoma in the irradiated area: 11 of these had received a dose of 20 Gy. Histological observation indicated that fibrosis was present only in a limited number of animals, mostly after irradiation with a dose of 15 or 20 Gy. All anastomoses were functional and showed normal histology. The hydroxyproline content of the anastomotic segment was increased--with respect to the control group--only in the 20/2 group after 6 months. After 12 months, the hydroxyproline concentration in the (irradiated) segment distal to the anastomosis proper was higher in the 10/1 and 15/1 groups than in the control group. Otherwise, there were no differences between groups. CONCLUSION: Intraoperative irradiation with a single dose of 10-20 Gy, delivered to the distal limb used for anastomotic construction, does not appear to constitute a threat to anastomotic integrity. Dose-related changes included formation of adenocarcinomas and fibrosis, but function and histology of the anastomosis proper remained unaffected.

Moderate doses of intraoperative radiation severely suppress early strength of anastomoses in the rat colon.

Intraoperative irradiation appears to be a valuable addition to the modalities available to treat patients with large bowel cancer. However, its potential effect on healing of anastomoses has not been investigated extensively. For this purpose, male Wistar rats underwent colonic resection. Subsequently, 1 cm of each bowel end was irradiated with doses of 10, 15, 20 or 25 Gy and intestinal continuity was restored. After 3 or 7 days, animals were killed and the anastomoses were analyzed for bursting pressure (intraluminal force), breaking strength (longitudinal force) and hydroxyproline content. Intraoperative irradiation led to a massive (40-70%) and significant (P < 0.025) reduction in bursting pressure 3 days after operation compared to the control group for every dose used. After 7 days, the bursting site was outside the area of the anastomosis in all groups. The breaking strength at day 3 was also reduced, even after 10 Gy. At day 7, when tearing still occurred in the wound area, the breaking strength was still significantly lower in the 15- and 25-Gy groups than in the control group. The hydroxyproline content of the anastomoses was significantly reduced only after irradiation with the higher doses. Thus intraoperative irradiation constitutes a threat to early strength of anastomoses in the rat colon, and even at moderate doses it may threaten the integrity of the anastomosis.

A semiquantitative histological analysis of repair of anastomoses in the rat colon after combined preoperative irradiation and local hyperthermia.

Hyperthermia is a promising method for increasing the efficacy of radiation therapy of colorectal cancer. To study the histological aspects of healing of an anastomosis in the colon, after combined preoperative (sham) irradiation and (sham) hyperthermia treatment, 48 male Wistar rats were divided randomly into four groups. In each animal, a segment of the colon was treated successively by (sham) irradiation (single dose of 25 Gy X rays) and/or (sham) hyperthermia (44 degrees C, 30 min). After 5 days, a resection of the colon was performed by construction of an anastomosis: The distal limb consisted of (sham-) irradiated and/or (sham-) hyperthermia-treated bowel. Rats were killed 3 or 7 days after the surgical procedure. Evaluation of healing of the anastomosis was made by: (1) histological analysis of sections stained with hematoxylin and eosin, (2) semiquantitative measurement of collagen in the area of the anastomosis and (3) semiquantitative analysis of the number of macrophages by immunocytochemistry. Healing of the anastomoses in animals receiving irradiation or hyperthermia alone and in control animals was relatively uneventful. There were no differences between groups in formation of collagen or infiltration by macrophages in the area of the anastomosis. Animals treated with both radiation and hyperthermia showed marked necrosis, infiltration by polymorphonuclear leukocytes and rupture of the anastomosis. It is concluded that preoperative irradiation with a single dose of 25 Gy in combination with local hyperthermia at 44 degrees C for 30 min leads to disturbed repair of anastomoses.

Intra-operative irradiation prolongs the presence of matrix metalloproteinase activity in large bowel anastomoses of the rat.

There exists a growing interest in intra-operative radiation therapy as a treatment modality for large bowel cancer. In a previous experimental study we showed that high-dose intra-operative irradiation delays the healing of colonic anastomoses. However, the contribution of proteases is unknown. In the present study, the gelatinolytic and collagenolytic activity in the healing anastomoses is investigated. After a resection of a 1-cm length of colon (uninjured colon), the rats were irradiated with a single dose of 25 Gy, either to the proximal limb, referred to as the proximal group, or to both proximal and distal limbs of the bowel, referred to as the combined group, before anastomotic construction. Both groups were compared to a control group with anastomoses which were sham-irradiated. The animals were killed 1, 3 or 7 days after operation. The gelatinolytic activity in uninjured and anastomotic tissue was quantified by gelatin zymography and the collagenolytic activity by an assay using a fibrillar rat collagen substrate. Compared with resected uninjured colon, most of the gelatinolytic activities were markedly increased in anastomotic tissue of all groups during the first postoperative week, and new additional activities were detected. The additional metalloproteinases (the 95-kDa family) of both irradiated groups were significantly elevated compared to the anastomoses of the sham-irradiated control group at 7 days after operation. In anastomotic tissue of all groups, the collagenolytic activity of the tissue was also significantly increased at 1 and 3 days after construction with respect to the resected, uninjured colon. After 7 days this effect had disappeared for the sham-irradiated anastomoses, but the activity in the anastomoses in both the proximal and combined groups remained significantly elevated. The findings provide evidence that intra-operative irradiation prolongs the presence of elevated gelatinolytic and collagenolytic activities in colon anastomoses. It may contribute to a reduced or delayed accumulation of collagen and other matrix proteins that supply anastomotic strength.

Quantitation of angiogenesis in healing anastomoses of the rat colon.

An accurate and reliable quantitation of angiogenesis is an essential requirement for a detailed study of new blood vessel growth during healing of colon anastomoses. In the present study we applied a computer-based digital image processing system for quantitative analysis of the anastomotic vascularization and perfusion. Rats underwent colonic resection (1 cm) followed by construction of an end-to-end anastomosis. The animals were killed 3 or 7 days after operation. The vascularization and perfusion were analysed in cyrostat sections using an anti-collagen type IV antibody and the fluorescent marker bisbenzimide H 33342, respectively. If compared with 3-day-old anastomoses, a significant increase in the median vascular and perfusion areas was found 7 days after operation. Within the same anastomosis, the vessel density and perfusion were lower in the central region than in the peripheral region of the wound. The present computerized digital image analysis system is a reliable technique that is suited for quantitative measurements of microvascular changes occurring in colon anastomoses during the first postoperative week.

Divergent patterns of matrix metalloproteinase activity during wound healing in ileum and colon of rats.

BACKGROUND: Uncontrolled and increased extracellular matrix degradation during early anastomotic repair in the intestine may reduce wound strength increasing the risk of anastomotic dehiscence. AIMS: To characterise the metalloproteinases present in intact and anastomosed ileum and colon to study their role in matrix degradation after surgery. SUBJECTS: Tissue extracts of uninjured, and of anastomosed rat ileum and colon at postoperative days 1, 2, 3, 7, and 90, were used. METHODS: Metalloproteinases were identified by gelatin and casein zymography. Aminophenylmercuric acetate (APMA) treatment was used to activate latent metalloproteinases. RESULTS: Both uninjured ileum and colon contained a 60 and 67 kDa activity, but a 54 and 72 kDa gelatinase was present in ileum only, and a 51 kDa activity in colon only. APMA treatment converted the 60 kDa protease to 54 and 51 kDa forms and the 72 kDa protease to the 67 kDa form. These gelatinases may correspond to latent and active forms of MMP 1 and MMP 2, respectively. Additional metalloproteinases were observed after anastomotic construction. Both ileum and colon contained 95 and 230 kDa gelatinases, which were converted to 83 and 76 kDa forms by APMA. They may be the latent and active forms of MMP 9, respectively. Gelatinolytic activities of 25 and 28 kDa were only found in anastomosed ileum. Caseinolytic activities were only found in ileum extracts and those were most prominent at day 1, 2, and 3 after surgery. CONCLUSIONS: The metalloproteinase pattern in ileum and colon differ considerably suggesting that matrix degradation after anastomotic construction may also vary.

Intraoperative irradiation delays anastomotic repair in rat colon.

BACKGROUND: There exists a growing interest in intraoperative radiation therapy as a treatment modality for large-bowel cancer. Since such therapy could interfere with wound repair, we investigated its effects on early healing of colonic anastomoses. METHODS: After resection of 1 cm of colon, rats were irradiated with a single dose of 25 Gy, either to the proximal limb (P group) or to both proximal and distal limbs of the bowel (PD group) before anastomotic construction. Both groups were compared with a sham-irradiated control group. Animals were killed 3, 7, or 14 days after operation, and healing was assessed by mechanical and biochemical (collagen) parameters. RESULTS: Three days after operation, bursting pressure was significantly lowered in the P group, whereas in the PD group both bursting pressure and breaking strength were strongly reduced. At day 7, the breaking strength was still reduced in the PD group, but not significantly so in the P group. The collagen synthetic capacity of the anastomotic segments was significantly lowered in both irradiated groups at day 3, resulting in a diminished collagen concentration in the actual wound area after 7 days. At 14 days after operation, no differences in strength were found between control and irradiated groups, while anastomotic hydroxyproline levels were significantly higher in both the P and PD groups than in the control group. CONCLUSIONS: High-dose intraoperative radiation therapy delays the healing of colonic anastomoses; it transiently reduces strength, probably as a result of a diminished accumulation of collagen.

Beta-carotene (provitamin A) decreases the severity of CCl4-induced hepatic inflammation and fibrosis in rats.

Earlier data from experiments in rats have shown that administration of retinyl esters (vitamin A) strongly influences the effects of CCl4 on the liver. The accumulation of collagen was inhibited, but an increase in CCl4-toxicity with high mortality was observed. The present study was conducted to examine the effects of beta-carotene (provitamin A) on CCl4-related general and hepatic toxicity in rats. Oral administration of beta-carotene during CCl4-treatment resulted, biochemically, in a significantly lower increase in the hydroxyproline liver content and, histopathologically, in less severe liver fibrosis as compared with the liver of rats not treated with beta-carotene. The study also showed that beta-carotene administration could prevent the long-term loss of retinoids from the CCl4-injured liver. No significant toxic effects of beta-carotene, as previously found with retinyl esters (vitamin A), were observed. This experimental study suggests that beta-carotene has the therapeutic potential to decrease the severity of liver fibrosis without marked toxicity.

Role of vitamin A in liver fibrosis.

The relationship between vitamin A and liver fibrosis was studied with a CCl4-induced fibrosis model in rats. Depending on the time of administration, vitamin A can potentiate or reduce fibrosis: when present during CCl4-treatment parenchymal cell damage and fibrosis were enhanced, whereas vitamin A post-treatment strongly reduced fibrosis. Enhanced fibrosis was also found in rats with low hepatic retinoid levels. Administration of beta-carotene during CCl4-treatment reduced several signs of fibrosis. The notion that liver retinoids play an important role in hepatic fibrogenesis is discussed.

Alcohol in combination with malnutrition causes increased liver fibrosis in rats.

Rats were malnourished for 12 months with a highly inadequate fat-rich, calorie-sufficient but otherwise poly-deficient liquid diet composed of mashed potatoes with mayonnaise, comparable with the nutritional intake of many chronic alcoholics. When alcohol was incorporated into this diet, administered as whisky in drinking water available ad libitum, the livers of all eight rats showed increased fibrosis and cirrhosis as compared to the livers of the eight non-alcohol-treated, isocalorically fed, paired control rats. Alcohol-treated rats developed fibrosis and cirrhosis on a dietary fat content of 38% of total caloric intake and low blood alcohol levels, ranging from 50 to 126 mg/dl, due to gradual intake over the day and to low absolute intake (mean 11.9 +/- 0.6 g/kg per day). None of the rats died spontaneously. Malnutrition is likely to be an important factor in the development of the fibrosis of alcoholic liver disease, and this rat model may be used to study aspects of the pathogenesis.

Fat-storing cells and myofibroblasts are involved in the initial phase of carbon tetrachloride-induced hepatic fibrosis in BN/BiRij rats.

This study on the appearance, distribution and kinetics of fibroblast-like cells (fat-storing cells, transitional cells, myofibroblasts and fibroblasts) after CCl4-treatment was undertaken to delineate further the respective roles of these cell types in liver fibrogenesis. The different cell types were distinguished on the basis of their immunophenotypic pattern with a combination of marker antibodies and on the basis of ultrastructural characteristics. Combined staining for alpha-smooth muscle actin (sma) and desmin (Des) revealed perisinusoidal fat-storing cells (FSC) as d+ sma- and myofibroblasts around the central veins of the normal rat liver as d+ sma+. During the initial phase of CCl4-induced hepatic fibrosis (week 1 and 2), the number of d+ sma+ cells increased in the degenerating area around the central veins and d+ sma+ cells appeared in the very thin fibrotic septa at week 2. Ultrastructural examination of the affected central areas showed the presence of myofibroblasts. These sma+ cells proliferated, as shown by double staining for bromodeoxyuridine (BrdU) and sma. In degenerating parenchymal areas, d+ sma- FSC were present. The FSC in the perisinusoidal space of areas which were not affected by CCl4 intoxification, remained d+ sma-. These immunostaining findings support the electron microscopical results, which show the presence of cells with the typical ultrastructural characteristics of FSC in both the degenerating areas and the perisinusoidal space of unaffected areas. After one week of CCl4-treatment, enhanced deposition of procollagen type III was observed around the central veins. Enhanced deposition of collagen type IV was seen subendothelially along the sinusoids, notably in degenerating parenchymal areas where the septa were later formed. FSC appear to be the principal source of collagen type IV during fibrogenesis. These observations further support and specify the role of FSC in early fibrogenesis. With the progression of the CCl4-induced fibrosis, d+ sma+ myofibroblasts remained localized in the fibrotic septa, but now along their outer edge. The majority of the cells in the septa were formed by d- sma- cells indicating a prominent role of fibroblasts in the septal formation. Septal fibroblasts are not only likely to produce matrix components, but also were shown to degrade collagen, as evidenced by the increased number of collagen-containing vacuoles during the course of fibrosis. In conclusion, myofibroblasts and FSC appear to be the main cell types involved in the initial phase of liver fibrogenesis induced by CCl4. Both myofibroblasts and FSC divide and transform.(ABSTRACT TRUNCATED AT 400 WORDS)

Vitamin A deficiency potentiates carbon tetrachloride-induced liver fibrosis in rats.

Earlier studies have shown that retinoid administration suppresses the generation of hepatic fibrosis and stimulates its regression in normal (i.e., vitamin A-sufficient) carbon tetrachloride-treated rats. This study focuses on the possible role of a marginal or deficient vitamin A status on carbon tetrachloride-induced fibrosis. This experimental study in rats shows that vitamin A status, reflected by hepatic retinoid content (retinol and retinyl esters), modulates the development of hepatic fibrosis induced by carbon tetrachloride. In rats with low hepatic retinoid levels (12 +/- 0.9 micrograms/gm liver), carbon tetrachloride-induced liver fibrosis was more pronounced than in rats with sufficient hepatic retinoid levels (1,065 +/- 327 micrograms/gm liver). Enhanced liver fibrogenesis was confirmed both morphologically and by a higher hydroxyproline content of the liver. It was associated with a reduced liver weight and the development of parenchymal regeneration nodules. Furthermore, carbon tetrachloride treatment itself reduced the hepatic retinoid content in rats independently of the liver vitamin A status before treatment and increased serum retinol levels in vitamin A-sufficient rats. The results show that the vitamin A status of the liver plays an important role in hepatic fibrogenesis. Low hepatic vitamin A levels, which can be the result not only of low dietary intake but also of interference with vitamin A metabolism by agents such as ethanol and carbon tetrachloride, may be a risk factor for the development of liver fibrosis. We suggest that retinoids modulate collagen synthesis and deposition irrespective of the degree of hepatocellular necrosis induced by carbon tetrachloride.(ABSTRACT TRUNCATED AT 250 WORDS)

Chronic administration of ethanol with high vitamin A supplementation in a liquid diet to rats does not cause liver fibrosis. 1. Morphological observations.

Rats of two strains (BN/BiRij and WAG/Rij) were fed the ethanol-containing Lieber-De Carli liquid diet supplemented with high amounts of vitamin A for 16 months. In contrast to Lieber and co-workers, who showed liver fibrosis developing within 9 months on the same diet in Sprague-Dawley rats, we were unable to demonstrate a histological and biochemical increase in liver collagen in either strain. Steatosis was present to a varying degree in both strains in ethanol-treated rats, but also in control animals. Considerable liver inflammation with focal necrosis accompanied by severe systemic inflammation was observed in 60% of the ethanol-treated WAG rats. This suggests that, at least in rats, the main effects of chronic ethanol consumption on the liver may be secondary to interference with host resistance to infections. The ethanol-high vitamin A Lieber-De Carli liquid diet does not necessarily elicit fibrosis or other characteristic histological abnormalities of human alcoholic liver disease.

Chronic administration of ethanol with high vitamin A supplementation in a liquid diet to rats does not cause liver fibrosis. 2. Biochemical observations.

The inability of the 'ethanol/high vitamin A Lieber-DeCarli diet' to induce liver fibrosis in two different rat strains was further evaluated by determining changes in parameters of liver cell damage and of retinoid and lipid metabolism. In the ethanol/vitamin A-treated group, slight but constant hepatic cell damage, as indicated by elevated alanine aminotransferase, aspartate aminotransferase and glutamate dehydrogenase activities in blood, was already observed at 6 months and maintained until the time of death at 16 months. Serum gamma-glutamyl transaminase activities were not raised. Moderate parenchymal liver cell damage was not accompanied by fibrosis. Hypertriglyceridemia or hypercholesterolemia were observed at 6-16 months of chronic alcohol administration. This response was strain dependent. In ethanol-treated rats of both strains, total liver retinoids and serum retinol concentrations were not altered. Therefore, the hypothesis that interaction between alcohol and retinoids is a major factor in the pathogenesis of alcoholic liver disease, needs to be reconsidered.

Synergism between ethanol and carbon tetrachloride in the generation of liver fibrosis.

In this study, alcohol-induced histological lesions in a short-term experimental rat model were compared with those characteristic of human alcoholic liver disease. In the rat model used, pretreatment with carbon tetrachloride (CCl4) for 6 weeks was employed possibly to sensitize the liver for the effects of alcohol and shorten the time of induction of alcoholic liver disease. After 6 weeks of CCl4 treatment, subsequent maintenance on drinking water containing up to 10 per cent alcohol for 7 weeks potentiated liver fibroplasia as compared with non-alcohol-treated rats. However, steatosis and alcoholic hepatitis, as histological evidence for alcoholic liver disease as seen in humans, were not observed. In non-CCl4-pretreated control animals, alcohol administration had no effect on liver histology. It can be concluded that in the model used, CCl4 pretreatment sensitizes the liver to increase collagen deposition following alcohol administration, but not to steatosis or alcoholic hepatitis as seen in human alcoholic liver disease. In this experimental set-up, direct metabolic interaction of CCl4 with alcohol as a cause of the increased fibroplasia can be excluded.

Catecholamines and endogenous opioids in ventral tegmental self-stimulation reward.

Midbrain dopaminergic pathways and opioid receptor systems have been implicated in the reward experienced in electrical intracranial self-stimulation behavior. In the present experiment, the influence of graded doses of the dopamine antagonist haloperidol and of the agonist cocaine were investigated on electrical self-stimulation reward, elicited by electrodes located in the ventral tegmental area. A threshold method, which is rather insensitive for aspecific motor effects, was applied to determine the reward of self-stimulation. The method allowed to determine simultaneously the rate of lever pressing for self-stimulation. All doses of haloperidol and cocaine were administered with and without the opioid antagonist naloxone, in order to investigate the interaction between dopaminergic and opioid modulation of reward. Haloperidol lowered and cocaine tended to increase the response rate, whereas cocaine but also haloperidol lowered the self-stimulation threshold. The effects appear to be dose-dependent. Naloxone did not interact with the effect of the drugs on threshold and it lowered the response rate, but in the haloperidol-treated rats only. It is concluded that dopamine is involved in the reward of electrical self-stimulation elicited from the ventral tegmental area and that this involvement is independent of endorphin systems, suggesting the existence of separate catecholamine and opioid mechanisms modulating brain reward.

Spontaneous IgA synthesis by blood mononuclear cells in alcoholic liver disease.

Changes in the concentration and composition of serum immunoglobulin A (IgA) and deposits of IgA in tissues are well-known characteristics of alcoholic liver disease. We investigated whether these changes also accompany IgA synthesis by peripheral blood mononuclear cells (MNC), by counting immunoglobulin-producing cells using a solid-phase enzymatic 'spot' test, and by analysis of immunoglobulin content in lysed cells with culture supernatant using conventional enzymatic methods. Patients with alcoholic liver disease exhibited a significantly higher number of spontaneously IgA-producing cells than did normal healthy controls (1.7 X 10(6) cells/blood and 0.5 X 10(6) cells/blood, respectively, P less than 0.01). The IgA content of MNC directly after isolation was also higher (38 and 13 ng/10(6) MNC, respectively, P less than 0.01), as was the IgA production during an unstimulated 6-day culture period (520 and 95 ng/10(6) MNC put into culture, respectively, P less than 0.001). The spontaneously IgA-producing cells assessed directly after isolation of mononuclear cells correlated with the IgA production during an unstimulated culture (P less than 0.01). We conclude that in alcoholic liver disease, B lymphocytes circulate which may have been activated in vivo.