Natural killer-cell cytotoxicity in HIV-positive and HIV-negative patients with and without severe course of hepatitis B virus infection.

Natural killer (NK) cells represent the first line of defence against viral infections but, in the case of hepatitis B virus (HBV), may also be involved in liver injury. We here compared NK-cell activity of 11 patients with acute HBV infection, either HIV-positive or HIV-negative, with that of 11 healthy subjects. One of the HIV-positive patients, characterized by a severe immunodeficiency, died 3 weeks after hospitalization for HBV-related fulminant hepatitis (FH). He displayed a remarkable NK-cell cytotoxicity against both cell lines and autologous dendritic cells, whereas the NK-cell activity of the remaining patients was significantly reduced as compared with healthy individuals. Our findings suggest that NK-cell-mediated cytotoxicity could contribute to the development of HBV-related acute liver failure in HIV-positive patients with severe immunodeficiency. An immunopathological model of FH in immunocompromised patients was proposed.

Natural killer cell function in HIV-1 infected patients.

A cross-talk between dendritic cells (DC) and resting natural killer (NK) cells leads to the activation of both cell populations, a process requiring cell-cell contact. When the number of activated NK cells overwhelms surrounding DC, they became able to kill specifically immature DC, a feedback mechanism to shut off DC-mediated immune responses. DC, at the mucosal site, can capture HIV and transfer it to CD4+ T lymphocytes present in the regional lymph node thus giving rise to a productive infection; on the other hand, NK cells represent the first line of defence against viral infection. Our preliminary results suggest that during the early phases of an HIV infection, NK cell activity is not functionally compromised, but that infected cells might escape natural immune surveillance through several mechanisms, including a reduced lysis of autologous DC.

A prospective, randomized clinical trial on perioperative feeding with an arginine-, omega-3 fatty acid-, and RNA-enriched enteral diet: effect on host response and nutritional status.

BACKGROUND: The use of immune-enhancing enteral diets in the postoperative period has given contrasting results. The purpose of this prospective, randomized, double-blinded clinical study was to evaluate the effect of immunonutrition given perioperatively on cytokine release and nutritional parameters. METHODS: Patients with cancer of the stomach or colo-rectum were eligible. Subjects consumed 1 L/d of either a control enteral formula (n = 25; control group) or a formula supplemented with arginine, omega-3 fatty acids, and RNA (n = 25; verum group) for 1 week before surgery. Both formulas were given by mouth. Six hours after the operation, jejunal infusion with the same diets was started and maintained for 7 days. Blood was drawn at different time points to assess albumin, prealbumin (PA), transferrin, cholinesterase activity, retinol binding protein (RBP), interleukin-2 receptors alpha (IL-2Ralpha), IL-6, and IL-1 soluble receptors (IL-1RII). The composite score of delayed hypersensitivity response (DHR) to skin test also was determined (the higher the score, the lower the immune response). RESULTS: During the 7 days of presurgical feeding, none of the above parameters changed in either group. Eight days after operation, in the control group, the concentration of PA and RBP was lower than in the verum group (0.18 vs 0.26 g/L for PA and 30.5 vs 38.7 mg/L for RBP; p < .05). IL-2Ralpha concentration was 507 pg/mL in the verum group vs 238 pg/mL in the control group (p < .001), whereas IL-6 and IL-1RII were higher in the control group than in the verum group (104 vs 49 and 328 vs 183 pg/mL, respectively; p < .01). The DHR score was 0.68 in the control group vs 0.42 in the verum group (p < .05). CONCLUSIONS: Perioperative feeding with a supplemented enteral diet modulates cytokine production and enhances cell-mediated immunity and the synthesis of short half-life proteins.

Dual role of TNF-alpha in NK/LAK cell-mediated lysis of chronically HIV-infected U1 cells. Concomitant enhancement of HIV expression and sensitization of cell-mediated lysis.

The U937-derived chronically HIV-infected U1 cell line and uninfected U937 cell clones were efficiently lysed by both unstimulated (NK) and IL-2-stimulated (lymphokine-activated killer; LAK) peripheral blood mononuclear cells (PBMC) of healthy HIV-seronegative donors. Pretreatment of target cells with IFN-gamma down-modulated killing of both U1 cells and two U937 cell clones, and up-regulated MHC class I expression. In contrast, TNF-alpha enhanced the sensitivity of infected U1 cells, but not of U937 cell clones to NK / LAK cell lysis. Co-cultivation of IL-2-stimulated PBMC with U1 cells triggered expression and replication of HIV by cell-cell contact, and this effect was inhibited by anti-TNF-alpha antibodies (Ab); virus production was partially inhibited by zidovudine. Of interest, anti-TNF-alpha Ab protected U1 cells from LAK cell activity. Thus, TNF-alpha can induce HIV expression from chronically infected U1 cells, but also plays an important role in sensitizing these cells to lysis.

Radical oncologic surgery affects the circulatory levels of interleukin 10.

BACKGROUND AND OBJECTIVE: Interleukin 10 (IL-10) has been shown to be elevated in the plasma of cancer-bearing patients. The source of systemic IL-10 may be the tumor microenvironment. We therefore tried to evaluate if ablative surgery for gastrointestinal cancer could affect the levels of circulating IL-10. METHODS: Plasma IL-10 concentration was measured in 45 patients with adenocarcinoma of the gastrointestinal tract. Forty healthy subjects, 15 women undergoing hysterectomy for uterine fibroma, and 15 patients undergoing palliative operation for pancreatic cancer were used as control groups. Plasma IL-10 was assessed 1 day before surgery (baseline) and 1, 4, and 8 days after operation. RESULTS: The baseline concentration of IL-10 was significantly higher in cancer patients than in healthy subjects and in women with fibroma (8.6 ng/mL, 2.1 and 1.8 respectively; P = 0.015). After radical surgery, the IL-10 levels significantly dropped in cancer patients (from 8.6 ng/mL to 3.8; P = 0.024), whereas in subjects undergoing palliative operation, the concentration remained elevated (8.5 ng/mL baseline versus 7.9 on day + 1). CONCLUSIONS: The origin of circulating IL-10 may be the tumor microenvironment.

Plasma nitrate plus nitrite changes during continuous intravenous infusion interleukin 2.

Nitric oxide (NO), a biologically active mediator generated in many cell types by the enzyme NO synthase, may play an important role in cardiovascular toxicity that is frequently observed in cancer patients during intravenous (i.v.) interleukin 2 (IL-2) therapy. The induction of NO synthase and the production of NO seem to be involved in the pathogenesis of the vascular leakage syndrome, as well as in the regulation of myocardial contractility. In the present study, we evaluated the pattern of plasmatic NO changes during multiple cycles of continuous i.v. infusion (CIVI) of IL-2 in ten advanced cancer patients (five males, five females, median age 59 years, range 33-67 years; eight affected by renal cell cancer and two affected by malignant melanoma). The patients received IL-2 at 18 MIU m-2 day-1 (14 cycles) or 9 MIU m-2 day-1 (seven cycles) for 96 h, repeated every 3 weeks. Interferon alpha (IFN alpha) was also administered subcutaneously (s.c) during the 3 week interval between IL-2 cycles. For each cycle, plasma samples were collected before treatment (t0), 24 h (t1), 48 h (t2), 72 h (t3) and 96 h (t4) after the start of IL-2 infusion, and 24 h after the end of the cycle. NO concentration was determined spectrophotometrically by measuring the accumulation of both nitrite and nitrate (after reduction to nitrite). The following observations may be drawn from data analysis: (1) plasma nitrate + nitrite significantly raised during treatment (P = 0.0226 for t0 vs t3), but statistical significance was retained only when cycles administered with IL-2 18 MIU m-2 day-1 are considered (P = 0.0329 for t0 vs t3; P = 0.0354 for t0 vs t2 vs t4) (dose-dependent pattern); (2) during subsequent cycles a significant trend toward a progressive increase of plasma nitrate + nitrite levels, with increasing cumulative dose of IL-2, was observed (linear regression coefficient r = 0.62, P = 0.0141 for t0; r = 0.80, P = 0.0003 for t1; r = 0.62, P = 0.013 for t2; r = 0.69, P = 0.045 for t3); (3) plasma nitrate + nitrite levels peaked earlier in subsequent cycles than in the first cycle; (4) all patients experienced hypotension. The mean of the systolic blood pressure values was significantly lower at the time of plasma nitrate + nitrite peak than at t0 (P = 0.0004); (5) the two cases of grade III hypotension occurred in patients with the higher mean and peak plasma nitrate + nitrite values. We conclude that determination of plasma nitrate + nitrite levels during CIVI IL-2 can usefully estimate, in a dose-dependent pattern, the degree of peripheral vascular relaxation and capillary leakage associated with cytokine action, clinically manifested as hypotension. However, isolated cardiac toxicity that continues to represent a relevant problem during IL-2 therapy, does not appear to correlate with plasma nitrate + nitrite levels; therefore, further studies are required to understand adequately the mechanisms underlying IL-2-induced cardiac toxicity.

Cell activation and death (apoptosis) induced by IL-2: ultrastructural evidence.

Peripheral blood cells of cancer patients with advanced renal carcinoma and treated with interleukin-2 by intravenous application were studied by electron microscopy at different intervals from the beginning of rIL-2 administration. Morphofunctional modifications of lymphocytes, circulating hystiocytes containing phagocitated bodies and polimorphonucleated cells undergoing phagocytosis were observed with a time dependent increase of altered cells. The analysis of blood cells cultured in the presence of recombinant IL-2 confirmed the in vivo results. Our data suggest that IL-2 induces apoptotic phenomena in the peripheral blood cells of treated patients.

Increased interleukin-10 serum levels in patients with solid tumours.

In 40 out of 99 patients (40.4%) with solid tumours of different tissue, but the same stage (IV), elevated serum levels of interleukin-10 were observed. The mean levels of the cytokine in patients with malignant melanoma (24.3 ng/ml), pancreatic (6.8 ng/ml) or gastric (6.3 ng/ml) adenocarcinoma were significantly higher than in healthy subjects (3.4 ng/ml) or in patients with uterine fibroma (1.7 ng/ml). Patients with colon (6.8 ng/ml) and renal (5.7 ng/ml) carcinoma had similar values of interleukin-10 but did not significantly differ from controls. Interleukin-10 is known to suppress the functions of both T lymphocytes and macrophages, working as a general dampener of the immune and inflammatory responses. The observation of increased circulating levels of interleukin-10 in cancer patients may have important implications for future investigations, immunological monitoring and therapeutic intervention on neoplastic patients, and suggests a mechanism for tumour cells escaping from immune surveillance.

Salvage immunotherapy with subcutaneous recombinant interleukin 2 (rIL-2) and alpha-interferon (A-IFN) for stage D3 prostate carcinoma failing second-line hormonal treatment.

Immunotherapy with subcutaneous rIL-2 and alpha IFN was administered to stage D3 prostate cancer patients after failure of secondary treatment with oral estramustine phosphate. Of a total of 15 patients, 2 are in partial response, with estramustine maintained after 44+ and 36+ weeks, respectively. Response to estramustine was observed initially in 7 of 13 patients, with a median duration of 12 weeks (range 8-20). No response to estramustine was observed in the remaining 6 patients. After the failure of estramustine, 13 patients were treated with immunotherapy. After the first cycle, progression of disease no therapy was given to those patients. A reduction of PSA levels was observed during the first cycle in 2 patients (15.3%); levels subsequently increased during the second cycle of treatment. A partial response was observed in 4 patients (30.7%), with a reduction of PSA levels in 3. The duration of response was 28 and 32 weeks in 2 patients who survived after failure for 18 and 21 weeks, respectively. Two patients are still alive, with continued partial response at 62+ and 42+ weeks. Side effects were represented mainly by a flu-like syndrome, associated with fever and nausea in all patients. The serum concentration of IL-10 was measured in 8 patients under study and in 11 matched controls. Levels higher than mean + 2D of controls before, during, or after immunotherapy were correlated with treatment failure, whereas levels below 6 ng/ml were encountered among the patients who showed a clinical response and a reduction of PSA during treatment. Within the limitations of this pilot study, it appears difficult to distinguish between a spontaneously slowly progressing disease and a true response to therapy.

Thymopentin modulates Th1 and Th2 cytokine response and host survival in experimental injury.

OBJECTIVE: To investigate the impact of thymopentin (Thy) on mortality and in vivo cytokine release in an animal model of gut-derived sepsis which includes different combinations of allogeneic blood transfusion (T) and burn injury plus bacterial gavage (BG). DESIGN: Randomly controlled experiments. MATERIAL: Two hundred sixteen Balb/c (H-2d) and 50 C3H/HeJ (H-2k) mice. INTERVENTIONS: In the first study 60 Balb/c mice were given Thy (1 mg/kg). The same day of therapy onset, 40 mice were transfused with allogeneic blood (from C3H/HeJ mice). The remaining 20 mice received aliquots of saline. Five days post-T, 20 of the 40 transfused mice were subjected to a 20% TBSA thermal injury and simultaneous gavage with 1 x 10(9) Escherichia coli and the other 20 mice underwent a sham burn. The 20 nontransfused mice also received a 20% burn plus bacterial gavage. In all animals Thy was administered for 15 days. Three control groups (n = 20 each) entered the same protocol design, but they did not receive Thy. In the second study 96 animals were randomized to six groups (n = 16 each) according to the above experimental design. Animals were sacrificed by exsanguination after burn or 5 days post-T in nonburned mice to measure TNF-alpha, IL-2, and IL-4 plasma levels. RESULTS: The highest mortality (70%) occurred when T was combined with BG. Thy significantly reduced mortality in both groups that underwent BG, regardless of the association with T. TNF-alpha was detectable in 30% of the tested samples, IL-2 in 50%, and IL-4 in 70%. Thy significantly reduced the levels of IL-4 and increased the production of IL-2. CONCLUSIONS: The protective effect of Thy in this experimental model may be mediated by modulation of cytokine release.

Serum levels of soluble cell adhesion molecules (ICAM-1, VCAM-1, E-selectin) and of cytokine TNF-alpha increase during interleukin-2 therapy.

Circulating levels of soluble intracellular adhesion molecule-1, vascular cell adhesion molecule-1, and E-selectin were measured in 20 advanced cancer patients at different times during recombinant interleukin-2 immunotherapy. The concentration of all three molecules progressively increased as did the levels of circulating tumor necrosis factor-alpha (TNF-alpha), which is known to induce endothelial cell activation. A fair direct relationship (but not statistically significant) between the raised concentration of soluble cell adhesion molecules and TNF-alpha was observed. We suggest that elevated levels of soluble adhesion molecules and TNF-alpha in the blood of IL-2-treated patients may arise from a systemic inflammatory reaction producing endothelial cell activation.

Reversible anergy in circulating lymphocytes of cancer patients during interleukin-2 therapy.

Interleukin-2 plays a crucial role in enhancing the antitumor immune response. Clinical trials, mainly in renal cell carcinoma and melanoma patients, have been carried out with encouraging results. Recent reports demonstrated that interleukin-2 therapy may depress the immune response either in vitro or in vivo. We decided to monitor, in nine renal cancer patients, the proliferative responses and the parallel variations in Ca2+ homeostasis of peripheral blood lymphocytes collected before, during and after the first cycle of a 3-day interleukin-2 systemic administration. The proliferative response to phytohemagglutinin or concanavalin A significantly dropped early during interleukin-2 infusion. Consistently, an impairment in mobilizing Ca2+, either from internal stores or via influx from outside, was observed. Results obtained with a mAb-alpha CD3 molecular complex strongly suggested that the TCR/CD3 signal transduction pathway was defective. In contrast, no major variations were observed in the general machinery controlling Ca2+ homeostasis nor in the total Ca(2+)-releasable pool. Patients' lymphocytes, cultured in vitro for 3 days in medium alone, showed an almost complete recovery in their ability to respond to mitogens. In conclusion, we show that interleukin-2 administration in cancer patients induces a reversible state of anergy in circulating lymphocytes, assessed both by the reduction in the proliferative response and the block of the mitogen-activated intracellular Ca2+ signalling.

Intrapleural administration of interleukin-2 and LAK cells in locally advanced non-small-cell lung cancer. A case report.

AIMS AND BACKGROUND: The systemic administration of recombinant interleukin-2 (rIL-2) and lymphokine-activated killer (LAK) cells is ineffective in non-small-cell lung cancer (NSCLC). However, there is some evidence that their intrapleural administration could be effective, since it increases the concentrations of the cytokine and the effector cells in the tumor area, thereby obtaining greater antitumor activity. STUDY DESIGN: We report the case of a patient affected by a locally advanced lung adenocarcinoma with pleural effusion (T4 N0 M0-stage IIIb) treated with repetitive courses consisting of a priming continuous i.v. infusion (48 h) of rIL-2 (18 MIU/m2/day) intraplural administration of LAK cells (3-9 x 10(9)/day), in a single daily bolus, for 3 consecutive days and concomitant administration of rIL-2 (1.8-7.2 x 10(6) IU/day), for 5 days. RESULTS: We observed early disappearance of neoplastic cells in the pleural effusion, progressive decrease until disappearance of the pleural effusion, cavitation of the primary lesion during the treatment, and its stabilization for 9 months until progression. Radiologic changes were accompanied by a marked eosinophilia (up to 50 x 10(9)/L), and the intrapleural route of administration of rIL-2 induced a relevant increase in eosinophil count in peripheral blood. Immunologic changes in lymphocyte subpopulation phenotypes were also observed. The performance status of the patient improved, and she was still alive and eupnoic 25 months from the diagnosis and 23 months from the start of treatment. CONCLUSIONS: This case suggests a therapeutic role for intrapleural rIL-2, and we believe that the relationship among intrapleural administration of rIL-2 and LAK cells, the development of peripheral eosinophilia, and clinical response should be further investigated.

Thymopentin increases the survival of mice after allogeneic blood transfusion, bacterial gavage, and burn injury.

OBJECTIVE: To investigate the effect of blood transfusion on mortality and the incidence of bacterial translocation in mice subjected to thermal burn or bacterial gavage, or both, and to assess the influence of thymopentin on mortality. DESIGN: Randomly controlled experiments. SETTING: University departments of surgery, immunology and nuclear medicine. MATERIAL: 235 Balb/c (H-2d) and C3H/HeJ (H-2k) mice. INTERVENTIONS: 8 groups of 20 mice each received: saline infusion (controls), blood transfusion (BT) alone, 20% burn alone, gavage with 1 x 10(10) Escherichia coli alone, BT and gavage, BT and burn, burn and gavage, or BT, burn, and gavage. A further 3 groups of 10 mice were all gavaged with 111In-biotin labelled E coli and randomised to additional BT and burn, BT alone, or burn alone. 98 mice that had had BT, burn, and gavage, were then randomised to receive thymopentin 0, 0.1, 1, or 5 mg/kg/day for 15 days. The impact of the pretreatment with thymopentin on PGE2 concentration was also evaluated in a separate group of 45 mice that received BT, burn, and gavage; or burn and gavage. MAIN OUTCOME MEASURES: Survival, degree of translocation. RESULTS: The highest mortality (75%) was in the BT, burn, and gavage group. BT alone significantly reduced survival in burned mice, whereas BT alone or associated with gavage had no effect. Thermal injury had the most influence on bacterial translocation, whereas BT did not increase it. Thymopentin significantly improved survival, particularly in the higher doses. The pretreatment with thymopentin significantly reduced PGE2 concentration after BT, burn and gavage. CONCLUSION: Burn injury significantly increased mortality in the presence of immune deficiency caused by BT. Thymopentin reduced mortality, possibly by immunomodulation.

Treatment of advanced renal cell cancer with sequential intravenous recombinant interleukin-2 and subcutaneous alpha-interferon.

Starting from in vitro studies suggesting synergistic antitumour activity against renal cell cancer (RCC) of recombinant interleukin-2 (rIL-2) and alpha-interferon (IFN), a phase II trial was initiated to test the clinical activity of this combination. The two cytokines were administered sequentially, with the aim of reducing the risk of additive toxicity and enhancing the immunological reaction against the tumour. The original treatment schedule consisted of rIL-2 18 x 10(6) U/m2/day by continuous intravenous infusion for 120 h days 1-5, and alpha-IFN 2b, at a flat dose of 9 x 10(6) U by subcutaneous or intramuscular injection thrice in a week, from day 8 to 28. Treatment was planned to be continued for six or more 28-day cycles, depending on clinical response. 12 patients were treated according to this schedule; as some cardiovascular toxicity was experienced in this set of patients, 11 further patients were treated with half-dose rIL-2 (i.e. 9 x 10(6) U/m2/day). 17 out of 23 enrolled patients completed at least one cycle of treatment and were evaluated for response. We observed six major responses [one complete response (CR) + five partial responses (PR)] for an objective response rate of 35% [95% confidence interval (CI) 17-59%]. 5 additional patients achieved stabilisation of disease; one of them reached CR after surgical extirpation of a lung mass. Sites of response included lung, nodes and bone. Duration of response is 12+ months for CR; 17, 16, 12+, 9 and 9 months for PRs. Median survival is 16 months. Response was not significantly different between full-dose and half-dose rIL-2. Considering stable disease (SD) as responses, there seemed to be a higher chance of response for patients with smaller tumour burden (P = 0.032). The toxicity of rIL-2 treatment, mainly cardiovascular, was substantial; 9 patients experienced severe cardiotoxicity, consisting of major arrhythmias, myocardial ischaemia, reduction of ejection fraction measured with heart radionuclide scan, and were excluded from continuing treatment. Other rIL-2-related toxicities forcing exclusion from the study were severe thrombocytopenia (1 case), and generalised exfoliative dermatitis requiring steroids (1 case). Otherwise, treatment was well tolerated; rIL-2-related toxicities promptly recovered after rIL-2 discontinuation in the majority of cases, and no treatment-related deaths were reported. The half-dose rIL-2 regimen was significantly less toxic in terms of hypotension (P = 0.014), fever (P = 0.014), oliguria (P = 0.042), serum creatinine elevation (P = 0.009) and prothrombin time elongation (P = 0.038).(ABSTRACT TRUNCATED AT 400 WORDS)

Recombinant interleukin-2 and lymphokine-activated killer cells in renal cancer patients: II. Characterization of cells cultured ex vivo and their contribution to the in vivo immunomodulation.

Peripheral blood mononuclear cells, obtained from patients with renal cell cancer and cultured ex vivo, exhibit high natural killer (NK) and lymphokine-activated killer (LAK) activity (also against allogeneic fresh tumour cells), which is transcribed into the hosts' immune status after reinfusion. Phenotypic analysis shows a slight increase in the percentage of CD56+ and CD8+ lymphocytes, while CD4+ lymphocytes decrease slightly. As a sign of activation an increase of cells expressing DR and CD25 antigens is observed. At the peripheral blood level, mononuclear cells show an increase, compared to basal values, of NK and LAK activity, especially at the end of the first infusion cycle. Phenotypic analysis of the patients' PBMC shows a decrease of CD3+CD4+ T lymphocytes and an increase of NK cells (CD3-CD56+CD16+) and of cells expressing activation markers (DR and CD25), particularly evident by the end of the second infusion cycle. Finally, in addition to the changes induced by IL-2 alone, reinfusion of incubated cells results in an activation of CD56+ and LeuM3+ cells.

LAK1 antigen defines two distinct subsets among human tumour infiltrating lymphocytes.

Both lymphokine activated killer (LAK) cells and specific cytotoxic T lymphocytes appear to play a role in tumour immunity. Tumour infiltrating lymphocytes (TIL) which display a CD56+ phenotype (both CD3+ and CD3-) are also likely to possess anti-tumour activity. We have previously described a 120 kDa surface antigen, termed LAK1, expressed on a subset of human peripheral blood lymphocytes (20-50%) with both NK and LAK activity. The aim of the present study was to determine whether LAK1 antigen is able to distinguish among TIL two populations of effector cells displaying either specific or non MHC-restricted (NK/LAK) activity. We showed that about 25% of freshly derived TIL were weakly stained with anti-LAK1 monoclonal antibody and most of them were also CD3+ CD56-. After culture in recombinant interleukin-2 the majority of TIL were CD3+ CD56- and the percentage of LAK1+ cells increased up to 50%. Among cloned TIL, only those lacking LAK1 antigen displayed a specific cytotoxicity against the autologous tumour, whereas the non-lytic clones were able to produce both tumour necrosis factor and gamma-interferon. Moreover, when TIL from a renal cell carcinoma were fractionated into LAK1- and LAK1+ populations, the specific lytic activity was mainly evident when LAK1- lymphocytes were used as effector cells. Conversely, LAK activity was confined to the LAK1+ subset.

Recombinant interleukin-2 and lymphokine-activated killer cells in renal cancer patients: I. Phenotypic and functional analysis of the peripheral blood mononuclear cells.

The efficacy of recombinant interleukin-2 (rIL-2) or rIL-2 plus lymphokine-activated killer (LAK) cells in cancer therapy has been demonstrated by several groups both in experimental models in animals and clinical trials in humans, but their effects in vivo have yet to be clarified. Starting February 1988, we have treated 12 patients affected by advanced renal cancer with rIL-2 + LAK cells according to an open, non-randomized, phase II trial. Immediately before each rIL-2 infusion and during the last day of infusion, immunological tests were performed on the patients' peripheral blood mononuclear cells. During rIL-2 infusion we have observed a slight increase of the spontaneous cell proliferation and of natural killer (NK) and LAK activity; phenotypic analysis showed a significant decrease in the CD4+ T-lymphocyte subset, both in percentage and in absolute number. Conversely, before each cycle CD4+ cells increased when compared to basal values. No significant variations were observed in the CD8+ T-lymphocyte subset. Furthermore, a significant increase of the NK cells (CD3- CD56+ CD16+) was evident during rIL-2 infusion.