Interleukin 2 therapy: current role in surgical oncological practice

The psychological and psychiatric effects of rIL-2 therapy: a controlled clinical trial

It has been suggested that recombinant interleukin-2 (rIL-2) may cause pyschological and psychiatric problems, although the effects of rIL-2 on its own have not been well documented. To evaluate these effects, 17 patients with advanced colorectal cancer took part in a randomised, parallel group study of rIL-2 with chemotherapy (5-fluorouracil and leucovorin) versus chemotherapy alone. Patients were assessed regularly by means of various psychometric tests including the Hospital Anxiety and Depression Scale, the Mood Rating Scale, the Mini-Mental State Examination, the Digit Symbol Substitution Test, the Trail-Making Test and the Benton Revised Visual Retention Test. Rigorous discontinuation criteria were applied to ensure that the effect of time-related variables did not influence the results. Compared with patients who were given chemotherapy alone, patients receiving immunochemotherapy reported reduced energy, impaired confidence, higher depressed mood and more confusion. Immunochemotherapy was rated as more distressing than chemotherapy alone and patients reported a greater incidence of appetite impairment, weight loss, poor concentration and fever. Cognitive assessments indicated that brain dysfunction can be caused by rIL-2. Compared with the control group, patients receiving immunochemotherapy showed significant impairment on Trail Making Test B and the Digit Symbol Substitution Test. One patient developed repeated transient psychotic episodes associated with rIL-2 infusions and another regularly became confused. These effects were not due to sleep deprivation or pyrexia. Treatment with rIL-2 should not be discarded on psychosocial grounds, although in each case the psychological morbidity and adverse effects on quality of life need to be balanced carefully against potential therapeutic benefits.

Modulation in vitro of human natural cytotoxicity, lymphocyte proliferative response to mitogens and cytokine production by essential fatty acids

Essential fatty acids (EFA) have been shown in animal studies to have a differential effect on various aspects of immune reactivity. However, there have been few studies in humans. Therefore, we elected to investigate the effects of a variety of EFA [gamma-linolenic acid (GLA), eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)] in vitro on human blood lymphocyte reactivity, cytokine secretion and natural cytotoxicity. The proliferative response to polyclonal mitogens (phytohaemagglutinin, pokeweed mitogen, concanavalin A), as measured by [3H]thymidine incorporation into newly synthesized lymphocytes, was inhibited (P < 0.05) by all EFAs tested, in a dose-dependent manner (3-15 micrograms/ml). The greatest inhibition of proliferation was caused by EPA and DHA. Similarly, EPA, DHA and GLA significantly reduced cytotoxic activity [expressed as lytic units, using 51 chromium-release assays natural killer (NK) (K562 cells) and lymphokine-activated (LAK) (Daudi cells) cells] (P < 0.05) in a concentration-dependent manner (5-50 micrograms/ml), without affecting cell viability. EPA and DHA exhibited greater suppression than GLA. Furthermore, the inhibition of cell proliferation and suppression of natural cytotoxicity was associated with marked decrease in cytokine [interleukin-1 (IL-1), IL-2, tumour necrosis factor-alpha (TNF-alpha) and interferon-gamma (IFN-gamma)] production in vitro. Our findings demonstrate that EFAs (GLA, EPA, DHA) have the potential to inhibit significantly various aspects of human lymphocyte cell-mediated and humoral immune reactivities.

The cognitive effects of recombinant interleukin-2 (rIL-2) therapy: a controlled clinical trial using computerised assessments

It has been suggested that patients undergoing treatment with recombinant interleukin-2 (rIL-2) may develop cognitive impairment. To evaluate these effects, 17 patients with advanced colorectal cancer took part in a randomised, parallel group study of rIL-2 with chemotherapy (5-fluorouracil and leucovorin) and chemotherapy alone. Assessments were carried out daily whilst patients were in hospital and regularly between cycles of treatment using state-of-the-art computerised cognitive assessment, as well as traditional psychometric tests. Rigorous discontinuation criteria were applied to ensure that the effect of time-related variables did not influence the results. One patient developed repeated transient psychotic episodes associated with rIL-2 infusions and another regularly became confused. Computerised cognitive assessments revealed that immunochemotherapy produced significant impairment in various tasks, especially reaction time, picture recognition and vigilance. These effects were not due to sleep deprivation or pyrexia. For most patients, cognitive functioning was restored to the baseline level within 10 days following the cessation of rIL-2. In conclusion, during infusions of rIL-2, some patients experience severe confusion and amnesia which resembles some of the major cognitive impairments associated with dementias such as Alzheimer's disease. Computerised cognitive assessment using the Cognitive Drug Research system provides a feasible, sensitive and reliable method of evaluating cognitive changes in patients with cancer. It could usefully be included in quality of life assessments in clinical trials where treatment-related cognitive changes need to be evaluated.

Repeated immunotherapy using intratumoural injection with recombinant interleukin-2 and tumour-infiltrating lymphocytes inhibits growth of breast cancer and induces apoptosis of tumour cells

This study tested the effect of repeated intratumoural injection with recombinant interleukin-2 (rIL-2) and tumour-infiltrating lymphocytes (TILs) on inhibition of growth of breast cancer and on induction of apoptosis of tumour cells. The tumour cell line LDLX43 was used to induce breast cancer in Wistar rats. Group I (10 rats) was the control. Group II (12 rats) received repeated intratumoural injection with rIL-2 and TILs. rIL-2 at the dose of 5 x 10(5) IU/day was given for 7 days, and 1 x 10(7) TILs were injected on the second day of each rIL-2 therapy, for a treatment session. Overall, two treatment sessions of immunotherapy with rIL-2 and TILs were given in all treated animals. Rapid increased tumour volume was found in the control group. In the treated group the total response rate was 42%, of which 25% tumours showed partial regression and 17% tumours reached complete remission where infiltration of plenty of T lymphocytes was detected, indicating that T cell-mediated antitumour immunity is primarily responsible for tumour rejection. Further investigation showed the repeated immunotherapy using intratumoural injection with rIL-2 and TILs could induce the development of apoptosis of breast cancer cells.

Natural killer cells and cancer

BACKGROUND

Natural cytotoxicity, mediated by natural killer (NK) cells and cell with lymphokine-activated killer (LAK) activity, is believed to play an important role in host anti-cancer mechanisms.

METHODS

The authors critically review recent publications on the role of natural cytotoxicity in patients with cancer.

RESULTS

In patients with cancer, several studies have noted variations in the numbers and activity of NK and cells with LAK activity in different body compartments. NK cell activity in the peripheral blood lymphocytes (PBLs) is higher than that found in lymph nodes and within tumors, and this appears to be due to the presence of suppressor factors. The natural cytotoxicity of PBLs in patients with different types of cancers varies. However, there appears to be a trend for natural cytotoxicity to be reduced in certain cancer patients, possibly related to tumor volume or dissemination. Anti-cancer treatments (e.g., surgery, hormonal modulation, radiotherapy and chemotherapy) can also result in suppression of natural cytotoxicity, although the long-term effect on response to treatment and development of metastases is at present unknown.

CONCLUSIONS

NK and LAK cells, through the use of immune biologic modifiers, have been demonstrated to have a therapeutic role in the treatment of human cancers. Further studies are required to determine the optimal dosages and combinations of chemotherapeutic agents, the timing of surgery, and the adjuvant use of immune biologic response modifiers. An increasing awareness and understanding of this field, may allow for the future development of anti-cancer therapies.

Effect of ranitidine and low-dose interleukin-2 in vitro on NK-cell activity in peripheral blood from patients with liver metastases from colorectal cancer

Peripheral venous blood from 12 patients with colorectal cancer and eight healthy volunteers was used to identify the lowest in vitro dose of human, recombinant interleukin-2 (rIL-2) with immunoactivity on NK-cell lysis of K562 tumour cells. Subsequently, this dosage of 200 units/ml rIL-2, which may respond to 10(6) units in vivo, was used alone or in combination with ranitidine (0.02 mg/ml, which may correspond to 100 mg in vivo) to improve in vitro NK-cell activity in peripheral blood from 25 patients with liver metastases from colorectal cancer. A standard 4-hour Cr51-release assay of K562 tumour cells was used for the analyses. Spontaneous NK-cell activity was 19.0% (6.5-33.2), while ranitidine-induced NK-cell activity was 23.6% (7.8-46.2), and without statistical difference from spontaneous activity. Recombinant IL-2-induced NK-cell activity was 37.1% (11.1-71.7) (P < 0.05 compared to spontaneous activity), and rIL-2 plus ranitidine-induced NK-cell activity was 52.7% (18.9-85.6) (P < 0.05 compared to spontaneous and to rIL-2-induced activity, respectively). These results suggest a synergistic increase of low-dose rIL-2-induced NK-cell activity by ranitidine. Therefore, the combination of low-dose rIL-2 and ranitidine may be beneficial to improve post-operative immune competence, and should be considered in future adjuvant treatment regimens of cancer patients.

Effect of essential fatty acids on natural cytotoxicity in patients with colorectal cancer

Essential fatty acids (EFAs), have been shown to modulate lymphocyte reactivity and destroy various tumour cells in vitro. Natural cytotoxicity, mediated by NK and LAK cells, is believed to play an important anti-cancer role in vivo. The effect of EFAs, given orally as dietary supplementation, on NK and LAK cell cytotoxic activity in patients with localized (n = 10) and advanced (n = 20) colorectal cancer has been studied, using in-vitro 51Cr release cytotoxicity assays with K562 (NK) and DAUDI (LAK) cells. The activity of effector cells was expressed as lytic units. NK cell activity showed no significant change following 15 days ingestion of EFAs in the group with localized cancer, but was significantly reduced in the group with advanced disease and continued to decline, reaching minimal levels following 6 months supplementation (P = 0.017). LAK cell activity showed no overall alteration after 15 days ingestion of EFAs in patients with localized cancer, but in the group with advanced disease, the reduction in the activity occurred at day 15 and steadily declined on prolonged intake, reaching significant minimal levels after 6 months of supplementation (P = 0.05). Cell surface marker analysis (FACS-CD MABs) revealed reduced absolute numbers of CD16+, CD56+ and CD57+ lymphocytes (P < 0.05) in the patients with advanced colorectal cancer. More importantly, the cytotoxicity of NK and LAK cells returned to the pre-supplementation values, 3 months after cessation of EFA intake. Furthermore, there was no alteration in the cytotoxic activity of NK and LAK cells in the control group (advanced colorectal cancer without EFA supplementation) during the 6 months period of evaluation. These results suggest that prolonged EFA supplementation, in the doses used in this study, may have detrimental effects on natural anti-cancer cytotoxic mechanisms in patients with malignant disease.

Acute phase proteins and recombinant IL-2 therapy: prediction of response and survival in patients with colorectal cancer

Twenty-four patients with metastatic colorectal cancer were treated with recombinant IL-2 (rIL-2) by continuous intravenous infusion for 5 days (18 x 10(6) U/m2 per 24 h), followed by three injections of 5-fluorouracil (600 mg/m2) and folinic acid (25 mg/m2) at weekly intervals. The response to treatment was assessed using standard UICC criteria (partial or complete response, stasis or progression of disease). The serum concentrations of the acute phase proteins; C-reactive protein (CRP), retinol binding protein (RBP), alpha 1-antitrypsin (alpha 1-AT), transferrin (TF) and albumin were measured. A response to therapy occurred in the tumours of seven (29%) of the 24 patients (two complete and five partial responses). All patients who demonstrated a response to treatment had a serum albumin level of > 37 g/l and a CRP level of < or = 10 mg/l. In contrast, of the 17 patients who did not respond to therapy, 12 (71%) had a serum albumin of less than 37 g/dl and a CRP of greater than 10 mg/l. Examination of the survival times of the 12 patients who had a pretreatment serum albumin level of less than 37 g/l revealed that all had died within 12 months of cessation of therapy. However, 58% of patients with pretreatment serum albumin levels of greater than 37 g/l survived for longer than 12 months. These results have shown that (i) patients who respond to rIL-2-based therapy and (ii) those patients who have prolonged survival times, can be identified by pretreatment measurement of serum levels of acute phase proteins.

Flow cytometric determination of intracellular calcium changes in human peripheral blood mononuclear cells during conjugation to tumour cell lines

Using a flow cytometric assay, conjugate formation between human peripheral blood mononuclear cells (PBMC) and three different human tumour cell lines has been analysed. Changes in the intracellular calcium levels of PBMC were monitored using the calcium sensitive dye Fluo-3. Target cell populations were distinguished by forward scatter or following loading with the fluorescent dye, SNARF-1. Intracellular calcium was expressed as a ratio of fluorescence of conjugated to unconjugated PBMC and followed for ten minutes after initiation of conjugation. The results demonstrate an apparent increase in intracellular calcium in PBMC conjugated to the NK-sensitive cell line K562, and that the kinetics and magnitude of this response varied considerably between individuals. Tumour cells which were resistant to lysis (as determined in a 4 h chromium release assay) were also capable of eliciting a calcium response from PBMC. Although the induction of a rise in intracellular calcium was therefore not correlated with cytotoxicity, it was greater in IL-2-activated PBMC upon exposure to the same target cell lines as PBMC.

Immunotherapy in liver tumors: III. A new experimental model of metastatic liver tumors from colorectal carcinoma for cytokine therapy

A new model of metastatic liver tumors in Wistar/Furth rats is introduced. A colorectal adenocarcinoma cell line (LDLX40) induced by 1,2-dimethylhydrazine was injected through one of the branches of the ileal mesenteric vein to develop metastatic liver tumors in rats. On day 30 after the inoculation of tumor cells, micrometastases were detected under microscopy in all animals that received tumor inoculation. Macrometastases in 87.7% of animals were found by either the tumor staining test or ultrasonography. No extrahepatic tumor developed in this tumor model. To observe the effects of different treatment strategies on metastatic liver tumors, 35 animals were randomly divided into four groups. Group I served as control. Group II underwent hepatic artery ligation (HAL). Group III received intraportal administration of recombinant interleukin-2 (rIL-2) and interferon-alpha (IFN-alpha). Group IV had intraportal medication of rIL-2 and IFN-alpha + HAL (the IIH protocol). Results indicated that rapid tumor growth was seen in the control tumors. HAL produced little response to metastatic liver tumors as compared to the control group (P > 0.05). The combined application of rIL-2 and IFN-alpha showed an improved result, with 22% of tumor growth inhibition or regression (P < 0.05 compared to the control group). Twenty-eight percent of tumor growth restraint or regression was found in the group treated with the IIH protocol (P < 0.05 compared to the control group). We conclude that this new experimental model of metastatic liver tumors is reproducible, and that the IIH protocol is effective in the treatment of metastatic liver tumors in rats. These beneficial effects from the IIH protocol may be introduced into patients with metastatic liver tumors.

A phase III study of recombinant interleukin-2, 5-fluorouracil and leucovorin versus 5-fluorouracil and leucovorin in patients with unresectable or metastatic colorectal carcinoma

135 patients with locally advanced or metastatic colorectal cancer were entered into a phase III trial evaluating the efficacy of chemoimmunotherapy [recombinant interleukin 2 (rIL2)/5-fluorouracil (5-FU) and leucovorin (LV)] versus chemotherapy alone (5-FU/LV). A cycle of chemoimmunotherapy comprised a constant intravenous infusion of rIL2 at a dose of 18 x 10(6) U/m2/24 h for 120 h, followed by three bolus injections of 5-FU (600 mg/m2) and LV (25 mg/m2) at weekly intervals. Patients receiving chemotherapy alone received 5-FU/LV at the same dose at weekly intervals for 6 weeks followed by a rest period of 2 weeks, constituting one cycle of therapy. A maximum of 6 months therapy was given in both arms of the study. The response rates (complete and partial responses) were 17% in patients receiving rIL2/5-FU/LV versus 16% in those in the 5-FU/LV arm of the study. Median survival and progression-free survival were comparable for the two groups of patients, although there was a trend for a prolongation of survival in patients receiving chemoimmunotherapy compared with chemotherapy alone, beyond 12 months. Retrospective subgroup analyses revealed a significantly increased survival in poor prognosis patients (ECOG 1) treated with rIL2/5-FU/LV when compared to those receiving chemotherapy alone. Therefore, further studies evaluating the dose and duration of chemoimmunotherapy in patients with metastatic colorectal cancer seem warranted.

Human, recombinant interleukin-2 induces in vitro histamine release in a dose-dependent manner

We previously observed that human, recombinant interleukin-2 in a pharmacologic dose (200 u/ml) induced histamine release from monocyte-depleted peripheral blood mononuclear cells in vitro. Therefore, we studied the role of various pharmacologic doses of rIL-2 on in vitro histamine release. Peripheral blood mononuclear cells (5 x 10(6) cells/ml), which also contain basophils, from 13 patients scheduled for elective colorectal cancer surgery and 10 age and sex matched healthy volunteers were stimulated with rIL-2 in concentrations of 0, 50, 100, 200, 450, 900, 1,800 and 3,600 u/ml, respectively, for 1, 24 and 48 hours under standard conditions. Histamine was analysed in supernatants using the glass fiber method. Simultaneously, total cell-bound histamine was analysed in lysate from 5 x 10(6) mononuclear cells from all patients and volunteers, thus allowing determination of percent histamine release. Supernatant histamine concentration from unstimulated cells was 17.2 +/- 1.5 ng/ml in patients compared to 7.9 +/- 1.0 ng/ml in volunteers (#p < 0.05) after 1 hour stimulation, and no further increase was observed after 24 and 48 hours, respectively. Histamine concentration increased significantly in the supernatant from cells stimulated by rIL-2 in a dose-dependent manner both in patients and volunteers. Total cell-bound histamine was 49.3 +/- 4.1 ng/ml in patients compared to 78.5 +/- 7.7 ng/ml in volunteers (p < 0.05). Therefore, both spontaneous and rIL-2-induced histamine release was significantly enhanced in cancer patients compared to volunteers (*p < 0.05). These data suggest that rIL-2 in high pharmacologic doses stimulates in vitro histamine release in a dose-dependent manner in both cancer patients and volunteers. This may in part explain the severe toxicity observed during high-dose rIL-2 therapy.

Immunotherapy in liver tumours: I. Combined administration of recombinant interleukin-2 (IL-2) and interferon-alpha through the subcutaneous transposed spleen prolongs the half-life of IL-2 in vivo and enhances antitumor effects

Two modifications in this study, including the use of subcutaneous transposed spleen (STS) as a port for administration of recombinant interleukin-2 (IL-2) and interferon-alpha (IFN-alpha), and the mixture of IL-2 and IFN-alpha with degradable starch microspheres (IIM), for the treatment of rat liver tumor are introduced. Group I is the control. Group II received the IIM schedule through the STS. Group III and group IV received IL-2 and IFN-alpha, diluted with normal saline and injected through the STS or a peripheral vein. The comparative studies indicated that the best result was seen in group II where the elevated concentration of IL-2 in portal blood and massive tumor necrosis with lysis were observed. Inhibitions of tumor growth of 33%, 20% and 13% in group II, III, and IV, respectively, were observed. We conclude that administration of the IIM schedule through the STS is an effective method for the treatment of liver tumor in rats.

Immunotherapy in liver tumors: II. Intratumoral injection with activated tumor-infiltrating lymphocytes, intrasplenic administration of recombinant interleukin-2 and interferon alpha causes tumor regression and lysis

This study tested the effect of intratumoral injection with activated tumor-infiltrating lymphocytes (TIL), and simultaneous administration of recombinant interleukin-2 (rIL-2) and interferon alpha (IFN-alpha) (LII protocol), on mouse liver tumor. Group I (n = 10) served as the controls. Group II (n = 17) received rIL-2 + IFN-alpha schedule. Group III (n = 20) received the LII protocol. A total of 5 x 10(6) of TIL were injected into 4 sites of a tumor in a single treatment. rIL-2 (1 x 10(6) IU) on the first day and IFN-alpha (1 x 10(5) IU) on the second day were alternately given with a total of 10 treatment doses that were completed in 20 days. Tumor remission or regression rates of 29% and 40% were obtained in groups II and III, respectively, but no remission was obtained in the controls. A large number of TIL were also observed in the tumors treated with the LII protocol. Making comparisons between the control group and IL-2 + IFN-alpha schedule, and the control group and LII protocol, the ratios of cytolytic activity of TIL in vitro were 0:32 and 0:57, respectively. We conclude that the LII protocol appears to be more effective in the treatment of mouse liver tumor than the IL-2 + IFN-alpha schedule, and that it may be a new promise for the treatment of patients with liver malignancies.