Riks A. Maas

Interleukin-2 in cancer treatment : disappointing or (still) promising ? a review

The central question to discuss in this review is whether the results of interleukin-2 (IL-2) treatment are still disappointing or again promising. Although in the (recent) past application of high doses of systemically applied rIL-2 has led to some success, the overall results are not as one had hoped. Considering these poor results it seems clear that the application of high systemic doses rIL-2 was not a good choice. IL-2 has been used more or less as a chemotherapeutic compound in the highest tolerable dose. This has led to a great number of unwanted toxic side-effects. In addition, these doses mainly stimulated nonspecific lymphokine-activated killer activity through low-affinity IL-2 receptors, which does not lead to systemic immunity. On the other hand, several groups have shown that application of intratumoral low doses of IL-2 can be highly effective against cancer and without toxic side-effects. Significant tumor loads constituting up to 6% of the total body weight of a mouse were eradicated after treatment with low-dose rIL-2 given locally. Furthermore local treatment can lead to eradication of a tumor at a distant site. This type of therapy is effective in many systems namely against different tumor types in mice, hepatocellular carcinoma in guinea-pigs and vulval papilloma and carcinoma and ocular carcinoma in cattle. Low-dose IL-2 is very effective in experimental animals if it is given relatively late after inoculation of the tumor cells. In other words, it seems necessary that some sort of immune reaction has started or is developing before low doses of rIL-2 effectively stimulate it. In fact there is strong evidence that T lymphocytes, both CD4+ and CD8+ cells, are directly involved in the process leading to induction of specific immunity. In our opinion rIL-2 therapy should therefore aim at the stimulation of such (originally weak) specific immune reaction. Under these conditions also systemic immunity can be induced. In conclusion, application of rIL-2 as a modality for cancer treatment is still promising. High priority should be given to a further delineation of the mechanisms involved after local application. The method of giving IL-2 systemically in the highest tolerable dose should be abandoned. Specific stimulation of the immune system by low-dose rIL-2 is a much more promising option.

Intratumoral low-dose interleukin-2 induces rejection of distant solid tumour

SummaryThis study shows that local tumour treatment with low-dose recombinant interleukin-2 (IL-2) can mediate rejection of a large distant solid tumour. When SL2 lymphoma cells were injected intraperitoneally (i.p.) in syngeneic DBA/2 mice on day 0, 70% of these mice were cured by daily i. p. injections with 20 000 units IL-2 on days 10–14. After injecting mice with SL2 both i.p. and subcutaneously (s. c.) on the flank, 50% of the mice treated i.p. with low-dose IL-2 rejected both the i.p. tumour and the large distant s.c. tumour. In contrast, i.p. IL-2 treatment on days 10–14 cured fewer than 10% of the mice bearing only a s. c. SL2 tumour. The described IL-2 immunotherapy also caused systemic tumour rejection in mice bearing both ascitic and solid P815 mastocytoma. Thus it was shown that low-dose IL-2 can induce systemic tumour rejection, when injected at a site of tumour growth. Interleukin-2-induced rejection of s. c. SL2 tumour was highly specific, as mice that were rejecting i.p. and solid s. c. SL2 lymphoma did not reject solid P815 mastocytoma, which was injected s.c. simultaneously on the other flank. Furthermore, solid s.c. tumours consisting of mixtures of SL2 and P815 were not rejected in mice that rejected i.p. SL2 or P815. We conclude that intratumoral injections of low-dose IL-2 can enhance an ongoing weak immune reaction against the tumour resulting in systemic tumour rejection.

Optimal regimes for local IL‐2 tumour therapy

In this report we present studies on optimal regimes for regional IL‐2 therapy, focused on dose, schedule and site of injection. Original data obtained in 2 murine tumour models show that all 3 factors are of importance. Anti‐tumour responses were most effective when IL‐2 was administered regionally 5 to 10 times, at doses ranging from 7,000 to 33,000 IU/day every day or every other day. This resulted in cure rates of more than 40% in mice bearing ascitic tumour that had also disseminated to liver and lungs. The importance of these data is discussed in the light of previous results of our group. These results illustrate that the doses and schedules used in this study are not effective exclusively in these 2 tumour models but may have a more general applicability. ©1996 Wiley‐Liss, Inc.

Growth arrest associated changes of mRNA levels in breast cancer cells measured by semi-quantitative RT-PCR: potential early indicators of treatment response

To find early and sensitive indicators of treatment response in breast cancer, we studied the mRNA levels of proliferation-related genes during growth arrest of the human breast cancer cell lines T47D and MCF7. A sensitive reverse transcriptase-PCR (RT-PCR) technique was used in order to monitor gene expression in small samples of cells. Estrogen-depletion and treatment with tamoxifen effectively induced a G1-arrest in both cell lines, accompanied by a decrease of the mRNA levels of histone H4, cyclin A, cyclin D1, and c-myc. Cyclin A expression decreased most strongly: up to 32-fold within 7 days. The expression of c-fos and WAF1 increased during growth arrest. In conclusion, significant changes of the levels of proliferation-related mRNAs, induced by growth arrest, can be measured in small samples of breast carcinoma cells using RT-PCR. Especially the decrease of the cyclin A mRNA level seems a potential early indicator of clinical response to tamoxifen therapy in breast cancer patients.

Effector cells of low-dose IL-2 immunotherapy in tumor bearing mice: Tumor cell killing by CD8+ cytotoxic T lymphocytes and macrophages

The presence and cytotoxicity of tumor infiltrating cells is described in mice during effective immunotherapy with interleukin 2 (IL-2). DBA/2 mice were inoculated i.p. with 2 x 10(4) tumor cells on day 0 and treated with daily i.p. injections with 20,000 units IL-2 on days 10-14. Mice bearing a large syngeneic i.p. tumor burden (SL2 lymphoma, P815 mastocytoma, L5178Y lymphoma, or L1210 lymphoma) could be cured by i.p. immunotherapy with these low doses of IL-2. In the peritoneal cavity of these mice an infiltrate of mononuclear cells was present. Similar numbers of lymphocytes (10(6)-10(7)) and macrophages (+/- 10(7)) were present in control tumor bearing mice and IL-2 treated tumor bearing mice. The ratio of CD4+/CD8+ T lymphocytes in the peritoneal cavity of mice rejecting the SL2 tumor was smaller than 0.5, whereas this ratio is about 2 in naive mice. In the spleens of IL-2 treated tumor bearing mice only a minor decrease of CD4+/CD8+ ratio was observed from 2.1-2.4 to 0.9-1.9. T cells isolated from the peritoneal cavity of mice inoculated with SL2 tumor cells and treated with IL-2, were highly cytotoxic to SL2 cells: at E:T ratio 2:1 the cytotoxicity index was 37 +/- 3. This cytotoxicity was specific and mediated by CD8+ T lymphocytes. Macrophages that were present in the peritoneal cavity of mice treated with IL-2 were also highly cytotoxic. The C.I. of these cells was 63-76% at E:T ratio 1:1. Cytotoxic macrophages were also present in untreated tumor bearing mice. The i.p. injections of IL-2 (20,000 units/day) caused a four-fold increase in the local NK-activity in the peritoneal cavity in naive mice. These IL-2 injections did not generate LAK-activity in vivo. Specificity of the in vivo tumor rejection was tested by injection SL 2 i.p. on day 0 and P815 i.p. on day 10, or vice versa, followed by IL-2 treatment. Only the tumor cells that were injected on day 0 were rejected. These in vivo experiments point to specific tumor rejection. In conclusion, both cytotoxic macrophages and CTLs are present in a sufficient number and with sufficient cytotoxicity to explain the killing of tumor cells in the peritoneal cavity. The CTL-activity seems of decisive importance for tumor rejection as this is induced by IL-2.

A method to monitor mRNA levels in human breast tumor cells obtained by fine-needle aspiration

A method based on the reverse transcriptase-polymerase chain reaction (RT-PCR) was developed that allows the determination of relative mRNA expression levels in fine-needle aspirates from human tumors. The method was developed for the c-erbB-2 gene, using the porphobilinogen deaminase (PBGD) gene as an internal standard. It was validated for mRNA isolated from cell lines and for material obtained by fine-needle aspiration from human breast cancer. Gene expression levels were determined by measuring the activity of radiolabeled RT-PCR-amplified gene-specific bands with a phosphor imager. At least four points are measured on the log-linear part of the amplification cycle versus signal intensity curves, and subsequently the distance between the curves of the gene of interest and that of an internal standard gene is used to calculate the relative expression levels. The method worked equally well with the BRCA1 gene, illustrating that it can be generalized to other genes. The method is suitable to measure or monitor semiquantitively gene expression levels in accessible human tumors in situ.

Transfer of tumor immunity by both CD4+ and CD8+ tumor infiltrating T lymphocytes activated in vivo by IL-2 therapy of tumor bearing mice.

DBA/2 mice were inoculated i.p. with syngeneic SL2 lymphoma or P815 mastocytoma on day 0 and treated i.p. with 20,000 units IL-2/day on day 10-14. This treatment is curative for 70-90% of the tumor bearing mice. Peritoneal cells and/or spleen cells were isolated from responding mice at the last day of IL-2 therapy. The in vivo antitumor activity of these cells was tested in Winn Assays (i.p.) and by adoptive transfer (i.v.) into mice injected s.c. with tumor previously. Peritoneal exudate cells isolated on day 14 (PEC14) from mice cured of SL2 tumor were highly effective in Winn Assays. Up to 5 x 10(7) SL2 cells could be eliminated in naive mice when injected i.p. together with 2 x 10(7) PEC14. Adoptive transfer (i.v.) of PEC14, without the addition of IL-2, into mice s.c. injected with SL2 tumor cells 1 or 3 days earlier, also prevented outgrowth of the tumor cells. T cells isolated from the spleens were less effective. Only at E:T ratios of 100:1 and 10:1 was tumor outgrowth inhibited. The adoptive transfer of PEC14 resulted in a long lasting immunity of the recipient mice. Furthermore, it was shown that depletion of both the CD8+ and CD4+ T cells from the suspensions used in the transfer studies, resulted in a significant decrease of tumor inhibition. However, the effect was not abrogated completely. PEC14 isolated from IL-2-treated DBA/2 mice cured of P815 tumor, protected naive mice against P815 tumor at E:T ratio 20:1. Adoptive transfer (i.v.) of these PEC14 into mice bearing s.c. P815 did not have an antitumor effect. In conclusion, low dose i.p. IL-2 therapy predominantly induces locally both CD4+ and CD8+ T cells with a strong antitumor activity in vivo. The potency of the IL-2-induced immunity seems related to the type of tumor used.

Histological analysis of IL-2 induced regression of murine solid SL2-tumors

When DBA/2 mice are inoculated both intraperitoneally (i.p.) and subcutaneously (s.c.) with syngeneic SL2 lymphoma cells and treated i.p. on day 10–14 with 20,000 units IL-2/day, about 50% of the mice reject both the ascitic tumour and the s.c. tumour. During IL-2 therapy large areas of necrosis appear in the solid SL2 tumours between day 12 and 15. Immunohistochemical studies show that only a small number of infiltrating cells is present in the tumours. The percentage of macrophages (MHC-II+)in the tumours is about 1 and the percentage of T-lymphocytes (αβ-TCR+) about 0.5. No differences in the numbers of infiltrating cells are seen in untreated and IL-2 treated tumour bearing mice. The tumoursurrounding infiltrate consists mainly of mononuclear cells: about 50% macrophages, 20% CD8+ cells, and 15% CD4+ cells. No tumour-infiltrating cells were found that express the IL-2 receptor.We conclude that direct cytotoxic activity of tumour infiltrating cells cannot account for the rapid occurrence of necrosis.When L3T4+ cells were eliminated by treating the mice withα-L3T4 monoclonal antibodies before tumor inoculation and treatment with rIL-2, tumor eradication did not occur. So, L3T4+ helper T-cells are essential for IL-2-mediated tumour regression. Exogenous rIL-2 is not directly responsible for the induced tumour regression. A significant stagnation of intratumoural bloodflow is observed after histological analysis; yet it still needs to be determined whether this is the primary cause or consequence of the observed necrosis.

Lymphokine-activated killer-cell-mediated killing of WiDr colon carcinoma cells is inhibited by K562 erythroleukemia cells

Lymphokine activated killer (LAK) activity was induced in human peripheral mononuclear blood cells by human recombinant interleukin-2. Monocytes were required for optimal rapid proliferation of cells with LAK activity. They had no influence on the expression of tumoricidal activity by the LAK cells. The effector cells killed K562 erythroleukemia cells and WiDr colon cells differently, i.e. contact areas with WiDr cells were limited, whereas the contact areas between effector cells and K562 cells were much longer. Using mixtures of hot and cold target cells it was shown that effector cells preferably bind with K562 cells, impeding the binding of WiDr cells. Differences in expression of cytotoxicity of LAK cells against WiDr and K562 cells respectively was also observed after culturing the LAK cells for a relatively longer period. Cytotoxicity against WiDr was maximal at 3-16 days after starting LAK cell generation, whereas cytotoxicity against K562 was kept constantly high for at least 21 days. The addition of biological response modifiers [PHA and anti-CD3 antibody (OKT3)] during the LAK cell induction also had different effects on the expression of LAK activity against WiDr and K562 cells. Whereas PHA, in combination with rIL-2 had no significant influence on the cytotoxicity against WiDr cells, the cytotoxicity against K562 was significantly inhibited. Addition of anti-CD3 antibody diminished the cytotoxicity against WiDr target cells and had no influence on the cytotoxicity against K562 cells.