Lysis by interleukin 2-stimulated tumor-infiltrating lymphocytes of autologous and allogeneic tumor target cells

The immunology and immunotherapy of breast cancer: an update

Adenocarcinomas of the breast behave clinically and epidemiologically in ways that show host resistance factors are important for outcome in addition to grade and stage of malignancy. Immune reactivity to autologous tumors is indicated by the general presence of lymphoid infiltration (LI) and regional lymph node changes; however, these changes predict favorable outcome only in non-metastatic disease. LI is characterized by CD4+ and CD8+ tumor infiltrating lymphocytes reflecting latent cell-mediated immunity (CMI). CMI and humoral immune reactivity have been demonstrated to autologous tumor and a variety of tumor-associated antigens (TAA) have been implicated including CEA, HER-2/neu, MAGE-1, p53, T/Tn and MUC-1. Immune incompetence involving CMI is progressive with the stage of breast cancer and is prognostically significant. Immunotherapy of several types has been designed to address this immunodeficiency and the TAAs involved. Animal models have employed drug therapy, cytokine transfection, vaccines with autologous tumor, cytokines like interferon alpha (IFN-alpha) and interleukin-2 (IL-2), TAA tumor vaccines, and immunotoxins with evidence of tumor regression by immunologic means. Immunotherapy of human breast cancer is a rapidly growing experimental area. Positive results have been obtained with natural IFN and interleukins, particularly in combination strategies (but not with high dose recombinant IFN or IL-2), with autologous tumor vaccine (but not yet with transfected autologous tumor); with a mucin carbohydrate vaccine (Theratope) in a combination strategy (but not with mucin core antigen) and with several immunotoxins. Combination strategies involving immunorestoration, contrasuppression, adjuvant, and immunotoxins are suggested for the future.

Efficient expansion of tumor-infiltrating lymphocytes from solid tumors by stimulation with combined CD3 and CD28 monoclonal antibodies

Combined CD3 and CD28 monoclonal antibodies (mAb) may initiate efficient activation and expansion of tumor-infiltrating lymphocytes (TIL). In this study we compared phenotypical and functional characteristics of TIL from a group of 17 solid human tumors, stimulated either by high-dose recombinant interleukin 2 (rIL-2, 1000 IU/ml) or by a combination of anti-CD3 and anti-CD28 monoclonal antibodies in the presence of low-dose rIL-2 (10 IU/ml). Compared to activation with high-dose rIL-2, stimulation of TIL with CD3/CD28 mAb induced significantly stronger proliferation and yielded higher levels of cell recovery on day 14. Following the CD3/CD28 protocol, expansion of an almost pure population of CD3+ cells was obtained. Whereas CD4+ cells dominated in the first week of culturing, within 4 weeks the CD8+ population increased to over 90%. The specific capacity to kill autologous tumor cells was not increased as compared to the high-dose rIL-2 protocol, but all cultures showed high cytotoxic T cell activity as measured in a CD3-mAb-mediated redirected kill assay. These studies show that combined CD3 and CD28 mAb are superior to rIL-2 with respect to the initiation of expansion of CD8+ cytolytic TIL from solid tumors. Stimulation with specific tumor antigens at a later stage of culturing may further augment the expansion of tumor-specific cytolytic T cells.

Adoptive immunotherapy of advanced melanoma patients with interleukin-2 (IL-2) and tumor-infiltrating lymphocytes selected in vitro with low doses of IL-2

Freshly isolated tumor-infiltrating lymphocytes (TIL) from stage IV melanoma patients were cultured for 2 weeks with low doses of interleukin-2 (IL-2; 120 IU/ml), to select potentially for tumor-specific lymphocytes present in the neoplastic lesion, followed by high doses (6000 IU/ml) to achieve lymphocyte expansion. TIL were serially analyzed for their expansion, phenotype and cytotoxic activity against autologous and allogenic tumor cells. A preferential lysis of autologous melanoma cells was obtained in long-term cultures of 7/13 cases (54%), while the remaining ones showed a major-histocompatibility-complex-unrestricted, lymphokine-activated-killer(LAK)-like activity at the time of in vivo injection. Sixteen patients with metastatic melanoma were infused with TIL (mean number: 6.8 x 10(9), range: 0.35 x 10(9)-20 x 10(9)) and IL-2 (mean dose: 130 x 10(6) IU, range: 28.8 x 10(6)-231 x 10(6) IU); 1 complete and 3 partial responses were observed in 12 evaluable patients (response rate 33%). In all responding patients, injected TIL showed an in vitro preferential lysis of autologous tumor cells, while in no cases were TIL with LAK-like activity associated with a clinical response. The mean autologous tumor cytotoxic activity of TIL at the time of in vivo injection was significantly higher in responding patients in comparison to nonresponding ones, suggesting that a marked and preferential cytolysis of autologous tumor cells is associated with the therapeutic efficacy of TIL.

Immunosuppression derived from human B-lymphoblastoid and melanoma cell lines

Previous work conducted by the authors, using a murine model, suggested that soluble factors secreted by tumor cells suppress lymphocyte responses. To apply this premise to human tumors, the effects of UC729-6 (lymphoblastoid B-cell) and M21-HPB (malignant melanoma) conditioned media (CM) on normal lymphocyte proliferation, as well as on tumor cell growth in autologous CM was studied. The CM was collected at 2-5 day intervals from cultures of UC729-6 and M21-HPB cells in serum-free media. Phytohemagglutinin- and concanavalin A-stimulated mononuclear peripheral blood cells from healthy human donors showed decreased [3H]thymidine ([3H]Tdr) uptake in the presence of each CM when compared with controls. In assays using 100% CM, mitogen stimulation was 68-85% less than that of controls and 40-50% less using 50% CM. The suppression was more pronounced with UC729-6 CM than with M21-HPB CM. In mixed lymphocyte cultures (MLC), addition of 50% CM from either tumor cell line resulted in 40-50% reduction in [3H]Tdr uptake by lymphocytes. Incubation of UC729-6 cells in 5% to 100% of UC729-6 fCM (filter-concentrated) produced a decrease in [3H]Tdr uptake which was directly proportional to the amount of fCM present. In contrast, M21-HPB cell growth in autologous fCM was dependent on cell number, as well as on the amount of fCM used. Treatment of the UC729-6 fCM using acid (pH 4.5), trypsin (100 micrograms/ml), and heat (56 degrees C) did not restore mitogen-stimulated lymphoproliferation. However, the inhibition observed with UC729-6 fCM was partially reversed after dialysis with membranes having M(r) limits of 2.5 x 10(4), 1.5 x 10(4), or 1 x 10(4).

Production and characterization of tumor infiltrating lymphocyte clones derived from B16-F10 murine melanoma

The adoptive transfer of tumor infiltrating lymphocytes (TIL) in conjunction with recombinant interleukin-2 (rIL-2) for the treatment of advanced cancer has recently been under intense investigation. Despite extensive research, the precise surface phenotype of TIL remains to be fully defined. To elucidate this unsolved problem, we established 11 TIL clones derived from rIL-2 expanded TIL obtained from B16-F10 murine melanoma tumors. These clones could be divided phenotypically into four groups: CD8 (+) T-cell clones, natural killer (NK)-cell clones, NK-like CD8 (+) T-cell clones, and double negative T-cell clones. Functionally, CD8 (+) T-cell clones demonstrated specific cytotoxic activity against B16-F10 melanoma cells, whereas NK-cell clones and double negative T-cell clones demonstrated only non-specific cytotoxic activity against NK-sensitive YAC-1 cells. NK-like CD8 (+) T-cell clones showed dual cytotoxic activity. Clones T1 [a CD8 (+) T-cell clone] and T2 [an NK-like CD8 (+) T-cell clone] which had cytotoxic activity against B16-F10 melanoma cells, demonstrated a proliferative response against immunoblotted B16-F10 melanoma antigens, whereas clones T7 (an NK-cell clone) and T10 (a double negative T-cell clone), which had no cytotoxic activity against B16-F10 cells, demonstrated no proliferative response against them. Winn assays revealed that only the CD8 (+) T-cell clone (T1) had an antitumor effect in vivo, whereas the double negative T-cell clone (T10) and NK-like CD8 (+) T-cell clone (T2) stimulated tumor growth in vivo. Adoptive immunotherapy using tumor-specific, highly cytotoxic TIL clones may represent a useful future immunotherapeutic option for the treatment of human tumors.

Activation and in vitro expansion of tumor-reactive T lymphocytes from lymph nodes draining human primary breast cancers

The feasibility of in vitro activation of lymphocytes from the draining lymph nodes (DLN) of breast cancer patients was examined. Lymphocytes isolated from 48 DLN from 12 patients were examined for their proliferative responses to rIL-2, autologous tumor cells, or rIL-2 plus tumor cells. Three general patterns of cellular responses were observed. Cells from some DLN (17%) were unresponsive to any stimuli. Lymphocytes from 52% of the DLN responded moderately to rIL-2 alone. The combination of rIL-2 and tumor antigen had a synergistic effect on the proliferation of cells from 31% of the DLN assayed. Phorbol dibutyrate and ionomycin plus rIL-2 stimulated expansion of DLN lymphocytes by up to 850-fold after 35 days. These expanded cell populations, as well as those stimulated with antigen plus rIL-2, were predominantly CD3+ and CD16- cells, varying in proportions of CD4+ and CD8+ subsets. Both populations were cytotoxic against autologous tumor, MCF-7, and K562 target cells.

Activation and expansion of tumour-infiltrating lymphocytes by anti-CD3 and anti-CD28 monoclonal antibodies

Cytotoxic T lymphocytes from healthy donors can be expanded to high numbers from the peripheral blood using combinations of anti-CD3 and anti-CD28 monoclonal antibodies (mAb). We investigated whether these antibodies could also be used to induce outgrowth of tumour-infiltrating lymphocytes (TIL) from tumour tissue. In the initiation phase of TIL culture immobilized anti-CD3 antibodies together with anti-CD28 mAb and low-dose interleukin-2 induced a rapid expansion of T cells from various human tumour tissues. The cultured cells showed high levels of cytotoxic T lymphocyte activity, but low levels of lymphokine-activated killer cell activity were generated. This study shows that TIL can be efficiently expanded from tumour tissue by combinations of anti-CD3 and anti-CD28 antibodies. This protocol for cell expansion in vitro may substantially reduce the time required to reach sufficient numbers of TIl for re-infusion to the patient.

Cancer patients' lymphocytes contain CD3+ CD4+ cells that proliferate in response to autologous tumor cells in the presence of exogenous low-dose interleukin-2 and autologous accessory cells

To see whether cancer patients possess CD3+ CD4+ lymphocytes able to proliferate in response to autologous tumor cells (Auto-Tu), this lymphocyte subset was isolated either by positive or negative selection, both methods resulting in highly enriched CD4+ populations. Unseparated and isolated CD3+ CD4+ lymphocytes were then assayed for proliferating activity in the presence or absence of various amounts of Auto-Tu, with or without recombinant interleukin-2 (IL-2) (1.5-15 U/ml) and DR+ adherent cells or E- lymphocytes as autologous accessory cells (Auto-AC). Isolated CD3+ CD4+ lymphocytes were stimulated by Auto-Tu alone in only 1 out of 12 cases. CD3+ CD4+ cells failed to proliferate significantly in response to low doses of IL-2 alone but the addition of Auto-Tu caused stimulation in 8 out of 12 cases (67%). The further addition of Auto-AC to Auto-Tu + IL-2 resulted in enhanced response of isolated CD3+ CD4+ lymphocytes in 6 out of 8 cases tested. When reactivities to Auto-Tu in the presence of IL-2 and IL-2 + Auto-AC were considered together, positive responses of CD3+ CD4+ lymphocytes were seen in 11 out of 12 cases (92%). On the other hand, unseparated lymphocytes were stimulated by Auto-Tu alone in none out of 12 cases. Unseparated lymphocytes, however, responded to IL-2 in 11 out of 12 cases; such a response was increased by the addition of Auto-Tu in only 2 cases. Moreover, the IL-2 proliferation of unseparated lymphocytes was suppressed in 4 and in 3 out of 12 cases tested when Auto-Tu or Auto-Tu + Auto-AC were added respectively. These data indicate that lymphocytes of cancer patients contain CD3+ CD4+ cells that are usually unable to proliferate in response to Auto-Tu only. This proliferation, however, occurs when low doses of exogenous IL-2 are present and can be further amplified by the addition of Auto-AC. No response of CD4+ cells is observed in the presence of DR+ Auto-AC + IL-2 except in 2 out of 7 cases tested (28%), suggesting an Auto-Tu-restricted reactivity of CD3+ CD4+ lymphocytes in the majority of cases.

Effect of indomethacin on tumor-infiltrating lymphocytes of a spontaneously developed murine mammary adenocarcinoma

The effect of indomethacin on tumor-infiltrating lymphocytes (TIL) was investigated in a spontaneously developed and weakly immunogenic murine mammary adenocarcinoma (designated JC) in syngeneic immunocompetent BALB/c mice, a tumor model mimicking human disease. Unlike other chemically and virally induced tumors, the expansion of TIL was only possible with an enriched population of lymphocytes, isolated on a discontinuous density gradient then cultured in complete medium containing recombinant human interleukin-2 (rIL-2). The freshly isolated TIL exhibited no cytotoxicity against either the natural-killer-sensitive YAC-1 or the natural-killer-resistant JC cells lines. After culture in rIL-2, the TIL of the JC tumor lysed both YAC-1 and JC. The cytotoxicity of the TIL reached a maximum between the 2nd and 3rd week of culture and decreased thereafter. Antibody- and complement-depletion tests revealed that the cells bearing asialo-GM1 antigen represented the major precursor cells of the cytotoxic TIL, which may explain its nonspecific cytotoxicity. Indomethacin was shown to accelerate the cell proliferation of the rIL-2-activated TIL, but only in the initial 2 weeks of culture and not in later culture. The addition of indomethacin to the rIL-2-containing medium at the beginning of culture resulted in a fast-acting and long-lasting enhancement in cytotoxicity. These results provided a basis for the clinical use of indomethacin, i.e. acceleration in proliferation and augmentation in cytotoxicity. However, the addition of indomethacin at the end of the fourth week after rIL-2 culturing produced neither accelerated proliferation nor augmented cytotoxicity. This study also suggested that a prolonged administration of indomethacin may not be advantageous in clinical trials, since the long-term continuous presence of indomethacin in the culture has resulted in a negative effect on the growth of TIL.