Immunotherapy: the last 25 years

Immunotherapy: the last 25 years

C ANCER TREATMENT REVIEWS 1999; 25: 355–363 Ar ticle No. ctr v. 1998.0135, available online at http://www.idealibr ar on Immunotherapy: the las...

136KB Sizes 0 Downloads 37 Views

C ANCER TREATMENT REVIEWS 1999; 25: 355–363 Ar ticle No. ctr v. 1998.0135, available online at http://www.idealibr ar on

Immunotherapy: the last 25 years T.A. Plunkett and D.W. Miles ICRF Breast Cancer Biology Group, Guy’s Hospital, London SE1 9RT



The past 25 years have witnessed extraordinary advances in our understanding of the immune system. The basis for immune recognition via the major histocompatibility complex (MHC) has been elucidated (1). The structures of the MHC class I (2) and class II (3) molecules have been defined. The mechanisms for antigen processing and presentation are being determined (4, 5). The T-cell receptor has been identified and the basis for T-cell recognition and activation is being unravelled (6Ð9). The mechanisms for the generation of antibody and T-cell receptor diversity have been defined. Monoclonal antibody technology was unknown until 1975 (10), yet now is involved in countless aspects of clinical practice. An ever increasing number of cytokines and chemokines have been identified and their role in the immune response is being determined. The pace of discovery has been breathtaking. Many of these advances have been heralded as offering the potential for novel treatments of human disease, including cancer. The concept of tumour immunotherapy is not new. It is more than 100 years since William Coley observed that tumour regression could be induced by stimulating the immune system with bacterial toxins (11). Interest has waxed and waned in the intervening years, but recently tumour immunotherapy has entered mainstream clinical practice. No review of the past 25 years could hope to be comprehensive; instead we will use a few examples to demonstrate the translation of this research from the laboratory to the clinic.

Cytokines have been used to enhance MHCrestricted and unrestricted tumour cytotoxicity. Interferons have shown potent immunomodulatory effects on the expression of MHC class I and class II, as well as tumour-associated antigens resulting in increased NK and T-cell cytotoxicity (12). It was not until the early 1980s that sufficient amounts of interferons were available for clinical trials. They were originally produced from cell cultures, but recombinant DNA technology soon superseded this method.

Hairy cell leukaemia The most notable early results were seen in the treatment of hairy cell leukaemia (13), a disease previously unresponsive to chemotherapy. Seven patients with progressive disease were treated with 3 ´ 106 IU of partially purified leukocyte interferon daily. Three attained complete bone marrow remission and three showed less than 5% leukaemic cells in the bone marrow aspirate. The blood indices returned to normal in all cases in which they were originally deranged. Remissions lasted up to more than 10 months (13). Further studies demonstrated that only 5Ð10% of patients achieved a complete remission, and newer agents, such as pentostatin, an adenosine deaminase inhibitor, have been demonstrated to prolong relapse-free survival compared to interferon-a (IFNa) (14).

Malignant melanoma T.A. Plunkett and D.W. Miles, ICRF Breast Cancer Biology Group, GuyÕs Hospital, London SE1 9RT 0305-7372/99/060355 +9 $12.00/0

Studies of IFNa have also been performed in patients with malignant melanoma. Immunological © 1999 HARCOURT PUBLISHERS LTD


factors may modify the course of the disease. Evidence for this is found in the profuse lymphocytic infiltrate sometimes seen in secondary lymph nodes and the instances of spontaneous remission or stabilisation of disease. Early reports showed evidence of activity for IFNa in the treatment of metastatic disease (15). However, the overall response rates were no better than the most active cytotoxic drugs. Some studies have suggested a benefit for combination therapy using standard chemotherapy and cytokines in the treatment of metastatic disease, but a lack of well-designed, prospective randomized trials precludes definite recommendations. The North Central Cancer Treatment Group (NCCTG) investigated high doses of IFNa given intramuscularly for 3 months in patients with intermediate-risk or high-risk node-positive stage III resected melanoma (16). This study showed a benefit in relapse free interval in node-positive patients, but the trial results were not significant overall, perhaps because only 160 of 262 patients were node-positive and at high risk. Studies of lower doses of IFNa in patients with intermediaterisk melanoma demonstrated a prolongation of relapse-free survival, but this effect was gradually lost following discontinuation of therapy (17, 18). There was no benefit in overall survival from treatment. Recent evidence suggests a possible benefit for high-dose IFNa as an adjuvant therapy in patients with high-risk malignant melanoma. In a randomized controlled trial, compared to observation alone, adjuvant therapy with IFNa significantly increased 5-year disease-free (26 vs 37%) and overall survival (2.8 vs 3.8 years) in patients with high-risk resected cutaneous malignant melanoma (19). It is the first adjuvant treatment shown to have such an effect in melanoma. The toxicity from the high-dose IFNa was probably excessive; more than two thirds of patients overall required a dose reduction or delay and two patients died from liver failure. Notably, when the original study data were re-analysed taking into account toxicity, overall survival adjusted for quality of life was no longer statistically significant using a two-tailed t-test (20). The results of a confirmatory study of high-dose IFN are awaited with interest; preliminary results suggest that there is a benefit in relapse-free survival, but not in overall survival. The apparent discrepancy between the two trials could result from the fact that many patients in the observation arm were treated with high-dose IFNa at relapse. Further studies of low doses of IFNa given for longer periods or high doses given for shorter periods are required.


Renal cancer In patients with renal cancer, prolonged stabilization of advanced disease and rare spontaneous regressions in the absence of systemic treatment suggested that host immune responses may be important in regulating tumour growth and have led to the study of immunotherapy for this malignancy. The overall proportion of responses to IFNa in over one thousand patients with metastatic renal cancer was 12% (21), although in patients whose predominant site of metastatic disease was pulmonary, the response rate has been reported to be as high as 30% (22). A recent study compared IFNa to medroxyprogesterone acetate in 335 patients with metastatic renal cell cancer. The results demonstrated a 28% reduction in the risk of death, a 12% improvement in 1year survival and a 2.5 month improvement in median survival for patients treated with IFNa (23). However, there was considerable early toxicity from IFNa treatment which, although less apparent by the end of treatment, should be set against the survival benefits. The use of IFNa in the management of early renal cell cancer, and in combination with other agents such as interleukin-2 (IL-2) and 5-fluorouracil, require further investigation. A study of patients with progressive metastatic disease suggested a benefit for combination therapy with IL-2 and IFNa (24). However, the benefits were achieved at the cost of substantial toxicity. The use of IFNg showed no benefit compared to placebo (25). IL-2 affects tumour growth by activating lymphoid cells in vivo without affecting tumour growth directly. The intravenous infusion of IL-2 combined with autologous lymphokine activated killer (LAK) cells resulted in objective responses in over 30% of patients (26). Subsequent studies showed that LAK cells could be omitted from the treatment schedule without affecting the response rate (27). In a multicentre study, 255 patients with metastatic renal cancer were treated with high-dose IL-2 (28, 29): 14% of these patients had complete or partial responses lasting for a median of 23 months. The rate and duration of these responses led the Food and Drug Administration in the USA to approve the regimen as standard therapy for metastatic renal cell carcinoma. However, inpatient monitoring is required, often in an intensive care unit, and there is a 4% incidence of treatment-related death. More recently, similar response rates were obtained with lower dose IL-2 given subcutaneously (27). The toxicity was considerably less, but the duration of the responses is not yet known. New combinations of IL-2, IFNa, and fluorouracil; IFNa and 13-cis-retinoic acid have been tested and shown promising results (30).


CANCER VACCINES The objective of active-specific immunotherapy (ASI) is to produce tumour-specific immunity by combining a non-specific immune adjuvant with a tumour-associated antigen. The advances in the understanding of immunology have coincided with the identification of tumour-associated antigens (TAA) (31, 32). These TAA are potential targets for specific immunotherapy and include viral antigens, mutated proteins and oncogene products.

Tumour cell vaccines In the past 25 years, tumour cell vaccines have moved from experimental animal models to phase III clinical trials. Autologous or allogeneic tumour cells have been used as immunogens in the hope that an immune response to putative tumour antigens might result. Autologous vaccines have the advantage that, by definition, they are MHC-matched with the recipient. However, they require the patient to have surgically accessible disease that will yield sufficient cells to prepare vaccine (50Ð100 ´ 106 cells; generally a mass of 2.5 cm diameter is required). This criterion is most likely to be fulfilled at the primary surgical treatment of cancers. A recent trial of autologous tumour cell vaccination was reported in patients with colon cancer (33). Two hundred and fifty-four patients were randomized to ASI or no further adjuvant treatment. ASI was intradermal vaccination with viable, irradiated autologous tumour cells with Bacille Calmet-Guerin (BCG) as an adjuvant. Patients were treated at weekly intervals for 3 weeks, with a final immunization at 6 months. At a median follow-up of 5.3 years, overall there were no significant benefits in recurrence-free or overall survival. In a sub-group analysis of patients with stage II disease, there was a significant increase in recurrence-free survival. However, this finding is based on a total of only 150 patients, and further investigation is required. The study of Vermoken et al. employed one of only two centres currently able to prepare autologous vaccines; this emphasizes the difficulties in accruing large numbers of patients to such trials. Allogeneic vaccines can be developed from cell lines selected to provide multiple TAA and a broad range of MHC expression. Also allogeneic vaccines are more immunogenic (34); the immune response against the foreign MHC antigens may induce a strong helper response against cross-reacting TAA. The allogeneic cell-based vaccines provide exposure to multiple known tumour antigens, as well as yet


unidentified tumour antigens. Therefore, allogeneic vaccines have been used more widely. Recent insights into antigen presentation and Tcell activation provide further rationales for allogeneic tumour cell vaccines. In animal models, bone marrow-derived antigen presenting cells were able to endocytose tumour antigens and present them to both CD4+ and CD8+ T cells, a process known as Ôcross-primingÕ (35). In this way, a single antigen-presenting cell is able to facilitate the provision of T-cell help to antigen-specific CD8+ T cells. This occurs via CD4+ T-cell derived lymphokines and also through the interaction of CD40 ligand on T cells and CD40 on the APC (7Ð9). This process suggests that the vaccinating tumour cells need not be compatible with the MHC haplotype of the recipient in order to generate a tumour-specific immune response. Allogeneic tumour vaccines have been used in the treatment of malignant melanoma and we will use this tumour type as an example of the advances in the past 25 years. The most extensively studied vaccine is the so-called CancerVax. This is an allogeneic, viable, antigen-enriched melanoma cell vaccine developed from three melanoma cell lines, chosen for their high expression of immunogenic antigens (36). The CancerVax contains an MHC haplotype match with 95% of melanoma patients (37). The vaccination schedule consists of an induction phase with injections every two weeks for five doses in a 2month period. BCG is given as a non-specific immunostimulatory agent with the first two doses. The induction phase is followed by a maintenance phase of injections every four weeks for 1 year, every two months for the second year, and then every three months to five years. There has been minimal toxicity from treatment. Treatment with CancerVax demonstrated impressive prolongation of survival compared with historical controls (38). For example, phase II trials in patients with stage IV disease demonstrated a fiveyear survival of 25% for 157 patients treated with CancerVax, compared to 6% for 1521 historical controls. These studies cannot eliminate selection bias or other unforseen bias, and although a matched-pair analysis also demonstrated a benefit for CancerVax, phase III randomized trials are required. These are now underway, comparing CancerVax to high-dose IFNa as post-surgical treatment for patients with stage III disease, and comparing CancerVax to placebo plus BCG in post-surgical patients with stage IV disease. Other phase III studies using different vaccines are also ongoing. A polyvalent shed antigen vaccine has been prepared from melanoma cell lines (39). Use of the vaccine conferred a survival benefit compared to historical controls, and there was a correlation


between vaccine-induced immune responses and improved clinical outcome (40, 41). The median overall survival time was 3.7 years longer for patients with a strong cellular immune response. The 5-year survival rate was 71% for antibody responders compared to 44% for non-responders. Preliminary results from a phase III study demonstrated that mortality at 2 years in the vaccine-treated group was half that of the placebo-treated group (42). Unfortunately, the trial was interrupted early because of slow patient accrual and the data are based on a total of only 38 patients. Further trials of this vaccine are awaited. Other ongoing trials include use of virus-induced lysates of allogeneic melanoma cells. Interim analysis of a randomized, controlled study of such a vaccine in 250 post-surgical patients demonstrated no clear benefits (43). Attempts have been made to enhance the immunogenicity of tumour cell vaccines by engineering the tumour cells to secrete cytokines. A variety of cytokines have been employed to enhance the immune response either by stimulating a local inflammatory infiltrate or by attracting or activating effector cells. These studies have shown that such modified vaccines are safe, but have not yet demonstrated any obvious clinical benefits (44).

Peptide vaccines Many TAA have now been cloned. As a result, instead of using tumour cells as immunogens, peptides can be synthesized whose sequences correspond to known epitopes of TAA recognised by T cells. The elucidation of the structure of class I and class II MHC, and the definition of critical MHCbinding residues within epitopes, has made it possible to identify putative MHC-binding regions from the DNA sequence of known TAA. The corresponding peptides, with an appropriate adjuvant, could be used as vaccines. Phase I clinical studies of peptide vaccines have been reported in patients with melanoma. These have included an HLA-A1 restricted MAGE3 peptide (45), a combination of three melanoma-associated HLA-A2 restricted peptides and an HLA-A2 restricted gp100 peptide analogue (46), modified to increase the binding affinity to HLA-A2 (47). In all the studies, toxicity was minimal, and immunological and clinical responses were detected. Recently, the modified gp100 peptide was used either alone or followed by high-dose IL-2 (48). The modified peptide induced T-cell responses in 91% of patients. The administration of high-dose IL-2 reduced the frequency of T-cell responses to 16%, yet in these patients a clinical response rate of 42% was


observed, with no responses seen in the patients receiving modified peptide alone. As an IL-2 treatment alone arm was not included, it is not clear whether the modified peptide contributed to the responses seen. However, the results demonstrate the difficulty of equating immunological responses with clinical responses. Peptide vaccines are by definition restricted to use in patients with the appropriate MHC haplotype. Also, at least in animal models, altering the dose, schedule or route of administration can result in tolerance rather than immunity (49).

Carbohydrate antigen vaccines Carbohydrate antigens aberrantly or overexpressed on tumour cells are further targets for immunotherapy. Gangliosides, particularly GM2, have been used as immunogens in patients with cancer (50). Randomized clinical trials are underway comparing high-dose IFNa to immunization with a GM2 conjugate in patients with high-risk melanoma. Expression of the carbohydrate moiety sialyl Tn (STn) is associated with a worse prognosis in colonic (51), gastric (52) and breast cancers (53). Circulating antigen has been detected in gastrointestinal and ovarian malignancies, and raised levels have been associated with a worse prognosis (54). A prospective, randomized clinical trial using STn as a target for active specific immunotherapy in patients with breast cancer has recently been reported (55). All patients were immunized subcutaneously with STn conjugated to keyhole limpet haemocyanin (KLH), with DETOX as an adjuvant on weeks 0, 2, 5 and 9. STn has been detected in the circulation and soluble antigens have been demonstrated to induce tolerance or anergy, rather than an effective immune response. In mice this apparent ÔsuppressorÕ activity can be overcome by pre-treatment with cyclophosphamide to allow active specific immunotherapy (56, 57). To determine its effectiveness in man, patients were randomized to receive, before the first immunisation, either cyclophosphamide 300 mgÐ2 intravenously (IV) on day Ð3 or cyclophosphamide 50 mg orally days Ð14 to Ð3, or no pre-treatment with cyclophosphamide. The treatment had minimal toxicity. All patients generated an antibody response to STn, STn-positive mucin and KLH. The highest antibody titres were in patients pre-treated with intravenous cyclophosphamide. The median survival for the group pretreated with IV cyclophosphamide was significantly longer than that for the other groups (19.7 months vs 12.6 months, P = 0.0176). The patients receiving IV


cyclophosphamide were less likely to have progressive disease, and there was a negative correlation between the growth of measurable tumours and antibody titre to STn. There was no correlation between progression and antibody titres to KLH As there were no differences between the groups in terms of the natural history of their disease or the number and type of previous treatments, the results suggest a therapeutic effect for pre-treatment with IV cyclophosphamide followed by immunisation with STn-KLH. A similar association between antibody titres and survival has been demonstrated in patients with colorectal cancer immunized with STn-KLH (58). A large international trial comparing IV cyclophosphamide and STn-KLH/DETOX with IV cyclophosphamide and KLH/DETOX in the treatment of patients with advanced breast cancer has begun.

Dendritic cell vaccines ÔCellular adjuvantsÕ have been advocated for use with peptide vaccines. Antigen-presenting cells (APC) include macrophages, activated B cells and dendritic cells. Although all these cells are capable of presenting antigen to T cells, dendritic cells are (i) the most efficient APC in the activation of resting T cells, (ii) the major APC for the activation of naive T cells in vivo (59) and (iii) the only APC known to induce antigen-specific CTLs in vivo (60). Dendritic cells (DC) arise from a CD34+ precursor common to granulocytes and macrophages. In undamaged tissue DCs present antigen poorly, but they mature in response to a variety of stressors, resulting in the increased surface expression of MHC and co-stimulatory molecules such as CD80 (B7.1) and CD86 (B7.2). These effects can be reproduced in vitro by coincubation with cytokines such as TNFa and IL-1b (61). The maturation of DCs is completed by interaction with T cells, and is characterized by further expression of co-stimulatory molecules and the synthesis of cytokines (62Ð64). DCs are motile, an important attribute for their role as an APC, allowing migration from peripheral tissues to lymphoid organs. Studies in animals and in man (65) have shown that autologous DCs pulsed with antigen in vitro and then re-infused, can induce immune responses that led to inhibition of tumour growth in vivo. A limitation to these studies had been the generation of sufficient DCs. It is now possible to generate DCs from peripheral blood by using GM-CSF and IL4 (66, 67). In animals, DCs derived from peripheral blood have been shown to migrate from the site of injection to draining lymph nodes (68). These find-


ings suggest that DCs differentiated in vitro from peripheral blood may be useful vehicles for immunotherapy. The first clinical trial testing peptide and tumour cell lysate-pulsed autologous DCs has recently been reported. The pulsed DCs were administered by direct injection into lymph nodes under ultrasound guidance. Clinical responses were noted in five of 16 patients with metastatic melanoma. Two patients had complete responses lasting more than 1 year; all clinical responses were accompanied by antigen-specific skin test reactivity. Notably, two of the five responding patients received tumour cell lysate-pulsed DCs in which the identity of the tumour antigens was not known. Therefore, this approach is potentially applicable to other human cancers lacking well-characterized antigens. A similar approach, particularly if a tumour antigen has not been cloned, is to fuse tumour cells with DCs (69).

Recombinant viral and bacterial vaccines Viruses transfected with cDNA encoding the tumour antigen have also been developed. The virus acts as an adjuvant by altering the intra- and extracellular trafficking of antigen and provides an additional substrate for specific and non-specific immune recognition. In animals, vaccination with tumour cells expressing a model antigen resulted in a negligible immune response. Vaccination with recombinant viral vectors expressing the same antigen resulted in specific cell-mediated immunity and caused tumour regression (70). It is also possible to engineer recombinant viral immunogens that express immunomodulatory molecules (such as CD80 or IL-2) together with a TAA, and this can enhance their immunogenicity (71,72). A recombinant HPV16 and 18 E6- and E7-expressing vaccinia virus has been tested in eight patients with metastatic cervical cancer (73). There was no significant toxicity, but no significant clinical responses either. All patients developed an anti-vaccinia antibody response. Immunization and repetitive boosting with the same recombinant virus can induce a strong immune response to the viral vector itself (74). These responses limit the immunogenicity of the TAA, perhaps by rapidly eliminating the recombinant virus (75). Alternatively, the response to immunodominant epitopes on the vector may suppress those to the weaker determinants of the tumour-associated antigens (76). This problem may possibly be overcome by using different viral vectors that express the same TAA.


DNA-based vaccines A potentially safer and easier method of vaccination is to use naked cDNA without a viral carrier system as an immunogen. The intramuscular injection of naked cDNA resulted in foreign protein expression in mice (77) and non-human primates (78). In mice, immunisation with cDNA encoding the influenza A nucleoprotein led to the generation of specific cellular and humoral responses, and protected the mice from subsequent challenge with the virus (79). Immunization with cDNA encoding the carcinoembryonic antigen (CEA) showed anti-tumour activity (80) as have experiments using cDNA encoding for other TAA (81). This method of vaccination is under further investigation. In animal models of malaria, a combination of viral and cDNA vaccines has shown considerable promise (82). Similar combinations may be effective for tumour immunotherapy.

MONOCLONAL ANTIBODY-BASED THERAPY Monoclonal antibodies were developed in the seventies and those developed against TAAs are being tested in a variety of forms in different clinical settings. Initial studies were hampered by the development of human antibody against the constant regions of xenogeneic monoclonal antibodies. The availability of human and humanized antibodies has revolutionised their use. These antibodies are produced by molecular engineering, including grafting antibody genes (83) and transgenic mice that produce human lgG in response to immunization (84). Antibodies can activate immune effector functions. Antibody that binds to the target antigen can mediate cell killing by complement fixation, opsonization or antibody-dependent cell-mediated cytotoxicity. B cells recognize native antigen and IgM secretion may be Tcell independent. However, long-lasting immunity, involving IgG, IgA or IgE, requires activated T cells (85). The T cell stimulates the B cell not only by secreting cytokines which activate and aid its multiplication and differentiation, but also by means of cognate costimulatory interactions (86). Concomitant treatment with cytokines can enhance the effector function. Bispecific antibodies have been generated that bind a TAA at one antigen-binding site and T cells at the other antigen-binding site. These antibodies can activate T cells adjacent to tumours and enhance cytotoxicity (87). In a randomized trial, 189 colorectal cancer patients who had undergone curative resection for


Dukes C cancer received either monoclonal antibody to the tumour-associated antigen 17Ð1A (a 34-kd glycoprotein of the cell membrane of epithelial cells), or were assigned to an observation arm (88). At a median follow-up of 5 years, the mortality rate and recurrence were reduced by 30% and by 27% respectively in the treatment group compared to controls. By 7 years of follow-up, the survival benefits were maintained (89). Interestingly, the local recurrence rate was not affected, whilst the distant relapse rate was significantly lower in the treatment arm. This suggests that the effect of treatment may be on disseminated tumour cells rather than local occult disease. The overall survival benefits were comparable to that of adjuvant leucovorin-primed 5-fluorouracil, but were associated with minimal toxicity. The two treatments alone or in combination are now being tested in a randomised trial. Encouraging results have also been demonstrated with antibody-based therapy in leukaemias and lymphomas (90). An unconjugated chimeric monoclonal antibody to CD20 was recently licensed for use in relapsed low grade or follicular non-Hodgkins B cell lymphoma (91). In patients with metastatic breast cancer who had received extensive prior anti-cancer therapy, treatment with anti-p185HER2 (Herceptin) produced response rates comparable to third or fourth-line chemotherapy, but with minimal toxicity (92). The antibody binds to HER2/c-erbB2, a transmembrane tyrosine kinase receptor that is overexpressed in approximately 30% of primary breast cancers. These studies have been extended and a multicentre, controlled trial of 469 women with HER2-positive metastatic breast cancer were randomized to receive doxorubicin and cyclophosphamide or paclitaxel (in those women that had received prior anthracycline therapy) with or without Herceptin. At a median follow-up of 25 months there was a significant survival benefit for women receiving concurrent Herceptin (25.4 months versus 20.9 months). Since there was considerable crossover to Herceptin treatment from the control arm of the study, the benefits may have been under-estimated. Further study is required, but the results to date are encouraging. Monoclonal antibodies directed against TAA when conjugated with a therapeutic agent, such as a radioisotope, can selectively deliver therapy to cancer cells. A novel means of drug delivery is by antibodydirected enzyme prodrug therapy (ADEPT), in which an anti-tumour antibody conjugated to an enzyme is administered intravenously (93). The antibody binds to tumour cells, and as a result the enzyme concentration is increased around the tumour compared to normal tissues. A prodrug is then given, and is converted to an active cytotoxic agent adjacent to tumour cells by


the enzyme. A phase I trial using such a system is planned in patients with advanced colorectal cancer (93). The high molecular weight of antibodies (approx mol wt 150 000kDa) can impair their passage into cell aggregates. Single chain Fv antibodies comprise the variable heavy and variable light regions of an IgG molecule linked by a short peptide (approx mol wt 27 kDa). Studies in patients with advanced colorectal cancer demonstrated that single chain Fv antibody specific for CEA can be more sensitive for tumour imaging than computed tomography (94). These antibodies may have greater potential for the delivery of anti-tumour therapy.

CONCLUSIONS The past 25 years have seen tumour immunotherapy move from the laboratory to the clinic and immunotherapy is now an established treatment for certain tumour types, such as renal cell cancer and low grade B cell non-Hodgkins lymphoma. The results of trials already underway may lead to further indications for immunotherapy in the management of patients with cancer. Immunotherapy is beginning to fulfil its much-vaunted early promise. Although advances in our understanding of the immune system have provided opportunities for new therapies, they have also provided insights into the myriad ways in which tumour cells evade attack. Tumour immunotherapy is certainly not a panacea and may not be relevant to every tumour type or perhaps to every patient with a particular cancer. In the future, it is likely that we will define patient or tumour characteristics that predict for potential benefit. Until then, progress depends on continued basic research and well-designed clinical trials to determine the route for the next 25 years.

REFERENCES 1. Zinkernagel R, Doherty P. MHC restricted cytotoxic T cells: studies on the biological role of polymorphic major transplantation antigens determining T cell restriction, specificity, function and responsiveness. Adv Immunol 1979; 26: 51Ð177. 2. Bjorkman P, Saper M, Samraoui B, Bennett W, Strominger J, Wiley D. Structure of the human class I histocompatibility antigen HLA-A2. Nature 1987; 329: 506Ð512. 3. Brown J, Jardetzky T, Gorga J, Stern L, Urban R, Strominger J, Wiley D. Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1. Nature 1993; 364: 33Ð39. 4. Townsend A, Rothbard J, Gotch F, Bahadur G, Wraith D, McMichael A. The epitopes of influenza nucleoprotein recognized by cytotoxic T lymphocytes can be defined with short synthetic peptides. Cell 1986; 44: 959Ð965.

361 5. Lechler R, Aichinger G, Lightstone L. The endogenous pathway of MHC class II antigen presentation. Immunol Rev 1996; 151: 51Ð79. 6. Viola A, Lanzavecchia A. T cell activation determined by T cell receptor number and tunable thresholds. Science 1996; 273: 104Ð106. 7. Schoenberger S, Toes R, van der Voort E, Offringa R, Melief C. T-cell help for cytotoxic T lymphocytes is mediated by CD40CD40L interactions. Nature 1998; 393: 480Ð483. 8. Ridge J, Di Rosa F, Matzinger P. A conditioned dendritic cell can be a temporal bridge between a CD4+ T-helper and a Tkiller cell. Nature 1998; 393: 474Ð478. 9. Bennett S, Carbone F, Karamalis F, Flavell R, Miller J, Heath W. Help for cytotoxic-T-cell responses is mediated by CD40 signalling. Nature 1998; 393: 478Ð480. 10. Kohler G, Milstein C. Continuous cultures of fused cells secreting antibody of pre-defined specificity. Nature 1975; 256: 495Ð7. 11. Coley W. The treatment of malignant tumors by repeated inoculations of erysipelas: with a report of ten original cases. Am J Med Sci 1893; 105: 487Ð511. 12. Kirkwood J. Biologic therapy with interferon alpha and beta: clinical applications-melanoma. In De Vita VJ, Hellman S, Rosenberg S, eds. Biologic therapy of cancer: principles and practice. Philadelphia, PA, Lippincott, 1995; 388Ð411. 13. Quesada J, Reuben J, Manning J, Hersh E, Gutterman J. Alpha interferon for induction of remission in hairy cell leukaemia. N Engl J Med 1984; 310: 510Ð8. 14. Grever M, Kopecky K, Foucar M, et al. Randomised comparison of Pentostatin versus interferon alpha-2 in previously untreated patients with hairy cell leukaemia: an intergroup study. J Clin Oncol 1995; 13: 974Ð979. 15. Nethersell A, Sikora K. Interferons and malignant disease. In Taylor-Papadimitriou J, ed. Interferons: their impact in biology and medicine. Oxford, Oxford University Press, 1985; 127Ð144. 16. Creagan E, Dalton R, Ahmann D, et al. Randomised, surgical adjuvant clinical trial of recombinant interferon alfa-2a in selected patients with malignant melanoma. J Clin Oncol 1995; 13: 2776Ð83. 17. Grob J, Dreno B, Salmoniere dl, Delaunay M, Cupissol D, Guillot B, Souteyrand P, Sassolas B, Cesarini J-P, Lionnet S, Lok C, Chastang C, Boneradi J. Randomised trial of interferon alfa-2a as adjuvant therapy in resected primary melanoma thicker than 1.5 mm without clinically detectable lymph node metastases. Lancet 1998; 351: 1905Ð10. 18. Pehamberger H, Soyer H, Steiner A, et al. Adjuvant interferon alfa-2a treatment in resected primary stage ii cutaneous melanoma. J Clin Oncol 1998; 16: 1425Ð29. 19. Kirkwood J, Strawdermann M, Ernstoff M, Smith T, Borden E, Blum R. Interferon alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma: the Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 1996; 14: 7Ð17. 20. Cole B, Gelber R, Kirkwood J, Goldhirsch A, Barylak E, Borden E. Quality-of-life-adjusted survival analysis of interferon alfa-2b adjuvant treatment of high-risk resected cutaneous melanoma: an Eastern Cooperative Oncology Group study. J Clin Oncol 1996; 14: 2666Ð2773. 21. Wirth M. Immunotherapy for metastatic renal cell carcinoma. Urol Clin N Am 1993; 20: 283Ð295. 22. Neidhart J, Anderson S, Harris J, et al. Vinblastine fails to improve response of renal cancer to interferon alfa-n1: high response rate in patients with pulmonary metastases. J Clin Oncol 1991; 9: 832Ð836. 23. MRC Collaborators. Interferon-alfa and survival in metastatic renal carcinoma: early results of a randomised controlled trial. Lancet 1999; 353: 14Ð17.

362 24. Negrier S, Escudier B, Lasset C, et al. Recombinant human interleukin-2, recombinant human interferon alfa-2a, or both in metastatic renal cell cancer. N Engl J Med 1998; 338: 1272Ð8. 25. Gleave M, Elhilali M, Fradet Y, et al. Interferon gamma-1b compared with placebo in metastatic renal cell carcinoma. N Engl J Med 1998; 338: 1265Ð71. 26. Rosenberg S, Lotze M, Muul L, et al. A progress report on the treatment of 157 patients with advanced cancer using lymphokine-activated killer cells and interleukin-2 or high-dose interleukin-2 alone. N Engl J Med 1987; 316: 889Ð897. 27. Law T, Motzer R, Mazumdar M, et al. Phase III randomized trial of interleukin-2 with or without lymphokine-activated killer cells in the treatment of patients with advanced renal cell cancer. Cancer 1995; 76: 824Ð832. 28. Fyfe G, Fisher R, Rosenberg S, Sznol M, Parkinson D, Louie A. Results of treatment of 255 patients with metastatic renal cell carcinoma who received high-dose recombinant interleukin-2 therapy. J Clin Oncol 1995; 13: 688Ð696. 29. Fyfe G, Fisher R, Rosenberg S, Sznol M, Parkinson D, Louie A. Long-term response data for 255 patients with metastatic renal cell carcinoma treated with high-dose recombinant interleukin-2 therapy. J Clin Oncol 1996; 14: 2410Ð11. 30. Motzer R, Bander N, Nanus D. Renal-cell carcinoma. N Engl J Med 1996; 335: 865Ð875. 31. Boon T, Gajewski T, Coulie P. From defined human tumor antigens to effective immunization? Immunol Today 1995; 16: 334Ð336. 32. Van Pel A, Van der Bruggen P, Coulie P, Brichard V, Lethe B, Van den Eydne B, Uyttenhove C, Renauld J, Boon T. Genes coding for tumor antigens recognized by cytolytic T lymphocytes. Immunol Rev 1995; 145: 229Ð250. 33. Vermoken J, Claessen A, van Tinteren H, Gall H, Ezinga R, Meijer S, Scheper R, Meijer C, Bloemena E, Ransom J, Hanna M, Pinedo H. Active specific immunotherapy for stage II and stage II human colon cancer: a randomised trial. Lancet 1999; 353: 345Ð50. 34. Toes R, Blom R, van der Voort E, et al. Protective antitumour immunity induced by immunization with completely allogeneic tumor cells. Cancer Res 1996; 56: 3782Ð3787. 35. Huang A, Golumbek P, Ahmadzadeh M, et al. Role of bone marrow derived cells in presenting MHC class I-restricted tumor antigens. Science 1994; 264: 961Ð5. 36. Conforti A, Ollila D, Kelley M. Update on active specific immunotherapy with melanoma vaccines. J Surg Oncol 1997; 66: 55Ð64. 37. Hoon D, Okamoto T, Wang H, et al. Is the survival of melanoma patients receiving polyvalent melanoma cell vaccine linked to the human leukocyte antigen phenotype of patients? J Clin Oncol 1998; 16: 1430Ð37. 38. Chan A, Morton D. Active immunotherapy with allogeneic tumour cell vaccines: present status. Semin Oncol 1998; 25: 611Ð22. 39. Bystryn J, Oratz R, Henn M, et al. Preparation and characterisation of a polyvalent human melanoma antigen vaccine. J Biol Response Mod 1986; 5: 211Ð224. 40. Bystryn J, Oratz R, Henn M, et al. Relationship between immune response to melanoma vaccine and clinical outcome in stage II malignant melanoma. Cancer 1992; 69: 1157Ð1164. 41. Miller K, Abeles G, Oratz R, et al. Response to immunization to a polyvalent melanoma vaccine. Cancer 1995; 75: 495Ð502. 42. Bystryn J, Oratz R, Shapiro R, et al. Phase III, double-blind, trial of a shed polyvalent melanoma vaccine in stage III melanoma. Proc Am Soc Clin Oncol 1998; 17: 434a (abstr). 43. Wallack M, Sivanandham M, Balch C, et al. A phase III randomized, double-blind multi-institutional trial of vaccinia melanoma oncolysate-active specific immunotherapy for patients with stage II melanoma. Cancer 1995; 75: 34Ð42.

T.A. PLUNKETT AND D.W. MILES 44. Nanni P, Forni G, Lollini P-L. Cytokine gene therapy: hopes and pitfalls. Annals Oncol 1999; 10: 261Ð66. 45. Marchand M, Weynants P, Rankin E, et al. Tumor regression responses in melanoma patients treated with a peptide encoded by MAGE3. Int J Cancer 1995; 63: 883Ð5. 46. Jager E, Ringhoffer E, Dienes H, et al. GM-CSF enhances immune responses to melanoma peptides in vivo. Int J Cancer 1996; 67: 54Ð62. 47. Parkhurst M, Salgaller M, Southwood S, et al. Improved induction of melanoma-reactive CTL with peptides from the melanoma antigen gp100 modified at HLA-A*0201-binding residues. J Immunol 1996; 157: 2539Ð2548. 48. Rosenberg S, Yang J, Schwartzentruber D, et al. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma. Nature Med 1998; 4: 321Ð7. 49. Toes R, Blom R, Offringa R, Kast W, Melief C. Functional deletion of tumor-specific CTLs induced by peptide vaccination can lead to the inability to reject tumours. J Immunol 1996; 156: 3911Ð3918. 50. Livingston P. Ganglioside vaccines with emphasis on GM2. Semin Oncol 1998; 25: 636Ð645. 51. Itzkowitz S, Bloom E, Kokal W, Modin G, Hakamori S, Kim Y. Sialosyl-Tn: a novel mucin antigen associated with prognosis in colorectal cancer patients. Br J Cancer 1990; 66: 1960Ð1966. 52. Chun M, Terata N, Kodami M, Jancic S, Hosokawa Y, Hattori T. Expression of sialyl-Tn antigen is correlated with survival time of patients with gastric carcinomas. Eur J Cancer 1993; 29A: 1820Ð1823. 53. Kinney A, Sahin A, Vernon S, Frankowski R, Annegers J, Hortobyagi G, Buzdar A, Frye D, Dhingra K. The prognostic significance of sialyl-Tn antigen in women with breast carcinoma treated with adjuvant chemotherapy. Cancer 1997; 80: 2240Ð2249. 54. Kobayashi H, Terao T, Kawashima Y. Serum sialyl-Tn as an independent predictor of poor prognosis in patients with epithelial ovarian cancer. J Clin Oncol 1992; 10: 95Ð101. 55. MacLean G, Miles D, Rubens R, Reddish M, Longenecker B. Enhancing the effect of THERATOPE STn-KLH cancer vaccine patients with metastatic breast cancer by pre-treatment with low-dose intravenous cyclophosphamide. J Immunother 1996; 19: 309Ð316. 56. Fung P, Madej M, Koganty K, Longenecker B. Active specific immunotherapy of a murine mammary adenocarcinoma using a synthetic tumour-associated glycoconjugate. Cancer Res 1990; 50: 4308Ð4314. 57. Fung P, Longenecker B. Specific immunosuppressive activity of epiglycanin, a mucin-like glycoprotein secreted by a murine adenocarcinoma (TA3-HA). Cancer Res 1991; 51: 1170Ð1176. 58. MacLean G, Reddish M, Koganty R, Longenecker B. Antibodies against mucin-associated sialyl-Tn epitopes correlate with survival of metastatic adenocarcinoma patients undergoing active specific immunotherapy with synthetic STn vaccine. J Immunother Immunology 1996; 19: 59Ð68. 59. Steinman R. The dendritic cell and its role in immunogenicity. Ann Rev Immunol 1991; 9: 271Ð296. 60. Porgador A, Gilboa E. Bone marrow-generated dendritic cells pulsed with a class I-restricted peptide are potent inducers of cytotoxic T lymphocytes. J Exp Med 1995; 182: 255Ð260. 61. Lutz M, Girolomoni G, Ricciardi-Castagnoli P. The role of cytokines in functional regulation and differentiation of dendritic cells. Immunobiology 1996; 195: 431Ð455. 62. Cella M, Scheidegger D, Palmer-Lehmann K et al. Ligation of CD40 on dendritic cells triggers production of high levels of IL12 and enhances T cell stimulatory capacity: T-T help via APC activation. J Exp Med 1996; 184: 747Ð752.

IMMUNOTHERAPY: THE LAST 25 YEARS 63. Heufler C, Koch F, Stanzl U, et al. Interleukin-12 is produced by dendritic cells and mediates TH1 development as well as interferon gamma production by TH1 cells. Eur J Immunol 1996; 26: 659Ð668. 64. Winzler C, Rovere P, Resigno M, et al. Maturation stages of mouse dendritic cells in growth factor-dependent long-term cultures. J Exp Med 1997; 185: 317Ð328. 65. Hsu F, Benike C, Fagnoni F, et al. Vaccination of patients with B cell lymphoma using autologous antigen-pulsed dendritic cells. Nature Med 1996; 2: 52Ð58. 66. Romani N, Lanzavecchia A. Proliferating dendritic cell progenitors in human blood. J Exp Med 1994; 180: 83Ð93. 67. Sallusto F, Lanzavecchia A. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony stimulating factor plus interleukin-4 and downregulated by tumour necrosis factor alpha. J Exp Med 1994; 179: 1109Ð1118. 68. Barratt-Boyes S, Watkins S, Finn O. In vivo migration of dendritic cells differentiated in vitro: a chimpanzee model. J Immunol 1997; 158: 4543Ð47. 69. Gong J, Chen D, Kashiwaba M, Kufe D. Induction of antitumour activity by immunization with fusions of dendritic and carcinoma cells. Nature Med 1997; 3: 558Ð561. 70. Wang M, Bronte V, Chen P, Gritz L, Panicali D, Rosenberg S, Restifo N. Active immunotherapy of cancer with a nonreplicating recombinant fowlpox virus encoding a model tumorassociated antigen. J Immunol 1995; 154: 4685Ð4692. 71. Bronte V, Tsung K, Rao J, Chen P, Wang M, Rosenberg S, Restifo N. IL-2 enhances the function of recombinant poxvirus-based vaccines in the treatment of established pulmonary metastases. J Immunol 1995; 154: 5482Ð5492. 72. Chamberlain R, Carroll M, Bronte V, Warren S, Yang J, Nishimura M, Moss B, Rosenberg S, Restifo N. Costimulation enhances the active immunotherapy effect of recombinant anti-cancer vaccines. Cancer Res 1996; 56: 2832Ð2836. 73. Borysiewicz L, Fiander A, Nimako M, et al. A recombinant vaccinia virus encoding human papillomavirus types 16 and 18, E6 and E7 proteins as immunotherapy for cervical cancer. Lancet 1996; 347: 1523Ð1527. 74. Murata K, Garcia Sastre A, Tsuji M, Rodrigues M, Rodriguez D, Rodriguez J. Characterization of in vitro primary and secondary CD8+ T cell responses induced by recombinant influenza and vaccinia viruses. Cellular Immunol 1996; 173: 96Ð107. 75. Battegay M, Moskophidis D, Waldner H, Brundler M, FungLeung W, Mak T, et al. Impairment and delay of neutralizing antiviral antibody responses by virus-specific cytotoxic T cells [published erratum appears in J Immunol 1994;152:1635]. J Immunol 1993; 151: 5408Ð15. 76. Deng Y, Yewdell J, Eisenlohr L, Bennink J. MHC affinity, peptide liberation, T cell repetoire, and immunodominance all contribute to the paucity of MHC class-I restricted peptides recognized by anti-viral CTL. J Immunol 1997; 158: 1507Ð15. 77. Wolff J, Malone R, Williams P, Chong W, Acsadia G, Jani A, Felgner P. Direct gene transfer into muscle muscle in vivo. Science 1990; 247: 1465Ð1468.

363 78. Jioa S, Williams P, Berg R, Hodgeman B, Liu L, Pepetto G, Wolff J. Direct gene-transfer in non-human primate myofibres in vivo. Human Gene Ther 1992; 3: 21Ð33. 79. Ulmer J, Deck R, DeWitt C, Friedman A, Donnelly J, Liu M. Protective immunity by intramuscular injection of low doses of influenza virus DNA vaccines. Vaccine 1994; 12: 1541Ð1544. 80. Conry R, Lobuglio A, Loechel F, Moore S, Sumerel L, Barlow D, Curiel D. A carcinoembryonic antigen polynucleotide vaccine has in vivo antitumour activity. Gene Ther 1995; 2: 59Ð65. 81. Graham R, Stewart L, Peat N, Beverley P, TaylorPapadimitriou J. MUC1-based immunogens for tumour therapy: development of murine model systems. Tumour Targeting 1995; 1: 211Ð221. 82. Schneider J, Gilbert S, Blanchard T, et al. Enhanced immunogenicity for CD8+ T cell induction and complete protective efficacy of malaria DNA vaccination by boosting with modified vaccinia virus Ankara. Nature Med 1998; 4: 397Ð402. 83. Winter G, Milstein C. Man-made antibodies. Nature 1991; 349: 293Ð299. 84. Jakobovits A, Grenn L, Hardy M, et al. Production of antigenspecific human antibodies from mice engineered with human heavy and light chain YACs. Ann NY Acad Sci 1995; 764: 525Ð535. 85. Miller J. Lymphocyte interactions in immune responses. Int Rev Cytol 1972; 33: 77Ð130. 86. Clark E, Ledbetter J. How B and T cells talk to each other. Nature 1994; 367: 425Ð428. 87. Renner C, Jung W, Sahin U, et al. Cure of xenografted human tumors by bispecific monoclonal antibodies and human T cells. Science 1994; 264: 833Ð835. 88. Riethmuller G, Schneider-Gadicke E, Schlimok G, et al. Randomised trial of monoclonal antibody for adjuvant therapy of resected DukesÕ C colorectal carcinoma. Lancet 1994; 343: 1177Ð1183. 89. Riethmuller G, Holz E, Schlimok G, et al. Monoclonal antibody therapy for resected Dukes C colorectal cancer:seven year outcome of a multicentre randomized trial. J Clin Oncol 1998; 16: 1788Ð94. 90. Kaminski M, Zasadny K, Francis I, et al. Iodine-131-anti-B1 radioimmunotherapy for B-cell lymphoma. J Clin Oncol 1996; 14: 1974Ð1981. 91. Maloney D, Press O. New treatments for non-Hodgkins lymphoma:monoclonal antibodies. Oncol 1998; 12: 63Ð76. 92. Baselga J, Tripathy D, Mendelsohn J, Baughman S, Benz C, Dantis L, Sklarin N, Seidman A, Hudis C, Moore J, Rosen P, Twaddell T, Henderson I, Norton L. Phase II study of weekly intravenous recombinant humanized anti-p185HER2 monoclonal antibody in patients with HER2/neu-overexpressing metastatic breast cancer. J Clin Oncol 1996; 14: 737Ð744. 93. Leonard P, Begent R. Immunotherapy options for colorectal cancer. Oncol in Practice 1997; 8Ð11. 94. Begent R, Verhaar M, Chester K, et al. Clinical evidence of efficient tumour targeting based on single chain Fv antibody selected from a combinatorial library. Nature Med 1996; 2: 979Ð983.