Immunotherapy for cancer is best thought of as part of a broader subject, namely biologic therapy, or the application of biologic-response modifiers (BRMs). These agents can act through one or more mechanisms (1) to stimulate the host’s antitumor response by increasing the number of effector cells or by producing one or more soluble mediators (eg, lymphokines); (2) to decrease host-suppressor mechanisms; (3) to alter tumor cells to increase their immunogenicity or make them more susceptible to damage by immunologic processes; and (4) to improve the host’s tolerance to cytotoxic drugs or radiotherapy (eg, by stimulating bone marrow function with granulocyte colony-stimulating factor or other hematopoietic factors). The first three mechanisms represent manipulation of immunologic processes and are considered immunotherapy. A given BRM may have immunologic and nonimmunologic effects; eg, IFN-α- enhances the expression of TAA on tumor cells and increases NK cell activity, but it also inhibits tumor cell proliferation directly through nonimmunologic mechanisms.

Passive Cellular Immunotherapy

Passive cellular immunotherapy is a term used when activated, specific effector cells are directly infused into a patient and are not induced or expanded within the patient. Early attempts involve reinfusion of the patient’s lymphocytes after expansion in vitro by exposure to IL-2 (T-cell growth factor). These cells are termed lymphokine-activated killer cells (LAK cells). Sometimes the cells are first exposed to phytohemagglutinin, a lymphocyte mitogen, to expand a broad variety of peripheral lymphoid cells. Such approaches are an extension of work in which allogeneic lymphocytes were cross-transfused between tumor patients after attempted immunization with tumor grafts. The availability of purified recombinant IL-2 in large quantities has made the LAK cell plus IL-2 technique feasible, and some melanoma and renal carcinoma patients have shown objective responses.

Because infusion of IL-2 after LAK cell infusions is associated with significant toxicity, variations of these methods are under study. One approach is to isolate and expand populations of lymphocytes that have infiltrated tumors in vivo and thus may have tumor specificity (termed tumor-infiltrating lymphocytes [TILs]). Infusion of TILs permits use of lower levels of IL-2 with equal or greater antitumor effects. TILs can also be genetically modified to express tumoricidal molecules to increase their cytotoxicity.

Another variation of passive cellular immunotherapy is the concurrent use of interferon, which enhances the expression of MHC antigens and TAAs on tumor cells, thereby augmenting the killing of tumor cells by the infused effector cells. However, remissions have been infrequent.

Passive Humoral Immunotherapy

The use of antitumor antibodies as a form of passive immunotherapy (in contrast to active stimulation of the host’s immune system) is at least a century old. Hybridoma technology has increased the potential of this approach to human immunotherapy because it permits in vitro detection and production of monoclonal antitumor antibodies directed against a variety of animal and human neoplasms.

Antilymphocyte serum has been used in chronic lymphocytic leukemia and in T-cell and B-cell lymphomas, resulting in temporary decreases in lymphocyte counts or lymph node size. Some studies of murine monoclonal antibodies against various antigens associated with malignant melanoma and lymphomas have shown significant responses; now “humanized antibodies” are used to avoid an immune reaction against mouse immunoglobulin.

Another variation is conjugation of monoclonal antitumor antibodies with toxins (eg, ricin, diphtheria) or with radioisotopes, so that the antibodies will deliver these toxic agents specifically to the tumor cells. A new approach, using cellular and humoral mechanisms, is the development of bispecific antibodies, which links one antibody reacting with the tumor cell to a second antibody reacting with a cytotoxic effector cell, targeting the latter more specifically to the tumor.

Active Specific Immunotherapy

Approaches designed to induce therapeutic cellular immunity in the tumor-bearing host are more promising than passive immunotherapy techniques. Inducing immunity in a host that failed to develop an effective response in the first place requires special procedures to present the tumor antigens to the host effectors. Intact tumor cells, defined tumor antigens, or general immunostimulants are used.

Autochthonous tumor cells (cells taken from the host) have been used-after irradiation, neuraminidase treatment, hapten conjugation, or hybridization with long-term cell lines in vitro-in kidney carcinoma and malignant melanoma patients, among others. More recently, approaches using tumor cells genetically modified to produce immunostimulatory molecules (including cytokines such as granulocyte-macrophage colony-stimulating factor or IL-2, costimulatory molecules such as B7-1, and allogeneic class I MHC molecules) have been used successfully in animal studies and are being evaluated in human clinical trials.

Allogeneic tumor cells (cells taken from other patients) have been used in patients with acute lymphoblastic leukemia and acute myeloblastic leukemia. Remission is induced by intensive chemotherapy and radiotherapy; irradiated allogeneic tumor cells are then injected with bacille Calmette-Guerin (BCG) vaccine or other adjuvants (see below). Prolonged remissions or improved reinduction rates have been reported in some series, but not in most.

Defined tumor antigen-based vaccines are among the most promising approaches in cancer immunotherapy. One advantage of using defined antigens is that the immunization technique can be readily evaluated for effectiveness because a defined end point is available (ie, measurable responses to a specific peptide). An increasing number of tumor antigens have been unequivocally identified as the target of specific T cells grown from cancer patients. These include antigens that have a normal sequence but are inappropriately expressed in the tumor and antigens derived from genes that have mutated during tumor development (eg, oncogenes). B-cell lymphomas have a unique antigen derived from the variable region of the clonally expressed immunoglobulin sequence (the idiotype). This is unique to the tumor cells but varies among patients.

Cellular immunity (involving cytotoxic T cells) to specific, very well defined antigens can be induced using short synthetic peptides in adjuvant or bound to autologous antigen-presenting cells in vitro (antigen pulsing). These antigen-pulsed, antigen-presenting cells are reintroduced intravenously and stimulate the patient’s T cells to respond to the pulsed peptide antigen. Early results in clinical trials have shown significant responses. Immunization with the custom-synthesized idiotype sequences expressed by the patient’s B-cell lymphoma cells has shown significant response rates.

Antigen-specific immunity can also be induced with recombinant viruses (eg, adenovirus, vaccinia virus) expressing such TAAs as CEA. These antigen-delivery viruses are being tested for antitumor effectiveness.