Methods for the expansion of dendritic cells and other antigen-presenting cells are not as effec- tive as those for NK cells. However, it is possible to isolate such cells from both neoplasms and normal tissues and manipulate them in cell culture. In experimental systems, bone marrow–de- rived dendritic cells pulsed with synthetic peptides as epitopes presented on the surface of neo- plastic cells have been used to elicit protective and therapeutic immunity to neoplasms in mice (Mayordomo et al., 1995). Phagocytosis of apoptotic cells from neoplasms by antigen-present- ing cells (Henry et al., 1999) and fusion of dendritic and neoplastic cells (Gong et al., 1997) have also been methods tried with some success in experimental animals. A unique protocol that may have direct application in the human is the immunization of bone marrow transplantation donors with tumor antigens and the use of the transplant as adoptive therapy against the established
Figure 19.35 A diagram of the preparation of human adherent lymphokine-activated killer (A-LAK) cells. Condition medium is obtained from 24-hour incubation cultures and is added back to A-LAK cells at a final concentration of 50% (v/v) in fresh IL-2–containing medium. A-LAK cells are grown in culture for
2 to 3 weeks, allowing for significant expansion of the cell population to the numbers utilized in therapy
(see text). (Adapted from Whiteside and Herberman, 1990, with permission of the authors and publisher.)
neoplasm (Hornung et al., 1995). In fact, graft-versus-leukemia reactions can be induced in pa- tients with hematological cancers by allogeneic stem cell and bone marrow transplantation and also have shown some success with renal cell carcinoma (Childs et al., 2000). In theory, such technologies may be an ideal way to produce dendritic cells specific for antigens in neoplasms (Björck, 1999; Fong and Engleman, 2000), but there are still a number of logistical problems to overcome.
Cytokine Immunotherapy of Neoplasia
The known number of hormone-like proteins affecting the immune system has increased expo- nentially during the last two decades. More than 18 interleukins are presently known, as well as a variety of other proteins directly affecting cells of the immune system (Table 19.4). Tumor necrosis factor and its family of ligands (Chapter 17) have already been discussed; below are several other important cytokines that have found some place in immunotherapy.
Of the 18 or more known interleukins, the one that has been found most effective in the therapy of neoplasia is interleukin-2 (IL-2). As noted above, IL-2, the T-cell growth factor, has been used in conjunction with cellular immunotherapy as well as by itself. In the latter instance, responses
Figure 19.36 Diagrammatic model of immunotherapeutic interaction of LAK cells and TIL on a target neoplasm. TIL and LAK cells are obtained from the autologous neoplasm and peripheral blood of the pa- tient, expanded as described above, and then reinoculated into the patient in the presence of interleukin-2 administered simultaneously. On the left, TIL are removed from fresh neoplasm obtained at surgery from the patient, expanded, and then reinfused together with interleukin-2 with subsequent reinfiltration of the target neoplasm. On the right, NK and T cells in peripheral blood are removed from leukapheresis and the LAK cells after expansion infused into the patient along with interleukin-2, many of the cells returning to the neoplasm where they in turn produce interferon-γ (IFN) and tumor necrosis factor (TNF). (Adapted from Atzpodien and Kirchner, 1990, with permission of the authors and publisher.)
of up to 40% have been seen (Oleksowicz and Dutcher, 1999) in metastatic renal cell carcinoma, while only a 13% response may be seen in the therapy of melanoma by IL-2 (cf. Bishop, 1996). IL-2 therapy, like a number of other immunotherapies, also has profound toxic effects in pa- tients, probably because its effects are far more diverse than simply those of NK cells (Janssen et al., 1994). It should be noted that in a very small percentage of cases, 6% to 15% (Figlin et al.,
1997; Oleksowicz and Dutcher, 1999), a complete remission was achieved, which remained for a number of years or as long as the study continued. Such effects are reminiscent of the “sponta- neous” disappearance of neoplasms with no obvious explanation. The difficulty is determining the exact conditions within the individual that give rise to such complete responses. In addition to IL-2, IL-12 (Okuno et al., 1996) and IL-10 (Kundu and Fulton, 1997) have shown significant immunotherapeutic effects in both human and experimental situations.
As implied in the discussion above concerning transfection of genes into cells, attempts have also been made to transfect cytokine genes into neoplastic cells in vitro into the autologous host with the intention that such cells would stimulate an immune response in their immediate area, such a response being then translated to the primary and metastatic neoplasm of the host (Bubeník, 1993; Belli et al., 1997; Leong et al., 1997). Similar rationales have been utilized for transfecting hematopoietic growth factors, specifically granulocyte-macrophage colony-stimu- lating factor, with similar intent (Bubeník, 1999).
The interferons are a family of glycoproteins made up of between 145 and 166 amino acids and having molecular weights ranging between 17,000 and 25,000. There are at least three different forms—alpha, beta, and gamma. Alpha interferon is produced primarily by activated leukocytes and appears within 4 to 6 hours after viral stimulation. Beta interferon is also rapidly induced by viral stimulation, but from fibroblasts. Gamma interferon is produced by lymphocytes and monocytes 2 to 3 days after these cells have been stimulated by exposure to antigens or mito- gens. As noted earlier, interferon gamma activates macrophages and NK cells as well as induc- ing the expression of MHC antigens. Of these interferons, interferon alpha has been shown to have the most efficacy in treatment of a variety of neoplasms. Results of a large series is seen in Table 19.16. As noted from the table, while relatively large doses are required for effectiveness, some neoplasms—such as hairy cell leukemia, chronic myelogenous leukemia, and endocrine pancreatic neoplasms—appear to have a very reasonable response rate. As with other immuno- therapeutic modalities, certain toxicities are seen with this type of therapy, many of them imitat- ing a flu-like complex, but hematological, renal, and gastrointestinal toxicities have also been
noted (Hansen and Borden, 1992). In general, however, interferon alpha has now become ac- cepted as an immunotherapeutic agent not only for neoplasia but also for viral and other inflam- matory diseases (Gutterman, 1994).
The Future of the Immunotherapy of Neoplasia
As may be noted from the length of this chapter, the immunobiology of the host–tumor relation- ship is one of the most active areas of investigation directed toward an ultimate effective therapy of neoplasia. However, a number of considerations must be taken into account if one is to be successful in utilizing the immunobiological host–tumor relationship for the therapy of cancer. Some of these are listed in Table 19.17. Of these various considerations, the selection of patients for immunotherapy on the basis of their immunological competence as well as the immunobiol- ogy of the neoplasm itself will be a major factor. Of equal or of greater importance is the re- moval by standard methods of the majority of the neoplasm from the host by surgery, radiotherapy, or chemotherapy. Immunotherapy has as its greatest potential the elimination of micrometastases and small masses of neoplastic cells that are completely available to the im- mune system. However, the patient must not be in an immunocompromised state when active and passive immunotherapy is undertaken. In fact, establishment of a normal immunocompetent state in a tumor-bearing host may be one of the greatest hurdles to overcome if immunotherapy is to be successful. As explained below, the immunobiology of the host–tumor relationship tends toward immunosuppression of the host as a result of the continued extension of the deterioration of the host–tumor relationship, as noted in Chapter 17 regarding the phenomenon of cachexia.