Reinventing Bone Marrow Transplantation

Champlin et al review a most interesting topic that has emerged recently; namely, the use of nonmyeloablative conditioning regimens to induce immune-mediated graft-vs-leukemia (GVL) effects, or, in a more general sense, graft-vs-tumor (GVT) effects. Their article highlights the potential issues involved in shifting gears from classic high-dose chemotherapy and radiation therapy to more biologically sound, and probably more effective, immune-mediated GVT effects. The rationale for employing immunotherapy mediated by allogeneic donor lymphocytes, which can react against tumor cells as if they were a foreign graft through recognition of cell-surface alloantigens, stems from well-documented GVL effects induced in conjunction with graft-vs-host disease (GVHD).

However, the key event that focused the attention of hematologists and oncologists on the immunotherapeutic potential of allogeneic immunocompetent donor lymphocytes was our observation of a patient with resistant leukemia who had relapsed shortly after receiving maximum tolerated doses of myeloablative chemoradiotherapy. He was completely cured following donor lymphocyte infusion and is now alive and disease free for more than 13 years. This patient, and numerous others who followed him, many of whom have been in complete remission for more than 10 years, clearly documented the feasibility of curing patients with resistant disease, including those who had relapsed after allogeneic bone marrow transplantation (BMT).

A Working Hypothesis

Complete response to donor lymphocyte infusion, despite previous resistance to all alternative anticancer modalities, led to the hypothesis that perhaps Mother Nature’s tool, circulating alloreactive lymphocytes, may be much more effective in eradicating the “last tumor cell” than are maximum tolerated doses of chemoradiotherapy. The consequences of this working hypothesis were self-evident: If myeloablative conditioning, no matter how strong, is unlikely to eradicate the last tumor cell, whereas alloreactive donor lymphocytes can do so, then the latter should be our first choice.

Furthermore, in an article mentioned by Champlin et al but not discussed sufficiently, it was reported that patients resistant to donor lymphocyte infusion may still respond following further amplification of the cytokinetic potential of alloreactive donor lymphocytes with recombinant human interleukin 2 (rIL-2, aldesleukin(Drug information on aldesleukin) [Proleukin]). This observation suggested that the full potential of immunotherapy mediated by donor lymphocytes has not yet been exploited.[1]

Indeed, in a series of recent studies,[Ji, Weiss, and Slavin, unpublished observations] we demonstrated that specific immune lymphocytes may induce curative antitumor responses, even in circumstances in which donor lymphocyte infusions mediated by nonactivated lymphocytes are ineffective. This observation provides further support for the aforementioned working hypothesis.

Although not pertinent to oncology, we have also shown that nonmyeloablative stem-cell transplantation can eradicate genetically abnormal stem cells in a variety of genetic disorders.[2] This suggests that alloreactive lymphocytes of donor origin, operating in a host tolerant of donor alloantigens, can effectively eliminate all host-type hematopoietic stem cells, including genetically abnormal and malignant ones, down to the “last cell.” Thus, abnormal host stem cells can be replaced with normal donor stem cells in patients with a matched donor available, or even in those with a single-locus mismatched donor[3] or a matched unrelated donor.[4]

Furthermore, the possibility that nonmyeloablative stem-cell transplantation may achieve the same goals as allogeneic BMT without subjecting patients to myeloablative conditioning could be further exploited in patients who do not respond to myeloablative chemoradiotherapy following conventional autologous stem-cell transplantation. A safe, second allogeneic bone marrow or blood stem-cell transplant following nonmyeloablative conditioning could provide such patients with a second chance for cure, again, based on the same working hypothesis.[5]. Needless to say, previous studies have shown that, following the failure of primary autologous BMT preceded by myeloablative conditioning, classic myeloablative stem-cell transplantation has high rates of procedure-related toxicity and mortality.

Wider Potential Application of Adoptive Immunotherapy in Solid Tumors

Unquestionably, nonmyeloablative stem-cell transplantation has opened the door for wider application of adoptive allogeneic cell-mediated immunotherapy in a large number of potential clinical indications. This approach also may be able to displace malignant host-type hematopoietic cells and, possibly, solid tumors. If the latter possibility is confirmed, it may lead to the more effective treatment of patients with otherwise incurable disorders at an earlier stage, including patients in poor general condition who, until recently, were denied an allogeneic BMT procedure because of anticipated toxicity and mortality. Furthermore, the possibility of accomplishing successful nonmyeloablative stem-cell transplantation with limited toxicity holds out the promise of a potential cure for the large number of elderly patients who, until recently, were excluded from transplantation programs based on age limit alone.

The advantages and future wider clinical applications of nonmyeloablative stem-cell transplantation appear to be unquestionable. However, one key issue that has not yet been resolved is whether or not a similar approach may be applicable in patients with metastatic solid tumors, for whom no alternative cure is currently available.

Champlin et al fail to mention important experimental work in this field, which strongly suggests the existence of GVT effects against solid tumors. Although the authors cite earlier work in metastatic solid tumors, these studies provide less than convincing evidence of the efficacy of GVT effects. Also, the article omits more conclusive data supporting such effects, particularly the large volume of data in experimental animals with solid tumors (sarcomas and metastatic breast cancer).[6-8]

For example, using Winn-type co-transfer experiments, our group clearly showed that intravenous injection of lethal doses of metastatic breast cancer cells together with syngeneic or semiallogeneic lymphocytes failed to prevent tumor growth in mice, whereas similar inoculation of tumor cells together with allogeneic lymphocytes completely prevented tumor growth in all recipients.[8] Our data also suggest that allogeneic cell therapy may completely prevent hematogenous spread of metastases.[8] However, it still remains to be seen whether the intensity of GVT required to eradicate well-established metastases can be accomplished without severe GVHD. It also is not yet clear whether the targets of alloreactive lymphocytes and GVL- or GVT-specific lymphocytes are different, although indirect evidence suggests that, under specific circumstances, special T-cell clones may be involved.[9]

Use of Specific Immune Donor Lymphocytes

With respect to the use of donor lymphocytes to transfer specific donor immunity to the host, data from our group, also not mentioned by Champlin et al, suggest that donor immunization with hepatitis vaccine can effectively transfer host immunity against the hepatitis B virus, including clearance of a carrier state in patients who were fully tolerant to hepatitis antigens.[10,11] These data, together with even more recent work from our group on the use of specific immune donor lymphocytes, indicate that the use of specific immune cells may further improve the efficacy of GVT and GVL effects.

Potentially promising methods for maximizing GVL and GVT effects while minimizing GVHD include: (1) the use of allogeneic immunocompetent donor lymphocytes, with the option of eliminating them at will; and (2) the use of nonalloreactive donor immunocompetent T-lymphocytes. Rather than employing T-cell–subset depletion to maximize GVL and GVT effects and minimize GVHD, it may be possible to accomplish this goal more easily or more effectively by partially immunizing donor lymphocytes against normal host alloantigens by blocking post-stimulatory signals, since antigen presentation without co-stimulation may be the easiest recipe for inducing transplant tolerance.

Thus, the induction of transplant tolerance appears to be a necessary first step in the acceptance of donor stem cells and, subsequently, donor lymphocytes. This preliminary step prepares the patient for step 2—allogeneic cell therapy with additional donor lymphocyte infusions if required, or activated donor lymphocytes in case more intensive GVL effects are indicated. This two-step process may become a universal approach to enable the durable engraftment of antitumor effector cells of donor origin in a receptive host.

An alternative approach, which we have described previously,[12] involves the possible use of transient GVL effects in patients in whom a minimal residual disease state has been achieved with conventional or high-dose chemotherapy. As has been shown in mice[12] and humans,[13] transient engraftment of lymphocytes may be sufficient to eliminate residual tumor cells, provided that the tumor burden is minimal. Thus, tumor elimination may be accomplished without GVHD, since durable engraftment of immunocompetent donor lymphocytes may be prevented.

Summary

In summary, the future of adoptive allogeneic cell therapy seems quite promising, especially since newly available, innovative immunologic tools can induce host-vs-graft tolerance without displacing the immunohematopoietic compartment of the host. These “patient-friendly” approaches, thus, may provide more effective, less hazardous anticancer immunotherapy mediated by the immune system.