Non-Engraftment Haploidentical Cellular Immunotherapy for Refractory Malignancies: Tumor Responses without Chimerism

Background: Allogeneic bone marrow transplantation (alloBMT) relies on immunosuppression to control graft-versus-host disease (GVHD) and allow successful engraftment; this happens at the expense of graft versus-tumor (GVT) activity. Advances in hematologic transplantation have prompted the development of effective, less toxic regimens attempting to balance graft-versus-host and GVT immunoreactions Conventional dogma holds that sustained chimerism/engraftment is necessary for prolonged responses. However, our previous study showed long-term response with transient chimerism, suggesting it was not necessary for therapeutic effect. In addition, Deyet al. showed anti-tumor responses with chimerism loss in 9/22 patients after non-myeloablative conditioning and alloBMT [Brit J Haem 128:2005]. In this phase I/II study, we determined recipient response rates with HLA-haploidentical donors, while infusing varying numbers of G-CSF primed cells.

Methods: Forty-one patients with relapsed refractory cancers, significant co-morbidities, and ineligible for standard palliative or curative therapy, received 100-cGy total body irradiation (TBI) and infusion of G-CSF primed mobilized HLA-haploidentical cells. The product contained 1×10⁶−2×10⁸ CD3+ cells/kg. Twenty-nine patients received the highest dose.

Results: The two highest CD3+ infusion cohorts caused a post-infusional cellular graft rejection syndrome. Symptoms resembled engraftment syndrome with cytokine elevations of IL-6, IL-10, IL-8, IFN-g, IL-5, IL-7, MCP-1 and MIP-1b. In the 26 patients with hematological malignancies there were 14-responses, 9-major. Two of six patients with lymphoma are presently free of disease at 74 and 80-months respectively; there were five durable complete-responses and four partial-responses in 13-patients with AML. All responses occurred without the presence of donor chimerism. Two patients who engrafted died; one due to GVHD and the other disease progression. We speculate that the cells responsible for the cytokine-storm, possibly tied to donor cell rejection, altered host tolerance to tumor and allowed for a host immune response against tumor. This theory is further supported in murine models showing that host anti-donor immune responses causing spontaneous or intentionally induced graft-rejection by recipient lymphocyte infusions, gave anti-tumor responses against recipient tumors [Blood 102:2003].

Conclusion: Non-engrafting Haploidentical transplant with minimal myeloablation (100cGy) is a therapeutic strategy appropriate for older patients with significant comorbidities, and it provides a virtual universal donor pool. Side-effects include well tolerated myelotoxicity and an immediate post-infusional steroid responsive immunologic syndrome. TBI of 100cGy followed by HLA-haploidentical transplant is a biologically active therapy for refractory hematologic disease. The responses outside of chimerism are intriguing and may be related to marrow rejection and or a unique type of cytokinestorm. These data further indicate that persistent donor chimerism may not be necessary for durable anti-tumor response. The rejection of donor-derived cells, in concert with elevations of specific cytokines, may represent a novel therapeutic approach for hematological malignancies.