Background. The major obstacles that impair successful outcome after allogeneic hematopoietic stem cell transplantation (HSCT) for hematologic malignancies remain graft-versus-host disease (GvHD) and tumor relapse. Improved survival after allogeneic HSCT therefore requires more effective control of GvHD while preserving graft-versus-tumor (GvT) effects. Methods. Allogeneic parent-into-F1 murine transplant models (BALB/c or C57BL/6 → F1[BALB/c×C57BL/6]) were used to evaluate the interrelation of GvHD and GvT effects targeting tumor-specific antigens or alloantigens on MethA tumor cells. Results. Compared with syngeneic F1-into-F1 controls (F1[H-2b/d] → F1: MethA[H-2d]), significant T cell-mediated GvT effects occurred in both allogeneic transplant models, even in the absence of histoincompatibilities between donor cells and host tumor (BALB/c[H-2d] → F1: MethA[H-2d]). Selective inhibition of type-1 (Th-1/Tc1) immune responses with TAK-603 after HSCT nearly abolished GvHD in both allogeneic transplant models. While GvT effects directed against alloantigens (C57BL/6[H-2b] → F1: MethA[H-2d]) remained unaffected during type-1-immune suppression, GvT effects targeted against tumor-associated antigens (BALB/c[H-2d] → F1: MethA[H-2d]) were not evident. Conclusions. Our data show that GvHD and GvT effects are in principle separable from each other by selective type-1 inhibition in transplantation models with major histocompatibility complex disparities between tumor, host, and donor. In contrast, in situations that only allow for GvT effects that exclusively target tumor-associated antigens (TAAs), type-1 inhibition results in complete abrogation not only of GvHD but also desired GvT reactions. These differences in GvT effects targeting alloantigens or TAAs and their interrelation to GvHD should be considered in future studies aimed at separating GvT reactions from GvHD.

Seventy-one patients with acute myeloid leukemia (AML), most of them (63/71) considered ineligible for conventional allogeneic hematopoietic stem cell transplantation (HSCT), were enrolled into a phase 2 study on reduced-intensity myeloablative conditioning with fractionated 8-Gy total body irradiation (TBI) and fludarabine (120 mg/m2). Patients received mobilized peripheral blood stem cells (n = 68) or bone marrow (n = 3) from siblings (n = 39) or unrelated donors (n = 32). Thirty-six patients received a transplant in complete remission (CR) and 35 had untreated or refractory disease (non-CR). Median patient age was 51 years (range, 20-66 years). Sustained engraftment was attained in all evaluable patients. With a median follow-up of 25.9 months (range, 3.7-61.2 months) in surviving patients, probabilities of overall survival for patients who received a transplant in CR and non-CR were 81% and 21% at 2 years, respectively. Relapse-free survival rates were 78% and 16%. The cumulative incidence of nonrelapse mortality (NRM) in CR patients was 8% at 2 years and beyond but amounted to 37% at 2 years in non-CR patients. Outcome data in this poor-risk population indicate that allogeneic HSCT from related or unrelated donors with 8-Gy TBI/fludarabine conditioning is feasible with low NRM and preserved antileukemic activity in AML patients in first or later CR.

Immunotherapy of B-cell malignancies with genetically engineered human CD8 + natural killer T cells

BACKGROUND Cytokine-induced killer (CIK) cells are ex vivo expanded T cells with co-expression of CD3 and CD56 and NK activity. They have recently been evaluated in a phase I/II clinical trial against malignant lymphoma. Bispecific Ab (bsAb) redirect CIK cells to tumor targets, thus enhancing their cytotoxicity. While bsAb may improve T-cell mediated anti-tumor activity, little is known about the fate of effector cells upon redirection to tumor targets using a bsAb. METHODS Using ex vivo-activated CIK cells, Her2/neu expressing breast and ovarian cell lines and a F(ab')2 Her2/neu x CD3 bsAb, we investigated the anti-tumor activity and the proliferative and apoptotic outcome of CIK cells. RESULTS When redirected to tumor targets with bsAb, there was a significant increase in anti-tumor activity as well as an increase in both CIK cell proliferation and apoptosis. The addition of agonistic Ab against CD28 did not significantly increase proliferation or apoptosis of CIK cells redirected to CD80- and CD86- tumor targets. To attempt to reduce T-cell apoptosis, we incubated CIK cells in the presence of the pan-caspase inhibitor z-VAD-fmk, which led to a partial reduction in T-cell apoptosis without increasing cellular cytotoxicity. DISCUSSION bsAb are effective in redirecting activated T cells to tumor targets and such redirection leads to both T-cell proliferation and apoptosis that are not altered by co-stimulation through CD28. Effector cell apoptosis can be reduced by using a caspase inhibitor but this does not increase CIK cell cytotoxicity.

The way that allogeneic hematopoietic cells are rejected is not completely understood. Regimen-resistant populations, including natural killer (NK) cells and lymphocytes, are thought to mediate the allograft barrier. In this report, the mechanism by which recipient cell populations resist engraftment of purified allogeneic hematopoietic stem cells (HSCs) was examined in mice. To define the immunoregulatory pathways involved in allogeneic hematopoietic cell resistance, HSC transplantations were performed in immune-defective recipients. Recipients were wild-type mice treated with alpha-NK cell antibodies or knockout strain mice lacking expression of CD8, perforin, Fas ligand, or 1 of the following cytokines: tumor necrosis factor alpha, transforming growth factor beta, interferon gamma, interleukin 4, or interleukin 10. Elimination of a single cytotoxic pathway was ineffective in reducing engraftment resistance, although mice treated with a polyclonal antibody that recognizes NK-cell determinants or CD8 expression showed a profound reduction in the engraftment barrier. Posttransplantation chimerism analysis revealed regeneration of host hematopoiesis in some experimental groups. These studies show, for the first time, that elimination of selected cytokines does not alter allogeneic hematopoietic resistance. Furthermore, the chimerism data reinforce the importance of competition for HSC niches in conjunction with immune mechanisms in resistance to long-term HSC engraftment.

In multiple myeloma (MM), both vascular endothelial (VEGF) and basic fibroblast growth factor (bFGF) promote tumor growth and survival. We have used the novel indolinone BIBF 1000 to study effects of simultaneous inhibition of VEGF, FGF and transforming growth factor-beta on MM cells and their interactions with bone marrow stroma cells (BMSCs). Both, in the absence and presence of myeloma-stroma cell contacts, BIBF 1000 abrogated BMSC-derived secretion of interleukin-6 (IL-6). In addition, BIBF 1000 directly induced apoptosis in t(4;14)-positive cell lines as well as in CD138+ marrow cells from patients with t(4;14) myeloma. To a similar extent, BIBF 1000 induced apoptosis in MM.1S and MM.1R cells carrying the translocation t(14;16). In case of MM.1S and other dexamethasone-sensitive t(14;16) cell lines, BIBF 1000 and dexamethasone had additive proapoptotic effects. Induction of apoptosis by BIBF 1000 was associated with inhibition of the mitogen-activated protein kinases (MAPK) pathway in t(4;14) and inhibition of the phosphatidyl-inositol-3 kinase/AKT pathway in t(14;16) cells. Apoptotic effects did not occur in t(4;14)-or t(14;16)-positive MM cells carrying n- or k-Ras mutations. The data provide the rationale for clinical evaluation of this class of targeted kinase inhibitors in MM with focus on defined cytogenetic subgroups.

We have previously reported on the ex vivo generation of cytotoxic effector cells, termed cytokine-induced killer (CIK) cells, that have both in vitro and in vivo antitumor activity in murine models. We now report on our efforts for the large-scale expansion of CIK cells and also present preliminary results from a phase I clinical trial. Nine patients with advanced Hodgkin disease (n = 7) and non-Hodgkin lymphoma (n = 2), all of whom had relapsed after an autologous transplantation, were treated with escalating doses of CIK cells (3 patients at each dose level of 1 × 109, 5 × 109, or 1 × 1010 cells). The CIK cells were produced by culturing unselected cells from steady-state apheresis products with interferon γ, OKT3, and interleukin 2. After 21 days in culture, with the addition of fresh media and interleukin 2 every 3 to 4 days, the median culture was 97% viable (range, 61%–100%), 98% CD3+ (range, 66%–99%), 76% CD8+ (range, 27%–96%), 23% CD4+ (range, 6%–78%), 20% CD3+CD56+ (range, 8%–58%), and <1% CD16+56+ (range, 0.2%–7.7%). The CD3+CD56+ cells have previously been shown to exhibit the most cytotoxic activity. The absolute number of CD3+CD56+ cells typically expanded 290-fold (range, 3- to 4000-fold) under these culture conditions. In vitro cytotoxic activity was measured against a human B-cell tumor line (OCI-Ly8). At a 40:1 effector-target cell ratio, CIK cells killed 32% (range, 2%–69%) of the target cells. A total of 21 infusions were administered to 9 patients. The number of CIK cells infused ranged from 1.0 × 109 to 1.0 × 1010 per treatment. Toxicity was minimal, and there were no immediate adverse reactions to the infusions. Two patients had partial responses, and 2 patients had stabilization of disease: 1 for more than 18 months. Considering that these were heavily pretreated patients with advanced hematologic malignancies, we believe that CIK cells expanded in this fashion may have utility for the treatment of high-risk patients with evidence of minimal residual disease after autologous transplantation.

Cure of hematologic malignancies after allogeneic hematopoietic stem cell transplantation is partially attributable to immunocellular antitumor reactions termed graft-versus-tumor (GvT) effect. GvT effects are heterogeneous with respect to effector cell populations, target antigens, and their interrelation with graft-versus-host disease (GvHD). In the present study, allogeneic parent-into-F1 murine transplantation models (BALB/c or C57BL/6 --> [C57BL/6 x BALB/c]F1) with different tumors derived from either parental strain were used to evaluate tumor-specific GvT effects. Compared with syngeneic F1-into-F1 controls, significant CD8+ T cell-mediated GvT effects occurred in both allogeneic transplantation models, even in the absence of histoincompatibilities between donor cells and host tumor. Identical genetic background of donor and tumor precluded allorecognition of tumor cells, indicating that tumor-associated antigens (TAAs) were targeted. With allowance made for selective major histocompatibility complex (MHC) disparities between donor cells and normal host tissue, GvHD was identified as a driving force for TAA-specific GvT effects. Adoptive transfer of the effector cells into secondary tumor-bearing recipients confirmed sustained antitumor activity and specificity of the T-cell response. The results provide experimental proof of a donor CD8+ T cell-mediated TAA-specific antitumor response in vivo that is driven by GvHD. It may represent one of the mechanisms contributing to GvT effects observed in allogeneic transplant recipients.

Prediction of relapse of acute myeloid leukemia in allogeneic transplant recipients by marrow CD34+ donor cell chimerism analysis

In den 50 Jahren seit den ersten klinischen Transplantationsversuchen hat sich die hämatopoetische Stammzelltransplantation zu einem wichtigen Element der modernen Tumortherapie und der Behandlung nichtmaligner Stammzell-und Immundefekte sowie bestimmter hereditärer Stoffwechselerkrankungen entwickelt. Neue Indikationsgebiete werden erschlossen und Anwendungsgrenzen, bedingt durch Alter, Vorbehandlung und Komorbidität der Patienten, verschieben sich. Möglich wurde dieser Fortschritt durch eine Vielzahl paralleler Entwicklungen auf den Gebieten der Transplantationsimmunologie und Stammzellbiologie sowie der immunsuppressiven und supportiven Therapie.

Conventional chemotherapy is often unsatisfactory in high-risk hematologic malignancies (Carella et al., 1993) while myeloablative chemotherapy followed by allogeneic SCT can produce durable remissions (Forman et al., 1991). Here we studied a novel conditioning regimen including dose-reduced TBI, fludarabine and ATG followed by SCT from an allogeneic donor. Between May, 1999 and December, 2000, 16 patients (14 AML, 1 MDS, 1 CML) refractory to chemotherapy or complicated by severe treatment related events underwent allogeneic SCT from HLA-matched sibling donors (n=11), matched unrelated donors (n=4), or mismatched related donors (n=1). All patients had risk factors precluding them from conventional myeloablative transplantation. Ten of 16 patients had a history of pulmonary aspergillosis.

Myeloma cells express basic fibroblast growth factor (bFGF), an angiogenic cytokine triggering marrow neovascularization in multiple myeloma (MM). In solid tumors and some lymphohematopoietic malignancies, angiogenic cytokines have also been shown to stimulate tumor growth via paracrine pathways. Since interleukin-6 (IL-6) is a potent growth and survival factor for myeloma cells, we have studied the effects of bFGF on IL-6 secretion by bone marrow stromal cells (BMSCs) and its potential reverse regulation in myeloma cells. Both myeloma-derived cell lines and myeloma cells isolated from the marrow of MM patients were shown to express and secrete bFGF. Cell-sorting studies identified myeloma cells as the predominant source of bFGF in MM marrow. BMSCs from MM patients and control subjects expressed high-affinity FGF receptors R1 through R4. Stimulation of BMSCs with bFGF induced a time- and dose-dependent increase in IL-6 secretion (median, 2-fold; P <.001), which was completely abrogated by anti-bFGF antibodies. Conversely, stimulation with IL-6 enhanced bFGF expression and secretion by myeloma cell lines (2-fold; P =.02) as well as MM patient cells (up to 3.6-fold; median, 1.5-fold; P =.002). This effect was inhibited by anti-IL-6 antibody. When myeloma cells were cocultured with BMSCs in a noncontact transwell system, both IL-6 and bFGF concentrations in coculture supernatants increased 2- to 3-fold over the sum of basal concentrations in the monoculture controls. The IL-6 increase was again partially, but significantly, inhibited by anti-bFGF. The data demonstrate a paracrine interaction between myeloma and marrow stromal cells triggered by mutual stimulation of bFGF and IL-6.

To exploit alloreactive T-cell responses known as the graft-versus-tumor effect, allogeneic hematopoietic stem cell transplantation is being explored as experimental therapy in selected solid tumors, including metastatic melanoma. However, donor T-cell responses are often delayed and associated with severe graft-versus-host disease. Posttransplant adoptive immunotherapy using tumor-specific cytotoxic T lymphocytes (CTL) of donor origin might provide immediate graft-versus-tumor effects but not graft-versus-host disease. Therefore, the aim of the current study was to define in vitro conditions for the expansion of allogeneic major histocompatibility complex-matched CTLs targeting melanoma-associated antigens (MAA). The CTLs were generated from peripheral blood mononuclear cells (PBMCs) of HLA-A*0201+ healthy donors by repetitive stimulations with HLA-A*0201-restricted MAA-derived peptides. Melanoma reactivity, as determined by lysis of peptide-pulsed T2 cells and HLA-A2+/Ag+ melanoma cells, was detected using in vitro expanded CTL targeting MAA peptides AAGIGILTV(MT(27-35)), IMDQVPFSV(G(209-2M)), and YMDGTMSQV(T(368-376)). In contrast, FLWGPRALV(MAGE3(271)-(279)) and VLPDVFIRCV(GnT-V(nt38-67)) induced peptide-specific recognition of T2 target cells only, whereas ITDQVPFSV(G(209-217)), KTWGQYWQV(G9(154)), MLLAVLYCL(T(1-9)), and tumor lysate could not induce specific CTLs. Specific cytolytic activity was accompanied by interferon-gamma secretion. Peptide-pulsed dendritic cells were required only for the initial stimulation of CTLs and could be substituted by PBMCs during restimulations. The median expansion rate of CTL was five to six times, regardless of whether dendritic cells or PBMCs were used after the initial stimulation. The results delineate the conditions for effective ex vivo expansion of melanoma-specific CTLs from PBMCs of healthy donors to be used as an adjunct in allogeneic cell therapy of metastatic melanoma.