Expression of MUC-1 epitopes on normal bone marrow: implications for the detection of micrometastatic tumor cells.

PURPOSE: The expression of the carcinoma-associated mucin MUC-1 is thought to be restricted to epithelial cells and is used for micrometastatic tumor cell detection in patients with solid tumors, including those with breast cancer. Little is known, however, about the expression of MUC-1 epitopes in normal hematopoietic cells. MATERIALS AND METHODS: MUC-1 expression was analyzed by flow cytometry and immunocytology on bone marrow (BM) mononuclear cells and purified CD34+ cells from healthy volunteers, using different anti-MUC-1-specific monoclonal antibodies. In addition, Western blotting of MUC-1 proteins was performed. RESULTS: Surprisingly, 2% to 10% of normal human BM mononuclear cells expressed MUC-1, as defined by the anti-MUC-1 antibodies BM-2 (2E11), BM-7, 12H12, MAM-6, and HMFG-1. In contrast, two antibodies recognizing the BM-8 and the HMFG-2 epitopes of MUC-1 were not detected. MUC-1+ cells from normal BM consisted primarily of erythroblasts and normoblasts. In agreement with this, normal CD34+ cells cultured in vitro to differentiate into the erythroid lineage showed a strong MUC-1 expression on day 7 proerythroblasts. Western blotting of these cells confirmed that the reactive species is the known high molecular weight MUC-1 protein. CONCLUSION: Our data demonstrate that some MUC-1 epitopes are expressed on normal BM cells and particularly on cells of the erythroid lineage. Hence the application of anti-MUC-1 antibodies for disseminated tumor cell detection in BM or peripheral blood progenitor cells may provide false-positive results, and only carefully evaluated anti-MUC-1 antibodies (eg, HMFG-2) might be selected. Furthermore, MUC-1-targeted immunotherapy in cancer patients might be hampered by the suppression of erythropoiesis.

A new look at the role of p53 in leukemia cell sensitivity to chemotherapy.

A gene encoding the p53 val135 mutant, which assumes mutant conformation at 38.5 degrees C and wild-type conformation at 32.5 degrees C, was introduced into p53-deficient K562 myeloid leukemia cells. Forced expression of wild-type, but not mutant, p53 resulted in growth arrest, accumulation of p21 and Bax proteins, and delayed cell death. Wild-type p53 enhanced the cytotoxic effects of some drugs and attenuated those of others. Compared with wild-type p53, mutant p53 induced much stronger sensitization to drug cytotoxicity. This occurred in the absence of effects on cell cycle progression or activation of several p53 target genes. Although both mutant and wild-type p53 induced changes of immunophenotype, no specific pattern of differentiation was associated with enhanced chemosensitivity. Thus, (1) induction of growth arrest and activation of p53 target genes such as p21 and bax are linked to the wild-type conformation of p53; (2) p53 induces immunophenotypic changes of myeloid leukemia cells suggestive of multidirectional differentiation in a conformation-dependent manner; and (3) (so-called) mutant p53 induces chemosensitization in the absence of effects on cell cycle progression, activation of bax, p21, gadd45 and mdm-2, or a specific pattern of differentiation; and (4) chemosensitization mediated by wild-type p53 may be masked by transcription-dependent induction of growth arrest.

Differential expression of mast cell growth factor receptor (c-kit) by peritoneal connective tissue-type mast cells and tissue culture-derived mast cells.

The peritoneal cavity of the mouse is a major source of connective tissue-type mast cells (CTMCs). Flow cytometric analysis using biotinylated recombinant murine mast cell growth factor (rMuMGF) showed that 1 to 3% of the cells in the peritoneal cavity exhibited MGF receptor (MGFR) (or c-kit). CTMCs were the only cell types expressing MGFR in the peritoneal cavity, and every one of them expressed MGFR. More than half the peritoneal CTMCs retained the potential to proliferate in the presence of recombinant murine interleukin 3 (rMuIL-3), rMuIL-4, and rMuMGF and gave rise to pure, alcian blue-positive "mast cells," which actively expressed c-kit transcripts and MGFR. Flow cytometric analysis and receptor assay carried out at 4 degrees C showed that the number of MGFRs on culture-derived mast cells (CDMCs) was one-third that of peritoneal CTMCs (6 x 10(4) vs. 1.8 x 10(5) MGFR/cell). At 37 degrees C, the total number of membrane MGFRs detected in CDMC was two to three times more than that detected at 4 degrees C, indicating that nearly 70% of total MGFR in CDMCs, compared with only 40% in peritoneal CTMCs, existed as "cryptic sites" unable to interact with exogenous ligand at 4 degrees C. Thus, diminished expression of MGFR is one of the phenotypic characteristics associated with CDMCs.

Acute toxic effects of 3'-azido-3'-deoxythymidine (AZT) on normal and regenerating murine hematopoiesis.

The immediate hematopoietic response to 3'-azido-3'-deoxythymidine (AZT) was characterized in an unperturbed and regenerating mouse marrow model to identify the in vivo hematopoietic targets of AZT and test whether AZT toxicity is dependent on the proliferative activity of the hematopoietic targets. B6D2F1 mice received intravenous (IV) bolus injections of 30, 60, 120, and 240 mg/kg AZT. None of the doses induced consistent changes in the number of hematopoietic stem and progenitor cells. However, identical cumulative doses administered as an intravenous 24-hour infusion led to marked changes. Spleen colony-forming units (CFU-S) per femur were diminished to about 60%. Burst-forming units-erythroid (BFU-E) and colony-forming units-erythroid (CFU-E) were substantially reduced to about 15 to 35% at the two highest doses, whereas the femoral content of colony-forming units-granulocyte/macrophage (CFU-GM) and colony-forming units-megakaryocyte (CFU-Meg) was unchanged. When administered to mice whose marrow was regenerating from total-body irradiation (TBI) and subsequent bone marrow transplantation (high proliferative fraction), 240 mg/kg AZT caused considerable reductions of all hematopoietic cell stages even when given as a single IV injection. The results indicate that (1) the mode of application is critical for AZT hematotoxicity; (2) erythropoietic progenitors are the most sensitive to AZT toxicity; and (3) hematotoxicity increases with increasing proliferative activity.

How many myeloid post-progenitor cells have to be transplanted to completely abrogate neutropenia after peripheral blood progenitor cell transplantation? Results of a computer simulation.

Although hematopoietic recovery following high-dose chemotherapy (HD-CT) and peripheral blood progenitor cell (PBPC) transplantation is rapid, there is still a 5- to 7-day period of severe neutropenia which, theoretically, might be abrogated by an additional transplantation of more differentiated myeloid post-progenitor cells (MPPC). However, both the number of MPPC required to abrogate neutropenia as well as the optimum scheduling of MPPC infusions are currently unknown. Therefore, these questions were addressed by applying a computer model of human granulopoiesis. First, model calculations simulating varying levels of chemotherapy dose intensity were performed and compared with typical clinical neutrophil recovery curves. Using this approach, the data for HD-CT without PBPC transplantation could be reproduced by assuming a reduction of stem cells, committed granulopoietic progenitors and proliferating precursors to about 0.001% of normal. PBPC-supported HD-CT was reproduced by increasing the starting values to at least 0.1%, which corresponded to about 1 to 2 x 10(5)/kg transplanted CFU-GM. Interestingly, reproduction of PBPC-supported HD-CT data could be observed for a wide range of starting values (0.1%-10% of normal), thus confirming the clinical observation that hematopoietic recovery after PBPCT cannot be improved by increasing the dose of transplanted cells over a certain threshold. Using the same simulation model, we then studied the effects of an additional MPPC transplantation. The results showed, that at least 5.7 X 10(8) MPPC/kg have to be provided in addition to the normal PBPC graft to avoid neutropenia <100/microL, and that MPPC are best transplanted on days 0 and 6 after HD-CT. Assuming a 100- to 120-fold cellular ex-vivo expansion rate and MPPC representing about 70% of total expanded cells, 5.7 X 10(8) MPPC/kg could be generated starting from 1 to 2 leukapheresis preparations with about 7 to 8 x 10(6) CD34+ PBPC/kg. Considering furthermore, that only a fraction of ex-vivo generated cells will seed and effectively produce neutrophils in-vivo, the required number of MPPC is most likely even higher and, therefore, might be difficult to be achieved clinically. However, the validity of the model results remains to be proven in appropriate clinical studies.

Effective ex vivo generation of megakaryocytic cells from mobilized peripheral blood CD34(+) cells with stem cell factor and promegapoietin.

OBJECTIVE: The additional transplantation of ex vivo-generated megakaryocytic cells might enable the clinician to ameliorate or abrogate high-dose chemotherapy-induced thrombocytopenia. Therefore, the ex vivo expansion of CD34(+) PBPC was systematically studied aiming for an optimum production of megakaryocytic cells. MATERIALS AND METHODS: CD34(+) PBPC were cultured in serum-free medium comparing different (n = 23) combinations of stem cell factor (SCF) (S), IL-1beta (1), IL-3 (3), IL-6 (6), erythropoietin (EPO) (E), thrombopoietin (TPO) (T) and promegapoietin (PMP, a novel chimeric IL-3/TPO receptor agonist). Ex vivo-generated cells were assessed by flow cytometry, morphology, and progenitor cell assays. RESULTS: Addition of TPO to cultures stimulated with S163E, a potent progenitor cell expansion cocktail previously described by our group, effectively induced the generation of CD61(+) cells (day 12: 31.4 +/- 7.9%). The addition of PMP tended to be more effective than TPO +/- IL-3. Whereas EPO was not required to maximize TPO- or PMP-induced megakaryocytic cell production, the use of IL-6 and IL-1beta augmented cellular expansion as well as CD61(+) cell production rates in the majority of cytokine combinations studied. Thus, the most effective CD61(+) cell expansion cocktail consisted of S163 + PMP which resulted in 65.9 +/- 3.0% CD61(+) at day 12 and an overall production of 40.7 +/- 4.5 CD61(+) cells per seeded CD34(+) PBPC. However, the basic 2-factor combination S + PMP also allowed for an effective CD61(+) cell production (day 12 CD61(+) cell production: 15.1 +/- 1.6). Moreover, maximum amplification of CFU-Meg was observed after 7 days using this two-factor cocktail (12.9 +/- 2.6-fold). The majority of CD61(+) cells generated in TPO- or PMP-based medium were low-ploidy 4N and 8N cells, and ex vivo-generated CD61(+), CD41(+), and CD42b(+) cells were mainly double positive for FACS-measured intracellular von Willebrand Factor (vWF) (76.7 +/- 3.3%, 58.8 +/- 4.4%, and 82.7 +/- 2.5%, respectively). CONCLUSIONS: Taken together, this study demonstrates that megakaryocytic cells can be effectively produced ex vivo with as little as two-factors (SCF + PMP), an approach that might be favorably employed in a clinical expansion trial aiming to ameliorate high-dose chemotherapy-induced thrombocytopenia.

Effective ex vivo generation of granulopoietic postprogenitor cells from mobilized peripheral blood CD34(+) cells.

OBJECTIVE: Neutropenia following high-dose chemotherapy and peripheral blood progenitor cell (PBPC) transplantation might be abrogated by an additional transplantation of ex vivo generated granulopoietic postprogenitor cells (GPPC). Therefore, the ex vivo expansion of CD34(+) PBPC was systematically studied aiming for optimum GPPC production. MATERIALS AND METHODS: CD34(+) PBPC were cultured in serum-free medium comparing different (n = 32) combinations of stem cell factor (S), interleukin 1 (1), interleukin 3 (IL-3) (3), interleukin-6 (6), erythropoietin (E), granulocyte colony-stimulating factor (G), granulate-macrophage colony-stimulating factor (GM), daniplestim (D, a novel IL-3 receptor agonist), and Flt3 ligand (FL) under various culture conditions. Ex vivo generated cells were assessed by flow cytometry, morphology, and progenitor cell assays. RESULTS: Addition of G +/- GM but not GM alone to cultures stimulated with S163E effectively induced the generation of GPPC. GPPC production was maximum after 12 to 14 days. Best expansion rates were observed when cells were cultured at 1.5x10(4)/mL in 21% O(2). Modifications of culture conditions were either less or equally effective (i.e., modification of starting cell concentrations, low oxygen, addition of serum albumin or autologous plasma, repetitive feeding). Comparison of different cytokine combinations revealed that the optimum GPPC expansion cocktail consisted of S6GD+FL (day 12: 130-fold cellular expansion, 32% myeloblasts/promyelocytes, 49.4% myelocytes/metamyelocytes, 12.4% bands/segmented), which furthermore expanded CD34(+) cells (3.4-fold) and clonogenic progenitors (13.4-fold). CONCLUSION: Using the S6DG+FL expansion cocktail, GPPC could be effectively produced ex vivo starting from positively selected CD34 PBPC, possibly enabling amelioration or even abrogation of posttransplant neutropenia.

Ex vivo manipulation of hematopoietic stem and progenitor cells.

Ex vivo manipulations of hematopoietic stem and progenitor cells are Increasingly used in the context of autologous and allogeneic stem-cell transplantation. These manipulations include the positive selection of CD34+ cells for tumor-cell reduction and/or T-cell depletion, the ex vivo expansion of hematopoietic progenitor and stem cells under appropriate cytokine-stimulated culture conditions, and the ex vivo generation of myeloid or megakaryocytic postprogenitor cells and Immune effector cells. This article summarizes both the preclinical data on the ex vivo expansion of hematopoietic progenitor and stem cells from purified CD34+ cells and the Initial clinical studies with ex vivo-expanded stem and progenitor cells for hematopoietic support after high-dose chemotherapy.

Ex vivo expansion of hematopoietic progenitor cells for clinical use.

The development of efficient stem and progenitor cell selection methods in combination with the development of hematopoietic growth factors facilitated the development of ex vivo expansion techniques. Currently, this novel domain of cellular therapy aims to generate stem and progenitor cells, as well as more differentiated post-progenitor cells and antigen-presenting dendritic cells. The feasibility of generating and transplanting hematopoietic progenitor cells ex vivo (using various cytokine combinations) has been successfully shown preclinically as well as clinically. Furthermore, cytokines (eg, Flt-3-ligand; thrombopoietin) have been identified that play important roles with regard to amplification of undifferentiated early hematopoietic cells. The use of lineage-specific cytokines such as granulocyte colony-stimulating factor and thrombopoietin facilitated the generation of large numbers of myeloid and megakaryocytic post-progenitor cells. The clinical usefulness of such ex vivo generated cells, however, has not yet been convincingly shown. Last, ex vivo expansion techniques can be used to generate large numbers of antigen-presenting dendritic cells from CD34+ peripheral blood progenitor cells that might be ideally used for immunotherapeutic approaches.

Characterization of purified and ex vivo manipulated human hematopoietic progenitor and stem cells in xenograft recipients.

Research on the biology, regulation, and transplantation of human hematopoietic stem cells requires test systems for the detection, monitoring, and quantitation of these cells. Xenografted animal models provide suitable stem cell assays, since they allow long-term engraftment, multilineage differentiation, and serial transfer of human hematopoietic cells. Recent techniques for the separation of hematopoietic cells have provided highly purified cellular subsets selected on the basis of the surface marker phenotype. The stem cell content of these subsets, however, is still unclear. Also, innovative approaches for the induction of hematopoietic cell proliferation and differentiation have generated ex vivo manipulated cells whose biological properties and functions still remain to be assessed. This paper reports on the biological characterization of these cell populations by the use of xenograft models.

Expression of novel surface antigens on early hematopoietic cells.

The purpose of this report is to demonstrate the expression of very recently identified surface antigens on CD34+ and AC133+ bone marrow (BM) cells. Coexpression analysis of AC133 and defined antigens on CD34+ BM cells revealed that the majority of the CD164+, CD135+, CD117+, CD38low, CD33+, and CD71low cells resides in the AC133+ population. In contrast, most of the CD10+ and CD19+ B cell progenitors and a fraction of the CD71high population are AC133-, indicating that CD34+AC133+ cells are enriched in primitive and myeloid progenitor cells, whereas CD34+AC133- cells mainly consist of B cell and late erythroid progenitors. This corresponds to the highly reduced percentage of CD10+ B cells and the absence of CD71high erythroid progenitors on AC133+ selected BM cells. A portion of 0.2-0.7% of the AC133+ selected cells do not coexpress CD34. These cells are very small and define a uniform CD71-, CD117-, CD10-, CD38low, CD135+, HLA-DRhigh, CD45+ population with unknown delineation. Four color analysis on CD34+CD38- BM cells revealed that virtually all of these primitive cells express AC133. Using an improved liposome-enhanced labeling technique for the staining of weakly expressed antigens, subsets of this population could be identified which express the angiopoietin receptors TIE (67.6%) and TEK (36.8%), the vascular endothelial growth factor receptors FLT1 (7%), FLT4 (3.2%), and KDR (10.4%), or the receptor tyrosine kinases HER-2 (15.4%) and FLT3 (CD135; 77.6%). Our results suggest that the CD34+CD38- population is heterogeneous with respect to the expression of the analyzed receptor tyrosine kinases.

Approaches to dendritic cell-based immunotherapy after peripheral blood stem cell transplantation.

High-dose chemotherapy with peripheral blood progenitor cell transplantation (PBPCT) is a potentially curative treatment option for patients with both hematological malignancies and solid tumors, including breast cancer. However, based on a number of clinical studies, there is strong evidence that minimal residual disease (MRD) persists after high-dose chemotherapy in a number of patients, which eventually results in disease recurrence. Therefore, several approaches to the treatment of MRD are currently being evaluated, including treatment with dendritic cell (DC)-based cancer vaccines. DCs, which play a crucial role with regard to the initiation of T-lymphocyte responses, can be generated ex vivo either from CD34+ hematopoietic progenitor cells or from blood monocytes. They can be pulsed in vitro with tumor-derived peptides or proteins, and then used as a professional antigen-presenting cell (APC) vaccine for the induction of antigen-specific T-lymphocytes in vivo. This paper summarizes our preclinical studies on the induction of primary HER-2/neu specific cytotoxic T-lymphocyte (CTL) responses using peptide-pulsed DC. As HER-2/neu is overexpressed on 30-40% of breast and ovarian cancer cells, this novel vaccination approach might be particularly applicable to advanced breast or ovarian cancer patients after high-dose chemotherapy and autologous PBPCT.

Hematotoxic effects of benzene analyzed by mathematical modeling.

The hematopoietic cell response to benzene intoxication in mice (during and after long-term inhalation) was analyzed by a mathematical model of murine hematopoiesis. Two complementary methods, Time-Curve and Steady-State Analysis, were developed to identify target cells for benzene toxicity and to quantify the extent of damage in different stages of development of these target cells. We found that (i) erythropoietic cells were the most sensitive; (ii) granulopoietic cells were about half as sensitive as erythropoietic and (iii) hematopoietic stem cells exhibited a sensitivity that ranged between that of erythropoietic and granulopoietic cells. A dose-response relationship between benzene levels and damage in target cells (valid from 1 to more than 900 ppm) was derived that was linear for doses up to 300 ppm and plateaued thereafter. This relationship indicated that benzene-induced hematotoxicity is subject to a saturable process. Recovery of hematopoiesis following chronic benzene intoxication was simulated for different doses and preceding exposure periods. The impaired recovery following exposure periods greater than 8 weeks could be explained by a severe reduction in the maximum self-maintenance of stem cells. This study indicates that the present mathematical model represents a useful approach to investigate alternate hypotheses for the action of hematotoxic agents.

Ex vivo manipulation of hematopoietic stem and progenitor cells.

Approaches to manipulate peripheral blood progenitor cells (PBPC) ex vivo currently include the selection of CD34+ cells as a means to purge contaminating tumor cells from leukapheresis preparations or to provide a homogeneous starting population for the expansion of hematopoietic progenitor cells as well as the induction of postprogenitor cells of either the myeloid or megakaryocytic lineage. The latter cell populations might be used for an additional transplantation together with PBPC to possibly shorten the period of aplasia. In addition, ex vivo expansion of CD34+ cells can be used to generate autologous tumor-antigen-presenting dendritic cells for immunotherapeutic approaches aiming to treat minimal residual disease following high-dose chemotherapy.

A mathematical approach to benzo[a]pyrene-induced hematotoxicity.

Benzo[a]pyrene (BaP) has been reported to exert a differential effect on murine hematopoiesis that is mouse strain specific. Interpretation of these results based solely on experimental data is restricted and leaves important questions unanswered. Therefore, a mathematical model of murine hematopoiesis was applied in order to: (1) identify the targets of BaP, (2) quantify the damage to target cells and (3) based on these results, interpret differences in strain susceptibility. Model analysis of the hematopoietic response of D2 and BDF1 mice to a daily oral administration of 125 mg/kg BaP showed that proliferating hematopoietic cells are the targets of BaP. Within this group it was found that: (a) erythropoietic cells were the most susceptible to BaP, (b) granulopoietic cells showed a susceptibility half that of erythropoietic cells and (c) the susceptibility of stem cells ranged between that of erythropoietic and granulopoietic cells. This damage pattern was the same for both strains, indicating that the difference between the strains was quantitative. As cell destruction rates were about 3-fold higher for D2 than BDF1 mice, it was concluded that D2 mice were about three times as susceptible to BaP as BDF1 mice. The study showed that the mathematical model, in addition to experimental methods, provided an efficient tool for the analysis of BaP hematotoxicity.

Clinical applications of CD34(+) peripheral blood progenitor cells (PBPC).

Recently, a number of devices have been developed for the positive selection of CD34(+) peripheral blood progenitor cells (PBPC) for clinical use in autologous or allogeneic transplantation. The rationale for CD34(+) selection is based on clinical studies showing a two- to five-log reduction of contaminating tumor cells in patients with breast cancer, multiple myeloma and low-grade lymphoma. In addition, a three- to five-log reduction of T cells can be obtained by CD34(+) selection in both autologous grafts for patients with autoimmune disease resistant to conventional therapy and allogeneic grafts to reduce the incidence and severity of acute graft-versus-host disease. Transplantation of positively selected autologous CD34(+) PBPC results in a rapid and stable neutrophil and platelet engraftment in patients who received an infused dose of at least 2.0 x 10(6) CD34(+) cells/kg. Results from randomized trials suggest that time to engraftment is not different compared to unmanipulated PBPC autografts. However, close monitoring for infectious complications (e.g., cytomegalovirus disease) is required. Allogeneic CD34(+) PBPC have also been successfully transplanted and, using novel technologies, megadoses of purified CD34(+) PBPC can be obtained and used to overcome histocompatibility differences betweeen allogeneic donor and patient resulting in stable engraftment, even in a haploidentical setting. Additional randomized phase III trials are required to determine whether tumor cell purging or lymphocyte depletion by CD34(+) cell selection will have a significant impact on progression-free and overall survival in both autologous and allogeneic transplantation.

Clinical use for expanded peripheral blood stem cells.

The increasing use of haematopoietic stem and progenitor cells from the peripheral blood (PBPC) to restore haematopoiesis following high-dose chemotherapy has widely propagated the development of techniques for the ex vivo manipulation of haematopoietic cells. In particular, protocols for the ex vivo expansion of PBPC have been developed for different clinical purposes. Quantitative expansion of PBPC may provide a successful strategy for tumour cell purging of autologous grafts, or may generate sufficient cell numbers for sequential transplantation protocols. Furthermore, allogeneic transplantation of megadoses of PBPC may enable us to overcome immunological barriers, and may substantially increase the number of suitable donors for an individual patient. Clinical applications also include the use of ex vivo generated, partially differentiated, post-progenitor cells, antigen presenting cells for immunotherapy of minimal residual disease, and ex vivo transduced haematopoietic cells as attractive vehicles for genetic therapy.

Peripheral blood stem cells: in vivo biology and therapeutic potential.

Peripheral blood stem cells (PBSC) have been studied for their use after high-dose chemotherapy. The combination of a standard-dose chemotherapy [VIP: VP16 (etoposide), ifosfamide, cisplatin] in combination with hematopoietic growth factors was shown to provide effective anti-cancer activity as well as to enable sufficient stem cell mobilization for clinical use. Different growth factor regimens [granulocyte-colony-stimulating factor (G-CSF), granulocyte-macrophage (GM)-CSF, interleukin (IL)-3/GM-CSF] resulted in a differential induction of high levels of circulating PBSC after VIP chemotherapy, with the sequential combination of IL-3 and GM-CSF inducing maximal numbers of CD34+ cells as well as clonogenic progenitors. Our studies revealed a correlation between prior treatment and PBSC recruitment: the highest numbers of PBSC were mobilized in untreated patients whereas stem cell harvest was considerably impeded in heavily pretreated patients. Phase I/II trials demonstrated that transplantation of PBSC collected after VIP plus growth factor mobilization was safe, and engraftment was rapid and sustained. However, PBSC mobilization carried the risk of concomitant tumor cell recruitment in patients with detectable levels of tumor cells prior to therapy and in 21% of patients without circulating tumor cells. Positive selection of CD34+ cells by immunoadsorption that leads to an approximately three-log depletion of contaminating tumor cells therefore was investigated with regard to feasibility and capability as a source for PBSC transplantation. Twenty-one patients with advanced malignancies received autologous CD34+ cell transplantation after high-dose chemotherapy. Hematological recovery was as rapid as recorded for unseparated PBSC preparations, indicating that CD34+ cells can be safely used for autologous PBSC transplantation.(ABSTRACT TRUNCATED AT 250 WORDS)