Identifying content to improve risk assessment communications within the Risk Profile: Literature reviews and focus groups with expert and non-expert stakeholders.

OBJECTIVE: To improve consumer decision making, the results of risk assessments on food, feed, consumer products or chemicals need to be communicated not only to experts but also to non-expert audiences. The present study draws on evidence from literature reviews and focus groups with diverse stakeholders to identify content to integrate into an existing risk assessment communication (Risk Profile). METHODS: A combination of rapid literature reviews and focus groups with experts (risk assessors (n = 15), risk managers (n = 8)), and non-experts (general public (n = 18)) were used to identify content and strategies for including information about risk assessment results in the "Risk Profile" from the German Federal Institute for Risk Assessment. Feedback from initial focus groups was used to develop communication prototypes that informed subsequent feedback rounds in an iterative process. A final prototype was validated in usability tests with experts. RESULTS: Focus group feedback and suggestions from risk assessors were largely in line with findings from the literature. Risk managers and lay persons offered similar suggestions on how to improve the existing communication of risk assessment results (e.g., including more explanatory detail, reporting probabilities for individual health impairments, and specifying risks for subgroups in additional sections). Risk managers found information about quality of evidence important to communicate, whereas people from the general public found this information less relevant. Participants from lower educational backgrounds had difficulties understanding the purpose of risk assessments. User tests found that the final prototype was appropriate and feasible to implement by risk assessors. CONCLUSION: An iterative and evidence-based process was used to develop content to improve the communication of risk assessments to the general public while being feasible to use by risk assessors. Remaining challenges include how to communicate dose-response relationships and standardise quality of evidence ratings across disciplines.

Synthesis of granulocyte colony-stimulating factor and its requirement for terminal divisions in chronic myelogenous leukemia.

In this paper we demonstrate that maturing neoplastic cells from patients with chronic myelogenous leukemia (CML) constitutively produce G-CSF and are also receptive for this molecule. G-CSF functions as an autocrine growth factor in stable phase CML, and thus is responsible for divisions of maturing leukemic cells leading to an expansion of the compartment of mature cells. This observation is well in line with in vivo features of CML in stable phase, i.e., the hyperplasia of the mature granulocyte compartment. In acute blastic phase of CML expression of the G-CSF gene seems to be less common and not related to autonomous blast growth.

Targeting the DNA Damage Response in OSCC with TP53 Mutations.

Oral squamous cell carcinoma (OSCC) is the most common type of oral cancer worldwide and in the United States. OSCC remains a major cause of morbidity and mortality in patients with head and neck cancers. Tobacco and alcohol consumption alone or with chewing betel nut are potential risk factors contributing to the high prevalence of OSCC. Multimodality therapies, including surgery, chemotherapy, biologic therapy, and radiotherapy, particularly intensity-modulated radiotherapy (IMRT), are the current treatments for OSCC patients. Despite recent advances in these treatment modalities, the overall survival remains poor over the past years. Recent data from whole-exome sequencing reveal that TP53 is commonly mutated in human papillomavirus-negative OSCC patients. Furthermore, these data stressed the importance of the TP53 gene in suppressing the development and progression of OSCC. Clinically, TP53 mutations are largely associated with poor survival and tumor resistance to radiotherapy and chemotherapy in OSCC patients, which makes the TP53 mutation status a potentially useful molecular marker prognostic and predictive of clinical response in these patients. Several forms of DNA damage have been shown to activate p53, including those generated by ionizing radiation and chemotherapy. The DNA damage stabilizes p53 in part via the DNA damage signaling pathway that involves sensor kinases, including ATM and ATR and effector kinases, such as Chk1/2 and Wee1, which leads to posttranscriptional regulation of a variety of genes involved in DNA repair, cell cycle control, apoptosis, and senescence. Here, we discuss the link of TP53 mutations with treatment outcome and survival in OSCC patients. We also provide evidence that small-molecule inhibitors of critical proteins that regulate DNA damage repair and replication stress during the cell cycle progression, as well as other molecules that restore wild-type p53 activity to mutant p53, can be exploited as novel therapeutic approaches for the treatment of OSCC patients bearing p53 mutant tumors.

Interleukin 1 stimulates T lymphocytes to produce granulocyte-monocyte colony-stimulating factor.

T lymphocytes are thought to cooperatively interact with monocytes to produce colony-stimulating factors (CSF). However, little is known about monocyte-mediated signals leading to CSF-secretion by T lymphocytes, although soluble monocyte products have been implicated. We have employed monoclonal antibody anti-T3B covalently coupled to CnBr-activated Sepharose 4B beads, to show that multimeric ligation of T cell antigen receptor leads to T cell receptiveness to interleukin 1 (IL-1), as indicated by T cell production of CSF, which induces growth of myeloid progenitor cells into neutrophil, eosinophil, and monocyte colonies. To investigate the molecular basis of these findings, total RNA was extracted from T3B Sepharose-primed and IL-1-stimulated T lymphocytes and probed for granulocyte-monocyte-CSF (GM-CSF), granulocyte-CSF (G-CSF), and monocyte-CSF (M-CSF) mRNA. GM-CSF, but not G-CSF or M-CSF, messages were detected. Nuclear "run on" assays revealed that IL-1 action is effective primarily at the level of GM-CSF gene transcription. These results suggest a previously unrecognized role of IL-1 in the regulation of GM-CSF secretion by T cells.

Granulocyte-macrophage colony-stimulating factor induces cytokine secretion by human polymorphonuclear leukocytes.

Granulocyte-macrophage colony-stimulating factor (GM-CSF) is known as an inducer of proliferation and functional activation of myeloid cells. This study was carried out to characterize the effects of GM-CSF on polymorphonuclear leukocytes (PMN) more extensively. Using Northern blot analysis, we show that PMN are able to accumulate mRNAs for different cytokines, including tumor necrosis factor-alpha (TNF-alpha); G-CSF, and M-CSF, all of which are involved in inflammation and hematopoiesis. Biological assays and immunoassays demonstrate that PMN translate these mRNAs, except TNF-alpha, into secretory proteins. However, the expression of these cytokines is dependent on stimulation by exogenous signals, preferentially provided by the T cell-derived lymphokine GM-CSF. Stimulation of hematopoiesis and amplification of defense mechanisms after T cell activation thus might involve not only monocytes but also PMN, a cell type previously believed to be biosynthetically inactive.

Participation of the cytokines interleukin 6, tumor necrosis factor-alpha, and interleukin 1-beta secreted by acute myelogenous leukemia blasts in autocrine and paracrine leukemia growth control.

Autonomous in vitro growth of myeloid leukemic colony-forming cells may in part result from autocrine production of colony-stimulating factors (CSF). Some acute myeloid leukemia (AML) samples, however, fail to synthesize CSF despite growing autonomously in agar, and are therefore believed to bypass CSF requirements. Cytokines such as IL-6, tumor necrosis factor (TNF)-alpha, and IL-1, products of cells of the myeloid lineage, are known to be involved in growth control of myeloid progenitor cells. Since these molecules may also contribute to autocrine and paracrine growth regulation of myeloid leukemias, we screened a series of AML for cytokine production. In addition, possible roles of IL-6, TNF-alpha, and IL-1 in growth control of AML were investigated in vitro. We show that a substantial proportion of AML cells produce IL-6, TNF-alpha, and IL-1-beta and use these mediators to stimulate their growth by disparate mechanisms: IL-6 acts as a costimulator to enhance CSF-induced clonogenicity of AML blasts. TNF-alpha induces CSF production by endothelial cells and may therefore provide a paracrine loop to support leukemia growth.

Immunostimulatory cytokines in somatic cells and gene therapy of cancer.

The use of immunostimulatory cytokines has become an increasingly promising approach in cancer immunotherapy. The major goal is the activation of tumour-specific T lymphocytes capable of rejecting tumour cells from patients with low tumour burden or to protect patients from a recurrence of the disease. Strategies that provide high levels of immunostimulatory cytokines locally at the site of antigen have demonstrated pre-clinical and occasional clinical efficacy. Animal models using poorly immunogenic tumours revealed that tumour cells genetically engineered to produce cytokines like IL-2, IL-4, IL-7, IL-12, IFNs, GM-CSF or TNF-alpha were found to be effective in eradicating disseminated tumours. Experimental data obtained from these different animal models are reviewed here to provide an overview of this rapidly evolving field. The data obtained so far from clinical trials involving cytokine gene-modified cells have provided important information regarding the feasibility, safety, immunological effects and occasional clinical responses.

Biologic effects of recombinant human interleukin-3 in vivo.

The biologic in vivo effects of recombinant human interleukin-3 (rhIL-3) were assessed in a phase I clinical study of 30 patients with advanced malignancy. On day 1 rhIL-3 was administered by a single intravenous (IV) bolus injection, followed by subcutaneous (SC) injections once daily from day 2 to 15; at least three patients were treated at each dose level (60, 125, 250, and 500 micrograms/m2). A transient decrease of eosinophil and monocyte counts was observed immediately after IV injection of rhIL-3, whereas the neutrophil count remained unaffected. Total WBC counts and neutrophil counts increased dose dependently up to threefold, whereas a 10-fold to 50-fold rise was observed in levels of circulating eosinophils and basophils. Platelet counts increased up to twofold. Patients developed moderate increases of serum levels of soluble interleukin-2 receptors, beta 2-microglobin, and immunoglobulin M (IgM), and of the acute phase reactants, C-reactive protein (CRP), fibrinogen, and haptoglobin. An increase in interleukin-6 (IL-6) serum levels was detected in patients treated by IV bolus rhIL-3. The serum half-life of IV injected rhIL-3 was 20 +/- 3 minutes; after SC administration, 210 +/- 15 minutes. Administration of rhIL-3 was generally well tolerated, with mild fever, headache, and local reactions at the injection site being the most frequent side effects. The primary course of the underlying malignant diseases was not significantly altered by administration of rhIL-3. The results indicate that rhIL-3 acts in vivo as a multilineage hematopoietic growth factor and a weak inflammatory mediator that may be used successfully to improve states of hematopoietic failure.

Hematopoietic responses in patients with advanced malignancy treated with recombinant human granulocyte-macrophage colony-stimulating factor.

The in vivo effect of yeast-derived recombinant human granulocyte-macrophage colony-stimulating factor (rh GM-CSF) was investigated in 30 patients with advanced malignancy in a phase Ib trial. Patients were treated at four different dose levels (120 to 1,000 micrograms/m2/d) by either daily intravenous (IV) bolus injection or 24-hour continuous infusion. Administration of rh GM-CSF resulted in a broad spectrum of dose- and schedule-dependent hematopoietic effects. Sustained infusion of rh GM-CSF elicited a maximum 17-fold average peak increase of the total WBC count with mainly neutrophils, eosinophils, and monocytes accounting for this rise, and increases in bone marrow cellularity with a shift to immature myeloid elements. Elevation of lymphocytes, platelets, and reticulocytes was not induced. Within five days after discontinuation of treatment the leukocytosis had disappeared. Adverse reactions encountered with rh GM-CSF seen in 65% of the patients studied were never life-threatening and always rapidly reversible. They included mild myalgias, facial flushing, low-grade fever, headache, bone discomfort, nausea, dyspnea, and transient decline of platelet counts. These results suggest that rh GM-CSF can be safely administered at the doses and schedules used and that it can induce in vivo some of the biological effects reported in in vitro studies. Although no objective antitumour responses have been seen, the ability of rh GM-CSF to increase number and function of leukocytes in vivo may prevent neutropenia and infections when GM-CSF is added to cytotoxic cancer therapy.

Elevated circulating levels of tumor necrosis factor predict unresponsiveness to treatment with interferon alfa-2b in chronic myelogenous leukemia.

PURPOSE: The study was undertaken to analyze circulating tumor necrosis factor (TNF) levels in patients with chronic-phase chronic myelogenous leukemia (CML) undergoing interferon (IFN) alfa-2b therapy, and to correlate pretreatment serum levels of TNF with response to IFN alfa-2b therapy. PATIENTS AND METHODS: Fourteen patients with CML in chronic phase were treated with recombinant human IFN alfa-2b for 7 to 39 months. RESULTS: In eight patients IFN alfa-2b treatment failed due to lack of hematologic response. A complete or partial hematologic remission was achieved in the remaining six patients, of whom two patients experienced a complete cytogenetic response. Retrospective analysis of serum samples obtained from all patients before the onset of IFN alfa-2b administration revealed that levels (mean +/- SEM) of circulating TNF were higher (P less than .001) in the group of patients who did not respond to IFN alfa-2b treatment (157 +/- 15 U/mL) than in the responders (10.3 +/- 4 U/mL) or healthy control subjects (9.1 +/- 3 U/mL). However, there was no correlation between TNF serum levels and other patient characteristics at study enrollment including age, sex, duration of disease, performance status, splenomegaly, WBC count, platelet count, hemoglobin value, prior therapy, and prognostic category. CONCLUSION: These findings indicate that circulating levels of TNF are increased in a subset of patients with chronic-phase CML and that this elevation is associated with poor response to IFN alfa-2b therapy.

Erythropoietin for the treatment of anemia of malignancy associated with neoplastic bone marrow infiltration.

This clinical trial was performed to study the effects of intravenously (IV) administered recombinant human (rh) erythropoietin (EPO) at escalating doses (150, 300, and 450 U/kg, administered as an IV bolus injection, twice weekly, for 6, 4, and 4 weeks, respectively) in five patients with low-grade non-Hodgkin's lymphoma (Ig NHL) and bone marrow involvement and one patient with multiple myeloma (MM). All patients were anemic due to underlying disease. None of the patients had a history of bleeding, hemolysis, renal insufficiency, or other disorders causing anemia in addition to bone marrow infiltrating malignancy. Endogenous EPO serum levels were significantly increased in all patients (74 to 202 mU/mL). Five patients (one MM, four small-cell lymphocytic [SCLC] NHL) showed a dramatic increase of hemoglobin (Hb), hematocrit (Hk) and RBC count becoming obvious on the second EPO dose level. Initial ferritin serum values, which were high mostly due to polytransfusion, were significantly reduced in responding patients. Erythropoiesis of one patient with extensive follicular mixed (fm) NHL did not respond to EPO treatment. Platelet (PLT) count increase (greater than 75% above starting levels) during and following EPO therapy was observed in one patient with MM. Adverse events due to EPO therapy have not been recorded. These findings point out a previously unrecognized capacity of EPO given at pharmacologic doses to stimulate erythropoiesis in patients with anemia due to bone marrow infiltration by neoplastic lymphocytes in spite of enhanced endogenous EPO expression.

Expression of leukemia inhibitory factor is regulated in human mesenchymal cells.

A protein variously termed leukemia inhibitory factor (LIF), differentiation-inducing factor, differentiation inhibitory activity or human interleukin for DA cells can control the differentiation and proliferation of hematopoietic cells as well as of several other cellular lineages. In order to further elucidate the spectrum of LIF-producing cells, we examined different cell types for the expression of LIF mRNA using Northern blot analysis. LIF mRNA was detected in activated normal human T-cells and in two T-cell lines but was undetectable in a B-lymphoid cell line, in both resting and activated normal human granulocytes and monocytes and in human myeloid cell lines K562 and HL-60. In human lung fibroblasts and in human umbilical vein endothelial cells, LIF was constitutively expressed and its accumulation was increased in a time-dependent manner following treatment with the phorbol ester TPA and in the presence of the two immediate response cytokines tumor necrosis factor (TNF)-alpha and interleukin (IL)-1-beta. We conclude that mRNA for LIF is not only expressed by T-cells but also in human mesenchymal cells. Expression of LIF transcripts in these cells is constitutive and can be significantly enhanced by phorbol ester, TNF-alpha and IL-1-beta.

Screening for expression of cytokines with hematopoietic growth factor activity by permanent human B-cell lines.

Using colony assays in semi-solid media, several investigators have shown that supernatants (SN) of normal and malignant human B-cells can stimulate the growth of granulocyte-macrophage (GM) progenitor cells. So far macrophage colony-stimulating factor (M-CSF) and interleukin-6 (IL-6) have been identified as potential colony-stimulating activity (CSA) present in B-cell SN. However, other CSAs such as GM-CSF, G-CSF, IL-1-beta, IL-3, and IL-4 may also be candidates in this respect. Several human B-cell lines (CL) were screened for the expression of the respective genes at the mRNA and protein level. Constitutive production of GM-CSF was detected in the lymphoblastoid CL Wi-L2-729-HF2 and in the Burkitt line Raji. The signal intensity of specific transcripts and the amount of protein being secreted increased upon exposure to the phorbol ester PMA. The hybridoma line HB-564 also expressed the GM-CSF gene, but required prior stimulation with PMA. 3H-thymidine incorporation of Raji and Wi-L2-729-HF2 cells was unchanged in the presence or absence of a specific neutralizing sheep anti-GM-CSF serum, suggesting that GM-CSF did not serve as an extracellular autocrine growth factor. The expression of the GM-CSF gene was independent of the proliferative state (log phase growth versus plateau phase growth) and of the presence of serum in cultures of the respective CL. The expression of G-CSF, IL-1-beta, IL-3, and IL-4 genes was not detectable in the CL at the mRNA level.

Poor prognosis of prethymic phenotype acute lymphoblastic leukemia (pre-T-ALL).

Pre-T-ALL is an important subgroup of ALL with clinical features different from adult T-ALL. Expression of intracytoplasmic CD3 represents the earliest marker for the prethymic phenotype. We studied four consecutive adult patients with this phenotype. Three of the four patients did not respond to the induction chemotherapy with vincristine, daunorubicin, prednisone and asparaginase. They reached a delayed remission only after chemotherapy with cyclophosphamide and cytosine arabinoside. All four patients relapsed 3, 9, 10 and 13 months after diagnosis. One patient died 2 months after relapse, another one 2 months after allogeneic BMT performed in second relapse. We conclude that patients with early T-cell precursor leukemia do not respond adequately to conventional chemotherapy and should be considered as a high-risk subgroup within the T-lineage ALL.

In vivo recruitment of GM-CSF-response myelopoietic progenitor cells by interleukin-3 in aplastic anemia.

Previous studies have indicated that colony-stimulating factors may stimulate myelopoiesis and thus increase the number of circulating white blood cells in patients with hematopoietic failure including aplastic anemia. However, long-term administration of the factor was required to maintain its response. In the present article we report on a patient with severe aplastic anemia undergoing treatment with recombinant human (rh) granulocyte-macrophage colony-stimulating factor (GM-CSF). After an initial response, the patient became refractory to GM-CSF. However, treatment with interleukin (IL)-3 restored responsiveness to GM-CSF, suggesting that IL-3 may have replenished the bone marrow with myelopoietic progenitor cells sensitive to the action of GM-CSF. This observation suggests the value of application of sequentially acting hematopoietic growth factors in aplastic anemia patients.

Gamma-interferon interrupts growth stimulation in chronic myelogenous leukemia established by endogenous granulocyte colony-stimulating factor.

We have previously shown that maturing neoplastic cells from patients with stable phase chronic myelogenous leukemia (SP CML) constitutively produce granulocyte colony-stimulating factor (G-CSF) and are also receptive for this molecule. G-CSF functions as an endogenous growth factor in SP CML, and thus is responsible for divisions in maturing leukemic cells leading to an expansion of the compartment of mature cells. In the investigations to be reported below, the effects of various hematopoietic inhibitor molecules on the expression of the G-CSF gene by SP CML bone marrow cells enriched for promyelocytes/myelocytes were examined at the mRNA and protein level. We show that exposure of SP CML bone marrow promyelocytes/myelocytes to recombinant human (rh) interferon (IFN)-gamma but not to rh IFN-alpha, rh tumor necrosis factor (TNF)-alpha, and rh lymphotoxin (LT) leads to downregulation of G-CSF expression and interruption of the G-CSF-mediated endogenous growth stimulation. The action of G-CSF takes place at the posttranscriptional level and involves an acceleration of decay of steady-state levels of G-CSF transcripts in the malignant cell population.

Effect of recombinant human granulocyte-macrophage colony-stimulating factor in patients with myelodysplastic syndrome with excess blasts.

As part of a broad phase I study of recombinant human granulocyte-macrophage colony-stimulating factor (rh GM-CSF), four patients were treated who had myelodysplastic syndrome (MDS) with excess blasts. The GM-CSF was given daily as an intravenous injection over a period of 30 min for 5 days. A total of 11 cycles were conducted. Each patient received at least two different dose levels. In three patients, three different dosages were delivered. The treatment course was interrupted by a 10-day rest period. Rh GM-CSF was well tolerated, with only minor side effects seen, which included bone discomfort at the lower back, sternum and ribs, and constitutional symptoms such as low grade fever, nausea/vomiting, and mild myalgias. Whereas no increases in platelet and reticulocyte counts were recorded, elevations of absolute neutrophil counts above 100 cells/microliters occurred in all patients. The most striking finding was, however, the development of increases in the number of circulating and bone marrow blast counts that were observed particularly when doses of greater than or equal to 500 micrograms/m2 of body surface area were administered. In line with data demonstrating in vitro induction of proliferation of leukemic blast cells by rh GM-CSF, one may take advantage of blastogenesis induced in vivo that may favor the use of a therapeutic strategy by recruiting quiescent cells into the mitotic cycle which would then represent optimum targets for a subsequent cycle-specific cytotoxic chemotherapy. Such an approach could form the basis for new clinical trials in MDS.

Effect of long-term treatment with recombinant human interleukin-3 in patients with myelodysplastic syndromes.

In a phase II study, involving nine patients with refractory anemia or refractory anemia with ring sideroblasts, the effects of treatment with recombinant human interleukin-3 (IL-3) on hematopoietic function were assessed. Doses of IL-3 ranging from 60 micrograms/m2 during weeks 1-6 to 125 micrograms/m2 during weeks 7-12 were administered as subcutaneous bolus injections three times per week for 12 weeks. Platelet counts increased in six patients. Platelet increase correlated with stable or decreased serum tumour necrosis factor alpha (TNF-alpha) levels, while an increase of TNF-alpha levels during IL-3 therapy occurred in patients with no change or a decrease of platelet counts. Leukocyte counts increased in two patients and reticulocytes in three, without an effect on hemoglobin levels. Morphological analysis of the bone marrow revealed an expansion of the myeloid compartment in seven of eight evaluable patients, mainly due to stimulation of the precursor cells. No improvement of the in vitro growth of hematopoietic progenitor cells was observed. Sequential cytogenetic analyses indicate that IL-3 treatment does not act preferentially on either the cytogenetically abnormal or the normal clones. These results suggest that long-term treatment with low-dose IL-3 stimulates megakaryopoiesis with increase of platelet counts, but that additional later-acting cytokines probably will be required to augment neutrophil and erythrocyte counts.

GM-CSF in a double-blind randomized, placebo controlled trial in therapy of adult patients with de novo acute myeloid leukemia (AML).

We investigated whether GM-CSF given concomitantly with chemotherapy (CT) and thereafter, improves the outcome of adults with de novo AML by increasing the efficacy of CT and reducing infections. CT included Ara-C, daunorubicin and etoposide (DAV) for induction and early consolidation therapy and one cycle with high-dose (patients aged < or = 50 years) or intermediate dose Ara-C (patients aged > 50 years)/daunorubicin for late consolidation therapy. Eight patients were randomized after DAVI to receive either GM-CSF (E. coli, 250 micrograms/m2/day, s.c.) or placebo starting 48 h prior to DAVII and the subsequent courses and given throughout CT until the absolute neutrophil count had recovered to > 500/microliters. The CR rate was 81% in the GM-CSF and 79% in the placebo group (p = 0.57; Fisher's exact test). The probability of relapse-free survival at 41 months after a median follow-up of 35 months was 42% in the GM-CSF and 41% in the placebo group (p = 0.89; log rank test). GM-CSF did not shorten the period of neutropenia < or = 500/microliters, while it prolonged the duration of thrombocytopenia < or = 25,000/microliters. The incidence and severity of infections as well as the non-hematological toxicity were similar in both groups. In summary, in this randomized trial GM-CSF as an adjunct to AML therapy was feasible. For the present GM-CSF does not have a significant effect on treatment outcome.

Generation and purification of CD8+ melan-A-specific cytotoxic T lymphocytes for adoptive transfer in tumor immunotherapy.

Tumor antigens that might serve as potential targets for adoptive T-cell therapy have been defined in different tumor entities, especially in malignant melanoma. To generate conditions to induce primary T-cell responses against different HLA-A*0201-restricted melanoma peptides and to allow further expansion of peptide-specific T cells for adoptive transfer, CD8+-purified T cells from healthy donors were stimulated with Melan-A-pulsed autologous dendritic cells. Dendritic cells were generated in vitro from monocytes with granulocyte macrophage colony-stimulating factor, interleukin-4, and transforming growth factor-beta1. After 3-4 weekly stimulation cycles with Melan-A-pulsed DCs, we were able to induce a strong peptide-specific CTL response in vitro. MHC-peptide tetramer staining revealed a frequency of up to 3.5% CD8+/Melan-A+ T cells. Additional antigen-independent expansion with anti-CD3/anti-CD28 monoclonal antibodies together with interleukin-2 gave rise to 600-fold expansion of CD8+ CTLs that maintained Melan-A specificity and were able to efficiently lyse Melan-A-expressing melanoma cells. To enrich antigen-specific T cells in vitro, we used a recently established technology for analysis and sorting of live cells according to secreted cytokines. In the present study, we demonstrated that Melan-A-specific T cells can be purified by magnetic separation according to secreted IFN-gamma. These cells revealed a very potent monospecific CTL response, even at low E:T ratios, against Melan-A-pulsed and Melan-A-expressing target cells. Altogether, our study demonstrated that we have developed an efficient method for generating large numbers of peptide-specific T cells in vitro that may be used for adoptive T-cell transfer in tumor immunotherapy.