K. Pfizenmaier

Intralesional application of recombinant human tumor necrosis factor alpha induces local tumor regression in patients with advanced malignancies

Fourteen patients with different advanced solid tumors were treated by intratumoral application of recombinant human tumor necrosis factor alpha. In five patients, local tumor regression occurred. However, the duration of response was short, implying a rapid development of resistance to rTNF-alpha application. The main clinical side-effects, including chills, fever, anorexia and fatigue, were similar to systemic rTNF-alpha treatment. Cardiovascular, pulmonary or metabolic toxicities were not observed. This study demonstrates that a high concentration of rTNF-alpha at the tumor site has the potential to induce local tumor regressions and, therefore, seems more reasonable for further clinical investigations, especially in combination with other cytokines.

Tumor necrosis factor (TNF) alpha: control of TNF-sensitivity and molecular mechanisms of TNF-mediated growth inhibition

Tumor necrosis factor (TNF) alpha is a macrophage derived cytokine, elicited during host responses to various microbial infections. TNF-alpha exerts direct cytotoxicity towards some tumor cells in vitro and produces in vivo hemorrhagic tumor necrosis, whereas normal tissues remain unaffected (for review see [1-5]). Although the mechanisms underlying this apparent tumor-selective action remain to be defined, the prospect of a potentially powerful and selective antitumoral agent has created much interest in the identification of this biological activity, which eventually resulted in purification of the protein [6] and subsequent molecular cloning of TNF-alpha cDNA [7]. Human TNF-alpha is a non-glycosylated protein comprised of 157 amino acids (molecular weight ~ 17 kDalton), which exhibits a high specific activity ( ~ 6× 107 U/rag) in a standard in vitro cytotoxicity assay using actinomycin D-pretreated mouse fibrosarcoma cell lines [8, 9]. Sequence analysis revealed approximately 30% homology, at the amino acid level, to another cytotoxin, the lymphocyte product TNF-beta, formerly called lymphotoxin [10]. TNF-beta is a glycosilated protein comprised of 171 amino acids with a total molecular weight of approximately 25 kDalton [10, 11]. The genes for TNF-alpha and TNF-beta are tandemly arranged within the HLA gene complex on human chromosome 6 and probably are derived by duplication of a common ancestral gene [12, 13]. TNF-alpha and TNF-beta bind to the same cell membrane receptor and possess a similar, if not identical, spectrum of biological activities [14-19]. Highly purified recombinant TNF-alpha and TNF-beta produced in bacteria are nonglycosilated molecules with a molecular weight of 17 and 18.6 kDalton, respectively, which exert the same specific biological activity as their natural counterparts [7, 10]. With the availability of large amounts of highly purified recombinant human TNF (rTNF)-alpha, both clinical trials on the role of TNF-alpha in tumor therapy and extensive laboratory investigations on mechanisms of TNF-action were begun. It soon became evident that rTNF-alpha is more than a cytotoxic or tumor necrotising effector molecule. Rather, TNF-alpha appears to be a cytokine with pleiotropic activi-

Human chromosome 21 is necessary and sufficient to confer human IFN gamma responsiveness to somatic cell hybrids expressing the cloned human IFN gamma receptor gene

The human interferon (IFN) gamma receptor cDNA has been stably expressed in human/mouse somatic cell hybrids, which differ in their content of human chromosome 21. Despite high affinity IFN gamma binding-capacity of all receptor transfectants, biological responsiveness to IFN gamma, as determined by enhancement of mouse-MHC class I gene expression, required the presence of chromosome 21. These data suggest complementation of at least two functionally distinct components in order to create a biologically active IFN gamma receptor.

IFN-γ Receptors on Human Tumor Cells: Relationship between Receptor Ligand Interactions and Induction of IFN-γ Response

Abstract Scatchard analysis of 125 I-IFN-³ binding on fresh lymphoid and myeloid tumor cells derived from 34 leukemia patients and on normal cells obtained from 14 healthy individuals revealed similar high affinity binding with a mean Kd of around 2 x 10 -11 M in 14/14 normal and 30/34 malignant cells, but large quantitative differences in the receptor number of malignant cells with a range of 300 to 12,000 receptors/cell. In contrast, normal lymphoid and myeloid cells expressed consistantly low numbers of receptors with a mean of 300 and 1,000 receptors/cell, respectively. Kinetic studies of IFN-γ binding in relation to induction of HLA-DR antigens in established cell lines revealed the existence of close correlations between the quantity of receptor ligand interaction and induction of IFN-γ response, indicating that at limiting IFN-γ concentrations the height of response is controlled by the number of expressed membrane receptors. In the light of the observed quantitative differences in IFN-γ receptors on various tumors, this finding may have implications for the definition of therapeutically effective IFN-γ doses.

Clonal rearrangements of T-cell receptor and immunoglobul in genes as sensitive marker for bone marrow involvement in non-Hodgkin's lymphoma

The c l i n i ca l stage of pat ients with Non-Hodgkins Lymphoma (NHL) not only is a prognostic parameter but also has important impl icat ions with regard to the therapeutic protocol. As to the staging of NHL, the involvement of the bone marrow indicates ear ly dissemination of disease. However, the detect ion of small numbers of i n f i l t r a t i n g neoplastic Tor B-cel ls often provides great d i f f i c u l t ies because diagnostic procedures including immunophenotyping are not speci f ic to transformed lymphocytes. Recently, i t has been shown that rearrangements of T-cel l receptor (TCR) and immunoglobulin (Ig) genes can be used to i den t i f y tumors of Tand B-cel l l ineage and to demonstrate monoclonality. We have used cDNA probes for the TCR B-chain and Ig heavy chain J region genes to document clonal expansions of neoplast ic Tor B-cel ls in the bone marrow of select cases, where h is to logica l and cyto logical examination was inconclusive. In four of s ix pat ients Southern b lot analysis of bone marrow DNA revealed rearranged TCR or Ig genes ind icat ing the presence of Tand B-cell tumors, respect ive ly . Rearrangements were never detected in the bone marrow of healthy donors or pat ients with CML. In th is study, we demonstrate that the detection of clonal TCR and Ig gene rearrangements unequivocally documents bone marrow involvement in NHL pat ients. These unique DNA rearrangements promise to be useful for confirming the pathological diagnosis and for assessing the stage of NHL.

Biological effects and specific membrane receptors for IFN-γ and for tumor necrosis factor on human tumor cells

Acutely transforming retroviruses contain oncogenic sequences which are derived from counterparts in the normal cellular genome. Only little is known about the function of these cellular homologs termed c-eric genes. We have investigated the expression of c-src, the cellular counterpart of the Rous sarcoma virus transforming gene v-src, c-src is highly conserved during phylogenesis and appears first in the sponges (Barnekow and Schartl, Mol.Oell.Biol.~1179,1984) The expression of the c-src gene was found to be tissue-specific and age-dependent(Barnekow and Bauer, BBA 782, 94,1984). Our data suggest that the physiological function of c-src appears to be more closely related to differentiation processes than to proliferation processes. This theory is supported by data obtained from experiments with the promyelocytic cell line HL-60. Induction of monocytic and granulocytic differentiation of HL-60 cells by 12-O-tetradecanoylphorbol-13-acetate(TPA) and dimethylsulfoxide (DMSO) is associated with an activation of the pp60 C~src tyrosine kinase, but not with increased c-src gene expression. Control experiments exclude an interaction of TPA and DMSO themselves with the pp60 C-src kinase. Using embryonal carcinoma cell lines~ which also can be induced to differentiate in vitro, we are currently analyzing whether the differentiation-dependent expression of c-src is restricted to monomyelocytic cells or can be generally observed during cellular differentiation processes.