Both T-cell colony formation and T-cell growth in liquid culture requires phytohemagglutinin, either adherent cells or conditioned medium from mononuclear cells, and perhaps other cooperating cells. Despite these similarities, there are some data that the two systems may differ. While it is clear that a source of T-cell growth factor (TCGF) is an absolute requirement for the growth of T cells in liquid culture, less is known of the factors influencing T-cell colony formation. In these studies, production of TCGF during T-cell colony formation was determined. High levels of TCGF can be extracted from methylcellulose when colony formation is rapidly increasing. When colony size is at its maximum, measurable levels of TCGF decline and after 2 weeks little or no TCGF is detectable, suggesting that the factor is used by colony-forming cells. The only requirement for these colony-forming cells to form secondary colonies in replating experiments is exogenous TCGF- The need for adherent cells in primary colony formation can be substantially replaced by interleukin 1, a factor which is produced by adherent cells and functions by amplifying TCGF production. Other promoters of T-cell colony formation may also amplify TCGF production. Addition of dexamethasone to the assay decreases the number and size of the T-cell colonies formed by 75-90%, and colony formation can be completely restored by the addition of purified TCGF. The addition of antigen, lectin, 12-O-tetradecanoyl-phorbol-13-acetate or interleukin 1 cannot overcome the specific dexamethasone-mediated inhibition of colony formation. Although there may be additional requirements for colony formation depending upon the T-cell subset and state of differentiation of the cells, it appears that TCGF is the ultimate proliferative signal in both liquid and colony T-cell growth.

Mononuclear cells (MNCs) were isolated from the endometrium of cattle by enzymatic digestion followed by centrifugation on a density gradient. Of the cells isolated, 32 +/- 3.2% rosetted AET-treated sheep red blood cells, 9 +/- 0.4% possessed surface membrane immunoglobulin, and 15 +/- 6.1% were esterase positive. The cells possessed the ability to proliferate in the presence of T cell mitogens and produced T cell growth factor (TCGF) after in vitro stimulation with Concanavalin A. Endometrial MNCs did not appear to suppress the blastogenesis of peripheral blood MNCs.

Polyclonal activation of T-cells with concanavalin A has been used as an in vitro test system to study immunosuppression induced by the avian retrovirus MAV-2-O. The mitogenic responsiveness of peripheral blood lymphocytes from infected chickens was only weakly suppressed, that of spleen lymphocytes was highly suppressed. Addition of conditioned media rich in T-cell growth factor (TCGF) activity to cultures of infected birds resulted in a reconstitution of the suppressed mitogenic responsiveness up to the level found in lymphocyte cultures from normal chickens. This indicates that the defect of the observed immunopathology is not at the level of responding T-cells. Measurements of TCGF production always showed reduced TCGF levels in the suppressed cultures suggesting that there is a defect at the TCGF production level. This is further supported by the failure to reconstitute the suppressed mitogenic response with normal macrophages, which are involved in activation of TCGF producer T-cells.

The properties of T-cell growth factor (TCGF), obtained by diucifon (Dc) stimulation of human mononuclear cells (MNC) (TCGF-Dc) have been studied. Taking into account the fact that Dc alone does not, like other TCGF inductors, cause proliferation, differences between TCGF-Dc and TCGF were suggested. Partial purification of supernatant from cells, activated by Dc was performed on Sephadex G-100 column. TCGF-Dc biological activity in these fractions was determined in the system of mitogen activated human MNC and mice thymocytes, as well as in the system of concanavalin A transformed cells. 2 peaks of TCGF-Dc activity have been revealed that are indicative of TCGF-Dc molecular mass heterogeneity. In contrast to TCGF, low molecular mass TCGF-Dc (8000-12000) and TCGF-Dc from the whole supernatant were capable of absorbing on intact human MNC. TCGF-Dc may be constantly present on MNC membrane, but TCGF-Dc fixation is not sufficient for proliferation induction, the receptor activation is necessary as well. Receptors to TCGF-Dc were suggested to consist of fixing and triggering sites.

Adherent peritoneal cells (APC) have often been used as a pure and effective macrophage population. Using partition analysis and small numbers of lymphoid cells activated by mitogens (concanavalin A for T cells (in the presence of TCGF) and LPS + DxS for B cells) we found that APC were accessory cells for T cell activation and growth but were not effective for B cells. Although APC were effective in assisting T-cell mitogenesis, they were not especially efficient. However, when APC were mixed with irradiated WEHI-3 cells (a tissue culture line previously shown to exhibit accessory cell activity in vitro for mitogenic activation T and B cells), the APC and WEHI-3 showed apparent synergy. One reason for failure of APC to assist B-cell mitogenesis was traced to the presence of a suppressor cell population which overcame the accessory cell help given by irradiated WEHI-3 cells to LPS-DxS stimulated murine B cells. It is thus possible to find "helper" effects (synergy of APC and WEHI-3 assisting the mitogenesis of T cells), as well as suppressor effects within the range of cells found in adherent accessory cells.

The functional and phenotypic characteristics of carcinomatous pleural or peritoneal lymphoid cells cultivated with either rIL 2 or TCGF have been investigated. The cultivation of the lymphoid cells with cytokines was initiated by a mixture of coexisting, viable carcinoma cells for 14 days. Results have indicated that cytokine-activated lymphoid cells from malignant pleural and peritoneal effusions showed considerable cytolytic activity against K562 and Daudi cells. The cell population responsible for LAK and/or CTL effector cells of TCGF-activated lymphoid cells were CD8+ CD11- cells. Further, in rIL 2-expanded cultures from pleural and peritoneal lymphoid cells, the CD4+ Leu 8- population was found to contain effector cells of cytotoxic activity against the tumor cells. It further was seen that the TCGF-activated CD8+ CD11- T cells possessed a more potent killing activity, in comparison to the rIL 2-activated CD4+ Leu8- T cells. However, rIL 2-activated lymphoid cells from ascites in liver cirrhosis (used for a control) showed a higher tumoricidal activity.

Analysis of immune regulation has been greatly facilitated by the development of clones of antigen reactive T cells. We have been analysing the regulation of these clones both by antigen, and by other cells, and review here the salient features of regulation of helper clones. Antigen at high concentrations may induce tolerance in a helper clone, in the absence of other T cells or accessory cells. This process was markedly antigen dose and time dependent, specific, and lasted at least 7 days. A key features was that response to TCGF was not altered. These experiments indicate that suppressor cells need not be involved in all forms of tolerance, and that antigen can interact directly with T cells. Regulation by an autologous anticlone raised by in vitro immunization of PBL with an irradiated clone is discussed. Its specificity for the immunizing clone suggests that it may recognize receptors, and suppressor clones of this type may be a possible strategy for regulating unwanted clones of lymphoid cells in autoimmunity or leukaemias. The implications of these findings for cancer in general are discussed.

Cutaneous T-cell lymphomas define a spectrum of disorders associated with T-lymphocytic proliferation with clinical manifestations occurring in the skin during the course of the disease. This review has dealt with two rather uncommon disorders, namely mycosis fungoides and Sezary syndrome which are indolent malignant lymphomas, occurring primarily in the fifth decade, and affecting males most frequently. Historically, mycosis fungoides and Sezary syndrome have been described for a relatively short time. As witnessed by Table 2, little was known concerning these disorders, other than clinical and pathologic features, until the application of immunologic, cell biologic, and cytogenetic technology which burgeoned a multitude of questions. The discovery of TCGF has allowed for both continuous growth of normal and neoplastic T cells and for the clonal expansion of some malignant clones. The establishment of these continuous cultures allowed for: (1) investigation of the mechanism of TCGF production and stimulation of T-cell growth, and (2) identification of HTLV, a retrovirus found in cell cultures from two patients with CTCL, and subsequently from patients with Japanese adult T-cell lymphoma. In addition, the HTLV has been related to a more virulent form of T-cell malignancy. The exact etiologic role of this virus in the CTCL is presently the subject of intense investigation. Through the use of immunologic methods the malignant cell of CTCL has been pheno-typically and functionally characterized as a "helper/inducer" subtype (E rosette+, anti-T-cell antisera+, T11+, T1+, T3+, 3A1-, T6-, T8-) and usually Ia-, HLADR-. Clinical manifestations of the phenotype may be clinically apparent in the serologic abnormalities present in these disorders. Utilizing these methods to investigate these disorders may provide a key to the understanding of T-cell function and cellular immunity much as myeloma provided a model for the understanding of B cells and immunity. Clinically and pathologically, these disorders behave as malignant indolent lymphomas with spread from localized cutaneous lesions to extracutaneous involvement of the blood, lymph nodes, and viscera culminating in the death of the patient from either organ dysfunction or infectious complications. At autopsy, this extracutaneous involvement is more pronounced than what was expected ante-mortem. Application of prospective staging techniques employing such special procedures as E-rosette cytology, cytogenetics, and electron microscopy in addition to usual light microscopy studies has demonstrated a greater percentage of extracutaneous involvement than otherwise expected.(ABSTRACT TRUNCATED AT 400 WORDS)

Cloned murine mammary tumor virus (MuMTV) sequences allowed us to search for murine mammary tumor virus related sequences in the DNA of surgically removed human breast tumors. Out of 28 tumors so far examined two were found to contain an Eco RI DNA fragment homologous to the long terminal repeat-group antigen (LTR-Gag) and the Envelope (Env) sequences of MuMTV. We have taken the lymphocytes of these patients and cultured them. Rapid growth of lymphocytes, mostly of T origin, occurred in the presence of T cell growth factor (TCGF). Whereas DNA extracted from fresh lymphocytes is negative, that extracted from the 3-day cultured lymphocytes showed MuMTV related sequences. Long term cultures of T cells and a similar culture derived from a healthy person donor were negative at all stages. DNA extracted from the Ebstein Barr Virus-transformed B cells of the patient does not contain the MuMTV related sequences.

Immunosuppressive substances which interfere with lymphocyte blastogenesis are released in vitro by embryos and endometrium from mares in early pregnancy. Immunosuppression was not evident when tissues were cultured in the presence of indomethacin (a prostaglandin-synthesis inhibitor). Various prostaglandins (PGs) were added to equine lymphocytes and lymphocyte proliferation was measured after the addition of concanavalin A (Con A) or phytohaemagglutinin A (PHA). PGE2 and PGF2 alpha inhibited Con A-induced blastogenesis down to final concentrations of 1.8 x 10(-9) M and 1.3 x 10(-6) M, respectively. Other PGs tested (6-keto-PGF1 alpha and 13,14-dihydro-15-keto-PGF2 alpha) did not affect Con A-induced blastogenesis. PHA-induced blastogenesis was inhibited by concentrations down to 1.8 x 10(-9) M PGE2, 3.3 x 10(-7) M PGF2 alpha and 2.8 x 10(-4) M 6-keto-PGF1 alpha. At all concentrations, 13,14-dihydro-15-keto-PGF2 alpha only slightly reduced PHA-induced blastogenesis. Therefore, PGE2 was the only prostaglandin tested which, at physiological concentrations, significantly inhibited incorporation of [3H] thymidine. The mechanism of PGE2-mediated suppression was studied by adding PGE2 and T cell growth factors (TCGF) to TCGF-responsive lymphocytes. PGE2 reduced the TCGF-mediated blastogenic response in a dose-dependent manner. Furthermore, culture supernatant from embryos and endometrium from 14-day pregnant mares inhibited lymphocyte blastogenesis induced by TCGF. These results show that PGE2 interferes with lymphocyte blastogenesis by TCGF-dependent mechanisms. It is suggested that the PGE2 present in the uterus of the early pregnant mare may be one of the factors involved in immunosuppression at the time of maternal recognition of pregnancy.

A 1000-fold purification of human T-cell growth factor (TCGF) was achieved starting from supernatants of human spleen cells stimulated with phytohaemagglutinin (PHA) in culture medium containing 0.5% serum. The purification scheme involved precipitation with ammonium sulphate, gel filtration and blue-Sepharose chromatography. The use of polyethylene glycol 6000 (PEG 6000) was critical during the chromatographic steps in order to obtain high final recoveries or activity (40-50%). Purified preparations of TCGF labelled with 125I by the chloramine T method revealed that the activity co-migrated with 2 molecular species of 14,000-17,000 daltons in SDS-PAGE under non-reducing conditions.

Fluid was aspirated from the preovulatory follicles of mares before and 12, 24 and 36 h after intravenous administration of hCG. Follicular fluid significantly (P less than 0.001) reduced lymphocyte blastogenesis in vitro and, at a dilution of 1:100, fluid collected at 36 h after administration of hCG was significantly more suppressive (P less than 0.01) than fluid collected before 36 h. Suppression of blastogenesis was reduced by extracting the follicular fluid with ether or by charcoal treatment (P less than 0.01) or by heating at 56 degrees C for 30 min (P less than 0.05). Preincubation of lymphocytes with 2 of 5 follicular fluid samples expressed subsequent blastogenesis. Follicular fluid inhibited blastogenesis of T-cell growth factor (TCGF)-dependent Con A lymphoblasts (P less than 0.05) and the degree of inhibition was related to time of addition of the TCGF and time of collection of the follicular fluid. These results indicate that preovulatory follicular fluid in the mare is increasingly suppressive to lymphocytes as time of ovulation approaches and that this immunosuppression is associated with an alteration of the response to lymphokine stimulation.

Rabbit T lymphocytes with cytotoxic activity for mouse target cells can be maintained in culture for several weeks by periodically restimulating with fresh mouse cells or by culturing with a source of TCGF. The former procedure greatly enriches the cultured cells for cytotoxic activity but has the disadvantage of introducing cellular debris. Cell growth and cytotoxic activity waned after 4-6 weeks using either procedure. However, T lymphocyte lines (TLL) could be isolated from limiting dilution cultures containing mitomycin C-treated mouse 'feeder' cells and a source of TCGF. These cells grew out with a frequency of one in 50-150 and approximately 15% of these growing cells had cytotoxic activity. Some cytotoxic cells lost their lytic activity after 4-5 months but others have persisted for over 7 months. Some TLL have now been growing for over 20 months.

The production of and responsivity to leukocyte-derived lymphokine-rich culture supernatants (SNs) was examined during the ontogeny of the frog, Xenopus. Thymocytes and splenocytes from adult frogs are capable of responding to the T-cell mitogen, phytohemagglutinin-P (PHA). Larval thymocytes are unresponsive to this lectin, whereas larval splenocytes are not. PHA-unresponsive thymocytes can be costimulated with PHA plus either a T-cell growth factor (TCGF)-rich SN or an interleukin-1 (IL-1)-rich SN (from cultures of lipopolysaccharide (LPS)-treated macrophage-enriched peritoneal cells (PCs). Stimulation of larval thymocytes with PHA does not produce a TCGF-rich SN as assayed by proliferation of lymphoblasts. Larval as well as adult splenocytes treated with PHA, however, do produce TCGF. These data are consistent with the hypothesis that the factor limiting mitogen responsiveness of larval thymocytes is the ability of cell populations in the thymus to produce rather than respond to either IL-1 or IL-2.

Human T-lymphoblasts maintained in culture for 1-2 months in T-cell growth factor conditioned medium continue growth when placed in fresh culture medium in the presence of phytohemagglutinin and irradiated adherent cells. In the absence of lectin, no growth occurs. Neither long-term cells nor irradiated adherent cells plus lectin produce soluble factors that stimulate long-term cell growth, but small amounts of this biologic activity can be detected in the culture medium when the two cell sets are mixed together. The data indicate that human long-term cell populations retain the ability to make enough TCGF to stimulate proliferation, but require physical contact with monocytes in the presence of phytohemagglutinin for this response to occur.

The increasing interest in the generation of lectin or antigen-activated (human) T cells by means of Interleukin-2 (TCGF) has led to numerous attempts to produce this substance. The simplest procedure is to use conditioned medium (CM) from IL-2 producing primary cells or cell lines with considerable activity. We describe a method of continuous CM-production in a 2-litre fermentation apparatus using a primate lymphoid cell line (MLA 144) which excretes IL-2 spontaneously. It was possible to harvest large amounts of homogeneous, lectin free material with similar production rates under different culture conditions. The activity of our CM was determined in various specific systems and proven to be comparable to that of supernatants of lectin-stimulated human primary lymphocytes.

Murine splenic cells were separated on Ig-anti-Ig columns to yield cell populations depleted of IgM-FcR (Fc receptor)-expressing (TM) and IgG-FcR-expressing (TG) cells and also cells expressing neither of these two FcR (TO). The proliferative response to the lectin concanavalin A (Con A) was shown to be present mainly in cell populations depleted of TM cells (T cells lacking IgM FcR)--that is, TM and TO. The low Con A response in the other cell populations, namely TG and TTOT, was due to a selection against T-cell growth factor (TCGF)-reactive cells, since TCGF production within these subsets was not impaired and the low proliferative response was not reconstituted by preformed TCGF. TG-cells were able to inhibit the Con-A-induced proliferation of TO and TM cells dose-dependently. These data strongly suggest the presence of a step-one regulatory, suppressive mechanism comprised within the TM population.

T-cell growth factor (TCGF) produced by the MLA 144 gibbon ape T-lymphosarcoma cell line was biosynthetically radiolabelled with [35S]methionine and isolated by reverse-phase h.p.l.c. [Milstone & Parker (1983) Biochem. Biophys. Res. Commun. 115, 762-768]. Two predominant species, of Mr 16300 and pI 6.8 and of Mr 14300 and pI 7.5, were resolved. Analysis of TCGF labelled with a mixture of [3H]glucosamine and [14C]methionine demonstrated that only the 16300-Mr protein contained detectable carbohydrate label, approx. 50% of which was present in sialic acid residues, which were largely responsible for the charge difference observed between the two forms of TCGF. The 14300-Mr protein was labelled within 5 min after addition of [35S]methionine and was the predominant intracellular form of TCGF, whereas the 16300-Mr protein was not detected until 25 min later and was the predominant extracellular form of TCGF. Pulse-chase experiments were consistent with the hypothesis that the 14300-Mr protein is an intracellular precursor that is glycosylated to form the 16300-Mr protein, which is then preferentially secreted from the cell.

L4-PHA (L4) and E4-PHA (E4) lectins isolated from Phaseolus vulgaris have different mitogenic properties. The mechanisms of the differences in mitogenic behavior were sought in the interaction of lectin, lymphocyte subsets, and T-cell growt factor (TCGF) also known as interleukin 2 (IL-2). TCGF activity in culture supernatants ( L4S ; E4S ) from L4- and E4-stimulated, freshly isolated lymphocytes was assayed as stimulation of DNA synthesis in TCGF-dependent continuous T-cell cultures (CTC). E4S contained less TCGF than did L4S . Addition of partially purified TCGF does not increase the stimulation of fresh lymphocytes by L4 or E4. L4 and E4 equally stimulate both helper (OKT4+) and suppressor (OKT8+) cells. The ability of L4 to further stimulate CTC is slowly lost (15 greater than 30 greater than 45 days). It is concluded that production of TCGF is not rate limiting in E4 and L4 stimulation of lymphocytes. The growth of CTC, which requires the presence of TCGF, remains sensitive to, but not dependent on, L4 for at least 30 days.

Interleukin 2 (IL2) or T cell growth factor (TCGF) has been characterized in a number of species but not in porcines. Porcine IL2 was detected in supernates (SN) of cultures of pig lymphocytes by: 1) the stimulation of the IL2-sensitive murine T cell line, CT6; 2) a costimulator assay involving porcine thymocytes; and 3) by the in vitro maintenance of antigen or mitogen-induced porcine lymphoblastoid cells. Porcine IL2 production by pig lymphocytes was induced by the mitogens Concanavalin A (Con A) Phytohemagglutiniin (PHA), and Pokeweed mitogen (PWM), but not by lipopolysaccharide (LPS). IL2 activity was demonstrated in the SN of mitogen-stimulated lymphocyte cultures as early as 24 hr after initiation of culture, reached peak levels at 48 hr, and decreased by 72 hr. Mitogens induced IL2 secretion by pig peripheral blood mononuclear cells, lymph node cells, and spleen cells, but not thymus cells. The cells responsible for IL2 production are presumptive T cells because: 1) they are nylon wool non-adherent; and 2) are non-surface-Ig bearing. In contrast, SN from cultures of surface Ig-positive cells had minimal IL2 activity. Porcine IL2 resembles rat and human IL2 in that it has an apparent molecular weight of approximately 15,000, and does not bind to DEAE-cellulose (DE-52) ion exchange columns equilibrated in 0.05 M sodium phosphate buffer (pH 7.6).

The authors have established a murine malignant glioma-specific cytotoxic T lymphocyte clone (G-CTLL 1) by T cell growth factor (TCGF) using 203-glioma (a methylcholanthrene-induced ependymoblastoma of C57BL/6 mouse origin). The cloned cells were found to release a large amount of gamma interferon (IFN) in response to glioma-associated antigen-specific stimulation. The authors have investigated whether the IFN produced can contribute to killing the target cells. Adding anti-mouse gamma IFN antibody to the mixed clone-target cell culture inhibited IFN production by the cloned cells but the toxicity of the cells was minimally diminished. Therefore, it is suggested that the endogenous gamma IFN produced by the TCGF-dependent cloned cytotoxic T lymphocyte line does not have direct cytotoxic action on the target cells. Furthermore, IFN production as well as cytotoxicity was blocked by anti-Lyt-2 monoclonal antibody in the absence of complement. This suggests that IFN plays a role in the process of antigen recognition of target cells because the Lyt-2 molecule is involved in an antigen-specific function on the cytotoxic T lymphocyte receptor. The role of TCGF in gamma IFN production was also investigated. The spontaneous production of gamma IFN by the cloned cells paralleled the amounts of exogenous TCGF added to the cultures, but TCGF had no synergistic effect on IFN production in the presence of mitogen or tumor antigen. Accordingly, it is possible that TCGF stimulates the cloned cells to proliferate, causing IFN release.

Supernatants from concanavalin A (Con A)-stimulated chicken spleen cells were used to generate a long-term cultured cell line from antigen-primed chicken peripheral blood leukocytes. This line has been kept in continuous proliferation in vitro for more than 25 weeks. Morphologically these cells were lymphoblastoid and expressed class I and class II antigens of the major histocompatibility complex as well as T-cell (but not B-cell or macrophage) antigens. In addition they contained no peroxidase or non-specific esterase activity, neither were they phagocytic. Proliferation of the line was totally dependent on exogenous T-cell growth factor (TCGF) activity provided by the Con-A-stimulated spleen cell supernatant, comparable with the proliferation of Con-A-induced T-cell blasts. TCGF activity from the supernatant was absorbed both by the long-term cultured T cells and by Con A blasts, demonstrating the presence of receptors for the same TCGF species on the two populations. We have used the long-term cultured cell line to characterize chicken TCGF further. The molecular weight of the biologically active fractions found by gel filtration on Sephadex G-100 was approximately 13,000 and isoelectric focusing showed chicken TCGF to have a pI of pH 5.9. We propose that the TCGF described here is the chicken analogue to the mammalian interleukin 2.

T-cell growth factor (TCGF) activity was studied in phytohemagglutinin (PHA)-stimulated peripheral blood mononuclear cells (PBMC) from 10 type-1 diabetic patients who had been diagnosed within the previous 12 months (group A), from 9 diabetic patients in whom the duration of disease was more than 1 year (group B) and from 12 healthy controls (group C). The effects of indomethacin on PHA-induced TCGF activity and the effects of adherent cells (macrophages) from group A and group C on TCGF production of normal group-matched non-adherent cells (lymphocytes) were also studied. TCGF activity was assayed on TCGF-dependent blast cells and calculated as a stimulation index (SI). TCGF activity in group A (SI 0.86 +/- 0.8) was significantly different from that in group B (SI 1.75 +/- 1.02; P = 0.037) and in group C (SI 1.91 +/- 1.29; P = 0.023). Following the addition of indomethacin, TCGF SI was 1.35 +/- 0.74 in group A, 1.85 +/- 0.73 in group B and 2.06 +/- 1.19 in group C. The responses to indomethacin were found to correlate with the basal TCGF activity in all subjects (r = -0.48; P = 0.006) independently of the disease process studied or its duration. No correlation was found between TCGF activity and parameters of metabolic control (HBA1c and fructosamine). Interestingly, a significant inverse correlation was found between TCGF activity and the required dose of insulin only in group A (r = -0.66; P < 0.05). Adherent cells from diabetic patients were found not to inhibit TCGF production.(ABSTRACT TRUNCATED AT 250 WORDS)

Cytotoxic T cells (CTL) have been known to be one of the effector cells responsible for regression of tumors. In tumor-bearing hosts, CTL insufficiently attack the tumor because the suppressor cells inhibit the induction and activation of CTL. In our in vitro study, CTL were induced from peripheral blood lymphocytes (PBL) of cancer patients. These CTL showed the specific cytotoxic activity against autologous tumor in one case and broad specificity against tumors in other cases. A patient with breast cancer was treated with adoptive immunotherapy of CTL because she had severe side effect from antineoplastics . Her breast cancer was histologically scirrhous type adenocarcinoma which was resistant to antineoplastics. Patient's PBL were cocultured with mitomycin C treated-autologous tumor, and they were proliferated with interleukin 2 or T-cell growth factor (TCGF). Then, these CTL were injected to this patient intravenously at the interval of one or two weeks. Before serial injection of CTL, antineoplastics was prescribed in order to inhibit the function of suppressor cells and cancer cells. The sizes of tumors were gradually reduced, suggesting clinical regression. We suggest that combined treatment of adoptive immunotherapy using CTL and antineoplastics (CTL therapy) is very useful in the treatment of cancer patients.