Type I, II, III, IV, V, and VI collagens serve as extracellular ligands for the isoforms of platelet-derived growth factor (AA, BB, and AB)

The biological activities of several growth factors/cytokines have been shown to be modulated by binding to molecules of the extracellular matrix. Here, the interactions of PDGF (isoforms AA, BB, and AB), a potent mitogen for mesenchymal cells, with collagens were investigated. All radiolabeled PDGF isoforms specifically interacted with type I, II, III, IV, V, and VI collagens (preferential binding to types III, I, VI, and IV) and their constituent chains, either when immobilized on polystyrene or blotted to nitrocellulose. PDGF-collagen interactions were of medium affinity (KD between 4 and 22 nM) and were inhibited by different soluble collagen chains suggesting a collagenous consensus binding site(s) for the PDGF isoforms investigated. Scatchard analysis revealed molar ratios of up to 3-4 PDGF molecules bound/triple-helical (native) collagen. Biological activity of collagen-bound PDGF was demonstrated by a 1.5-3-fold stimulation of proliferation of human fibroblasts and mouse 3T3 cells. Furthermore, a preferential association of PDGF with the collagenous extracellular matrix of cirrhotic liver could be shown by immunostaining. Our data are in accord with previous studies that localized PDGF in the extracellular matrix of fibroproliferative lesions and suggest that binding of PDGF to collagens may localize and modulate its biological activities.

Anti-idiotype cancer vaccines: past and future

Anti-idiotypic antibodies (Ab2) binding to the antigen-combining site of antitumor antibodies (Ab1) can induce anti-anti-idiotypic antibodies (Ab3) that specifically bind to the tumor antigen recognized by Ab1. Furthermore, Ab2, mimicking tumor antigens, have been shown to induce anti-anti-idiotypic proliferative T lymphocytes of the helper and suppressor type, as well as cytotoxic lymphocytes. The immunomodulatory activities of Ab2 have been demonstrated both in animals and in patients. The demonstration of tumor growth inhibition by anti-idiotypes in preclinical and phase I clinical studies emphasizes that randomized control trials should be performed to demonstrate clinical efficacy of Ab2 vaccines.

Monoclonal anti-idiotypic antibody functionally mimics the human gastrointestinal carcinoma epitope GA733

Anti-idiotypic antibodies (Ab2) that bind to the antigen-combining region of anti-tumor antibodies (Ab1) may functionally, and even structurally, mimic tumor antigen. We have previously demonstrated that polyclonal goat Ab2 directed against anti-human gastrointestinal carcinoma Ab1 GA733 induces anti-anti-idiotypic antibodies (Ab3) in animals that are Ab1-like in their binding specificity and idiotope expression. To obtain more defined Ab2 vaccines with potentially increased specificity and efficacy, a monoclonal Ab2 (FG1) was produced against Ab1 GA733 in rats. The monoclonal Ab2 FG1, similar to the polyclonal Ab2 described previously, induced Ab3 in rabbits that were Ab1-like in their idiotope expression and binding specificity to tumor cells and antigen. Antigen-specific Ab3 induced by Ab2 FG1 were easily detected in unprocessed rabbit sera, whereas the demonstration of such Ab3 after polyclonal Ab2 immunization required purification of the Ab3 from the rabbit sera. In addition, Ab2 FG1 induced antigen-specific humoral and cellular immunity in mice. Murine Ab3 bound specifically to antigen-positive tumor cells. Ab2-immunized mice showed antigen-specific delayed-type hypersensitivity (DTH) reaction, and cultured splenocytes from the immune mice demonstrated specific proliferation and cytokine (interferon-gamma and interleukin-4) secretion upon stimulation with GA733 antigen. However, immune mice were not protected against a challenge with syngeneic GA733 antigen-expressing colon carcinoma cells.

Induction of antigen-specific T and B cell immunity in colon carcinoma patients by anti-idiotypic antibody

Polyclonal goat anti-idiotypic Abs directed against anti-human gastrointestinal carcinoma mAb GA733 were administered to 13 colon cancer patients who had their primary tumor and lymph node metastases removed before immunotherapy. Patients received four s.c. doses (0.5 to 8 mg each) of alum-precipitated anti-idiotypic Ab. Seven of the 13 patients produced anti-anti-Ids that bound specifically to the GA733 epitope on tumor cells and shared idiotopes with mAb GA733. In four of the seven responding patients, anti-Id therapy specifically modulated T cell responses. In two patients who did not demonstrate GA733 Ag/anti-Id-reactive T cells before therapy, anti-Id administration induced CD4+, MHC class II-dependent T cells that specifically proliferated in culture in response to stimulation with either anti-Id or GA733 Ag. In two other patients who did demonstrate Ag/anti-Id-reactive T cells before therapy, anti-Id administration transiently induced lymphocytes that suppressed the proliferative responses of cultured pretherapy lymphocytes to stimulation with anti-Id or GA733 Ag. Nine of the 13 treated patients showed no evidence of disease after 39 to 86 mo of observation. Five of these patients developed Ag-specific Ab3 and one had, in addition, a T cell response.

Induction of antigen-specific T and B cell immunity in colon carcinoma patients by anti-idiotypic antibody

Polyclonal goat anti-idiotypic Abs directed against anti-human gastrointestinal carcinoma mAb GA733 were administered to 13 colon cancer patients who had their primary tumor and lymph node metastases removed before immunotherapy. Patients received four s.c. doses (0.5 to 8 mg each) of alum-precipitated anti-idiotypic Ab. Seven of the 13 patients produced anti-anti-Ids that bound specifically to the GA733 epitope on tumor cells and shared idiotopes with mAb GA733. In four of the seven responding patients, anti-Id therapy specifically modulated T cell responses. In two patients who did not demonstrate GA733 Ag/anti-Id-reactive T cells before therapy, anti-Id administration induced CD4+, MHC class II-dependent T cells that specifically proliferated in culture in response to stimulation with either anti-Id or GA733 Ag. In two other patients who did demonstrate Ag/anti-Id-reactive T cells before therapy, anti-Id administration transiently induced lymphocytes that suppressed the proliferative responses of cultured pretherapy lymphocytes to stimulation with anti-Id or GA733 Ag. Nine of the 13 treated patients showed no evidence of disease after 39 to 86 mo of observation. Five of these patients developed Ag-specific Ab3 and one had, in addition, a T cell response.

Animal models of human-derived cancer vaccines

Preclinical cancer vaccine studies must address vaccine safety, immunogenicity, and efficacy, as well as mechanism of vaccine action. Animal models of vaccines employing human tumor-associated antigen or epitopes (TAA, TAE) differ fundamentally from those employing tumor-specific antigens or epitopes (TSA, TSE). TSA and TSE vaccines will most likely demonstrate similar toxicity, immunogenicity, and efficacy in both tumor-bearing animals and patients. In contrast, TAA/TAE immunizations may have to overcome a host's immunological tolerance to TAA/TAE expressed not only on tumor, but also on normal tissues; immunity to TAA/TAE will potentially target normal tissues and thus may induce autoimmunity. Various experimental models for human-derived TAA/TAE vaccines have been developed. These models include transgenic mice, mice with severe combined immunodeficiency (SCID), and non-human primates. Recently, unique animal models of TAA/TAE cancer vaccines have been developed, taking advantage of the discovery of animal tissue antigens with significant sequence homologies to human TAA/TAE. These models mimic perhaps most closely the situation in cancer patients.

Tetanus toxoid-specific T cell responses in severe combined immunodeficiency (SCID) mice reconstituted with human peripheral blood lymphocytes

SCID mice reconstituted with human peripheral blood lymphocytes (PBL) have repeatedly been shown to produce antigen-specific B cell responses. We have derived tetanus toxoid (TT)-specific human T cell lines from cells of the peritoneal cavity, spleen and lymph nodes of SCID mice reconstituted with human PBL and boosted with TT. Establishment of these cell lines was dependent on the time interval between reconstitution of the mice with human PBL and initiation of lymphocyte cultures in vitro. When lymphocytes were collected from the mice 8 weeks after reconstitution, human lymphocytes with TT-specific proliferative activity in vitro were isolated from the peritoneal cavity and spleen, but long-term cell lines could not be established after repeated lymphocyte stimulation with TT, IL-2 and autologous Epstein-Barr virus-transformed B cells. In contrast, three long-term (> 10 months) TT-specific human T cell lines were established from lymphocytes collected from two of the eight mice in the group 4 weeks after reconstitution. The T cell lines were either CD4+ (two lines derived from peritoneal cavity and lymph node, respectively) or CD8+ (one line derived from spleen) and all expressed CD3, T cell receptor alpha/beta, and human histocompatibility leucocyte class I antigen. The T cell lines, however, lacked cytotoxic, helper and suppressor activities. Thus, SCID mice can support human T cells that actively migrate to various organs and respond to antigenic stimuli both in vivo and in vitro, but these T cells lack characteristic functions.

Low prevalence of hepatitis C virus infection in porphyria cutanea tarda in Germany

Previous studies from Spain, Italy, and France have demonstrated a high prevalence (71% to 91%) of antibodies against hepatitis C virus in patients with porphyria cutanea tarda (PCT). To determine the role of hepatitis C virus (HCV) in PCT in Germany, we have assessed the prevalence of antibodies against HCV and hepatitis B virus (HBV) in 106 patients (mean age, 60 +/- 14 years) with the disease. Eight of 106 patients (8%) were positive for HCV antibodies and HCV RNA using second-generation enzyme-linked immunosorbent assay (ELISA), recombinant immunoblot assay, and polymerase chain reaction. Antibodies against HBV core antigen were found in 14 patients (13%). Of the patients with antibodies against HCV alanine transaminase (ALT) (aspartate transaminase [AST]) levels above normal occurred in 71% (86%). Because elevated ALT (AST) levels were also found in 51% (64%) of 88 patients without markers of HCV or HBV, we suggest that liver damage in PCT may exist in absence of these viruses. This is supported by the finding that in patients without HCV or HBV markers, higher serum ALT and AST activities were found in patients with overt disease or relapse (ALT, 59 +/- 44 U/L; AST, 37 +/- 21 U/L), whereas patients in remission displayed significantly lower serum enzyme activities (ALT, 16 +/- 8 U/L; AST, 16 +/- 7 U/L), (P < .001). These results indicate that HCV infection does not play a major role in the pathogenesis of PCT in Germany.

Cloned antigens and antiidiotypes

Both monoclonal and polyclonal antiidiotypic antibodies mimicking the human colorectal carcinoma (CRC) associated antigen CO17-1A/GA733 have induced antigen-specific humoral and cellular immunity in CRC patients. The immune responses may underlie the clinical responses observed in some of the treated patients. Recently, the CO17-1A/GA733 antigen has been molecularly cloned and expressed in baculo-, adeno-, and vaccinia viruses. In preclinical studies, these recombinant antigen preparations elicited specific humoral immunity (cytotoxic antibodies) and cellular immunity (DTH-reactive and proliferative T cells). Antibody titers elicited in animals by recombinant antigen were significantly higher than those elicited by antiidiotypes. The recombinant antigen has a potential as a vaccine for CRC patients.

Limitations of the severe combined immunodeficiency (SCID) mouse model for study of human B-cell responses

Mice lacking functional T and B lymphocytes offer an in vivo animal model for the study of human immune functions. We have attempted to optimize the reconstitution of severe combined immunodeficiency (SCID) mice with human peripheral blood lymphocytes (PBL) using radiation, anti-asialo GM1 antibody or cyclophosphamide (Cy) treatment of the mice and in vitro stimulation of human PBL with interleukin (IL)-2 prior to their transfer to the mice. Total human IgG and tetanus-toxoid (TT)-specific human IgG responses of the mice were used as parameters of successful reconstitution. Treatment of the mice with anti-asialo GM1 antibody significantly enhanced total human IgG levels, but not TT-specific antibody responses, whereas irradiation or Cy treatment of the mice had no effect on human antibody production. In vitro treatment of human PBL with IL-2 prior to engraftment significantly decreased total human IgG responses of human PBL-grafted SCID mice. The immune responses of individual mice within a group were highly variable, which constitutes a major disadvantage of this model.

Immunomodulatory activity of monoclonal anti-idiotypic antibody to anti-colorectal carcinoma antibody CO17-1A in animals and patients

Monoclonal anti-idiotypic antibody (Ab2) VF2 was derived from rats immunized with anti-colorectal carcinoma (anti-CRC) monoclonal antibody (Ab1) CO17-1A. In rabbits the Ab2 induced anti anti-idiotypic antibodies (Ab3) that shared idiotopes with the Ab1, bound to the same epitope on CRC cells as Ab1, and bound to the isolated CO17-1A antigen. Monoclonal Ab2 VF2 was superior to the previously described polyclonal goat Ab2 against Ab1 CO17-1A in its capacity to elicit humoral immunity in animals. Ab2 VF2 also induced a specific delayed-type hypersensitivity (DTH) response to challenge with irradiated CO17-1A antigen-positive human CRC cells in mice. Of nine CRC patients immunized with aluminum hydroxide-precipitated Ab2 VF2, six developed antibodies that bound to Ab2, but only three patients developed Ab3 that bound to idiotypic determinants on Ab2. However, the Ab3 did not bind to CO17-1A antigen-positive CRC cells. In contrast, in a previously described trial with polyclonal goat Ab2 to Ab1 CO17-1A, most of the patients developed anti-CRC antibodies. Four of the nine patients immunized with Ab2 VF2 developed DTH responses to intradermal challenge with the Ab2, and in one patient DTH was both Ab2- and antigen-specific. Peripheral blood mononu-clear cells of the four DTH-reactive patients did not proliferate in response to in vitro stimulation with either Ab2 or antigen. These studies demonstrate that the immunomodulatory activity of monoclonal Ab2 VF2 in animals is only in part predictive of its activity in patients.

Anti-idiotype and recombinant antigen in immunotherapy of colorectal cancer

The CO17-1A/GA733 antigen (Ag), bound by monoclonal antibodies (MAb) CO17-1A and GA733 that define two different epitopes on the Ag, has proven a useful target in passive and active immunotherapy of colorectal carcinoma (CRC). Previous studies suggest that the antitumor effects demonstrated in MAb-treated patients may be mediated by idiotypic cascades. In approaches to active immunotherapy against the Ag, polyclonal goat and monoclonal rat anti-idiotypic antibodies (Ab2) directed against MAb CO17-1A or GA733 (Ab1) were administered as alum precipitates to 54 patients with CRC (stage Dukes' B, C, and D). The majority of the patients treated with the various Ab2 preparations developed anti-anti-idiotypic antibodies (Ab3) that specifically bound to the CO17-1A or GA733 epitope and shared idiotopes with the corresponding Ab1. Approximately 30% of the patients tested developed specific cellular immunity, i.e., Ag-specific T-cells mediating delayed-type hypersensitivity (DTH) reaction in vivo or proliferating on stimulation with the Ag in vitro. The humoral and cellular immune responses may underlie the clinical responses observed in some of the treated patients. Recently, the CO17-1A/GA733 Ag has been molecularly cloned and expressed in baculo-, adeno-, and vaccinia viruses. In preclinical studies, these recombinant Ag preparations elicited specific humoral immunity (cytotoxic antibodies) and cellular immunity (DTH-reactive and proliferative T-cells), similar to the native Ag. Antibody titers elicited in experimental animals by recombinant Ag were significantly higher than those elicited by Ab2, presumably because Ag expresses numerous epitopes, whereas Ab2 mimics a single epitope. Recombinant CO17-1A/GA733 Ag has potential as a vaccine for CRC patients.

Anti-idiotype cancer vaccines: pre-clinical and clinical studies

We have previously shown that anti-idiotypic antibodies (Ab2) that functionally mimic the epitope defined by anti-colorectal carcinoma (CRC) monoclonal antibody (Ab1) CO17-1A induce in cancer patients highly specific anti-anti-idiotypic antibodies (Ab3) which are Ab1-like in their binding specificity to tumor cells and antigen. Ab1 GA733, originally produced against gastric carcinoma, binds to the same antigen as Ab1 CO17-1A, but to a different epitope. Ab2 against Ab1 GA733 produced in goats have been previously shown in experimental animals to functionally mimic the epitope defined by the Ab1. For the purposes of studying immune responses, this Ab2 preparation was administered to 12 patients who had previously been diagnosed with CRC but whose tumors were excised prior to Ab2 therapy. Patients were injected subcutaneously with escalating doses (0.5-4 mg) of Ab2 precipitated to alum. Ten of the 12 patients produced antibodies to the administered Ab2. In six patients a fraction of these antibodies bound specifically to the Ab2 and not to normal goat IgG. These anti-anti-idiotypic antibodies (Ab3) shared idiotopes with the Ab1 and bound to antigen-positive, but not antigen-negative, cultured tumor cells. The Ab3 specifically inhibited binding of the Ab1 to tumor cells and therefore may bind to the same epitope as Ab1. Our studies demonstrate that Ab2 are highly specific modulators of cancer patients immune responses to their tumors.

Antibody dependent cellular cytotoxicity and complement mediated cytotoxicity on leukemic cells mediated by anti K562 monoclonal antibodies

Three anti K562 monoclonal antibodies (MAb) 4.3, 4.6 and 4.8 reacting predominantly with cells of myeloid lineage, were tested for antibody dependent cellular cytotoxicity (ADCC) and complement dependent cytotoxicity (CDC). MAb 4.6 (IgG3k) effectively mediated ADCC against K562 cells and fresh leukemic targets with effectors from healthy donors. However, for ADCC on chronic myeloid leukemia (CML) targets, effectors from CML patients in remission needed modulation with IL-2. All MAb showed significant CDC against peripheral blood (PB) and bone marrow (BM) cells obtained from CML patients in chronic phase, and untreated acute myeloid leukemia (AML) patients. MAb displayed no CDC against PB and BM cells from CML patients in remission and BM cells of Hodgkin's Disease (HD) patients with normal BM cellularity. In clonogenic assay, colony forming units (CFU) in the BM aspirate obtained from CML patients in chronic phase were significantly reduced by treatment with MAb and complement.

Monocyte/macrophage functions in patients with squamous cell carcinoma of the oral cavity

Peripheral blood monocytes and draining lymph node macrophages from patients with squamous cell carcinoma of the oral cavity, monocytes from patients with oral leukoplakia and those from healthy donors were assessed for FcR. HLA-DR expression and interleukin-1 (IL-1) production after activation with LPS or IFN gamma. Monocyte cytotoxicity was also tested after activation with LPS, IFN gamma, IL-2 singly, or in combinations at suboptimal concentrations. The results showed that the percentage of activated monocytes expressing FcR was significantly low in untreated oral cancer patients, however, the proportion of HLA-DR positive cells was normal. The unstimulated monocytes from oral cancer patients showed spontaneous generation of IL-1. Upon activation, few patients could produce IL-1 to normal levels. The unstimulated monocytes from untreated patients and treated patients with recurrence also exhibit significantly higher tumoricidal activity. Treatment of monocytes with combinations of two modulators (IFN gamma, LPS and IL-2) induced significantly higher cytotoxicity.

Interleukin-2 mediated regulation of mitogen-activated T cell reactivity from different lymphoid sources in patients with squamous cell carcinoma of the oral cavity

Peripheral blood lymphocytes (PBL) from patients with oral cancer (treated and untreated), oral leukoplakia and healthy donors; lymphocytes from metastatic and non-metastatic lymph nodes (met LNL and non-met LNL); and tumor infiltrating lymphocytes (TIL) were tested for proliferative response to mitogen PHA (phytohemagglutinin) and its augmentation by recombinant Interleukin-2 (rIL-2), for expression of Tac antigen (CD25) and for production of IL-2. Depressed PHA responses were found in PBL of treated and untreated patients, and in TIL. Addition of IL-2 could bring about 16% to 31% augmentation in lymphocyte response to PHA from all the three sources. PBL from 50% of healthy donors, 45% of patients with leukoplakia, 25% untreated oral cancer patients and 35% treated oral cancer patients showed IL-2 mediated augmentation of PHA response. While, 40% non-met LNL samples, 70% met LNL samples and only 23% TIL samples showed increased mitogen induced proliferation by IL-2. The augmented levels of PHA response of PBL from treated and untreated patients, and of TIL, were still below those of normal PBL. PBL from patients with leukoplakia, treated oral cancer patients and TIL showed depressed CD25 antigen expression. Depressed IL-2 production was observed only in PBL of leukoplakia patients. Thus the IL-2 mediated events of T cell activation from different lymphoid sources in patients with oral cancer did not correlate with their proliferative responses.

Propagation of cytotoxic effectors from chronic myeloid leukemia patients and cloning of cytotoxic T cells

Cytotoxic cells (CTCs) generated from peripheral blood lymphocytes of 5 chronic myeloid leukemia (CML) patients in remission on stimulation with autologous leukemic cells and allogeneic lymphocytes (3-cell assay), were propagated in vitro in interleukin-2 (IL-2)-containing medium and periodic stimulation with autologous leukemic cells, for a period of 4 to 6 months. During this period, the cells were assessed for phenotype and for cytotoxic responses in a 4-h 51Cr release microcytotoxicity assay. The CTCs continued to show specific lysis of autologous leukemic cells and bone marrow (BM) cells. However, the nonspecific lysis of natural killer (NK) targets and the proportion of cells showing NK phenotype (HNK-1 antigen) increased progressively on cultivation in IL-2-containing medium. Therefore cells showing CD8 phenotype and specific cytotoxic function were segregated by cloning CTCs under the condition of limiting dilution in the presence of allogeneic feeder cells and IL-2-containing medium. Three cytotoxic T cell (CTL) clones expressing CD3+, CD8+, and HLA DR+ phenotypes were obtained from CTCs of 2 CML patients. These clonoid populations, maintained in IL-2-containing medium and periodic antigenic stimulation with autologous leukemic cells, showed specific lysis of autologous leukemic cells and BM cells even at lower (10:1) effector:target ratios. They did not kill K562 (erythroblastoid leukemic NK target cell line) cells and autologous phytohemagglutinin-induced blasts. These clones apparently functioned in an MHC-restricted manner as they did not lyse allogeneic CML cells which would also express a similar set of maturation antigens if sensitization was, as it appeared, against these antigens. Finally, interaction of autologous BM cells with CTL clones reduced the colony forming potential of BM cells only to the extent of 18%-30%. The results therefore indicate that such CTL clones can possibly be used in adoptive immunotherapy as they showed minimal BM toxicity.

In vitro generation of effector cells cytotoxic to autologous targets from chronic myeloid leukemia patients in remission

Peripheral blood lymphocytes (PBL) from chronic myeloid leukemia (CML) patients in remission were stimulated in vitro, in a 3-cell assay with autologous leukemic cells or autologous bone marrow (BM) cells alone, or each in combination with allogeneic PBL. The responder cells were used as effectors in a 4-h 51Cr release cytotoxicity assay using autologous targets such as leukemic cells, BM cells, phytohemagglutinin-induced lymphoblasts, and allogeneic K562 (erythroblastoid leukemic cell line) target cells. Sensitization of lymphocytes from CML patients with either autologous leukemic cells or BM cells generated cytotoxic cells (CTCs) capable of killing both the targets. These results suggested that in CML, the PBL may have been sensitized to myeloid maturation-releated antigens in vivo, which, on secondary stimulation in vitro, may result in differentiation of CTCs cytotoxic to immature myeloid cells, either from autologous leukemic cells or autologous BM. The inability of PBL from patients with oral cancers to lyse autologous BM cells upon in vitro stimulation, supported this possibility. Clonogenic assays conducted to assess the colony forming potential of BM cells which had interacted with CTCs indicated that there was about 37% reduction in committed granulocyte stem cell colony formation without an appreciable change in committed granulocyte/monocyte stem cell units and clusters. Therefore, since the BM toxicity of the CTCs is not very high, these cells may have a potential clinical use in CML.

Effect of BM 12 531 on in vitro induction of suppressor cells by concanavalin A in cancer patients and normal healthy donors

The effects of BM 12 531, a 2-cyanaziridinyl derivative, on in vitro generation of Con A induced suppressor cells as well as generation of spontaneous suppressor cells in peripheral blood lymphocytes of 9 chronic myeloid leukemia (CML) patients in remission, 9 patients with active Hodgkin's disease (HD) and 12 normal healthy donors were studied. The suppressor cells generated spontaneously and with Con A, in the presence and absence of the drug, were tested for their modulating effect on mitogenic (PHA) response of autologous lymphocytes. The results indicate that the addition of the drug at the time of generation of suppressor cells decreased the spontaneous suppressor cell activity in 2/4 healthy donors, 3/7 CML patients and 4/6 HD patients. Con A induced suppressor cell activity generated in presence of the drug was significantly reduced in 7/12 healthy donors, 6/9 CML patients and 3/9 HD patients.

Concanavalin A induced suppressor cell activity and autorosette forming cells in chronic myeloid leukemia patients

In the present paper attempts have been made to investigate suppressor cell activity in CML patients in first and subsequent remissions in order to study the relationship between suppressor cell activity and progression of the disease. For this purpose, the ability of Con A activated suppressor cells from peripheral blood of CML patients in 1st, 2nd and 3rd remission to suppress PHA response of autologous lymphocytes is investigated and compared with that of normal healthy donors. The ability of Con A activated cell population to form rosettes with autologous RBCs (ARFC) is also investigated. The results indicate that lymphocytes from CML patients in 1st (61.8 +/- 6.1%), 2nd (62.6 +/- 3.0%) and 3rd (55.3 +/- 4.8%) remissions show significantly high suppressor cell activity than normal healthy donors (36.5 +/- 1.9%) when activated with Con A. Similarly, generation of spontaneous suppressor cell activity was also higher in 1st (23.3 +/- 4.7%) and 2nd (25.3 +/- 4.2%) remission lymphocytes than controls (10.1 +/- 2.5%). In the 3rd remission however, the spontaneous suppressor cell activity (14.5 +/- 3.2%) was comparable to controls. Thus it appears that a higher suppressor cell precursor population is present in CML patients in remission. However, this could not be correlated with the progression of the disease. CML patients in 1st remission also revealed an increased percentage of ARFC which correlated with the suppressor cell function. The ARFC activity tested in a few patients in subsequent remissions was comparable with controls although functional suppressor activity was increased.