Cell interactions in the suppression of in vitro antibody responses.

Normal T and immune B lymphocytes interact in a fashion that leads to suppression of the immune response. Normal spleen cells added to cultures of primed spleen cells specifically suppressed both the IgM and IgG secondary antibody response of the primed cells to less than 30% of the response of the immune cells cultured alone. Cell crowding as a possible in vitro artifact was ruled out. The suppression was specific for the priming antigen, even when the specific and nonspecific antigens were included in the same cultures. Suppression required both normal T and immune B cells to be present in culture. We suggest that the immune population produces a signal that can induce normal T cells to become specific suppressor cells. This form of interaction may represent an important regulatory (homeostatic) mechanism in the immune system.

Evolution of the expression of fetal acinar antigens during carcinogenesis of the pancreas in hamsters: individual follow-up by open biopsy.

Individual lesions in the pancreas and the presence of fetal acinar antigens along with carcinogenesis induced by N-nitrosobis(2-oxopropyl)amine (CAS: 60599-38-4) were studied by open biopsy in 16 Syrian golden hamsters 13, 22, and 40 weeks after initiation of treatment. At 13 weeks, cystadenoma and regular ductal hyperplasia were noted in 3 animals and 1 animal, respectively. Staining for fetal acinar antigens in the pancreas was found in 69% of the hamsters. At 22 weeks, cystadenoma and hyperplastic ducts were common (60 and 53%), and 3 hamsters developed pancreatic adenocarcinoma. Fetal acinar antigens persisted in the acini and extended to irregular hyperplastic ducts and tumor cells. At 40 weeks, ductal proliferation was the main lesion in all pancreatic tissue, and 9 animals had adenocarcinoma. Acinar antigens were found in the remaining acini, in irregular hyperplastic ducts, and in tumor cells. Thus, once reexpressed, fetal acinar antigens persist in pancreatic lesions and pancreatic carcinomas in the hamster.

Enhanced activity of murine peritoneal cells after aclacinomycin injection: characteristics of the enhanced respiratory burst.

After the i.p. injection into normal mice, of 4 mg/kg of aclacinomycin (ACM), a dose which prolongs the survival of tumor-bearing mice, the zymosan-elicited chemoluminescence (CL) of the peritoneal cells (PC) is greater than that of control cells. The volume in which the drug is administered plays an important role in the intensity of the response. ACM also stimulated the CL of PC from tumor-bearing mice. It is known that CL can also be elicited by soluble stimuli such as 4 beta-phorbol-12-myristate-13 alpha-acetate or Ca2+ ionophore A 23187, which, however, act in different ways. The response of ACM cells to these stimuli is also greater than in control cells. The enhanced CL of ACM-treated cells can be inhibited by incubating in vitro the zymosan-triggered PC with superoxide dismutase (300 units/ml) and catalase (2750 units/ml), but not with ethanol (20 microM) or potassium cyanide (100 microM). This indicates the participation of O2- and H2O2 in the CL of ACM-treated cells, whereas mitochondrial respiration does not appear to be involved. Furthermore, the following facts suggest the participation of arachidonic acid metabolism in the control of CL: (a) the in vitro addition of nordihydroguaiaretic acid (7 x 10(-6) M) and indomethacin (10(-3) M) inhibits the CL, while indomethacin (10(-6) M) has the opposite effect; (b) the PC from normal or ACM-treated mice when stimulated with zymosan secrete high amounts of prostaglandin (PG); (c) treated cells secrete the same amounts of PGE2 and 6-keto-PGF1 alpha but the secretion of PGF2 alpha and particularly of thromboxane B2 is greater in treated cells than in control cells and indomethacin (10(-6) M) strongly inhibits PG secretion in all groups; (d) in vitro addition of PGE2 at a concentration of 10(-6) M has an inhibitory effect on the CL emission of control and of treated cells, but it does not have this effect at lower concentrations (10(-8) M). These data suggest that the lipoxygenase pathway of arachidonic acid metabolism may be involved in the triggering of CL of ACM-treated cells, as well as that of normal cells, whereas products of the cyclooxygenase pathway may act as feedback inhibitors.

Enhanced activity of mouse peritoneal cells after aclacinomycin administration.

We have investigated the activation of mouse peritoneal macrophages by injection of aclacinomycin (ACM). Macrophages from ACM-treated mice have an increased phagocytic activity as measured by Candida ingestion. The microbicidal activity indirectly evaluated by chemiluminescence and superoxide determination in response to stimulation with zymosan and 4 beta-phorbol-12-myristate-13 alpha-acetate is also greater in the cells from treated mice. Direct measurement of the cytostatic function, and of in vitro and in vivo cytotoxicity shows comparable significant increases against L1210 or P815 target cells. The enhanced antitumoral activity could not be attributed to the residual presence of ACM in the peritoneal cells since no drug was detected by high-performance liquid chromatography and since their freeze-thaw lysates incubated with P815 cells did not modify the growth of tumor cells as measured by [3H]-thymidine incorporation. We also checked that the presence of ACM did not influence the intensity of the chemiluminescence. In all tests performed, only i.p. ACM administration could stimulate the peritoneal cells. Since the doses of ACM inducing an increase in macrophage activity are effective on the survival of tumor-bearing mice, the participation of this mechanism in tumor control might be suggested.

Enhanced activity of peritoneal cells after aclacinomycin injection: effect of pretreatment with superoxide dismutase on aclacinomycin-induced cytological alterations and antitumoral activity.

Peritoneal macrophages from mice injected with aclacinomycin (ACM) (4 mg/kg, i.p.) showed increased functional activity, as assessed by increased antitumoral activity in vitro and in vivo and zymosan-triggered chemoluminescence. They also showed ultrastructural signs of activation (increased number of cytoplasmic organelles), and atypical alterations (giant vacuoles and giant lysosomes containing heterogenous myelinoid bodies, lipofuscine-like substance, cytoplasmic debris, and a fine granular material). As these atypical alterations could be due to the generation of superoxide following ACM injection, superoxide dismutase (SOD) was injected 1 h prior to ACM administration. Neither the morphological characteristics of activation, nor the enhanced metabolic and antitumoral activities induced by ACM were affected by SOD pretreatment, but the atypical alterations were inhibited in a dose-dependent manner. Heat-inactivated SOD did not prevent their appearance. The atypical alterations were not found in peritoneal macrophages from talc or lipopolysaccharide-injected mice, but they were present in Adriamycin-treated mice and were also prevented by SOD pretreatment, indicating that the alterations are due to anthracycline treatment. Finally, [125I]SOD was phagocytized by peritoneal macrophages in vitro and in vivo and not by L1210 tumoral cells, explaining why the atypical alterations induced by ACM were no longer seen after SOD pretreatment. The unchanged direct oncostatic activity of ACM following SOD pretreatment suggests that this combination may have some wider perhaps clinical, potential.

In vivo modification of adriamycin-induced apoptosis in L-5178Y lymphoma cell-bearing mice by (+)-alpha-tocopherol and superoxide dismutase.

Apoptosis was followed in L5178Y lymphoma cell-bearing mice at different times after intraperitoneal injections of adriamycin (ADM). Apoptosis was determined morphologically and confirmed by DNA laddering on electrophoresis. Apoptosis was observed 36h after injection of 5mg/kg ADM (apoptotic cell index 64.2+/-5.6 vs. 1.5+/-2.1 from the untreated group) and confirmed by DNA electrophoresis. However, when the animals were pretreated with (+)-alpha-tocopherol acid succinate or superoxide dismutase before ADM administration apoptotic index significantly diminished (P<0.05) and the DNA electrophoresis did not show fragmentations. We conclude that in ADM-treated mice, tumour cell death occurs in two ways: first by necrosis, then later by apoptosis. These observations are likely to be associated with or caused by the generation of reactive oxygen species.

Urokinase-type plasminogen activator and plasminogen activator inhibitors (PAI-1 and PAI-2) in extracts of invasive cervical carcinoma and precursor lesions.

In a previous study we reported a direct correlation between the degree of total proteolytic activity and the natural history of cervical carcinoma. The present work examined whether an increase in the urokinase-type plasminogen activator (uPA) and the plasminogen activator inhibitors (PAIs) is correlated with the natural history of cervical carcinoma. We measured uPA and PAI-1 activities and uPA, PAI-1 and PAI-2 antigen concentrations in cervical extracts from normal, squamous intraepithelial lesions (SIL) or invasive carcinoma patients. The uPA activity in invasive carcinoma extracts was 8.46 times that of normal extracts and 4.9 times that of SIL extracts. The PAI-1 activity in invasive carcinoma extracts was 1.3 times that of normal extracts and 1.24 times that of SIL extracts. uPA, PAI-1 and PAI-2 amounts were 25.7-, 12.1- and 7.9-fold higher, respectively, in invasive carcinoma than in SIL, and 39.1-, 21.38- and 27.3-fold higher, respectively, than in normal extracts. uPA and PAI-1 activities were 2.02- and 1.42-fold higher in extracts from patients with stages II-IV than those from stage I extracts, respectively. uPA, PAI-1 and PAI-2 amounts were 3.06-, 4.2- and 1.4-fold higher in extracts from patients with stages II-IV than those from stage I extracts, respectively. The increase in uPA activity and the antigen levels of uPA and PAIs (PAI-1 and PAI-2) in stages II-IV of invasive carcinoma of the cervix suggests that these components play an important role in invasion and metastasis in advanced stages of this tumour.

Direct measure of the destruction of bone marrow cells, after their injection into variously pretreated syngeneic hosts.

125IUdR-labeled bone marrow cells have been injected into syngeneic mice and their destruction measured by counting daily the animals in toto in a gamma counter. In lethally irradiated recipients, the immediate cell loss was the greatest when the irradiation was delivered just prior to the cell transfer and the cellular multiplication in the spleen on day 7, the smallest. The slopes of the destruction between days 1 and 4 are closely comparable, whether the animals were irradiated 45, 21, or 4 hr before the cell transfer. The immediate cell loss was greater in untreated animals than in the cyclophosphamide-treated recipients and the slope of the destruction curve was higher in cyclophosphamide-treated than in untreated animals. The intensity of the restoration measured by the 125IUdR and 59Fe incorporation on day 6 is inversely correlated to the intensity of the early destruction.

Specific resistance to local graft-versus-host reaction in F1 hybrids pretreated intravenously with parent-strain spleen cells. Role of cytotoxic T lymphocytes directed against receptors for the major histocompatibility complex.

Pretreatment of B6D2F1 hybrids by injection of B6 spleen cells by subcutaneous or intravenous routes induces specific resistance to the local graft versus host reaction provoked by s.c. engraftment of B6 spleen cells. This resistance has been attributed to the presence in the pretreated F1 hybrids of cytotoxic T lymphocytes directed against receptors that recognize the D2 alloantigens (anti-D2-receptor CTL). However, this hypothesis would seem to be challenged, at least partially, by our previously published results showing that a) when tested before induction of GVHR, the anti-D2-receptor CTL are detectable in F1 hybrids pretreated only by the s.c. route but not by the i.v. route; and b) specific resistance to GVHR observed in i.v.-pretreated F1 hybrids is mediated by a nylon-adherent, Thy-1-, radioresistant (2000 rads) suppressor cell of B6 origin that does not manifest any anti-D2-receptor CTL activity. However, these results did not allow us to exclude the possibility of the presence, in the i.v.-pretreated F1 hybrids, of anti-D2-receptor precursor CTL that could be reactivated during the GVHR by the D2-receptors expressed on the proliferating clone of grafted B6 cells, then differentiate to the receptor-specific CTL effectors that control the development of the GVHR. That is why we have studied in the present work the CTL activity developed against D2-receptors after induction of GVHR in either normal or resistant F1 hybrids. Our results show that F1 hybrids protected against GVHR by i.v. pretreatment with B6 cells or by a transfer of nylon-adherent spleen cells from i.v.-pretreated syngeneic F1 mice do not manifest enhanced anti-D2-receptor CTL activity. When considered along with our previous observations, these results favor our hypothesis that anti-D2-receptor CTL are not involved in the specific resistance to GVHR observed in the i.v.-pretreated F1 hybrids.

Inhibition of hybrid resistance by 5-fluorouracil. Destruction of a Thy-1+ Lyt-1+2- spleen cell implicated in the expression of hybrid resistance.

Hybrid resistance designates the poor proliferation of parent-strain bone marrow cells injected into lethally irradiated hybrids. While testing a variety of oncostatic drugs for their capacity to inhibit hybrid resistance, we found that 5-fluorouracil (5-FU) injected into B6D2F1 hybrids three days before grafting enhances the growth of parent-strain B6 bone marrow cells. Hybrid resistance can be restored in these 5-FU-treated B6D2F1 recipients by the injection of spleen cells from untreated (normal) B6D2F1 mice. Treatment of these normal B6D2F1 spleen cells with anti-Thy-1 or anti-Lyt-1 antibodies plus complement abolishes their capacity to restore resistance, whereas treatment with anti Lyt-2 antibody plus complement has no effect. It is known that the rejection of parent-strain bone marrow cells by the F1 hybrid is a multistep process in which more than one cell type is involved, and that the final effector cell is asialo-GMI+, cyclophosphamide-sensitive, and responsive to IFN (possibly the natural killer cell). The Thy-1+Lyt-1+2- spleen cell we describe here, which is eliminated by the 5-FU, doesn't seem to be the final effector cell for hybrid resistance since transfer of spleen cells from cyclophosphamide-treated (final-effector-cell-depleted) B6D2F1 hybrids into 5-FU treated B6D2F1 mice restores hybrid resistance just as well as do normal B6D2F1 spleen cells; and, when injected into 5-FU-treated B6D2F1 hybrids, the IFN inducer pI:pC also restores resistance. The cell we now describe would be a third cell type, probably responsible for the initial recognition of grafted parent-strain bone marrow cells, which triggers the mechanism of hybrid resistance. This hypothesis could explain the observed specificity of parent-strain bone marrow graft rejection by F1 hybrids.

Specific and nonspecific resistance to local graft-versus-host reaction in F1 hybrids pretreated intravenously with parent-strain spleen cells. I. Two distinct mechanisms.

B6D2F1 hybrid mice pretreated i.v. with 5 X 10(7) spleen cells from B6 donors seven days earlier (B6-pretreated B6D2F1 hybrids) develop resistance to local GVHR induced by the subcutaneous injection of spleen cells of either B6 (GVHR-B6) or D2 (GVHR-D2) origin. This resistance has specific and a nonspecific components that concern the GVHR-B6 and the GVHR-D2, respectively. The two types of resistance to GVHR are neither induced under the same conditions nor mediated by the same mechanism. Specific resistance to GVHR is observed in B6D2F1 hybrids pretreated with unseparated, anti-Lyt-1.2+C' treated or 1000 rads-irradiated B6 cells, but not in B6D2F1 hybrids pretreated with anti-Thy-1.2+C' or anti Lyt-2.2+C'-treated B6 cells. In contrast, nonspecific resistance to GVHR is induced only by pretreatment with unseparated B6 cells. Treatment of B6 cells with anti-Thy-1.2, anti-Lyt-1.2, or anti-Lyt-2.2 moAb plus C', or their irradiation at 1000 rads completely abolishes their capacity to induce the nonspecific resistance to GVHR. Moreover, specific resistance to GVHR can be transferred to normal B6D2F1 mice by injection of nylon-adherent, anti-Thy-1.2+C'-treated or 2000-rads-irradiated, but not unseparated or nylon-nonadherent, B6-pretreated B6D2F1 spleen cells. Treatment of nylon-adherent B6-pretreated B6D2F1 cells with anti H-2d antiserum plus C' does not affect their capacity to transfer specific resistance to GVHR. Nonspecific resistance to GVHR can be transferred by unseparated, anti-Lyt-1.1+C' or anti Lyt-2.1+C'-treated, but not by anti-Thy-1.2+C' anti-Lyt-1.2+C', anti-Lyt-2.2+C'-treated or 2000-rads-irradiated B6-pretreated B6D2F1 spleen cells. Both types of resistance are observed in B6D2F1 hybrids pretreated with more than 2.5 X 10(7) B6 spleen cells.

Suppression of the in vitro parent antihybrid reaction by spleen cells from F1 hybrids pretreated with spleen cells from the opposite parent strain.

Spleen cells from B6C3F1 hybrid mice pretreated i.v. with 5 X 10(7) C3H spleen cells seven days earlier (C3H-pretreated B6C3F1) suppress the in vitro B6 anti-B6C3F1 proliferative and cytotoxic responses, when they are added to cultures of B6 responding and B6C3F1 stimulating spleen cells. This suppression is mediated by a Thy-1+Lyt-1+2+ cell of C3H origin that is radiosensitive at 2000 rads. This suppressor cell is not induced by the injection to B6C3F1 hybrids of spleen cells from the other parent strain (B6) or an allogeneic strain (D2). It does not suppress either the response of the other parent (C3H) or an allogeneic strain (D2) to B6C3F1 antigens, or the response of B6 cells to an allogeneic strain (D2). Its induction depends upon the number and the subpopulation of C3H spleen cells injected since suppression is observed after the injection of more than 2.5 X 10(7) C3H cells, and the suppression inducing cells have the phenotype Thy-1+Lyt-1+2+. This phenomenon is not limited to the C3H-B6C3F1 genetic combination, since it has been observed in all parent hybrid combinations tested to date.

Decreased natural killer cell activity after prolonged administration of interferon in cancer patients.

Two patients with metastatic neoplastic disease received 2-3 X 10(6) IU alpha recombinant interferon (IFN) 3 times/wk, every other week, for 3-6 mth. The natural killer (NK) activity of their peripheral blood leukocytes, was followed during the course of the treatment. A significant decrease was observed in the NK activity, which returned to normal values at the end of IFN administration. The treatment did not modify the evolution of metastasis.

Increased phagocytic activity of peripheral blood monocytes after intravenous injection of phospholipase A2 to monkeys.

One of the expressions of the activity of phagocytic cells such as monocytes or macrophages is a burst of increased oxidative activity on stimulation. The free oxygen radicals liberated, mainly O2- and H2O2, lead to chemoluminescence, which is thus a measure of activation. Chemoluminescence also depends on arachidonic acid metabolism, and this depends on phospholipase A2 (PLA2). We modified monocyte activity in monkeys by injecting them i.v. with this enzyme and observed that 30 min after injection, the phagocytic activity of peripheral blood monocytes and the chemoluminescence they emitted was greater than that of controls. We suggest that PLA2 may act as an in vivo immunomodulator in mammals.

The bactericidal capacity of peripheral blood monocytes from HIV positive patients may collapse very soon after the infection.

The activity of peripheral blood monocytes from HIV positive patients was measured by the intensity of the chemoluminescence those cells emit when they ingest a foreign particle. In most patients (84%) that value was impaired. Such an alteration may occur very early in the course of the disease. The anti-p24 antibody titer was correlated with monocyte phagocytic potential as measured by the chemoluminescent value thus indicating the need for adequate monocyte activity in order to obtain antibody formation. The severity of opportunistic infections that HIV positive subjects may develop is a clear indication that their immune systems are abnormal. The most frequently affected cells are those which bear the viral receptor, the CD4 antigen. Those cells are mainly the helper T cells and monocytes. The monocytes are the immune system phagocytic cells which actively control infections, in part by the release oxidative radicals. Those radicals can easily be measured by the chemoluminescence (CL) the cells emit when they ingest a foreign particle. This study examines the CL emitted by peripheral blood monocytes from HIV-positive and control subjects while they phagocytose opsonized zymosan in vitro and correlates these values with other laboratory parameters.

Enhanced production of interleukin 1 by mouse peritoneal macrophages after aclacinomycin administration.

The peritoneal cells of mice injected with aclacinomycin (ACM), an oncostatic drug of the anthracyclin family, were found to secrete more interleukin (IL-1), after two successive 24-h periods of in vitro LPS stimulation than those of control mice. This measured IL-1 production is one of the signs of enhanced macrophage activity. The cells of ACM-injected mice also contained more intracellular IL-1 than those of controls. In contrast, macrophages from ACM-injected mice only increased their IL-1 production after the first 24-h incubation with PMA, and not after the second 24-h incubation. The response to ACM was dose- and time-dependent. We have also compared the IL-1 production by macrophages from mice injected with other anthracyclins, at doses equimolar to that of 4 mg/kg ACM and we have observed that adriamycin, 4'-epiadriamycin and aclacinomycin had similar activity, while THP-adriamycin an daunorubicine were slightly more active. Exploitation of this increased IL-1 production by macrophages could be beneficial in the design of tumor treatment protocols.

Phospholipase A2, an in vivo immunomodulator.

Arachidonic acid (AA) can be released from membrane phospholipids by the action of phospholipase A2 (PLA2). There is evidence that unsaturated fatty acids, particularly AA, released from membrane phospholipids are required to activate the respiratory burst of macrophages. The data reported here indicate that peritoneal macrophages harvested 30 min after i.p. injection of PLA2 can phagocytose Candida albicans more efficiently and emit more chemoluminescence (CL) than normal cells when stimulated by zymosan. PLA2 injection also enhances the CL of peritoneal cells from mice already stimulated by immunomodulators such as trehalose dimycolate (TDM), bestatin, or oncostatic drugs such as aclacinomycin (ACM). CL is not sensitive to potassium cyanide (KCN), but is inhibited by catalase, superoxide dismutase (SOD), nordihydroguaiaretic acid (NDGA) and high doses of indomethacin (10(-3) M). In vivo PLA2 treatment stimulates the synthesis of both cyclooxygenase and lipoxygenase derivatives of AA metabolism (PGE2, 6-keto, PGF1 alpha TXB2 and LTC4). Inhibitors of AA metabolism (NDGA, indomethacin) modulate the production of free oxidizing radicals in this experimental model, partly because of their effect on AA metabolism, as determined by the measuring immunoreactive products. However, this work indicates that the effects of these inhibitors, which have been extensively used in CL studies, should be interpreted with caution, since their specificity for AA metabolism is relative.

Alterations of biological parameters in mice chronically exposed to low-frequency (50 Hz) electromagnetic fields.

In an experimental study we measured changes in hematological, biochemical and cortisol parameters in 6-week-old Swiss mice continuously exposed to ELF generated by a transformer station and high current bus bars. Mean daily exposure of 5.0 microT was maintained for 350 days. Hematological parameters were compared to those of control mice (n=12) exposed to a field level lower than 0.1 microT. Serum biochemical parameters (sodium, potassium, chloride, calcium, magnesium, phosphorus, amylase, creatine phosphokinase, and lactate dehydrogenase) were measured after 28 days of exposure and serum cortisol after 90 and 190 days. Granulocyte/macrophage colony-forming cells (GM-CFC) were counted at the end of the 350-day exposure. On day 20, exposed animals showed a significant decrease in leukocyte, erythrocyte, lymphocyte and monocyte counts and in hemoglobin and hematocrit values, while MCV increased. On days 43 and 63 no significant difference was observed in leukocyte and erythrocyte values, as if hemopoiesis had recovered. On day 90, a significant fall in the leukocyte, polynuclear neutrophil and eosinophil counts was observed in the exposed animals. No significant difference was noted in the biochemical parameters studied. On day 190, exposed animals had neutropenia and a decrease in the cortisol value. On day 350, no significant difference in hematological parameters was noted. Individual differences in sensitivity were observed, as 8 mice in the exposed group showed a significant decrease in the leukocyte, polymorphonuclear neutrophil and GM-CFC counts, while in two mice there was a significant increase in these same values compared to those unexposed mice.

Charcoal suspension for tumor labelling modifies macrophage activity in mice.

We have previously developed a charcoal suspension for injection into human breast cancers in order to facilitate their location during surgery. We observed that charcoal particles were ingested by intra and peritumoral macrophages, some of which carried the particles at some distance from the injection site. We studied the influence of the formulation parameters of the charcoal suspension for intratumoral injection on in vitro and in vivo activation and in vivo mobilization of mouse peritoneal macrophages after intra-peritoneal injection of 2 mL of each preparation. The influence of the charcoal origin (peat vs wood), granulometry, suspension vehicle (water for parenteral injection, vs saline), concentration and excipients were studied. Micronized peat charcoal in water for injection at the highest studied concentration reduced macrophage activation in vitro and in vivo. However, macrophage mobilization was weaker than after thioglycolate injection and did not seem to be charcoal dose-dependent. The additives incorporated in the charcoal suspension led in vivo to increased peritoneal macrophage activation and mobilization (mannitol, and glucose), only increased activation (polysorbate 80 and pluronic F68) or mobilization (dextran 40, egg lecithin, and cabosil), or inhibited both activation and mobilization (cremophor EL).

Kinetics of in vivo tumor cell destruction after specific immunization.

The natural cytotoxic activity in vivo can be evaluated by measuring in vivo tumor cell destruction after injecting mice with 125IUdR-labeled tumor cells, measuring their total body radioactivity and calculating the % radioactivity lost. We have studied the in vivo destruction of 125IUdR-labeled L1210 leukemic cells by B10.D2 mice previously immunized 4 times with heavily irradiated L1210 leukemic cells. Mathematical analysis of our results indicates that the radiolabel loss on day 1 is similar in normal and immunized animals, but that it stays greater over the following days in immunized animals, indicating that the difference between the two groups is not the extent of the initial cell destruction but the durability of the response. There is a good correlation between the eventual survival, the % of 125IUdR lost and the number of tumor cells present in the peritoneal cavity three hours after their injection. Such a methodology provides a very early prediction of the survival of each mouse, thus identifying animals with a poor prognosis.