Kinetics of divalent monoclonal antibody binding to tumour cell surface antigens using flow cytometry: standardization and mathematical analysis

Mathematical modeling the kinetics of cell distribution in the process of ligand-receptor binding

A statistical approach is presented to model the kinetics of cell distribution in the process of ligand-receptor binding on cell surfaces. The approach takes into account the variation of the amount of receptors on cells assuming the homogeneity of monovalent binding sites and ligand molecules. The analytical expressions for the kinetics of cell distribution have been derived in the reaction-limited approximation. In order to demonstrate the applicability of the mathematical model, the kinetics of binding the rabbit, anti-mouse IgG with Ig-receptors of the murine hybridoma cells has been measured. Anti-mouse IgG was labeled with fluorescein isothiocyanate (FITC). The kinetics of cell distribution on ligand-receptor complexes was observed during the reaction process by real-time measuring of the fluorescence and light-scattering traces of individual cells with the scanning flow cytometer. The experimental data were fitted by the mathematical model in order to obtain the binding rate constant and the initial cell distribution on the amount of receptors.

Reduced blood accumulation of biotinylated monoclonal antibody A7 after the subsequent administration of avidin

Clear immunoscintigraphy with radiolabeled monoclonal antibodies (MAbs) requires a high tumor tissue/blood ratio of radioactivity. In this study, we attempted to obtain a high tumor tissue/blood ratio by the active removal of radiolabeled MAb from the circulation, using the avidin-biotin system. Biotinylated 125I-labeled MAb A7 was injected intravenously into nude mice bearing a human colon cancer (WiDr) xenograft. Avidin was injected 24 h later. The tumor tissue/blood ratio of radioactivity was almost four times that of controls. These results suggest that biotinylated 125I-labeled MAb A7 and avidin are potentially useful for the rapid immunodetection of human colon cancer.

Regression analysis of simulated radio-ligand equilibrium experiments using seven different mathematical models

The objective of this study was to investigate the conditions for regression analysis of data from equilibrium experiments. One important issue was to recognize that Kd and the binding site concentration (A) are not of equal nature, although both are parameters in the regression analysis. Whereas Kd approximates to a true constant, A is subject to experimental variation due to pipetting errors and in solid-phase experiments also to uneven coating properties. While recognizing that the ideal assumptions for ordinary regression analysis are poorly satisfied, different regression models were evaluated by extensive simulations. It was first established by a 'worst case' investigation that a limited error (8%) in the dependent variable is not critical for the results obtained at curve-fitting to Langmuir's equation. Seven different equations were compared for the calculation of data representing a solid-phase equilibrium experiment with statistical but no systematic errors. All the equations are rearrangements of the law of mass action. In this setting the Scatchrd plot gave the best result, but also the double reciprocal and the Woolf plots worked well in weighted analysis. Langmuir's equation gave the best result of the 4 nonlinear regression models tested. The influence of one type of systematic error was also investigated. This assumed that 10% of the label was positioned on particles other than the functional ligand molecules. This systematic error was amplified, which resulted in a substantial bias. The calculated Kd-values varied slightly with the regression method used and were almost 24% too high in the best methods.

Correlation between CD29 density on CD8+ lymphocytes and serum IgG in systemic lupus erythematosus

The purpose of this paper is to establish whether there is increased lymphocyte adhesion molecule density in systemic lupus erythematosus (SLE), which could alter the migration pathways and activation thresholds of lymphocytes and thus contribute to the pathogenesis of the disease. We analysed the CD11a, CD29 and CD2 bound antibody molecule (bam) density on the CD4+ and CD8+ CAMhigh (primed) lymphocytes of 28 SLE patients (8 active and 20 inactive by BILAG), using reproducible flow cytometric measurements, standardized with fluorescent beads and antibodies of known fluorescein: protein ratios. In a second patient cohort (17 patients), we investigated whether CD29 density on CD8+ cells correlated with measures of humoral (serum IgG) or cellular (urine neopterin) activation. In the first cohort, 36% of patients had elevated CD29 (beta 1 integrin) density on CD8+ cells. In the second cohort, CD29 density on CD8+ cells was found to be closely associated with total plasma IgG (r = 0.71, P = 0.001), but not with urine neopterin, disease activity (BILAG) or drug treatment. We conclude that CD29 on CD8+ cells is associated with B cell activation in SLE.

Immunotoxins in the therapy of cancer: from bench to clinic

This review presents only those contributions that have progressed from the bench to the clinic using murine monoclonal antibodies coupled chemically to toxins, their subunits or ribosome-inactivating proteins. The rationale and progress in the development, characterization, preclinical testing and clinical trials are discussed.

Screening of monoclonal antibodies recognizing oncofetal antigens for isolation of trophoblasts from maternal blood for prenatal diagnosis

Many monoclonal antibodies have been produced against tumour-associated cell surface antigens for cancer therapy. They have therefore been selected for minimal reactivity with normal tissues and in particular for lack of binding to blood cells or serum components. Many of the antigens recognized are of fetal origin. These monoclonal antibodies may therefore be ideal candidates to recognize and sort fetal trophoblasts from maternal blood for prenatal diagnosis of genetic abnormalities. A panel of 19 anti-tumour antibodies were therefore screened for reactivity with early trimester placenta and two, 340 and 154, were shown to stain trophoblasts. If MAb 340 is linked to magnetic beads, it can efficiently sort trophoblast cell lines from whole blood.

Enhanced tumor localization of radiolabeled Fab fragments of monoclonal antibody A7 in nude mice bearing human pancreatic carcinoma xenografts

Much recent research has been directed toward the use of monoclonal antibodies (MAb) for the immunodetection of solid tumors. In pancreatic cancer, the results of conventional immunoscintigraphy using intact MAb remain disappointing. Clear immunoscintigraphy with radiolabeled MAb requires a high tumor tissue/blood ratio of radioactivity and a low normal tissue/blood ratio of radioactivity. In this study, 125I-labeled Fab fragments produced by papain digestion of MAb A7 were injected intravenously into nude mice bearing a human pancreatic cancer (HPC-YS) xenograft previously shown to react specifically with MAb A7. The radioactivity of tumors and normal organs was subsequently measured. The tumor tissue/blood ratio of 125I-labeled Fab fragments of MAb A7 was 1.00 +/- 0.24 and 9.68 +/- 2.54 at 2 and 24 h after injection, respectively. The tumor tissue/blood ratio of radioactivity was significantly higher than those of normal organs at 24 h after injection. Moreover, the tumor tissue/blood ratio of 125I-labeled Fab fragments of MAb A7 was greater than that of intact MAb A7, although the 125I-labeled Fab accumulation level was much less than that of 125I-labeled intact MAb A7 in the tumor. When mice bearing tumors which did not react with MAb A7 were studied, 125I-labeled Fab fragments did not specifically localize to the tumors. These results suggest that Fab fragments of MAb A7 may be suitable carriers of radionuclides for the immunodetection of human pancreatic cancer.

Treatment of leukemia with radiolabeled monoclonal antibodies

In contrast to radioimmunotherapy of solid disease, wherein the primary obstacle to success is access of radiolabeled antibody to antigen-positive cells, in the treatment of leukemia delivering a lethal absorbed dose to the isolated cell appears to be the primary obstacle. The isolated cell is defined as one that is exposed only to self-irradiation (from internalized or surface-bound radiolabeled antibody) and to irradiation from free antibody in the blood. It is isolated in the sense that the particulate (beta, electron, alpha) emissions from its nearest neighboring antigen-positive cell do not contribute to its absorbed dose. Disease in the bone marrow and other tissues, since it is confined to a smaller volume, is more easily eradicated because the absorbed dose to a given cell nucleus is enhanced by emissions from adjacent cells (a smaller fraction of the emission energy is 'wasted'). The optimization simulations presented above for the M195 antibody suggest that the optimum dose of antibody that should be administered is that required to yield a concentration within the distribution volume of the antibody that is approximately equal to the concentration of antigen sites as determined by the tumor burden. Although not specifically considered in the modeling example presented above, antibody internalization and catabolism may be expected to play an important role in radioimmunotherapy treatment planning of leukemia. Depending upon the kinetics of internalization and catabolism, the absorbed dose to the red marrow and to antigen-positive cells may be reduced considerably, since catabolism, assuming that it is followed by rapid extrusion of the radioactive label, would decrease the cells' exposure time considerably. The recently demonstrated effectiveness of radioimmunotherapy in certain cases of B-cell lymphoma and in reducing tumor burden in acute myelogenous leukemia suggests that radioimmunotherapy is beginning to fulfill the promise held when it was initially conceived. The long delay in achieving reproducible success has, in large part, been the result of the conceptual simplicity of using agents that specifically 'target' tumor cells and they may thus selectively deliver cytotoxic agents. Emboldened by this apparent simplicity, early trials of radioimmunotherapy failed to consider the many variables involved in its implementation. As has been recently demonstrated using mathematical models of antibody delivery to solid tumor, chief among these may have been the failure to select the appropriate tumor type. By significantly reducing the problems associated with antibody delivery, hematopoietic malignancies offer the optimum conditions for successful radioimmunotherapy. As evinced by the wide range of antibody and radioactivity doses administered in the B-cell lymphoma trials, the case-specific nature of radioimmunotherapy requires an understanding of the relationship between the various input parameters and patient response. The complexity and interrelationship of these parameters precludes an experimental trial-and-error approach to their optimization. A stepwise approach to radioimmunotherapy treatment planning is proposed in which a model of antibody kinetics is developed and validated.(ABSTRACT TRUNCATED AT 400 WORDS)

Monoclonal antibody targeting of methotrexate (MTX) against MTX-resistant tumour cell lines

Several Methotrexate (MTX)-resistant sublines of the osteogenic sarcoma cell line 791T were derived by continuous selection in the presence of MTX and 12-O-tetradecanoylphorbol-13-acetate (TPA). Studies including assays of the uptake and binding of [3H]MTX and fluoresceinated-MTX, determined that these sublines showed diminished MTX transport, and that none of them appeared to overproduce the MTX-target enzyme dihydrofolate reductase. Conjugates of the anti-791T monoclonal antibody 791T/36 linked to MTX via human serum albumin (HSA) were prepared by Dr M.C. Garnett. These were cytotoxic selectively for cells bearing the 791T/36-defined antigen (gp72), and were found to be as cytotoxic to most of the MTX-resistant 791T sublines as they were to parental 791T cells. Furthermore, an anti-MTX/anti-gp72 bispecific antibody 516 augmented the cytotoxicity of HSA-MTX conjugate to the MTX-resistant 791T variant R120 apparently as efficiently as for parental 791T cells. It is suggested that acquired drug resistance caused by deficient transport mechanisms may be partially or wholly overcome by targeting the drug to a readily-internalised cell surface antigen.

Equilibrium binding characteristics of monoclonal antibodies recognizing melanoma cell surface antigens

The equilibrium binding characteristics of a panel of six monoclonal antibodies (MAb) recognizing melanoma cell surface antigens (125 kdal cell surface melanoma associated glycoprotein antigen, 125kD-MAA; high molecular weight melanoma associated antigen, HMW-MAA; and a non-protein melanoma associated antigen, NP-MAA) were investigated using the cell lines SK-MEL-2, SK-MEL-5, and M21. The MAbs displayed equilibrium association constant (K) values ranging from 10(7) M-1 to 10(10) M-1 and maximum MAb binding values (Qmax) from 2 x 10(4) to 2 x 10(6) MAb molecules bound per cell. High trypsin concentrations were shown to have deleterious effects on Qmax values obtained for antibodies recognizing the 125kD-MAA, and even low trypsin concentrations affected Qmax values obtained for MAbs recognizing the HMW-MAA (although a complete linear recovery of HMW-MAA antigen was observed in 20-25 hours). Significant changes in Qmax were also noted for different cell passages. Except for MAb 43.2, little variation in K was observed when different cell lines were used. Linear Scatchard plots were obtained for all MAbs except 43.2 in which case concave down behavior was observed suggesting the existence of positive cooperativity between the binding sites of this MAb.

The effect of temperature on the binding kinetics and equilibrium constants of monoclonal antibodies to cell surface antigens

The effect of temperature on the kinetic association and dissociation binding parameters, and equilibrium constants of four monoclonal antibodies to the murine Ly-2.1 and Ly-3.1 antigens has been studied using flow cytometry. All four monoclonal antibodies were conjugated to FITC and their association to, and dissociation from, the surface of murine thymoma cells was observed at 15 sec intervals, at temperatures between 1 and 37 degrees C. The initial association rate constant and the dissociation rate constant for each antibody at each temperature were calculated from graphs of the first-order reactions and it was demonstrated that an increase in temperature caused an increase in both association rate and dissociation rate of the antibodies. Generally the increase in association rate with temperature was less than the increase in dissociation rate. Differences between antibodies to the same antigen (Ly-2.1) suggest that changes in membrane fluidity were not solely responsible for the changes in association rate. However, the equilibrium constants (Keq) did not always show a simple relationship of increasing temperature causing decreasing values for Keq. For one antibody the highest value for Keq was seen at 17 degrees C rather than at 37 degrees C and differences in Keq between individual antibodies were greater at 1 degree C than at 37 degrees C. Kinetic rate constants are usually measured at 4 degrees C or room temperature, therefore for antibodies under consideration for in vivo use, measurements at 37 degrees C are more appropriate.

A bispecific monoclonal antibody against methotrexate and a human tumour associated antigen augments cytotoxicity of methotrexate-carrier conjugate

A bispecific monoclonal antibody, reactive with methotrexate (MTX) and a tumour associated antigen (gp72) has been produced by fusing spleen cells from MTX immunised mice with 791T/36/3 (anti-gp72) hybridoma. The hybrid antibody was purified from anti-MTX and anti-gp72 antibodies present in the hybridoma culture supernatant by combinations of affinity chromatography on a MTX-agarose immunoabsorbent and stepwise acid elution from Sepharose-Protein A. A particular feature of the present antibody is that it reacts with conjugated MTX; this would allow in vivo targeting of conjugates, increasing many fold the number of molecules of drug carried or localising to pre-targeted antibody. Dual binding between tumour cell surface antigen and MTX was demonstrated by the ability of the hybrid antibody to bridge between tumour cells and MTX as MTX-HSA conjugate, reaction here being detected by flow cytofluorimetry. Purified hybrid antibody specifically enhanced the in vitro cytotoxicity of MTX-HSA for gp72 positive tumour cells.

Human non-lineage antigen, CD46 (HuLy-m5): purification and partial sequencing demonstrates structural homology with complement-regulating glycoproteins

CD46, until recently known as HuLy-m5, is a non-lineage restricted surface antigen ubiquitously expressed by almost all human cells except erythrocytes. The CD46 antigen is identified by the E4.3 monoclonal antibody (mAb) and exists at the surface of human peripheral blood lymphocytes (PBLs) as two acidic, non-disulfide bonded chains, alpha and beta, of Mr 66,000 and 56,000. Receptor density analysis showed that CD46 was of moderately low abundance on PBLs with 7.5 x 10(3) molecules present on each cell. The two chains of CD46 were purified (144,000-fold) by immunoaffinity-chromatography with E4.3 mAb from the plasma membranes of a human spleen infiltrated with chronic myelogenous leukemia cells. Amino acid sequence analysis of the NH2-terminal of both alpha and beta chains yielded the same sequence; XEEPPQ/TFEAMELIGKPKPYYEIGE. Peptide mapping studies confirmed that both CD46 chains were closely related, except for one peptide fragment. This amino acid sequence is identical to that of the NH2-terminal of the recently cloned membrane co-factor protein (MCP), a membrane protein that binds the C3b and C4b fragments of complement and acts as a co-factor for I protein-mediated decay of the complement convertases. CD46 shares a cross-reactive epitope with some primate retroviruses, and this may indicate that some retroviruses mimic the mechanisms used by autologous human cells to evade complement-mediated immune clearance.

Measurement of antibody affinity for cell surface antigens using an enzyme-linked immunosorbent assay

We present a fast, simple, and accurate method to determine the affinity constants of antibodies that bind to cell surface antigens. This procedure utilizes intact cells and native, unmodified antibody in a conventional enzyme-linked immunosorbent assay. Target cells are incubated with serial dilutions of antibody and allowed to reach equilibrium. Cells are then pelleted by centrifugation, and aliquots of unbound antibody in the supernatant are added to a microtiter plate precoated with capture antibody and measured in a conventional enzyme-linked immunosorbent assay (ELISA). We measured the affinity constant of murine monoclonal antibody CLB-1H-gran2, which binds to K562 cells (a human erythroleukemia line), and compared the ELISA-based results to those obtained by flow cytometric determination of antibody affinity. The affinity constants obtained by the two methods are in good agreement. The affinity constant is calculated utilizing only the concentrations of bound and free antibody, so that the actual antigen concentration (or number of antigenic sites per cell) need not be known. However, the number of antibody molecules bound per cell can be estimated from the results.

Purification and properties of Myxococcus xanthus cell surface antigen 1604

A cell surface antigen complex from Zwittergent-solubilized Myxococcus xanthus has been purified by immunoaffinity chromatography with monoclonal antibody (MAb) 1604 and by subsequent gel filtration. We propose that the cell surface antigen (CSA) 1604 complex participates in intercellular interactions. The apparent total molecular mass of the CSA 1604 complex is 200 kilodaltons (kDa), as determined by gel filtration and by electrophoresis and Western immunoblot probing with MAb 1604. The antigen epitope recognized by MAb 1604 is on a 51-kDa polypeptide. The CSA complex also contains 14% neutral carbohydrate and a 23-kDa polypeptide that lacks the 1604 epitope. The carbohydrate is most likely part of a lipopolysaccharide (LPS) associated with the CSA, because an MAb recognizing an O antigen epitope from the LPS of M. xanthus also reacted with CSA 1604 on Western immunoblots. Thus, the 200-kDa CSA complex consists of 97 +/- 6 kDa of protein and many associated LPS molecules. The LPS evidently produces the multiplicity of bands observed on Western immunoblots between 100 and 200 kDa. The association with LPS may contribute to the negative charge of the CSA 1604 complex, which has a pI of 4.3. The CSA was clustered on the surface of intact M. xanthus cells after labeling with MAb 1604 and immunogold. Furthermore, fractionation studies indicated that cells grown on a plastic surface had 50% of their total CSA 1604 in the cytosol, 39% in the membrane fraction, and 8% in the periplasm. Saturable binding studies with 125I-MAb 1604 indicated that there were 2,400 CSA 1604 sites per cell. The Kd for MAb 1604 binding to the cell was 9 nM.

Specific targeting and toxicity of sulphonated aluminium phthalocyanine photosensitised liposomes directed to cells by monoclonal antibody in vitro

A partially purified fraction of the water soluble photosensitive dye sulphonated aluminium phthalocyanine (AlSPc) was encapsulated in liposomes which were then linked to a targeting monoclonal antibody 791T/36 using a heterobifunctional linking agent. The photocytotoxic effects of the liposomes were determined on two cell lines bearing an antigen with which the targeting antibody binds: 791T, an osteosarcoma and C170, a colorectal carcinoma; and a control cell line not bearing the antigen; DW-BCL, an Epstein-Barr virus immortalised B-cell line. Antibody dependent cytotoxicity was observed in 791T and C170 cells and was proportional to the number of antigens on the cells, the AlSPc concentration and the time of exposure to activating red light. No significant toxicity was seen using untargeted liposomes, control cells or free AlSPc fraction under similar conditions. Targeted cells and controls kept in the dark also showed no significant toxicity. A possible mechanism of action is postulated and simple adaptations which demonstrate the versatility of the model are discussed. Some suggestions as to the clinical situations to which this system might be applied in the form of photodynamic therapy (PDT) are made.

Dissociation kinetics of antigen-antibody interactions: studies on a panel of anti-albumin monoclonal antibodies

Kinetic parameters and equilibrium association constants (K) are reported for a panel of anti-bovine serum albumin (BSA) monoclonal antibodies (MAb) immobilized onto agarose particles. For 12 covalently immobilized MAb of moderate affinity (K = 0.25 x 10(8)-1.2 x 10(8) M-1) measured dissociation time constants varied two orders of magnitude, from 2.1 to 410 min. Directly measured association rate parameters agree with values calculated from measured equilibrium and dissociation rate parameters. Dissociation time constants and equilibrium association constants were also determined for eight MAb immobilized biospecifically (via their Fc regions). A significantly lower K was observed with those MAb which were covalently immobilized as opposed to biospecifically immobilized. These decreases in K appear to reflect decreased association rates rather than increased dissociation rates. The data suggest that, for the MAb described herein, dissociation rates do not correlate with equilibrium association constants.

Kinetics of antigen-antibody reactions at solid-liquid interfaces

The kinetics of antigen-antibody reactions is reviewed with special attention paid to the specific properties at solid-liquid interfaces. Theories of possible diffusion limitation in forward reaction rates are compared to experiments. It is found that the intrinsic forward reaction rate in the bimolecular antigen-antibody reaction is normally not limited by diffusion either in solution or at the solid-liquid interface. However, reactions at the solid-liquid interface can be diffusion limited due to depletion of reactants close to the surface. This effect depends on geometry, intrinsic reaction rate and surface concentration of receptor molecules. Normally cell surface reactions are not diffusion limited whereas reactions at artificial surfaces often are limited by diffusion. When not limited by diffusion it is also found that the intrinsic forward and reverse reaction rates are lower for surface reactions compared to reactions in solution. Antigen-antibody reactions at solid-liquid interfaces can often be considered as practically irreversible and limited by mass transport or steric interactions.

Transferrin receptor expression of AML blasts is related to their proliferative potential

Expression of the transferrin receptor (TfR) was studied on peripheral blood blast cells from 11 patients with acute myeloid leukaemia (AML). Using a monoclonal anti-TfR antibody (OKT9) and a polyclonal antibody against surface membrane-bound transferrin, a proportion of blasts from all the patients was found to express receptors for transferrin. Further analysis of OKT9 expression using a fluorescent activated cell sorter (FACS) showed that the TfR was heterogeneously distributed in the blast cell population. In five out of six samples studied, stimulation of DNA synthesis following short-term culture induced a several-fold increase in TfR display as analysed by flow cytometry using OKT9 or FITC-conjugated transferrin. Blasts from seven patients stained with OKT9 were separated on the FACS into positive and negative or weakly positive fractions. Culture of the TfR negative population in a blast cell colony assay produced no colonies in either of two patients. In a further five patients the colony forming cells were predominantly associated with the TfR strongly positive fraction (52 +/- 25 colonies/10(4) cells) rather than the TfR weakly positive fraction (12 +/- 11 colonies/10(4) cells). Analysis of colony size showed that clones derived from the weakly positive fraction were smaller than clones from the TfR strongly positive fraction. These results suggest that TfR display by AML blasts is related to their proliferative capacity and is expressed by the leukaemic stem cell fraction.

Sensitivity of newly established colorectal cell lines to cytotoxic drugs and monoclonal antibody drug conjugates

A major problem in the chemotherapy of colorectal cancers is their resistance to most cytotoxic drugs which may be due to insufficient cellular transport. Drugs conjugated to monoclonal antibodies recognising tumour antigens may overcome these difficulties by providing access of active agents to the tumour cells. The anti-tumour monoclonal antibody shown to localise in patients with colorectal cancer, 791T/36, has been investigated as a potential targeting antibody. Eight cell lines were established from surgically resected material and were shown to bind 791T/36 antibody. They were screened for their sensitivity to methotrexate, 5-fluorouracil and daunomycin. Although 5-fluorouracil is the drug of choice for chemotherapy of colorectal cancer it was the most cytotoxic drug in only 2 of the 8 cell lines. Only the 4 cell lines which were resistant to methotrexate showed less cytotoxicity with methotrexate than 5-fluorouracil. The cell lines which were resistant to methotrexate were more sensitive to 791T/36-methotrexate conjugates. Daunomycin was the most cytotoxic drug in 4 of the 8 cell lines. However, a similar cytotoxicity was observed for free drug and 791T/36 daunomycin in the two lines tested. Selective monoclonal antibody drug conjugates may offer a solution to treatment of tumours which are resistant to classical chemotherapeutic agents. This is the first report to show that newly established cell lines that are resistant to classical chemotherapeutic agents are rendered sensitive when the drug enters the cell as a drug monoclonal antibody carrier.

Quantitation of MHC antigen expression on colorectal tumours and its association with tumour progression

A flow cytometric technique has been established for accurately quantitating the cell surface density of MHC antigens and the percentage of cells expressing MHC antigens in 38 colorectal tumours. Thirty-four percent of tumours were partially or completely negative for HLA-ABC antigen expression. Although the quantity of HLA-ABC antigens varied widely, there was no correlation between the density of HLA-ABC antigens, or the percentage of cells expressing these antigens and clinicopathological stage. Fifty percent of the colorectal tumours expressed HLA-DR with varying antigen densities. All of the poorly differentiated tumours expressed HLA-DR but there was no correlation between expression of HLA-DR and clinicopathological stage. The aneuploid tumours expressed more HLA-ABC and HLA-DR antigens on a higher percentage of cells than the diploid tumours. Abnormal expression of the tumour associated antigens CEA, Y haptenic blood group and 791T p72 also correlated with expression of HLA-ABC and HLA-DR antigens on colorectal tumours. The majority of early derived in vitro dividing cells failed to express both HLA-ABC and HLA-DR antigens although they expressed high levels of tumour associated antigens. If there is a correlation between in vitro and in vivo growth perhaps tumours are maintained and seeded by MHC antigen negative cells.

Theoretical aspects of screening for high affinity monoclonal antibody to cell surface antigens

The relative affinity of the cell-antibody interaction can be determined by finding the highest titer of one reactant which binds effectively to a fixed, dilute concentration of the other. There are two ways to perform these experiments: the antibody dilution method and the cell dilution method. The antibody dilution approach requires measuring the highest antibody dilution which can react with a small, fixed concentration of cells. If the endpoint antibody concentration is small, the antibody has a high binding constant. In the cell dilution method the investigator must find the highest cell dilution which can bind a small, fixed antibody concentration. If the antibody has a high affinity constant, the endpoint cell concentration will be small. When the antibody and cell receptor concentrations are known, functional binding constants can be calculated directly from binding data using equations presented in this paper. Binding constants measured by these methods will help predict the behavior of monoclonal antibodies in immunoassays and in experiments where monoclonal antibody is injected in vivo for imaging or pharmacologic delivery.

Comparative analysis of monoclonal antibody-drug conjugate binding by flow cytometry

Flow cytometric methods for the evaluation of the cell surface binding properties of monoclonal antibody (MoAb)-drug/toxin conjugates defining tumor-associated antigens are presented. In these techniques, suspension cultures of solid human tumor cell lines are incubated with either varying dilutions of MoAb or MoAb-drug conjugates followed by FITC-conjugated anti-mouse immunoglobulin antibodies in an indirect assay or with FITC-conjugated MoAbs specific for the tumor target cell line in a competition assay. The amount of fluorescent probe bound is measured by flow cytometry and the mean fluorescence intensity determined. The relative binding capacity is quantified by linear regression of the mean fluorescence versus the concentration of primary antibody or antibody conjugate. The application of these techniques to several drug and toxin conjugates of MoAb KS1/4, which defines a human adenocarcinoma-associated antigen, demonstrates that these assays can be effectively utilized to monitor the effects of covalent chemical modification on a MoAb's antigen binding reactivity.

Mapping of monoclonal antibody-defined epitopes associated with carcinoembryonic antigen, CEA

Six immunoglobulin G monoclonal antibodies reactive with carcinoembryonic antigen (CEA) were evaluated with respect to parameters implicated in their potential diagnostic application and use as tumor targeting agents for cytotoxic drugs or plant or bacterial toxins. Antibody reactivity with surface antigens of the MKN-45 gastric tumor cell line was demonstrated by flow cytofluorimetry. In a subcellular membrane binding assay, each antibody reacted preferentially with membranes isolated from colorectal tumor tissue in comparison with their reaction with membranes from adjacent, apparently normal colonic mucosa. Three of the antibodies (NCRC-23, C228, and 11.285.14) reacted specifically with CEA with little or no reaction with the cross-reacting antigen, NCA. The remaining three antibodies (C24, C161, and C198) were reactive with both CEA and NCA. Analysis of the epitopes defined by these antibodies was performed by competitive binding inhibition assays evaluating the capacity of unlabeled antibodies to compete with 125I-labeled antibodies in their binding to CEA. In addition, double determinant or 'sandwich' radioimmunoassays were employed to examine the coexpression of epitopes on CEA molecules. These studies permitted an epitope map to be constructed which describes the coincidence, overlapping, or independent expression of both CEA specific epitopes and epitopes shared between CEA and NCA. The map may be employed for the selection of antibodies for diagnostic and therapeutic use.

Measurement of tumour reactive antibody and antibody conjugate by competition, quantitated by flow cytofluorimetry

Binding of unlabelled monoclonal antibody preparations has been assessed by competition at saturation with fluorochrome labelled homologous antibody for binding to antigen bearing target cells. The extent of competition was measured by quantitative flow cytofluorimetry, and simple mathematical procedures have been developed to allow the interpretation of competition data in terms of antibody binding activity. In the system studied, non-specific (non-competitive) fluorescence was minimal, but an iterative method to calculate its contribution to the measured signal is given. This approach has the advantage that the antibody preparation to be tested does not need to be labelled or modified; this is particularly important when evaluating the binding activity of therapeutic antibody conjugates. Comparison with a well characterized standard antibody preparation provides a rapid, sensitive and accurate quality control procedure. This test is also simple to perform, requiring only the mixing of labelled and unlabelled antibodies with target cells, a single incubation, followed by analysis without washing of the target cells.

Antigenicity of newly established colorectal carcinoma cell lines

Cells from two adenocarcinomas, an adenoma and a metastatic node were isolated in soft agar. Expression of antigens, CEA, Y haptenic blood group and 791T-p72, defined by a range of candidate antibodies for tumour targeting was assessed. All of the cells expressed low levels of CEA but high levels of the Y haptenic blood group antigen although there was enormous inter and intraclonal variation. Of particular interest was the membrane expression of 791T-p72 antigen on all of the dividing tumour cells as previous studies had shown that 791T/36 antibody reacted with tumour stromal elements rather than malignant cell surfaces. The DNA content was abnormal in all of the cells whether they were derived from diploid or aneuploid primary tumours. They all grew readily in athymic mice and at least one monoclonal antibody, 791T/36, localised efficiently within these xenografts. Clonogenic cells therefore expressed the three tumour-associated antigens, several at higher levels than observed in the primary tumour. Monoclonal antibody 'cocktails' should therefore allow antibody mediated drug cytotoxicity to be effective at eradicating rapidly dividing tumour cells.