Influence of Affinity and Antigen Density on Antibody Localization in a Modifiable Tumor Targeting Model1

Home , Avidity

[CANCER RESEARCH 60, 7008–7013, December 15, 2000] Influence of Affinity and Antigen Density on Antibody Localization in a Modifiable Tumor Targeting Model1

Lionel S. Zuckier,2 Eric Z. Berkowitz, Ronald J. Sattenberg, Quing Hua Zhao, Hou Fu Deng, and Matthew D. Scharff Departments of Nuclear Medicine [L. S. Z., E. Z. B., R. J. S., Q. H. Z., H. F. D.] and Cell Biology [M. D. S.], Albert Einstein College of Medicine, Bronx, New York 10461 [L. S. Z., E. Z. B., R. J. S., Q. H. Z., M. D. S.], and First University Teaching Hospital, West China University of Medical Sciences, Chengdu, 610041, P. R. China [H. F. D.]

ABSTRACT Administration-approved for the imaging of tumors with other prom- ising agents for radioimmunoimaging and therapy in the regulatory A persistent question in the field of antibody imaging and therapy is pipeline (1). Despite this achievement, controversy still persists re- whether increased affinity is advantageous for the targeting of tumors. We garding several fundamental issues relating to imaging and therapy have addressed this issue by using a manipulatable model system to investigate the impact of affinity and antigen density on antibody local- with antibody-based radiopharmaceuticals. One important but unre- ization. In vitro enzyme-linked immunosorbent assays and bead-binding solved question is whether increased antibody affinity is advantageous assays were carried out using BSA conjugated with high and low densities for the targeting of tumors (2–4). (HD and LD, respectively) of the chemical hapten ␳-azophenyl-arsonate as Various provocative mathematical models have been used to probe an antigen. Isotype-matched monoclonal antibodies (mAbs) 36-65 and specific theoretical issues relating to antibody targeting, including the 36-71, with identical epitope specificity but 200-fold differences in affinity, influence of affinity on binding (5–10). Fundamental behavior of were chosen as targeting agents. The relative in vitro binding of 36-65 and antibodies has also been investigated in experimental in vitro studies 36-71 was compared with an artificial “tumor” model in vivo using (9–11). Nonetheless, the extrapolation of these findings to clinical antigen-substituted beads s.c. implanted into SCID mice. Nonsubstituted situations has proven uncertain (12–14), and experimental validation BSA beads were implanted in the contralateral groin as a nonspecific in animal studies is needed. In fact, a limited number of in vivo studies control. The efficacy of the targeting of [125I]-labeled antibodies was assessed by the imaging of animals on a gamma-scintillation camera using have been performed to investigate factors influencing antibody bind- quantitative region-of-interest image analysis over the course of 2 weeks ing (8, 13, 15–18), often with conflicting or unclear conclusions. It has and by postmortem tissue counting. In vitro, both antibodies bound well to proved difficult to identify antibodies with significantly different the HD antigen, whereas only the high-affinity mAb 36-71 bound effec- affinity but with the same class and subclass that bind to an identical tively to the LD antigen. In vivo, high-affinity mAb 36-71 bound appre- epitope. Another frequent limitation of in vivo models is a restricted ciably to both LD and HD beads. In contrast, there was no specific ability to modify relevant variables. For example, the influence of localization of low-affinity mAb 36-65 to LD antigen beads, although the antigen density on antibody targeting has not been extensively studied antibody did bind to the beads with the HD antigen. Whereas the high- in experimental models, largely because it is difficult to systematically affinity mAb 36-71 increased its binding to HD beads throughout the 14 vary this parameter in tumor xenografts. This is regrettable in that the days of observation, binding of the high affinity antibody to LD beads and density of an antigen determines the ability of a multivalent antibody of the low affinity antibody to HD beads plateaued between 10–14 days. These in vitro and in vivo findings demonstrate that the need for a to engage multiple antigenic sites simultaneously. In contrast with high-affinity antibody is dependent on the density of the target antigen. “affinity,” which measures the binding of an individual Fab-binding High-affinity antibodies bind effectively even with a single antigen-Fab site and antigen, the overall strength of interaction between a multi- interaction, irrespective of the antigen density. In contrast, low-affinity valent antibody and antigen is termed “avidity” (19). This parameter antibodies, because of weak individual antigen-Fab interactions, require is of paramount importance in predicting antibody binding in a given the avidity conferred by divalent binding for effective attachment, which context (9–11). can only occur if antigen density is above a certain threshold. An under- The goal of this paper is to investigate the relationship between standing of the differential behavior of high- and low-affinity antibodies antibody affinity, antigen density, and antibody targeting by applying and the impact of avidity is useful in predicting the binding of monovalent a highly modifiable, well-controlled antibody system to in vitro and in antibody fragments and engineered antibody constructs and underlies the vivo experimental models. For this purpose, we have chosen to study trend toward development of multivalent immunological moieties. Con- sideration of the relative density of the antigen on the tumor and the antibodies that bind to the well-characterized chemical hapten Ars on background tissues may enable and even favor targeting with low-affinity the basis of the spectrum of affinities available and the ability to antibodies in selected situations. modify the density of the antigen target. Investigations were initially performed in vitro using ELISA and bead-binding assays. This system was then extended into an in vivo targeting model using radiolabeled INTRODUCTION antibodies and an artificial “tumor” implanted s.c. in mice where variables relating to the immunoglobulin and antigen could be rigor- Significant progress has been achieved in the use of radiolabeled ously controlled. The efficacy of targeting was assessed by the serial mAbs3 for the imaging and therapy of malignancy. At present, four imaging of live animals on a gamma-scintillation camera with quan- murine antibody-derived radiopharmaceuticals are Food and Drug titative image analysis and by postmortem tissue counting.

Received 5/18/00; accepted 9/1/00. The costs of publication of this article were defrayed in part by the payment of page MATERIALS AND METHODS charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Antibodies. 36-65 and 36-71 are IgG1 mouse mAbs that bind the chemical 1 L. S. Z. was partially supported by NCI 1K11 CA01503, and M. D. S. was supported by R35CA39838, R01CA72649, and the Harry Eagle Chair from the National Women’s hapten Ars and are encoded by the same heavy- and light-chain V(D)J regions ϫ 5 Ϫ1 Division of the Albert Einstein College of Medicine. (20). mAb 36-65 uses germline V regions and has a Ka of 2.5 10 M as 2 To whom requests for reprints should be addressed, at Department of Nuclear determined by fluorescence enhancement, whereas mAb 36-71 has undergone Medicine, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461. 3 The abbreviations used are: mAb, monoclonal antibody; Ars, ␳-azophenyl-arsonate; region of interest; SCID, severe combined immunodeficiency; scFv, single-chain variable HD, high-density; LD, low-density; ELISA, enzyme-linked immunosorbent assay; ROI, region. 7008

ϫ 7 Ϫ1 ␮ somatic mutation and has a Ka of 4.5 10 M (21). Subcloned hybridoma 50 l of antigen-negative control BSA beads were introduced into the con- cell lines were used to make ascites in pristane-primed SCID mice, which lack tralateral right side. Mice were prepared at least 1 week before targeting studies potentially confounding endogenous immunoglobulins. To avoid bias that to allow any acute changes surrounding the beads to subside. Thyroid uptake might be introduced if the antigen were used to purify high- and low-affinity of radioiodine was blocked by the addition of three drops of Strong Iodine antibodies, pooled ascites from each of the cell lines were purified by protein (Lugol’s) Solution USP (Regional Service Center, Inc., Woburn, MA) per A or G affinity chromatography (HiTrap Protein A or G 1-ml columns; bottle of drinking water from this time and throughout the duration of the Pharmacia LKB Biotechnology, Piscataway, NJ) according to standard proto- study. cols. Purified MOPC 21, used as a murine IgG1 control, was obtained from a Two separate investigations were performed as described below. The first commercial vendor (The Binding Site, Birmingham, United Kingdom). study characterized the time course of antibody binding and validated choice Antigen. Attachment of the hapten Ars to carrier BSA at specific substi- of an optimal time point for statistical comparison of the groups. The second tution ratios was performed as described previously (22), and the resultant investigation studied a large number of mice at a single time point after binding Ars-BSA was stored at Ϫ70°C until use. For the purpose of the current studies, had peaked to generate statistically valid comparisons of antibody binding in antigen with a ratio of 1.8 Ars molecules per BSA molecule was selected for the different groups. use as a HD antigen, whereas antigen with a 0.15 Ars:BSA molar substitution Temporal Analysis of In Vivo Bead-Binding. To determine the time ratio was used as a LD antigen. Unconjugated BSA served as an Ars-negative course of antibody binding to antigen, the four permutations of antigen density control target. (HD and LD) and antibody affinity (mAbs 36-65 and 36-71) were studied in ELISA. Interaction of antibody affinity and antigen density was illustrated groups of four mice each. Mice were injected by tail vein with 2 ␮gof by ELISA based on the variation of a technique published previously (22). [125I]-labeled antibody. Over the course of 14 days, two mice per each group Binding of the specific antibodies (mAbs 36-65 and 36-71) to HD and LD (one for the final image set) were periodically anesthetized with i.m. Pento- antigen plates was studied in duplicate. Unless otherwise indicated, dilutions barbital, and imaged for 5 min in ventral projection using a dedicated animal were performed in a solution of 1 mg of BSA/100 ml of PBS, incubations were gamma-camera imaging system (Picker Dynacamera 4C; Picker International, at 37°C, and volumes of 50 ␮l/well were used. Plates were washed three times Highland Heights, OH; and NucLear MAC computer system, Scientific Im- between each reagent (Microwash II; Skatron Instruments, Sterling VA) with aging, Inc., Littleton, CO). The camera was peaked on 125I using a 20% energy PBS containing 0.05% Tween 20. window, with collimation achieved by perpendicularly placed radiography Antigen was diluted in PBS to a concentration of 10 ␮g/ml and distributed grids (Model 278032; Liebel Flarsheim, Cincinnati, OH). Identically sized into disposable 96-well ELISA plates (Corning Glassworks, Corning NY). ROIs were placed over the specific and control antigen beads, and a ROI- Two h thereafter, plates were washed, blocked with a solution of 1 mg binding ratio was generated on the basis of the ratio of left to right groin BSA/100 ml of PBS (100 ␮l/well) and stored at 4°C until used. At that time, counts. Values from duplicate animals were averaged for the initial six time plates were rewashed and Ars-binding antibodies, at concentrations of 20 points whereas the solitary value was used for the seventh and final point. ␮g/ml, were serially diluted (1:2) across the plate and incubated for 1 h. In addition, a single mouse per combination of antibody and antigen density ␥ Alkaline-phosphatase conjugated goat antimouse 1 (Southern Biotechnology was sacrificed at each of four time points over the 14-day period, and specific Associates, Inc., Birmingham AL) at a concentration of 1 ␮g/ml was then and control beads were dissected out of the inguinal areas, weighed, and added to the washed plates with a repeat washing 2 h later and the addition of counted by ␥-counter (Compugamma 1282; LKB Wallac, Turku, Finland). A substrate (␳-nitrophenyl phosphate, disodium) dissolved in a bicarbonate relative binding ratio was calculated based on the ratio of counts bound per g

buffer [0.001 M MgCl2/0.05 M Na2CO3 (pH 9.8)]. Absorbance of the plates was of specific Ars-beads (left groin) divided by those bound per g of nonspecific read at 405 ␭, after color developed (Multiskan MS; Labsystems). control beads (right groin). Iodination. For the in vitro and in vivo bead studies, antibodies (mAbs Statistical Comparison of In Vivo Bead-Binding. The effect on the 36-65, 36-71, and MOPC 21) were iodinated using the Iodogen method binding of antibody affinity and antigen density was studied in a total of 28 (Pierce, Rockford Il; Ref. 23) under mild conditions with resultant specific animals consisting of two cohorts of mice, which were pooled to increase activity of between 3–5 ␮Ci/␮g. Typically, 150 ␮Ci of [125I]NaI were incu- sample size. Mice were prepared with specific (HD or LD) and control beads bated with 24 ␮g of antibody in a volume of 100 ␮l for 40 min. Free iodine as above, injected with one of three radiolabeled antibodies (mAbs 36-65, was then separated from the iodinated proteins by size-exclusion chromatog- 36-71, or MOPC 21), and sacrificed after 10 or 13 days, after binding had raphy over a Sephadex G-25 column (Pharmacia, Piscataway NJ). Typical generally plateaued. Specific and control antigen beads were removed, reaction yields of 45–50% were obtained and trichloroacetic acid precipitation weighed, and counted by ␥-counter as above. The percent injected dose/g of revealed 93–98% protein-bound activity. beads was calculated as was the relative binding ratio between specific and Conjugation of Ars-BSA to Beads. HD and LD Ars-BSA and control control beads. Relationships between the relative binding ratio, antibody BSA lacking Ars were each conjugated to commercially available, chemically affinity, and antigen density were analyzed by two-way ANOVA. activated beads (Reacti-Gel HW-65F; Pierce, Rockford IL; nominal diameter of 32–63 ␮m) under identical reaction conditions. Twenty-one mg of protein RESULTS in 5 ml of 0.1 M borate buffer (pH 9.0) were added per 5 ml of beads. After a 72-h incubation (4°C), beads were separated from solution and 1 M Tris buffer The present studies contrast the targeting of two murine IgG1 was added for 24 h to block remaining active sites. Beads were then spun down antibodies that bind to the chemical hapten Ars. On the basis of and stored in PBS and 0.02% azide at 4°C. In Vitro Antibody Binding Assay. Binding of the three radiolabeled serology and sequencing (21), both anti-Ars antibodies are encoded by antibodies (mAbs 36-65, 36-71, and MOPC 21) to beads bearing HD, LD, and the same germline V(D)J genetic elements for their heavy- and control antigen was studied using duplicate samples per combination of anti- light-chain variable regions and bind to the identical epitope. mAb body and antigen. Within the upper chamber of a 0.45-␮m cellulose acetate 36-65 is a relatively low-affinity germline-encoded antibody with a ␮ ϫ 5 Ϫ1 SPIN-X microtube (Costar, Cambridge MA), 0.004 g of radiolabeled anti- true Ka of 2.5 10 M (21). mAb 36-71 has a markedly higher ␮ ␮ ϫ 7 Ϫ1 body, 5 l of freshly washed antigen beads, and 200 l of 1% BSA/PBS were affinity with true Ka of 4.5 10 M as a result of somatic mutation combined and agitated on a rotating platform at room temperature. Seventy- and is among the highest affinity antibodies that have been elicited by two h later, supernatant was separated from the beads by spinning the micro- this hapten (21). ␮ tubes at 1500 rpm for 5 min, rinsing with 300 l of 1% BSA/PBS, and In Vitro ELISA. Binding of mAbs 36-71 and 36-65 to HD and LD respinning. Duplicate supernatant and bead samples were counted on a ␥ antigen is illustrated by ELISA in Fig. 1. Duplicate measurements are counter (Compugamma 1282; LKB Wallac, Turku, Finland), and the mean fraction of bound antibody was calculated. shown and were highly consistent. On HD antigen, both antibodies Murine Targeting Model. SCID mice were chosen for these studies be- bound effectively, with the higher-affinity mAb 36-71 binding to a cause they lack any potentially confounding endogenous immune response quantitatively greater degree than mAb 36-65. On LD antigen, a (24). Fifty ␮l of antigen-specific beads (HD or LD) were introduced into the marked qualitative difference in binding between the two antibodies left inguinal region of anesthetized adult female mice by s.c. injection, whereas was noted. mAb 36-71 bound effectively to the LD antigen plate, 7009

inguinal region by s.c. injection, whereas 50 ␮l of antigen-negative BSA beads were similarly introduced on the right side. One week later, mice were injected with 2 ␮gof125I-labeled mAbs 36-65 or 36-71. Two animals per group (one in the final session) were periodically anesthetized and imaged (Fig. 2). Identically sized ROIs were placed over the specific and control antigen beads, and a ROI binding ratio was generated based on an average value per time point (Fig. 3). On sequential gamma-camera images (Fig. 2), activity is initially noted in the vascular structures, with prominence of the heart, great vessels, and viscera, especially the liver. Over time, this activity lessens as the blood pool activity decreases. It is interesting to note a mild amount of activity which was localized to the control beads (right groin) in all groups of animals. This nonspecific accumulation

Fig. 1. ELISA of low-affinity mAb 36-65 (squares) and high-affinity mAb 36-71 (circles). Duplicate measurements are shown and are highly consistent. Left panel,HD antigen. Both antibodies bind to the HD target, mAb 36-71 to a quantitatively greater degree than mAb 36-65. Right panel, LD antigen. mAb 36-65 fails to bind significantly to the LD antigen because of the ability of the antibodies to bind divalently only to HD antigen.

Table 1 Mean Percentage of Antibody Bound to Antigen Percent binding retention of high affinity, low affinity and control mAbs to HD, LD, and control (BSA) antigen is compared in an in vitro bead assay as described in “Materials and Methods.” Values are the average of duplicate measurements, which were always consistent to within 2.5% Antibody HD antigen LD antigen Control (BSA) antigen High affinity (36-71) 75.8 56.6 9.7 Low affinity (36-65) 57.8 17.6 9.0 Control (MOPC 21) 2.7 2.0 2.6 whereas, for all intents and purposes, the low-affinity mAb 36-65 did not bind. The necessity of a HD target for effective binding of a low-affinity antibody reflects the fact that individual antigen-Fab Fig. 2. Sequential images of a single mouse per combination of antibody type and interactions are weak and highly reversible and require the divalent antigen density at representative time points. In ventral projection, control antigen has been introduced into the right groin (left), whereas specific Ars-BSA antigen (either HD attachment of both Fabs from each IgG molecule. This higher-avidity or LD) has been injected into the left groin (right). Initial activity in the blood pool (heart, interaction can occur only if antigen density is above a certain thresh- vessels, and liver) tends to decrease over the course of time. A mild degree of nonspecific old. High-affinity antibodies, in contrast, are able to bind effectively localization is noted in the control beads (right groin), consistent with that noted in the in vitro bead experiments previously described (Table 1). With respect to the antigen beads even with a single antigen-Fab interaction, and therefore are retained (left groin), no specific localization of low-affinity mAb 36-65 to LD antigen is apparent well on both HD and LD targets. (Column 1), although significant mAb 36-65 binding is observed to HD antigen (Column 3). High-affinity mAb 36-71 binds noticeably to LD antigen (Column 2) and binds to a In Vitro Bead-Binding Study. To develop a system for evaluating visually superior degree to HD antigen (Column 4). the role of affinity and avidity in the targeting of antibody, we initially coupled HD and LD Ars-BSA to microbeads, as described in “Mate- rials and Methods” and studied antibody-binding in vitro using radiolabeled antibodies. The relative binding of mAbs 36-65 and 36-71 to the beads was compared by determining the percentage of radioiodi- nated antibody bound to the beads after a 72-h incubation. Mean binding results for the nine permutations of antibody affinity and the antigen target density are shown in Table 1. Duplicate measurements were all within 2.5%. The slightly higher binding of mAbs 36-65 and 36-71 to control BSA beads as compared with the MOPC 21 isotype- matched control antibody was a consistent finding, suggesting weak cross-reactivity of the specific antibodies for the BSA-beads. Both high- and low-affinity antibodies bound effectively to HD antigen beads with retention of 75.8 and 57.8%, respectively. With respect to the LD target, a much greater fraction of high-affinity mAb 36-71 bound (56.6%) than did the low-affinity mAb 36-65 (17.6%), the latter approaching the level of background binding. These findings Fig. 3. ROI analysis and postmortem counting of beads were used to generate relative are similar to the ELISA observations and reflect the necessity of binding ratios of antibody per g of specific beads as compared with control beads. divalent attachment for successful binding of antibodies of lower Although generally similar in trend, ratios generated by ROI analysis were relatively affinity. blunted as compared with the absolute ratios obtained by postmortem counting. On the basis of the postmortem bead ratios, the binding of low-affinity mAb 36-65 to LD antigen In Vivo Targeting. The ultimate goal of our study was to examine (ARS5) was similar to its binding to control antigen (ratios of 1), although low-affinity antibody behavior in vivo. The time course of binding was demon- mAb 36-65 did bind specifically to the HD (ARS2) antigen target. High-affinity mAb ␮ 36-71 bound specifically to the LD antigen target and in a superior manner to the HD strated over a 14-day period in adult female SCID mice. Fifty lof antigen. As a general rule, binding appeared to plateau by 10–13 days, with the exception antigen-specific beads (HD or LD) were introduced into the left of high-affinity mAb 36-71 and HD antigen, where the binding ratio continued to increase. 7010

Table 2 Percent Injected Dose Bound to Antigen Beads SCID mice were prepared with HD- or LD-substituted antigen beads implanted S.C. in the left groin and control BSA-substituted beads in the right. Ten to 13 days post-administration of control (MOPC 21), high (36-71) or low (36-65) affinity mAb, absolute and relative binding of radiolabeled mAbs to specific and control beads were recorded. Percent injected dose (%ID) and ratio of specific:control values are expressed as mean Ϯ 1 SD. mAb/bead %ID specific beads %ID control beads Ratio Statistical significance of ratio 36-71/LD 54.7 Ϯ 17.6 16.6 Ϯ 5.7 3.3 Ϯ 0.3 Similar to 36-65/HD (p ϭ 0.80); different than other groups (all P Ͻ 0.05) 36-65/LD 16.1 Ϯ 4.4 16.6 Ϯ 8.8 1.1 Ϯ 0.3 Indistinguishable from 1 (P ϭ 0.85) MOPC 21/LD 8.2 Ϯ 0.8 7.6 Ϯ 0.5 1.1 Ϯ 0.1 Indistinguishable from 1 (P ϭ 0.50) 36-71/HD 201.9 Ϯ 80.3 19.0 Ϯ 5.1 10.5 Ϯ 3.0 Superior to all other groups (P ϭ 0.0001) 36-65/HD 45.2 Ϯ 16.2 14.1 Ϯ 3.9 3.2 Ϯ 0.4 Similar to 36-71/LD (P ϭ 0.80); different than other groups (all P Ͻ 0.05) MOPC 21/HD 7.3 Ϯ 0.4 7.6 Ϯ 0.3 0.96 Ϯ 0.0 Indistinguishable from 1 (P ϭ 0.25)

may be attributable in part to increased vascular and interstitial MOPC 21 control antibody to the HD or the LD antigen target did not volume surrounding the beads as well as to a mild degree of cross- differ appreciably from binding to contralateral control BSA beads, reactivity of the antibodies with control BSA beads, as noted in the in with localization ratios statistically indistinguishable from 1 (P ϭ 0.25 vitro bead study (Table 1). The presence of nonspecific uptake high- and 0.50, respectively). Binding of low-affinity antibody to the LD lights the need to compare specific localization of antibody to antigen- antigen was also indistinguishable from the control groups and was positive beads in the left groin with the mild degree of nonspecific statistically similar to 1, indicating a lack of effective targeting (all Ps localization in the right groin. No specific localization of low-affinity Ͼ0.85). Binding of the high-affinity mAb 36-71 to the HD antigen antibody to LD antigen was noted (column 1), although specific was statistically superior to all other groups (P ϭ 0.0001) with a binding was apparent on HD antigen (column 3). High-affinity mAb localization ratio of 10.5 Ϯ 3.0. Binding of high-affinity mAb 36-71 36-71 bound appreciably to both LD antigen (column 2) and espe- to LD antigen beads was intermediate in strength and similar to the cially to HD antigen (column 4). Visual impression was quantitated binding of the low-affinity antibody on HD beads (P ϭ 0.80), al- and validated by the ROI analysis (Fig. 3A). though different from each of the other four groups (Ps Ͻ0.05). The At each of four serial time points over 14 days, one animal per qualitative change in the binding of mAb 36-65 when binding LD group was killed, and the extirpated beads weighed and counted. A versus HD antigen, as contrasted with the relatively similar behavior relative binding ratio was determined based on the ratio of counts per of mAb 36-71, indicates a statistically significant interaction between g of specific beads (left groin) divided by counts per g of nonspecific antibody type and antigen density. control beads (right groin; Fig. 3B). Binding of low-affinity mAb 36-65 to the LD antigen was similar to control antigen, with a ratio of DISCUSSION 1, whereas this antibody did bind effectively to the HD antigen target. High-affinity mAb 36-71 bound specifically to both the LD and In the present study, we have addressed questions regarding optimal especially to the HD antigen. As a general rule, antibody binding antibody affinity for targeting by using a highly manipulatable, though appeared to plateau between 10–14 days, validating the choice of this admittedly artificial, model system. Findings made on ELISA plates, interval for use in the quantitative analysis described below. mAb an in vitro bead assay, and an in vivo animal model consistently 36-71 binding to HD beads was not only the highest for each obser- demonstrate that the need for a high-affinity antibody is in fact vation, but continued to increase throughout the duration of the study. dependent on the density of the target antigen. On HD antigen, both This is not surprising, in that the high affinity and avidity of interac- antibodies bound effectively, whereas on the LD target, only the tion between 36-71 and the beads would dictate that any available high-affinity antibody was able to bind, with insignificant binding of antibody would continue to bind to the beads, whereas effectively the low-affinity antibody. This relationship between the antibody none would be lost by elution. Visual (Fig. 2) and ROI observations affinity, the density of the target, and binding can best be understood (Fig. 3A) were consistent with the quantitative postmortem bead as reflecting different minimal requirements needed for the binding of measurements (Fig. 3B), though somewhat blunted in magnitude. the high and low-affinity antibodies. For low-affinity antibodies, such Decreased ratios for the imaging data are not surprising based on the as mAb 36-65, the strength of a monovalent interaction between a degrading effects of overlapping background tissues, limited camera single Fab binding arm and antigen is insufficient for retention. spatial resolution, and the attenuation by overlying soft tissues. These However, if the density of the antigen is high enough so that both arms technical factors may explain the inability to obtain correlation be- of the IgG1 molecule can bind simultaneously, even if one Fab tween absolute counts and imaging results in a previous study by temporarily dissociates, the antibody can again bind by both arms and Andrew et al. (16). is retained. With high-affinity antibodies, such as mAb 36-71, the Binding of control, low-, and high-affinity antibodies was also strength of the interaction between a single Fab binding site and the compared in a larger group of 28 mice to enable valid statistical antigen is sufficient to retain antibody with only monovalent binding, analysis. SCID mice bearing HD or LD antigen, prepared as above, and the antibody is bound effectively, irrespective of the antigen were injected with 2 ␮gof125I-labeled mAbs 36-65, 36-71, or MOPC density. The findings in our model system clearly reiterate the need to 21. After 10 or 13 days, mice were sacrificed and specific and control consider antigen density in in vivo as well as in vitro antibody binding antigen beads were dissected, weighed, and counted. The absolute studies. binding of antibody to the antigen and control beads was expressed as It is interesting to consider how the present findings can be gener- a percentage of injected dose/g of beads (%ID/g; Table 2). A relative alized to actual clinical tumor targeting. In the present paper, a binding ratio was also calculated based on the ratio of counts bound/g difference in the ability to bind divalently was noted, although the HD of specific beads (left groin) divided by those bound to the nonspecific and LD antigens only varied in density 12-fold. These concentrations control beads (right groin). were, in fact, chosen to demonstrate a dramatic difference in divalent Percent ID/g and absolute localization ratios for the six combina- binding, even within a small range of densities, without confounding tions of antibody and bead-density are listed in Table 2. Binding of secondary factors such as the total number of binding sites or mass- 7011

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2000 American Association for Cancer Research. INFLUENCE OF ANTIBODY AFFINITY ON TUMOR TARGETING action. In more extreme examples of HD and LD antigen, antibody relative density of antigen on target and nontarget tissues, it could be would still be only able to bind to LD antigen monovalently, whereas possible to select antibodies of specific affinity and valence so as to divalent binding could occur with HD antigen. In vivo, it is difficult to better differentiate tissues bearing HD and LD antigen. For example, theorize what concentration of antigen would enable divalent binding. if an epitope is present at a high density on the tumor target and more One could compare the radius of the binding arms of an IgG molecule sparsely on benign tissues, then it would be advantageous to choose a with the average density of antigen on the cell surface, but this would lower-affinity multivalent antibody so that avidity can be used to ignore the ability of antigen to migrate in the fluid cell membrane. discriminate between binding on the two tissues. In an analogous Many antigens, such as mucins, also contain repetitive epitopes, manner, avidity may be the key to understanding the initially surpris- which would have to be analyzed. Furthermore, issues such as acces- ing absence of interference by some circulating antigens when target- sibility of antigen, steric hindrance, and the flexibility of the IgG ing tumors with radiolabeled antibodies. Because interaction between molecule would be difficult to predict. Perhaps the best proof regard- an antibody and many soluble antigens is often monovalent, overall ing applicability of these concepts to actual tumor targeting is the fact avidity of binding is low. This allows antibodies of low-to-moderate that a large array of multivalent constructs, discussed below, has been affinity to readily dissociate from circulating antigen and to bind in a shown to be qualitatively superior to univalent molecules in clinical more irreversible manner to the target tumor where antigen density, and experimental studies. and hence avidity, is greater (22). This difference in binding would be In fact, the present findings provide interesting insights into clinical lessened in the case of high-affinity antibodies, which could presum- studies and suggest potential novel strategies for improving target-to- ably bind irreversibly to circulating antigen even monovalently, or background localization. Clearly, the critical parameter for predicting when targeting with monovalent fragments, which do not bind mul- antibody binding is avidity rather than affinity, because it takes into tivalently to the tumor antigen. In these cases, circulating antigen may account the effect of multivalent interaction. A decrease in binding conceivably represent a greater problem for tumor targeting. caused by reduced valence has been shown experimentally (25) and When targeting with monovalent antibody fragments, the present has long been clinically recognized when targeting enzymatically observations suggest that binding would succeed best with higher generated Fab and FabЈ antibody fragments. This observation has affinity antibodies. This is in accordance with the observations of taken on new relevance as novel immunological molecules are de- Adams et al. (43), who generated a set of scFv molecules with signed using recombinant DNA techniques. Many of the initial con- increased affinity by site-directed mutagenesis. Using this well- structs developed, such as scFvs (26, 27) were monovalent, and controlled family of reagents to target a SCID mouse xenograft model, consequently exhibited relatively poor binding. To compensate, diva- increased affinity resulted in improved tumor targeting. In his model lent constructs were devised, such as the divalently linked scFvs (28), system, Adams has also noted that, above a given threshold, no diabodies (29), and minibodies (30). In carefully controlled animal additional benefit is realized (44). The existence of a ceiling of affinity studies, the advantages of divalent constructs have been shown to be above which no additional advantage is obtained should be even more independent of the size and rate of vascular clearance (28, 31). evident when targeting with multivalent immunological molecules. Beyond producing small divalent constructs, other efforts have been Whereas a small experimental literature comparing divalent antibod- expended to artificially increase the valence of immunologically de- ies of differing affinity has appeared (13–18, 45–47), it is difficult to rived molecules to supranormal values such as trimers or tetramers of extrapolate from many of these studies because of unclear results or Fab fragments (32–34) or multimers of intact IgG molecules (35–37). the presence of confounding factors, such as differences in the epitope This trend, in fact, recalls an earlier observation by one of the current targeted. authors (M. D. S.) regarding mutant antibodies spontaneously derived Much has also been made of the provocative, yet persuasive, from the 36-65 cell line which demonstrated increased antigen bind- arguments of Weinstein et al. (5) and Fujimori et al. (6), who ing to low-density Ars-BSA. Analysis showed that the mutations were discussed the effect of elevated affinity on homogeneity as well as on not in the variable region increasing affinity, as might have been total uptake by tumor nodules. Their modeling work suggested that an expected, but were, rather, in the constant region, leading to poly- overly high affinity might be detrimental in therapy applications merization (22, 38). The consequent increase in span of the immuno- because the antibody would tend to accumulate at the periphery of gobulin and valence resulted in Ͼ100-fold increases in antigen bind- tumor deposits. This phenomenon, which they termed the “binding ing, as a manifestation of increased avidity. site barrier,” would thereby deprive the innermost tumor of therapeu- The benefit of increasing the valence of the binding molecule tic effect. In fact, this concept has been supported by experimental would only be effective in the presence of antigen density adequate data (48) and certainly would be relevant in situations of elevated for multivalent binding. In clinical practice, a low-affinity divalent avidity attributable to multivalent binding. antibody may be quite adequate for targeting a specific tumor ex- Conclusion. We have investigated the relationship between anti- pressing a high density of antigen on its surface, but totally inadequate body affinity, antigen density, and antibody targeting using modifi- for a different tumor with a lower density of antigen. Experimental able, well-controlled in vitro and in vivo experimental models based strategies to improve antibody targeting by increasing antigen density, on the chemical hapten Ars and isotype and epitope-matched mono- such as biological response modifiers (reviewed in Refs. 39 and 40), clonal antibodies that vary 200-fold in affinity. In contrast to high- can therefore be understood not only quantitatively as a means of affinity antibody, which binds effectively to both HD and LD antigen, increasing the total number of antigen-binding sites, but also qualita- low-affinity antibody only binds appreciably to HD antigen because tively, as a strategy to increase the avidity of binding, which is of its requirement for divalent attachment. In the rational design of especially important when antibody affinity and baseline antigen immunological reagents for specific in vivo applications, the impact of density are low. The importance of total antigen in driving the kinetics antibody affinity must be viewed in the context of the antigen density of antibody-antigen interaction has been considered previously in of the target and background tissues. theoretical discussions (8, 41) and in animal experiments (42), although the qualitative effect of antigen density on avidity has not been singled out in these discussions. REFERENCES It is interesting to contemplate tailoring the antibody affinity and 1. Zuckier, L. S., and DeNardo, G. L. Trials and tribulations: oncological antibody valence to the specific clinical usage at hand. Taking into account the imaging comes to the fore. Semin. Nucl. Med., 27: 10–29, 1997. 7012

2. Jurcic, J. G., and Scheinberg, D. A., Recent developments in the radioimmunotherapy engineering of antibody binding sites: recovery of specific activity in an anti-digoxin of cancer. Curr. Opin. Immunol., 6: 715–721, 1994. single-chain Fv analogue produced in Escherichia coli. Proc. Natl. Acad. Sci. USA, 3. Adams, G. P. Improving the tumor specificity and retention of antibody-based 85: 5879–5883, 1988. molecules. In Vivo, 12: 11–22, 1998. 27. Bird, R. E., Hardman, K. D., Jacobson, J. W., Johnson, S., Kaufman, B. M., Lee, 4. Chester, K. A., and Hawkins, R. E. Clinical issues in antibody design. Trends S-M., Lee, T., Pope, S. H., Riordan, G. S., and Whitlow, M. Single-chain antigen- Biotechnol., 13: 294–300, 1995. binding proteins. Science (Washington DC), 242: 423–426, 1988. 5. Weinstein, J. N., Eger, R. R., Covell, D. G., Black, C. D. V., Mulshine, J., 28. Adams, G. P., McCartney, J. E., Tai, M-S., Oppermann, H., Huston, J. S., Stafford, Carrasquillo, J. A., Larson, S. M., and Keenan, A. M. The pharmacology of mono- W. F., III, Bookman, M. A., Fand, I., Houston, L. L., and Weiner, L. M. Highly clonal antibodies. Ann. NY Acad. Sci., 507: 199–210, 1987. specific in vivo tumor targeting by monovalent and divalent forms of 741F8 anti-c- 6. Fujimori, K., Covell, D. G., Fletcher, J. E., and Weinstein, J. N. Modeling analysis of erbB-2 single-chain Fv. Cancer Res., 53: 4026–4034, 1993. Ј the global and microscopic distribution of immunoglobulin G, F(ab )2, and Fab in 29. Holliger, P., Prospero, T., and Winter, G. “Diabodies”: small bivalent and bispecific tumors. Cancer Res., 49: 5656–5663, 1989. antibody fragments. Proc. Natl. Acad. Sci. USA, 90: 6444–6448, 1993. 7. Thomas, G. D., Chappell, M. J., Dykes, P. W., Ramsden, D. B., Godfrey, K. R., Ellis, 30. Hu, S-z., Shively, L., Raubitschek, A., Sherman, M., Williams, L. E., Wong, J. Y. C., J. R. M., and Bradwell, A. R. Effect of dose, molecular size, affinity, and protein Shively, J. E., and Wu, A. M. Minibody: a novel engineered anti-carcinoembryonic binding on tumor uptake of antibody or ligand: a biomathematical model. Cancer antigen antibody fragment (single-chain Fv-CH 3) which exhibits rapid: high-level Res., 49: 3290–3296, 1989. targeting of xenografts. Cancer Res., 56: 3055–3061, 1996. 8. Sung, C., Shockley, T. R., Morrison, P. F., Dvorak, H. F., Yarmush, M. L., and 31. Adams, G. P., McCartney, J. E., Wolf, E. J., Tai, M-S., Schier, R., Stafford, W. F., Dedrick, R. L. Predicted and observed effects of antibody affinity and antigen density Marks, J. D., Bookman, M. A., Huston, J. S., and Weiner, L. M. Influence of avidity on monoclonal antibody uptake in solid tumors. Cancer Res., 52: 377–384, 1992. on the tumor retention of monospecific and bispecific anti-c-erB-2 single-chain Fv 9. Crothers, D. M., and Metzger, H. The influence of polyvalency on the binding dimers. Proc. Amer. Assoc. Cancer Res., 37: 472, 1996. properties of antibodies. Immunochemistry, 9: 341–357, 1972. 32. Schott, M. E., Frazier, K. A., Pollack, D. K., and Verbanac, K. M. Preparation, 10. Kaufman, E. N., and Jain, R. K. Effect of bivalent interaction upon apparent antibody characterization, and in vivo biodistribution properties of synthetically cross-linked affinity: experimental confirmation of theory using fluorescence photobleaching and multivalent antitumor antibody fragments. Bioconjugate Chem., 4: 153–165, 1993. implications for antibody binding assays. Cancer Res., 52: 4157–4167, 1992. 33. King, D. J., Turner, A., Farnsworth, A. P. H., Adair, J. R., Owens, R. J., Pedley, R. B., 11. Hornick, C. L., and Karush, F. Antibody affinity. III. The role of multivalence. Baldock, D., Proudfoot, K. A., Lawson, A. D. G., Beeley, N. R. A., Millar, K., Immunochemistry, 9: 325–340, 1972. Millican, T. A., Boyce, B. A., Antoniw, P., Mountain, A., Begent, R. H. J., Shochat, 12. McCready, D. R., Balch, C. M., Fidler, I. J., and Murray, J. L. Lack of comparability D., and Yarranton, G. T. Improved tumor targeting with chemically cross-linked between binding of monoclonal antibodies to melanoma cells in vitro and localization recombinant antibody fragments. Cancer Res., 54: 6176–6185, 1994. in vivo. J. Natl. Cancer Inst., 81: 682–687, 1989. 34. Werlen, R. C., Lankinen, M., Offord, R. E., Schubiger, P. A., Smith, A., and Rose, K. 13. Buchsbaum, D., Lloyd, R., Juni, J., Wollner, I., Brubaker, P., Hanna, D., Spicker, J., Preparation of a trivalent antigen-binding construct using polyoxime chemistry: Burns, F., Steplewski, Z., Colcher, D., Schlom, J., Buchegger, F., and Mach, J. P. improved biodistribution and potential for therapeutic application. Cancer Res., 56: Localization and imaging of radiolabeled monoclonal antibodies against colorectal 809–815, 1996. carcinoma in tumor-bearing nude mice. Cancer Res., 48: 4324–4333, 1988. 35. Wolff, E. A., Schreiber, G. J., Cosand, W. L., and Raff, H. V. Monoclonal antibody 14. Gallinger, S., Reilly, R. M., Kirsh, J. C., Odze, R. D., Schmocker, B. J., Hay, K., homodimers: enhanced antitumor activity in nude mice. Cancer Res., 53: 2560–2565, Polihronis, J., Damani, M. T., Shpitz, B., and Stern, H. S. Comparative dual label 1993. study of first and second generation antitumor-associated glycoprotein-72 monoclonal 36. Smith, R. I. F., and Morrison, S. L. Recombinant polymeric IgG: an approach to antibodies in colorectal cancer patients. Cancer Res., 53: 271–278, 1993. engineering more potent antibodies. Biotechnology, 12: 683–688, 1994. 15. Sakahara, H., Endo, K., Koizumi, M., Nakashima, T., Kunimatsu, M., Watanabe, Y., 37. Poon, P. H., Morrison, S. L., and Schumaker, V. N. Structure and function of several Kawamura, Y., Nakamura, T., Tanaka, H., Kotoura, Y., Yamamuro, T., Hosoi, S., anti-dansyl chimeric antibodies formed by domain interchanges between human IgM Toyama, S., and Torizuka, K. Relationship between in vitro binding activity and in and mouse IgG2b. J. Biol. Chem., 270: 8571–8577, 1995. vivo tumor accumulation of radiolabeled monoclonal antibodies. J. Nucl. Med., 29: 235–240, 1988. 38. French, D. L., Pollock, R. R., Aguila, H. L., and Scharff, M. D. The molecular and 16. Andrew, S. M., Johnstone, R. W., Russell, S. M., McKenzie, I. F. C., and Pietersz, biochemical characterization of mutant monoclonal antibodies with increased antigen G. A. Comparison of in vitro cell binding characteristics of four monoclonal anti- binding. J. Immunol., 146: 2010–2016, 1991. bodies and their individual tumor localization properties in mice. Cancer Res., 50: 39. Buchsbaum, D. J. Experimental tumor targeting with radiolabeled ligands. Cancer 4423–4428, 1990. (Phila.), 80 (Suppl.): 2371–2377, 1997. 17. Schlom, J., Eggensperger, D., Colcher, D., Molinolo, A., Houchens, D., Miller, L. S., 40. Roselli, M., Buonomo, O., Piazza, A., Guadagni, F., Vecchione, A., Brunetti, E., Hinkle, G., and Siler, K. Therapeutic advantage of high-affinity anticarcinoma radio- Cipriani, C., Amadei, G., Nieroda, C., Greiner, J. W., and Casciani, C. U. Novel immunoconjugates. Cancer Res., 52: 1067–1072, 1992. clinical approaches in monoclonal antibody-based management in colorectal cancer 18. Shockley, T. R., Lin, K., Sung, C., Nagy, J. A., Tompkins, R. G., Dedrick, R. L., patients: radioimmunoguided surgery and antigen augmentation. Semin. Surg. Oncol., Dvorak, H. F., and Yarmush, M. L. A quantitative analysis of tumor-specific mono- 15: 254–262, 1998. clonal antibody uptake by human melanoma xenografts: effects of antibody immu- 41. Fischman, A. J., Khaw, B. A., and Strauss, H. W. Quo vadis radioimmune imaging. nological properties and tumor antigen expression levels. Cancer Res., 52: 357–366, J. Nucl. Med., 30: 1911–1915, 1989. 1992. 42. Blumenthal, R. D., Sharkey, R. M., Kashi, R., Natale, A. M., and Goldenberg, D. M. 19. Rosen, F. S., Steiner, L. H., and Unanue, E. R. (eds.). Macmillan Dictionary of Influence of animal host and tumor implantation site on radio-antibody uptake in the Immunology. New York: The Macmillan Press, Ltd., 1989. GW-39 human colonic cancer xenograft. Int. J. Cancer, 44: 1041–1047, 1989. 20. Marshak-Rothstein, A., Siekevitz, M., Margolies, M. N., Mudgett-Hunter, M., and 43. Adams, G. P., Schier, R., Marshall, K., Wolf, E. J., McCall, A. M., Marks, J. D., and Gefter, M. L. Hybridoma proteins expressing the predominant idiotype of the anti- Weiner, L. M. Increased affinity leads to improved selective tumor delivery of azophenylarsonate response of A/J mice. Proc. Natl. Acad. Sci. USA, 77: 1120–1124, single-chain Fv antibodies. Cancer Res., 58: 485–490, 1998. 1980. 44. Adams, G. P., Schier, R., McCall, A. M., Simmons, H., Horak, E., Marks, J. D., and 21. Rothstein, T. L., and Gefter, M. L. Affinity analysis of idiotype-positive and idiotype- Weiner, L. M. What are the determinants of antibody-based targeting? Proc. Amer. negative Ars-binding hybridoma proteins and Ars-immune sera. Mol. Immunol., 20: Assoc. Cancer Res., 39: 436, 1998. 161–168, 1983. 45. Sharkey, R. M., Gold, D. V., Aninipot, R., Vagg, R., Ballance, C., Newman, E. S., 22. Pollock, R. R., French, D. L., Gefter, M. L., and Scharff, M. D. Identification of Ostella, F., Hansen, H. J., and Goldenberg, D. M. Comparison of tumor targeting in mutant monoclonal antibodies with increased antigen binding. Proc. Natl. Acad. Sci. nude mice by murine monoclonal antibodies directed against different human colo- USA, 85: 2298–2302, 1988. rectal cancer antigens. Cancer Res., 50 (Suppl): 828s–834s, 1990. 23. Fraker, P. J., and Speck, J. C., Jr. Protein and cell membrane iodinations with a 46. Sharkey, R. M., Primus, F. J., Shochat, D., and Goldenberg, D. M. Comparison of sparingly soluble chloroamide, 1,3,4,6-tetrachloro-3a,6a-diphenylglycoluril. Bio- tumor targeting of mouse monoclonal and goat polyclonal antibodies to carcino- chem. Biophys. Res. Commun., 80: 849–857, 1978. embryonic antigen in the GW-39 human tumor-hamster host model. Cancer Res., 48: 24. Bosma, G. C., Custer, R. P., and Bosma, M. J. A severe combined immunodeficiency 1823–1828, 1988. mutation in the mouse. Nature (Lond.), 301: 527–530, 1983. 47. Colcher, D., Minelli, M. F., Roselli, M., Muraro, R., Simpson-Milenic, D., and 25. Chapman, P. B., Lonberg, M., and Houghton, A. N. Light chain variants of an IgG3 Schlom, J. Radioimmunolocalization of human carcinoma xenografts with B72.3 anti-GD3 monoclonal antibody and the relationship among avidity, effector functions, second generation monoclonal antibodies. Cancer Res., 48: 4597–4603, 1988. tumor targeting, and antitumor activity. Cancer Res., 50: 1503–1509, 1990. 48. Juweid, M., Neumann, R., Paik, C., Perez-Bacete, M. J., Sato, J., van Osdol, W., and 26. Huston, J. S., Levinson, D., Mudgett-Hunter, M., Tai, M-S., Novotny, J., Margolies, Weinstein, J. N. Micropharmacology of monoclonal antibodies in solid tumors: direct M. N., Ridge, R. J., Bruccoleri, R. E., Haber, E., Crea, R., and Oppermann, H. Protein experimental evidence for a binding site barrier. Cancer Res., 52: 5144–5153, 1992.

7013

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2000 American Association for Cancer Research. Influence of Affinity and Antigen Density on Antibody Localization in a Modifiable Tumor Targeting Model

Lionel S. Zuckier, Eric Z. Berkowitz, Ronald J. Sattenberg, et al.

Cancer Res 2000;60:7008-7013.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/60/24/7008

Cited articles This article cites 46 articles, 29 of which you can access for free at: http://cancerres.aacrjournals.org/content/60/24/7008.full#ref-list-1

Citing articles This article has been cited by 6 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/60/24/7008.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/60/24/7008. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.