Extracorporeal Immunoadsorption Compared to Avidin Chase: Enhancement of Tumor-to-Normal Tissue Ratio for Biotinylated Rhenium-188- Chimeric BR96
JianQing Chen. Sven-Erik Strand. Jan Tennvall, Lars Lindgren, Cecilia Hindorf and Hans-Olov Sjögren Departments of Radiation Physics. Oncology and Tumor Immunology, Lund University, Lund, Sweden
antigen (8). The anticancer drug doxorubicin-coupled chiBR96 Based on the biodistribution and kinetics of biotinylated ^Re chimene BR96 (chiBR96), the enhancement of the tumor-to-normal has been extensively investigated preclinically for solid tumor tissue (T/N) radioactivity ratio using extracorporeal immunoadsorp- therapy, resulting in complete regression and cure of xe- tion (ECIA) was evaluated and compared with that of avidin "chase" nografted human lung, breast and colon carcinomas growing in colon carcinoma-isografted Brown Norwegian rats. Methods: subcutaneously in athymic mice (6,9). Doxorubicin is coupled Extracorporeal immunoadsorption (ECIA) was performed 6 or 12 hr to chiBR96 by an acid-labile thioether linker that is stable in after intravenous administration of biotinylated 188Re-chiBR96. Ra plasma and liberates doxorubicin, once it is internalized into the dioactivity redistribution was investigated just after ECIA [8 or 14 hr acidic environment of lysosomes and endosomes. Also, postinjection of the antibody] and 40 or 34 hr after ECIA perfor chiBR96-doxorubicin is currently in Phase I clinical trials for mance (48 hr postinjection). Avidin was administered intraperitone- the treatment of solid tumors (6). However, the limitation of ally at 6 hr postinjection. Tumor radioactivity uptake and T/N activity ratios of biotinylated 188Re-chiBR96 were compared using ECIA and MAb penetration and tumor heterogeneity are thought to be the main factors hampering the success of drug immunoconjugates avidin chase and were also compared with controls. Results: Both ECIA and avidin chase can rapidly increase the radioactivity tumor- in human cancer therapy. Radionuclide-coupled chiBR96, which can be used at least as an amplifier due to the "cross fire" to-blood ratio without significantly interfering with the tumor uptake. ECIA at 8 hr postinjection increased the T/N ratios in blood-rich from the targeted antigen-positive cells to the neighboring tissues, such as liver and bone marrow, from 6 and 8 to 10 and 18, antigen-negative cells, is being investigated preclinically by our respectively. Corresponding T/N ratios at 14 hr increased from 9 and group (¡0-12). 10 to 24 and 36. In contrast, when avidin chase was used, there Our previous experiments have demonstrated that a high were no T/N improvements, except in blood. Moreover, avidin chase tumor uptake can be obtained as early as a few hours after the caused a significant accumulation of radioactivity in the liver. Con injection of iodine- or indium-labeled chiBR96 (11,13). Also, clusion: The ECIA approach, with direct removal of unbound circulating biotinylated 188Re-chiBR96, thus can rapidly improve and Jure-Kunkel et al. (14) showed that chiBR96-doxorubicin maintain T/N ratios without overloading the liver with radioactivity, in already can be detected in various areas of the tumor 1 hr after contrast to avidin chase. injection into a mouse model with human lung carcinoma. Its early tumor targeting capacity encouraged us to explore the Key Words: extracorporeal immunoadsorption; avidin chase; bio labeling of chiBR96 with IKXRe(t,,2 = 16.98 hr; Eßmiix= 2.12 tinylated rhenium-188-chiBR96; tumor-to-normal tissue ratio Mev) for use in radioimmunotherapy (15). J NucÃMed 1997; 38:1934-1939 The main problem in radioimmunotherapy is the low clear ance rate of intact MAb from the circulation, causing radiation Ai,Ithough impressive advances have been made in the treat exposure to normal tissue such as myelosuppression (16,17). A ment of hematopoietic neoplasm (1,2) and in the adjuvant general approach, extracorporeal immunoadsorption (ECIA), treatment of colorectal carcinomas (3), only modest improve based on the high binding affinity of biotin-avidin, has been ments in the treatment of solid manifest tumors have been investigated for various kinds of antibodies and animal models obtained with immunoconjugates, mainly due to their low by our group (18-20). Briefly, biotinylated and radiolabeled tumor-to-normal tissue [or target-to-nontarget (T/N)] activity MAb is intravenously administered, allowing the antibody to ratio (4,5). However, recent achievements concerning immuno accumulate in the tumor. Unbound circulating MAb is removed conjugates, especially chimeric BR96 (chiBR96)-doxorubicin, by pumping the blood out of the body and passing it through an in preclinical studies have created a wave of enthusiasm among avidin-gel absorption column (for details see "Materials and scientists and clinicians working in this field (4,6,7). Methods"). Other groups also have used ECIA for clinical Monoclonal antibody (MAb) chiBR96, which consists of studies but with an antiantibody adsorption column (21,22). murine variable regions and human IgG, constant regions, Biotinylated and radiolabeled antibody combined with avidin strongly binds to a carbohydrate determinant epitope, Le-Y, "chase" is another approach that can be used to increase blood expressed (>200,000 molecules per cell) on most human colon, radioactivity clearance (23-25). Avidin-biotin-antibody com lung, breast and ovarian carcinomas. Chimeric BR96 has a toxic plexes are formed in the circulation after avidin chase. These effect on antigen-positive tumor cells, by mediating antibody- complexes are rapidly accumulated and catabolized in liver dependent cellular cytotoxicity and complement-dependent cy- tissue. totoxicity, and is rapidly internalized and degraded in lyso- The kinetics of biotinylated IKXRe-chiBR96 in colon carcino somes and endosomes after binding to cells expressing the ma-isografted Brown Norwegian (BN) rats, show not only maximal tumor uptake (%/g) at 6 hr postinjection but also a Received Oct. 9, 1996; revision accepted Dec. 19, 1996. For correspondence or reprints contact: Sven-Erik Strand, PhD, Department of high background in both blood and blood-rich normal tissues. Radiation Physics, Lund University, SE-221 85 Lund, Sweden. The starting time for ECIA was determined according to the
1934 THEJOURNALOFNUCLEARMEDICINE•Vol. 38 •No. 12 •December 1997 uptakes in both tumors and normal tissue. The ECIA procedure Tumor tissue, serially passaged in syngeneic BN rats, was homog was performed to remove unbound circulating biotinylated enized in buffer solution and used for tumor inoculation. lsxRe-chiBR96 at 6 hr or 12 hr postinjection. Tumor uptakes Fifty male BN rats, 4-6 mo of age, with a mean weight of 320 ± and T/N activity ratio enhancements with ECIA were evaluated 18 g, were grafted with 50 ju.1tumor suspension (tumor tissue and compared with those of avidin chase, as well as with those suspended in 16 volumes of buffer) intramuscularly in the thigh, of controls. To our knowledge, such a comparison between beneath the subrenal capsule and intrahepatically. Eight days after ECIA and avidin chase to increase T/N activity ratios in the grafting, the tumors reached a weight of about 100 mg, and the rats same model has not been previously described. were used for the in vivo studies of biotinylated IKSRe-chiBR96.
MATERIALS AND METHODS In Vivo Studies Monoclonal Antibody Experiment I: Extracorporeal Immunoadsorption, Twenty tu- Mouse human chiBR96 is a human IgG, isotype of murine mor-isografted BN rats were used for studies of the ECIA proce BR96 and binds to a tumor-associated antigen Le-Y (26,27). dure to enhance the T/N activity ratio. Two days before the ECIA, According to immunohistology, BR96 binds to the majority the rats were catheterized with an arterial (aorta carotis) and a vein (>75%) of human carcinomas of colon, lung, breast, ovary, (vena jugularis) catheter to gain blood access. Blood was pumped stomach, pancreas, esophagus and cervix (25). As with all other out from the arterial catheter, passed through an absorption column tumor-associated MAbs so far described, it is not completely at a flow rate of 1.5 ml/min and then returned by the venous tumor-specific, and it reacts with some normal tissues from man, catheter to the rat. The procedure of ECIA was completed within 2 rat, dog and monkey. However, only a limited number of normal hr. During this period, more than three blood volumes had passed human tissues express the BR96 antigen, such as differentiated through the avidin-absorption column, which contained 1.2 ml cells within the gastrointestinal epithelium and the acinar cells of avidin-Sepharose 6MB gel, with a concentration of 2 mg/ml avidin. the pancreas. In addition, both immunohistology and biodistribu- Before it was connected to the ECIA system, the column was tion of radiolabelcd chiBR96 show a very high tumor selectivity, rinsed with an aqueous solution of 0.9% sodium chloride to clear and the binding to normal tissues tends to be very low (7,11). the column of any traces of free avidin. A detailed description of our ECIA system can be found elsewhere (29,30). Chimeric BR96 Biotinylation and Rhenium-188 Labeling One hundred micrograms of biotinylated IKXRe-chiBR96,corre The biotinylation and lxsRe labeling of chiBR96 were performed sponding to 11-13 MBq of activity, with a specific activity of in accordance with our previously reported method (/5) using a 110-130 kBq per (xg of antibody, were administered to the rat bifunctional eliciate agent of tetrafluorophenyl mercaptoacetylgly- cylglycyl-y-aminobutyrate (TFP MAG2-GABA). Rhenium-188 through the vein catheter. Blood samples of about 0.5 ml were was obtained by eluting a 100 mCi lxxW/lxxRegenerator with 0.9% taken from the arterial catheter at 5 min postinjection, just before and after the ECIA procedure. ECIA was performed at 6 or 12 hr NaCl solution. Briefly, chiBR96. in bicarbonate buffer (pH 9.5), was mixed with biotin reagent (N-hydroxysuccinimido-biotin) at after the antibody injection. Groups of five rats were dissected just after the ECIA (8 or 14 hr postinjection) and at 40 or 34 hr after this 18 /ng per mg of antibody and incubated at room temperature for 1 hr. The biotin-coupled chiBR96 was separated from excess biotin procedure (48 hr postinjection). Tumors and normal tissues of regent using small-column centrifugation. The numbers of biotin interest were dissected, the blood was sponged away with gauze and the contents of the gastrointestinal tract were removed. The molecules coupled to each antibody molecule were determined tissues were weighed, and their radioactivity was measured in an (2fi). For '8KRc labeling, a solution containing 11 mg citric acid, 370 automatic Nal(Tl) scintillation well counter. Corrections were made for the decay. Uptakes in tumor and normal tissue are /j,g gentisic acid and 420 /j,g SnCU in 100 /nl H2O, together with 6 ju.g TFP MAG2-GABA in 4 ¡j,\isopropanol, was added to 1 ml expressed as a percentage of the injected activity per g of tissue fresh lxxRe elution, corresponding to an activity of 350-900 MBq. (%/g). The mixture was incubated at 90°Cfor 30 min and then rapidly Experiment 2: Avidin Chase. Twelve tumor-isografted rats were used for the study of avidin chase. One hundred micrograms of cooled to room temperature in an ice bath. After the pH of the biotinylated lxxRe-chiBR96 with the same specific activity as in mixture had been adjusted with 25% (v/v) bicarbonate buffer (1.0 Experiment I were injected into the superficial femoral vein of the M; pH 10.0), biotinylated chiBR% was immediately added, with a final pH of about 9.0. After a 15-min incubation at room temper rat. At 6 hr after the antibody administration, 300 ¿tgavidin, about a l M ratio of the administered antibody, in 0.9% NaCl solution ature, the radiolabeled antibody was purified on an equilibrated Sephadex pD-10 column eluted with 0.05 M PBS (pH 7.0). Bovine with a concentration of 1 mg/ml, was injected intraperitoneally within 1 min. The ratio of avidin to administered antibody was serum albumin and ascorbic acid were added to the purified biotinylated lxsRe-chiBR96, to final concentrations of 1 and 5 twice as high as that used for patients by Paganelli (22). The choice mg/ml, respectively. The immunoreactivity and avidin-binding of intraperitoneal route instead of intravenous injection was to capacity of the biotinylated lsxRe-chiBR96 were determined using prolong the biological half-life of avidin in the blood. A slight cultured cells of BN 7005 rat colon carcinoma and an avidin- improvement in blood activity clearance with intraperitoneal injec agarose gel (Iti). The stability of the biotinylated lxxRe-chiBR96 tion compared with intravenous injection was demonstrated by Marshall et al. (23). In the clinical application, intravenous infusion was checked by scrum incubation. of avidin would be more feasible. Blood samples were obtained for Animal Tumor Model radioactivity measurements from the periorbital venous plexa with Brown Norwegian rats, expressing the chiBR96 epitope in an EDTA tube just before and 30 min after the chase injection. normal tissues, such as pancreas and gastrointestinal tract, can Groups of four rats were dissected at 8 (2 hr after the avidin chase), imitate the human situation. Brown Norwegian 7005 rat colon 24 and 48 hr postinjection. Tumors and normal tissues of interest carcinoma expressing the chiBR96 epitope was originally induced were dissected and weighed, and their activities were measured. by 1,2-dimethylhydrazine. In this study, BN rats bearing isografts Experiment 3: Controls. Eighteen tumor-isografted rats given of BN 7005 colon carcinoma were used to avoid the limitations in 100 ¿igof biotinylated IKI EXTRACORPORKAL IlMMUNOADSORPTION AND AVIDIN CHASE •Chen et al. 1935 • A: before ECIA B:afterECIA C: 0.5 h after avidin chase FIGURE 1. Whole-body imaging. BN rats, carrying isografts of BN 7005 20 30 40 colon carcinoma intramuscularly in the thigh, beneath the subrenal capsule Time pi (h) and intrahepatically, were given 100 /xg biotinylated 188Re-chiBR96. (A) Before ECIA. (B) After ECIA. (C) At 0.5 hr after avidin chase. Both ECIA and FIGURE 2. Activity clearance of biotinylated 188Re-chiBR96from the blood. the avidin chase were started at 6 hr postinjection. Arrows represent tumors. Comparison between ECIA (starting at 6 or 12 hr postinjection) (n = 5), avidin chase (starting at 6 hr postinjection) and controls up to 48 hr postinjection The specific activity in tumor and normal tissues as well as the (n = 4). T/N activity ratio were evaluated for all three experiments. volume (around 100 mg), the intramuscular tumor could be Scintillation Camera Imaging clearly seen in all the images. The subrenal and intrahepatic Whole-body imaging was performed using a scintillation camera tumors could only be visualized after removal of the back equipped with a medium-energy collimator. A 25% energy window ground (Fig. IB), whereas it was impossible to distinguish was centered over the 155-keV photopeak of lxl 1936 THEJOURNALOFNUCLEARMEDICINE•Vol. 38 •No. 12 •December 1997 TABLE 1 Specific Activity (%/g) in Tumors and Normal Tissue for ECIA*, Avidin ChaseT and Controls1 at Three Different Times Postinjection Specific activity 8 hr p.i. 14 hr p.i. 48 hr p.i. chase chase TissueTumor (6hr)1 Zta)+ (6hr)0.87 iti0901.231.450.050.06(6hr)1 (intramuscular)Tumor 055+ 573.4443 + 0 +0414.17 882.643.740.050.23038+ 0.215.3308 + 0.231.51+ 009+ 0.412.34.31 + 0.832.14.81 + (subrenal)Tumor 1.18+ 0.953+ 0.874.81+ 1.12+ 0.624+ 0.581+ 0.60+ 0.462.19+ 0.732.78+ (intrahepatic)MuscleTesticle 36+1 570.0456 + 0 850.04+ 0 067+ 0.600.0536 + 0.410.03.39 + 0.18+ 0.460.07+ 0.680.07+ 0.00±0.040.010.27+ 0.020.41+ 0.02±0.010.010.32+ 0.020.05+ 0.01±0.000.010.1+ 0.020.15+ ±0.02 ±0.04 ±0.03 ±0.01 1 ±0.02 ±0.02 LiverSpleenKidneyPancreas0490 +013+ 2201 25 + 0 025089 + 0 0160090290.11+002+ 003045 + 0 003008 + 0 0110070250.07+002+ 00401424 + 0 00301723 + 0 190450.13003+ 0208835 + 0 0507943 + 0 001+ 0307128 + 0 0201706 + 0 001+ 0020 + 0020 + 007±0.020120.29+ 100.23+ 0 004±0.01+0060.27 0060.05+ 003±0.01060.1538 + 0 050.144 + 0 ±0.09 ±0.02 ±0.02 ±0.02 ±0.03 8 ±0.03 StomachSmall 013023020021040026045CIAihr)+002++ 004025 + 0 00204122 + 0 0120 +002+ 0190010 + 001003 + 0 004006014007015009041CIA'hr)++000+ 00201508 + 0 012003016+ intestineColonHeartLungBone 005+ 1042 + 0 1 17038+ 0 14028015027010032CIA004+ 02037 + 0 0200905 + 0 001+ 0030 + 0020 + 009+ 43078 + 0 0140 + 020+ 05045 + 0 0060 + 001+ 0.0701422 + 0.0701724 + 002+ 08139 + 0 07043 + 0 001+ 02043 + 0 030.1106 + 0 001+ 0030 + 003034+ 013+ 5706000 + 0 105777 + 0 1 009+ 04077 + 0 +0040 001+ 0401426 + 0 0040 + marrowBloodE(62.173.795130.030.30005+ 0171.99+ 0132.26+ 001+ 061.9347 + 0 020.3006 + 0 001+ 0040.61+ 020.8020 + 0 0.17Avidin + 0.13Controls231+ 0.31E(i:1 0.06Controls2+ 0.11ECIA + 0.11E 0.16Advidin + 0.13Controls1+ 0.09 •n= 5. fn = 4. Values are expressed as mean ±s.d. Radioactivity Distribution pared with controls, a similar radioactivity distribution was Specific activities (%/g) in tumors and normal tissues of found in all normal tissues except for a significantly higher interest at different time points with ECIA and avidin chase are activity accumulation in the liver 2 hr after the avidin chase (8 listed together with those of controls in Table 1. High tumor hr postinjection). The avidin chase could not significantly uptake was obtained at both 8 hr and 14 hr postinjection for the reduce the radioactivity in blood and blood-rich tissues (Table three different tumors. The specific activity in blood was about 1), although the blood radioactivity was greatly reduced within 2%/g, but it was lower than that of tumors. Extracorporeal the first 30 min (Fig. 2). immunoadsorption removed 40%-70% of the radioactivity The radioactivity ratio of intrahepatic tumor-to-normal tissue from blood-rich tissues, such as bone marrow, liver, spleen, using the ECIA procedure and the avidin chase was compared kidney and lung, without significantly interfering with tumor to that of controls (Table 2). Because of the direct removal of uptake. The radioactivity in organs such as the pancreas and the background activity without significantly interfering tumor gastrointestinal tract, which contain the epitope of chiBR96, uptake, the T/N activity ratio was significantly increased by the could also be partly removed by the ECIA procedure. Corn- ECIA procedure for most normal tissues, especially for the TABLE 2 Enhancement of Radioactivity Ratios of Intrahepatic Tumor-to-Nonmal Tissue Using ECIA* and Avidin Chasef Compared with Controls* p.i.Avidinh TissueMuscleTesticleLiverSpleenKidneyPancreasStomachSmallhr)178.5(6 hr)106.8chase (6 44.4§15.8± 71.213.3± 30.01± 24.517.0± 5.2. ± 2.9§aa+1.7125.4± 2.52.7± 3.35.61.3 ± 1.223.7± 14.4±3.39.2 0.3§10.3± 2.710.6± 6.2§40.3± ±1.215.4 '10.8 ±7.1 1.24.4± 3.35.7± 8.8§12.8± ±4.06.1 3.2§38.3± 0.615.3± 1.319.2± 3.2§32.5± ±1.516.7 13.3§35.7± 4.516.0± 3.721± 5.5s31.1± ±1.323.7 12.120.4± 1.59.8± ±7.313.3.4 ±7.128.4 3.712.2± intestineColonHeartLungBone3.833.3± 3.77.8± 8.114.1± 6.9§21± 1.110.3± 25.4§22.4± 5.110.3± 8.710.6± 6.825.0.4 ± 2.310.2± 7.3§13.1± 3.56.2± ±2.75.9 5.01114.4± 2.05.7± 5.5§18.2± ±1.27.2 ±1.68.1 4.2§35.6± ±1.29.9 marrowBloodECIA ±4.2"11.3 0.51.8± 2.62.0± 8.3"11.8±3.411P.LControls84.7± 1.42.3± ±5.1118 ±0.4Controls115.1 ±0.514hECIA(12hr)78.7 ±0.4 *n = 5. fn = 4. *n = 4. §p< 0.01. flp < 0.05. Values are expressed as mean ±s.d. EXTRACORPOREAL IMMUNOADSORPTION AND AVIDIN CHASE •Chen et al. 1937 TABLE 3 long waiting period for the blood concentration to fall because Comparison of Activity Distributions of Biotinylated Rhenium-188- any biotin-MAb in circulation must be saturated before any chiBR96 at 24 hr postinjection for the Avidin Chase and Controls radiolabeled streptavidin can reach the tumor (37). Hence, Avidin chase* ControlsT three-step pretargeting with an avidin infusion to clear the circulating biotin-MAb before administering the radiolabeled Tumor(intramuscular)Tumor 0.323.17± 0.253.50± streptavidin was investigated. A high liver activity accumula (subrenal)Tumor 0.533.56± 0.813.82± tion has been demonstrated in the three-step strategy (3X.39), (intrahepatic)MuscleTesticleLiverSpleenKidneyPancreasStomachSmall0.570.06± 0.670.06± perhaps due to the presence of the biotin-MAb still in the 0.030.19± 0.020.19± circulation. These two- or three-step pretargeting approaches 0.040.38± 0.020.46± require the MAb to be accumulated and maintained on the ±0.110.20 0.170.27± 0.030.54± 0.040.59± targeted cell for radiolabeled component binding. Possible 0.090.13± 0.040.21± disadvantages of these approaches are degradation of MAb due 0.060.14± 0.010.16± to the long interval between injection of MAb and radiolabeled 0.030.25± 0.000.26± component targeting (40), as well as antigenicity of avidin/ intestineColonHeartLungBone 0.010.37± 0.020.43± streptavidin (31). However, although two- and three-step pre 0.090.25± 0.120.24± targeting approaches are under active investigation and im 0.030.46± 0.020.40± proved results have been obtained (36), it is extremely difficult 0.080.27± 0.020.32± to use them for internalized MAbs because of the lack of the marrowBlood2.06 0.071.11± 0.051.21± biotin- or avidin-MAb remaining on the tumor cell surface for ±0.082.24 ±0.08 the amplified targeting of the radiolabeled avidin or biotin. 'Performed at 6 hr postinjection Contrary to the two- and three-step pretargeting, ECIA and fn = 4. avidin chase are the methods that might be used to improve T/N Values are expressed as mean ±s.d. ratios of internalizing MAbs. These two methods have been explored separately, both preclinically (IX,23) and clinically (21,22,41). A comparison between these two strategies, how blood-rich tissues. No T/N ratio enhancement could be demon ever, has not been performed earlier for T/N ratio enhancement strated for the avidin chase. of the radioimmunoconjugates in the same animal model. Although a reduction of activity in the blood and blood-rich In the present investigation, the ECIA procedure could organs could also be seen at 48 hr postinjection, when the avidin directly remove more than 80% of the blood radioactivity or chase and controls were compared, the tumor uptake (%/g) was 50%-55% of whole-body activity and consequently decrease lower in the three different tumors at 48 hr postinjection for the activity in the blood and blood-rich organs without signif ECIA treatment (Table I ). icantly interfering with tumor uptake, which resulted in an To determine the kinetics of the radioactivity accumulated in improved T/N ratio. In contrast, the avidin chase could not the liver, the activity distribution 18 hr after the avidin chase (24 significantly improve the T/N ratio. The relatively low chase hr postinjection) was compared with that of the controls. The efficacy (Fig. 2) could be due to the low biotinylation level of accumulated activity in liver after the avidin chase decreased to MAb and the short biological half-life of the avidin (42). Nine the level of the controls with a similar activity distribution in and 22 biotin molecules coupled to each anti-CEA antibody both tumors and normal tissues at 24 hr postinjection (Table 3). have been used for avidin chase by Marshall et al. (24). Linkers with space between biotin and MAb allowing biotin to interact DISCUSSION more effectively with avidin can improve the efficacy. Al A way of increasing the activity ratio of T/N is to dramati though more biotin coupling to each chiBR96 could increase cally decrease the level of radioimmunoconjugates in the blood the blood radioactivity clearance by avidin chase, the immuno- while retaining the level in the tumor. A review of the strategies reactivity of the biotinylated lsxRe-chiBR96 was significantly used to enhance the clearance of blood immunoconjugates to depressed resulting in low tumor uptake (data not shown). An gain an increased T/N ratio, such as avidin-biotin complex increase in blood radioactivity within 2 hr after the avidin chase formation, immunocomplex formation with a second antibody, also has been reported by Kobayashi et al. (43), and repeated plasmapheresis and ECIA, has been published recently (32). avidin chase was seen to further reduce the blood activity. The availability, ease of conjugation, extremely high affinity Our previous results showed that the radiolabeled (with I25I (IO15 A/"') and specificity of the biotin-avidin interaction have and '"in) and biotinylated chiBR96 has a high tumor selectiv recently prompted several preclinical and clinical investigations ity and that the biotinylation procedure does not significantly into their usefulness in MAb targeting. Avidin, a 66-kDa glyco- change the biodistribution and antigen binding capacity of protein present in egg white, shows tetravalency for biotin, a chiBR96 (11,13). The intrahepatic tumor uptake of biotinylated 244-Da vitamin H. More than one radiolabelcd biotin molecule 125I-chiBR96 reached 7.4%/g, whereas the intrahepatic tumor can bind to a pretargeted avidin-MAb. thereby providing the uptake of biotinylated I25I-L6 (an IgG2a as a control) was only possibility of amplifying tumor targeting. Also, through the 0.6%/g (12). The interfering of activity retention in the tumor accumulation of pretargeted biotin- or avidin-MAb, the activ due to the ECIA procedure depends on several factors, such as ity coupled to a small molecule of avidin or biotin is rapidly blood content in the tumor tissue, the affinity between the excreted. Therefore, the T/N ratio can be increased with a antigen and antibody, and shedding of the antibody from the reduced systemic toxicity, if the tumor uptake is not signifi binding site (17). Fast internalization of chiBR96 after binding cantly reduced. Various strategies based on the biotin-avidin to the epitope, minimizes the redistribution of the antibody concept have been examined using either two- or three-step within the tumor tissue after ECIA treatment. The specific pretargeting procedures (33.34). For instance, the T/N ratio was activity in the tumor tissues was significantly lower 40 hr and improved by two-step pretargeting, in which radiolabeled 34 hr after the ECIA (48 hr postinjection), compared with the streptavidin was administered after the pretargeting of the cold control and avidin chase, whereas the tumor uptake was not biotinylated MAb (35.36). Two-step pretargeting requires a significantly reduced immediately after the ECIA procedure 1938 Tin: JOURNALOFNUCLKARMEDICINE;•Vol. 38 •No. 12 •December 1997 (Table 1). The decreased specific activity in the tumor tissues a (tri-n-butylstannyl) benzoate conjugate. Cancer Biother Radiopharm 1996:11:217- 226. longer period of time after ECIA is probably due to the fast 14. Jure-Kunkel MN, Willner D, Casazza AM, et al. Distribution of BR96-DOX growth of the tumor cells combined with the cutting off of the immunoconjugates in human tumor xcnografts using laser scanning confocal micros MAb supply by the ECIA for the newly grown tumor cells. In copy. Proc Am Assoc Cancer Res 1994:35:509. 15. Chen J. Strand S-E. Tennvall J, Hindorf C. SjögrenH-O. Biodistribution, pharmaco the present tumor model, the tumor volume doubling time is kinetics and dosimetry of biotinylaled l*"Re-chiBR96 in colon carcinoma ¡sografted about 52 hr for the subrcnal tumor, with extensive necrosis, rats. Cancer 1997; in press. 16. Sgouros G. Plasmapheresis in radioimmunotherapy of micrometastases: a mathemat within 2.5 wk after the inoculation (10). The ECIA procedure ical modeling and dosimetrica! analysis. J NucÃMed 1992:33:2167-2179. did not remove the radioactive catabolites, which accumulated 17. DeNardo GL, Maddock SW, Sgouros G, Scheibe PO. DeNardo SJ. Immunoadsorption. in the colon content and would be excreted in feces (30). Hence, An enhancement strategy for radioimmunotherapy. J NucÃMed 1993:34:1020 1027. ECIA can further decrease the whole-body activity if the 18. Strand S-E, Norrgrcn K. Garkavij M, et al. A general extracorporeal immunoadsorp tion method to increase tumor-to-tissue ratio. Cancer 1994;73{suppl):H)33 1037. stability of the radioimmunoconjugate is increased. 19. Norrgren K, Strand S-E. Ingvar C. Contrast enhancement in RII and modification of the therapeutic ratio in RIT: a theoretical evaluation of simulated extracorporeal immunoadsorption. Antibody Imtmmoconj Radiopharm 1992:5:61 73. CONCLUSION 20. Garkavij M. Tennvall J. Strand S-E, Norrgren K. Nilsson R, Lindgren L. Improving The present investigation has shown that ECIA can improve radioimmunotargeting of tumors: variation in the amount of L6 MAb administered, and maintain the T/N activity ratio by direct removal of combined with an immunoadsorption system (ECIA). Acia Oncol 1993:32:853-859. unbound circulating biotinylated '8*Re-chiBR96 without signif 21. Lear JL, Kasliwal RK. Feyerabend AJ. et al. Improved tumor imaging with radiola- bcled monoclonal antibodies by plasma clearance of unbound antibody with anti-body icantly interfering with the tumor uptake. Also, the liver is not column. Radiolog\- 1991:179:509-512. overloaded with radioactivity as is the case in the avidin chase. 22. Dienhart DG, Kasliwal R. Lear JL, et al. Extracorporeal immunoadsorption of radiolabcled monoclonal antibody: a method for reduction of background radioactivity The additional burden of performing ECIA in clinical cancer and its potential role during the radioimmunotherapy of cancer. Antibody Imnntnocon] therapy would be warranted, if it were proven to be successful Radiopharm 1994:7:225-252. in treating distant métastases.The therapy would be comparable 23. Paganelli G, Stella M, Zito F. et al. Radioimmunoguidcd surgery using iodine-125- labeled biotinylated monoclonal antibodies and cold avidin. J NucÃMed I994;35: to the dialysis of uremia patients. 1970-1975. 24. Marshall D. Pcdley RB. Boden JA. Boden R. Begent RHJ. Clearance of circulating ACKNOWLEDGMENTS radio-antibodies using streptavidin or second antibodies in a xenograft model. Br J Cancer 1994:69:502-507. We thank Professor Ingegerd Hellströmand Professor Karl Erik 25. Kobayashi H, Sakahara H, Hosono M. et al. Improved clearance of radiolabcled Hellström for their gift of MAb chiBR96; NeoRx Corporation biotinylated monoclonal antibody following the infusion of avidin as a chase without (Seattle, WA) for the ligand of TFP MAG2-GABA; and Karin decreased accumulation in the target tumor../ NucÃMed 1994:35:1677 1684. 26. Hellström 1, Garrigues HJ. Ciarrigues U, Hellström KE. Highly tumor-reactive, Wingârdh(Department of Radiation Physics, Lund University) and internalizing, mouse monoclonal antibodies to Ley-related cell surface antigens. Eva Gynnstam (Department of Tumor Immunology, Lund Univer Cancer Res 1990:50:2183-2190. 27. Yarnold S. Fell HP. Chimerization of antitumor antibodies via homologous recombi sity) for their assistance. This work was supported by Swedish nation conversion vectors. Cancer Re* 1994:54:506-512. Cancer Foundation Grant 2353-B95-09 X AB, John and Augusta 28. Green NMA. Spectrophotometric assay for avidin and biotin based on binding of dyes Persson's Foundation, Mrs. Berta Kamprad's Foundation, Gunnar by avidin. Biochem J 1965;94:23c-24c. Nilsson's Foundation, Bristol Myers Squibb Research Institute and 29. Norrgrcn K, Strand S-E. Nilsson R. Lindgren L. SjögrenH-O. A general extracorpo real immunoadsorption method to increase tumor-to-normal tissue ratio in radioim- the Medical Faculty of Lund University. munoimaging and radioimmunotherapy. J NucÃMed 1993:34:448- 454. 30. Garkavij M. Tennvall J, Strand S-E, et al. Extracorporeal immunoadsorption from whole-blood based on avidin-biotin concept: evaluation of a new method. Acta Oncol REFERENCES 1996:53:309-312. 1. Kaminski MS. Zasadny KR. Francis IR. et al. Radioimmunotherapy of B-ccll 31. Goldrosen MH. Biddle WC. Pancook J. et al. Biodistribution, pharmacokinetics and lymphoma with I-131-antiBl (ANTI-CD20) antibody. .V Engl J Med 1993:329:459- imaging studies with Re-186-labeled NR-LU-IO whole antibody in LS I74T colonie 465. tumor-bearing mice. Cancer Res 1990:50:7973-7978. 2. Press OW, Eary JF, Appelbaum FR. et al. Phase II trial of I-131-B1 (anti-CD20) 32. Schreiber GJ. Kerr DE. Strategies to enhance the localization of anticancer immuno antibody therapy with autologous stem cell transplantation for relapsed B cell conjugates. Curr Med Chem 1995:2:616-629. lymphomas. Lancet 1995:346:336-340. 33. Dosio F. Magnani P. Paganelli G. Samuel A. Chiesa G, Fazio F. Three-step tumor 3. Riethmuller G. Sehneider-Gädicke li. Schlimok G, et al. Randomised trial of monoclonal antibody for adjuvant therapy of respected Dukes' C colorectal carcinoma. prc-targeting in lung cancer immunoscmtigraphy. J NucÃBiol Med 1993:37:228-232. Lancet 1994:343:1177-1183. 34. Hnatowich DJ. Virzi F. Rusckowski M. Investigations of avidin and biotin for imaging 4. C'anevari S. Colombarii M, Colnaghi Ml. Immunoconjugates. Lessons from animal applications. J NucÃMed 1987:28:1294-1302. models. Ann Oncal 1994:5:698-701. 35. Khawli LA, Alauddin MM, Miller GK, Epstein AL. Improved immunotargeting of tumors with biotinylated monoclonal antibodies and radiolabeled streptavidin. Ami- 5. Goldenherg DM. Cancer radioimmunotherapy: passing through adolescence. Tumor Targeting 1995:1:233-225. body llrinilinocotij Rudiopharm 1993:6:13- 27. 36. Saga T, Weinstein JN, Jeong JM, et al. Two-step targeting of experimental lung 6. Trail PA. Willner D. Lasch SJ. et al. Cure of xenografted human carcinomas by BR96 doxorubicin immunoconjugates. Science 1993:261:212 215. métastaseswith biotinylated antibody and radiolabeled streptavidin. Cancer Res 7. Dincalci M. Immunoconjugate. The magic bullet revisited or future tumor-targeting 1994:54:2160 2165. 37. Goodwin DA. Tumor pre-targeting. Almost the bottom line. J NucÃMed 1995:36: therapy? Ann Oncol 1994:5:697. 876-879. 8. Garrigues J. Garrigues U. Hellström I, Hellström KE. Ley specific antibody with potent anti-tumor activity is internalized and degraded in lysosomes. Am J Puthol 38. Paganelli G. Magnani P. Zito F, et al. Three-step monoclonal targeting in carcinoem- 1993:142:607-622. bryonic antigen-positive patients. Cancer Res 1991:51:5960-5966. 9. Casazza AM, Trail PA. HcllströmKE. Drug immunotargeting for carcinoma: a reality 39. Zhengsheng Y, Melie Z, Hisataka K. et al. Improved targeting of radiolabeled at last? Ann Oncol 1994:5:703-708. streptavidin in tumors pretargeted with biotinylated monoclonal antibodies through an 10. Chen J. Strand S-E. Sjögrcn H-O, et al. Biodistribution and pharmacokinetics of avidin chase. J NucÃMed 1995:36:837-841. 125I/'"l pair-labeled biotinylated chimeric BR96 in colon carcinoma ¡sotransplanted 40. Blumenthal RD. Sharkey RM, Natale AM, Hess J, Goldcnberg DM. Second antibody rals. Tumor Targeting 1997:2:204-214. administration affects the immunogenicity of a primary xenogenetic antibody. Anti 11. Chen J. Strand S-E, Brechbiel MW. Gansow OA. Sjögren H-O. Combination of body Immunoconj Radiopharm 1994:7:43-55. biotinylation and In-Ill labeling chimeric BR96 with SCN-Bz-CHX-A-DTPA. 41. Paganelli G. Stella M. DeNardi P. et al. A new method for faster blood clearance in biodislribution and pharmacokinetics in colon carcinoma isografted rats. Tumor radioimmuno-guided surgery. J NucÃMed Allied Sci 1990:35:88-89. Targeting 1996:2:66-75. 42. Rosebrough SF. Pharmacokinetics and biodistribution of radiolabeled avidin, strepta 12. Garkavij M. Tennvall J. Strand S-E, et al. Enhanced tumor radioimmunotargeting of vidin and biotin. NucÃMed Biol 1993:20:663-668. chimeric l-125-BR96-biotin in a syngeneic rat tumor model using whole blood 43. Kobayashi H. Sakahara H. Endo K. Yao Z, Toyama S. Konishi J. Repeating the avidin extracorporeal immunoadsorption. J NucÃMed 1997:38:895-901. chase markedly improved the biodistribution of radiolabeled biotinylated antibodies 13. Chen J. Strand S-E. SjögrcnH-O. Optimization of the radioiodination and biotinylation and promoted the excretion of additional background radioactivity. Eur J Cancer of monoclonal antibody chimeric BR96: an indirect labeling using N-succinimidyl-3- 1995:31:1689-1696. EXTRACORPOREALIMMUNOADSORPTIONANDAVIDINCHASE•Chen et al. 1939