Purpose: A disadvantage of conventionally radioiodinated monoclonal antibodies (mAb) for cancer therapy is the short retention time of the radionuclide within target cells. To address this issue, we recently developed a method in which radioiodine is introduced onto antibodies using an adduct consisting of a nonmetabolizable peptide attached to the aminopolycarboxylate diethylenetriaminepentaacetic acid, designated IMP-R4. This adduct causes the radioiodine to become trapped in lysosomes following antibody catabolism. Clinical-scale production of 131I-IMP-R4-labeled antibodies is possible using a recently developed facile method.

Experimental Design: The properties of 131I-IMP-R4-labeled anticarcinoembryonic antigen (CEA) humanized mAb hMN-14 were compared with the directly radioiodinated hMN-14 (131I-hMN-14) in CEA-expressing human colon cancer cell lines, LoVo and LS174T, and in nude mice bearing established LoVo tumor xenografts.

Results:125I-IMP-R4-hMN-14 retention in the cell lines was significantly increased (61.5% after 3 days) compared with 125I-hMN-14. In vivo, a significant improvement in tumor accretion of radiolabel was obtained using 131I-IMP-R4-hMN-14, which led to a marked improvement in therapeutic efficacy. Eight weeks post-treatment, mean tumor volumes were 0.16 ± 0.19 and 1.99 ± 1.35 cm3 in mice treated with 131I-IMP-R4-hMN-14 and 131I-hMN-14, respectively, with complete remissions observed in 27% of mice treated with 131I-IMP-R4-hMN-14 and none using 131I-hMN-14.

Conclusion:131I-IMP-R4-hMN-14 provides a significant therapeutic advantage in comparison to the conventionally 131I-labeled antibody. The ability of this labeling method to lend itself to clinical-scale labeling, the broad applicability of a humanized anti-CEA mAb for CEA-expressing cancers, and the clinical benefits of radioimmunotherapy with anti-CEA mAb shown recently for small-volume and minimal residual disease combine to make 131I-IMP-R4-hMN-14 a promising new agent for radioimmunotherapy.

With its success in lymphoma patients (1), radioimmunotherapy is evolving into a conventional modality for cancer therapy. Ibritumomab tiuxetan and Tositumomab are two radiolabeled antibody products directed against the CD20 antigen that have been approved by the U.S. Food and Drug Administration for radioimmunotherapy of non-Hodgkin's lymphoma. Although radioimmunotherapy of hematologic malignancies has exhibited a certain measure of success, solid tumors have been less responsive, especially when large bulky disease is present. Our goal was to make 131I, a readily available and popular radioisotope with imaging and therapeutic properties, more effective for radioimmunotherapy by a new facile method applicable to virtually any antibody.

Radioimmunotherapy, a therapeutic modality in which radiolabeled monoclonal antibodies (mAb) are used to selectively target ionizing radiation to tumor sites, has principally involved the β emitters 131I and 90Y. Each of these radiolabels has advantages and disadvantages warranting the use of each, possibly for different applications. 90Y mAbs exhibit greater accretion to tumor targets than 131I mAbs due to the trapping of chelated radiometals within target cells after protein catabolism. In contrast, the catabolic product of radioiodinated antibody, 131I-iodotyrosine, is able to diffuse out of target cells leading to shortened residence times and correspondingly lower radiation doses delivered to the tumor target (2).

To make 131I more effective in radioimmunotherapy, there has been an ongoing effort to develop new methods of radioiodination designed to trap radioiodine inside a tumor cell following delivery by a labeled mAb. Such intracellulary retained labels are called residualizing radiolabels. The underlying principle behind these approaches is the inclusion of 131I as a component of a nonmetabolizable moiety, chemically designed to be lysosomally trapped after catabolism of the carrier mAb. Early work was based on the use of nonmetabolizable carbohydrates as linking agents. Dilactitoltyramine (3, 4) and tyramine cellobiose (5) are two substrates examined for this purpose. Using dilactitoltyramine, radioiodine accretion in tumor cells increased over 3-fold in vitro (4), and pronounced dosimetric and therapeutic advantages were obtained in animal experiments using nude mice bearing human lung tumor (6) and lymphoma xenografts (7, 8). Using tyramine cellobiose, radioiodine accretion was over twice greater than that of conventionally iodinated antibody, and plasma clearance and uptake in normal tissues was not changed (5). An alternative technique for preparing residualizing labels involved the use of positively charged moieties, including pyridine-based N-succinimidyl 5-[131/125I]iodo-3-pyridinecarboxylate (9) and N-succinimidyl 4-guanidinomethyl-3-iodobenzoate (10). Another approach involves the use of nonmetabolizable peptide adducts. We described the use of diethylenetriaminepentaacetic acid–appended radioiodinated peptides containing D-amino acids as a class of residualizing 131I label (1113). Foulon et al. reported the use of a radioiodinated basic pentapeptide (Lys-Arg-Tyr-Arg-Arg), containing all D-amino acids, as a residualizing label for anti-EGFRvIII mAb, L8A4 (14). A recent report by Shankar et al. (15) described the use of a negatively charged acylating agent, N-succinimidyl 3-131Iiodo-4-phosphonomethylbenzoate (131I-SIPMB), for labeling peptides and mAbs.

All the above approaches for the production of residualizing iodine have yielded enhanced tumor accretion and, where evaluated, substantial improvements in therapeutic efficacy. In comparison with the radiometal-labeled mAbs, the residualizing iodine-labeled mAbs examined by us yielded similar tumor uptake, retention, and efficacy in human tumor xenografts (4, 12, 16, 17). Given that most of the β-energy of 131I is absorbed within a small radius, 131I may be more cytotoxic for small and micrometastatic tumors than a higher energy β emitting isotope such as 90Y, in which a higher proportion of the dose would be dissipated outside the tumor, possibly in neighboring normal tissues. In addition, the longer half-life of 131I should be an advantage for radioimmunotherapy, where there is relatively slow uptake and accumulation of the radiolabel within tumors. The 2.7-day half-life of 90Y may not be long enough to take full advantage of the retention of the label at the tumor site. Another difference between 131I and 90Y is that 131I has a γ emission permitting localization and quantitation of radioactivity in tissues by imaging. Finally, 131I has an ∼10-fold cost advantage over 90Y, assuming the cost of the manufacture of the radiolabeled mAb to be the same for both nuclides, for respective levels of radioactivity approximately corresponding to nonmyeloablative doses used in the clinical radioimmunotherapy of hematologic malignancies.

Despite the compelling advantages of residualizing iodine for radioimmunotherapy these agents have not met with widespread acceptance for clinical use, mainly because of limitations imposed by the chemical syntheses. Low conjugation efficiency and unacceptable antibody aggregation precluded the clinical use of dilactitoltyramine (4) and tyramine cellobiose (5), respectively. In addition, production of the residualizing agents described above involved multistep procedures of radioiodination, activation, conjugation to mAb, and purification. For example, with the pentapeptide of Foulon et al., an intermediate purification step at the radioiodination stage was reported as well as purification of the radioiodinated ligand-mAb conjugate on a size exclusion column done at the end of the process (to remove unincorporated radioiodide and unconjugated 131I-radioiodinated small molecular mass moieties; ref. 14). The use of 131I-SIPMB necessitated a high-performance liquid chromatography purification at the radioiodination stage and a PD10 column purification after mAb conjugation (15).

We recently reported the development of 131I-IMP-R4, an improved residualizing form of 131I that overcomes many of the limitations that have impeded the development of residualizing iodine for clinical use (17, 18). Importantly, a practical method for producing residualizing 131I-IMP-R4-labeled mAb at levels needed clinically, which does not involve column purifications and is not more complex than that used in direct radioiodinations, has been developed (19). IMP-R4 contains D-amino acids in the peptide to confer proteolytic stability in lysosomes, D-tyrosine to provide a radioiodination site, diethylenetriaminepentaacetic acid moieties to increase hydrophilicity and aid in intracellular retention, and two maleimide groups for protein binding. Using this adduct to label hRS7, a humanized mAb recognizing EGP-1, an antigen highly expressed in carcinomas of breast, lung, ovary and prostate, marked improvement in tumor uptake and retention and therapeutic efficacy was observed in preclinical models of human lung and breast cancers as compared with conventional radioiodine labeling (17, 18).

In this article, we show the potential of 131I-IMP-R4-labeled mAb for use in the therapy of carcinoembryonic antigen (CEA)–expressing cancers using colon cancer models. The humanized high-affinity anti-CEA monoclonal antibody hMN-14 was used for radioimmunoconjugate production. Although CEA is well known as a secreted tumor-associated antigen, present in the circulation, and used as a marker of disease progression, it is also present on the surface of many tumor cells, and there have been reports of internalization of anti-CEA antibodies (2022). We show that the residualizing adduct IMP-R4 can improve radioiodine retention in colon cancer cells, leading to increased radioiodine accretion in human tumor xenografts and marked improvement in therapeutic efficacy in the model system. With its advantage established previously using a rapidly internalizing mAb in two other cancer models, IMP-R4-based 131I-labeling technology is now evolving into a general method for improving therapeutic outcomes with 131I-radioimmunotherapy.

Monoclonal antibodies and cell lines. Human colon carcinoma cell lines, LoVo and LS174T, were purchased from the American Type Culture Collection (Rockville, MD). The cells were grown in DMEM (Life Technologies, Gaithersburg, MD) supplemented with 10% fetal bovine serum, penicillin (100 units/mL), streptomycin (100 μg/mL), and l-glutamine (2 mmol/L). The cells were routinely passaged after detachment with trypsin, 0.2% EDTA.

MN-14 is a class-III anti-CEA (CD66e, CEACAM5) mAb, reacting with CEA and unreactive with the normal cross-reactive antigen, NCA, biliary antigen, and meconium antigen (23). The construction and characterization of the humanized form of MN-14 (hMN-14, labetuzumab) has been described previously (24). hMN-14 was provided by Immunomedics, Inc. (Morris Plains, NJ).

Radioiodination. The peptide IMP-R4 {4-(N-maleimidomethyl)-cyclohexane-1-carbonyl-Lys[4-(N-maleimidomethyl)-cyclohexane-1-carbonyl]-Lys(1-[(p-thiocarbonylamino)benzyl)diethylenetriaminepentaacetic acid])-d-Tyr-d-Lys(1-[(p-thiocarbonylamino)benzyl)diethylenetriaminepentaacetic acid])-OH} was used for preparation of residualizing radioiodine-labeled mAb. Preparation of the IMP-R4 peptide, and the procedures for radioiodination of IMP-R4 and conjugation to reduced mAb were as described previously (11, 17). Direct radioiodination of hMN-14 was carried out by the chloramine-T method (25). Assessment of immunoreactivity was done by size exclusion high-performance liquid chromatography analysis following incubation of the radioiodinated antibody with the antigen, CEA (Scripps Laboratory, Inc., San Diego, CA).

Binding and processing of labeled hMN-14 by colon cancer cell lines.In vitro processing experiments were carried out using the published procedure (26). Cells were plated at 50,000 cells per well in a 96-well tissue culture plate. After 24 hours, cells were incubated in triplicate with 5 × 105 cpm of radioiodinated antibody (125I-IMP-R4-hMN-14 or conventionally iodinated 125I-hMN-14) in 0.1 mL of tissue culture medium for 2 hours at 37°C, washed four times to remove unbound antibody, and the radioactivity associated with the cells was counted. This was the total cell-bound radioactivity at “zero” time, which was taken as 100%. Media was added to all other wells, and the incubation was continued for 69 hours. Samples were taken at 4, 21, 45, and 69 hours. At each time point, the supernatant was removed and counted in a γ counter to determine radioactivity released to the medium. After counting, trichloroacetic acid was added to the supernatant in the presence of a carrier protein. Samples were centrifuged, and the radioactivity associated with the pellet was counted to determine the level of undegraded antibody in the supernatant. Cells were solubilized in 2 mol/L NaOH, and the cell-bound radioactivity was determined. All data were expressed as percent of “zero” time cell-bound radioactivity. Nonspecific binding was determined by adding an excess of unlabeled hMN-14 before the addition of the radiolabeled antibodies.

In vivo experiments. For targeting, 5 μCi 125I-hMN-14 (CT) and 12.5 μCi of 131I-IMP-R4-hMN-14 were combined in a final injectate volume of 0.15 mL/animal, and injected i.v. into 20 tumor-bearing mice. Groups of five animals were sacrificed at indicated times. Tumor and normal tissues were counted for radioactivity in a γ counter, and corrections were made for backscatter of 131I counts into the 125I counting window. The results were expressed as percent injected dose per gram of tissue. Radiation dose estimates were determined from 131I-IMP-R4-hMN-14 and 125I-hMN-14 (131I surrogate) biodistribution data. Calculations were carried out by first integrating the trapezoidal regions (for tumors) or exponential regions (for normal organs) defined by the time-activity data corrected for physical decay. To avoid overestimation of the tumor cumulative dose, a zero-time value of zero was assumed for the trapezoidal fit of the tumor. For other tissues, the zero time point was extrapolated according to the exponential curve. The resulting integral for each organ was converted to cGy/mCi using S values appropriate for organ weight and calculated by assuming uniformly distributed activity in small unit-density spheres (27). For therapy, seven groups of 11 mice each, consisting of the following, were used: untreated group, 250, 275, and 300 μCi 131I-hMN-14 (CT) dose groups and 225, 250, and 275 μCi of 131I-IMP-R4-hMN-14 dose groups. Protein doses in the therapy groups were adjusted with unlabeled hMN-14 to account for the different specific activities of the products of the IMP-R4 and chloramine-T methods. Protein doses given were all within the range of 60 to 90 μg/dose. The groups were randomized to contain similar assortment of initial tumor volumes. Baseline data were compared with weekly measurements of body weights, tumor volumes, and WBC counts. Animals with progressive tumor growth reaching 3 cm3 were euthanized. All animal experiments were carried out in accord with Institutional Animal Care and Use Committee–approved protocols.

Statistical analyses. For in vitro processing and in vivo targeting experiments, statistical significance was determined by Student's t test. For therapy experiments, statistical analyses on different treatment groups were done by the Student's t test on the area under the growth curves. Two-sided tests were used.

In vitro studies

Cellular retention of radionuclide, after labeled mAbs were bound to the cell surface, was studied in colon cancer cell lines using an in vitro assay. Retention of hMN-14 labeled with 125I using IMP-R4 was compared with that of directly labeled hMN-14 by measuring the percent of originally bound radionuclide still bound to the cells at indicated times after removal of excess labeled mAb. Data are shown for LS174T and LoVo colon carcinoma cell lines in Table 1 and Fig. 1. In both cell lines, a greater percentage of the radioiodine remained associated with cells at the later time points when the residualizing label, 125I-IMP-R4, was used. At 45 and 69 hours 125I-IMP-R4-hMN-14 showed 48.3% greater (P = 0.002) and 61.5% greater (P = 0.0003) levels of cell-bound radioactivity than the directly labeled antibody in LoVo. In the LS174T cell line, 125I-IMP-R4-hMN-14 also yielded higher levels of cell-bound radioactivity compared with the directly labeled antibody at 45 and 69 hours (P = 0.02 and 0.03, respectively). The percent of intact antibody, as measured by trichloroacetic acid–precipitable radioactivity in the supernatant, was the same for both radiolabels, indicating that the differences observed were due to internalization and catabolism of the antibody, rather than loss of intact mAb. Nonspecific binding was <10% for both LoVo and LS174T cell lines.

Table 1.

In vitro processing by LoVo and LS174T cell lines

% Cell-bound cpm (mean ± SD)
Cell lineRadiolabel45 h69 h
LoVo 12548.2 ± 5.0 40.3 ± 3.3 
 125I-IMP-R4 71.5 ± 6.6 65.1 ± 1.8 
LS174T 12557.7 ± 7.5 51.6 ± 6.1 
 125I-IMP-R4 73.2 ± 1.9 66.1 ± 4.5 
% Cell-bound cpm (mean ± SD)
Cell lineRadiolabel45 h69 h
LoVo 12548.2 ± 5.0 40.3 ± 3.3 
 125I-IMP-R4 71.5 ± 6.6 65.1 ± 1.8 
LS174T 12557.7 ± 7.5 51.6 ± 6.1 
 125I-IMP-R4 73.2 ± 1.9 66.1 ± 4.5 
Fig. 1.

Processing of radioiodinated hMN-14 by LoVo (A) and LS174T (B) cell lines. Composite data on the cell-bound (▪, □) and supernatant (▴, ▵) radioactivity, as percent of initially bound cpm, for both hMN-14 radiolabeled with 125I-IMP-R4 (▪, ▴) and directly labeled 125I-hMN-14 (□, ▵). Trichloroacetic acid–precipitable radioactivity for each label (•).

Fig. 1.

Processing of radioiodinated hMN-14 by LoVo (A) and LS174T (B) cell lines. Composite data on the cell-bound (▪, □) and supernatant (▴, ▵) radioactivity, as percent of initially bound cpm, for both hMN-14 radiolabeled with 125I-IMP-R4 (▪, ▴) and directly labeled 125I-hMN-14 (□, ▵). Trichloroacetic acid–precipitable radioactivity for each label (•).

Close modal

In vivo studies

Paired-label biodistribution. Targeting of hMN-14 labeled with 131I-IMP-R4 was compared with that of directly radioiodinated hMN-14 in paired labeled biodistribution studies. Figure 2A summarizes the accretion of radioiodine in tumor and normal tissues over a 14-day time period. 131I-IMP-R4-hMN-14 yielded a progressively enhanced retention in tumor compared with the directly radioiodinated antibody prepared using the conventional chloramine-T method [125I-hMN-14 (CT)]. For example, % ID/g values in LoVo human colon cancer xenografts 7 days post-injection of dual-labeled hMN-14 were 33.1 ± 6.3 and 26.6 ± 4.8 (P = 0.005) for 131I-IMP-R4-hMN-14 and 125I-hMN-14 (CT), respectively. The percent injected dose (ID)/g in tumor and blood for the two labels are shown for individual mice in Fig. 2B. In each animal, IMP-R4 labeling yielded an increase %ID/g in tumor in the range of 16% to 36% (mean = 25 ± 8%) on day 7 and 23% to 31% (mean = 26 ± 3%) on day 14 (data not shown). In addition, the residualizing label cleared from blood more quickly (range = 26-30%, mean = 28 ± 1%) on day 7 and 18% to 22% (mean = 20 ± 1%) on day 14 (data not shown). Resulting mean tumor/blood ratios on day 7 were 4.78 ± 0.88 (range = 3.96-6.19) and 2.76 ± 0.42 (range = 2.32-3.29) for the residualizing and conventional labels, respectively (P < 0.001), a mean increase of 73 ± 10%.

Fig. 2.

Biodistribution of radiolabeled hMN-14 in nude mice bearing LoVo tumors. Double-label biodistribution study done in which tumor-bearing mice were given 5 μCi hMN-14 labeled with 131I-IMP-R4 and 12.5 μCi hMN-14 labeled with 125I using the conventional chloramine-T labeling method. Mice with established tumors received injections i.v. on day 0 with the radiolabels and were sacrificed successively. A, mean values for %ID/g over the 14-day time course (points); SD (bars). Solid lines and filled symbols, 131I-IMP-R4-hMN-14; dashed lines and open symbols, conventionally iodinated 125I-hMN-14; ⧫ and ◊, tumor; • and ○, blood; ▪ and □, liver; ▴ and △, spleen; ×, kidney. B, %ID/g values for individual mice euthanized 7 days after radiolabeled mAb injections. Black columns, 125I-hMN-14 in tumor; dark gray columns,131I-IMP-R4-hMN-14 in tumor; white columns, 125I-hMN-14 in blood; light gray columns,131I-IMP-R4-hMN-14 in blood.

Fig. 2.

Biodistribution of radiolabeled hMN-14 in nude mice bearing LoVo tumors. Double-label biodistribution study done in which tumor-bearing mice were given 5 μCi hMN-14 labeled with 131I-IMP-R4 and 12.5 μCi hMN-14 labeled with 125I using the conventional chloramine-T labeling method. Mice with established tumors received injections i.v. on day 0 with the radiolabels and were sacrificed successively. A, mean values for %ID/g over the 14-day time course (points); SD (bars). Solid lines and filled symbols, 131I-IMP-R4-hMN-14; dashed lines and open symbols, conventionally iodinated 125I-hMN-14; ⧫ and ◊, tumor; • and ○, blood; ▪ and □, liver; ▴ and △, spleen; ×, kidney. B, %ID/g values for individual mice euthanized 7 days after radiolabeled mAb injections. Black columns, 125I-hMN-14 in tumor; dark gray columns,131I-IMP-R4-hMN-14 in tumor; white columns, 125I-hMN-14 in blood; light gray columns,131I-IMP-R4-hMN-14 in blood.

Close modal

Cumulative absorbed radiation doses were calculated from the biodistribution data shown in Fig. 2. Cumulative absorbed doses for 131I-hMN-14 (CT) were calculated using the 125I-hMN-14 (CT) distribution data. Doses were normalized to blood and are presented in Fig. 3. As shown in Fig. 3, the residualizing label, 131I-IMP-R4-hMN-14, is estimated to increase the dose to tumor dose by 47% in comparison to the conventional label, 131I-hMN-14 (CT). Although doses to organs involved in protein processing, including liver, spleen, and kidney, are also elevated for the residualizing label compared with the direct label, the absolute doses are expected to be in the nontoxic range at the estimated MTD. Toxic levels for these organs are taken to be above 2,000 cGy for spleen and kidney and above 3,000 cGy for liver (28).

Fig. 3.

Mean cumulative absorbed dose in tissues following injection of radiolabeled hMN-14 in nude mice bearing LoVo tumors. Results are calculated from data described in Fig. 2 and are presented as cGy to tissue normalized to cumulative absorbed dose to blood.

Fig. 3.

Mean cumulative absorbed dose in tissues following injection of radiolabeled hMN-14 in nude mice bearing LoVo tumors. Results are calculated from data described in Fig. 2 and are presented as cGy to tissue normalized to cumulative absorbed dose to blood.

Close modal

Radioimmunotherapy. The therapeutic efficacy of 131I-IMP-R4-hMN-14 was compared with that of directly labeled 131I-hMN-14 in nude mice bearing LoVo human colon cancer xenografts. Three escalating doses of each agent were given to groups of 11 mice with established tumors. Initial tumor volumes were 0.220 ± 0.098 cm3 (range = 0.041-0.388 cm3). Doses given were 225, 250, and 275 μCi in the 131I-IMP-R4-hMN-14 dose groups and 250, 275, and 300 μCi in the 131I-hMN-14 (CT) groups. The doses selected for therapy thus had μCi match in the 250 and 275 μCi categories of the two labels but a higher 300 μCi dose in the conventionally treated group, because historical data indicated that the directly iodinated mAb could be tolerated up to this dose. Figure 4A shows the tumor volumes of individual mice in each group over a 10-week period. The results are also shown in terms of mean tumor volumes, in Fig. 4B. Radioimmunotherapy with any radiolabeled hMN-14 preparations used in the study produced significant growth control compared with the untreated group (P for area under the curve = 0.006 at 4 weeks). However, the residualizing 131I-IMP-R4-hMN-14 label at all doses was markedly better than any of the conventional 131I-hMN-14 doses (P for area under the curve < 0.0004 versus the 300 μCi conventional 131I dose at 8 weeks). Overall, the mean tumor volume in the 131I-IMP-R4-hMN-14-treated mice was 0.16 ± 0.19 versus 1.99 ± 1.35 cm3 in the 131I-hMN-14-treated mice at 8 weeks. Complete remissions were observed in six mice in the residualizing 131I groups compared with none in the conventional 131I groups.

Fig. 4.

Comparative radioimmunotherapy in LoVo bearing nude mice. Nude mice bearing established LoVo tumors (mean tumor volume, 0.22 cm3) were treated using increasing doses of 131I-IMP-R4-hMN-14 or conventionally iodinated 131I-hMN-14. Mice were given a single injection of the labeled mAbs. A, tumor volumes of individual animals. B, mean tumor volumes of the respective treatment groups. C, toxicity evaluation. Total WBC counts were evaluated weekly. Points, means; bars, SD (only in one direction). Solid lines,131I-IMP-R4-hMN-14; dashed lines, conventionally iodinated 131I-hMN-14.

Fig. 4.

Comparative radioimmunotherapy in LoVo bearing nude mice. Nude mice bearing established LoVo tumors (mean tumor volume, 0.22 cm3) were treated using increasing doses of 131I-IMP-R4-hMN-14 or conventionally iodinated 131I-hMN-14. Mice were given a single injection of the labeled mAbs. A, tumor volumes of individual animals. B, mean tumor volumes of the respective treatment groups. C, toxicity evaluation. Total WBC counts were evaluated weekly. Points, means; bars, SD (only in one direction). Solid lines,131I-IMP-R4-hMN-14; dashed lines, conventionally iodinated 131I-hMN-14.

Close modal

Toxicity of the radioimmunotherapy treatments was monitored by weekly WBC counts and body weight measurement. All of the tested doses were tolerated, with no resulting treatment-related deaths. The WBC counts for the various groups are shown in Fig. 4C. The nadirs for the residualizing radioiodine-treated mice were lower than that for conventional groups, although for the 225 μCi 131I-IMP-R4-hMN-14 group, differences in WBC counts from the directly iodinated 131I-hMN-14 groups were not statistical significantly at most time points. The WBC counts recovered, albeit more slowly in the case of residualizing iodine labeled groups. None of the treatments caused significant body weight loss.

In this study, the superiority of the IMP-R4 residualizing labeled mAb over directly radioiodinated antibody was shown by in vitro assays and in vivo animal biodistribution and therapy studies. In vitro, after 3 days, cellular retention of radioiodine was increased by 61.5% using 125I-IMP-R4-anti-CEA mAb hMN-14 compared with conventionally iodinated hMN-14. In biodistribution studies in nude mice bearing colon cancer xenografts, 131I-IMP-R4-hMN-14 yielded a mean increase in %ID/g in tumor of 25%, a significant enhancement of retention in tumor compared with the conventionally iodinated antibody. Although this increase is modest compared with the enhancement of radionuclide accretion obtained with the more rapidly internalizing anti-EGP-1 mAb, RS7, previously evaluated in human lung and breast cancer models (where the %ID/g in tumor on day 7 with the IMP-R4-labeled mAbs were 4- and 7-fold the values of directly iodinated mAb), a marked difference in therapeutic efficacy was obtained. At 8 weeks post-treatment, mean tumor volumes were 0.16 ± 0.19 and 1.99 ± 1.35 cm3 in human colon cancer-bearing nude mice treated with 131I-IMP-R4-labeled and directly iodinated-hMN-14, respectively, with complete remissions observed in 27% of mice treated with 131I-IMP-R4-hMN-14 and none using the directly iodinated antibody. The marked tumor growth control achieved using the residualizing 131I radiolabel was not anticipated from dosimetry calculations, which estimated that the radiation dose increase due to the residualizing iodine labeled mAb (47%) was more modest than that observed with the more rapidly internalizing anti-EGP-1 mAb. Apparently, even a small portion of the label internalized and residualized, as a result of membrane turnover, for example, can result in a significant difference in the therapeutic potentials of the two labels.

Colorectal cancer was among the first diseases for which radiolabeled antibodies were investigated as targeting agents. This early application stemmed from the identification of CEA as a suitable target antigen. CEA, also known as CD66e or CEACAM5 (29, 30), was first described in 1965 as a gastrointestinal oncofetal antigen (31) but is now known to be overexpressed in the majority of carcinomas including those of the gastrointestinal tract, the respiratory and genitourinary systems, and breast cancer (3236). Radiolabeled antibodies to CEA were first applied to the detection of cancer using purified polyclonal anti-CEA immunoglobulin G labeled with 131I (37). Despite this long history, successful radioimmunotherapy of colon cancer, as well as of other solid tumors, is still limited (3845), probably due to the low specific accretion of the radiolabeled antibody in the tumor target compared with normal tissues and the relative radioresistance of these tumors.

Nevertheless, because tumor uptake, and thus the radiation dose to the tumor, increases with decreasing tumor size, it is likely that radioimmunotherapy can be applied with greater success in small volume and minimal residual disease, as has recently been shown in clinical studies with 131I-labeled humanized anti-CEA antibody hMN-14 (46, 47). An overall response rate of nearly 60% (∼20% objective, 40% minor responses) was obtained in patients with known metastatic disease. In addition, seven of nine patients treated in an adjuvant setting after resection of liver metastases remained disease free for up to 36 months, whereas the relapse rate in a corresponding historical control group receiving chemotherapy was 67% over the same time frame. Thus, it is likely that the benefits of radioimmunotherapy can be optimized by using 131I-labeled mAbs in small-volume and minimal residual disease.

These encouraging results in solid tumors can most likely be improved by increasing radioiodine accretion in the tumor by using a residualizing methodology as described here. Although the rate and extent of antibody internalization differs for various mAbs, both rapidly and slowly internalizing mAbs are taken into the cell and catabolized. The rapidly internalizing mAbs presumably enter their target cells by way of receptor-mediated mechanisms, whereas membrane turnover apparently accounts for the slower internalization. Thus, residualizing radiolabels affect the radiation dose delivered to target cells for slowly internalizing mAbs, such as anti-CEA and anti-epithelial glycoprotein-2 (EGP-2) mAb, RS11 (4), as well as the more rapidly internalizing mAbs, such as anti-epithelial glycoprotein-1 (EGP-1) mAb, RS7 (2, 4, 26), CD22 mAb (8), and anti-HER-2/neu mAbs (48).

Additional methods aimed at improving tumor accretion of radionuclides are under active investigation. Among the approaches taken are pretargeting methods (49), the use of genetically engineered immunoconstructs with molecular mass intermediate between that of intact immunoglobulin G and Fab′ fragments (50, 51), and combination of radioimmunotherapy with other therapeutic modalities such as external beam irradiation or chemotherapy (5254). Chelated radiometals represent an alternate source of residualizing labels available for radioimmunotherapy, remaining trapped in lysosomes in the form of lysine adducts of the respective metal chelates (55). Our previous studies predicted an increase in absorbed dose to tumor using residualizing 131I-labeled mAbs in comparison to 90Y, due to the longer physical half-life of 131I (4). This advantage would also be predicted for radiometals of longer half-life, assuming that they are trapped inside target cells as well as the residualizing iodine and 90Y. 177Lu is one such radiometal, with radiophysical properties resembling those of 131I, having a 6.7-day half-life, moderate 496-keV maximum β emissions, and low abundance γ emissions. However, in a direct comparison of 90Y, 177Lu, and residualizing 131I-labeled RS7 in a nude mouse-human lung cancer xenograft model, tumor targeting, therapeutic efficacy, and toxicity of the three radionuclides were found to be similar (16). 131I-IMP-R1 was used in that study, an earlier version of the residualizing 131I adduct differing from IMP-R4 in labeling efficiency due to its having only one maleimide and one benzyl-diethylenetriaminepentaacetic acid rather than two of each group in IMP-R4. Similar results were also observed in a comparison of 90Y and residualizing 131I-IMP-R4 labeled RS7 in the lung cancer xenograft model (17). The similarity of therapeutic efficacies of 90Y and residualizing 131I in these may be due to the s.c. tumor models used, wherein the tumor sizes are relatively large. It would be interesting to compare these radionuclides in micrometastatic tumor models or in adjuvant settings after debulking surgery, situations which are considered particularly amenable to radioimmunotherapy and which may provide a particular advantage for 131I with its shorter tissue range of β emission. A direct comparison of this new method with other β-emitting isotopes, including 90Y and 177Lu, in additional tumor models would also be of interest in future studies.

In summary, at the present time for solid tumor therapy, radioimmunotherapy is expected to have the greatest potential when applied to patients with small tumor volume or as an adjuvant therapy. Based on its β-particle energy, tissue range, half-life, and ready availability, 131I is the radionuclide of choice for this application. The development of a practical residualizing 131I method, in the form of 131I-IMP-R4, overcomes the problem of reduced tumor dose due to a directly radioiodinated mAb. Using the anti-CEA antibody, hMN-14, in an in vivo tumor model of colon carcinoma, we showed a marked therapeutic advantage of residualizing radioiodine. It is noteworthy that this effect was observed although the anti-CEA mAb is not considered to be a rapidly internalizing mAb and the dosimetric improvement was not as large as seen in the lung and breast cancer models using rapidly internalizing antibody. Thus, the benefit of residualizing labels has now been shown for rapidly internalizing and nonrapidly internalizing mAbs and in three cancer models, and is expected to be applicable to other cancer types.

Grant support: NIH grant CA103312.

Note: Presented in part at the AACR/National Cancer Institute/European Organization for Research and Treatment of Cancer International Conference on Molecular Targets and Cancer Therapeutics, Boston, November 17, 2003 and at the 51st Annual Meeting of the Society of Nuclear Medicine, Philadelphia, June 19-23, 2004.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Susan Chen, Adriane Rosario, and Philip Andrews for their excellent technical assistance.

1
Goldenberg DM. Advancing role of radiolabeled antibodies in the therapy of cancer.
Cancer Immunol Immunother
2003
;
52
:
281
–96.
2
Mattes MJ, Griffiths GL, Diril H, Goldenberg DM, Ong GL, Shih LB. Processing of antibody-radioisotope conjugates after binding to the surface of tumor cells.
Cancer
1994
;
73
:
787
–93.
3
Thorpe SR, Baynes JW, Chroneos ZC. The design and application of residualizing labels for studies of protein catabolism.
FASEB J
1993
;
7
:
399
–405.
4
Stein R, Goldenberg DM, Thorpe SR, Mattes MJ. Effects of radiolabeling monoclonal antibodies with a residualizing iodine radiolabel on the accretion of radioisotope in tumors.
Cancer Res
1995
;
55
:
3132
–9.
5
Ali SA, Warren SD, Richter KY, et al. Improving the tumor retention of radioiodinated antibody: arylcarbohydrate adducts.
Cancer Res
1990
;
50
:
783
–8s.
6
Stein R, Goldenberg DM, Thorpe SR, Mattes MJ. Advantage of a residualizing iodine radiolabel for radioimmunotherapy of xenografts of human non-small cell carcinoma of the lung.
J Nucl Med
1997
;
38
:
391
–5.
7
Mattes MJ, Shih LB, Govindan SV, et al. The advantage of residualizing radiolabels for targeting B-cell lymphomas with a radiolabeled anti-CD22 monoclonal antibody.
Int J Cancer
1997
;
71
:
429
–35.
8
Sharkey RM, Behr TM, Mattes MJ, et al. Advantage of residualizing radiolabels for an internalizing antibody against the B-cell lymphoma antigen, CD22.
Cancer Immunol Immunother
1997
;
44
:
179
–88.
9
Reist CJ, Garg PK, Alston KL, Bigner DD, Zalutsky MR. Radioiodination of internalizing monoclonal antibodies using N-succinimidyl 5-iodo-3-pyridinecarboxylate.
Cancer Res
1996
;
56
:
4970
–7.
10
Vaidyanathan G, Affleck DJ, Bigner DD, Zalutsky MR. Improved xenograft targeting of tumor-specific anti-epidermal growth factor receptor variant III antibody labeled using N-succinimidyl 4-guanidinomethyl-3-iodobenzoate.
Nucl Med Biol
2002
;
29
:
1
–11.
11
Govindan SV, Mattes MJ, Stein R, et al. Labeling of monoclonal antibodies with DTPA-appended radioiodinated peptides containing D-amino acids.
Bioconjug Chem
1999
;
10
:
231
–40.
12
Stein R, Govindan SV, Mattes MJ, et al. Targeting human cancer xenografts with mAbs labeled using radioiodinated, DTPA-appended, peptides.
Clin Cancer Res
1999
;
5
:
3079
-87s.
13
Stein R, Govindan SV, Chen S, et al. Successful therapy of a human lung cancer xenograft using mAb RS7 labeled with residualizing radioiodine.
Crit Rev Oncol Hemat
2001
;
39
:
173
–80.
14
Foulon CF, Reist CJ, Bigner DD, Zalutsky MR. Radioiodination via D-amino acid peptide enhances cellular retention and tumor xenograft targeting of an internalizing anti-epidermal growth factor receptor variant III monoclonal antibody.
Cancer Res
2000
;
60
:
4453
–60.
15
Shankar S, Vaidyanathan G, Affleck D, Welsh PC, Zalutsky MR. N-succinimidyl 3-[([131])I]iodo-4-phosphonomethylbenzoate ([([131])I]SIPMB), a negatively charged substituent-bearing acylation agent for the radioiodination of peptides and mAbs.
Bioconjug Chem
2003
;
14
:
331
–41.
16
Stein R, Govindan SV, Chen S, et al. Radioimmunotherapy of a human lung cancer xenograft with monoclonal antibody RS7: evaluation of 177Lu and comparison of its efficacy with that of 90Y and residualizing 131I.
J Nucl Med
2001
;
42
:
967
–74.
17
Stein R, Govindan SV, Mattes MJ, et al. Improved iodine radiolabels for monoclonal antibody therapy.
Cancer Res
2003
;
63
:
111
–8.
18
Govindan SV, Stein R, Qu Z, et al. Preclinical therapy of breast cancer with a radioiodinated humanized anti-EGP-1 monoclonal antibody: advantage of a residualizing iodine radiolabel.
Breast Cancer Res Treat
2004
;
84
:
173
–82.
19
Govindan SV, Griffiths GL, Stein R, et al. Clinical-scale radiolabeling of a humanized anti-CEA mAb, hMN-14, with residualizing 131I for use in radioimmunotherapy.
J Nucl Med
2005
;
46
:
153
–9.
20
Tsaltas G, Ford CH, Gallant M. Demonstration of monoclonal anti-carcinoembryonic antigen (CEA) antibody internalization by electron microscopy, Western blotting and radioimmunoassay.
Anticancer Res
1992
;
12
:
2133
–42.
21
Ford CH, Tsaltas GC, Osborne PA, Addetia K. Novel flow cytometric analysis of the progress and route of internalization of a monoclonal anti-carcinoembryonic antigen (CEA) antibody.
Cytometry
1996
;
23
:
228
–40.
22
Shih LB, Goldenberg DM, Xuan H, Lu HW, Mattes MJ, Hall TC. Internalization of an intact doxorubicin immunoconjugate.
Cancer Immunol Immunother
1994
;
38
:
92
–8.
23
Hansen HJ, Goldenberg DM, Newman ES, Grebenau R, Sharkey RM. Characterization of second-generation monoclonal antibodies against carcinoembryonic antigen.
Cancer
1993
;
71
:
3478
–85.
24
Sharkey RM, Juweid M, Shevitz J, et al. Evaluation of a complementarity-determining region-grafted (humanized) anti-carcinoembryonic antigen monoclonal antibody in preclinical and clinical studies.
Cancer Res
1995
;
55
:
5935
-45s.
25
McConahey PJ, Dixon FJ. A method of trace iodination of proteins for immunologic studies.
Int Arch Allergy Immunol
1966
;
29
:
185
–9.
26
Kyriakos RJ, Shih LB, Ong GL, Patel K, Goldenberg DM, Mattes MJ. The fate of antibodies bound to the surface of tumor cells in vitro.
Cancer Res
1992
;
52
:
835
–42.
27
Siegel JA, Stabin MG. Absorbed fractions for electrons and β particles in spheres of various sizes.
J Nucl Med
1994
;
35
:
152
–6.
28
Rubin P, Casarett GW. Clinical radiation pathology. Philadelphia: W.B. Saunders; 1968.
29
Beauchemin N, Draber P, Dveksler G, et al. Redefined nomenclature for members of the carcinoembryonic antigen family.
Exp Cell Res
1999
;
252
:
243
–9.
30
Hammarstrom S, Olsen A, Teglund S, Baranov V. The nature and expression of the human CEA family. In: Stanners CP, editors. Cell adhesion and communications mediated by the CEA family basic and clinical perspectives. Amsterdam: Harwood Academic Publishers; 1998. p. 1–30.
31
Gold P, Freedman SO. Specific carcinoembryonic antigens of the human digestive system.
J Exp Med
1965
;
122
:
467
–81.
32
Gold P, Goldenberg NA. The carcinoembryonic antigen (CEA): past present, and future.
McGill J Med
1997
;
3
:
46
–66.
33
Goldenberg DM, Sharkey RM, Primus F. Carcinoembryonic antigen in histopathology: immunoperoxidase staining of conventional tissue sections.
J Natl Cancer Inst
1976
;
57
:
11
–22.
34
Hammarstrom S. The carcinoembryonic antigen (CEA) family: structures, suggested functions and expression in normal and malignant tissues.
Semin Cancer Biol
1999
;
9
:
67
–81.
35
Shively JE, Beatty JD. CEA-related antigens: molecular biology and clinical significance.
Crit Rev Oncol Hemat
1985
;
2
:
355
–99.
36
Thompson JA, Grunert F, Zimmermann W. Carcinoembryonic antigen gene family: molecular biology and clinical perspectives.
J Clin Lab Anal
1991
;
5
:
344
–66.
37
Goldenberg DM, Deland FH, Kim E, et al. Use of radiolabeled antibodies to carcinoembryonic antigen for the detection and localization of diverse cancers by external photoscanning.
N Eng J Med
1978
;
298
:
1384
–8.
38
Lane DM, Eagle KF, Begent RH, et al. Radioimmunotherapy of metastatic colorectal tumours with iodine-131-labelled antibody to carcinoembryonic antigen: phase I/II study with comparative biodistribution of intact and F(ab′)2 antibodies.
Br J Cancer
1994
;
70
:
521
–5.
39
Murray JL, Macey DJ, Kasi LP, et al. Phase II radioimmunotherapy trial with 131I-CC49 in colorectal cancer.
Cancer
1994
;
73
:
1057
–66.
40
Juweid ME, Sharkey RM, Behr T, et al. Radioimmunotherapy of patients with small-volume tumors using iodine-131-labeled anti-CEA monoclonal antibody NP-4 F(ab′)2.
J Nucl Med
1996
;
37
:
1504
–10.
41
Behr TM, Sharkey RM, Juweid ME, et al. Phase I/II clinical radioimmunotherapy with an iodine-131-labeled anti-carcinoembryonic antigen murine monoclonal antibody IgG.
J Nucl Med
1997
;
38
:
858
–70.
42
Tempero M, Leichner P, Dalrymple G, et al. High-dose therapy with iodine-131-labeled monoclonal antibody CC49 in patients with gastrointestinal cancers: a phase I trial.
J Clin Oncol
1997
;
15
:
1518
–28.
43
Ychou M, Pelegrin A, Faurous P, et al. Phase-I/II radio-immunotherapy study with iodine-131-labeled anti-CEA monoclonal antibody F6 F(ab′)2 in patients with non-resectable liver metastases from colorectal cancer.
Int J Cancer
1998
;
75
:
615
–9.
44
Tempero M, Leichner P, Baranowska-Kortylewicz J, et al. High-dose therapy with 90yttrium-labeled monoclonal antibody CC49: a phase I trial.
Clin Cancer Res
2000
;
6
:
3095
–102.
45
Wong JYC, Chu DZ, Yamauchi DM, et al. A phase I radioimmunotherapy trial evaluating 90yttrium-labeled anti-carcinoembryonic antigen (CEA) chimeric T84.66 in patients with metastatic CEA-producing malignancies.
Clin Cancer Res
2000
;
6
:
3855
–63.
46
Behr TM, Salib AL, Liersch T, et al. Radioimmunotherapy of small volume disease of colorectal cancer metastatic to the liver: preclinical evaluation in comparison to standard chemotherapy and initial results of a Phase I clinical study.
Clin Cancer Res
1999
;
5
:
3232
-42s.
47
Behr TM, Liersch T, Greiner-Bechert L, et al. Radioimmunotherapy of small-volume disease of metastatic colorectal cancer.
Cancer
2002
;
94
:
1373
–81.
48
Zalutsky MR, Xu FJ, Yu Y, et al. Radioiodinated antibody targeting of the HER-2/neu oncoprotein: effects of labeling method on cellular processing and tissue distribution.
Nucl Med Biol
1999
;
26
:
781
–90.
49
Chang CH, Sharkey RM, Rossi EA, et al. Molecular advances in pretargeting radioimunotherapy with bispecific antibodies.
Mol Cancer Ther
2002
;
1
:
553
–63.
50
Wong JYC, Chu DZ, Williams LE, et al. Pilot trial evaluating an 123I-labeled 80-kilodalton engineered anticarcinoembryonic antigen antibody fragment (cT84.66 minibody) in patients with colorectal cancer.
Clin Cancer Res
2004
;
10
:
5014
–21.
51
Adams GP, Shaller CC, Dadachova E, et al. A single treatment of yttrium-90-labeled CHX-A-C6.5 diabody inhibits the growth of established human tumor xenografts in immunodeficient mice.
Cancer Res
2004
;
64
:
6200
–6.
52
Buchegger F, Roth A, Allal A, et al. Radioimmunotherapy of colorectal cancer liver metastases: combination with radiotherapy.
Ann NY Acad Sci
2000
;
910
:
263
–9.
53
Stein R, Chen S, Reed L, Richel H, Goldenberg DM. Combining radioimmunotherapy and chemotherapy for treatment of medullary thyroid cancer; effectiveness of dacarbazine.
Cancer
2002
;
94
:
51
–61.
54
DeNardo SJ, Kukis DL, Kroger LA, et al. Synergy of Taxol and radioimmunotherapy with yttrium-90-labeled chimeric L6 antibody: efficacy and toxicity in breast cancer xenografts.
Proc Natl Acad Sci U S A
1997
;
94
:
4000
–4.
55
Franano F, Edwards W, Welch M, Duncan J. Metabolism of receptor targeted 111In-DTPA-glycoproteins: identification of 111In-DTPA-ε-lysine as the primary metabolic and excretory product.
Nucl Med Biol
1994
;
8
:
1023
–34.