A preclinical ¹⁸⁸Re tumor therapeutic investigation using MORF/cMORF pretargeting and an antiTAG-72 antibody CC49

Abstract

The utility of MORF/cMORF pretargeting for the radiotherapy of cancer requires further validation in tumored mice before clinical trials. We now report on a therapeutic study in mice pretargeted with MORF-CC49 (an anti-TAG-72 antibody CC49 conjugated with MORF, a phosphorodiamidate morpholino oligomer) and then targeted by 188Re-cMORF (a 188Re labeled complementary MORF). Before the dose-escalating therapeutic study, a pretargeting study in LS174T tumored mice was performed at tracer levels. By both necropsy and imaging, the tracer study showed that the whole body radioactivity was largely restricted to tumor in the mice pretargeted 48 h earlier with MORF-CC49 and the tumor radioactivity was retained over 90 h. After decay correction, a best-fit to the biodistribution provided the areas under the radioactivity curves (AUCs) used for the radiation dose estimates. The tumor to normal organ AUC ratios in all cases were greater than unity and ranged from 3 (kidneys) to 48 (muscle). Tumor growth was inhibited in the therapy study. At the highest 188Re dose of 1.40 mCi, a complete but temporary tumor remission was evident in 3 out of the 5 animals. Histological examination of tissues from these animals showed no evidence of cytotoxicity to normal tissues but obvious radiation damage to tumor. In conclusion, effective radiotherapy was achieved in a mouse model by MORF/cMORF pretargeting using 188Re as the therapeutic radionuclide and CC49 as the pretargeting antibody.

Introduction

As an alternative to conventional radiotherapy using radiolabeled antitumor antibodies, radiotherapy by pretargeting can reduce the radiation burden to normal organs while maximizing toxicity to tumor by separating the antibody from its radiolabel.1–5 The MORF/cMORF pretargeting approach uses two complementary phosphorodiamidate morpholino oligomers as the recognition pair. A number of imaging studies using technetium-99m as the radiolabel6–9 have demonstrated sufficiently rapid tumor accumulations and normal tissue clearance, suggesting that, if a therapeutic radionuclide is used, effective radiotherapy may be possible. This laboratory has already reported on a therapeutic response using a rhenium-188 (¹⁸⁸Re) labeled cMORF and a MORF conjugated antiCEA antibody MN14.10 Recently, again using ^99mTc as the label, another antibody CC49 targeting TAG-72 has been shown to be a suitable alternative antibody for this therapeutic application.11,12

The attractive properties of ¹⁸⁸Re for radiotherapy have been previously appreciated and the ¹⁸⁸Re has been used in tumored mice either directly labeled to antibodies and peptides13–20 or by pretargeting.21–23 Thus far, few instances of size reduction have been reported.10,13–20 We now report that a temporary tumor remission was achieved using the anti TAG-72 antibody CC49 and the MORF/cMORF pretargeting strategy.

Results

Pharmacokinetics and radiation dose estimates.

The ¹⁸⁸Re accumulations in tumor and eight normal tissues obtained in the tracer study are listed in Table 1 and are plotted against time in Figure 1 with the solid lines showing the non-linear best fits. When presented in %ID/g, the tumor accumulation (Fig. 1, part I) shows a steady decrease. However, this decrease is due to tumor growth rather than loss of label, as is evident from the virtually unchanged tumor accumulation when presented in % ID/organ (Fig. 1, part J).

Since the radioactivity decreased rapidly in blood while the accumulation in tumor remained fairly constant, the tumor to blood (T/B) ratio increased rapidly as shown in Figure 2A. The T/B ratio reached 5 immediately and increased steadily to 20 over 90 h. The tumor to normal tissue (T/NT) ratios increased also fairly rapidly for most organs except for liver and spleen (data not presented). Fortunately, the accumulations in these two organs were minimal.

After decay correction, the AUCs for tumor and organs of interest were calculated from the best fits to the biodistribution data. As shown in Figure 2B, the AUC ratios of tumor to normal tissues were always greater than unity, ranging from 3 (kidneys) to 48 (muscle). The absorbed radiation doses calculated from these AUC values are listed in Table 2 in rads per µCi of ¹⁸⁸Re. It should be noted that since tumor accumulations are strongly related to tumor size, the AUC ratios and the radiation dose ratios will also be related to tumor size. In addition to any differences in initial tumor sizes, the tumor size reduction resulting from a therapeutic effect is another factor adding to the uncertainty inherent in absorbed radiation dose estimates.

Tumor imaging.

Two animals were imaged on a NanoSPECT/CT small animal camera (Bioscan, Washington DC) at 3.5 and 10 h post ¹⁸⁸Re administration and the acquisitions were reconstructed using the InvivoScope 1.35 program (Bioscan). The anterior, left lateral and posterior projections of the SPECT images at 3.5 and 10 h of one animal are shown in Figure 3. As shown, the ¹⁸⁸Re was largely restricted to tumor and there was no important change in tumor accumulation over the imaging period. Similar results were obtained for the other animal. As shown in Table 1, the biodistribution results by necropsy of the two imaged animals are in general agreement with those of other animals sacrificed at similar time points.

Dose escalation study.

The results in Table 3 indicated that administering 30 µg of the MORF-CC49 antibody had no detectable influence on tumor size. Analysis of variance for repeated measures (RMANOVA) showed no significant difference in tumor growth between the animals receiving MORF-CC49 alone and the untreated animals (p = 0.921). We had earlier shown that the ¹⁸⁸Re-cMORF has no effect on tumor growth.10 The subsequent dose escalation study was therefore performed without either of these controls.

Prior to radioactivity administration, the average tumor weights were generally similar among the five groups (estimated at 0.26 ± 0.15, 0.46 ± 0.16, 0.35 ± 0.09, 0.46 ± 0.14 and 0.26 ± 0.14 g respectively for n = 0–4). Figure 4A and B present the average increases in tumor weight at day 5 and day 11 respectively post radioactivity administration. Indicated by the increase in tumor weight at day 5, the tumor growth was clearly slower in all animals receiving radioactivity but more evident in animals receiving 0.7 mCi or greater (tumors were barely palpable in three of the five mice in the highest dose group). Tumor growth in animals receiving either no radioactivity or the lowest dose was sufficiently rapid such that two untreated animals had to be euthanized at day 5 as the tumors had reached their size limit. The remaining untreated animals and most of the animals receiving the lowest dose were sacrificed prior to day 11 for the same reason. At day 11, the tumors in animals of the three remaining groups increased in size but at a rate inversely correlated to the dose (0.70 mCi group vs. 1.4 mCi group, p < 0.05). In Figure 4C, the estimated average size of tumor over time for untreated animals and animals receiving the highest dose are presented. The figure shows a definite dose-response relationship for the highest dose group. Tumor growth inhibition for this group extended to at least day 20.

Histopathology.

Figure 5 presents representative H&E stained sections obtained from untreated mice and mice receiving the highest ¹⁸⁸Re dose. The percentage of live cells in tumor was reduced to about 30% in the treated mice from about 90% in untreated mice. Furthermore, while necrosis in the untreated mice was mainly localized to tumor cells remote from blood vessels, suggestive of ischemic necrosis resulting from insufficient blood supply in a rapidly growing tumor, the necrosis in the treated mice were in addition localized to cells in close proximity to the blood vessels, suggestive of ¹⁸⁸Re toxicity. In the femur of treated mice, a slight decrease in the overall bone marrow cellularity to about 70% was evident compared to about 95% in untreated mice. However, there was no evidence of toxicity in either myeloid, erytroid or megakaryocytic cell lines in the bone marrow. Similarly, no morphologic abnormality was evident in the white and red pulp of the spleen. The histopathologic examination of tissues with the highest radioactivity levels (kidney, liver and lung) in the treated mice showed no significant morphologic alterations.

Discussion

Given the wide variety of tumor-associated antigens, clinical relevance requires that a variety of antibodies be available for pretargeting. For this reason, the antiTAG-72 CC49 antibody was earlier considered for imaging as an alternative to the antiCEA MN14 antibody for MORF/cMORF pretargeting.12 By both imaging and necropsy, the CC49 antibody was considered to be equally useful to the MN14 antibody in the LS174T tumor model. These previous results also confirmed our earlier prediction that the maximum percent tumor accumulation of the radiolabeled cMORF is independent of the pretargeting antitumor antibody,11 despite the differences in pharmacokinetic properties.

Prior to a therapy study with the CC49 antibody in this investigation, it was necessary to ensure that the ¹⁸⁸Re label on the cMORF was retained in the pretargeted tumor for at least several half lives of ¹⁸⁸Re. As shown in Figure 1 part J, there was no detectable release from tumor over 90 h, a period sufficiently long to provide valid AUCs for this radionuclide. In addition, all normal tissues except for liver and spleen showed an extensive decrease in radioactivity levels over this period, but the radio-activity levels in these two organs were low throughout. Thus except for kidney, the T/NT AUC ratios in all normal tissues considered were all over 6. Even in kidneys where a lower AUC ratio was expected because of its role in clearing the effector, the AUC ratio was 3.

The results obtained in the pharmacokinetic study also provided a set of data usable to estimate the absorbed radiation dose to tumor and normal tissues. As earlier,10 a “self-absorbed” model was again used and the results are presented in Table 2. A comparison of the results in this study with those from the previous study using MN14,10 shows significantly lower estimated radiation doses in all tissues apart from tumor. However, we earlier confirmed the observations of others29–31 that tumor accumulation often decreases with increasing tumor size.32 Thus the approximately two-fold lower radiation dose for tumor is due to the larger size (average 0.96 g) in this CC49 study compared to the earlier MN14 study (average 0.56 g).

Selecting the optimum dosages and timing in pretargeting is difficult, but can be simplified following certain rules.25 Thus, a convenient antibody dosage below that just to saturate the tumor antigen sites was selected, a two day pretargeting interval was selected based on the pharmacokinetics of IgG antibodies,12,33 and the dosage of the radiolabeled cMORF effector was selected to just saturate the MORF-antibody in tumor. The dosage of the labeled cMORF was 2.5 µg in tumored mice pretargeted with 30 µg of MORF-CC49 (gpm = 0.68),12 corresponding to a cMORF/MORF molar ratio of 3.1. If this ratio was kept constant, the percent tumor accumulation of labeled cMORF would be the same.25 For this reason, a cMORF/MORF ratio as close to 3.1 as achievable was selected (2.6 for the pharmacokinetic study and 3.0 for the therapy study).

Compared to the relatively modest therapeutic response in our earlier study,10 the higher doses used in this study provided more striking results. At day 5 post ¹⁸⁸Re injection when the untreated control mice were euthanized, tumors in three mice receiving the highest dose became barely palpable. The number of reports documenting a temporary tumor remission following intravenous administration of ¹⁸⁸Re is quite limited.13–20

The maximum ¹⁸⁸Re dose in this study was confined by the injectate volume that was restricted to below 250 µL and by the age of the ¹⁸⁸Re generator used. As shown in Figure 4, a clear dose response was apparent at doses above 0.35 mCi. It was also gratifying that the results of the histopathology examination showed only minimal damage to normal tissues. In addition, no decrease in average body weight was observed at any dose (data not presented). Taken together, these data suggested radiotherapeutic efficacy coupled with minimal normal tissue toxicity, serving as a rationale for future studies with higher radiation doses.

Materials and Methods

The MORF and cMORF sequences were respectively 5′-TCT TCT ACT TCA CAA CTA-linker-amine (6,060 kDa) and 5′-TAG TTG TGA AGT AGA AGA-linker-amine (6,318 Da) and were purchased from GeneTools (Philomath, OR). The C6-SANH [N-(6′-succinimidyl hexanoyl) 6-hydrazinonicotinamide acetone hydrazone] (403 Da) and C6-SFB [N-(6′-succinimidyl hexanoyl) 4-formylbenzamide] (360 Da) were from Solulink (San Diego, CA). Commercial 9-mL Sephadex G-25 columns were from NeoRx Corp. (Seattle, WA), Sephadex G-100 was from Amersham Biosciences (Uppsala, Sweden) and P-4 Gel (medium) was from Bio-Rad Laboratories (Hercules, CA). The ¹⁸⁸W/¹⁸⁸Re generator was purchased from Oak Ridge National Laboratory (Oak Ridge, TN).

Conjugations and radiolabeling.

The antiTAG-72 antibody CC49 (160 kDa) was prepared by Strategic Biosolutions (Ramona, CA) from the CC49 murine hybridoma cell line (a gift from Dr Jeffrey Schlom, National Cancer Institute, NIH). Conjugation of MORF to the CC49 was performed as described previously.12,24 The MAG₃-cMORF effector was prepared and radiolabeled with ¹⁸⁸Re as previously described.25 A shift assay by size exclusion HPLC, in which the labeled cMORF was analyzed before and after adding excess MORF-CC49, showed a 90% shift of the radioactive peak to that of a higher molecular weight as evidence of effective radiolabeling and preserved hybridization ability.

Pharmacokinetics of ¹⁸⁸Re-cMORF in pretargeted mice.

For any therapeutic nuclide, the areas under the radioactivity curves (AUCs) over about six half-lives are required for an accurate estimates of the radiation doses to tumor and organs. Therefore a ¹⁸⁸Re (T_1/2 = 17 h) tracer study was performed prior to the therapy study. Tumored animals were sacrificed at regular time points over the 90 h period and the results subjected to a non-linear regression analysis to provide more accurate radioactivity-time curves for tumor and organs of interest.

Since the pharmacokinetics of ^99mTc and ¹⁸⁸Re labeled cMORFs are similar,25 optimal pretargeting parameters (the dosage of the MORF-CC49, the pretargeting interval and the dosage of labeled cMORF) obtained from earlier pretargeting studies with ^99mTc-cMORF and MORF-CC49 were again used.12,26 To initiate the study, about 10⁶ LS174T colon cancer cells were inoculated in the left thigh of each of the 16 NIH Swiss nude mice (Taconic Farms, Germantown, NY). On day 12, each mouse received via a tail vein 30 µg MORF-CC49 (MORF groups per molecule, gpm = 0.68) and 48 h later 2.1 µg (182 µCi) of ¹⁸⁸Re-cMORF. The pharmacokinetics of the ¹⁸⁸Re-cMORF in normal organs and tumor of the pretargeted mice was followed by necropsy after cardiac puncture under halothane anesthesia. The excised tissues were prepared and counted as previously.9 To estimate the AUCs, the percent of injected dosage per gram (%ID/g) values were fitted over time using SigmaPlot data processing program (San Jose, CA).

Radiation dose estimates.

The AUCs for tumor and organs of interest were calculated from the %ID/g-time best-fit curves after decay correction. A self-absorbed model was again used in which only beta radiation was considered and 100% of the radiation energy was assumed to be deposited in the source organ.10,21,27 The absorbed radiation doses in Rads were calculated using 2.143* 0.766 (MeV)*AUC (µCi*h/g), where 2.143 is a unit conversion factor and 0.766 MeV is the average energy of the ¹⁸⁸Re beta particles.

Tumor imaging.

To supplement the biodistribution results of the tracer study, two pretargeted animals receiving ¹⁸⁸Re-cMORF were imaged on a NanoSPECT/CT camera (Bioscan, Washington DC) at 3.5 and 10 h post radioactivity administration. To facilitate the imaging, the dosages were increased to 60 µg for MORF-CC49 and 4.8 µg (1.4 mCi) for ¹⁸⁸Re-cMORF with the antibody/effector dosage ratio held constant and equal to that in the tracer study. We earlier showed experimentally and theoretically that the biodistribution of the effector remains unchanged if this ratio (and the timing schedule) is not varied (provided that the tumor was not saturated by the antibody).25,28 After imaging, each animal was sacrificed for biodistribution.

Dose escalation study.

The ¹⁸⁸Re-cMORF effector had earlier been shown to have no observable influence on tumor growth in the LS174T model.10 Prior to the dose escalation study, it was also necessary to determine whether the MORF-CC49 itself had any therapeutic potential. One group of four tumored mice were each administered 30 µg of the MORF-CC49 while another group of four mice were left untreated as control. Tumor sizes were measured over time. Differences were evaluated using analysis of variance for repeated measures (RMANOVA).

After confirming that the MORF-CC49 had no therapeutic effect similar to that of ¹⁸⁸Re-cMORF, a study with five groups of mice (n = 0–4, N = 5) at five doses was initiated 11 days post tumor inoculation. Mice in group n (n = 1–4) each received (n × 15) µg of MORF-CC49 intravenously and 48 h later (n × 1.20) µg and (n × 350) µCi of ¹⁸⁸Re-cMORF. The control group (n = 0) received neither MORF-CC49 nor ¹⁸⁸Re-cMORF. Thus, except for the control group, mice each received 15–60 µg of MORF-CC49 and 1.2–4.8 µg (up to 1.40 mCi) of labeled cMORF. Once again, the ratio of antibody/effector dosage was held constant and equal to that used in the tracer and imaging studies. The tumor size and body weight were monitored and each mouse was sacrificed when the square root of the product of width and thickness of the tumor thigh exceeded 1.50 cm. The tumor weight was estimated from this product by the empirical formula: tumor weight (g) = 0.94* the width*thickness of tumor thigh (cm²) − 0.63.28

Histopathology.

Preliminary information about toxicity to tissues and tumor was collected after euthanization from the mice receiving the highest radiation dose of 1.40 mCi by histopathological examination of liver, lung, kidney, spleen and tumor. The femur was also included because of the radiosensitivity of bone marrow. Except for the femur, all specimens were first fixed in 10% neutral buffered formalin and then embedded in paraffin blocks. The femur was decalcified and fixed overnight in a commercial solution (Surgipath Decalicifier I, Surgipath Medical Ind. Inc., Richmond, IL) before being embedded in a paraffin block. Tissues were then cut into 5 µm thick slices and fixed on glass slides for Hematoxylin and Eosin (H&E) staining. The findings were compared to those from the untreated control mice.

Conclusion

The predominant localization and sustained retention of ¹⁸⁸Re in tumor by MORF/cMORF pretargeting observed in the tracer study and the successful subsequent radiotherapy in mice suggest that further therapeutic studies of MORF/cMORF pretargeting are justified.

Abbreviations

MORF	=	an 18 mer phosphorodiamidate morpholino oligomer
cMORF	=	another oligomer complimentary to the MORF
AUC	=	area under the radioactivity-time curve

Figures and Tables

Figure 1 Biodistributions of the ¹⁸⁸Re-cMORF effector from Table 1 plotted individually in %ID/g (parts A to I) and, in the case of tumor only, also in %ID (part J)

Figure 2 (A) The tumor to blood ratios with time since radioactivity administration. The solid line represents the linear best fit. (B) Histograms showing tumor to normal tissue AUC ratios for listed organs.

Figure 3 Anterior, left lateral and posterior projections for each SPECT/CT acquisition of one pretargeted mouse imaged at 3.5 h (A) and 10 h (B) after injection of ¹⁸⁸Re-cMORF. Except for the intense urine radioactivity at 3.5 h, the radioactivity is essentially restricted to tumor at both time points.

Figure 4 Histograms presenting the increase in tumor weight at day 5 (A) and day 11 (B) after ¹⁸⁸Re administration. The estimated tumor dimensions over time for groups 0 (0 mCi) and 4 (1.40 mCi) are also presented (C). Average, N = 5. Error bars signify one standard deviation.

Figure 5 Representative H&E stained tissue sections of tumor and several normal tissues from untreated mice and mice receiving the highest ¹⁸⁸Re dose.

Table 1 Individual biodistributions in %ID/g and, for stomach and intestines, %ID/organ from 1–90 h post IV injection of ¹⁸⁸Re-cMORF to tumored mice pretargeted 48 h earlier with MORF-CC49

Mouse no	Sacrifice time (h)	Tumor weight (g)	%ID/g									%ID
			Tumor	Blood	Kidneys	Liver	Spleen	Lung	Heart	Muscle	Salivary	Stomach	Sm. int.	Lg. int.
1	1.00	0.76	5.91	1.66	4.52	0.70	0.38	0.98	0.40	0.26	0.50	0.22	3.30	0.18
2	6.05	0.64	4.54	0.83	2.68	0.48	0.34	0.41	0.20	0.11	0.25	0.30	0.87	2.91
3 imaged	10.87	0.76	4.20	0.92	1.28	0.59	0.31	0.61	0.31	0.09	0.23	0.20	0.20	0.55
4	12.03	1.04	2.49	0.42	1.36	0.25	0.24	0.20	0.09	0.05	0.17	0.20	0.30	0.79
5	18.00	0.67	4.38	0.56	0.96	0.53	0.31	0.25	0.13	0.07	0.15	0.06	0.18	0.85
6 imaged	21.92	1.46	2.07	0.46	0.82	0.54	0.36	0.27	0.12	0.05	0.15	0.07	0.12	0.28
7	24.00	0.96	3.46	0.42	1.02	0.55	0.25	0.20	0.13	0.07	0.15	0.05	0.12	0.17
8	30.00	0.74	4.29	0.43	0.81	0.75	0.45	0.21	0.10	0.07	0.13	0.08	0.14	0.54
9	36.00	1.16	2.69	0.29	0.65	0.52	0.35	0.18	0.08	0.05	0.12	0.06	0.05	0.14
10	42.00	0.97	3.51	0.27	0.39	0.59	0.27	0.14	0.07	0.05	0.11	0.03	0.07	0.12
11	48.00	0.82	3.58	0.24	0.50	0.60	0.28	0.28	0.08	0.06	0.15	0.04	0.09	0.09
12	54.35	0.85	4.18	0.25	0.71	0.66	0.43	0.64	0.09	0.04	0.15	0.05	0.08	0.14
13	60.00	1.24	2.28	0.18	0.39	0.57	0.34	0.21	0.08	0.04	0.12	0.03	0.05	0.07
14	66.15	0.68	3.83	0.27	0.36	0.86	0.38	0.14	0.07	0.04	0.14	0.04	0.08	0.08
15	72.05	1.16	2.73	0.18	0.58	0.71	0.39	0.22	0.04	0.03	0.12	0.06	0.11	0.16
16	79.03	1.17	2.42	0.14	0.23	0.41	0.25	0.13	0.06	0.03	0.08	0.03	0.05	0.04
17	84.00	1.39	2.16	0.14	0.26	0.57	0.44	0.21	0.07	0.06	0.12	0.03	0.06	0.05
18	90.15	1.15	3.61	0.17	0.29	0.60	0.30	0.13	0.06	0.03	0.09	0.03	0.04	0.08

The tumor weights at the time of necropsy are also listed. Data for the two imaging mice are also included.

Table 2 The AUCs and absorbed radiation doses for tumor and organs

Organs	Tumor	Blood	Kidney	Liver	Spleen	Lung	Heart	Muscle	Salivary gland
AUC(uCi*h/g)	0.96	0.15	0.40	0.15	0.08	0.08	0.04	0.02	0.05
Dose (Rad/µCi)	1.58	0.24	0.65	0.24	0.13	0.13	0.06	0.04	0.08

Table 3 Tumor dimension (width × thickness of the tumor thigh, cm²) of mice receiving only 30 µg of the MORF-CC49 compared to untreated mice

Days after activity injection	1	2	3	4	5
Treated	0.89 ± 0.09	1.01 ± 0.09	1.13 ± 0.14	1.27 ± 0.14	1.51 ± 0.08
Non-treated	0.91 ± 0.08	0.99 ± 0.14	1.08 ± 0.22	1.37 ± 0.26	1.51 ± 0.33

Average ± sd, N = 4. The results show no influence on tumor growth by the antibody alone.

Acknowledgements

The authors are grateful to Dr. Jeffery Schlom (Laboratory of Tumor Immunology and Biology, Center for Cancer Research, NCI, NIH, Bethesda, MD) for providing the CC49 hybridoma. Financial support was from the National Institutes of Health (CA94994 and CA107360).