Combined Radioimmunotherapy and ... - ACS Publications

We performed studies on nude mice bearing xenografts from MCF7, a cell line that has low Her2 and CEA expression, to more accurately reflect the more ...
1 downloads 0 Views 532KB Size
Download PDF
Bioconjugate Chem. 2005, 16, 1117−1125

1117

Combined Radioimmunotherapy and Chemotherapy of Breast Tumors with Y-90-Labeled Anti-Her2 and Anti-CEA Antibodies with Taxol Desiree M. Crow,† Lawrence Williams,‡ David Colcher,† Jeffrey Y. C. Wong,† Andrew Raubitschek,† and John E. Shively*,§ Department of Radioimmunotherapy, Division of Radiation Oncology, City of Hope National Medical Center, Duarte, California 91010, Division of Radiology, City of Hope National Medical Center, Duarte, California 91010, and Division of Immunology, Beckman Research Institute of the City of Hope, Duarte, California 91010. Received March 23, 2005; Revised Manuscript Received July 20, 2005

Because breast cancer cells often express either Her2/neu or carcinoembryonic antigen (CEA) or both, these tumor markers are good targets for radioimmunotherapy using Y-90-labeled antibodies. We performed studies on nude mice bearing xenografts from MCF7, a cell line that has low Her2 and CEA expression, to more accurately reflect the more usual situation in breast cancer. Although uptake of In-111 anti-CEA into tumors was lower than that for In-111-labeled anti-Her2, radioimmunotherapy (RIT) with Y-90 anti-CEA was equivalent to that of Y-90 anti-Her2. When either Y-90 antibody was combined with a split-dose treatment with Taxol, the antitumor effect was greater than with either agent alone. When Y-90 anti-CEA was combined with a single dose of Taxol, the results were equivalent to the split-dose regimen. RIT plus cold Herceptin had no additional effects on tumor size reduction over RIT alone. When animals were first treated with Y-90 anti-Her2 and imaged 1-2 weeks later with In-111 anti-CEA or anti-Her2, tumor uptake was higher for anti-CEA and improved over tumor uptake with no prior RIT. These results suggest that a split dose of RIT with anti-Her2 antibody followed by anti-CEA antibody would be more effective than a single dose of either. This prediction was partially confirmed in a controlled study comparing single- vs split-dose anti-Her2 RIT followed by either anti-Her2 or anti-CEA RIT. These studies suggest that combined RIT and Taxol therapy are suitable in breast cancers expressing either low amounts of Her2 or CEA, thus expanding the number of eligible patients for combined therapies. They further suggest that split-dose RIT using different combinations of Y-90-labeled antibodies is effective in antitumor therapy.

INTRODUCTION

Despite the use of combination chemotherapy, metastatic breast cancer is invariably a fatal disease (1). The advent of monoclonal antibodies as therapeutic approaches has made it possible to directly target cell surface receptors that are required for tumor cell growth. One of the most studied antibodies in this approach is Herceptin (trastuzumab) that targets Her2, a member of the EGFR family (2, 3). Unfortunately, anti-Her2 therapy is limited to tumors that overexpress Her2, limiting this therapy to a small subset (15-20%) of breast cancer patients (4). A further problem is that Herceptin as a single agent is cytostatic (5) and must be used in combination with chemotherapeutic agents to achieve maximum therapeutic benefit (6). Among the combination therapies, it has been used with doxorubicin/cyclophosphamide (AC), Taxol (7), and more recently, docetaxel plus platinum salts (8). Although response rates of up to 79% have been reported (8), the median time to progression of 9.9-15.5 months suggests that further improvement is warranted. In addition, prolonged use of Herceptin with some drugs, especially doxorubicin, * To whom correspondence should be addressed. † Department of Radioimmunotherapy, Division of Radiation Oncology, City of Hope National Medical Center. ‡ Division of Radiology, City of Hope National Medical Center. § Division of Immunology, Beckman Research Institute of the City of Hope.

have led to cardiac toxicity in as many as 27% of treated patients (9). In view of these problems, we have been exploring the use of Y-90 radioimmunotherapy (RIT) as an alternative to cold antibody therapy. Y-90 is a high-energy beta emitter with a half-life of 2.7 days and a maximum path length of 11 mm in tissue. Antibodies conjugated to chelates such as DTPA (10) and DOTA (11) readily bind Y-90 just prior to injection, thus making this approach attractive in the clinic. A major advantage of RIT is that the once the antibody is targeted to the tumor, it exerts a direct cytotoxic effect. Disadvantages of this approach are the logistical production of radiolabeled antibodies with associated waste disposal and the time required for effective targeting to the tumor because while the radiolabeled antibody is in circulation there is accumulated bone marrow toxicity (12, 13). While many researchers have tried to decrease blood circulation time by engineering small-sized antibody fragments, the net effect is usually faster renal excretion reducing tumor uptake accordingly (14). As a result most RIT approaches have relied on whole antibodies engineered as chimeric (1517) or humanized (18-21) versions to maximize tumor uptake while reducing immunogenicity. The excellent targeting properties of radiolabeled antiHer2 antibodies has long been appreciated and shown to be effective in reducing tumor size in animal models (11, 22). However, these studies, like therapy with cold Herceptin, have been focused on tumors that overexpress

10.1021/bc0500948 CCC: $30.25 © 2005 American Chemical Society Published on Web 08/26/2005

1118 Bioconjugate Chem., Vol. 16, No. 5, 2005

Her2. Similarly, we have focused our efforts on the use of radiolabeled anti-carcinoembryonic antigen (anti-CEA) antibodies on tumors that overexpress CEA (23). In clinical trials in which Y-90 anti-CEA Mab cT8.66 was administered as a single agent to patients with colorectal cancer, only modest effects on tumors were seen, while in CEA-positive breast cancer patients the effects were more impressive (16, 24). The conclusion from these studies is that combination therapies of RIT plus chemotherapy will be required to produce more lasting results. CEA is an attractive cell surface target in that it is expressed in about 50% of breast cancers (25) and is a reliable marker for breast cancer recurrence (26-28). Since most breast cancers express Her2 at some level, the combined expression of Her2 and CEA in breast cancers is about 50%. In this study, we now ask the question, can RIT, especially RIT combined with chemotherapy, be effective against breast tumors that have low to moderate amounts of target antigens such as Her2 and CEA? To answer this question in a preclinical setting, we have grown dual Her2 and CEA positive MCF7 cells as xenografts in nude mice and treated the animals with either Y-90-labeled Herceptin or cT84.66 in the presence or absence of Taxol. This cell line has low amounts of cell surface Her2 (29) that fall below the amount required for successful cold Herceptin therapy and has moderate to low amounts of cell surface CEA (30). The results suggest that in combination with Taxol, the amount of RIT to produce sustained tumor reductions can be reduced 50% or more over that required producing equivalent results as a single agent. Quite surprising, while In-111-labeled cT84.66 tumor targeting was low (tumor-to-blood ratio of 1.0), RIT with Y-90-labeled cT84.66 was as effective as Y-90-labeled Herceptin in this animal model. The results are encouraging in that the use of RIT with these antibodies may be effective even in tumors with low amounts of cell surface antigen, thus opening up this therapy to a larger group of patients compared to the current scenario with cold Herceptin. A second intriguing result from this study was that animals treated with Y-90 Herceptin exhibited an increased uptake of In-111 cT84.66 compared to In-111 Herceptin, suggesting that sequential RIT first with Y-90 Herceptin and then with Y-90 cT84.66 may be possible in tumors that express both Her2 and CEA. This approach would be especially attractive in treating tumors that downregulate a target antigen after initial therapy. MATERIALS AND METHODS

Materials. MCF-7 cells were obtained from ATCC and maintained in sterile growth media consisted of Eagle’s Minimal Essential media 1× (EMEM) (Cellgro, Herndon, VA) supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Omega Scientific, Tarzana, CA), and 1% L-glutamine, 10 mM sodium pyruvate, 0.1 mM nonessential AA. Media was sterile filtered through a 0.2 µm vacuum filter. Herceptin was obtained from the City of Hope Pharmacy and dialyzed vs PBS prior to conjugation. 1B4M chelate was a generous gift from Dr. Martin Brechbiel (NCI) and was conjugated to Herceptin according to established methods (10). DOTA-conjugated chimeric anti-CEA antibody cT84.66 has been previously described by us (31, 32). The number of chelates per antibody which were determined using In-111 isotope dilution in InCl3 essentially according to Lewis et al. (32) was 3 and 2.4 for 1B4M-Herceptin and DOTA-cT84.66, respectively. Chelate conjugated antibodies were radiolabled with In-111 (Amersham, 1-2 mCi per 0.1 mg of protein) as previously described (32).

Crow et al.

MCF-7 Xenograft Model. Female, athymic nu/nu mice (Charles River), 10-12 weeks old, were exposed to 200 RADS external beam Co-60 irradiation 3 days prior to MCF-7 implantation and given sulfatrim antibiotic water for two weeks. Two days prior to MCF-7 implantation mice were administered 0.8 mg of Delestrogen IM injection in the thigh. MCF-7 cells (5 × 106) in EMEM and Matrigel were implanted into each mouse subcutaneous in the flank. Tumors were established within 1014 days postinjection. Before therapy studies began, tumor-bearing animals were separated into treatment groups based on tumor size (100-200 mg), so all groups average tumor volumes were similar at the onset of the therapy studies. Tumor volumes were estimated twice weekly using the formula length × width2/2. Relative tumor volume was calculated as the ratio of the current measurement to the initial value. Weights of the animals, measured in grams, were recorded once a week; no animal had more than a 2% weight loss during the course of the experiment. Radioimaging and Biodistribution Studies. Athymic mice bearing MCF-7 xenografts were injected with 200 µL (2.8 µg of 0.9 µCi/µg) of In-111-labeled 1B4MHerceptin via tail vein. Five mice per time point were sacrificed at 0, 2, 24, 48, 72, and 96 h postinjection and biodistributions performed (blood, liver, spleen, kidneys, lungs, tumor, and carcass). Alternatively, mice were tail injected with 200 µL (3.2 µg of 1.25 µCi/µg) of In-111DOTA-cT84.66. Results were calculated as percent injected dose per gram (%ID/g) vs time. For radioimaging, mice were injected with 15 µCi In-111-1B4M-Herceptin (6 µg) and imaged using a BIOSPACE γ imager instrument with settings of 154-188 keV and 220-270 keV. RIT and Combined Therapy Studies. RIT and Cold Herceptin. Four Y-90-DOTA cT84.66 RIT treatment groups (eight Athymic mice per group bearing MCF-7 xenografts) were as follows: 50 µCi RIT, 100 µCi RIT, 50 µCi RIT with cold Herceptin, and 100 µCi RIT with cold Herceptin. A fifth group of mice received only cold Herceptin, which was used as a non-RIT control. The three groups that received the cold Herceptin were administered two IP injection doses (200 µg each), 9 days apart. The sixth group of mice received a single IP injection of sterile saline (no treatment control). The six groups of mice were monitored daily for 5 weeks. RIT and Single Dose Taxol. Four Y-90-DOTA-cT84.66 RIT treatment groups (eight athymic mice per group bearing MCF-7 xenografts) were as follows: 20 µCi RIT, 40 µCi RIT, 20 µCi RIT with Taxol, and 40 µCi RIT with Taxol. The fifth group received Taxol only. The three groups that received Taxol were administered a single bolus IP of 600 µg Taxol admininistered 24 h post RIT. The sixth group received a single IP injection of sterile saline (no treatment control). The six groups of mice were monitored daily for 5 weeks. RIT and Two Doses of Taxol. Four Y-90-DOTA-cT84.66 RIT treatment groups (eight athymic mice per group bearing MCF-7 xenografts) were as follows: 40 µCi RIT, 80 µCi RIT, 40 µCi RIT with Taxol, or 80 µCi RIT with Taxol. The fifth group received Taxol only. The three groups that received Taxol were administered two IP injections of 300 µg Taxol 48 and 72 h after RIT. The sixth group received a single IP injection of sterile saline (no treatment control). The six groups of mice were monitored daily for 5 weeks. An identical study was performed with Y-90-1B4M-Herceptin. RIT Followed by In-111 Antibody Biodistribution. Eight treatment groups (8 athymic nude mice bearing MCF-7 xenografts per group) were as follows: four

Bioconjugate Chem., Vol. 16, No. 5, 2005 1119

Combined RIT and Taxol in a Breast Cancer Model

groups received 60 µCi Y-90-1B4M-Herceptin, two of which received 200 µL either In-111-labeled cT84.66 (1.7 µg of 2.4 µCi/µg) or Herceptin (1.7 µg of 2.4 µCi/µg) 7 days after initial therapy and two groups of which received either In-111-labeled DOTA-cT84.66 (2.5 µg of 2.4 µCi/ µg) or In-111 1B4M-Herceptin (2 µg of 2.35 µCi/µg) 14 days after initial therapy. The other four groups were as follows: no RIT, two groups on day 7 received either In111-labeled DOTA-cT84.66 or In-111-labeled 1B4M-Herceptin. The final two groups received either In-111labeled DOTA-cT84.66 or 1B4M-Herceptin on day 14. The In-111-labeled anibody was injected via tail vein, and animals were sacrificed 72 h postinjection and biodistributions were performed (blood, liver, spleen, kidneys, lungs, tumor, and carcass). RIT Followed by Second-Dose RIT. Six treatment groups (10 athymic nude mice bearing MCF-7 xenografts per group) were as follows: three groups received 60 µCi (8 µg) of Y-90-1B4M-Herceptin of which one group received no second therapy, one received a second therapy with 40 µCi (5.6 µg) of Y-90-1B4M-Herceptin and the final group received a second therapy with 40 µCi (6 µg) of Y-90-DOTA-cT84.66 administered 14 days after the first RIT. Two other groups received no initial RIT but on day 14 received either 40 µCi (5.6 µg) of Y-90-1B4M-Herceptin or 40 µCi (6 µg) of Y-90-DOTA-cT84.66. The final group received only saline via tail vein and served as the notreatment therapy control. All RIT was administered via tail vein in 200 µL total volume injected. Data Analysis. For biodistribution studies, the results were reported as %ID/g of tissue vs time. Comparisons were made by t-test and p values were calculated. For RIT or combined RIT and Taxol studies, average relative tumor volumes vs time were reported and the curves were analyzed by ANOVA (two-factor without replication). RESULTS

Biodistributions of In-111-Labeled Anti-CEA and Anti-Her2 Antibodies in MCF-7 Xenografts. CEA and Her2 are expressed in breast cancer in variable amounts (2, 25). The overall positivity for CEA is about 50%, but the expression level is usually much less than in colon cancer. Her2 is overexpressed in 15-20% of breast cancers but is moderately expressed in higher numbers, perhaps up to 30-40% (33). However, Herceptin therapy is only appropriate in the tumors that overexpress Her2. We were therefore interested in testing the possibility of using RIT or combined RIT and Taxol therapy to treat tumors that had low to moderate amounts of either antigen. For these reasons, we chose MCF-7 cells which express both Her2 and CEA in low to moderate amounts. To determine the potential for RIT, we first compared the uptake of In-111-labeled anti-CEA (cT84.66) and antiHer2 (Herceptin) in nude mice bearing MCF-7 xenografts. As shown in Table 1 tumor uptake reached a maximum in 48-72 h with values of 11.6 ((2.3) and 25.0 ((5.6) %ID/g for anti-CEA and anti-Her2, respectively. The tumor-to-blood ratios at maximum uptake were 1.0 and 2.0 for anti-CEA and anti-Her2, respectively. The low tumor-to-blood ratio for the anti-CEA study was especially worrisome, since bone marrow toxicity is the major side effect in high dose RIT. In general, tumor-toblood ratios of at least 2:1 are preferred to minimize this toxicity. Thus, from the results of the biodistribution data, we predicted that anti-Her2 would be the superior antibody for RIT in this tumor model. Nonetheless, we proceeded to compare both antibodies in RIT since the

Table 1. Biodistributions for Nude Mice Bearing MCF7 Xenografts with In-111 Anti-Her2 or In-111 Anti-CEA Antibodiesa In-111 Anti-Her2 Antibodies 0h blood liver spleen kidney lung tumor carcass

blood liver spleen kidney lung tumor carcass

2h

31.84 (1.48 6.98 (1.63 6.31 (1.97 4.95 (0.52 8.31 (1.00 0.51 (0.11 2.11 (0.32

24 h

48 h

25.15 12.26 10.81 (2.40 (1.23 (0.95 5.85 4.60 4.48 (0.87 (1.49 (1.56 5.50 3.62 3.20 (1.33 (0.86 (0.32 5.65 3.36 3.60 (0.83 (0.40 (0.69 8.41 5.05 5.03 (1.04 (0.56 (0.92 2.25 16.41 20.47 (0.24 (5.40 (2.67 3.02 3.32 3.43 (0.09 (0.31 (0.23 In-111 Anti-CEA Antibodies

72 h

96 h

11.42 (1.24 3.96 (0.18 4.71 (1.18 3.04 (0.29 4.66 (0.43 25.01 (5.64 3.37 (0.21

8.25 (2.35 4.83 (1.08 3.22 (1.34 3.16 (0.39 3.81 (0.44 23.08 (4.44 3.19 (0.10

0h

24 h

48 h

72 h

33.15 (2.30 6.39 (0.50 5.09 (1.41 6.42 (0.67 9.59 (1.60 0.75 (0.11 1.76 (0.15

10.49 (2.46 3.99 (1.19 2.65 (1.21 3.11 (0.40 4.06 (0.98 7.02 (1.31 3.12 (0.35

11.24 (1.64 3.78 (0.59 2.71 (0.46 3.39 (0.83 4.05 (1.41 11.58 (2.30 3.45 (0.43

7.88 (3.36 3.96 (1.42 4.42 (1.84 3.49 (1.12 3.94 (1.18 10.78 (2.27 3.36 (0.11

a Five animals per each time point were analyzed for the indicated tissues. Results are reported as %ID/g with the standard deviation given below.

tumor model provided a unique opportunity to see if the biodistribution results would correctly predict the RIT results. RIT and Taxol Therapy with Y-90-Labeled AntiCEA and Anti-Her2 Antibodies in MCF-7 Xenografts. Since previous studies have shown that only high-dose RIT with or without autologous blood or marrow transplant (BMT) gave superior responses for single agent therapy (34), we considered from the outset to perform combined RIT plus chemotherapy in the tumor model system. Taxol was chosen as the agent of choice because of its prevalent use in breast cancer therapy especially when used with Herceptin cold antibody therapy. On the basis of the results of DeNardo and co-workers in which RIT with various taxol combinations were explored (35), we chose to administer single-dose RIT followed by two doses of taxol IP (300 µg). Since previous experience established that 100-120 µCi/mouse of Y-90-cT84.66 therapy was sufficient to cause >50% tumor growth inhibition (11), we explored the use of lower-dose RIT (40 and 80 µCi) so that the effect of Taxol (if any) would not be obscured by the RIT. As shown in Figure 1A, the relative tumor volume of the control animals increased by 2-2.5-fold over 35 days, while Taxol-only treated animals showed a brief period of tumor growth inhibition but ultimately reached the same size as control animals. Animals treated with 40 µCi of Y-90-cT84.66 exhibited a similar reduction in tumor growth compared to Taxol alone (p ) 0.29), but by the end of 35 days tumor regrowth was apparent. When the RIT dose was increased to 80 µCi/animal a substantial tumor growth inhibition was seen, even at the end of 35 days (p )

1120 Bioconjugate Chem., Vol. 16, No. 5, 2005

Crow et al.

Figure 2. Radioimmunotherapy of established MCF-7 tumors in nude mice with Y-90-labeled anti-CEA antibody in combination with a single dose of Taxol. Groups of eight mice were injected IV with Y-90-labeled anti-CEA antibody. Taxol (600 µg) was administered IP 48 h post RIT. The tumor volume for each mouse at the time indicated was normalized to the tumor volume at the start of treatment (average relative tumor volume) and was plotted versus days after the start of treatment.

Figure 1. Radioimmunotherapy of established MCF-7 tumors in nude mice with Y-90-labeled anti-Her2 or anti-CEA antibodies in combination with a split dose dose of Taxol. Groups of eight mice were injected IV with either (A) Y-90-labeled antiHer2 antibody or (B) Y-90-labeled anti-CEA antibody. Taxol (300 µg) was administered IP 48 and 72 h post RIT. The tumor volume for each mouse at the time indicated was normalized to the tumor volume at the start of treatment (average relative tumor volume) and was plotted versus days after the start of treatment.

0.0006). The addition of Taxol to the 40 µCi RIT and 80 µCi RIT further improved the tumor effect (p ) 0.003 for both). Essentially similar results were obtained for

Y-90-labeled anti-Her2 RIT at two dose levels with or without Taxol (Figure 1B). At first inspection, the results surprised us because based on the biodistribution results shown in Table 1 we expected superior results with anti-Her2 RIT vs anti-CEA RIT. Since the tumor cell line and animal model are identical, we tentatively conclude that there is an inherent advantage to anti-CEA RIT over anti-Her2 RIT. We therefore decided to explore this phenomenon more thoroughly by performing other RIT studies with Y-90 anti-CEA, one using RIT plus a single dose of Taxol and another using RIT plus cold Herceptin. RIT with Y-90-Labeled Anti-CEA and Taxol or Cold Herceptin in MCF-7 Xenografts. Since DeNardo and co-workers (35) had shown that a single dose of 600 µg of Taxol IP 24 h after RIT was equivalent to a split dose of 300 µg of Taxol at 48 and 72 h post RIT, we chose the single-dose regimen to repeat the study and chose the 40 µCi dose of RIT with Taxol to determine if the single dose was equivalent to the split-dose regimen. The results shown in Figure 2 demonstrate a more prolonged tumor suppression with the single dose of Taxol compared to the split-dose regimen (compare Figure 1 to Figure 2, taxol only, p ) 0.01) while the 40 µCi RIT only demonstrated tumor reduction followed by regrowth, similar to the results shown in Figure 1. Single-dose Taxol plus 40 µCi of RIT resulted in prolonged tumor suppression, greater than that observed for either Taxol or RIT alone (p ) 0.00007 and 0.0003, respectively). In this case, the results were essentially equivalent to the higher-dose RIT combined therapy results shown in Figure 1 (p ) 0.56). Thus, it is likely either single or split doses of Taxol plus single-dose RIT are promising therapies for CEA-positive breast cancers.

Bioconjugate Chem., Vol. 16, No. 5, 2005 1121

Combined RIT and Taxol in a Breast Cancer Model

Table 2. Biodistributions of In-111-Labeled Anti-Her2 or In-111-Labeled Anti-CEA Antibodies in Nude Mice Bearing MCF7 Xenografts 7 or 14 Days after Treatment with Y-90-Labeled Anti-Her2 Antibodya 7 Days In-111 anti-Her2, 72 h blood liver spleen kidney lung tumor carcass

In-111 anti-CEA, 72 h

no RIT

RIT

no RIT

RIT

11.96 2.87 5.52 1.56 4.35 0.72 6.79 0.69 5.14 0.87 21.14 3.45 3.08 0.37

11.89 2.76 5.87 2.52 5.46 1.08 6.24 0.58 5.64 0.96 19.84 1.89 3.17 0.26

11.29 1.74 5.07 1.96 3.25 0.91 3.62 0.60 4.35 1.01 29.72 8.05 2.47 0.31

15.83 2.02 5.48 1.25 5.35 1.52 4.83 0.64 6.95 0.60 39.31 7.87 2.98 0.31

14 Days In-111 anti-Her2, 72 h

Figure 3. Radioimmunotherapy of established MCF-7 tumors in nude mice with Y-90-labeled anti-CEA antibody in combination with a split dose of cold anti-Her2 antibody. Groups of eight mice were injected IV with Y-90-labeled anti-CEA antibody. Anti-Her2 antibody (200 µg) was administered IP at day 0 and day 9 post RIT. The tumor volume for each mouse at the time indicated was normalized to the tumor volume at the start of treatment (average relative tumor volume) and was plotted versus days after the start of treatment.

It has been previously shown that cold Herceptin has a minimal effect on tumors with a low expression level of Her2 (36). Since the expression level of Her2 is low and the gene is not amplified in MCF-7 cells (29), we expected that treatment of the animals with cold Herceptin would have no effect on tumor growth. Nonetheless, it is evident that many patients with low levels of Her2 expression are still treated with Herceptin, and we felt it prudent to determine if RIT plus cold Herceptin had an additive tumor effect. Therefore, MCF-7 xenograft-bearing animals were treated with Herceptin alone (200 µg IP at days 0 and 9), RIT alone (50 or 100 µCi), and Herceptin plus RIT (Figure 3). Tumor growths curves for untreated, day 0-33, or Herceptin-treated, day 1950, animals after initial tumor suppression were essential identical (p ) 0.28), while either 50 or 100 µCi of RIT with Y-90 anti-CEA produced substantial tumor growth inhibition (p ) 0.000004 for both) with a good separation between the two doses of RIT. The addition of cold Herceptin had no additive effect on either dose of RIT (p ) 0.57, p ) 0.14, respectively). Effect of Y-90 Anti-Her2 RIT on Anti-CEA and Anti-Her2 Uptake. The above results suggest that either anti-Her2 or anti-CEA RIT would have antitumor effects even in tumors expressing low levels of one or the other antigen. However, we were also interested in the possibility of treating tumors expressing both antigens by sequential RIT first with one then the other Y-90labeled antibody. The rationale for this approach is that single agent RIT against a tumor antigen may cause selective outgrowth of tumors that downregulate the antigen even further, thus rendering the complete ablation of the tumor unfeasible even with subsequent doses. If the tumor were treated first with one RIT agent and then a second, one would predict that the sequential dual therapy would be more efficacious than repeated admin-

blood liver spleen kidney lung tumor carcass

In-111 anti-CEA, 72 h

no RIT

RIT

no RIT

RIT

13.06 2.20 6.07 1.80 3.43 0.73 4.27 0.44 5.21 1.11 22.33 2.35 2.62 0.25

13.71 1.11 3.73 0.75 3.85 0.47 4.54 0.45 5.32 0.33 21.77 4.53 2.65 0.18

13.77 2.37 7.00 1.67 3.22 0.69 4.46 0.42 5.59 1.14 27.63 7.34 2.56 0.34

16.35 1.68 4.09 1.25 3.74 0.37 4.70 0.58 6.43 0.96 38.95 8.60 2.74 0.22

a Five animals were analyzed for the indicated tissues at the 72 h time point. Results are reported as %ID/g with the standard deviation given below.

istration of the single agent. As a first step in testing this approach, we treated MCF-7 xenograft-bearing animals with Y-90 anti-Her2 (60 µCi) for 7 or 14 days at which point the tumor growth reduction had reached its maximal effect and then administered In-111-labeled anti-Her2 or anti-CEA to determine if either or both antibodies have altered tumor uptake. Because the number of animals required to perform this experiment, we limited the biodistribution data to 72 h after administration of the In-111-labeled antibody in which the maximum tumor uptake is know to occur (see Table 1). We also performed whole animal imaging at various times to verify that tumor uptake of In-111 antibody was uniform over the time course of RIT. The results shown in Table 2 demonstrate that the tumor and blood values are statistically similar for In-111 anti-Her2 with or without Y-90 anti-Her2 RIT (p ) 0.06); however, for In111 anti-CEA, there is a substantial increase in tumor uptake after Y-90 anti-Her2 therapy (p ) 0.0005). The results are even more striking because the In-111 antiHer2 results serve as a control for the In-111 anti-CEA results. Whole body images of the mice after 7 days of RIT and 74 h after administration of In-111 anti-CEA revealed about a 2-fold increase in tumor uptake, comparing animals imaged with In-111 anti-Her2 or those receiving

1122 Bioconjugate Chem., Vol. 16, No. 5, 2005

Crow et al.

Figure 4. Radioimmunoimaging of established MCF-7 tumors in nude mice with In-111-labeled anti-Her2 or anti-CEA antibody, pretreated with Y-90-labeled anti-Her2 antibody. Groups of eight mice were treated IV with 60 µCi (6 µg) of Y-90-labeled anti-Her2 antibody followed by either 15 µCi (6 µg) of In-111-labeled anti-Her2 or 15 µCi (12 µg) of In-111-labeled anti-CEA antibody 7 days after the start of RIT. Images were acquired 74 h after injection of In-111-labeled antibodies. The central organ is liver and SC tumors are on located on the right side. A and C: In-111-labeled anti-Her2 antibody. B and D: In-111-labeled anti-CEA. A and B: images taken 10 days after start of RIT. C and D: images taken 10 days after injection of saline (no treatment controls).

no therapy (Figure 4). Similar results were obtained for animals treated for 14 days and then imaged with either In-111-labeled antibody. These images agree with the results of the animal biodistributions and demonstrate that prior RIT did not affect the time course of In-111 antibody uptake in tumor or normal tissues. Split-Dose RIT with Y-90-Labeled Anti-Her2 Followed by Y-90-Labeled Anti-CEA or Anti-Her2 Antibodies. Since In-111-labeled anti-CEA uptake increased after Y-90-labeled anti-Her2 RIT, we proceeded to test the effect of split-dose RIT, first with anti-Her2 and second with anti-Her2 or anti-CEA RIT. The first dose of RIT was administered at 60 µCi as suggested by the optimal anti-CEA uptake seen in Figure 4. The second dose of RIT was reduced to 40 µCi and administered 7 days after the first dose to minimize bone marrow toxicity. As seen in Figure 5, the single dose of Y-90labeled anti-Her2 RIT resulted in significant tumor reduction compared to controls, but the tumors continued to grow at a reduced rate. In animals that received no first dose of RIT but received the “second” dose of either Y-90-labeled antibody, tumor growth followed the controls until the “second” dose was administered at which point tumor growth was retarded but remained above 1.25 in relative tumor volume units. In animals that received the first dose of RIT followed by the second dose of either Y-90-labeled antibody, tumor volume showed a substantial reduction below 1.00 relative tumor volume units out to 60 days. It also appears that second RIT treatment with Y-90-labeled anti-CEA is more effective than that with Y-90-labeled anti-Her2 antibody. The curves are significantly different when analyzed by ANOVA (p < 0.05) but not significantly different when analyzed by the t-test (p ) 0.05). Thus, we conclude that a split dose of RIT with either antibody is effective and that the increased uptake of In-111-labeled anti-CEA observed in Figure 4 (Table 2) may predict a more effective split-dose RIT for anti-Her2/anti-CEA vs antiHer2/anti-Her2. To confirm this prediction additional experiments would be necessary with different doses of

Figure 5. Radioimmunotherapy of established MCF-7 tumors in nude mice with Y-90-labeled anti-Her2 antibody followed by either Y-90-labeled anti-Her2 or anti-CEA antibody. Groups of 10 mice were injected IV with 60 µCi (8 µg) of Y-90-labeled antiHer2 antibody followed by either 40 µCi (5.6 µg) of Y-90-labeled anti-Her2 or 40 µCi (6 µg) of Y-90-labeled anti-CEA 14 days later. The tumor volume for each mouse at the time indicated was normalized to the tumor volume at the start of treatment (average relative tumor volume) and was plotted versus days after the start of treatment.

RIT for each antibody. We believe that experiments of this nature are beyond the useful limits of this animal model.

Combined RIT and Taxol in a Breast Cancer Model DISCUSSION

Since it is unlikely that single agent therapies will be effective against metastatic breast cancer, new approaches to combined therapies need to be explored. In this regard, the combination of RIT and chemotherapy is relatively untried, and as previously mentioned, may allow the treatment of a greater number of patients than is possible for cold antibody/chemotherapy combinations. Our initial hypothesis that RIT would be more effective than cold antibody even against a tumor with relatively low to moderate amounts of target antigen was verified in the case of Her2 expression on MCF-7 cells. Surprisingly, in the case of CEA expression, which is also low on MCF-7 cells and led to disappointing tumor targeting as judged by a tumor-to-blood ratio of 1.0, RIT was equally effective for anti-CEA as for anti-HER2. Furthermore, we were able to reduce the amount of RIT required to produce an effective reduction in tumor size by the addition of Taxol as a chemotherapeutic agent. In our study, we were guided by the previous work of DeNardo and co-workers [35] who showed that the optimal combination of Taxol with RIT was either a single IP dose 24 h post RIT or a split IP dose at 48 and 72 h post RIT. Since in their study administration of Taxol prior to RIT was equivalent to single agent therapy, we tested only the single or split-dose regimen. Likewise, since the blood levels of Taxol were previously studied by DeNardo and co-workers (35) and shown to be within the effective range for 3-4 days, we did not measure blood levels. Our results closely mimic those obtained by DeNardo and co-workers (35) who performed their studies in a different tumor model (HBT 3477) with a different Y-90-labeled antibody (ChL6). Thus, it appears that the increased efficacy of RIT plus Taxol can be generalized, especially in the case of breast tumors. In our study, we had roughly equivalent results between the single or split-dose Taxol plus RIT regimens, suggesting that the largest antitumor effect occurred within the first few days of treatment, with tumor reductions of greater than 50% lasting up to 35 days. Since tumor growth was not monitored beyond 35 days, it is possible that some of the tumors would eventually regrow, requiring a second round of treatment. However, in one of the groups (80 µCi 90Y-cT84.66 ( Taxol), no detectable tumors were seen in 6/15 animals out to 35 days. Although it is generally recognized that only tumors that overexpress Her2 respond to cold Herceptin therapy, we felt compelled to determine if cold Herceptin therapy plus RIT was effective in this animal model. In this study, a split IP dose of cold Herceptin (200 µg) was administered at day 0 and day 9 after RIT. This regimen was adopted from our previous study in which we administered either 50 or 200 µg of cold 4D5, the parent monoclonal antibody for Herceptin, to nude mice bearing MCF-7/Her2 xenografts [11]. In that study, there was a 50% reduction in tumor growth at the end of 28 days. However, in our current study no long-term antitumor effect was observed for cold Herceptin only, as expected for a low Her2 expressing tumor (compared to the Her2 transfected MCF-7 cells used in the previous study). Furthermore, cold Herceptin therapy did not increase the antitumor effect of RIT, demonstrating the complete lack of an antitumor effect for Herceptin even in the presence of other cytotoxic agents. The findings of this study as a control for the next study (Herceptin RIT followed by antibody uptake at later times) are relevant in that even high-dose Herceptin has no antitumor tumor effect on MCF-7 xenografts. Thus, the antitumor effect is due to RIT alone.

Bioconjugate Chem., Vol. 16, No. 5, 2005 1123

When animals bearing MCF-7 xenografts were treated with Y-90 anti-Her2 and followed by In-111 anti-Her2 or anti-CEA 7-14 days later, a substantial increase in anti-CEA but not anti-Her2 uptake was observed compared to untreated controls. These results suggest that sequential treatment of tumors positive for both Her2 and CEA is feasible and may have an advantage over repeated RIT with the same agent. The mechanism for the increase in anti-CEA uptake was not studied. Possibilities include that anti-Her2 RIT induced expression of CEA or that prior therapy made CEA produced by the tumor more accessible. In consideration, of the latter possibility, antibody uptake is limited by tumor perfusion, which is known to be inhibited by high local intratumoral pressure (37-39). It is likely that some of this barrier is disrupted by tumor and vascular necrosis caused by RIT. Furthermore, in the untreated tumors (no RIT), tumor size has increased by 20% (Table 2), while tumor size has been reduced by 15% in the RIT-treated animals. Since uptake is measured as %ID/g, and we have observed that smaller tumors have an apparent advantage in uptake over larger tumors (23, 40), the increase in uptake is not surprising. However, the same effect is not observed for anti-Her2 uptake. Again, the mechanism was not studied. Thus, in this case the above arguments are unlikely to apply. In fact, one possibility is that antigen expression actually decreased in treated animals and was offset by the expected increase in uptake for a smaller tumor size. While further studies are warranted to unravel the underlying mechanisms, we conclude that sequential RIT therapy using different antibodies, especially combined with Taxoids such as paclitaxel and docetaxel are worth exploring. ACKNOWLEDGMENT

This research was supported by NCI Grant CA43904 and the Army Medical Research and Materiel Command DAMD 17-98-1-8031. LITERATURE CITED (1) O’Shaughnessy, J., Miles, D., Vukelja, S., Moiseyenko, V., Ayoub, J. P., Cervantes, G., Fumoleau, P., Jones, S., Lui, W. Y., Mauriac, L., Twelves, C., Van Hazel, G., Verma, S., and Leonard, R. (2002) Superior survival with capecitabine plus docetaxel combination therapy in anthracycline-pretreated patients with advanced breast cancer: phase III trial results. J. Clin. Oncol. 20, 2812-2823. (2) Albanell, J., and Baselga, J. (1999) The ErbB receptors as targets for breast cancer therapy. J. Mammary Gland Biol. Neoplasia 4, 337-351. (3) Baselga, J., Norton, L., Albanell, J., Kim, Y. M., and Mendelsohn, J. (1998) Recombinant humanized anti-HER2 antibody (Herceptin) enhances the antitumor activity of paclitaxel and doxorubicin against HER2/neu overexpressing human breast cancer xenografts. Cancer Res. 58, 2825-2831. (4) Baselga, J., Seidman, A. D., Rosen, P. P., and Norton, L. (1997) HER2 overexpression and paclitaxel sensitivity in breast cancer: therapeutic implications. Oncology 11, 4348. (5) Baselga, J. (2000) Clinical trials of single-agent trastuzumab (Herceptin). Semin. Oncol. 27, 20-26. (6) Baselga, J. (2001) Herceptin((R)) Alone or in Combination with Chemotherapy in the Treatment of HER2-Positive Metastatic Breast Cancer: Pivotal Trials. Oncology 61, 1421. (7) Slamon, D. J., Leyland-Jones, B., Shak, S., Fuchs, H., Paton, V., Bajamonde, A., Fleming, T., Eiermann, W., Wolter, J., Pegram, M., Baselga, J., and Norton, L. (2001) Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N. Engl. J. Med. 344, 783-792.

1124 Bioconjugate Chem., Vol. 16, No. 5, 2005 (8) Pegram, M. D., Pienkowski, T., Northfelt, D. W., Eiermann, W., Patel, R., Fumoleau, P., Quan, E., Crown, J., Toppmeyer, D., Smylie, M., Riva, A., Blitz, S., Press, M. F., Reese, D., Lindsay, M. A., and Slamon, D. J. (2004) Results of two openlabel, multicenter phase II studies of docetaxel, platinum salts, and trastuzumab in HER2-positive advanced breast cancer. J. Natl. Cancer Inst. 96, 759-769. (9) Theodoulou, M., and Seidman, A. D. (2003) Cardiac effects of adjuvant therapy for early breast cancer. Semin. Oncol. 30, 730-739. (10) Blend, M. J., Stastny, J. J., Swanson, S. M., and Brechbiel, M. W. (2003) Labeling anti-HER2/neu monoclonal antibodies with 111In and 90Y using a bifunctional DTPA chelating agent. Cancer Biother. Radiopharm. 18, 355-363. (11) Tsai, S. W., Sun, Y. Y., Williams, L. E., Raubitschek, A. A., Wu, A. M., and Shively, J. E. (2000) Biodistribution and radioimmunotherapy of human breast cancer xenografts with radiometal-labeled DOTA conjugated anti-her2/neu antibody 4D5. Bioconjugate Chem. 11, 327-334. (12) O’Donoghue, J. A., Baidoo, N., Deland, D., Welt, S., Divgi, C. R., and Sgouros, G. (2002) Hematologic toxicity in radioimmunotherapy: dose-response relationships for I-131 labeled antibody therapy. Cancer Biother. Radiopharm. 17, 435-443. (13) DeNardo, S. J., Williams, L. E., Leigh, B. R., and Wahl, R. L. (2002) Choosing an optimal radioimmunotherapy dose for clinical response. Cancer 94, 1275-1286. (14) Williams, L. E., Liu, A., Wu, A. M., Odom-Maryon, T., Chai, A., Raubitschek, A. A., and Wong, J. Y. (1995) Figures of merit (FOMs) for imaging and therapy using monoclonal antibodies. Med. Phys. 22, 2025-2027. (15) Behr, T. M., Griesinger, F., Riggert, J., Gratz, S., Behe, M., Kaufmann, C. C., Wormann, B., Brittinger, G., and Becker, W. (2002) High-dose myeloablative radioimmunotherapy of mantle cell non-Hodgkin lymphoma with the iodine-131-labeled chimeric anti-CD20 antibody C2B8 and autologous stem cell support. Results of a pilot study. Cancer 94, 1363-1372. (16) Wong, J. Y. C., Chu, D. Z., Yamauchi, D. M., Williams, L. E., Liu, A., Wilczynski, S., Wu, A. M., Shively, J. E., Doroshow, J. H., and Raubitschek, A. A. (2000) A phase I radioimmunotherapy trial evaluating 90yttrium-labeled anticarcinoembryonic antigen (CEA) chimeric T84.66 in patients with metastatic CEA-producing malignancies. Clin. Cancer Res. 6, 3855-3863. (17) O’Donnell, R. T., DeNardo, S. J., DeNardo, G. L., Miers, L., Lamborn, K. R., Kukis, D. L., and Meyers, F. J. (2000) Efficacy and toxicity of radioimmunotherapy with (90)YDOTA-peptide-ChL6 for PC3-tumored mice. Prostate 44, 187192. (18) Colnot, D. R., Roos, J. C., de Bree, R., Wilhelm, A. J., Kummer, J. A., Hanft, G., Heider, K. H., Stehle, G., Snow, G. B., and van Dongen, G. A. (2003) Safety, biodistribution, pharmacokinetics, and immunogenicity of 99mTc-labeled humanized monoclonal antibody BIWA 4 (bivatuzumab) in patients with squamous cell carcinoma of the head and neck. Cancer Immunol. Immunother. 52, 576-582. (19) Milenic, D., Garmestani, K., Dadachova, E., Chappell, L., Albert, P., Hill, D., Schlom, J., and Brechbiel, M. (2004) Radioimmunotherapy of human colon carcinoma xenografts using a 213Bi-labeled domain-deleted humanized monoclonal antibody. Cancer Biother. Radiopharm. 19, 135-147. (20) Postema, E. J., Frielink, C., Oyen, W. J., Raemaekers, J. M., Goldenberg, D. M., Corstens, F. H., and Boerman, O. C. (2003) Biodistribution of 131I-, 186Re-, 177Lu-, and 88Ylabeled hLL2 (Epratuzumab) in nude mice with CD22positive lymphoma. Cancer Biother. Radiopharm. 18, 525533. (21) Richman, C. M., and DeNardo, S. J. (2001) Systemic radiotherapy in metastatic breast cancer using 90Y-linked monoclonal MUC-1 antibodies. Crit. Rev. Oncol. Hematol. 38, 25-35. (22) De Santes, K., Slamon, D., Anderson, S. K., Shepard, M., Fendly, B., Maneval, D., and Press, O. (1992) Radiolabeled Antibody Targeting of the HER-2/neu Oncoprotein. Cancer Res. 52, 1916-1923.

Crow et al. (23) Williams, L. E., Beatty, B. G., Shively, J. E., and Beatty, J. D. (2001) Accumulation of radiolabeled anti-CEA antibody (mT84.66) in the case of multiple LS174T tumors in a nude mouse model. Cancer Biother. Radiopharm. 16, 147157. (24) Wong, J. Y., Shibata, S., Williams, L. E., Kwok, C. S., Liu, A., Chu, D. Z., Yamauchi, D. M., Wilczynski, S., Ikle, D. N., Wu, A. M., Yazaki, P. J., Shively, J. E., Doroshow, J. H., and Raubitschek, A. A. (2003) A Phase I trial of 90Y-anticarcinoembryonic antigen chimeric T84.66 radioimmunotherapy with 5-fluorouracil in patients with metastatic colorectal cancer. Clin. Cancer Res. 9, 5842-5852. (25) Esteban, J. M., Felder, B., Ahn, C., Simpson, J. F., Battifora, H., and Shively, J. E. (1994) Prognostic relevance of carcinoembryonic antigen and estrogen receptor in breast cancer patients. Cancer 74, 1575-1583. (26) Soletormos, G., Nielsen, D., Schioler, V., Mouridsen, H., and Dombernowsky, P. (2004) Monitoring different stages of breast cancer using tumour markers CA 15-3, CEA and TPA. Eur. J. Cancer 40, 481-486. (27) Given, M., Scott, M., McGrath, J. P., and Given, H. F. (2000) The predictive of tumour markers CA 15-3, TPS and CEA in breast cancer recurrence. Breast 9, 277-280. (28) Lumachi, F., Ermani, M., Brandes, A. A., Basso, S., Basso, U., and Boccagni, P. (2001) Predictive value of different prognostic factors in breast cancer recurrences: multivariate analysis using a logistic regression model. Anticancer Res 21, 4105-4108. (29) Ballangrud, A. M., Yang, W. H., Palm, S., Enmon, R., Borchardt, P. E., Pellegrini, V. A., McDevitt, M. R., Scheinberg, D. A., and Sgouros, G. (2004) Alpha-particle emitting atomic generator (Actinium-225)-labeled trastuzumab (Herceptin) targeting of breast cancer spheroids: efficacy versus HER2/neu expression. Clin. Cancer Res. 10, 4489-4497. (30) Meyer, D. L., Law, K. L., Payne, J. K., Mikolajczyk, S. D., Zarrinmayeh, H., Jungheim, L. N., Kling, J. K., Shepherd, T. A., and Starling, J. J. (1995) Site-specific prodrug activation by antibody-beta-lactamase conjugates: preclinical investigation of the efficacy and toxicity of doxorubicin delivered by antibody directed catalysis. Bioconjugate Chem. 6, 440446. (31) Lewis, M. R., Raubitschek, A., and Shively, J. E. (1994) A facile, water-soluble method for modification of proteins with DOTA. Use of elevated temperature and optimized pH to achieve high specific activity and high chelate stability in radiolabeled immunoconjugates. Bioconjugate Chem. 5, 565576. (32) Lewis, M. R., Kao, J. Y., Anderson, A. L., Shively, J. E., and Raubitschek, A. (2001) An improved method for conjugating monoclonal antibodies with N-hydroxysulfosuccinimidyl DOTA. Bioconjugate Chem. 12, 320-324. (33) Bacus, S. S., Chin, D. M., Stern, R. K., Ortiz, R., Ruby, S. G., and Weinberg, D. S. (1992) HER-2/neu oncogene expression, DNA ploidy and proliferation index in breast cancer. Anal. Quant. Cytol. Histol. 14, 433-445. (34) Mayer, A., Tsiompanou, E., Flynn, A. A., Pedley, R. B., Dearling, J., Boden, R., and Begent, R. H. (2003) Higher dose and dose-rate in smaller tumors result in improved tumor control. Cancer Invest. 21, 382-388. (35) Denardo, S. J., Richman, C. M., Kukis, D. L., Shen, S., Lamborn, K. R., Miers, L. A., Kroger, L. A., Perez, E. A., and Denardo, G. L. (1998) Synergistic therapy of breast cancer with Y-90-chimeric L6 and paclitaxel in the xenografted mouse model: development of a clinical protocol. Anticancer Res. 18, 4011-4018. (36) Lewis, G. D., Figari, I., Fendly, B., Wong, W. L., Carter, P., Gorman, C., and Shepard, H. M. (1993) Differential responses of human tumor cell lines to anti-p185HER2 monoclonal antibodies. Cancer Immunol. Immunother. 37, 255-263. (37) Hori, K., and Saito, S. (2003) Microvascular mechanisms by which the combretastatin A-4 derivative AC7700 (AVE8062) induces tumour blood flow stasis. Br. J. Cancer 89, 13341344. (38) Brekken, C., Bruland, O. S., and de Lange Davies, C. (2000) Interstitial fluid pressure in human osteosarcoma xeno-

Bioconjugate Chem., Vol. 16, No. 5, 2005 1125

Combined RIT and Taxol in a Breast Cancer Model grafts: significance of implantation site and the response to intratumoral injection of hyaluronidase. Anticancer Res. 20, 3503-3512. (39) Rohde, D., Wiesner, C., Graf, D., Wolff, J., Fuzesi, L., and Jakse, G. (2000) Interstitial fluid pressure is increased in renal cell carcinoma xenografts. Urol. Res. 28, 1-5.

(40) Williams, L. E., Duda, R. B., Proffitt, R. T., Beatty, B. G., Beatty, J. D., Wong, J. Y., Shively, J. E., and Paxton, R. J. (1988) Tumor uptake as a function of tumor mass: a mathematic model. J. Nucl Med. 29, 103-109.

BC0500948