Cancer Diagnosis and Therapy
Combined 90Yttrium-DOTA-labeled PAM4 antibody radioimmunotherapy and gemcitabine radiosensitization for the treatment of a human pancreatic cancer xenograft
Article first published online: 14 JAN 2004
Copyright © 2004 Wiley-Liss, Inc.
International Journal of Cancer
Volume 109, Issue 4, pages 618–626, 20 April 2004
How to Cite
Gold, D. V., Modrak, D. E., Schutsky, K. and Cardillo, T. M. (2004), Combined 90Yttrium-DOTA-labeled PAM4 antibody radioimmunotherapy and gemcitabine radiosensitization for the treatment of a human pancreatic cancer xenograft. Int. J. Cancer, 109: 618–626. doi: 10.1002/ijc.20004
- Issue published online: 24 FEB 2004
- Article first published online: 14 JAN 2004
- Manuscript Accepted: 15 OCT 2003
- Manuscript Revised: 24 SEP 2003
- Manuscript Received: 4 JUN 2003
- National Institutes of Health. Grant Number: CA54425
- New Jersey Commission on Cancer Research. Grant Number: 03-1105-CMM-N0
- monoclonal antibody;
- pancreatic cancer;
We have examined the application of 90Y-DOTA-cPAM4, anti-MUC1 IgG, in combination with the front-line drug gemcitabine as a potential therapeutic for pancreatic cancer. Athymic nude mice bearing CaPan1 human pancreatic cancer xenografts were administered 2 mg of gemcitabine on days 0, 3, 6, 9 and 12 with concurrent 90Y-DOTA-cPAM4 (100 μCi) provided on day 0. A second group of mice received a second cycle of treatment 5 weeks after the start of the first cycle. Control groups of mice included those that received either treatment arm alone, the combined modality treatment employing a nontargeting control antibody (hLL2, anti-B-cell lymphoma) and a final group that was left untreated. Gemcitabine administered as a single agent provided no antitumor effect. A single cycle of the combined 90Y-DOTA-cPAM4 and gemcitabine treatment provided greater inhibition of tumor growth than was observed for any of the other treatment procedures. Tumor growth was delayed for a period of 7 weeks. Two cycles of gemcitabine with concomitant 90Y-DOTA-cPAM4 yielded significant tumor regression and increased median survival to 21 weeks vs. 12 weeks for mice receiving a single cycle of therapy (p<0.024). Median tumor volume doubling-times were 18 weeks in mice treated with 2-cycles of therapy vs. 7 weeks in mice given only 1-cycle (p<0.001), and 3.5 weeks for the group that received 2-cycles of gemcitabine concomitant with equitoxic nontargeting 90Y-DOTA-hLL2 (p<0.001). These data suggest that addition of 90Y-DOTA-cPAM4 RAIT to a gemcitabine treatment regimen may provide enhanced antitumor efficacy for the treatment of pancreatic cancer. © 2004 Wiley-Liss, Inc.
Gemcitabine, a pyrimidine analog of deoxycytidine, is one of the most frequently used drugs for the treatment of pancreatic cancer. When used alone or in combination with other chemotherapeutic agents, this drug provides a moderate increase in median survival time.1 Increasingly, clinical investigations have turned towards the application of combined chemo- and radio-therapy for the treatment of pancreatic cancer. Gemcitabine may be useful in this regard since it is a potent radiosensitizer.2 External beam radiotherapy combined with gemcitabine chemotherapy has been studied in patients with both resectable3, 4, 5, 6, 7, 8, 9 and nonresectable tumors,10, 11, 12 with minimal toxicity. Although palliation of disease symptoms has been beneficial, only a moderate increase in median survival has been observed.
A major problem in the application of chemoradiation procedures for pancreatic cancer is the local nature of the treatment. While growth of the primary tumor may be controlled for a length of time, virtually all patients succumb to either local recurrence or metastatic disease. Limitations in providing directed beam radiation to all tumor sites, metastatic as well as primary, is difficult at best. One potential means for overcoming this problem is the use of radiolabeled antibodies that are able to target metastatic as well as primary tumor sites with minimal reactivity to normal tissues. Radiolabeled antibodies have been used with some success in numerous cancers, such as metastatic colorectal cancer13, 14, 15, 16 and several forms of hematologic malignancy.17, 18, 19, 20, 21, 22, 23 Although mechanistically complex, the inherent radiosensitivity of many hematologic malignancies, as well as the biological activity of the antibodies employed, have drawn considerable attention to radioimmunotherapy as a potential front-line therapeutic procedure.24, 25
Although several antibodies exist that are reactive with pancreatic cancer, little attention has been directed towards their application for radioimmunotherapy. We have developed MAb PAM4 that recognizes the MUC1 glycoprotein expressed selectively by human pancreatic cancer.26 By immunohistology, PAM4 was shown to be reactive with 85% of pancreatic cancer specimens examined. Except for a weak reactivity within the normal gastrointestinal tract, the antibody was not reactive with normal tissues, including normal pancreas. Targeting studies with radiolabeled PAM4 in patients with pancreatic cancer confirmed the presence of cancer in 8 of 10 patients.27, 28 One of the 2 nontargeted patients had a poorly differentiated form of pancreatic cancer that did not express MUC1, while the other patient did not have pancreatic cancer, but rather pancreatitis, providing further evidence as to the specificity of the PAM4 antibody for pancreatic cancer.
In preclinical studies, radiolabeled PAM4 provided significant antitumor effects in mice bearing large human pancreatic cancer xenografts.29, 30 We demonstrated recently that gemcitabine administered as a radiosensitizer significantly enhanced the anti-tumor efficacy of 131I-labeled PAM4 radioimmunotherapy (RAIT).31 However, the main effect was stabilization of disease with improvement in median survival, rather than tumor regression or cure. A major consideration in the application of RAIT is the choice of radionuclide coupled to the targeting antibody. In another recent study, we reported that PAM4 that was labeled with the high-energy β-emitter, 90Y, yielded a significantly improved anti-tumor effect as compared to 131I-labeled PAM4.29 The goal of the present study was to test the hypothesis that administration of 90Y-DOTA-PAM4 RAIT concurrently with gemcitabine, a radiosensitizing drug already approved for use in the treatment of pancreatic cancer, could provide a more effective treatment than either agent used alone.
MATERIAL AND METHODS
Experimental animal model
CaPan1 is a human pancreatic carcinoma that was obtained as a cell culture from the ATCC (Manassas, VA). It was established as a solid tumor by injecting 107 cells subcutaneously into the right flank of 5-week-old, female athymic nu/nu mice (Taconic, Germantown, NY). Once tumors had grown to approximately 1 cm,3 they were propagated serially by subcutaneous injection of 0.2 ml of a 20% (w/v) tumor suspension prepared by mincing the tumors in 0.9% saline with subsequent passage through a 40-mesh wire screen. CaPan1 cells produced moderate- to well-differentiated tumors. Tumors used in our study were passaged less than 10 times. Animal studies were approved by the Garden State Cancer Center's Institutional Animal Care and Use Committee and performed in accordance with the American Association of Laboratory Animal Care (AALAC), United States Department of Agriculture (USDA) and Department of Health and Human Services (DHHS) regulations.
Radiolabeling of antibody with 90Y
The purification and characterization of chimeric PAM4 (cPAM4) and isotype-matched, control, humanized LL2 (hLL2) anti-CD22, B-cell lymphoma MAb have been described;26, 32 hLL2 was provided by Immunomedics, Inc. (Morris Plains, NJ). DOTA (Macrocyclics, Inc., Richardson, TX) was conjugated to cPAM4 IgG by a previously described method that provides for prolonged, stable chelation of 90yttrium.33, 3490Y-chloride (NEN Life Science Products, Inc., Boston, MA) was added to a reaction vial with DOTA-cPAM4 IgG at a ratio of 5 mCi per mg cPAM4. The reaction was incubated at 45°C for 15 min and then quenched by removal from the heat source and addition of 0.1 volume of 100 mM DTPA (diethylenetrianinepentaacetic acid) to capture any free 90Y that remained unbound to the DOTA conjugated MAb. This mixture was then incubated at room temperature for 5 min. A volume of 1% human serum albumin in PBS was added to give a final activity concentration of approximately 5 mCi/ml. Conjugation of DOTA did not alter the immunoreactivity of the PAM4 antibody as determined by enzyme immunoassay with MUC1 antigen.26 The specific activities for the 90Y-DOTA-cPAM4 were in the range of 2.64–3.21 μCi/μg with less than 6% unbound material, while the specific activities for the 90Y-DOTA-hLL2 were in the range of 3.39–4.43 μCi/μg with less than 12.5% unbound material. The final labeled antibody product was placed in a dose calibrator (Capintec CRC15R, Ramsey, NJ) and the time and activity recorded. The immunoreactive fraction of the radiolabeled product was examined by reaction of antibody (approximately 10 ng) with 25 μg of purified MUC1 antigen at 37°C for 1 hr, followed by gel filtration chromatography on a column filled with Sepharose 4B-CL (1 × 50 cm) and eluted with 0.1 M sodium phosphate, pH 7.2, containing 0.15 M sodium chloride. Fractions (0.5 ml) were collected and then counted in a Packard Cobra-II Auto-Gamma Counter (Meriden, CT). Peaks of radioactivity were determined, area under the curve calculated and percent immunoreactive fraction derived. For the current studies, these values were 85% or better.
One-cycle gemcitabine radiosensitization and radioimmunotherapy
Initial tumor volume was determined by caliper measurements in 3 dimensions and calculated by length × width × depth. Mice were divided into groups (10 to 13 mice/group with starting tumor size of 1.0 ± 0.2 cm3) that were treated with 90Y-DOTA-cPAM4 with and without gemcitabine or 90Y-DOTA-hLL2 with and without gemcitabine, or gemcitabine alone. Radiolabeled antibody was augmented with unlabeled antibody to ensure that each mouse received a total of 50 μg of antibody. Those mice receiving gemcitabine (Gemzar, Eli Lilly and Company, Indianapolis, IN) were administered 5 intraperitoneal injections of 2 mg each on days 0, 3, 6, 9 and 12. This dose was chosen to approximate the human dose of 1000 mg/m2/week. Based on the formula to calculate the surface area of a mouse,35 m2=7.9 × (body weight)2/3/10,4 2 mg given to a 20 gram mouse would be similar to 333 mg/m2 or ∼1,000 mg/m2/week when given 3 times per week. This schedule for administration of gemcitabine was chosen to provide continuous radiosensitization of the tumor during the time period for maximum tumor uptake of radiolabeled antibody (first 2 weeks). In addition, it should be noted that others have reported an MTD for gemcitabine (in mice) of 2 mg given every third day for a total of 6 injections.36 The mice were weighed prior to each injection of gemcitabine to ensure that their body weight remained above 80% of starting weight. Weights were then taken once a week after the last injection. The maximum tolerated dose was defined as the dose at which all mice survived for at least 4 weeks with a body weight loss of no more than 20% of starting weight. Measurements for determination of tumor size were likewise performed on a weekly basis. Mice were sacrificed at a point in time when tumor size was equal to, or greater than, 5.0 cm3 (∼2.5 g).
Two-cycle gemcitabine radiosensitization and radioimmunotherapy
The mice were set up and treated the same as described above for 1 cycle of therapy. A second cycle of treatment was begun 5 weeks after the start of the first cycle. To reduce potential toxicity, mice received only 3 injections of gemcitabine (2 mg) on days 35, 38 and 41 and RAIT (100 μCi 90Y-DOTA-cPAM4 or equitoxic 50 μCi 90Y-DOTA-hLL2) on day 35 or RAIT alone. A second cycle of gemcitabine alone was not given since by 5 weeks post-first cycle therapy only 1 mouse was still alive in this group.
Statistical analyses for the tumor growth data were based on area under the curve (AUC), tumor doubling time and survival time. Profiles of individual tumor growth were obtained through linear and exponential curve modeling. The t-test was used to assess statistical significance between any 2 groups. As a consequence of incompleteness of some of the growth curves (due to deaths), statistical comparisons of AUC were performed only up to the time at which the first animal within a group was sacrificed. AUC analyses were supported by statistical comparisons of survival data; in this case, survival was defined as the time for a tumor to reach 5 cm.3 At the termination of the study, some of the animals had not yet experienced the endpoint and their observations were considered as censored. The Mantel-Haenszel, log-rank test was then used for comparison of treatment arms.37
Maximum tolerated dose (MTD)
The MTD for 90Y-DOTA-labeled cPAM4 administered as a single agent was previously shown to be 260 μCi.29 For the current work, a dose escalation study was performed to determine the MTD of 90Y-DOTA-labeled MAb that could be given concurrently with gemcitabine. Groups of CaPan1 tumor-bearing mice were administered gemcitabine (2 mg by intraperitoneal injection on days 0, 3, 6, 9 and 12) with 90Y-DOTA-labeled cPAM4 (50, 100 or 150 μCi) given on day 0. The MTD (dose at which none of the mice died within the first 4 weeks as a consequence of the treatment procedure and none of the mice lost greater than 20% of body weight) was determined to be 100 μCi. Likewise, 90Y-DOTA-labeled control hLL2 was administered concurrently with gemcitabine and the MTD determined to be 50 μCi. The estimated, potential radiation dose provided by the MTD of each 90Y-DOTA-radiolabeled MAb was calculated employing biodistribution data reported elsewhere.38 It should be noted that gemcitabine did not alter the pattern of tissue uptake for 90Y-DOTA-cPAM4 from what we had previously reported.29, 31, 38 At their respective MTD, both cPAM4 and hLL2 would deliver a similar radiation dose to the blood (608 cGy and 592 cGy, respectively). When normalized to these dose levels as an estimate of dose-limiting myelotoxicity, cPAM4 would provide a radiation dose to the tumor of 2,216 cGy (3.6-fold higher than that received by the blood), whereas control hLL2 would provide a dose to the tumor of only 758 cGy (1.3-fold higher dose than delivered to the blood). The only nontumor tissue that would receive a radiation dose higher than blood would be the liver (745 cGy with cPAM4 and 633 cGy with hLL2, at their respective MTDs).
Combined 90Y-DOTA-labeled cPAM4 RAIT and gemcitabine therapy of pancreatic cancer
Mice bearing large (∼1 cm3) subcutaneous human pancreatic cancer xenografts (CaPan1) received a single cycle of either cPAM4 RAIT (100 μCi 90Y-DOTA-cPAM4 on day 0) with or without gemcitabine (2 mg on days 0, 3, 6, 9 and 12) or equitoxic hLL2 RAIT (50 μCi 90Y-DOTA-hLL2 on day 0) with or without gemcitabine. Another group of animals received only gemcitabine, while a final group remained untreated. The mean tumor volumes from the various treatment groups, normalized to day 0, are presented in Figure 1a with survival curves provided in Figure 1b (survival was defined as the time required for the tumor size to become equal to or larger than 5.0 cm3 or for the animal to lose greater than 20% of body weight). Values for area under the curves of individual animals, response rates and survival were used to compare the antitumor effects of the different treatment procedures.
Gemcitabine, when provided in a regimen designed for radiosensitization as indicated above, had no effect upon the growth of these large tumors. The untreated group, as well as the groups treated with gemcitabine alone and 90Y-DOTA-hLL2 alone, exhibited rapid progression of tumor with median survival times of 6, 4 and 6 weeks, respectively. No statistically significant differences were observed between these 3 groups. The combined application of gemcitabine and nontargeting 90Y-DOTA-hLL2 demonstrated an early antitumor effect in comparison to either treatment arm alone (p<0.005 and p<0.007 at week 3 for AUC comparisons with gemcitabine alone and 90Y-DOTA-hLL2 alone, respectively). Tumor growth was delayed approximately 2 weeks (Fig. 1a); however, no significant differences were observed in median survival among these 3 groups (Fig. 1b).
Administration of 100 μCi 90Y-DOTA-cPAM4 without gemcitabine provided significant inhibition of tumor growth as compared to the untreated group. At week 4, the last assessable time point for the untreated mice, the group receiving cPAM4 treatment had tumors that were approximately 3-fold smaller than the untreated group [1.60 ± 0.58 cm3vs. 4.56 ± 2.66 cm,3 respectively; area under the curve, p<0.033]. No significant differences (AUC or median survival) were noted when comparing the cPAM4 treatment group to the gemcitabine alone, 90Y-DOTA-hLL2 alone or 90Y-DOTA-hLL2 with gemcitabine groups.
When 90Y-DOTA-cPAM4 was administered in combination with gemcitabine, we observed significantly greater inhibition of tumor growth than was provided by any of the other treatment procedures. At week 4, this group of mice had tumors that were 4.8-fold smaller than tumors in the untreated mice (0.95 ± 0.52 cm3vs. 4.56 ± 2.66 cm,3 respectively; AUC, p<0.002). Furthermore, the tumors in mice treated with combined cPAM4 RAIT and gemcitabine were 3.0-fold smaller than the tumors in mice that received only gemcitabine (1.26 ± 0.31 cm3vs. 3.83 ± 1.70 cm3 at week 3; AUC, p<0.001), and 1.7-fold smaller than those in mice that received only cPAM4 RAIT (1.44 ± 0.88 cm3vs. 2.46 ± 1.54 cm3 at week 7; AUC, p<0.002). Comparisons to the hLL2 RAIT groups, with or without gemcitabine, demonstrated the specificity of the PAM4 antibody targeted antitumor effect. These treatment groups were each significantly less tumoricidal than the cPAM4 RAIT/gemcitabine treatment group [AUC, p<0.002 at week 3 for comparison with hLL2 RAIT alone (3.45 ± 1.75 cm3) and p<0.001 at week 6 for comparison with hLL2 RAIT/gemcitabine (4.76 ± 1.87 cm3)].
These results translated into a significantly extended survival time (Fig. 1b) for the group of mice that received the combined 90Y-DOTA-cPAM4 RAIT with gemcitabine treatment. Median survival time for this group of mice was 12 weeks versus only 6 weeks for untreated mice (p<0.022), 4 weeks for gemcitabine-treated mice (p<0.001), 9 weeks for cPAM4 RAIT-treated mice (p<0.005) and 7 weeks for hLL2 RAIT/gemcitabine-treated mice (p<0.009).
Although only 1 out of 13 mice in the cPAM4 RAIT/gemcitabine group had an objective response to the treatment (tumor volume regressed to less than 50% of initial size) and was still alive at week 26 (the end of the study), 8 mice (62%) had stable disease (tumor volume between 50% and 125% of starting size) for a median of 7 weeks before disease progression. Four mice (31%) did not respond to the treatment. Similar response rates were observed in mice that received only cPAM4 RAIT, in which 7 of 12 mice (58%) had stable disease for a median of 7 weeks, while 5 of 12 mice (42%) did not respond. None of these mice were alive at the end of the study. In contrast, 7 of 8 mice (88%) that received hLL2 RAIT were nonresponders with 1 having an objective response for 6 weeks before disease progression. In those mice that received hLL2 RAIT/gemcitabine, 9 of 10 (90%) were nonresponders, with 1 animal disease-free at the end of the study.
The administration of 90Y-DOTA-cPAM4 RAIT concomitant with gemcitabine was well tolerated. The nadir, in terms of treatment related body weight loss, was 12 days poststart of therapy, with a mean 5.1 ± 4.2% drop in weight (range of 1.4% weight gain to 9% weight loss). Those mice that received only cPAM4 RAIT likewise reached their nadir on day 12, with a mean 1.1 ± 4.2% loss in weight (range of 6.1% weight gain to 9.2% weight loss). The difference between these 2 groups was not significant. Finally, the mice that received 90Y-DOTA-hLL2 RAIT and gemcitabine reached their nadir on day 7, with a mean 8.2 ± 1.6% loss in weight (range of 5.2% to 9.8% loss of weight). There were no treatment-related deaths in any of the groups.
Two-cycle 90Y-DOTA-cPAM4 RAIT with gemcitabine
We next examined the effect of administering multiple cycles of combined 90Y-DOTA-cPAM4 RAIT and gemcitabine. The first cycle utilized the same regimen as described above. A second cycle was administered 5 weeks (day 35) after the first and consisted of 100 μCi 90Y-DOTA-cPAM4 or 50 μCi 90Y-DOTA-hLL2 (equitoxic doses) concomitant with gemcitabine. A second cycle was not administered to those mice receiving gemcitabine alone, since only 1 out of 13 mice had not succumbed to disease by week 5. As further controls, 1 group of mice received a second cycle of cPAM4 RAIT alone while a second group of mice received a second cycle of hLL2 RAIT alone.
The mean normalized tumor growth curves of the various treatment groups are shown in Figure 2a. Two treatment cycles of 90Y-DOTA-labeled cPAM4 RAIT/gemcitabine provided significantly enhanced inhibition of tumor growth as compared to the untreated (p<0.001), gemcitabine alone (p<0.001), 2 cycles of 90Y-DOTA-cPAM4 (p<0.001), 2 cycles of 90Y-DOTA-hLL2 (p<0.001) or 2 cycles of 90Y-DOTA-hLL2 RAIT/gemcitabine (p<0.003) treatment groups. Survival curves for each of the groups are presented in Figure 2b. The median survival for animals treated with 2 cycles of cPAM4 RAIT/gemcitabine was 21 weeks. This was a 2.1-fold increase in median survival over animals administered 2 cycles of equitoxic hLL2 RAIT/gemcitabine (median survival=10 weeks, p<0.004). No significant difference was observed for survival curves between 2-cycle cPAM4 RAIT/gemcitabine and 2-cycle cPAM4 RAIT alone (median survival=16 weeks, p<0.23). However, response rates indicated that the combined cPAM4 RAIT/gemcitabine treatment group of mice fared better than the 2-cycle cPAM4 RAIT treatment group (Table I). Although both of these groups had similar numbers of mice with stable disease and similar median time to progression of tumor, the number of responders in the combined cPAM4 RAIT/gemcitabine group was almost double that of the cPAM4 RAIT alone group. Tumors from the cPAM4 RAIT/gemcitabine treated mice regressed to less than 50% of their initial tumor volume in approximately half the time (median=4 weeks) in comparison to tumors from the cPAM4 RAIT-treated mice (median=7.5 weeks, p<0.031). In addition, the combined treatment group maintained this positive response for more than twice the length of time (median=11 weeks) than the cPAM4 RAIT group (median=5 weeks) before the disease began to progress (p<0.048).
|Treatment||(N)||Number of mice with stable disease1 (percent)||Median time to disease progression (weeks)||Number of positive responders2 (percent)||Median Time to Response in Weeks (range)||Median Time of Response in Weeks (range)|
|50 μCi90Y-hLL2||13||1 (8%)||53||0||—||—|
|50 μCi90Y-hLL2 plus Gemcitabine||13||3 (23%)||5||1 (8%)||43||>233|
|100 μCi90Y-cPAM4||12||3 (25%)||12||4 (33%)||7.5 (7 to 9)||5 (2 to 10)|
|100 μCi90Y-cPAM4 plus Gemcitabine||12||4 (33%)||11||7 (58%)||4 (3 to 9)||11 (6 to >23)|
Some toxicity was observed in the group of mice that received the cPAM4 RAIT/gemcitabine treatment (Fig. 3). During the first cycle of treatment, 2 out of 13 mice lost more than 20% of their starting body weight by day 14 (23% loss for both). This was in contrast to a mean weight loss of only 7.4 ± 3.6% for the remaining mice in the group (n=10). One of the 2 mice died while the second mouse recovered and was not adversely affected by a second cycle of treatment. None of the remaining mice in this group dropped below the 20% mark, even after the administration of the second cycle. The nadir for the second cycle was reached on day 41 with a mean 6.0 ± 5.0% loss in weight (range of 3.3% gain to 17.2% loss in weight). These data suggested that we were at the toxic level and may need to use slightly less RAIT and/or gemcitabine in the future to ensure less toxicity. The mice that received control MAb reached nadir on day 12 of the first cycle with a mean 10.2 ± 3.7% loss in weight (range of 5.7% to 16.3% drop in weight) with no overt signs of toxicity. However, the second cycle resulted in a 21% drop in weight by day 41 for 1 animal while the group as a whole (excluding this animal) lost only 3.7 ± 4.8% (day 41). The overall nadir was reached 3 days later (day 44) with a mean 6.2 ± 7.7% loss in weight (range of 2.4% to 21.8% loss in weight). At the nadir points there were no statistically significant differences between the 90Y-DOTA-labeled cPAM4/gemcitabine and control hLL2/gemcitabine groups. However, there were significant differences between each of the respective single treatment arms in comparison to the combined treatment group, in that they did not experience as great a loss of body weight (p<0.001 for both gemcitabine and 90Y-cPAM4 alone).
Comparison of 2-cycle vs. 1-cycle therapy
Adding a second cycle of 90Y-DOTA-cPAM4/gemcitabine significantly increased median survival (Table II) almost 2-fold from 12 weeks to 21 weeks (p<0.024). At 5 weeks after initiation of treatment (the time at which the second cycle was started), there was no significant difference in tumor volumes between the 1-cycle and 2-cycle treatment groups (p<0.10). By week 7 (2 weeks post-second cycle) the 2-cycle treated mice had significantly smaller tumors than those mice that received only 1 treatment cycle (0.74 ± 0.59 cm3vs. 2.10 ± 1.25 cm,3 respectively; AUC, p<0.014). This difference grew to a 4.5-fold advantage for the 2-cycle group at week 8, the last assessable time point for the 1-cycle treatment group of mice (0.60 ± 0.60 cm3vs. 2.69 ± 1.60 cm,3 respectively, AUC, p<0.003). An analysis of the time required for tumors to double in size showed that 2 cycles of 90Y-DOTA-cPAM4/gemcitabine treatment resulted in a median time of 18 weeks vs. 7 weeks for mice treated with 1-cycle (p<0.001) and 3.5 weeks for mice treated with 2 cycles of control 90Y-DOTA-hLL2/gemcitabine (p<0.001). The doubling time in mice treated with 2 cycles of 90Y-DOTA-cPAM4 alone was 12 weeks and approached significance when compared to 2 cycles of 90Y-DOTA-cPAM4/gemcitabine (p<0.059).
|Treatment 1 vs.||Treatment 2||Median Survival Time1||Normalized tumor growth (week 8)|
|(1×) 100 μCi90Y-cPAM4 plus Gemcitabine||Untreated||2.0||<0.022||N/A6||—|
|(2×) 100 μCi90Y-cPAM4 plus Gemcitabine||Untreated||3.5||<0.006||N/A6||—|
|(1×) 100 μCi90Y-cPAM4 plus Gemcitabine||1.8||<0.024||0.21||<0.003|
Pancreatic cancer is one of the most challenging forms of cancer to both diagnose and treat. Unfortunately, this is reflected in a high mortality rate; <5% at 1 year and <1% at 5 years.39 Early diagnosis would almost certainly improve the mortality rate; however, the nature of disease progression is such that by the time symptoms become evident, the tumor has already grown very large and has metastasized. Current treatment procedures for advanced local disease, as well as metastatic disease (the 2 stages of disease activity at which the overwhelming majority of patients present) have not been able to bring about a cure nor substantially improve survival times. The front-line drugs of choice, gemcitabine and 5-fluorouracil, and/or irradiation of the tumor via external and intra-operative beam procedures, are most often used for palliation of disease symptoms rather than as curative procedures.
In recent years, investigators have turned their attention towards a combined modality approach for the treatment of pancreatic cancer. Many of the chemotherapeutic agents that are routinely used for pancreatic cancer, in addition to their antitumor activities, are able to radiosensitize the tumor.2 Thus, the application of radiosensitizing drugs with concurrent irradiation of the tumor bed, that is, chemoradiation, is being investigated as a primary treatment for locally advanced, unresectable pancreatic cancer,10, 11, 12, 40, 41 as well as a neoadjuvant procedure for down-staging of tumor to the point where surgical resection may be feasible.42, 43, 44 Although few patients are deemed to have resectable disease, recent studies have suggested that these patients may also benefit from a program of preoperative chemoradiation.3, 4, 5, 6, 7, 8, 9 Those patients given preoperative chemoradiation had significantly higher numbers of disease-free margins and fewer numbers of involved lymph nodes at surgery than their counterparts treated with “curative resection” alone. Overall, the use of chemoradiation procedures has provided a modest increase in median survival time. However, long-term patient outcomes have not been altered, with the majority of patients succumbing to progression of tumor at metastatic sites.
As already noted, one potential means for overcoming this problem with chemoradiation is the use of radiolabeled antibodies that are able to provide irradiation of both primary and metastatic tumors. Applications for radioimmunotherapy in solid tumors, and in particular those tumors that usually present with bulky disease, must overcome the challenges of adequate tumor uptake of radiolabeled antibody and penetration of the tumor. For solid tumors, the most appropriate use of radioimmunotherapy, as the technology exists today, may be as neoadjuvant prior to surgical resection and/or as adjuvant for small volume disease. Preclinical studies of colorectal cancer have shown radiolabeled anti-CEA antibodies to be curative of small volume disease.45, 46 In an ongoing trial employing radiolabeled anti-CEA antibody for treatment of small volume metastatic colon cancer, an overall response rate of 58% was observed. In the adjuvant setting postresection of liver metastases, 7 of 9 patients remained disease free for up to 36 months, whereas the historical relapse rate for patients receiving chemotherapy was 67%.47
Studies designed to improve the prospects for radioimmunotherapy in solid tumors have for the most part focused on mechanisms that would improve radiation dose and dose rate to the tumor. In contrast, we, and others, have examined methods to increase radiosensitivity of the tumor, particularly with the application of chemotherapeutic drugs that may provide antitumor effects in and of themselves. Indeed, several preclinical studies of various tumor types (colorectal,48, 49, 50, 51 breast,52, 53, 54 medullary thyroid55, 56, 57 and the more radiosensitive hematologic cancers58, 59) have examined this issue, with the majority reporting enhanced antitumor activity of the combined treatment modality as compared to either treatment arm alone. To date, clinical experience in solid tumors is limited. However, a recent Phase I trial where the treatment of recurrent ovarian cancer with combined paclitaxel and 90Y-CC49 was well tolerated, an interim report noted an extended disease-free interval of 15–23 months in almost half of the patients receiving the combined modality approach.60
In an effort to develop procedures that will enhance the current clinical treatment of pancreatic cancer, we examined the ability of gemcitabine to sensitize pancreatic cancer to the antitumor effects of PAM4 RAIT. Gemcitabine, a nucleoside analog of cytosine, was chosen because it is one of the current front-line drugs of choice for pancreatic cancer as well as a potent radiosensitizer.2, 61 MAb PAM4 is reactive with the MUC1 antigen as produced by the majority of pancreatic adenocarcinomas.26 Clinical studies have demonstrated specific targeting of radiolabeled PAM4 to pancreatic tumors within an initial group of patients.27, 28 In preclinical studies, PAM4 targeted primary subcutaneous and orthotopic sites as well as metastatic tumors.62, 63 Experimental RAIT with PAM4 in mice bearing human pancreatic cancer xenografts demonstrated significant tumor regression and increased survival of treated animals.29, 30 In a previous study employing 131I-labeled cPAM4 with gemcitabine, we demonstrated that the combined modality treatment controlled tumor growth for several weeks, resulting in significantly longer survival.31 However, the combined treatment did not produce tumor regression.
In our study, the radiosensitizing properties of gemcitabine and the superior tumoricidal effects of 90Y-DOTA-labeled cPAM4 RAIT were combined. The dose of gemcitabine was chosen to approximate the amount administered to patients. Usually, gemcitabine is administered as a 30 min infusion of 1,000 mg/m2 once a week. For mice, this would translate to approximately 6 mg. Previous biodistribution data indicated that the 90Y-labeled cPAM4 would accumulate in the tumor over several days, with peak tumor uptake on days 3–4, before decaying to background with very little activity remaining after day 14.29 In an effort to maintain the gemcitabine in the tumor over this time, we administered the drug as a 2 mg i.p. injection given every 3 days over 12 days. In addition, our studies utilized large (∼1.0 cm3) CaPan1 human pancreatic tumor xenografts at the start to better approximate the clinical presentation of this disease.
The MTD for 90Y-DOTA-labeled PAM4 was found to be 100 μCi when administered with gemcitabine at the dose schedule used here, in comparison to an MTD of 260 μCi when used alone.29 The dose-limiting toxicity of RAIT is myelotoxicity and in mice the maximum absorbed radiation dose to the blood, as an estimate of myelotoxicity, is approximately 1500 cGy.64, 65, 66, 67, 68 At 100 μCi of 90Y-cPAM4, the estimated absorbed radiation to the blood was calculated to be 600 cGy. This radiation dose is the same as was observed with 131I-cPAM4 when given with gemcitabine at its MTD of 200 μCi.31
The combined modality treatment employing gemcitabine and 90Y-DOTA-cPAM4 yielded stable disease for 5 weeks before progression resumed. This resulted in a median survival of 12 weeks, which was not very different from the median survival of 13 weeks previously reported for 131I-cPAM4 in combination with gemcitabine.31 However, in preliminary MTD studies, smaller tumors (<0.5 cm3) did regress when the animals were treated with the combined modality regimen. Thus, tumor burden plays a role in the effectiveness of the treatment outlined in our study. As discussed above, the combined 90Y-DOTA-cPAM4/gemcitabine regimen might be especially effective for small volume disease.46, 47
Although a single cycle of combined gemcitabine and 90Y-DOTA-cPAM4 RAIT provided control of tumor growth with stable disease, administration of a second treatment cycle provided significant tumor regression and extended survival. The data suggest that multiple treatment cycles may provide further enhancement of antitumor therapy over that seen with gemcitabine chemotherapy alone. Several points must be considered when multiple treatment cycles are to be used. First, the scheduling of the cycles must take into account the time required for the individual to recover from toxicity arising from the previous cycle. Generally, the bone marrow of mice injected with 90Y-labeled MAbs rebounded faster than in mice injected with 131I-labeled MAbs. It was reported, and our studies confirmed, that mice recover from 90Y-induced myelotoxicity by week 4 post-injection.45 It has also been demonstrated that the marrow of mice will start to recover even during continual exposure to gemcitabine and will rebound to nearly 100% 7 days after treatment is stopped.69 A second consideration for administration of multiple cycles is the effect the first cycle may have on the ability of subsequent cycles to target and penetrate the tumors, including the effect on the tumor vasculature. A decreased vascular permeability should have a negative impact on the ability of gemcitabine and PAM4 to be taken up by the tumor. However, in a study investigating the effect of PAM4 RAIT on tumor vascular permeability, it was found that vascular permeability within the tumor was not significantly altered at absorbed radiation doses of less than 3,000 cGy.70 At the doses administered in the current studies, approximately 2,200 cGy were delivered to the tumor and thus the uptake of gemcitabine and cPAM4 when given in a second cycle should not have been affected.
A problem that may arise from combining radiosensitizers with radiotherapy is toxicity to normal tissues. The main toxicity associated with radiotherapy of the abdomen is GI syndrome. This condition can be mild with diarrhea to severe with enterobacterial infection leading to septic shock and death.71 In the crypt cells of mouse jejunum, gemcitabine has been shown to inhibit DNA synthesis within 3 hr of administration followed by a rapid recovery 6 hr later.72 During this time, the cells are sensitive to the effects of ionizing radiation. It has been reported in some clinical trials that the combination of gemcitabine and external beam radiotherapy has resulted in higher than expected toxicity to normal tissues.10, 12, 73 This has manifested itself in pancreatic cancer patients as intestinal ulcerations10 and complications at endobilliary stents.12 However, these toxicities did not lead to deaths and in the case of endobilliary stent problems, other studies refuted this observation.74 Moreover, a preponderance of clinical data show the combination to be well tolerated.3, 4, 5, 6, 7, 8, 9, 11, 12 Since we were administering 90Y-DOTA-cPAM4 concurrently with gemcitabine, toxicity to the intestine was a distinct possibility. However, 90Y-PAM4 is cleared rapidly from the blood (t1/2=28 hr) and our previous biodistribution data show that very little radiation is absorbed by the small intestine (135 cGy at 100 μCi injected dose), and therefore should not be toxic in these mice.31, 38 While there was 1 mouse that appeared to have died due to treatment-related causes in the 2-cycle study, overall the animals in our study tolerated the treatment well, with a nadir in body weight loss of the first cycle (7.4 ± 3.6%) reached on day 14 post-therapy and recovery by week 4. The nadir for the second cycle was reached on day 6 post-second treatment (study day 41), again without major loss of body weight (6.0 ± 5.0%).
To date, the most effective antitumor radioimmunotherapy-based treatment regimen that we have observed in the experimental pancreatic cancer model (human tumors xenografted in athymic nude mice) is the administration of the MTD of 90Y-DOTA-PAM4 provided as a single agent.29 However, in a clinical setting, tumor uptake of radiolabeled antibody is usually orders of magnitude less than is achieved in mice. Thus, the significance of the current work is that the combination of 90Y-DOTA-cPAM4 RAIT with a gemcitabine treatment regimen, designed to provide high levels of radiosensitization within the tumor, resulted in significant anti-tumor effects and prolonged survival over that observed by treatment of tumor-bearing mice with gemcitabine alone. Furthermore, the addition of a second treatment cycle proved more effective than administration of only a single cycle. It should be noted that we have not attempted to optimize the protocol as to timing between administration of treatment cycles, as well as dose of gemcitabine and cPAM4 RAIT within each cycle. Perhaps, by providing a lower dose of cPAM4 RAIT more frequently, with constant gemcitabine dosing, we could enhance antitumor efficacy similar to the increased anti-tumor effects observed with fractionated dose RAIT procedures.75, 76, 77 In a clinical setting, the major reason for treatment failure has been the occurrence and/or recurrence of disease at distant metastatic sites. As stated above, we have demonstrated that radiolabeled PAM4 can target both primary and metastatic pancreatic tumor sites both in preclinical animal models62, 63 and clinical imaging studies.27, 28 By combining PAM4 RAIT with current chemo- and/or radio-therapies, we hope to provide a more effective treatment for pancreatic cancer.
The authors thank Dr. D.M. Goldenberg for his review and advice in the performance of these studies, as well as Mr. T. Jackson and Mr. P. Andrews for their expert help in the radiolabeling of antibodies.
- 10Eastern Cooperative Oncology Group Phase I trial of protracted venous infusion fluorouracil plus weekly gemcitabine with concurrent radiation therapy in patients with locally advanced pancreas cancer: a regimen with unexpected early toxicity. J Clin Oncol 2000; 18: 3384–9., , , , , , .
- 11A Phase I trial of gemcitabine and radiation in locally advanced unresectable cancer of the pancreas. Proc Am Assoc Cancer Res 2000; 41: 613., , , , .
- 35Veterinary cancer medicine. Philadelphia: Lea and Feiger, 1987. 189., .