Advantage of Dose Fractionation in Monoclonal Antibody-Targeted Radioimmunothefapy Jeffrey Schlom, * Alfredo Molinolo, Jean F. Simpson, Kathleen Siler, Mario Roselli, George Hinkle, David P. Houchens, David Colcher
Numerous monoclonal antibodies (MAbs) to human solid tumors have been developed and are being evaluated to determine their potential use in the treatment of human carcinomas. Several of these MAbs have been coupled to radionuclides and have-been used in clinical trials for diagnostic tumor targeting. Phase I therapeutic trials with some of these MAbs have recently been initiated {1-11). As a prerequisite for most therapeutic clinical trials and for a Vol. 82, No. 9, May 2, 1990
Downloaded from http://jnci.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on March 15, 2015
Monoclonal antibody (MAb) B72.3 IgG was radiolabeled with 131I and administered to female athymic NCr-nu mice bearing the LS-174T human colon adenocarcinoma xenograft to determine if fractionation of MAb dose had any advantage in tumor therapy. In the LS-174T xenograft, only approximately 30%-60% of tumor cells express the B72.3-reactive TAG-72 antigen. The LS-174T xenograft was used to reflect the heterogeneity of the TAG-72 antigen often seen in biopsy specimens from patients. In contrast to a single 600-JJLG dose of 131I-B72.3 IgG where 60% of the animals died from toxic effects, two 300-y.Ci doses of 131I-B72.3 IgG (total of 600 ^Ci) reduced or eliminated tumor growth in 90% of mice, with only 10% of the animals dying from toxic effects. Dose fractionation even permitted escalation of the dose to three doses (each 1 wk apart) of 300 u,Ci of 131I-B72.3 IgG (for a total of 900 (xCi), resulting in even more extensive tumor reduction or elimination and minimal toxic effects. The use of an isotype-matched control MAb revealed a nonspecific component to tumor growth retardation, but the use of the specific B72.3 IgG demonstrated a much greater therapeutic effect. Tumors that had escaped MAb therapy were analyzed for expression of the B72.3-reactive TAG-72 antigen with the use of the immunoperoxidase method; they were shown to have the same antigenic phenotype as the untreated tumors. We verified tumor elimination by killing the test animals after a 7-week observation period and performing histologic examination of tumor sites. We also monitored toxic effects by histologic examination of numerous organs, including bone marrow. These studies thus demonstrate the advantage of dose fractionation of a radiolabeled MAb for tumor therapy. We anticipate that the concept of dose fractionation can be practically applied in radioimmunotherapeutic clinical trials with the development and use of recombinant-chimeric MAbs and modified constructs. [J Nat! Cancer Inst 82:763-771,1990]
better understanding of how radiolabeled MAbs kill in vivo, several MAbs have been analyzed for efficacy in reducing tumor growth in athymic mice bearing transplanted human tumor xenografts (12-23); often these MAbs have been labeled with radionuclides such as 131I and 90 Y. Two conclusions can be drawn from most of these studies: (a) the rate of tumor growth is reduced following one administration of radiolabeled MAb; and (b) with increasing levels of radionuclide bound to MAbs, more efficient tumor cell killing occurs, but there is a concomitant increase in bone marrow toxicity that leads to the death of some animals. One way of potentially overcoming this marrow toxicity would be to fractionate the dose of the administered MAb. In a previous study (72), employing F(ab')2 fragments of l3l I-labeled, anti-carcinoembryonic antigen MAbs, doses of MAb that were fractionated at days 10,12,24, and 38 after tumor transplantation showed only a slight advantage over a single dose. In another study (24), each of two F(ab') 2 MAbs was given at a dose of 2 mCi at 3-day intervals three times; this regimen was shown to be as therapeutic as a single 2-mCi dose of l31I-labeled IgG. These data suggest that multiple doses of F(ab') 2 can be as tumoricidal as a single dose of an intact IgG and can be less toxic to the host. Ceriani and Blank (25) reported that a second administration of 500 |xCi of l3l I-MAb (for a total dose of 1,000 p-Ci) further diminished tumor growth compared with tumor growth in mice treated with one 500-|iCi 131I-MAb dose. Hyams et al. (26) observed no significant improvement in xenograft-treated mice when given two cycles of 120 u,Ci of ^ - l a b e l e d MAbs versus one cycle of 120 u,Ci of ^ - l a b e l e d MAbs. However, Hyams et al. reported 100% and 91.7% disease-related deaths, respectively, in these two groups. Thus, these two recent studies (25,26) compared multiple administrations with a single administration of radiolabeled MAbs, but in these studies, the groups receiving multiple administrations also received higher doses of radiola- .
Received November 6,1989; revised February 6, 1990; accepted February 20, 1990. J. Schlom, A. Molinolo, J. F. Simpson, K. Siler, M. Roselli, D. Colcher, Laboratory of Tumor Immunology and Biology, National Cancer Institute, Betbesda.MD. G. Hinkle, The Ohio State University, Columbus, OH. D. P. Houchens, Battelle Memorial Institute, Columbus, OH. We thank D. Milenic, R. Nines, and D. Hill for expert technical assistance, and E. Glatstein and A. Raubitschek for insightful discussions. 'Correspondence to: Jeffrey Schlom, Ph.D., Bldg. 10, Rm. 8B07, National Institutes of Health, Bethesda, MD 20892.
ARTICLES
763
Materials and Methods Monoclonal antibodies. MAb B72.3 was obtained by immunization of BALB/c mice with a membrane-enriched fraction of a human breast tumor metastasis to the liver. The generation, characterization, and reactivities of MAb B72.3 were described elsewhere (46,47). The IgG was purified as previously reported (48) from ascitic fluid by ammonium sulfate precipitation and ion-exchange and size-exclusion chromatography. BL-3 is a murine IgGl anti-idiotype MAb to a human B-cell lymphoma previously shown (49) to be negative for reactivity to a wide range of cells from murine and human tissues, including LS- 174T colon carcinoma cells. Iodination of MAb B72.3. MAb B72.3 was labeled with Na l 3 l l using the Iodogen method (50). The purified IgG (1.7 mg) was labeled with approximately 18 mCi of Na l 3 l l in a glass vial coated with 250 (xg of Iodogen. After a 20-minute incubation, the free I3I I was removed by gel filtration through a Sephadex G-25 column. The immunoreactivity of the labeled MAb B72.3 was tested in a solid-phase radioimmunoassay, using extracts of a human colon carcinoma xenograft (57). Cell line. The LS-174T cell line (CL188) (45) was obtained from the American Type Culture Collection, Rockville, MD. This cell line was established from a human mucinous adenocarcinoma of the colon. The cells were grown in Eagle's minimum essential medium supplemented with 1% nonessential amino acids (100 vaM), including 1% glutamine (200 mM), 10% heat-inactivated fetal calf serum, and gentamicin (50 u,g/mL).
764
Cells were passaged weekly at a 1:10 dilution. They grew as a monolayer and were harvested with the use of 0.1 % trypsin in 0.5 mAf edetic acid (EDTA). The cells were washed twice in serum-free minimum essential medium before they were injected into athymic mice. Tumor growth in athymic mice. Female athymic mice (NCrnu) were obtained from Simonsen Laboratories, Gilroy, CA, at 4-6 weeks of age. The mice were given a subcutaneous injection in the right flank of 1 x 106 cells (0.1 mL). Seven days after inoculation, the mice bearing rumors of approximately 0.5 cm in diameter were selected and divided into treatment groups. Groups of 10-15 mice each received an intraperitoneal injection of various amounts of I3I I-B72.3 IgG at weekly intervals. The interval of 7 days between doses was chosen because we (52) have shown that, at day 7, greater than 90% of intraperitoneally administered MAbs are cleared from the blood pool, thus reducing the potential for marrow toxicity and enabling the marrow to regenerate. For the fluorouracil (5-FU) study, 5-FU was injected intraperitoneally daily for 5 days. The tumor growth was measured in two diameters weekly with a precision caliper, until the mice were killed. Mice receiving one injection of the 600-^Ci dose of' 3 ' I-B72.3 IgG were held for 42 days after injection. Mice receiving two injections of radiolabeled MAb B72.3 were held for 35 days after the last injection. Mice receiving three injections of radiolabeled MAb B72.3 were held for .45 days after the last injection. Mice receiving two injections of radiolabeled MAb BL-3 were held for 43 days after the last injection. Results are expressed as tumor volume as described previously (75-78). The control groups, mice given an injection of phosphate-buffered saline or unlabeled MAb B72.3, were similarly analyzed. Histologic studies. The tumor or site of tumor inoculation, spleen, bone marrow, lung, small intestine, and liver were obtained from selected animals. They were fixed in 10% buffered Formalin, routinely processed, and embedded in paraffin. Fivemicron sections were obtained from the paraffin blocks and stained with hematoxylin and eosin for examination by light microscopy. Blood smears were also obtained and stained with Wright's stain. Reactivity of the various tissues with biotinylated MAb B72.3 was determined using a modification of the avidinbiotin-peroxidase complex (ABC) method (Vectastain ABC kit; Vector Laboratories, Inc., Burlingame, CA) (53). The test concentration of MAb B72.3 was 80 u,g/mL; an isotype-identical control antibody, MOPC-21 (54), was used at an identical concentration on parallel sections; no reactivity was observed with the control MAb.
Results Single-Dose Administration of Radiolabeled MAb We (23) previously showed that administration of a single dose of either 300 or 500 |xCi of I3I I-B72.3 IgG to athymic mice bearing LS-174T colon carcinoma xenografts reduced tumor growth for 17 days after MAb administration. The weight of tumors in the controls was up to 3.7 times the weight of tumors in mice receiving 500 u.Ci. However, that same study also demonstrated that at 17 days after injection of the 500-jiCi dose of MAb, there was almost complete aplasia of the host bone marrow, with less than 5% of the marrow cellularity remaining. In addition, the Journal of the National Cancer Institute
Downloaded from http://jnci.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on March 15, 2015
beled MAbs. To our knowledge, no systematic comparison of various equivalent fractionated doses has been undertaken in which radiolabeled MAbs were used for tumor therapy. In the studies reported here, we have chosen MAb B72.3 IgG to conduct dose-fractionation studies. Radiolabeled B72.3 IgG has been administered intravenously to over 500 patients and has demonstrated efficient rumor targeting in approximately 70% of carcinomas of the colon and ovary (27-32). Recent studies involving intraperitoneal administration of l3lI-labeled MAb B72.3 have also demonstrated efficient tumor targeting (33). l3l I-MAb B72.3 is now being used in a phase I clinical trial (77). In the consideration of any MAb therapy using multiple dose scheduling in patients, a major concern is the development of human anti-mouse antibodies (HAMAs). The construction of recombinant-chimeric MAbs, such as 17-1A (34) and B72.3 (35,36) as well as others (37-42), should greatly reduce or eliminate such a response; this appears to have occurred in the first clinical trial using recombinant-chimeric 17-1A (43). In these studies, we investigated dose fractionation of 131 Ilabeled MAb B72.3 (44) in the human LS-174T colon tumor xenograft model (45). We believe that the LS-174T model is appropriate, since it mimics the heterogeneity of tumor cell expression of the TAG-72 antigen found in many human tumors. We fractionated otherwise lethal radiolabeled MAb doses and analyzed their effect on tumor growth. We also obtained biopsy specimens from tumor sites and analyzed them histologically. Moreover, to determine the degree of target antigen expression, we performed immunohistochemical assays on those tumors that escaped the MAb B72.3 therapy. In addition, we analyzed a variety of organs, including bone marrow, for toxic effects.
administration, six of 10 experimental mice had died. Thus, by the end of the 6-week observation period after administration of 600 jiCi of I3lI-labeled B72.3 IgG, two mice had no evidence of tumor growth, two mice-had tumors that had grown, and six mice had died (fig. 1, table 1). Dose Fractionation (2x) of Radiolabeled MAb
response was similar in the spleen, where erythroid, myeloid, lymphoid, and megakaryocytic precursors were rarely found. Subsequent studies in our laboratory (unpublished data) also showed that while administration of a single dose of 500,600, or 700 M,Ci of' 3 ' I-B72.3 IgG will reduce the growth of the LS-174T xenograft, 40% (four of 10), 40% (four of 10), and 70% (seven of 10) of the mice died, respectively, within the observation period of 36 days after MAb administration. The results of administration of a single dose of 600 |xCi of I31 I-B72.3 IgG to mice bearing the LS-174T xenograft are shown in figure 1. In the first 2 weeks after MAb administration, tumors in eight of 10 treated animals grew very little, if at all; this finding was in marked contrast to that in the 10 control mice receiving unlabeled antibody (fig. 1). However, by 4 weeks after MAb
Downloaded from http://jnci.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on March 15, 2015
Figure 1. Effect of administration of a single 600-jj.Ci dose of m I-B72.3 IgG on growth of LS-174T xenograft in athymic mice. Each line represents growth of tumor in an individual mouse. Open triangles represent results obtained from 10 mice that received an equal dose of unlabeled B72.3 IgG. Arrow represents time of MAb administration, which is 1 wk after tumor implantation. Cross represents death of mouse.
To obtain more efficient tumor killing with less toxic effects to the host, we undertook experiments in which the radiolabeled MAb was administered in two dose fractions. The results, shown in figure 2A-D and table 1, demonstrate the effects of the following two dose fractions of l31I-B72.3 IgG given 1 week apart on tumor growth in mice bearing the LS-174T xenograft: 200 jiCi (400 jxCi total), 300 jxCi (600 jiCi total), 400 n,Ci (800 [iCi total), or 500 (iCi (1,000 uXTi total), respectively. As figure 2A shows, there was little or no tumor growth for 4 weeks following the two 200-u,Ci MAb doses; however, most of the tumors were growing at the end of the observation period. There were no deaths due to toxic effects in this group. Ten mice bearing LS-174T xenografts received 300 |xCi of 131 I-B72.3 IgG at 1 week after tumor implantation and again at 2 weeks after tumor implantation. At the end of the observation period, three of these mice had no evidence of tumor growth, while another four mice had minimal tumor growth beginning at approximately 5 weeks after MAb administration (fig. 2B); one xenograft escaped the therapy relatively early. Only one mouse in this group died. Differences in responses among mice within the same group could not be attributed to differences in the actual amount of radiolabeled MAb administered, since mice were monitored by whole-body counts following MAb administration, and there was no correlation between the minor differences in total radioactive MAb received and antitumor responses. The results obtained with 10 mice receiving two weekly 400-|iCi injections of radiolabeled B72.3, for a total of 800 n.Ci, are shown in figure 2C. Two of these mice showed no evidence of tumor growth through the observation period. However, eight mice in this group died.
Table 1. Effect of dose fractionation of radiolabeled MAb B72.3 IgG on antitumor activity and toxicity Therapeutic agent I3I
I-B72.3 IgG I-B72.3 IgG
131
I3I
I-B72.3 IgG
Unlabeled B72.3 IgG Unlabeled control MAb BL-3 l3l I-BL-3 5-FUt
Dose or dose fractions CiCi) 600 200,200 300,300 400,400 500,500 150, 150, 150 200, 200, 200 250, 250, 250 300, 300, 300 0 0 300,300 0
Total dose OiCi)
No. of mice
No. of mice showing antitumor effect (%)*
No. of mice with NETG (%)t
No. of mice that died (%)
600 400 600 800 1,000 450 600 750 900 0
10 10 10 10 10 14 14 13 13
3(30) 10(100) 9(90) 2(20) 1(10) 12(86) 13 (93) 13(100) 12 (92)
2(20) 1(10) 3(30) 2(20) 0 3(21) 8(57) 6(46) 10(77)
23 8 7 44
1(4) 2(25) 4(57) 0
1(4) 2(25) 0 0
6(60) 0 1(10) 8(80) 9(90) 0 1(7) 0 1 (8) 0 0
0 600 0
0 0
* Antitumor effect is deflned as a
Numerous monoclonal antibodies (MAbs) to human solid tumors have been developed and are being evaluated to determine their potential use in the treatment of human carcinomas. Several of these MAbs have been coupled to radionuclides and have-been used in clinical trials for diagnostic tumor targeting. Phase I therapeutic trials with some of these MAbs have recently been initiated {1-11). As a prerequisite for most therapeutic clinical trials and for a Vol. 82, No. 9, May 2, 1990
Downloaded from http://jnci.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on March 15, 2015
Monoclonal antibody (MAb) B72.3 IgG was radiolabeled with 131I and administered to female athymic NCr-nu mice bearing the LS-174T human colon adenocarcinoma xenograft to determine if fractionation of MAb dose had any advantage in tumor therapy. In the LS-174T xenograft, only approximately 30%-60% of tumor cells express the B72.3-reactive TAG-72 antigen. The LS-174T xenograft was used to reflect the heterogeneity of the TAG-72 antigen often seen in biopsy specimens from patients. In contrast to a single 600-JJLG dose of 131I-B72.3 IgG where 60% of the animals died from toxic effects, two 300-y.Ci doses of 131I-B72.3 IgG (total of 600 ^Ci) reduced or eliminated tumor growth in 90% of mice, with only 10% of the animals dying from toxic effects. Dose fractionation even permitted escalation of the dose to three doses (each 1 wk apart) of 300 u,Ci of 131I-B72.3 IgG (for a total of 900 (xCi), resulting in even more extensive tumor reduction or elimination and minimal toxic effects. The use of an isotype-matched control MAb revealed a nonspecific component to tumor growth retardation, but the use of the specific B72.3 IgG demonstrated a much greater therapeutic effect. Tumors that had escaped MAb therapy were analyzed for expression of the B72.3-reactive TAG-72 antigen with the use of the immunoperoxidase method; they were shown to have the same antigenic phenotype as the untreated tumors. We verified tumor elimination by killing the test animals after a 7-week observation period and performing histologic examination of tumor sites. We also monitored toxic effects by histologic examination of numerous organs, including bone marrow. These studies thus demonstrate the advantage of dose fractionation of a radiolabeled MAb for tumor therapy. We anticipate that the concept of dose fractionation can be practically applied in radioimmunotherapeutic clinical trials with the development and use of recombinant-chimeric MAbs and modified constructs. [J Nat! Cancer Inst 82:763-771,1990]
better understanding of how radiolabeled MAbs kill in vivo, several MAbs have been analyzed for efficacy in reducing tumor growth in athymic mice bearing transplanted human tumor xenografts (12-23); often these MAbs have been labeled with radionuclides such as 131I and 90 Y. Two conclusions can be drawn from most of these studies: (a) the rate of tumor growth is reduced following one administration of radiolabeled MAb; and (b) with increasing levels of radionuclide bound to MAbs, more efficient tumor cell killing occurs, but there is a concomitant increase in bone marrow toxicity that leads to the death of some animals. One way of potentially overcoming this marrow toxicity would be to fractionate the dose of the administered MAb. In a previous study (72), employing F(ab')2 fragments of l3l I-labeled, anti-carcinoembryonic antigen MAbs, doses of MAb that were fractionated at days 10,12,24, and 38 after tumor transplantation showed only a slight advantage over a single dose. In another study (24), each of two F(ab') 2 MAbs was given at a dose of 2 mCi at 3-day intervals three times; this regimen was shown to be as therapeutic as a single 2-mCi dose of l31I-labeled IgG. These data suggest that multiple doses of F(ab') 2 can be as tumoricidal as a single dose of an intact IgG and can be less toxic to the host. Ceriani and Blank (25) reported that a second administration of 500 |xCi of l3l I-MAb (for a total dose of 1,000 p-Ci) further diminished tumor growth compared with tumor growth in mice treated with one 500-|iCi 131I-MAb dose. Hyams et al. (26) observed no significant improvement in xenograft-treated mice when given two cycles of 120 u,Ci of ^ - l a b e l e d MAbs versus one cycle of 120 u,Ci of ^ - l a b e l e d MAbs. However, Hyams et al. reported 100% and 91.7% disease-related deaths, respectively, in these two groups. Thus, these two recent studies (25,26) compared multiple administrations with a single administration of radiolabeled MAbs, but in these studies, the groups receiving multiple administrations also received higher doses of radiola- .
Received November 6,1989; revised February 6, 1990; accepted February 20, 1990. J. Schlom, A. Molinolo, J. F. Simpson, K. Siler, M. Roselli, D. Colcher, Laboratory of Tumor Immunology and Biology, National Cancer Institute, Betbesda.MD. G. Hinkle, The Ohio State University, Columbus, OH. D. P. Houchens, Battelle Memorial Institute, Columbus, OH. We thank D. Milenic, R. Nines, and D. Hill for expert technical assistance, and E. Glatstein and A. Raubitschek for insightful discussions. 'Correspondence to: Jeffrey Schlom, Ph.D., Bldg. 10, Rm. 8B07, National Institutes of Health, Bethesda, MD 20892.
ARTICLES
763
Materials and Methods Monoclonal antibodies. MAb B72.3 was obtained by immunization of BALB/c mice with a membrane-enriched fraction of a human breast tumor metastasis to the liver. The generation, characterization, and reactivities of MAb B72.3 were described elsewhere (46,47). The IgG was purified as previously reported (48) from ascitic fluid by ammonium sulfate precipitation and ion-exchange and size-exclusion chromatography. BL-3 is a murine IgGl anti-idiotype MAb to a human B-cell lymphoma previously shown (49) to be negative for reactivity to a wide range of cells from murine and human tissues, including LS- 174T colon carcinoma cells. Iodination of MAb B72.3. MAb B72.3 was labeled with Na l 3 l l using the Iodogen method (50). The purified IgG (1.7 mg) was labeled with approximately 18 mCi of Na l 3 l l in a glass vial coated with 250 (xg of Iodogen. After a 20-minute incubation, the free I3I I was removed by gel filtration through a Sephadex G-25 column. The immunoreactivity of the labeled MAb B72.3 was tested in a solid-phase radioimmunoassay, using extracts of a human colon carcinoma xenograft (57). Cell line. The LS-174T cell line (CL188) (45) was obtained from the American Type Culture Collection, Rockville, MD. This cell line was established from a human mucinous adenocarcinoma of the colon. The cells were grown in Eagle's minimum essential medium supplemented with 1% nonessential amino acids (100 vaM), including 1% glutamine (200 mM), 10% heat-inactivated fetal calf serum, and gentamicin (50 u,g/mL).
764
Cells were passaged weekly at a 1:10 dilution. They grew as a monolayer and were harvested with the use of 0.1 % trypsin in 0.5 mAf edetic acid (EDTA). The cells were washed twice in serum-free minimum essential medium before they were injected into athymic mice. Tumor growth in athymic mice. Female athymic mice (NCrnu) were obtained from Simonsen Laboratories, Gilroy, CA, at 4-6 weeks of age. The mice were given a subcutaneous injection in the right flank of 1 x 106 cells (0.1 mL). Seven days after inoculation, the mice bearing rumors of approximately 0.5 cm in diameter were selected and divided into treatment groups. Groups of 10-15 mice each received an intraperitoneal injection of various amounts of I3I I-B72.3 IgG at weekly intervals. The interval of 7 days between doses was chosen because we (52) have shown that, at day 7, greater than 90% of intraperitoneally administered MAbs are cleared from the blood pool, thus reducing the potential for marrow toxicity and enabling the marrow to regenerate. For the fluorouracil (5-FU) study, 5-FU was injected intraperitoneally daily for 5 days. The tumor growth was measured in two diameters weekly with a precision caliper, until the mice were killed. Mice receiving one injection of the 600-^Ci dose of' 3 ' I-B72.3 IgG were held for 42 days after injection. Mice receiving two injections of radiolabeled MAb B72.3 were held for 35 days after the last injection. Mice receiving three injections of radiolabeled MAb B72.3 were held for .45 days after the last injection. Mice receiving two injections of radiolabeled MAb BL-3 were held for 43 days after the last injection. Results are expressed as tumor volume as described previously (75-78). The control groups, mice given an injection of phosphate-buffered saline or unlabeled MAb B72.3, were similarly analyzed. Histologic studies. The tumor or site of tumor inoculation, spleen, bone marrow, lung, small intestine, and liver were obtained from selected animals. They were fixed in 10% buffered Formalin, routinely processed, and embedded in paraffin. Fivemicron sections were obtained from the paraffin blocks and stained with hematoxylin and eosin for examination by light microscopy. Blood smears were also obtained and stained with Wright's stain. Reactivity of the various tissues with biotinylated MAb B72.3 was determined using a modification of the avidinbiotin-peroxidase complex (ABC) method (Vectastain ABC kit; Vector Laboratories, Inc., Burlingame, CA) (53). The test concentration of MAb B72.3 was 80 u,g/mL; an isotype-identical control antibody, MOPC-21 (54), was used at an identical concentration on parallel sections; no reactivity was observed with the control MAb.
Results Single-Dose Administration of Radiolabeled MAb We (23) previously showed that administration of a single dose of either 300 or 500 |xCi of I3I I-B72.3 IgG to athymic mice bearing LS-174T colon carcinoma xenografts reduced tumor growth for 17 days after MAb administration. The weight of tumors in the controls was up to 3.7 times the weight of tumors in mice receiving 500 u.Ci. However, that same study also demonstrated that at 17 days after injection of the 500-jiCi dose of MAb, there was almost complete aplasia of the host bone marrow, with less than 5% of the marrow cellularity remaining. In addition, the Journal of the National Cancer Institute
Downloaded from http://jnci.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on March 15, 2015
beled MAbs. To our knowledge, no systematic comparison of various equivalent fractionated doses has been undertaken in which radiolabeled MAbs were used for tumor therapy. In the studies reported here, we have chosen MAb B72.3 IgG to conduct dose-fractionation studies. Radiolabeled B72.3 IgG has been administered intravenously to over 500 patients and has demonstrated efficient rumor targeting in approximately 70% of carcinomas of the colon and ovary (27-32). Recent studies involving intraperitoneal administration of l3lI-labeled MAb B72.3 have also demonstrated efficient tumor targeting (33). l3l I-MAb B72.3 is now being used in a phase I clinical trial (77). In the consideration of any MAb therapy using multiple dose scheduling in patients, a major concern is the development of human anti-mouse antibodies (HAMAs). The construction of recombinant-chimeric MAbs, such as 17-1A (34) and B72.3 (35,36) as well as others (37-42), should greatly reduce or eliminate such a response; this appears to have occurred in the first clinical trial using recombinant-chimeric 17-1A (43). In these studies, we investigated dose fractionation of 131 Ilabeled MAb B72.3 (44) in the human LS-174T colon tumor xenograft model (45). We believe that the LS-174T model is appropriate, since it mimics the heterogeneity of tumor cell expression of the TAG-72 antigen found in many human tumors. We fractionated otherwise lethal radiolabeled MAb doses and analyzed their effect on tumor growth. We also obtained biopsy specimens from tumor sites and analyzed them histologically. Moreover, to determine the degree of target antigen expression, we performed immunohistochemical assays on those tumors that escaped the MAb B72.3 therapy. In addition, we analyzed a variety of organs, including bone marrow, for toxic effects.
administration, six of 10 experimental mice had died. Thus, by the end of the 6-week observation period after administration of 600 jiCi of I3lI-labeled B72.3 IgG, two mice had no evidence of tumor growth, two mice-had tumors that had grown, and six mice had died (fig. 1, table 1). Dose Fractionation (2x) of Radiolabeled MAb
response was similar in the spleen, where erythroid, myeloid, lymphoid, and megakaryocytic precursors were rarely found. Subsequent studies in our laboratory (unpublished data) also showed that while administration of a single dose of 500,600, or 700 M,Ci of' 3 ' I-B72.3 IgG will reduce the growth of the LS-174T xenograft, 40% (four of 10), 40% (four of 10), and 70% (seven of 10) of the mice died, respectively, within the observation period of 36 days after MAb administration. The results of administration of a single dose of 600 |xCi of I31 I-B72.3 IgG to mice bearing the LS-174T xenograft are shown in figure 1. In the first 2 weeks after MAb administration, tumors in eight of 10 treated animals grew very little, if at all; this finding was in marked contrast to that in the 10 control mice receiving unlabeled antibody (fig. 1). However, by 4 weeks after MAb
Downloaded from http://jnci.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on March 15, 2015
Figure 1. Effect of administration of a single 600-jj.Ci dose of m I-B72.3 IgG on growth of LS-174T xenograft in athymic mice. Each line represents growth of tumor in an individual mouse. Open triangles represent results obtained from 10 mice that received an equal dose of unlabeled B72.3 IgG. Arrow represents time of MAb administration, which is 1 wk after tumor implantation. Cross represents death of mouse.
To obtain more efficient tumor killing with less toxic effects to the host, we undertook experiments in which the radiolabeled MAb was administered in two dose fractions. The results, shown in figure 2A-D and table 1, demonstrate the effects of the following two dose fractions of l31I-B72.3 IgG given 1 week apart on tumor growth in mice bearing the LS-174T xenograft: 200 jiCi (400 jxCi total), 300 jxCi (600 jiCi total), 400 n,Ci (800 [iCi total), or 500 (iCi (1,000 uXTi total), respectively. As figure 2A shows, there was little or no tumor growth for 4 weeks following the two 200-u,Ci MAb doses; however, most of the tumors were growing at the end of the observation period. There were no deaths due to toxic effects in this group. Ten mice bearing LS-174T xenografts received 300 |xCi of 131 I-B72.3 IgG at 1 week after tumor implantation and again at 2 weeks after tumor implantation. At the end of the observation period, three of these mice had no evidence of tumor growth, while another four mice had minimal tumor growth beginning at approximately 5 weeks after MAb administration (fig. 2B); one xenograft escaped the therapy relatively early. Only one mouse in this group died. Differences in responses among mice within the same group could not be attributed to differences in the actual amount of radiolabeled MAb administered, since mice were monitored by whole-body counts following MAb administration, and there was no correlation between the minor differences in total radioactive MAb received and antitumor responses. The results obtained with 10 mice receiving two weekly 400-|iCi injections of radiolabeled B72.3, for a total of 800 n.Ci, are shown in figure 2C. Two of these mice showed no evidence of tumor growth through the observation period. However, eight mice in this group died.
Table 1. Effect of dose fractionation of radiolabeled MAb B72.3 IgG on antitumor activity and toxicity Therapeutic agent I3I
I-B72.3 IgG I-B72.3 IgG
131
I3I
I-B72.3 IgG
Unlabeled B72.3 IgG Unlabeled control MAb BL-3 l3l I-BL-3 5-FUt
Dose or dose fractions CiCi) 600 200,200 300,300 400,400 500,500 150, 150, 150 200, 200, 200 250, 250, 250 300, 300, 300 0 0 300,300 0
Total dose OiCi)
No. of mice
No. of mice showing antitumor effect (%)*
No. of mice with NETG (%)t
No. of mice that died (%)
600 400 600 800 1,000 450 600 750 900 0
10 10 10 10 10 14 14 13 13
3(30) 10(100) 9(90) 2(20) 1(10) 12(86) 13 (93) 13(100) 12 (92)
2(20) 1(10) 3(30) 2(20) 0 3(21) 8(57) 6(46) 10(77)
23 8 7 44
1(4) 2(25) 4(57) 0
1(4) 2(25) 0 0
6(60) 0 1(10) 8(80) 9(90) 0 1(7) 0 1 (8) 0 0
0 600 0
0 0
* Antitumor effect is deflned as a
