Growth of Human Lung Tumor in the Brain of the Nude Rat As a Model to Evaluate Antitumor Agent Delivery Across the Blood-Brain Barrier1

[CANCER RESEARCH 45, 2827-2833, June 1985)

Growth of Human Lung Tumor in the Brain of the Nude Rat as a Model to Evaluate Antitumor Agent Delivery across the Blood-Brain Barrier1

Edward A. Neuwelt,2 Eugene P. Frenkel, Anthony N. D'Agostino, Desmond N. Carney, John D. Minna, Peggy A. Barnett, and Christopher I. McCormick

Department ol Surgery. Division of Neurosurgery, Oregon Health Sciences University, Portland, Oregon 97201 [E. A. N., P. A. B., C. I. M.]; Department of Internal Medicine, Division of Hematology -Oncology, University of Texas Health Science Center at Dallas, Dallas, Texas 75235 [E. P. F.¡;Department of Pathology, Good Samaritan Hospital and Medical Center, Portland, Oregon 97210 [A. N. D.]; and NCI-Navy Medical Oncology Branch, Division of Cancer Treatment/NIH, Naval Hospital, Bethesda, Maryland 20814 [D. N. C., J. D. M.]

ABSTRACT human tumors as xenografts. For instance, intracerebral growth of human glioma in immunosuppressed rats results in a highly We have developed a brain tumor model in the nude rat utilizing permeable tumor (4) unlike the situation clinically (6, 7) or in a NCI-N417D human small cell carcinoma of the lung grown both virally induced animal glioma model (8, 9). intracerebrally and s.c. The median latency period from the time The introduction of the athymic nude mouse (10) offered a of intracerebral tumor inoculation to the onset of neurological new opportunity to study the growth and behavior of human symptoms is 13 days with an intracerebral tumor take rate of tumor cells in vivo under controlled conditions (11). A variety of 91% (29 of 32). The median survival is 13 days, and all animals human tumors have been established and serially passed in the were dead by Day 26. The tumor is discrete, well circumscribed, athymic mouse (12). The histology and responsiveness for most with occasional leptomeningeal spread and with minimal evi tumors in the nude mouse are similar to those in humans, and dence of surrounding cerebral edema. Intracerebrally, this tumor validation of this model for the study of a variety of biological is usually impermeable to Evan's blue:albumin (M, 68,500) but features of human tumors has been achieved (13). Tumors so not fluorescein (M, 376). Although variable, the intracerebral grown show a differential sensitivity to cytoreductive chemo- tumor is less permeable to methotrexate than is the same tumor therapeutic agents that reflects the clinical effectiveness of these grown s.c. in the same animal (P < 0.005). The intraarterial and agents in the original patient. For instance, Giovanella et al. (14) i.v. routes of methotrexate administration in the presence and evaluated the efficacy of 9 chemotherapeutic agents on 3 types absence of blood-brain barrier opening were evaluated. Drug of human tumor xenografts implanted in the nude mouse (breast delivery to the intracerebral tumor and ipsilateral brain was cancer, colon cancer, and melanoma); tumor regressions were significantly (P < 0.025) greater when the methotrexate was consistent with the clinical experience in patients except for 2 given intraarterially and was significantly (P < 0.0025) increased false-positive and one false-negative values. Histological evalu after osmotic blood-brain barrier opening. After barrier opening, ation of human tumor xenografts has been reported to maintain methotrexate concentration was enhanced 3- to 4-fold in tumor morphological fidelity with the parent tumor (13). In addition, and 10- to 20-fold in brain around tumor. Thus, the nude rat several reports (15-17) have shown that xenografts of human provides a model to investigate the biology and therapeutic tumors can be consistently transplanted intracerebrally into the responsiveness of human small cell carcinoma of the lung grown nude mouse, and Chambers ef al. (18) have reported the intra- intracerebrally where it develops a blood-tumor barrier similar to cranial growth of cells from small cell carcinoma of the lung in that seen in humans. This model further provides the unique nude mice. One possible disadvantage of the nude mouse is that opportunity to investigate the role of osmotic blood-brain barrier human tumors which have a high propensity to metastasize do opening in the treatment of a tumor which is sensitive to chem- not metastasize in nude animals. otherapeutic agents and for which tumor-specific monoclonal Unfortunately, the small size of the nude mouse makes it a antibodies are available. very difficult model for the study of complex multimodality ther apeutic approaches to ¡ntracerebral tumors. The ideal model would permit human xenografts to be examined at both intra INTRODUCTION cerebral and systemic sites, and the intracerebral site should develop the biological characteristics of spontaneous primary or Although Greene (1, 2) reported heterotransplantation of hu metastatic tumors, i.e., to demonstrate at least a partial BBB3 man brain tumors in 1951, attempts to implant and grow spon (8, 9, 19, 20). In addition, animal size should permit a variety of taneous human tumors intracerebrally in laboratory animals have therapeutic manipulations, including i.a. (9, 21-23) infusions, been inconsistent (3) and generally unsuccessful. Neither condi opening of the BBB (6-8, 24, 25), and applicability for studies of tioning with immunosuppressive agents (4) nor utilization of the "targeted" therapy, such as with monoclonal antibodies (26-29). subarachnoid space of the animal as a growth chamber (5) The nude rat provided the appropriate structural requirements, provides a satisfactory model for the study of gliomas or other and human small cell (oat cell) xenografts provided the accept 1This work was supported by the Oregon Medical Research Foundation, able tumor for the development of a therapeutic model. The National Cancer Institute Grant CA31770, Good Samaritan Hospital Foundation, nude rat has immunological features like those of the nude NIH Grant CA 23115, Meadows Foundation, and the Southwestern Medical Foun- dation-Kinsler Williamson Brown Fund. mouse (30, 31). Thus, the congenitally athymic nude rat accepts 2 To whom requests for reprints should be addressed, at Division of Neurosur allografts and xenografts, and their splenic lymphocytes respond gery (L472), Oregon Health Sciences University, 3181 S.W. Sam Jackson Park Road, Portland, OR 97201. 3 The abbreviations used are: BBB, blood-brain barrier; MTX, methotrexate; Received 8/6/84; revised 11/14/84. 2/28/85; accepted 3/11/85. CNS, central nervous system; i.a., intraarterial(ly); i.e. intracarotid.

CANCER RESEARCH VOL. 45 JUNE 1985 2827

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1985 American Association for Cancer Research. NUDE RAT BRAIN TUMOR poorly to T-cell mitogens. Unlike the nude mouse, the nude rat BBB Opening in the Nude Rat. BBB opening in the nude rat was is less susceptible to infections and is thereby easier to maintain. performed and graded as previously described (8). Five min before BBB The nude rat does not provide as good a growth medium for opening, Evans blue (2%, 2 ml/kg, i.v.) and/or fluorescein (10%, 0.12 to tumor xenografts (31) as the mouse, perhaps as the result of a 0.25 ml, i.v.) was administered. MTX Delivery to Human Tumor in Nude Rats. Animals symptomatic higher level of natural killer cells (32, 33). However, a variety of from intracerebral tumor were randomly divided into 4 groups: Group I, different human cancers (e.g., carcinoma of the breast, gastroin i.v. MTX after i.e. 0.9% NaCI solution (saline) (BBB intact); Group II, i.e. testinal tract, melanoma, and rhabdomyosarcoma) have been MTX after i.e. saline (BBB intact): Group III, i.v. MTX after i.e. mannitol grown in the nude rat (34, 35). (BBB opened); or Group IV, i.e. MTX after i.e. mannitol (BBB opened). In The present studies focused on the small cell carcinoma (oat all groups, MTX (4 mg/kg) was administered immediately after i.e. saline cell) of the humarj lung, since this carcinoma is a common tumor or mannitol over 1 min. Thirty min after MTX administration, a serum which frequently metastasizes to the CNS and is responsive to sample was obtained, and the animals were sacrificed with an overdose a variety of chemotherapeutic agents. In addition, a number of of sodium pentobarbital. Tumor, brain around tumor (1 to 2 mm of brain tumor-specific monoclonal antibodies have been developed to immediately surrounding tumor), brain distant to tumor (coronal section this neoplasm (36,37). The human small cell tumors were grown of the occipital lobe) in the perfused hemisphere as well as in the contralateral brain (coronal section of occipital lobe), s.c. tumor, and in the nude rat, and histology, survival, and the delivery of a other organs and tissues were obtained for MTX determinations. MTX chemotherapeutic agent (MTX) were examined. These studies content in serum and tissue was determined by radioimmunoassay as have shown the feasibility of the nude rat as a xenograft model previously described (22). to examine this common human tumor and to evaluate strategies The effect of the intracerebral implantation technique on BBB integrity of therapy by comparing the delivery of drugs to the tumor in was examined in rats inoculated with lethally irradiated NCI-N417D cells the brain with the tumor at s.c. sites. In particular, this model (8000 to 9000 rads, using a 137Cssource). On Day 14 after intracerebral allows an evaluation of the role of BBB modification in anti-tumor inoculation, Evans blue, fluorescein, and MTX were administered i.v. as agent delivery. described above. Statistical Analysis. MTX levels are expressed in ng/g of tissue or ng/ml of serum as the mean ±SE. Comparisons of the log concentration MATERIALS AND METHODS of tissues in the irradiated inoculum control group (Table 1) and within the 4 treatment groups (Table 2) were made by the one-way analysis of variance followed by the Neuman-Keuls multiple comparisons when the Nude Rat Colony. The animals utilized were the athymic nude rat overall difference was statistically significant. The 2-factor analysis of (rnu/rnu) which were bred from an outbred pair (rnu/+) furnished from HaríanSprague Dawley (Indianapolis, IN). The stock and breeding ani variance was utilized to evaluate the effect of route of MTX administration mals were kept in laminar flow contaminant hoods. Homozygote males and/or osmotic BBB opening on delivery to brain tissues (Table 2, Groups were paired with hétérozygote(rnu¡+)females which resulted in 50% I to IV) (39, 40). homozygote nude (rnu/rnu) athymic animals. All animals received steri lized fat-enriched food and vitamin-supplemented water. Water and RESULTS bedding were changed twice weekly under sterile conditions. Animals were examined daily. Survival and Onset of Neurological Symptoms. For the study Human Tumor Cell Line. The human tumor cell line used was the of survival characteristics, 4 x 105 human small cell carcinoma NCI-N417D (38) which originated in the laboratory of Dr. John D. Minna. of the lung cells (serially passaged in tissue culture) were injected Cells were grown in 20 ml of RPMI Medium 1640 supplemented with 10% heat-inactivated fetal calf serum, penicillin (10,000 units/ml), strep intracerebrally into the right parietal lobe of 32 nude rats. The tomycin (10,000 M9/ml), arid gentamicin (50 ng/m\). Complete medium median survival was 13 days (range, 10 to 25 days), and all was filtered (0.2-Mm Millipore) prior to use. All cells were maintained in animals were dead by Day 25 (Chart 14). Because of the large 5% CO2 at 37CC and passaged in culture twice weekly. The NCI-N417D variation in time from death to autopsy, intracerebral tumor human tumor cells are free-floating cell suspensions that are easily confirmation was often not available in these rats. harvested by centrifugation at 250 x g (1200 rpm) for 3 to 5 min. Cell Of 32 additional rats given injections of intracerebral NCI- suspensions for inoculation were made by resuspending the cell pellet in N417D human tumor cells, the median latency period from medium to achieve a final concentration with a packed cell volume of 20 intracerebral inoculation to the development of neurological ±1% as measured in a microhematocrit tube. Cell viability was evaluated symptoms was also 13 days (range, 9 to 26 days) (Chart 1B). by trypan blue exclusion; viability was greater than 85%. Symptoms identifying the presence of tumor included cachexia, Implantation of Human Tumor Cells into the Nude Rat. Implantation s.c. was achieved by the s.c. injection into the right flank of 750 to 1000 weight loss, and dehydration. Animals died within 24 h of evident M! of tumor cell suspension with a packed cell volume of 20 ± 1% symptoms. Of these rats, 91% (29 of 32) developed verified (approximately 20 to 27 x 106 cells). Intracerebral implantation of tumor intracerebral tumors. In the remaining 3 animals, intracerebral cells was performed after the rats were anesthetized by i.p. injection of tumor could not be confirmed by visual inspection, but histology sodium pentobarbital (50 mg/kg). The head was shaved, sterilized with was not performed. iodine solution, and immobilized in a rat stereotaxic frame (David Kopf Pathological Characteristics. When the animals become neu- Instruments, Tunjunga, ÇA).A 2-cm midline incision was made to expose rologically symptomatic or at expiration, the tumors exhibited a the frontal bone on the right, where a 2-mm burrhole with a Stryker drill mass effect on the brain. The tumors were approximately 5 to 7 was placed just posterior to the coronal suture line and 4 mm lateral to mm in diameter with occasional leptomeningeal spread. No ani the sagittal suture. A micromanipulator (Brinkman Instruments, West- bury, NY) equipped with a 27-gauge needle and a 100-iil syringe was mal had multiple tumors. The tumors were usually subcortical used to inject 15 ^l of tumor cell suspension with a packed cell volume and often extended to the ependymal wall of the lateral ventricle. of 20 ±1% over 2 to 3 min at a depth of 1.5 mm. This concentration of The intracerebral tumors were round and had neovascularity. cells equals approximately 4 x 105 tumor cells. The wound was closed The tumor margin was clearly defined, and little surrounding in a single layer. cerebral edema was present (Fig. 1-4). Systemic spread (other

CANCER RESEARCH VOL. 45 JUNE 1985 2828

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1985 American Association for Cancer Research. NUDE RAT BRAIN TUMOR

A100"5> confirmed histologically. This led to the use of i.v. fluorescein (M, 376) when symptoms were evident, thus providing clearly de marcated, easily visible tumors identified by the yellow stain of 80 fluorescein (Fig. 2A). The relative integrity of the BBB was 'E 1 n=32 evaluated by the concurrent administration of i.v. fluorescein (M, W60§ 376) and Evans blue (Evans blue:albumin, M, 68,500). As shown in Fig. 2A, the intracerebral tumor usually stained with the low- 40 molecular-weight moiety fluorescein, but not with the high-mo O; lecular-weight Evans blue:albumin, whereas both markers gen 20 erally penetrated s.c. tumor (Fig. 26). By contrast, when these same molecular markers (fluorescein and Evans blue:albumin) were given i.v. and osmotic barrier opening of the tumor-bearing 2 6 10 14 18 22 26 hemisphere was performed, both Evans blue and fluorescein Days After Intracerebral Inoculation penetrated the tumor as well as the surrounding brain. A green discoloration (due to the combination of yellow and blue) of these tissues was seen in the area of brain where barrier opening was performed (Fig. 2, C and D). When only Evans blue was given i.v. at the time of barrier opening, the tumor and the brain around the tumor were stained blue. n=32 MTX Delivery to Intracerebral and s.c. Tumor. To determine the effect of the inoculation technique on BBB integrity, lethally irradiated tumor inoculum was implanted intracerebrally. There was no evident staining with either i.v.-administered Evans blue or fluorescein. Following i.v. MTX administration, MTX delivery to the area of inoculation resulted in levels ranging between 0.5 and 3% of serum levels, which was consistent with MTX levels 2 6 10 14 18 22 26 in the contralateral hemisphere of all inoculated animals (Table 1). There was a statistical difference (P < 0.05) in MTX concen Days After Intracerebral Inoculation tration between the inoculation site (94 ±54) and brain distant Chart 1. A, survival curve of 32 rats inoculated intracerebrallywith NCI-N417D to the inoculation (16 ±3) site and contralateral brain (17 ±3). humansmallcell carcinomaof the lung and followed until death. The medianlatency period from intracerebral inoculation to death was 13 days (range, 10 to 25 days), Despite an equivalent dose of MTX, the serum drug levels in and all animals were dead by Day 25. B, curve illustrating the development of tumor-bearing animals were higher than in animals receiving neurologicalsymptoms after intracerebralinoculationwith NCI-N417Dhuman small irradiated tumor cells, which probably reflects cachexia and cell carcinoma of the lung cells. The median latency period from intracerebral volume depletion of tumor-bearing animals. inoculation to onset of symptoms was 13 days (range, 9 to 26 days), and all animals had symptoms by Day 26. Ninety-one % (29 of 32) had confirmed Drug delivery studies were begun when the animals mani intracerebral tumor; the remaining 3 could not be confirmed by visual inspection, fested symptoms from their intracerebral tumor. Rats with both and histology was not performed. intracerebral and s.c. tumor were given MTX either i.v. or ¡.a. after either i.e. saline (Groups I and II) or mannitol infusion than the simultaneously inoculated s.c. flank tumor) has not been (Groups III and IV). After i.v. MTX, the intracerebral tumor in observed. Microscopic examination of the tumor cells shows Group I contained significantly (P < 0.005) less MTX than the them to be slightly larger than the characteristic human lung s.c. tumor, suggesting limited entry of drug at the intracerebral small cell tumor as identified in clinical material; mitoses were site (Table 2). The level of MTX in intracerebral tumor, although very frequent (Fig. ÌB). variable, was approximately 8% of that in serum, whereas the In the initial studies, the same size of tumor inoculum was level in the s.c. tumor was approximately 20% of the serum level used intracerebrally and s.c. However, at the time the animals at the time of sacrifice. The intracerebral tumor, however, con became symptomatic from their intracerebral tumor, no s.c. tained significantly more MTX than the brain around tumor, brain tumor was detected when the tumor inoculum was the same at distant to tumor, and contralateral brain (P < 0.005). both sites. When s.c. tumor inoculum was increased 50- to 65- fold, the diameter of the s.c. tumor was 2 to 3 cm at the point when the animals developed neurological symptoms from the Table1 MTX concentration in the intracerebral inoculation site of lethally irradiated small intracerebral tumor. The s.c. tumor was 4 to 5 times larger than cell carcinoma tumor cells the intracerebral tumor. In addition, in almost half the animals, MTX (ng/g)" the s.c. tumors were partially cystic and contained serosangui- 94 ±54" Inoculation site neous fluid. Brain around inoculation site 25 ±6 Tumor Staining with Fluorescein and Evans Blue:Albumin. Brain distant to inoculation site 16±3 Brain contralateral to inoculation site 17±3 During the preliminary studies, it was often difficult to identify Serum (ng/ml) 3033 ±33 discrete tumors at the time of death, even when Evans blue "MTX, 4 mg/kg, was administered i.v. on Day 14 after the intracerebral inoculation. Animals were sacrificed 30 min after MTX administration (n = 3). (which binds tightly to albumin, thereby identifying barrier integ " Mean±SE. rity) had been given. Thus, there were several animals in which 0Significantly (P < 0.05) greater than brain distant to the inoculation site and the intracerebral tumor was missed by visual inspection, but brain contralateral to the inoculation site.

CANCER RESEARCH VOL. 45 JUNE 1985 2829

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1985 American Association for Cancer Research. NUDE RAT BRAIN TUMOR

Table 2 MTX concentration in small cell lung carcinoma intracerebral and s.c. xenographs in the nude rat following i.v. or i.e. MTX administration with and without osmotic BBB opening MTX, 4 mg/kg, was administered i.v. or i.e. to nude rats symptomatic with NCI-N417D human small cell lung carcinoma intracerebral xenograft immediately after i.e. saline or mannitol (BBB opened). Animals were sacrificed 30 min after MTX administration. All animals infused with mannitol had 2+ or 3+ Evan's blueialbumin staining. Nonneural tissue values for all groups were combined since there were no statistical differences among the 4 groups. MTX (ng/g)a

Control barrier barrierGroupOpened

III (i.v. IV (i.e. MTX Group I (i.v. MTX Group II (i.e. MTX MTX after manni- after mannitol, n = 8)IntracerebralXenografts after saline, n = 6) after saline, n = 9) toi, n = 8)2275

79138±116* 923 ±299 843 ± TumorBrain ±6482631 tumorBrainaround 7897±34 316 ±99 784 ± ±8983725 tumorBraindistant to 8873±26 123 ±20 875 ± ±1377206 tumorSerumcontralateral to ±166950±14 109 ±21 113 ±668663 (ng/ml)Nonneural ±19722681378 9556 ±1178 621 3 ± ±873

(n= tissues from all groups 31)s.c. tumorLiverKidneyMuscleSpleenHeartLung530 +832000 ±48091 93±1395627 ±58934 ±95738 ±891922 ±162Group a MTX deliveries to tumor, brain around tumor, and brain distant to tumor were significantly affected by the route of MTX administration (P < 0.025) and by osmotic BBB opening (P < 0.0025) (2-factor analysis of variance). For further details of statistical analysis, see "Results." 6 Mean ±SE.

When the MTX was given i.a. after i.e. saline infusion (Group in previous studies with the avian sarcoma virus-induced glioma II), there was a 2-fold increase of MTX delivery to the tumor and in the rat, we have demonstrated that, by combining i.a. drug brain around tumor as compared to Group I. Similarly, when administration with reversible osmotic BBB opening, the delivery MTX was given i.a. after i.e. mannitol (Group IV), there was a 3- of MTX to the tumor and the brain around the tumor was fold increase to tumor and brain around tumor compared to increased over that achieved by routine (I.e., i.v.) routes of drug Group III. Thus, route of MTX administration significantly (P < administration (8). However, the avian sarcoma virus-induced 0.025) altered drug uptake in tumor and surrounding brain. glioma model is not sensitive to most chemotherapeutic agents, and it is not amenable to an investigation of "targeted" therapeu Osmotic BBB opening, independent of route of administration, also significantly (P < 0.0025) increased drug delivery to tumor, tic studies, since monoclonal antibodies have not been developed brain around tumor, and brain distant to tumor. MTX adminis against these tumors. tered i.v. after osmotic opening (Group III) resulted in 2- to 9-fold The present studies utilized the immunologically athymic increases compared to Group I. When MTX was admnistered (nude) rat for the study of drug delivery to an intracerebral tumor. i.a. after osmotic opening (Group IV), there was increased deliv The rat was selected because its size permitted cannulation of ery of MTX, ranging from 2.5-fold in tumor to 30-fold in brain the carotid artery necessary for i.a. drug administration and an distant to tumor as compared with Group II. The greatest MTX examination of the effect of osmotic BBB opening on drug delivery to tumor and ipsilateral brain was obtained after i.a. delivery. Although Colston ef a/. (31) reported that the nude rat administration in the presence of BBB opening. By this route of was less able than the nude mouse to sustain a human tumor administration, the MTX content in the intracerebral tumor was xenograft, we have had little difficulty in growing either human double that in the tumor at the s.c. site. The MTX levels in the carcinoma of the breast or small cell carcinoma of the lung in s.c. tumor and serum were not statistically different in any of the this model. The present studies focused on the characterization groups. of the growth of human small cell carcinoma of lung in the nude Thus, less MTX entered this small cell carcinoma of lung grown rat both intracerebrally and s.c. intracerebrally in contrast to the tumor grown in s.c. tissues. This human small cell carcinoma of lung tumor model has Osmotic BBB opening combined with i.a. drug administration provided several very important observations. The characteris achieved drug levels in intracerebral tumor and surrounding brain tics of this tumor model make it very suitable for the evaluation equal to or greater than that found in the systemic tissues or of antitumor agent (chemotherapeutic drug or monoclonal anti s.c. tumor. body) delivery in the presence or absence of BBB opening. A most important finding demonstrated in this model is that tumor DISCUSSION inoculated and grown in the brain almost totally excludes (via the BBB) the uptake of a protein marker (Evans blue:albumin), The critical requirement of an adequate xenograft model for even though there is partial uptake of smaller-molecular-weight the evaluation of complex multimodality therapeutic approaches markers (i.e., MTX and fluorescein). Evidence that this is not a to malignant intracerebral tumors is easily evident. For instance, property of the tumor cells is the greater uptake of Evans

CANCER RESEARCH VOL. 45 JUNE 1985 2830

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1985 American Association for Cancer Research. NUDE RAT BRAIN TUMOR

after osmotic blood-brain barrier opening. Neurosurgery, 72: 662-671,1983. blue:albumin or the cytotoxic agent MIX to the same tumor 7. Neuwelt, E. A., Hill, S. A., and Frenkel, E. P. Osmotic blood-brain barrier grown at a s.c. site. As in most human primary and metastatic modification and combination chemotherapy: concurrent tumor regression in malignant brain tumors, permeability to drug was variable. As areas of barrier opening and progression in brain regions distant to barrier opening. Neurosurgery, 75: 362-366, 1984 reported by Gazdar et al. (38), the dose of tumor cells necessary 8. Neuwelt, E. A., Bamett, P., Bigner, D. D., and Frenkel, E. P. The effects of to produce a s.c. tumor was much greater than the dose nec adrenal cortical steroids and osmotic blood-brain barrier opening on metho- essary to produce an intracerebral tumor. The reason for this trexate delivery to gliomas in the rodent: the factor of the blood-brain barrier. Proc. Nati. Acad. Sci. USA, 79: 4420-4423, 1982. difference is not apparent. 9. Neuwelt, E. A., Barnett, P. A., and Frenkel, E. P. Chemotherapeutic agent In spite of these observations, the significance of the BBB in permeability to normal brain and delivery to avian sarcoma virus-induced brain tumors in the rodent: observations on problems of drug delivery. Neurosurgery, drug delivery to CNS tumors is still a matter of some controversy. 74:154-160,1984. Hiesiger ef al. (24) recently stated that the blood tumor barrier in 10. Pantelouris, E. M. Absence of a thymus in a mouse mutant. Nature (Lond.), C6 rat gliomas is already open and cannot be further opened 277:370-371, 1968. 11. Povlsen, C. 0., and Jacobsen, G. K. Chemotherapy of a human malignant osmotically, thus explaining why MIX entry into such tumor melanoma transplanted in the nude mouse. Cancer Res., 35: 2790-2796, does not increase after i.e. hyperosmolar mannitol. These obser 1975. vations are in contrast to our tumor model, which has a differ 12. Ovejera, A. A., and Houchens, D. P. Human tumor xenografts in athymic nude mice as a preclinical screen for anticancer agents. Semin. Oncol. 8: 386-393, ential permeability between the intracerebral and the s.c. tumor. 1981. Analogously, initial clinical trials of i.v.-administered "Tc-gluco- 13. Bullard, D. E., Schold, C., Jr., Bigner, S. H., and Bigner, D. D. Growth and heptanate and, at a different time point, 131l-tumor-specific mono Chemotherapeutic response in athymic mice of tumors from human glioma- derived cell lines. J. Neuropathol. Exp. Neural., 40: 410-427, 1981. clonal antibody resulted in positive scans with the low-molecular- 14. Giovanella, B. C., Stehlin, J. S.. Shepard, R. C., and Williams, L. J., Jr. weight "Tc-glucoheptanate but negative serial scans with the Correlation between response to chemotherapy of human tumors in patients monoclonal antibody over a 10-day period. Thus, this patient's and in nude mice. Cancer (Phila.), 52: 1146-1152, 1983. 15. Horton, B. C., Basler, G. A., and Shapiro, W. R. Xenograft of human malignant glioma, as with our tumor model, displays more permeability to glial tumors into brains of nude mice. J. Neuropathol. Exp. Neural., 40: 493- low-molecular-weight than to high-molecular-weight markers.4 In 511, 1981. addition, BBB opening increases the concentration of MTX deliv 16. Houchens, D. P., Ovejera, A. A., and Riblet, S. M. Human brain tumor xenografts in nude mice as a chemotherapy model. Eur. J. Cancer Clin. Oncol., ered to the intracerebral tumor in both this model and our virally 79:799-805,1983. induced glioma model (8, 9). 17. Slagel, D. E., Feola, J., Houchens, D. P., and Ovejera, A. A. Combined modality treatment using radiation and/or chemotherapy in an athymic nude mouse- It must be emphasized that the current studies only examine human medulloblastoma and glioblastoma xenograft model. Cancer Res., 42: the question of drug delivery as a method to characterize this 812-816, 1982. tumor, not drug efficacy. However, this model provides an excit 18. Chambers, W. F., Pettengill, O. S., and Sorenson, G. D. Intracranial growth of pulmonary small cell carcinoma cells in nude athymic mice. Exp. Cell Biol., 49: ing opportunity to evaluate therapeutic efficacy of osmotic BBB 90-97,1981. opening as a factor in therapy of cancers in the CNS, since 19. Groothuis, D. R., Fischer, J. M., Lapin, G., Bigner, D. D., and Vick, N. A. human small cell carcinoma of the lung is very sensitive to Permeability of different experimental brain tumor models to horseradish peroxidase. J. Neuropathol. Exp. Neurol., 41: 164-185, 1982. chemotherapy (41), and since these are tumors to which highly 20. Yamada, K., Hayakawa, T., Ushio, Y., Anta, N., Kato, A., and Mogami, H. specific monoclonal antibodies have been produced (36, 37). Regional blood flow and capillary permeability in the ethylnitrosourea-induced Because a steep dose-response curve seems to characterize rat glioma. J. Neurosurg., 55: 922-928,1981. 21. Neuwelt, E. A., Bamett. P. A., Glasberg, M., and Frenkel, E. P. Pharmacology most tumors (42) and often the proliferating edge of these tumors and neurotoxicity of c/s-platinum, bleomycin, 5-fluorouracil, and cyclophospha- has an intact BBB (43), the utilization of osmotic modification to mide administration following osmotic blood-brain barrier modification. Cancer Res., 43: 5278-5285,1983. maximize antitumor agent delivery, be it drug or monoclonal 22. Neuwelt, E. A., Frenkel, E. P., Rapoport, S. E., and Barnett, P. A. Effect of antibody,56 seems very appropriate. The relatively brief and osmotic blood-brain barrier disruption on methotrexate pharmacokinetics in consistent latency period from the time of tumor inoculation to the dog. Neurosurgery. 7: 36-43, 1980. 23. Neuwelt, E. A., Pagel, M., Bamett, P., Glasberg, M., and Frenkel, E. P. the onset of neurological symptoms further enhances the value Pharmacology and toxicity of intracarotid Adriamycin administration following of this model. osmotic blood-brain barrier modification. Cancer Res., 47: 4466-4470, 1981. 24. Hiesiger, E., Voorhies, R., Basler, G. A., Shaprio, W. R., Posner, J. B., and Blasberg, R. Capillary permeability of experimental rat brain tumor (C6) and REFERENCES cortex after intracarotid hyperosmolar mannitol as measured by quantitative autoradiography (QAR). Neurology, 33: 108, 1983. 1. Greene, H. S. N. The transplantation of tumors to the brains of heterologous 25. Rapoport, S. I., Fredericks, W. R., Ohno, K.. and Pettigrew, K. D. Quantitative species. Cancer Res., 11: 529-534,1951. aspects of reversible osmotic opening of the blood-brain barrier. Am. J. 2. Greene, H. S. N. The transplantation of human brain tumors to the brain of Physiol., 235: 421-431,1980. laboratory animals. Cancer Res., 73; 422-426, 1953. 26. Day, E. D., Lassiter, S., Woodhall, B., Mahaley, J. L., and Mahaley, M. S. The 3. Stromblad, L. G., Burn, A., Salford, L. G., and Stenevi, U. A model for localization of radioantibodies in human brain tumors. I. Preliminary exploration. xenotransplantation of human malignant astrocytomas into the brain of normal Cancer Res., 25: 773-778,1965. adult rats. Acta Neurochir., 65: 217-226, 1982. 27. Gilliland, D. G., Steplewski, Z., Collier, R. J., Mitchell, K. F., Chang, T. H., and 4. Saris, S. C., Bigner, S. H., and Bigner, D. D. Intracerebral transplantation of a Koprowski, J. Antibody-directed cytotoxic agents: use of monoclonal antibody human glioma line in immunosuppressed rats. J. Neurosurg., 60: 582-588, to direct the action of toxin A chains to colorectal carcinoma cells. Proc. Nati. 1984. Acad. Sci. USA, 77: 4539-4543,1980. 5. Wilson, C. B., and Marker, M. Intrathecal transplantation of human neural 28. Miller, R. A., Maloney, D. G., Warnke, R., and Levy, R. Treatment of B-cell tumors to the guinea pig. J. Nati. Cancer Inst., 47: 1229-1240, 1968. lymphoma with monoclonal anti-idiotype antibody. N. Engl. J. Med., 306: 517- 6. Neuwelt, E. A.. Balaban, E., Diehl, J., Hill, S., and Frenkel, E. Successful 522, 1982. treatment of primary central nervous system lymphomas with chemotherapy 29. Moshakis, V., Mcllinney, R. A. J., Raghaven, D., and Neville, A. M. Localization of human tumor xenografts after i.v. administration of radiolabelled monoclonal 4 A. Coakham, personal communication. antibodies. Br. J. Cancer. 44: 91-99,1981. SE. A. Neuwelt, C. I. McCormick, E. P. Frenkel, J. D. Minna, and P. Bamett. 30. Brooks, C. G., Webb, P. J., Robins, R. A., Robinson, G., Baldwin, R W., and Osmotic blood-brain barrier modification: monoclonal antibody, albumin, and meth- Festing, M. F. W. Studies on the immunobiology of rnu/rnu "nude" rats with otrexate delivery to CSF and brain, submitted for publication. congenital aplasia of the thymus. Eur J. Immunol., 70: 58-65, 1980. "E. A. Neuwelt, J. Minna, P. Barnett, C. McCormick, and E. Frenkel. Pharma- 31. Colston, M. J., Fieldsteel, A. H., and Dawson, P. J. Growth and regression of cokinetics of monoclonal antibody delivery after osmotic blood-brain barrier modi human tumor cell lines in congenially athymic (rnu/rnu) rats. J. Nati. Cancer fication, submitted for publication. Inst., 66: 843-847, 1981.

CANCER RESEARCH VOL. 45 JUNE 1985

2831

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1985 American Association for Cancer Research. NUDE RAT BRAIN TUMOR

32 De Jong, E . Steerenberg, P. A., Ursem. P. S., Osterhaus, A. D. M. E., Vos, J. and Krasnow, S. Methods for production of monoclonal antibodies with spec G.. and Ruitenberg, E. J. The athymic nude rat. III. Natural cell-mediated ificity for human lung cancer cells. In Vitro (Rockville), 17: 1058-1070,1981. cytotoxicity. Immunol. Immunopathol., 17: 163-172. 1980 38 Gazdar, A. F.. Carrey, D. N., Sims, H. L, and Simmons, A. Heterotransplan- 33. Reynolds. C. W., Timonen, T. T., Holden. H. T.. Hansen. C. T., and Herberman, tation of small-cell carcinoma of the lung into nude mice: comparison of R. B. Natural killer cell activity in the rat. Analysis of effector cell morphology intracranial and subcutaneous route. Int. J. Cancer, 28: 777-783,1981. and effects of Interferon on natural killer cell function in the athymic (nude) rat. 39 Winer, B. J. In: N. Garmezy, R. Solomon, L. Jones, and H. Stevenson (eds). Eur. J. Immunol., 12: 577-582, 1982. Statistical Principles in Experimental Design. New York: McGraw-Hill, 1971. 34. Davies, G., Duke. D., Grant, A. G., Kelly, S. A., and Hermon-Taylor, J. Growth 40 Zar, J. H. In: W. D. McElroy and C. P. Swanson (eds.). Biostatistical Analysis. of human digestive-tumour xenografts in athymic nude rats. Br. J. Cancer, 43: Englewood Cliffs. NJ: Prentice-Hall, Inc., 1974. 53-58,1981. 41 Meijstrom, E. S., Capizzi, R. L.. Rudnick, S. A., Krisch, M., Delaney, D.. Kahn, 35. Dawson, P. J.. Kluskens. L. F., Colston, M. J., and Fieldsteel. A. H. Transplan L., Lipper, S., and Carney, C. High-dose methotrexate in small cell lung cancer. tation of human malignant tumors to the athymic rat. Cancer (Phila.), 50:1151- Cancer (Phila.), 51: 1056-1061, 1983. 1154,1982. 42 Frei, E., Ill, and Canellos, G. P. A critical factor in cancer chemotherapy. Am. 36. Cuttitta, F., Rosen, S., Gazdar, A. F.. and Minna, J. D. Monoclonal antibodies J. Med., 69: 585-594. 1980. that demonstrate specificity for several types of human lung cancer. Proc. 43 Levin, V. A., Clancy, T. P., Ausman. J. I., and Rail, D. P. Uptake and distribution Nati Acad. Sci. USA, 78. 4591-4595. 1981. of (3HJmethotrexate by the murine ependymoblastoma. J. Nati. Cancer Inst., 37 Minna, J. D., Cuttitta, F., Rosen, S., Bunn, P. A., Carney. D. N., Gazdar, A. F., 48: 875-883. 1972.

Fig. 2. A, coronal section of a nude rat brain 17 days after 4 x 105 NCI-N417D human small cell lung carcinoma cells were implanted in the right hemisphere. Prior to sacrifice, Evans blue (2%, 2 ml/kg, i.v.) and fluorescein (10%, 0.12 ml, i.v.) were administered, followed by i.e. saline. Note the tumor is stained with fluorescein (M, 376) but not with Evans blue:albumin (M, 68,500). B, s.c. tumor in a nude rat 17 days after 27 x 10* NCI-N417D human small cell lung carcinoma cells were implanted. Fluorescein and Evans blue were given as in A prior to i.e. saline. Note that the tumor is stained by both Evans blue albumin and fluorescein. C and D, intact (C) and coronal (D) section of a nude rat brain 18 days after 4 x 105 NCI-N417D human small cell lung carcinoma cells were implanted in the right hemisphere. Prior to sacrifice, Evans blue (2%, 2 ml/kg, i.v.) and fluorescein (10%, 0.12 ml, i.v.) were administered, followed by mannitol infusion into the right internal carotid artery. The tumor and surrounding brain in the distribution of barrier opening are stained with Evans Wue:albumin and fluorescein (green due to blue and yellow).

CANCER RESEARCH VOL. 45 JUNE 1985 2832

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1985 American Association for Cancer Research. Fig. 1. A, brain tumor interface of nude rat brain 23 days after 4x10" NCI-N417D human small cell lung carcinoma cells were implanted in the right hemisphere. There are neovascularity, a sharp tumor edge, and little surrounding cerebral edema. H & E, x 100. B, high magnification of section of Fig. 1/1. The tumor cells are larger than the typical human small cell lung tumor, and mitotic figures are frequent. H & E, x 450.

CANCER RESEARCH VOL. 45 JUNE 1985 2833 Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1985 American Association for Cancer Research. Growth of Human Lung Tumor in the Brain of the Nude Rat as a Model to Evaluate Antitumor Agent Delivery across the Blood-Brain Barrier

Edward A. Neuwelt, Eugene P. Frenkel, Anthony N. D'Agostino, et al.

Cancer Res 1985;45:2827-2833.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/45/6/2827

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/45/6/2827. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1985 American Association for Cancer Research.