experimental chemotherapy of human medulloblastoma cell ... · medulloblastoma cell lines and...

[CANCER RESEARCH 48,4189-4195, August 1, 1988|

Experimental Chemotherapy of Human Medulloblastoma Cell Lines andTransplantable Xenografts with Bifunctional Alkylating Agents1

Henry S. Friedman,2O. Michael Colvin, Stephen X. Skapek, Susan M. Ludeman, Gertrude B. Elion,

S. Clifford Schold, Jr., Phillip F. Jacobsen, Lawrence H. Muhlbaier, and Darell D. BignerDepartments of Pediatrics [S. X. S., H. S. F.], Medicine [G. B. E, S. C. SJ, Community and Family Medicine, Biometry Division [L. H. M.], and Pathology [D. D. B.,H. S. F., S. C. SJ, and the Preuss Laboratory for Brain Tumor Research [D. D. B.], Duke University Medical Center, Durham, North Carolina 2 7710; the Johns HopkinsOncology Center, Baltimore, Maryland 21205 [O.M.C.J; the Department of Chemistry, The Catholic University of America, Washington, DC 20064 [S. M. LJ; and theDepartment of Neuropathology, Royal Perth Hospital, Perth, Western Australia 6001 [P. F. J.J

ABSTRACT

A serÃesof bifunctional al kviators were tested against the genotypicallyand phenotypically heterogeneous continuous human medulloblastomacell lines, TE-671, Daoy, and D283 Med in vitro and against TE-671and Daoy growing as s.c. and intracranial xenografts in athymic mice.Drugs tested included melphalan, cyclophosphamide, iphosphamide,phenylketocyclophosphamide, thiotepa, l,3-bis(2-chloroethyl)-l-nitro-sourea (in vivo), and busulfan (in vivo). Melphalan and phenylketocyclophosphamide were the most active agents in vitro with drug doses atwhich there is a 90% reduction in the number of colonies in comparisonto controls of 2.13, 5.29, and 4.72 pM for melphalan and 4.60, 5.01, and434 MMfor phenylketocyclophosphamide against TE-671, D283 Med,and Daoy, respectively. Melphalan, cyclophosphamide, iphosphamide,phenylketocyclophosphamide, and thiotepa produced significant growthdelays against s.c. TE-671 and Daoy xenografts, while no activity couldbe demonstrated for l,3-bis(2-chlornuthyl)-l-nitrnsnurea or busulfan.Melphalan, cyclophosphamide, iphosphamide, and thiotepa also produced significant increases in median survival in mice bearing intracranialTE-671 and Daoy xenografts. These results extend our previous studiesdemonstrating the antitumor activity of nitrogen and phosphoramidemustard-based bifunctional alkylating agents in the treatment of humanmedulloblastoma continuous cell lines and transplantable xenografts, andsupport the continued use of these agents in clinical trials.

INTRODUCTION

Medulloblastoma, the most common malignant tumor of thecentral nervous system in childhood, continues to represent afrustrating therapeutic challenge (1, 2). Despite treatment withincreasingly sophisticated neurosurgical and radiotherapeuticintervention, the majority of children with medulloblastomawill eventually die of progressive disease. Chemotherapy hasproduced responses in recurrent disease, but has only providedminimal increases in disease-free survival for patients withadvanced disease (3).

Our laboratory has established and characterized a panel ofhuman medulloblastoma models allowing the design of chem-otherapeutic intervention for this tumor (4-6). Preliminarystudies with TE-671, a human medulloblastoma continuous cellline and transplantable xenograft, suggested the antitumor activity of a series of classical alkylating agents, particularlymelphalan and cyclophosphamide (7-9). Clinical trials havesupported a role for cyclophosphamide in producing objectiveresponses in patients with recurrent medulloblastoma (10, 11),and preliminary results suggest similar activity for melphalan.

Nevertheless, reliance on a chemotherapeutic profile for medulloblastoma based on studies with a single cell line and

Received 9/21/87; revised 2/15/88; accepted 4/25/88.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 LI.S.C. Section 1734 solely to indicate this fact.

1This work was supported by NIH Grants CAI 1898, CA44640, I NINCDS

NS 20023, l K07 NS 00958, NS 20581, and CA 37323.2To whom requests for reprints should be addressed at P.O. Box 2916,

Department of Pediatrics, Division of Hematology/Oncology, Duke UniversityMedical Center, Durham, NC 27710.

transplantable xenograft disregards the known marked geno-typic and phenotypic heterogeneity of this tumor. Data confirming this heterogeneity continue to mount, convincingly demonstrating the wide spectrum of histological malignancy, tissuepattern, neuronal and/or glial differentiation, and genotypesseen in medulloblastoma (12-19). Preclinical studies designedto identify chemotherapeutic intervention for this tumor mustreflect this heterogeneity if agents active in the treatment ofpatients with medulloblastoma are to be successfully chosen.

We now report the in vitro and in vivo responses of a panelof 3 genotypically and phenotypically heterogeneous humanmedulloblastoma cell lines and transplantable xenografts, TE-671, D283 Med, and Daoy, to a series of bifunctional alkylatingagents [5 of the 7 agents have been previously published forTE-671 (9)]. Agents studied included melphalan, cyclophosphamide, iphosphamide, phenylketocyclophosphamide, thiotepa,BCNU3 (in vivo), and busulfan (in vivo). Our data demonstrate

that despite differences in the quantitative response of TE-671,D283 Med, and Daoy to nitrogen and phosphoramide mustard-based alkylating agents, these compounds, particularly melphalan, cyclophosphamide, and thiotepa, are active in the treatment of all 3 human medulloblastoma cell lines grown in cultureor as s.c. or intracranial xenografts. These studies have strongimplications for the design of future Phase II and III chemotherapeutic trials for patients with medulloblastoma.

MATERIALS AND METHODS

Cell Lines. The human medulloblastoma-derived cell lines TE-671,Daoy, and D283 Med were used in these studies. Their establishment,characterization, and tissue culture techniques have been previouslydescribed (4, S, 20, 21). Chromosomal analysis of D283 Med revealeda stemline of 46,XY,i(17q),der(8)t(8,?)(ql3,?),der(20)t(5;20)(ql I;ql3)(5). TE-671 and Daoy contained near-tetraploid stemlines with distinctive marker chromosomes identical to those described originally (20,21). Phenotypic analysis revealed that TE-671 and Daoy demonstrateda glial-like profile, with reactivity with monoclonal antibodies directedagainst epidermal growth factor receptor, tenascin, 111,A-A,B epitopes,02-microglobulin, and 5 of 8 of glioma-associated antigens, but not withan antineurofilament protein antibody. D283 Med demonstrated aneuronal-like profile, with expression of neurofilament protein, butwithout reactivity with monoclonal antibodies against tenascin, epidermal growth factor receptor, 11LA A,B epitopes, ft-microglobulin, or 6of 8 of glioma-associated antigens (22).

Clonogenic Assay. All drug assays were performed by using a 1-hdrug incubation with a final cell concentration of 2.5 x 10' cells/dishfor TE-671 and 1.0 x 10* cells/dish for Daoy and D283 Med, as

previously described (9). Response was assessed by the ratio of treatedversus control colony formation. All drugs and solutions were providedand prepared as previously described (9).

Animals. Male or female athymic BALB/c mice (nu/nu genotype, 6weeks or older) were used for all in vivo studies and were maintainedas previously described (23).

3The abbreviations used are: BCNU, l,3-bis(2-chloroethyl)-l-nitrosourea;

DMSO, dimethyl sulfoxide.

4189

Research. on January 20, 2021. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

CHEMOTHERAPY OF HUMAN MEDULLOBLASTOMA LINES WITH ALKYLATORS

Xenograft Transplantation. TE-671 and Daoy grown as s.c. andintracranial xenografts were used for in vivo studies. Lack of homogeneous growth and survival curves with D283 Med prevented in vivo useof this xenograft. Xenograft transplantation (s.c.) was performed aspreviously described with inoculation volumes of 30 n\ of TE-671 and100 n\ for Daoy (8). Intracranial xenograft transplantation of TE-671was performed as previously described (8). Daoy cells for intracranialinoculation were harvested in vitro at 90% confluence, centrifuged at1000 rpm for 5 min, and resuspended in 1.0 ml minimum essentialmedium to a concentration of 4.0 x 107-6.75 x IO7cells/ml. Injections

(30 fiI)of the homogeneous suspension were given into the right cerebralhemisphere with a 500-^1 Hamilton syringe through a 27-gauge needleequipped with a sleeve allowing 4.5-mm penetration.

Tumor (s.c.) Measurements. The s.c. tumors were measured as previously described (8).

Drug Toxicity. The lethal toxicity of individual drugs was assessedby probit analysis (24). A minimum of 4 dose levels with 10 animals/dose was used to calculate the 10% lethal dose of each drug. Theregimen used in each experiment was 100% of the calculated 10%lethal dose administered as a single i.p. injection in a volume of 90 ml/m2. The doses used were: melphalan, 71.3 mg/m2; cyclophosphamide,1391 mg/m2; phenylketocyclophosphamide, 279.9 mg/m2; iphospham-ide, 2014.9 mg/m2; thiotepa, 61.8 mg/m2; BCNU, 100 mg/m2; andbusulfan, 60.3 mg/m2. Melphalan was administered in 16.7% DMSO,

phenylketocyclophosphamide in 33% DMSO, busulfan in 10% DMSO,and BCNU in 3.3% ethanol. Cyclophosphamide, iphosphamide, andthiotepa were administered in 0.9% NaCl solution.

Tumor Therapy. Therapy experiments for s.c. and intracranial xenografts were performed as previously described (8). Xenografts (s.c)were treated when the median tumor volume exceeded 200 mm3 (TE-671) and 350 mm3 (Daoy); intracranial xenografts were treated on day12 (TE-671) and day 21 (Daoy) after implantation.

Response of s.c. xenografts was assessed by mean treated versuscontrol tumor volume (T/C), the difference in days between the medianof individual treated animals' and individual control animals' tumorsto reach a volume of 5 times the treatment volume (Tâ€”C), and treated

versus control tumor regressions. Response of intracranial xenograftswas assessed by comparing median survival time and long term survivors (>90 days) between treated and control groups.

Statistical analyses were performed as previously described (7).

RESULTS

Clonogenic Assay. The in vitro dose response curves formelphalan, 4-hydroperoxycyclophosphamide, thiotepa, 4-hy-droperoxyiphosphamide, and phenylketocytophosphamideagainst each cell line are presented in Figs. 1-5. The curves foreach drug tested were similar in slope and demonstrated decreasing survival with increasing dose. The calculated in vitrodrug dose at which there is a 90% reduction in the number ofcolonies in comparison to controls for the 5 drugs tested in

vitro are presented in Table 1. Within each line, the relativeand absolute activity of each drug was similar, with melphalanand phenylketocyclophosphamide being most active on a molarbasis and thiotepa and 4-hydroperoxyiphosphamide least active.

Drug Toxicity. Twelve deaths in 220 and twenty-two deathsin 217 tumor-bearing animals for TE-671 and Daoy, respectively, were attributed to drug toxicity with the following distribution: melphalan, 1 of 40 and 5 of 50; cyclophosphamide,4 of 40 and 5 of 39; phenylketocyclophosphamide, 0 of 40 and5 of 29; thiotepa, 5 of 40 and 4 of 39; iphosphamide, 2 of 40and 2 of 40; busulfan, 0 of 10 and 0 of 10; and BCNU, 0 of 10and 0 of 10 for TE-671 and Daoy, respectively. Mean nadirweight loss was: melphalan, 15.9 and 20.9%; cyclophosphamide, 17.6 and 22.0%; phenylketocyclophosphamide, 17.7and 12.8%; thiotepa, 17.0 and 10.9%; iphosphamide, 12.6 and14.7%; busulfan, 3 and 1.1%; and BCNU, 2 and 0% for TE-671 and Daoy, respectively.

Tumor Therapy (s.c.). The response to chemotherapy for eachline is summarized in Table 2. Growth curves illustrating theresponse to melphalan, cyclophosphamide, thiotepa, iphosphamide, phenylketocyclophosphamide, BCNU, and busulfanfor Daoy are presented in Figs. 6 and 7 [growth curves for TE-671 to melphalan, cyclophosphamide, thiotepa, iphosphamide,and phenylketocyclophosphamide have been previously published (9)].

All agents except BCNU and busulfan produced statisticallysignificant (P < 0.01) growth delays in TE-671 and Daoy.Melphalan and cyclophosphamide were the only agents thatproduced a statistically significant (P < 0.01 ) number of regressions in TE-671 with neither agent producing cures (defined ascomplete regression without regrowth). All agents with theexception of phenylketocyclophosphamide which producedgrowth delays in Daoy also demonstrated consistent tumorregressions. There were no spontaneous regressions in controlanimals for either TE-671 or Daoy.

Intracranial Tumor Therapy. The five drugs which were activeagainst the s.c. xenografts were also tested against intracranialxenografts (Table 3). Mortality distribution for animals bearingDaoy xenografts treated with melphalan, cyclophosphamide,thiotepa, iphosphamide, and phenylketocyclophosphamide arepresented in Figs. 8 and 9 [mortality distribution curves for TE-671 to melphalan, cyclophosphamide, thiotepa, iphosphamide,and phenylketocyclophosphamide have been previously published (9)]. All agents tested produced statistically significant(P < 0.01) increases in median survival in both lines with theexception of phenylketocyclophosphamide against intracranial

Fig. 1. In vitro dose-response curves show- zing the chemosensitivity of TE-671, Daoy, and -D283 Med to melphalan; 95% confidence limits.

I'20

< 60

ZO6 20VJ

H

TE-671 Daov D283 MED

03 16 33 03 16 33 66

iiM

03

4190




Fig. 2. In vitro dose-response curves showing the chemosensitivity of TE-671, Daoy, andD283 Med to 4-hydroperoxycyclophospham-ide. Data from TE-671 are reproduced as published in Ref. 9, Fig. 1, p. 2828. It is reshownfor comparative purposes; 95% confidencelimits.

Doo,

0.3 1.7 3.4uM

D283 MED

140

Fig. 3. In vitro dose-response curves showing the chemosensitivity of TE-671, Daoy, andD283 Med to thiotepa. Data from TE-671 arereproduced as published in Ref. 9. Fig. 2, p.2828. It is reshown for comparative purposes;95% confidence limits.

60

40

20

TE-671 Door D283 MED

J l l53 264 52.Â» 52.9 105.7 0.5 5.3 112 244 52.9 105.7

Fig. 4. In vitro dose-response curves showing the chemosensitivity of TE-671, Daoy, andD283 Med to phenylketocyclophosphamide.Data from TE-671 are reproduced as publishedin Ref. 9, Fig. I, p. 2828. It is reshown forcomparative purposes; 95% confidence limits.

_I40|-

j 120

1 Â«O

Z 80O

S 60Ofo

Dooy

03 7.1 14.2 OÃ 0.25 05 1 2 0.1 0.25 0.5 1

Fig. 5. In vitro dose-response curves showing the chemosensitivity of TE-671, Daoy, andD283 Med to 4-hydroperoxyiphosphamide.Data from TE-671 are reproduced as publishedin Ref. 9, Fig. 1, p. 2828. It is reshown forcomparative purposes; 95% confidence limits.

D283 MED

Daoy. Melphalan was the most active drug against intracranial against intracranial Daoy with percentage of increased medianTE-671 with percentage of increased median survival of 42 and survival of 61 and 36% in duplicate trials. No agent produced64% in duplicate trials; thiotepa was the most active drug statistically significant numbers of long term survivors.

4191




Table 1 In vitro cytotoxicityIDÂ«,"(â€žM)

DrugMelphalanPhenylketocyclophos-phamide4-Hydroperoxycyclophos-phamide4-Hydroperoxyiphospham-ideThiotepaTE-671*2.134.6012.8020.8094.6D283Med5.295.0122.4747.13136.40Daoy4.724.3429.90104.9876.30

" II>.â€žâ€žin vitro drug dose at which there is a 90% reduction in the number of

colonies in comparison to controls.* Data from TE-671 are reproduced as published in Ref. 9, Table 1, p. 2829,

for phenylketocyclophosphamide, 4-hydroperoxycyclophosphamide, 4-hydrope-roxycyclophosphamide, and thiotepa, reshown for comparative purposes. Theresponse of TE-671 to melphalan represents new experiments, using a standardlot of melphalan tested against all three cell lines.

DISCUSSION

The role of chemotherapeutic intervention for medulloblas-toma remains poorly defined, limited by a lack of active agentswhich can be utilized in adjuvant therapy trials. Recent clinicaland laboratory studies have suggested the potential role of theclassical alkylating agent cyclophosphamide in the treatment ofhuman medulloblastoma (7, 10). More recent studies in ourlaboratory have documented the activity of a series of classicalbifunctional alkylators in the treatment of the human medulloblastoma continuous cell line and transplantable xenograftTE-671, and have directly led to a Phase II trial of melphalanfor recurrent medulloblastoma (9). This trial, now in progress,supports the potential role of melphalan, with 2 responses (1complete response and 1 partial response) in the first 7 patientstreated.

Nevertheless, clinical translation of experimental therapeuticprofiles for medulloblastoma based on studies limited to onecell line and xenograft ignore the known heterogeneity ofmedulloblastoma, and may well lead to inappropriate clinicaltrials. Studies demonstrating the wide spectrum of histolÃ³gica!malignancy, tissue pattern, neuronal and/or glial differentia

n, and genotypes seen in medulloblastoma (12-19) supportconsideration of this heterogeneity in the selection of chemotherapeutic agents for clinical trials. The 3 human medulloblastoma cell lines utilized in our current alkylator studies reflectthe wide spectrum of genotypic and phenotypic patterns seenin this tumor. TE-671 and Daoy, derived from cerebellar me-dulloblastomas, are adherence-dependent cell lines which demonstrate a glial-like profile with reactivity against epidermalgrowth factor receptor, tenascin, HLA-A,B epitopes, fa micro-globulin, and 5 of 8 of glioma-associated antigens, but not withan an t incuro filament protein antibody. TE-671 and Daoy alsodemonstrate marked differences in the transport and glutathi-one-mediated detoxification of melphalan. Karyotypically, bothcell lines contain near-tetraploid stemlines but each possessesdistinctive marker chromosomes. D283 MED, derived fromperitoneal medulloblastoma mÃ©tastases,is an adherence-independent cell line which demonstrates a neuronal-like profile,with expression of neurofilament protein, but without reactivitywith monoclonal antibodies against tenascin, epidermal growthfactor receptor, HLA-A,B epitopes, fo-microglobulin, or 6 of 8glioma-associated antigens. Chromosomal analysis reveals a

pseudodiploid stemline karyotype with i(17q) and unbalancedtranslocations involving chromosomes 5, 8, and 20. The mostcommon chromosomal abnormality seen in medulloblastomabiopsies is i(17q). This structural abnormality characterizes D-283 Med and TE-679 but is not present in Daoy. All three linesdemonstrate rearrangements involving chromosomes 5 and 8,but the breakpoints are in different regions. Thus, althougheach cell line contains abnormalities in common with othermedulloblastoma cell lines and medulloblastoma biopsies, eachline maintains an individual and distinctive karyotypic pattern.

The current studies extend our previous work with classicalalkylators, demonstrating the activity of nitrogen and phos-phoramide mustard-based alkylators in the treatment of thispanel of genotypically and phenotypically heterogeneous human medulloblastoma continuous cell lines and transplantablexenografts. The anti-neoplastic activity of melphalan, a watersoluble (log P = â€”1.70)bifunctional alkylator transported via a

Table 2 Chemotherapeutic response ofs.c. xenografts

DrugMelphalanCyclophosphamideIphosphamideThiotepaPhenylketocyclophosphamideBusulfanBCNUTIC(%)*36.7

24.312.616.5

15.524.5

20.036.1

15.260.5

66.298.7

(NS)120(NS)DaoyT-

C(days)'91+'

63.4104+/42.2

45.043.5

29.631.237.814.9

7.92.7

(NS)0(NS)Regressions'*6/8

10/1010/10â€¢ill

10/1010/10

9/99/9

8/82/8

(NS)0/7(NS)0/9

(NS)0/10

(NS)T/C(%)22.9

39.630.5

26.337.2

37.148.9

(NS)36.740.0

39.3118.0(NS)83.8

(NS)TE-67

1Â°T-

C(days)20.9

19.518.4

17.315.8

11.312.0

11.15.97.2-0.6

(NS)-0.35

(NS)Regressions10/10

10/107/9

7/70/10(NS)3/7

(NS)4/7(NS)0/10

(NS)1/10(NS)0/10

(NS)0/10

(NS)" Data from TE-671 to melphalan, cyclophosphamide, iphosphamide, thiotepa, and phenylketocyclophosphamide are reproduced exactly as published in Table 2,

p. 2830. Ref. 9, reshown for comparative purposes.* T/C (%); mean treated versus control tumor volume 7-12 days posttreatment.'/'-(' (days); the difference in days between the median of individual treated animals and individual control animals to reach a volume of 5 times the treatment

volume." Number of animals with regressing tumors/number of treated animals. Regressing tumors have at least 2 consecutively decreasing volume measurements.' Treated animals include 5 of 8 that were not palpable at â€¢120 days after inoculation (NS).â€¢^Treated animals include 5 to 10 that were not palpable at >120 days (NS).* P values for all experiments were <0.01 unless otherwise designated; NS, not significant.

4192




Daoy

Fig. 6. Groups of 10 randomly assignedmice bearing s.c. Daoy were treated with chem-otherapeutic agents or with an equivalent volume of drug vehicle. Drugs were delivered byi.p. injection at 100% of the 10% lethal dosewhen the median tumor volume exceeded 350

104

Â§

i io:

MELPHAIAN CYCIOPHOSPHAMIDE

30 40 50 60 70 80 90 BO 110 Â¿gJÂ«Â¿830 40 50 60 70

DAYS AFTER IMPLANTATION30 40 50 60 70 80

IO4Daoy

103

510

os

IO3

IO2

IO

IPHOSPHAMIDE

i i i \i i lTREATMENT!

PHENYLKETOCYClOPHOSPHA/vMDE

10 , 20TSE ATME NT I

J_30 40

j_

BCNU

i i lTREATMENT)

BUSULFAN

DAYS AFTER

50 60 70

IMPLANTATION

VEHIC t e TREATED

DRUG TKATfD

J_10 , 20

TREATMENT!30 40

Fig. 7. Groups of 10 randomly assigned mice bearing s.c. Daoy were treatedwith chemotherapeutic agents or with an equivalent volume of drug vehicle. Drugswere delivered by i.p. injection at 100% of the 10% lethal dose when the mediantumor volume exceeded 350 mm3.

carrier-mediated pathway(s), noted initially against TE-671,was also seen against Daoy and D283 Med. This agent consistently demonstrated the greatest activity of all agents testedagainst the 3 lines in vitro, and the 2 lines tested as s.c.xenografts. The results with melphalan against intracranialDaoy were less impressive than those with cylophosphamide orthiotepa, but these may not be biologically relevant differences.These results, coupled with our current Phase II study with

melphalan, indicate that the role of this agent in the treatmentof patients with medulloblastoma deserves further study. Laboratory studies in progress are evaluating potential mechanismsmodulating sensitivity and resistance to melphalan in humanmedulloblastoma, including transport, glutathione levels, andformation and repair of DNA interstrand cross-links.

Cyclophosphamide, previously shown to be active in clinicaland experimental studies with medulloblastoma (7-10) wasactive against all 3 lines in vitro and both xenografts in vivo.(Tables 1-3). Iphosphamide (4-hydroperoxyiphosphospham-ide) was less active on a molar basis in vitro than cyclophosph-amide (4-hydroperoxycyclophosphamide), but produced similargrowth delays and increases in median survival against s.c. andintracranial xenografts, respectively. Phenylketocyclophos-phamide was markedly more active than 4-hydroperoxycyclophosphamide against the 3 cell lines in vitro, but producedsmaller growth delays and increases in median survival againsts.c. and intracranial xenografts, respectively. This may be dueto dose limitations secondary to increased hematological tox-icity or more rapid plasma clearance (9). The stem cell protection from Cyclophosphamide afforded by aldehyde dehydrogen-ase will not be operational with phenylketocyclophosphamide,which instead undergoes a simple elimination reaction to generate phosphoramide mustard (25). This agent may provide anapproach to bypass aldehyde dehydrogenase-mediated resistance to Cyclophosphamide, although myelosuppression mayprove unacceptably severe.

Thiotepa is a phosphoramide mustard analogue that has beenminimally active in the intrathecal treatment of advanced medulloblastoma, quite possibly due to the low dosages used (26,27). Although it was less effective in vitro, it produced similargrowth delays (s.c. xenografts) and increases in median survival(intracranial xenografts) compared to the other agents tested.The disparity between in vitro, s.c. xenograft, and intracranialxenograft results demonstrates the complex interactions ofcellular sensitivity and drug delivery operational in each system,and supports inclusion of data from all three systems in determining selection of drugs for entry into clinical trials. On thebasis of our results thiotepa may be considered for new PhaseII trials of recurrent medulloblastoma.

Neither busulfan, a methane sulfonate derivative, nor BCNU,a highly lipophilic (log /' â€”1.53) nitrosourea demonstratedactivity against s.c. TE-671 or Daoy. Busulfan has not beenutilized in the treatment of patients with medulloblastoma, andour studies do not support clinical investigations with thisagent. BCNU has demonstrated moderate efficacy in the treatment of adults with glioma (28), but is not active in thetreatment of children with medulloblastoma (29). Recent laboratory studies suggest that the presence of high levels of O6-alkylguanine-DNA alkyltransferase in the medulloblastoma

4193



CHEMOTHERAPY OF HUMAN MEDULLOBLASTOMALINES WITH ALKYLATORS

Table 3 Chemotherapeutic response of intracranial xenografis

DrugMelphalanCyclophosphamideThiotepaIphosphamidePhenylketocyclophosphamide%IMS*33

1446

2961

3614

348(NS)DaoyMedian

survival(range)'64(52-126+)

64.5(59-91+)70

(52-100)73(53-81)88(31-117)

68(55-94)54.5

(40-73)67(54-86)59

(42-78)LTS*2

(NS)'

HNS)1(NS)

0(NS)3(NS)0(NS)0(NS)%

IMS64

4230

4442

42262619

30TE-67

1Â°Median

survival(range)59

(24-75+)54(38-72+)35

(39-38)49(45-54)49

(33-54)49(31-54)38

(32-42)38(31-48)32

(28-37)45 (38-54)LTS2(NS)

1(NS)0(NS)

0(NS)0(NS)

0(NS)0(NS)

0(NS)1(NS)

0(NS)" Data from TE-671 are reproduced as published in Ref. 9, Table 3, p. 2831, reshown for comparative purposes.* % IMS, percentage of increased median survival of treated animals versus control animals. Median day of death for control animals ranged from 27-38 for TE-

671 and 48-56.5 days for Daoy.' Median survival in days of treated animals (range of deaths of treated animals excluding toxic deaths secondary to drug).* LTS, long term survivors; Â£90days. No long term survivors were seen in control groups.'' P values are Â£0.01,unless otherwise designated not significant (NS).

DaoyMEIPHAIANCYCLOPHOSPHAMIOETh,o-TEPA-;i7ftonÂ«o30â€ž1011rS8&t

, .1Â¡h

iiL

iiL_j-

.iâ€¢-

ii-ll[_.

. . ,1

tj

, i-

1|-i

â€”iL

iii':iL

_|1

1111'-"

~1l11i11

!.;WtClf

TPfATfli '1--Â¿Mtc'Ã„KtfKP|>

i!..

DaoyIPHOSPHAMIDE PHENYLKETOCYCIO-

PHOSPHAMIDE

)30Â«5060*>8090 K30

DAYS AFTER IMPLANTATION

Fig. 8. Groups of 10 randomly assigned mice bearing intracranial Daoy weretreated with chemotherapeutic agents or with an equivalent volume of drug vehicleby i.p. injection of 100% of the 10% lethal dose on day 21 after tumor implantation.

xenografts TE-671, Daoy, and D283 Med may be responsiblefor minimizing the activity of BCNU (30). Our studies do notsupport inclusion of BCNU in new clinical trials for medullo-blastoma.

Our results demonstrate that specific bifunctional alkylators,notably melphalan, Cyclophosphamide, thiotepa, and iphos-phamide are active in all cell lines and xenografts tested, thusconfirming our previous studies with TE-671 and suggestingthat a homogeneous response profile of this tumor to classicalalkylators, paralleling its homogeneous radiosensitivity (1), maywell exist, similar to that seen in rhabdomyosarcoma or neuroblastoma. Clinical trials using alkylating agent therapy fornewly diagnosed patients prior to the use of radiotherapy willbe needed to determine if this alkylator response rivals thatseen to radiation therapy. Nevertheless, clinical trials in patientswith recurrent medulloblastoma have demonstrated high response rates, albeit in limited numbers of patients (10). Continued studies both in the laboratory with our newer medulloblastoma cell lines and xenografts (31), and clinically with largernumbers of patients with medulloblastoma treated with melphalan or other bifunctional alkylators (in Phase II and Phase

80 20TREATMENT!

DAYS AFTER IMPLANTATION

40 60 80

Fig. 9. Groups of 10 randomly assigned mice bearing intracranial Daoy weretreated with chemotherapeutic agents or with an equivalent volume of drug vehicleby i.p. injection of 100% of the 10% lethal dose on day 21 after tumor implantation.

Ill trials), will be needed to determine if the observed alkylatorresponse of the three cell lines and xenografts tested is indeedrepresentative of the response of medulloblastoma to specificbifunctional alkylating agents.

Future studies in our laboratory will continue to define thetherapeutic profile of medulloblastoma, to determine if syner-gistic interactions between specific alkylating agents are present, and to quantitate cross-resistance and collateral sensitiv

ity. Analysis of the antigenic profile and the biochemical differences in drug transport, drug metabolism, glutathionedetoxification, and formation/repair of DNA interstrand crosslinks seen in medulloblastoma may allow better drug designand selection for clinical trials based on tumor profiles obtainedat initial neurosurgical intervention. Clinical protocols willcontinue to attempt translation of these laboratory studies intoeffective Phase II and III trials.

4194




REFERENCES

1. Bloom, H. J. G. Medulloblastoma in children: increasing survival rates andfuture prospects. Int. J. RadiÃ¢t.Oncol. Biol. Phys., 8: 2023-2027, 1982.

2. Duffner, P., and Cohen, M. Brain Tumors in Children. New York: RavenPress, 1985.

3. Allen, J. C., Bloom, J., Ertel, I., Evans, A., Hammond, D., Jones, H., Levin,V., Jenkin. D., Sposto, R., and Wara, W. Brain tumors in children: currentcooperative and institutional chemotherapy trials in newly diagnosed andrecurrent disease. Semin. Oncol., 13:110-122, 1986.

4. Friedman, H. S., Bigner, S. H., McComb, R. D., Schold, S. C., Jr., Pasternak,J. F., Groothuis, D. R., and Bigner. D. D. A model for human medulloblas-toma: growth, morphology and chromosomal analysis in vitro and in athymicmice. J. Neuropathol. Exp. Neurol., 42:485-503, 1983.

5. Friedman, H. S., Burger, P. C., Bigner, S. H., Trojanowski, J. Q., Wikstrand,C. J., Halperin, E. C., and Bigner, D. D. Establishment and characterizationof the human medulloblastoma cell line and transplantable icnografi D283Med. J. Neuropathol. Exp. Neurol., 44: 592-605, 1985.

6. Friedman, H. S., Bigner, S. H., Schold, S. C., Jr., and Bigner, D. D. The useof experimental models of human medulloblastoma in the design of rationaltherapy. In: M. D. Walker and D. G. T. Thomas (eds.), Biology of BrainTumor, pp. 405-409. The Hague: Martinus Nijoff Publishing, 1986.

7. Friedman, H. S., Schold, S. C., Jr., Varia, M., and Bigner, D. D. Chemotherapy and radiation therapy of human medulloblastoma in athymic nudemice. Cancer Res., 43:3088-3093,1983.

8. Friedman, H. S., Schold, S. C., Jr., and Bigner, D. D. Chemotherapy ofsubcutaneous and intracranial human medulloblastoma xenografts inathymic nude mice. Cancer Res., 46: 224-228, 1986.

9. Friedman, H. S., Colvin, O. M., Ludeman, S. M., Schold, S. C., Jr., Boyd,V. I... Muhlbaier, L. H., and Bigner, D. D. Experimental chemotherapy forhuman medulloblastoma with classical alkylators. Cancer Res., 46: 2827-2833, 1986.

10. Allen, J. C., and Helson, L. High dose cyclophosphamide chemotherapy forrecurrent CNS tumors in children. J. Neurosurg., 55:749-756, 1981.

11. Friedman, H. S., Mahaley, M. S., Jr., Schold, S. C., Jr., Vick, N. A., Palletta,J. M., Bullard, D. E., D'Souza, B. J., Khandekar, D. J., Lew, S., Oakes, W.

J., and Bigner, D. D. Efficacy of vincristine and cyclophosphamide in thetherapy of recurrent medulloblastoma. Neurosurgery, 18:335-340, 1986.

12. Burger, P. C., Grahmann, P. C., Bliestle, A., and Kleihues, P. Differentiationin medulloblastomaâ€”a histolÃ³gica! and immunohistochemical study. ActaNeuropathol., 73: 115-123, 1987.

13. Rubinstein, L. J. The cerebellar medulloblastoma: its origin, differentiation,morphological variants and biological behavior. In: R. J. Vincken and G. W.Brown (eds.), Tumors of the Brain and Skull, Pan 3, Vol. 9, pp. 167-194.New York: Elsevier/North-Holland Biomedicai Press, 1975.

14. Rorke, L. B. The cerebellar medulloblastoma and its relationship to primitiveneuroectodermal tumors. J. Neuropathol. Exp. Neurol., 42:1-15, 1983.

15. Becker, L. E., and Hinton, D. Primitive neuroectodermal tumors of thecentral nervous system. Hum. Pathol., 19: 538-550, 1983.

16. Velasco, M. E., Ghobrial, M. W., and Ross, E. R. Neuron specific enolase

and neurofilament proteins as markers of differentiation in medulloblastoma.Surg. Neurol., Â¿J:177-182, 1985.

17. Barnard, R. O., and Pambakian, H. Astrocytic differentiation in medulloblastoma. J. Neurol. Neurosurg. Psychiatry, 43: 1041-1044, 1980.

18. Sawa, H., Takeshita, I., Kuramitsu, M., Egami, H., Fukui, M., and Sato, Y.Neuronal and glial proteins in medulloblastoma. II. Heterotransplantationof human medulloblastoma. Anticancer Res., 6:911-916, 1986.

19. Bigner, S. H., Mark. J., Friedman, H. S., Biegel, J. A., and Bigner, D. D.Structural chromosomal abnormalities in human medulloblastoma. CancerGenet. Cytogenet., JO: 91-101, 1988.

20. McAllister, R. M., Isaacs, H., Rongey, R., Peer, M., Au, W., Soukup, S. W.,and Gardner, M. B. Establishment of a human medulloblastoma cell line.Int. J. Cancer, 20: 206-212, 1977.

21. Jacobsen, P. F., Jenkyn, D. F., and Papadimitriou, J. M. Establishment of ahuman medulloblastoma cell line and its heterotransplantation into nudemice. J. Neuropathol. Exp. Neurol., 44:472-485, 1985.

22. He, X., Skapek, S. X., Wikstrand, C. J., Friedman, H. S., Trojanowski, J.Q.. Kemshead, J. T., Coakham. H. B.. Bigner, S. H., and Bigner, D. D.Phenotypic analysis of four human medulloblastoma cell lines and transplant-able xenografts. J. Neuropathol. Exp. Neurol., in press, 1988.

23. Hullard. D. E., Schold, S. C., Jr., Bigner, S. H., and Bigner, D. D. Growthand chemotherapeutic response in athymic mice of tumor arising from humanglioma-derived cell line. J. Neuropathol. Exp. Neurol., 40: 410-427, 1981.

24. Skipper, H. E., and Schmidt, L. H. A manual on quantitative drug evaluationin experimental tumor systems. Cancer Chemother. Rep., 17: 1-143, 1962.

25. Ludeman, S. M., Boyd, V. L., Regan, J. B., Gallo, K. A., Zon, G., and Ishii,K. Synthesis and antitumor activity of cyclophosphamide analogues. 4.Preparation, kinetic studies, and anticancer screening of "phenylketophos-phamide" and similar compounds related to the cyclophosphamide metabolite aldophosphamide. J. Med. Chem., 29: 716-727, 1986.

26. Gutin, P. H., Weiss, M. D., Wiernik, P. H., and Walker, M. D. Intrathecal.V..V N" trii'thyk-nc thiophosphate (thiotepa) in the treatment of malignantmeningeal disease: Phase I-II study. Cancer (Phila.), 38: 1471-1475, 1976.

27. Edwards, M. S., Levin, V. A., Saeger, M. L., and Wilson, C. Intrathecalchemotherapy for leptomeningeal dissemination of medulloblastoma. Child'sBrain, 8:444-451, 1981.

28. Green, S. B., Byar, D. P., Walker, M. D., Pistenmaa, D. A., Alexander, E.,Jr., Batzdorf, V., Brooks, W. H., Hunt, W. E., Mealey, J., Jr., Odom, G. L.,Paoletti, P., Ransohoff, J., Robertson, J. T., Selker, R. G., Shapiro, W. R.,Smith, K. R., Jr., Wilson, C. B., and Strike, T. A. Comparisons of carmustine,procarbazine, and high-dose methylprednisilone as additions to surgery andradiotherapy for the treatment of malignant glioma. Cancer Treat. Rep., 67:121-132,1983.

29. Friedman, H. S., and Oakes, W. J. The chemotherapy of posterior fossatumors in childhood. J. Neuro-oncol., J: 217-229, 1987.

30. Schold, S. C., Jr., Mitra, S., Swenberg, J., Friedman, H. S., and Kleihues, P.Alkyltransferase activity and alkylating drug sensitivity in human brain tumorxenografts in athymic mice. Neurology, 37: 311, 1987.

31. Friedman, H. S., Burger, P. C., Bigner, S. H., Trojanowski, J. Q., Brodeur,G. M., He, X., Wikstrand, C. J., Kurtzberg, J., Berens. M. E., Halperin, E.C., and Bigner, D. D. Phenotypic and genotypic analysis of a human medulloblastoma cell line and transplantable xenograft (D341 Med) demonstratingamplification of c-myc. Am. J. Pathol., 130:472-484, 1988.

4195



1988;48:4189-4195. Cancer Res Henry S. Friedman, O. Michael Colvin, Stephen X. Skapek, et al. Alkylating AgentsLines and Transplantable Xenografts with Bifunctional Experimental Chemotherapy of Human Medulloblastoma Cell

Updated version

http://cancerres.aacrjournals.org/content/48/15/4189

Access the most recent version of this article at:

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

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/48/15/4189To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



experimental chemotherapy of human medulloblastoma cell ... · medulloblastoma cell lines and...

Documents