Activity of a Novel Anti-folate (PDX, 10-propargyl10-deazaaminopterin) against Human Lymphoma is

Superior to Methotrexate and Correlates with TumorRFC-1 Gene Expression

EUNICE S. WANGa,b, OWEN O’CONNORa, YUHONG SHEc, ANDREW D. ZELENETZa, F.M. SIROTNAKc andMALCOLM A.S. MOOREb,*

aDepartment of Medicine, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY, 10021-6306, USA; bLaboratory ofDevelopmental Hematopoiesis, Memorial Sloan-Kettering Cancer Center, New York, NY, USA; cProgram of Molecular Pharmacology and Experimental

Therapeutics, Memorial Sloan-Kettering Cancer Center, New York, NY, USA

(Received 28 November 2002)

PDX (10-propargyl-10-deazaaminopterin) is a novel anti-folate with improved membrane transport andpolyglutamylation in tumor cells. In prior studies, PDX exhibited enhanced efficacy over methotrexate(MTX) in lung and breast carcinoma xenografts. Because MTX is active in the treatment of aggressivenon-Hodgkin’s lymphoma (NHL), we compared the efficacy of PDX and MTX against five lymphomacell lines: RL (transformed follicular lymphoma), HT, SKI-DLBCL-1 (diffuse large B cell), Raji(Burkitt’s), and Hs445 (Hodgkin’s disease). After 5-day continuous in vitro exposure, PDXdemonstrated .10-fold greater cytotoxicity than MTX in all cell lines (IC50 PDX ¼ 3–5 nM;IC50 MTX ¼ 30–50 nM). We then compared the in vivo effects of anti-folates against three establishedhuman NHL xenografts in NOD/SCID mice. Tumor bearing animals were treated with saline (control)or the maximum tolerated doses of MTX (40 mg/kg) or PDX (60 mg/kg) via an intraperitoneal routetwice weekly for 2 weeks. Almost 90% of HT lymphomas treated with PDX completely regressed,whereas, those treated with MTX treatment had only modest growth delays. In two other xenografts,tumor bearing mice had complete regression rates of 56% (RL) and 30% (SKI-DLBCL-1) after PDXtherapy. No regressions and only minor growth inhibition was noted after MTX therapy. RT-PCRanalysis for the expression of genes involved in folate metabolism demonstrated that increasedsensitivity to PDX correlated with higher RFC-1 gene expression with no difference in FPGS orFPGH levels, suggesting that measurement of tumor RFC-1 gene expression level may be a predictor ofresponse to PDX. These results demonstrate that the PDX has markedly greater potential activityagainst human NHL than MTX and warrants further preclinical and clinical evaluation.

Keywords: Anti-folate; Murine xenografts; Lymphoma; Non-Hodgkin’s

Abbreviations: BL, Burkitt’s lymphoma; DHFR, dihydrofolate reductase; DLBCL, diffuse large B-celllymphoma; FL, follicular lymphoma; FPGS, folylpolyglutamate synthetase; FPGH, folylpolyglutamatehydrolase; HD, Hodgkin’s disease; IC50, 50% inhibitory value; MSKCC, Memorial Sloan-KetteringCancer Center; MTD, maximum tolerated dose; MTX, methotrexate; NHL, non-Hodgkin’slymphoma; NOD/SCID, non-obese diabetic severe combined immunodeficient; PDX, 10-propargyl-10-deazaaminopterin

INTRODUCTION

Non-Hodgkin’s lymphomas (NHL) are presently the

fifth most common cancer in the United States and

account for 4–5% of cancer diagnoses with an estimated

54,900 new cases expected this year. Approximately

300,000 individuals in the U.S. are living with NHL, and

an estimated 26,000 will succumb to this disease this year.

Lymphomas are responsible for approximately 5% of

all cancer-related deaths with a case fatality rate of 47%.

Although lymphomas have long been recognized

as “curable” chemotherapy-sensitive malignancies, in

ISSN 1042-8194 print/ISSN 1029-2403 online q 2003 Taylor & Francis Ltd

DOI: 10.1080/1042819031000077124

*Corresponding author. Tel.: þ1-212-639-7090. Fax: þ1-212-717-3618. E-mail: m-moore@ski.mskcc.org

Leukemia & Lymphoma, 2003 Vol. 44 (6), pp. 1027–1035

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yal

e D

erm

atol

ogic

Sur

gery

on

10/0

1/14

For

pers

onal

use

onl

y.

actuality over 60% of all malignancies (including breast,

colon, and prostate cancers) demonstrate a relative 5-year

survival greater than that of NHL (52%). Over the past 20

years, the standard first line chemotherapy for most

patients has remained largely unaltered [1,2].

Methotrexate (MTX) has been long been known to be

an active drug in the treatment of NHLs. This classical

anti-folate has been a component of multiple drug

regimens for diffuse large B cell, Burkitt’s, cutaneous

T cell and primary central nervous system lymphoma

as well as cerebrospinal lymphomatous metastases [3–7].

However, in recent years, MTX has been dropped

from current upfront lymphoma programs because of

dose-limiting myelosuppression and mucositis.

The cumulative myelotoxicity associated with multiple

rounds of front line chemotherapy not only leaves patients

at high risk for infectious and bleeding complications but

also limits their ability to undergo second line

chemotherapy with autologous stem cell transplantation

and increases the risk of secondary myelodysplasia. The

introduction of novel non-myelotoxic agents for lym-

phoma offers the potential of preserving stem cell reserve

while expanding therapeutic options for this disease.

PDX (10-propargyl-10-deazaaminopterin) belongs to a

class of new folate analogues specifically designed to

possess greater anti-tumor effects than MTX. PDX is

structurally similar to MTX with only a propargyl group

substitution at carbon 10 [8] (Fig. 1). However, prior

pharmacokinetic studies have shown PDX to be a much

more efficient permeant for reduced folate carrier (RFC)

mediated internalization than MTX. In addition, PDX is a

more effective substrate for polyglutamylation by

folylpolyglutamate synthetase (FPGS) than MTX, leading

to greater intracellular drug retention [9–12]. These

biochemical properties may contribute to the superior

anti-tumor effects of PDX as compared to other anti-

folates in early preclinical analyses. In one study, PDX

was ten-to twenty-fold more potent than MTX against a

panel of human breast and lung cancer cell lines after pulse

in vitro exposure [13]. Treatment of human xenograft

models (representing MX-1 mammary carcinoma, LX-1

lung carcinoma, and A549 lung cancer) with MTX only

delayed tumor growth, whereas, PDX treatment resulted

in high complete regression rates (75–85% in MX-1 and

LX-1 tumors, 30% in A549 tumors) and cure rates of

20–30% [13]. A recent phase I trial of PDX was

performed in thirty-three heavily pretreated non-small

lung cancer patients. Although dose-limiting mucositis

was seen, no myelosuppression was observed in any

patients. Major clinical responses occurred in 2 patients,

and 5 patients had stabilization of disease lasting 7–13

months. [14].

Based on these results, we evaluated the preclinical

activity of PDX against aggressive NHL using a panel of

human lymphoma cell lines and three lymphoma

xenografts established in immunodeficient mice.

MATERIALS AND METHODS

Cell Lines

All lymphoma cell lines originated from human patients.

With the exception of the SKI-DLBCL1 line, the cell lines

were well-established lines obtained from the American

Tissue Culture Collection. The Raji cell line represents

Burkitt’s lymphoma with expression of EBNA [15].

The RL cell line represents a transformed follicular

lymphoma (FL) with a t(14;18) translocation, bcl-2 over-

expression, and a p53 mutation. RL cells express CD19,

CD20, CD21, CD22, HLADQ, and HLAR and do not

express CD25 or the T cell receptor [16]. The HT cell line

represents diffuse large B cell lymphoma (DLBCL) and

was obtained from ascitic fluid. These cells have tested

positive for the p53 mutation. HT cells express CD19,

CD20, CD21, CD22, HLADQ and HLADR do not express

CD25 of T cell receptors, and test negative for the Epstein

Barr virus genome [16]. The Hs602 cell line represents a

B cell lymphoma obtained from a cervical lymph node.

Hs602 cells express CD19, CD38, and CS45 with no

expression of CD56 or CD3 [17,18]. The Hs445 cell

line was obtained from the intra-abdominal lymph node

of a patient with Hodgkin’s disease (HD) [17,18].

FIGURE 1 Structural differences between methotrexate and PDX (10-propargyl-10-deaazaminopterin).

E.S. WANG et al.1028

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yal

e D

erm

atol

ogic

Sur

gery

on

10/0

1/14

For

pers

onal

use

onl

y.

The SKI-DLBCL-1 cell line is a cell line established from

the ascitic fluid of patient at our institute which has been

previously described [19,20]. SKI-DLBCL-1 cells

represent de novo extranodal DLBCL with a unique

translocation t(1:14) (q21;q32) resulting in over

expression of MUC-1, an adhesion molecule associated

with many malignancies including lymphoma.

SKI-DLBCL-1 cells express CD19, CD20, and CD45

with no expression of CD5, CD10, or CD23 [19,20].

In order to determine the transcript signature subtype of

the three cell lines, total RNA was extracted and submitted

to expression profiling using the Affymetrix HG-U95Av2

oligonucleotide array. The three DLBCL used in the

present study were of the GC-like DLBCL subtype based

on cluster analysis in comparison with the Lymphochip

cDNA array date (data not shown, R.K. Chaganti, personal

communication).

All cell lines were cultured in RPMI 1640 medium

supplemented with 10% fetal bovine serum, 2 mM

L-glutamine, 4.5 g/l glucose, 1.5 gm sodium bicarbonate,

1 u/ml penicillin/streptomycin, 10 mM HEPES, and 1 nM

sodium pyruvate. Cells were grown in humidified

incubators at 378C with 5% CO2.

Drug Solutions

The synthesis of PDX has been previously described [8].

Solutions of MTX and PDX were prepared in 0.9% NaCl

(pH 7.0) for in vitro and in vivo use.

In vitro Cytotoxicity Assays

Experiments were performed as described previously [9].

In brief, 2:5–5 £ 103 cells were plated per well in 96 well

flat bottom plates. Drug was added at increasing

concentrations, and cells were continuously exposed to

drug for 5 days. Colorimetric dye (XTT or Alamar blue)

was added for an additional time period (XTT dye 6 h,

Alamar blue 24 h). Each plate was then read on an

automated plate reader at 590 nm. The percentage of

inhibition was calculated as growth of cells exposed to

drug divided by growth of controls (cells incubated with

media only). IC50 values were defined as the drug

concentrations at which cell growth was inhibited by 50%

as compared to controls. Experiments were repeated at

least three times. Statistical evaluation was performed

using student’s t-test analysis. Experiments were also

conducted with continuous drug exposures lasting 3 and 4

days and yielded similar results (data not shown).

Xenograft Studies

Six to eight week-old non-obese diabetic severe combined

immunodeficient (NOD/SCID) mice were purchased

from the Jackson Laboratories, Bar Harbor, ME.

Animals were maintained in core animal facilities under

an institute approved animal protocol. All experi-

ments were performed in accordance with the “Principles

of Laboratory Animal Care” (NIH publication No. 85-23

revised 1985). Mice were sublethally irradiated with three

cGy from a gamma source and inoculated with 10 £ 106

lymphoma cells via a subcutaneous route. When tumor

volumes approached 100 mm3, mice were divided into

three groups averaging 3–8 mice per group. Mice were

treated with normal saline (diluent) or the maximum

tolerated doses (MTD) of MTX (40 mg/kg) or PDX

(60 mg/kg) via an intraperitoneal route twice weekly for

two weeks or four total doses [21,22]. The MTD of each

drug has been previously shown to result in ,10% weight

loss and no toxic deaths in nude mice [11,13]. The

schedule of administration has previously been demon-

strated to be effective when comparing these agents and

was chosen for convenience and to conserve drug [13].

Data was expressed as the average change in tumor

volume (mm3), mean tumor volume (mm3) per group, and

average tumor diameter (mm). Tumors were assessed

using the two largest perpendicular axes of the tumor

(l ¼ length, w ¼ width) as measured by calipers. Tumor

volume was calculated using the formula mm3 ¼ 4=3pr 3

where r ¼ ðl þ wÞ=4: Tumor diameter was calculated

from l and w. Animals with no palpable tumor were scored

as complete regressions. Tumor bearing mice were

assessed for weight loss and tumor volume at least twice

weekly for 40 days or until death. Animals were sacrificed

when moribund from progressive tumor related cachexia

and localized symptoms or when one dimensional

tumor diameter exceeded 2.0 cm (according to the

guidelines of the animal core facilities). Statistical

evaluation of the differences in tumor volumes between

treatment groups was performed using student’s t-test

analysis.

Induction of Tumor Apoptosis

Assessment of apoptosis in lymphoma xenografts was

performed using TUNEL (TdT-mediated dUTP-biotin

nick end labeling) immunohistochemical staining. Tissues

from sacrificed animals were fixed in 10% buffered

formalin for 12 h and stored in 70% ethanol prior to

processing and paraffin embedding. The slides were

de-paraffinized, rehydrated, and digested in 20mg/ml of

proteinase K solution for 15 min at room temperature. The

slides were then washed in PBS and refixed in 4%

formalin, washed again in PBS and equilibrated.

Biotinylated nucleotide mix and TdT enzyme were

added and incubated for 90 min at 378 C. Slides were

washed in PBS, blocked in 2% BSA in PBS, and incubated

in ABC vector reagent for 1 h at room temperature. Slides

were then developed in 3,30-diaminobenzidine and

counterstained with Harris hematoxylin prior to

dehydration and mounting.

Quantitative RT-PCR Analysis

RT-PCR analysis was performed as described previously

[23,24]. In brief, total RNA was prepared from cultured

PDX IN NON-HODGKIN’S LYMPHOMA 1029

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yal

e D

erm

atol

ogic

Sur

gery

on

10/0

1/14

For

pers

onal

use

onl

y.

lymphoma cell lines with Trizol regent (Gibco BRL).

Messenger RNA was purified using the Quick Prep mRNA

Purification Kit (Amersham Pharmacia) and then reverse

transcribed using M-MLV RT and random hexamers

(Gibco BRL). First strand cDNA was prepared using the

Superscript System (Gibco BRL). Quantitation of RFC-1,

FSGS, and FPGH gene expression was performed with the

aid of an ABI Prism 7700 Sequence Detection System

using the Taqmam assay [21]. The primer and probe sets

were designed using Primer Express Software (applied

Biosystems) with primer sequences as previously

described [24]. Relative quantiation was performed

using the comparative Ct method with the amount of

target gene normalized to a b actin reference [23].

RESULTS

Cytotoxicity Studies

The in vitro cytotoxicity of increasing concentrations of

PDX and MTX was evaluated against a panel of five

human lymphoma cell lines. As summarized in Table I,

PDX was consistently most potent than MTX in all lines

with IC50 PDX values at least ten-fold less than the

IC50 MTX values. No cell line exhibited enhanced

sensitivity or resistance to either anti-folate in these

studies with only incremental values differences between

all tested lines.

In vivo Studies of Antitumor Activity

In order to study the effects of PDX on lymphoma growth

in vivo, we generated subscutaneous xenotransplantation

models in NOD/SCID mice using three established

human lymphoma lines representative of aggressive

transformed FL (RL) and de novo extranodal DLBCL:

(HT, SKI-DLBCL-1) histologies. Engraftment rates

ranged from 80 to 90%. Palpable tumors formed under

the skin approximately 7–10 days after inoculation and

were readily measurable via calipers. Mice with

subcutaneous lymphoma growths survived an average of

40–50 days after inoculation. Tumor bearing animals

received saline (control), PDX, or MTX via an

intraperitoneal route twice weekly for four total doses

with assessment of tumor growth and weight loss. Review

of H and E stained slides of lymphomas xenografted into

mice were similar in appearance to surgical biopsies of

diffuse large cell lymphoma from patients (slides

reviewed by an experienced hematopathologist, data not

shown).

Treatment results in lymphoma xenografted mice are

summarized in Tables II–IV. In the RL (transformed FL)

and SKI-DLBCL-1 (DLBCL) xenografts, PDX treatment

resulted in much greater inhibition of lymphoma growth

than MTX. Tumors were only minimally sensitive to MTX

treatment with small reductions in growth and no

regressions. PDX treatment, however, decreased tumor

volumes by at least 50% from initial volumes and induced

complete tumor regressions in 56% (5 of 9 mice) and 30%

(3 of 10 mice) of RL and SKI-DLBCL-1 xenografts,

respectively. Although MTX inhibited some tumor growth

in SKI-DLBCL-1 xenografts ð p , 0:0001Þ; the differ-

ences in tumor volumes between control and MTX treated

RL xenografts were not statistically significant ð p ¼

0:046Þ: In contrast, PDX was consistently more effective

than MTX in suppressing tumor growth in both xenografts

( p ¼, 0:0001 and p ¼, 0:0001). At the nadir of tumor

regression, control and MTX treated mice had average

tumor diameters (TD) of 12 and 10.8 mm, whereas, PDX

treated mice had TD of 2.7 mm (^1.1). SKI-DLBCL-1

lymphoma xenografts had TD of 12, 9.5 and 3.5 mm after

saline, MTX and PDX therapy, respectively. The superior-

ity of PDX over MTX was most clearly seen in

experiments with HT (DLBCL) xenografts (Table IV).

Although MTX treatment resulted in modest growth delay

as compared to controls ðp ¼ 0:013Þ; there was no tumor

TABLE I Relative growth inhibition by folate analogues against human lymphoma cell lines. The cells were continuously exposed to variousconcentrations of each analogue for up to 5 days. Values are mean IC50 from two or more experiments. Text references denoted in squared parenthesis

Cell line Lymphoma type IC50 PDX (nM) IC50 MTX (nM) p value

Hs445 Hodgkin’s disease [17,18] 1.6 ^ 0.8 32 ^ 2.2 0.0455HT Diffuse large B cell [16] 3.0 ^ 0.4 35 ^ 5.0 0.0236Raji Burkitt’s [15] 2.0 ^ 0.3 16 ^ 0.8 0.0034RL Transformed follicular [16] 23 ^ 2.0 210 ^ 40 0.0429SKI-DLBCL-1 Diffuse large B cell [19,20] 5.1 ^ 0.1 48 ^ 2.5 0.0035

TABLE II Treatment of human RL (transformed follicular) non-Hodgkin’s lymphoma xenografts in NOD/SCID mice with folate analogues. Animalsreceived drug i.p. every 3–4 days for a total of four doses. Tumors are compared at the nadir of tumor regression and comprise two experimentsperformed with 3–8 mice per group

AgentDose

(mg/kg)Weight change

(%)Average tumor diameter

(mm ^ SE)Average change in tumor volume

(mm3 ^ SE)Tumor regression

(%)Complete regressions

(no./total)

Control – +15.9 12.5 ^ 1.3 +1228 ^ 238 – 0/7MTX 40 214.8 10.9 ^ 0.5 +618 ^ 108 – 0/12PDX 60 211.1 2.7 ^ 1.1 246 ^ 34 57 5/9

E.S. WANG et al.1030

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yal

e D

erm

atol

ogic

Sur

gery

on

10/0

1/14

For

pers

onal

use

onl

y.

regression in these animals (Fig. 2). In contrast, PDX

administration induced complete tumor regression in 89%

of mice with an average tumor regression of 99%. PDX

treatment was statistically better at inhibiting tumor

growth than MTX ð p ¼ 0:0032Þ: At the nadir of tumor

regression, PDX treated mice had a TD of 0.5 mm as

compared to MTX ðTD ¼ 8:7 mmÞ and control mice

ðTD ¼ 11 mmÞ:Mice were monitored throughout treatment. Average

weight changes from initial values were followed during

and after drug administration to confirm that animals

were receiving approximately equitoxic drug doses.

As shown in Tables II–IV, average maximum weight

losses were comparable between MTX and PDX groups

and were consistent with the prior toxicity data of these

analogues administered on an identical schedule in nude

mice [22]. Treatment with saline, MTX, and PDX was

halted in all animals after four doses regardless of

toxicity. Although no permanent cures of disease were

seen, this was expected given the limited duration of

treatment.

Induction of Apoptosis in Lymphoma Xenografts

Immunohistochemical staining for apoptosis as measured

by terminal deoxynucleotide transferase-mediated nick

end-labeling (TUNEL) was performed on SKI-DLBCL-1

xenografts from mice sacrificed at day 21 following

initiation of anti-folate therapy. Mice received four doses

of saline, MTX (40 mg/kg), or PDX (60 mg/kg) on days

0, 3, 7, and 10. As demonstrated in Fig. 3A and 3B,

tumors taken from saline and MTX treated animals

demonstrated little to no evidence of apoptotic cells

(as measured by brown staining). In contrast, systemic

treatment with PDX induced a high level of apoptotic

cells in the tumor tissue 11 days after the last drug dose

(Fig. 3C).

Expression Levels of Genes Involved in Folate

Metabolism

The expression levels of three genes [(RFC-1

transporter, folylpolyglutamate synthetase (FPGS) and

folylpolyglutamate hydrolase (EPGH)] were examined

in the three lymphoma cell lines utilized for xenograft

studies. We designed competitive templates for these

three genes and measured mRNA expression by

quantitative RT-PCR with normalization of expression

to beta-actin. We found that HT lymphoma cells

contained two to three-fold higher expression of the

RFC-1 folate transporter gene than RL or SKI-

DLBCL-1 ( p ¼ 0:02 and 0.007, respectively). The

expression of the FPGH and FPGH genes in the HT

cell line relative to the RL and SKI-DLBCL-1 lines

was similar (Table V).

DISCUSSION

While PDX had demonstrated improved anti-tumor

activity over MTX in other tumor types [13,21,22], this

paper is the first to document the preclinical efficacy

of PDX in aggressive and rapidly growing human

non-Hodgkin’s lymphoma. Because NHL are among the

most heterogeneous of human neoplasms with over forty

different subtypes, it is important to note that continuous

5-day exposure of PDX was uniformly at least ten-fold

more cytotoxic than MTX across a panel of lymphoma

cell lines representing different disease histologies.

To further investigate the therapeutic potential of PDX

in clinically relevant NHL, we developed novel xeno-

transplantation models of human NHL in sublethally

irradiated NOD/SCID mice. In the past, NHL have been

notoriously difficult to study in vivo because of the lack of

established cell lines of animal models representative of

the predominant clinical subtypes. Although mice bearing

TABLE III Treatment of human SKI-DLCL (de novo diffuse large B cell) non-Hodgkin’s lymphoma xenografts in NOD/SCID mice with folateanalogues. Animals received drug i.p. every 3–4 days for a total of four doses. Tumors are compared at the nadir of tumor regression. Shown are theresults of two experiments with 3–5 mice per group

AgentDose

(mg/kg)Weight change

(%)Average tumor

diameter (mm ^ SE)Average change in

tumor volume (mm3 ^ SE)Avg tumor

regression (%)Complete regressions

(no./total)

Control – +4.9 12 ^ 0.3 +786 ^ 64 – 0/8MTX 40 +1.9 9.5 ^ 0.4 +299 ^ 58 – 0/10PDX 60 21.2 3.5 ^ 0.7 281 ^ 16 54 3/10

TABLE IV Treatment of human HT (diffuse large B cell) non-Hodgkin’s lymphoma xenografts in NOD/SCID mice with folate analogues. Animalsreceived drug i.p. every 3–4 days for a total of three to four doses. Tumors are compared at the nadir of tumor regression. Shown are the results of twoexperiments with 3–5 mice per group

AgentDose

(mg/kg)Weight change

(%)Average tumor diameter

(mm ^ SE)Average change in

tumor volume (mm3 ^ SE)Average tumorregression (%)

Complete regressions(no./total)

Control – +13.3 11.2 ^ 1.3 +641 ^ 252 – 0/8MTX 40 29.8 8.7 ^ 2.0 +300 ^ 225 – 0/7PDX 60 28.9 0.5 ^ 0.3 295 ^ 0.8 99 8/9

PDX IN NON-HODGKIN’S LYMPHOMA 1031

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yal

e D

erm

atol

ogic

Sur

gery

on

10/0

1/14

For

pers

onal

use

onl

y.

the nude or SCID mutation have been extensively used to

evaluate human solid tumor neoplasms in vivo, these

strains often have residual NK and host immunity that

limits the engraftment and establishment of human

hematological malignancies. Recently, the SCID mutation

was backcrossed ten generations onto the NOD/Lt strain

background, resulting in an immunodeficient mouse strain

(NOD/LtSz-SCID/SCID) with multiple defects adaptive

as well as non-adaptive immunologic function.

NOD/LtSz-SCID/SCID mice lack functional lymphoid

cells, are hypogammaglobinemic, and have markedly

reduced NK cell activity with functionally immature

macrophages. These multiple defects in innate and

adaptive immunity are unique to the NOD/LtSz-SCID-

SCID mouse and provide an excellent in vivo environment

for reconstitution with human hematopoietic cells [25].

The NOD/SCID mouse has rapidly become the model

system for manipulation of human stem and leukemogenic

cells in vivo [25–28]. A direct comparison of the ability of

four strains of genetically immunodeficient mice

(NOD/SCID, SCID, Nude, Rag-1) to engraft subcutaneous

Burkitt’s lymphoma cell lines demonstrated optimal

growth in NOD/SCID animals receiving whole body

irradiation [29]. Most prior experimental lymphoma

systems employ cell lines derived from BL (i.e. Raji,

Daudi, Namalwa, and Ramos) or are known to be infected

with transforming viruses such as EBV, HHV8, and HIV.

These diseases comprise only a fraction of NHL in the

general population.

Although human systems remain the most reliable and

relevant system in which to study drug activity, we believe

that human xenotransplantation models, despite their

many shortcomings, still provide a valuable experimental

platform on which to rapidly assess and evaluate novel

biologic agents prior to clinical trials. The cell lines

employed in our xenografts were all obtained from human

patients and represent follicular transformed and diffuse

large B cell lymphoma. All these lines have been well

characterized via immunophenotyping, cytogenetic, and

gene expression microarray analysis (see “Materials and

Methods” section). Expression profiling data demonstrate

that all three DLBCL used in the present study were of

FIGURE 3 Measurement of apoptosis in tumor tissue samples fromNOD/SCID mice following treatment with PBS, MTX or PDX. Micereceived four doses of PBS (control) MTX (MTD 45 mg/kg), or PDX(MTD 60 mg/kg) on days 0, 3, 7, and 10. SKI-DLBCL-1 lymphomaxenografts were obtained from mice on day 21. Apoptotic (brownstaining) cells were determined using immunohistochemical staining forterminal deoxynucleotidyl transferase-mediated nick end labeling(TUNEL) assay. As seen in Fig. 3A and 3B, representative tumorsfrom control and MTX treated mice demonstrated little to no evidence ofapoptotic cells (as measured by brown staining in the TUNEL assay). Incontrast, systemic treatment with PDX induced a high level of apoptoticcells in the tumor tissue which was still measurable 11 days after last drugdose (Fig. 3C).

FIGURE 2 Subcutaneous HT lymphoma growth curve after treatmentwith folate analogues. Sublethally irradiated NOD/SCID mice wereinoculated via a subcutaneous route with 10 £ 106 HT (diffuse large Bcell) lymphoma cells. Mice were treated twice weekly with saline(control) or drug (MTX 40 mg/kg or PDX 60 mg/kg) i.p. every 3–4 daysfor a total of four doses. Data are expressed as the increase or decrease inmean tumor volume in mm3 calculated using the formula mm3 ¼ 4/3pr 3

where l ¼ length, w ¼ width, and r ¼ ðl þ wÞ /4. Curve shown is onerepresentative experiment.

E.S. WANG et al.1032

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yal

e D

erm

atol

ogic

Sur

gery

on

10/0

1/14

For

pers

onal

use

onl

y.

the GC-like DLBCL subtype described by Staudt et al.

(data not shown) [30,31].

The present study was aimed at determining the single

agent activity of a novel anti-folate PDX (10-propargyl-

10-deazaaminopterin) against human lymphoma as

compared to the classical anti-folate, MTX. The goal of

the in vivo experiments was to evaluate the relative

efficacy of each agent on inducing tumor regression in

three separate xenograft models. The results described

here demonstrate that PDX consistently possessed

statistically significant superior anti-tumor activity over

MTX in the treatment of aggressive NHL. Treatment with

PDX resulted in complete tumor regression rates of 30–

99% in all three xenografts, whereas, MTX treatment in

these same models induced only minimal reductions in

tumor growth. To confirm the increased cytotoxicity of

PDX vs. MTX on malignant cells in vivo, we performed

immunohistochemical staining of xenograft tissue taken

several days following anti-folate administration and

noted significant apoptotic changes in PDX (but not MTX)

treated tumors.

PDX (10-propargyl-10-deazaaminopterin) belongs to a

class of new folate analogues specifically designed to

posses greater anti-tumor effects than MTX. The

biochemistry and cellular pharmacology of MTX and

other classical folate analogues has previously been

well-described [32–34]. In most tumor cells, a one carbon

reduced folate transporter encoded by the RFC-1 gene

mediates internalization of folate analogues. Once inside

the cell, these analogues will either bind dihydrofolate

reductase (DHFR), thereby depleting intracellular reduced

folate pools needed for the purine and thymidine

biosynthesis, or will be metabolized to a polyglutamate

prior to binding to DHFR. Polyglutamylation is catalyzed

by folylpolygamma-glutamate synthetase (FPGS) and

results in the addition of several glutamic acid resides to

the g-carboxyl group of both folates and anti-folates. This

step is important for the cytotoxic effects of MTX because

polyglutamylated folate analogues are more retentive in

the cell, and polyglutamylated derivatives inhibit de novo

purine sythesis directly as well as dihydrofolate reductase

(DHFR). The folylpolyglutamate hydrolase (FPGH)

enzyme mediates cleavage and clearance of these

intracellular polyglutamated antifolates.

In prior experiments by our group, PDX was

compared with three other 4-aminofolate analogues

(AMT, MTX and EDX) as an inhibitor of DHFR. PDX

was found to have a Ki two to three-fold higher than

the other analogues, suggesting that the interaction of

PDX with this enzyme was less effective than the other

compounds. However, when compared with the other

analogues, PDX was the most efficient permeant for

one-carbon reduced folate transport in CCRF-CEM

leukemic cells with a calculated value for the first-order

rate constant (Vmax/Km) twelve-fold greater than for MTX.

As a substrate for FPGS, PDX was also the most effective

among the analogues examined. Assays of the poly-

glutamate content in the cytosol of cells exposed to the

analogues for 3 h demonstrated increased longer chain

polyglutamates after PDX versus MTX treatment [13].

Drug resistance to folate analogues may occur via

several mechanisms including: (a) decreased transport via

the reduced folate carrier (RFC), (b) altered levels of

target enzymes, e.g. DHFR and thymidylate synthase

(TS), and (c) decreased ratio of folylpolyglutamate

synthetase (FPGS)/folylpolyglutamate hydrolase (FPGH)

[35]. Multiple studies have documented that poor

long-term survival following chemotherapy for pediatric

acute lymphoblastic leukemia correlates with MTX

resistance in the leukemic blasts and may be predicted

by mRNA expression of folate metabolism-related

proteins in malignant cells. For example, investigators

have found up to 500-fold variation of FPGS in leukemic

blasts with evidence of two to three-fold lower FPGS

mRNA levels and three to four-fold higher DHFR and TS

mRNA levels in MTX resistant leukemia vs. MTX

sensitive disease [24,36]. Reports have also correlated the

effectiveness of MTX-based therapy in pediatric

leukemia patients with the amount of intracellular MTX

incorporated into blast cells at the time of diagnosis

[37,38]. Defective transport via decreased RFC-1

expression has been found to be a common mechanism

of acquired MTX resistance in leukemia [39]. Acquisition

of anti-folate resistance has also been attributed to

enhanced FPGH activity resulting in decreased levels of

intracellular polyglutamylated MTX derivative [40].

Because the same mechanisms affecting MTX

cytotoxicity on tumor cells could also affect the efficacy

of PDX, we employed quantitative RT-PCR to examine

the expression levels of three genes: RFC-1 transporter,

FPGS, and FPGH in the three lymphoma cell lines

utilized for xenograft studies. Our finding that increased

relative RFC-1 gene expression was present in the

lymphoma (HT) line with the highest complete

regression rate (89%) to PDX therapy suggests that one

reason for increased tumor sensitivity to PDX may be

improved intracellular transport into malignant cells due

to enhanced RFC-1 transporter activity. This is consistent

with prior reports demonstrating that in vivo antifolate

uptake into cells, rather than polyglutamyulation, pre-

dicts clinical response [39]. Of note, decreased RFC gene

expression also augments folate uptake, an effect which

may be equally important in antifolate metabolism. One

clinical study of pediatric leukemic patients showed that

while MTX uptake into lymphoblasts alone was not

predictive of treatment success, the ratio of MTX to

TABLE V Relative levels of RFC-1, FPGS, and FPGH mRNA geneexpression in lymphoma cell lines as determined by real-time RT-PCR.Levels were measured relative to b-actin control £ 1000 and wereperformed five times

RFC-1 ðn ¼ 5Þ FPGS ðn ¼ 5Þ FPGH ðn ¼ 5)

HT 0.96 ^ 0.2 4.92 ^ 0.6 1.06 ^ 0.2SKI 0.30 ^ 0.04 6.84 ^ 0.6 1.08 ^ 0.10RL 0.41 ^ 0.1 7.28 ^ 0.8 0.58 ^ 0.08

PDX IN NON-HODGKIN’S LYMPHOMA 1033

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yal

e D

erm

atol

ogic

Sur

gery

on

10/0

1/14

For

pers

onal

use

onl

y.

5-methyltetrahydrofolate correlated positively with

treatment outcome [41]. Direct comparative pharmaco-

kinetic analyses of reduced folate, MTX and PDX

internalization and efflux in these three lymphoma cell

lines are planned.

Alterations in RFC gene expression may occur in tumor

cells as a result of transforming genetic events or as

inherited variants. One study reported that wild type p53

expression in leukemic cells specifically repressed human

RFC-1 gene expression independent of effects on cell

cycle progression [42]. Clusters of mutations in the RFC

gene have also been found in leukemic cells resistant to

antifolates [43]. However, evaluation of single nucleotide

polymorphisms in human RFC-1 identified similar allelic

frequencies of a specific G to A polymorphims in both

leukemic and non-leukemic genomic. DNAs. Although

MTX transport was not affected, small differences in the

transport rates of other antifolates by different RFC-1

forms was detected [44]. The HT and RL human

lymphoma lines have both previously been described to

possess p53 mutations [16]. Whether p53 and other

transforming events, or mutations (inherited or acquired)

in the RFC gene affect folate metabolism in human

lymphoma cells is presently unknown.

The improved activity of PDX over MTX demonstrated

in these preclinical studies has supplied the rationale to

propose a phase II clinical trial of PDX at our Center for

lymphoma patients with refractory or relapsed disease.

The observation that PDX does not cause dose-limiting

myelotoxicity may allow for the enrollment of patients

ineligible for other therapeutic approaches due to prior

hematologic toxicity [14]. Lymphoma tissue from patients

on this study will be evaluated for expression of folate

metabolism genes to determine whether alterations in the

RFC-1 expression correlate with potential clinical

responses.

Further studies will study the efficacy of PDX in

combination with other agents to improve tumor

responses. The development of lymphoma xenotrans-

plantation models should prove particularly valuable in

evaluating multi-drug regiments. In a prior study, we

found that the coadministration of probenecid (an agent

which functions to inhibit the extrusion of folate

analogues from tumor cells) with PDX induced complete

regressions in lung, prostate, and mammary cancer

xenografts while PDX monotherapy was only inhibitory

[21]. We have also demonstrated that the combination

of platinum based chemotheapeutic agents with

PDX improved complete regression rates in human

mesothelioma xenografts [22]. Because PDX has not

been associated with myelotoxicity [14], attractive

candidates for possible combination regimens for

lymphoma include other myelosuppressive agents

known to synergize with MTX (i.e. cytarabine

analogues such as ara-C and gemcitabine), anti-CD20

monoclonal antibodies (with or without radioisotopes),

and novel therapies such as growth factor inhibitors and

anti-angiogenic agents.

Acknowledgements

We would like to acknowledge Kang Zhang for excellent

technical assistance. We also thank Haiying Ju, Jing Chen,

Nushmia Khokhar, Harry Satterwhite, William Tong, and

the Molecular Cytology Core Facility (MSKCC) for their

assistance. The Affymetrix gene expression microarrays

were performed by the Genomics Core Lab Microarray

Facility (MSKCC) with additional analyses performed by

Jane Houldsworth and R.K. Chaganti (Cytogenetics

Laboratory, MSKCC).

Supported in part by NIH CA-09207-24 (MASM, EW),

Experimental Therapeutics Center of MSKCC (MASM,

EW), ASCO Young Investigator Award (EW), NIH

CA-0172(OO) and the Leukemia and Lymphoma

Society (OO).

References

[1] DeVita, V., Hellman, S. and Rosenberg, S. (2001) Cancer:Principles and Practice of Oncology, 6th ed. (Lippincott Williamsand Wilkins, Philadelphia).

[2] Howe, H.L., Wingo, P.A., Thun, M.J., Ries, L.A., Rosenberg, H.M.,Feigel, E.G. and Edwards, B.K. (2001) “Annual report to the nationon the status of cancer (1973–1998), featuring cancers with recentincreasing trends”, J. Natl Cancer Inst. 93, 824–842.

[3] DeVita, V.T.J., Hellman, S. and Rosenberg, S.A., eds (1995)Biologic Therapy of Cancer (Lippincott Williams & Wilkins,Philadelphia).

[4] Dhaliwal, H.S., Rohatiner, A.Z., Gregory, W., Richards, M.A.,Johnson, P.W., Whelan, J.S., Gallagher, C.J., Matthews, J., Ganesan,T.S., Barnett, M.J., et al. (1993) “Combination chemotherapy forintermediate and high grade non-Hodgkin’s lymphoma”,Br. J. Cancer 68, 767–774.

[5] Dorigo, A., Mansberg, R. and Kwan, Y.L. (1993) “Lomustine,etoposide, methotrexate and prednisone (LEMP) therapy forrelapsed and refractory non-Hodgkin’s lymphoma”, Eur.J. Haematol. 50, 37–40.

[6] Sarris, A.H., Phan, A., Duvic, M., Romaguera, J., McLaughlin, P.,Mesina, O., King, K., Medeiros, L.J., Rassidakis, G.Z., Samuels, B.and Cabanillas, F. (2002) “Trimetrexate in relapsed T-celllymphoma with skin involvement”, J. Clin. Oncol. 20, 2876–2880.

[7] McGrath, I., Adde, M., Shad, A., Venzon, D., Seibel, N., Gootenber,J., Neely, J., Arndt, C., Neider, M., Jaffe, E., Wittes, R.A. and Horal,I.D. (1996) “Adults and children with small non-cleaved-celllymphoma have a similar excellent outcome when treated with thesame chemotherapy regimen”, J. Clin. Oncol. 12, 925–934.

[8] DeGraw, J.I., Colwell, W.T., Piper, J.R. and Sirotnak, F.M. (1993)“Synthesis and antitumor activity of 10-propargyl-10-deaza-aminopterin”, J. Med. Chem. 36, 2228–2231.

[9] Sirotnak, F.M., DeGraw, J.I., Moccio, D.M., Samuels, L.L. andGoutas, L.J. (1984) “New folate analogs of the 10-deaza-aminopterin series. Basis for structural design and biochemicaland pharmacologic properties”, Cancer Chemother. Pharmacol. 12,18–25.

[10] Sirotnak, F.M., DeGraw, J.I., Schmid, F.A., Goutas, L.J. andMoccio, D.M. (1984) “New folate analogs of the 10-deaza-aminopterin series. Further evidence for markedly increasedantitumor efficacy compared with methotrexate in ascitic andsolid murine tumor models”, Cancer Chemother. Pharmacol. 12,26–30.

[11] Schmid, F.A., Sirotnak, F.M., Otter, G.M. and DeGraw, J.I. (1985)“New folate analogs of the 10-deaza-aminopterin series: markedlyincreased antitumor activity of the 10-ethyl analog compared to theparent compound and methotrexate against some human tumorxenografts in nude mice”, Cancer Treat. Rep. 69, 551–553.

[12] Sirotnak, F.M., Schmid, F.A., Samuels, L.L. and DeGraw, J.I.(1987) “10-Ethyl-10-deaza-aminopterin: structural design andbiochemical, pharmacologic, and antitumor properties”, NCIMonogr. 5, 127–131.

E.S. WANG et al.1034

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yal

e D

erm

atol

ogic

Sur

gery

on

10/0

1/14

For

pers

onal

use

onl

y.

[13] Sirotnak, F.M., DeGraw, J.I., Colwell, W.T. and Piper, J.R. (1998)“A new analogue of 10-deaza-aminopterin with markedly enhancedcurative effects against human tumor xenografts in mice”, CancerChemother. Pharmacol. 42, 313–318.

[14] Krug, L.M., Ng, K.K., Kris, M.G., Miller, V.A., Tong, W., Heelan,R.T., Leon, L., Leung, D., Kelly, J., Grant, S.C. and Sirotnak, F.M.(2000) “Phase I and pharmacokinetic study of 10-propargyl-10-deaza-aminopterin, a new antifolate”, Clin. Cancer Res. 6,3493–3498.

[15] Pulvertaft, R. (1964) “Cytology of Burkitt’s tumour (Africanlymphoma)”, Lancet 1, 238–240.

[16] Beckwith, M., Longo, D.L., O’Connell, C.D., Moratz, C.M. andUrba, W.J. (1990) “Phorbol ester-induced, cell-cycle-specific,growth inhibition of human B-lymphoma cell lines”, J. Natl CancerInst. 82, 501–509.

[17] Arnstein, P., Taylor, D.O., Nelson-Rees, W.A., Huebner, R.J.and Lennette, E.H. (1974) “Propagation of human tumors inantithymocyte serum-treated mice”, J. Natl Cancer Inst. 52, 71–84.

[18] Ohsugi, Y., Gershwin, M.E., Owens, R.B. and Nelson-Rees, W.A.(1980) “Tumorigenicity of human malignant lymphoblasts:comparative study with unmanipulated nude mice, antilymphocyteserum-treated nude mice, and X-irradiated nude mice”, J. NatlCancer Inst. 65, 715–718.

[19] Goy, A., Gilles, F., Remache, Y., Filippa, D., Portlock, C.S.,Jhanwar, S.C. and Zelenetz, A.D. (2001) “Establishment of a humancell line (SKI-DLCL-1) with a t(1;14)(q21;q32) translocation fromthe ascites of a patient with diffuse large cell lymphoma”, Leuk.Lymphoma 40, 419–423.

[20] Gilles, F., Goy, A., Remache, Y., Shue, P. and Zelenetz, A.D. (2001)“MUC1 dysregulation as the consequence of a t(1;14)(q21;q32)translocation in an extranodal lymphoma”, Blood 95, 2930–2936.

[21] Sirotnak, F.M., Wendel, H.G., Bornmann, W.G., Tong, W.P., Miller,V.A., Scher, H.I. and Kris, M.G. (2000) “Co-administration ofprobenecid, an inhibitor of a cMOAT/MRP-like plasma membraneATPase, greatly enhanced the efficacy of a new 10-deazaaminop-terin against human solid tumors in vivo”, Clin. Cancer Res. 6,3705–3712.

[22] Khokhar, N.Z., She, Y., Rusch, V.W. and Sirotnak, F.M. (2001)“Experimental therapeutics with a new 10-deazaaminopterin inhuman mesothelioma: further improving efficacy through structuraldesign, pharmacologic modulation at the level of MRP ATPases,and combined therapy with platinums”, Clin. Cancer Res. 7,3199–3205.

[23] Eads, C.A., Danenberg, K.D., Kawakami, K., Saltz, L.B.,Danenberg, P.V. and Laird, P.W. (1999) “CpG island hypermethyla-tion in human colorectal tumors is not associated with DNAmethyltransferase overexpression”, Cancer Res. 59, 2302–2306.

[24] Rots, M.G., Willey, J.C., Jansen, G., Van Zantwijk, C.H.,Noordhuis, P., DeMuth, J.P., Kuiper, E., Veerman, A.J., Pieters, R.and Peters, G.J. (2000) “mRNA expression levels of methotrexateresistance-related proteins in childhood leukemia as determined bya standardized competitive template-based RT-PCR method”,Leukemia 14, 2166–2175.

[25] Shultz, L.D., Schweitzer, P.A., Christianson, S.W., Gott, B.,Schweitzer, I.B., Tennent, B., McKenna, S., Mobraaten, L., Rajan,T.V., Greiner, D.L., et al. (1995) “Multiple defects in innate andadaptive immunologic function in NOD/LtSz-scid mice”,J. Immunol. 154, 180–191.

[26] Shultz, L.D., Lang, P.A., Christianson, S.W., Gott, B., Lyons, B.,Umeda, S., Leiter, E., Hesselton, R., Wagar, E.J., Leif, J.H., Kollet,O., Lapidot, T. and Greiner, D.L. (2000) “NOD/LtSz-Rag 1 nullmice: an immunodeficient and radio resistant model for engraftmentof human hematolymphoid cells, HIV infection, and adoptivetransfer of NOD mouse diabetogenic T cells”, J. Immunol. 164,2496–2507.

[27] Piacibello, W., Sanavio, F., Severino, A., Dane, A., Gammaitoni, L.,Fagioli, F., Perissinotto, E., Cavalloni, G., Kollet, O., Lapidot, T.and Aglietta, M. (1999) “Engraftment in nonobese diabetic severecombined immunodeficient mice of human CD34(þ ) cord bloodcells after ex vivo expansion: evidence for the amplification andself-renewal of repopulating stem cells”, Blood 93, 3736–3749.

[28] Kollet, O., Peled, A., Byk, T., Ben-Hur, H., Greiner, D., Shultz, L.and Lapidot, T. (2000) “Beta2 microglobulin-deficient (B2m(null))NOD/SCID mice are excellent recipients for studying human stemcell function”, Blood 95, 3102–3105.

[29] Hudson, W.A., Li, Q., Le, C. and Kersey, J.H. (1998)“Xenotransplantation of human lymphoid malignancies is

optimized in mice with multiple immunologic defects”, Leukemia12, 2029–2033.

[30] Alizadeh, A.A., Eisen, M.B., Davis, R.E., Ma, C., Lossos, I.S.,Rosenwald, A., Boldrick, J.C., Sabet, H., Tran, T., Yu, X., Powell,J.I., Yang, L., Marti, G.E., Moore, T., Hudson, J., Lu, L., Lewis,D.B., Tibshirani, R., Sherlock, G., Chan, W.C., Greiner, T.C.,Weisenburger, D.D., Armitage, J.O., Warnke, R., Staudt, L.M., et al.(2000) “Distinct types of diffuse large B-cell lymphoma identifiedby gene expression profiling”, Nature 403, 503–511.

[31] Shipp, M.A., Ross, K.N., Tamayo, P., Weng, A.P., Kutok, J.L.,Aguiar, R.C., Gaasenbeek, M., Angelo, M., Reich, M., Pinkus, G.S.,Ray, T.S., Koval, M.A., Last, K.W., Norton, A., Lister, T.A.,Mesirov, J., Neuberg, D.S., Lander, E.S., Aster, J.C. and Golub, T.R.(2002) “Diffuse large B cell lymphoma outcome prediction by geneexpression profiling and supervised machine learning”, NatureMed. 8, 68–74.

[32] Sierra, E.E. and Goldman, I.D. (1999) “Recent advances in theunderstanding of the mechanism of membrane transport of folatesand antifolates”, Semin. Oncol. 26, 11–23.

[33] Kamen, B. (1997) “Folate and antifolate pharmacology”, Semin.Oncol. 24, S18-30–S18-39.

[34] Sirotnak, F.M. and Tolner, B. (1999) “Carrier-mediated membranetransport of folates in mammalian cells”, Annu. Rev. Nutr. 19,91–122.

[35] Mauritz, R., Peters, G.J., Priest, D.G., Assaraf, Y.G., Drori, S.,Kathmann, I., Noordhuis, P., Bunni, M.A., Rosowsky, A.,Schornagel, J.H., Pinedo, H.M. and Jansen, G. (2002) “Multiplemechanisms of resistance to methotrexate and novel antifolates inhuman CCRF-CEM leukemic cells and their implications for folatehomeostasis”, Biochem. Pharmacol. 63, 105–115.

[36] Lenz, H.J., Danenberg, K., Schnieders, B., Goeker, E., Peters, G.J.,Garrow, T., Shane, B., Bertino, J.R. and Danenberg, P.V. (1994)“Quantitative analysis of folylpolyglutamate synthetase geneexpression in tumor tissues by the polymerase chain reaction:marked variation of expression among leukemia patients”, Oncol.Res. 6, 329–335.

[37] Whitehead, V.M., Vuchich, M.J., Lauer, S.J., Mahoney, D., Carroll,A.J., Shuster, J.J., Esseltine, D.W., Payment, C., Look, A.T.,Akabutu, J., et al. (1992) “Accumulation of high levels ofmethotrexate polyglutamates in lymphoblasts from children withhyperdiploid (greater than 50 chromosomes) B-lineage acutelymphoblastic leukemia: a Pediatric Oncology Group study”, Blood80, 1316–1323.

[38] Whitehead, V.M., Payment, C., Cooley, L., Lauer, S.J., Mahoney,D.H., Shuster, J.J., Vuchich, M.J., Bernstein, M.L., Look, A.T.,Pullen, D.J. and Camitta, B. (2001) “The association of the TEL-AML1 chromosomal translocation with the accumulation ofmethotrexate polyglutamates in lymphoblasts and with ploidy inchildhood B-progenitor cell acute lymphoblastic leukemia: aPediatric Oncology Group study”, Leukemia 15, 1081–1088.

[39] Gorlick, R., Goker, E., Trippett, T., Steinherz, P., Elisseyeff, Y.,Mazumdar, M., Flintoff, W.F. and Bertino, J.R. (1997) “Defectivetransport is a common mechanism of acquired methotrexateresistance in acute lymphocytic leukemia and is associatedwith decreased reduced folate carrier expression”, Blood 89,1013–1018.

[40] Rhee, M.S., Wang, Y., Nair, M.G. and Galivan, J. (1993)“Acquisition of resistance to antifolates caused by enhancedgamma-glutamyl hydrolase activity”, Cancer Res. 53, 2227–2230.

[41] Mantadakis, E., Smith, A.K. and Kamen, B.A. (2000) “Ratio ofmethotrexate to folate uptake by lymphoblasts in children withB-lineage acute lymphoblastic leukemia: a pilot study”, J. Pediatr.Hematol. Oncol. 22, 221–226.

[42] Ding, B.C., Whetstine, J.R., Witt, T.L., Schuetz, J.D. and Matherly,L.H. (2001) “Repression of human reduced folate carrier geneexpression by wild type p53”, J. Biol. Chem. 276, 8713–8719.

[43] Drori, S., Jansen, G., Mauritz, R., Peters, G.J. and Assaraf, Y.G.(2000) “Clustering of mutations in the first transmembrane domainof the human reduced folate carrier in GW1843U89-resistantleukemia cells with impaired antifolate transport and augmentedfolate uptake”, J. Biol. Chem. 275, 30855–30863.

[44] Whetstine, J.R., Gifford, A.J., Witt, T., Liu, X.Y., Flatley, R.M.,Norris, M., Haber, M., Taub, J.W., Ravindranath, Y. and Matherly,L.H. (2001) “Single nucleotide polymorphisms in the humanreduced folate carrier: characterization of a high-frequency G/Avariant at position 80 and transport properties of the His(27) andArg(27) carries”, Clin. Cancer Res. 7, 3416–3422.

PDX IN NON-HODGKIN’S LYMPHOMA 1035

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yal

e D

erm

atol

ogic

Sur

gery

on

10/0

1/14

For

pers

onal

use

onl

y.