1. No category

From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
1995 85: 2537-2545
Biotherapy for xenografted human central nervous system leukemia
in mice with severe combined immunodeficiency using B43
(anti-CD19)- pokeweed antiviral protein immunotoxin
R Gunther, LM Chelstrom, L Tuel-Ahlgren, J Simon, DE Myers and FM Uckun
Updated information and services can be found at:
http://www.bloodjournal.org/content/85/9/2537.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
Biotherapy for Xenografted Human Central Nervous System Leukemia in
Mice With Severe Combined Immunodeficiency Using B43 (Anti-CD19)Pokeweed Antiviral Protein Immunotoxin
By Roland Gunther, Lisa M. Chelstrom, Lisa Tuel-Ahlgren, John Simon, Dorothea E. Myers, and Fatih M. Uckun
The study of central nervous system (CNS) leukemia has
been hampered by the lack of a suitable animal model.We
report that severe combined immunodeficiency (SCID) mice
invariably develop rapidly progressive fatal CNS leukemia
within 3 weeks after intravenous injectionof NALM-6 pre-B
acute lymphoblasticleukemia (ALL) cells. Colonization the
of
dura mater andsubarachnoidspace, usually of the distal
spinalcord with occasionalextension into the VirchowRobin spaces of blood vessels subjacent to the meninges,
followed involvement of bone marrow in the skull, vertebrae, and, occasionally, the appendicular skeleton. Occult
CNS leukemiawas detectable by polymerase chain reaction
amplification of human DNA as early as 8 days postinoculation of leukemia cells.We used this in vivo model of human
CNS leukemiato examine the therapeutic efficacy and toxicity of intrathecally administered B43 (anti-CDlS)-pokeweed
antiviral protein (PAP), an anti-B-lineage ALL immunotoxin
directedagainst the pan-B-cellantigenCD19/Bp95.Intrathecal therapy with B43 (anti-CD19)-PAPimmunotoxin at
nontoxic dose levels significantly
improved survival of SCID
mice andwas superior to intrathecal methotrexate therapy.
0 7995 by The American Society of Hematology.
P
intrathecal CNS therapy with B43 (anti-CD19)-pokeweed
antiviral protein (PAP) immunotoxin, a new anti-ALL agent
that does not cross the blood-brain b a r ~ i e r , ' ~was
. ' ~ more
effective in achieving long-term event-free survival (EFS)
than systemic therapy with the same agent, suggesting that
seeding of BM and other extramedullary sites by blasts from
the CNS plays a major role in progression of leukemia in
this model system.
The SCID mouse model of human CNS leukemia described in this report is devised to provide a basis for future
comparative studies of the toxicity and efficacy of new CNS
therapy regimens in B-lineage ALL. A major finding of clinical significance reported here is that leukemic blasts in the
CNS can be effectively eradicated by weekly intrathecal
administration of B43-PAP immunotoxin without side effects. These experiments provide a paradigm for the critical
evaluation of a new CNS therapy regimen in B-lineage ALL.
ROPHYLACTIC central nervous system (CNS) therapy,
which consists of intrathecal chemotherapy with or
without high-dose intravenous (IV) methotrexate or cranial/
craniospinal irradiation in conjunction with intrathecal chemotherapy, has reduced the incidence of CNS relapse in
patients with acute lymphoblastic leukemia (ALL) to 5% to
However, patients who develop a CNS relapse despite prior prophylactic therapy have only a 15% to 45%
chance of long-term disease-free survival (DFS).'s','o," The
primary cause of treatment failure in these patients is the
recurrence of leukemia in the CNS as well as other extramedullary sites and bone marrow (BM).',7,10,"Patients who sustain a second CNS relapse have a dismal outcome."." Furthermore, contemporary CNS radiochemotherapy regimens
are often associated with prolonged myelosuppression and
neurotoxicity."" Therefore, more effective and less toxic
CNS prophylaxis and treatment strategies to prevent seeding
of leukemia cells from the CNS to other extramedullary
sites and BM, along with more effective systemic therapy
to prevent reseeding of the CNS with clonogenic leukemia
cells, are needed to further improve long-term DFS after a
recurrence in the CNS.
Until recently, the study of new therapeutic modalities for
CNS leukemia has been hampered by the lack of a suitable
model system. Before clinical trials are initiated in patients
with CNS leukemia, information regarding an agent's in vivo
activity is desirable. Clonogenic assays are useful indicators
of the in vitro antileukemic efficacy of new drugs against
freshIy isolated primary ALL blasts. However, in vitro clonogenic assays do not provide direct evidence for the in vivo
activity of a new agent, nor do they provide an estimate
of toxicity. Furthermore, clonogenic assays do not offer an
opportunity to evaluate the toxicity and efficacy of
multiagent therapeutic regimens. Therefore, we were motivated to establish an in vivo model system of human CNS
leukemia.
We and others have used mutant CB.17 mice with severe
combined immunodeficiency (SCID) as a model system to
examine the in vivo homing, engraftment, and growth patterns of normal and malignant human hematopoietic celis.'2"s
In this report, we show that NALM-6 human pre-B ALL
cells cause overt and fatal CNS leukemia when injected
IV into SCID mice. Intriguingly, histologically detectable
engraftment of leukemia cells in skull or vertebral BM and
meninges preceded engraftment in other sites. Furthermore,
Blood, Vol 85,No 9 (May l), 1995: pp 2537-2545
MATERIALS AND METHODS
SCID mice. SCID mice were produced by specific pathogenfree (SPF)CB-l7 SCID/SCID breeders in the American Association
for the Accreditation of Laboratory Animal Care (AAALAC)-accredited Research Animal Resources Facility of the University of
Minnesota (Minneapolis, MN). SCID mice were housed inmicroisolator cages containing autoclaved food, water, and bedding. Trimethoprim-sulfamethoxazole was added to the drinking water for a 3day period each week.
From the Biotherapy Program, the Departments of Therapeutic
Radiology-Radiation Oncology, Pediatrics, Pharmacology, and
Comparative Medicine, Universiry of Minnesota, Minneapolis.
Submitted June 28, 1994; accepted December 15, 1994.
Supported by the Parker HughesTrust and US Public Health
Service Grants No. CA-42633, CA-44114, CA 42111,CA-21737,
CA-60437, CA-13539,and CA-61549 fromthe National Cancer Institute, Department of Health and Human Services. F.M. U. is a
Stohlman Scholar of the Leukemia Society of America.
Address reprint requests to Fatih M. Uckun, MD, PhD, University
of Minnesota Biotherapy Program, 2685 Patton Rd, Roseville, MN
55113.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 I995 by The American Society of Hematology.
0006-4971/95/8509-0021$3.00/0
2537
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
GUNTHER ET AL
2538
Cells and inoculation. The previously described human NALM6-UMI (NALM-6) pre-B ALL cell lineI7 was maintained by serial
passage in RPM1 1640 medium (GIBCO Laboratories, Grand Island,
NY) supplemented with 10% (vol/vol) heat-inactivated fetal bovine
serum (Hyclone Laboratories, Logan, UT), 50 pg/mL streptomycin,
50 IU/mL penicillin, 2 m m o l L L-glutamine, and 10 mmollL HEPES
buffer.Cellswereculturedintissuecultureflasksat
37°C ina
humidified 5% COn atmosphere. Before injection, cells were washed
twice in phosphate-buffered saline (PBS) and resuspended in PBS
at 2.5 X IO'/mL or S X IO'lmL. Inoculation was into the dorsal tail
vein.
Drugs. B43-PAPimmunotoxinwasproducedintheImmunotoxin Production Facility of the University of Minnesota Biotherapy
(25 mglmL) was
Program, as previously described." Methotrexate
obtained from Lederle Parenterals Inc (Carolina, Puerto Rico).
Intrathecal and systemic therapy
for xenografedhuman CNS leuuskemia. Intrathecal injections were made into the lumbar space
ing a syringe with a 25-gauge needle, and 20
ng substance P (total
volume, 1 pL) wascoadministered to confirmthesuccessofthe
intrathecalinjections;substancePinducesacaudallydirected
scratching and biting response when injected into the lumbar space
of mice." To evaluate the efficacy of intrathecal therapy with843PAP immunotoxin, 6-week-old male SCID mice were inoculatedin
thedorsaltail vein with 1 X lo6 NALM-6cellsandtreatedwith
weekly intrathecal injections of B33-PAP (0.5 pg, I .0 pg, 2.0 pg,
2.5 pg, or 10 p g per dose in IO p L normal saline) on days 14, 21,
28, and 3.5. For comparison, control mice were treated with intrathecal normal saline, methotrexate,
or systemic (ie, intraperitoneal) B43PAP. Intrathecal treatments of
15 pg methotrexate (total injection
volume, 10 pL) diluted with normal saline were delivered weekly
for 4 weeks according to the same schedule as for intrathecal B43PAP. Mice were observed daily for evidence of disease and killed
when moribund or unable to obtain foodor water because of paralysis. Long-termsurvivingmicewereelectivelykilled
at 11 I days
afterinoculation of NALM-6cells. For eachtreatmentgroup,at
least two mice that died early in the observation period and two
mice that diedlatewerehistologicallyexamined
to assesstheir
CNS and systemic leukemia burden. NALM-6-inoculated,untreated
control mice were killed at 14 days postinoculation to assess their
leukemiaburdenatthetime
of initiation of treatment.Afterthe
discovery of drug-related paralysis in SCID mice after intrathecal
administration of B43-PAP,single IO- to75-pgintrathecaldoses
were administered to Balb/c mice to further characterize the histologic lesions of the spinal cord.
Exatninarion of SCID mouse tissuesf o r xenografted huntan leukemiacells. Atthetime of death,necropsieswereperformed,and
tissues, including BM and brain, of SCID mice were collected for
evaluation of the human pre-B ALL leukemia burden. Tissues from
mice that died or wereeuthanized by exposure to CO2 gaswere
fixed with 10% neutral buffered formalin by perfusion via the lef
ventricle or by immersion.Theaxialandappendicularskeleton,
containing the CNS and BM, were subsequently decalcified i n 10%
formic acid for 48hours. Representative portions of tissues, including multiple sections of all regions of the spinal column, were dehydratedandembedded in paraffin by routinemethods.Six-micron.
hematoxylin and eosin-stained sections were examined. For immunophenotypic studies, leukemic cells were dissociated from the leptomeninges of the brain by gently vortexing the whole brain in PBS
in a test tube. BM suspensions were preparedby flushing the marrow
cavities of long bones with PBS using a syringe with a 25-gauge
needle.Othertissuesweregentlyhomogenized
in PBS.Multiparameter flow cytometric analyses using two-color
immunofluoreccence staining techniques and pairwise combinations
of a selected
panel of fluorochrome[fluoresceinisothiocyanate(FITC)
o r phycoerythrin (PE)]-labeled monoclonal antibodies (MoAbs) were peras previously described.""' The
formedonthecellsuspensions,
following MoAbs were included in the analyses: B43 (anti-CD19).
J-S (anti-CD10; Coulter Corp. Hialeah, FL), Leu- 12 (anti-CD19).
Leu- 16 (anti-CDW), Hle- 1 (anti-CD45; Becton-Dickinson, RutherHuford, NJ), and anti-lgM p-chain (Tago, Inc, Burlingame, CA).
man DNA was detected in mouse tissues by amplifying a
1 IO-bp
fragment from the first exon of the human P-globin gene
using a
previously published polymerase chain reaction (PCR)
Sfutisticalunulyses. Eventtimesweremeasuredfromtheday
of inoculation of NALM-6 cells to the day of paraplegia or death.
The probabilitiesof paraplegia and EFS were determined, and eventfree interval curves were generated using the Kaplan-Meier product
limit method, as previously r e p ~ r t e d . ' ~The
" ~ log-rdnk test was used
to assess the effectsof various treatment protocols on the probability
of EFS of SCID mice inoculated with NALM-6 cells.
RESULTS
Human pre-B ALL cell line N A M - 6 causes fatul CNS
leukemia in SCID mice. Of 83 SCID mice challenged with
an IV inoculum of 1 X 10' NALM-6 cells, 83 developed
symptomatic CNS leukemia. The first clinical sign of CNS
leukemia was a muscle weakness in the rear legs with a
slight lowering of the lower back at a median of 32 days
(mean ? SE, 33 ? 1 days; range, 26 to 58 days). Subsequently, mice developed a rapidly progressive paralysis of
the rear legs, and at a median of 39 days (range, 36 to 56
days), they were nolongerableto
obtain food or water
Table 1. Kinetics of Colonization of SCID Mouse Tissues by NALM-6 Cells
Days After Inoculation With 5 x lo6 NALM-6 Cells
Tissue
~~
_________~
BM
Brain
Dura mater of the brain
Leptomeninges of the brain
and/or
Assay
1
PCR
Histopathology
PCR
Histopathology
-
2
Histopathology
~
~
~
-
~
-
~
29
5
8
11
14
17
20
23
-
+
+
+
+
N
+D
N
+D
ND
+
+
+
+
t-
i
l
-
l
+
I
ND
ND
ND
_____
-
+
~
-
-
-
~
-
+
26
+
+
+
+
Groups of three SClD mice were killed o n the indicated days after inoculation with 5 x lo6 NALM-6 cells, and assays were performed as
detailed in Materials and Methods. PCR was performed on whole brain without dura mater. Histology of the CNS was performed on sections
of the brain and spinal cord.
Abbreviations: PCR,PCR amplification of human 0-globin gene sequences; ND, not done; f, detection byPCR or histopathology in three of
three mice; -, absence of leukemic cells byPCR or histopathology in three of three mice.
* First day any animal in a group was positive
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
BIOTHERAPY OF CNS LEUKEMIA
Fig l. Colonization of dura mater andsubarachnoid space of SClD
mice by human leukemia cells. (A) Caudal spinal column of a paralyzed SClD mouse showing extensive colonization of the B M space
of the vertebra (BM), theepidural space (ES), leptomeninges of spinal
cord, and sheaths of ventral nerve roots. Hematoxylin andeosin (H&
E) stain; original magnification (OM),
x 25. (B, C) Ventralcaudal spinal
cord andventralnerverootsshowingcolonization
of sheaths of
nerve roots as well as of subarachnoid space around the ventral
spinal artery (arrow) of spinal cord. H&E stain; O M (B), x 5 0 ; IC). x 100.
because of paralysis. Some of these mice suffered a generalized tonic-clonic seizure followed by sudden death. whereas
others either died without a preceding seizure or they were
electively killed in moribund condition. Postmortem histopatho-
2539
logic studies showed that the CNS was extensively involved,
with dense leukemic cell infiltrates in the subarachnoid space
and the superficial perivascular Virchow-Robin spaces (data
not shown). Large numbers of CDIO'CDI9'CD20- human
NALM-6 leukemia cells could be dissociated from the leptomeninges of the brain by gently vortexing the whole brain of
paralyzed SCID mice in PBS in a test tube.
To further study the pathobiology of xenografted human
CNS leukemia. SCID mice were inoculated in the dorsal tail
vein with S X 10" NALM-6 cells. Micewere observed daily,
and groups ofthreemicewerekilled
on days I , 2. and S
and at 3-day intervals fromR days to 30 days postinoculation
and subjected to pathologic analysis to assess their leukemia
burden.Xenograftedhumanleukemia
cells were first detected in the BM, always starting in the skull or vertebrae.
All micehadhistologically detectable BM involvement in
the skull or vertebrae and occasionally theappendicular skeleton by day I 1 (Table I). NALM-6 cells in hematoxylin and
eosin-stained sections of paraffin-embedded mouse BM were
easily distinguished from murine hematopoietic cells. Leukemic cells were 10- to 13-pm in diameter. with oval to irregularly polygonal nuclei with smooth borders, finely stippled
chromatin, anda single prominent eosinophilic nucleolus.
They hadscant.lightlybasophilic
cytoplasm anda high
mitotic rate. Colonizationof the dura mater andsubarachnoid
space,usuallyofthedistalspinal
cord,was observed as
early as 20 and 23 days postinoculation, respectively, and
colonizationof both sites wasaconstantfeatureafter
23
days (Table I , Fig 1 ). Histologically evident CNS involvement was limited to the dura mater and subarachnoid space,
with occasional extension into the Virchow-Robin spaces of
blood vessels subjacent to the meninges: no leukemic cells
wereseen in theVirchow-Robin spaces ofbloodvessels
deep within the parenchyma of the CNS.
The kineticsof engraftment of NALM-6 cells in SClD
mouse tissues was also examined by PCR analysis. Groups
of three mice were killed on days I . 2, and S and at 3-day
intervalsfrom day 8 to day 20. and theiroccultleukemia
burden was evaluated by PCR amplification of human DNA,
as described.".'* As shown in Fig 2A. no PCR evidence of
occult CNS leukemia was found in SClD miceat 1 to 5
days afterinoculation of NALM-6 cells. By contrast, positive
control SCID mice that died of disseminated leukemia with
CNS involvement showed strong PCR positivity for human
@globin gene sequences in DNA specimens from BM and
brain (Fig 2B). Similarly. PCR analysis of SClD mouse BM
and braininvariablyshowedpresenceofhumanDNA
at
8. 1 l , 14, 17, and 20 days (Fig 2C through E). Although
histologically detectable engraftmentofleukemia
cells in
skull or vertebral BM and meninges preceded engraftment
in other sites. PCR analysis of human DNA detected leukemic cell engraftment in extraneural sites (eg. liver) as early
as in the CNS and BM (Fig 2C).
Several features suggest that a major mechanism of colonization of the meninges by NALM-6 cells was by direct
extensionfromthe
BM ofthe calvariaand vertebrae. ( I )
Colonization of the BM of the calvaria and vertebrae always
occurred earlier in the course of disease than did colonization
of the meninges. (2) Meningeal disease was always accompanied by leukemicinvolvementoftheadjacent
BM. (3)
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
GUNTHER ET AL
2540
C
SClD 4287 SCID 4288 SClD 4293
(Ed)
I
D
'
E
(206)
7
-t-r
led)
t
,,
(Ed)
SCID 4249 SClD 4253
l 1 W
I,Id)(1
,pas
,,
NEG
BM BR LIV BM BR LIV BM BR LIV BM BR LIV BM BR LIVCON -CON
SC;y4r2
BM LIV
SCID4228
(17d)
BR
CH "BM
SClD 4231
Pod)
SClD 4933
(14a
PUS
LIV BR CH 'CONI
-
SClD
-
4802 BM
POS
CON 2
-
NEG
CON
SClDSClD4240
4237
P
-O
-S
BM BR LIV SPL BM BR LIV SPL BM BR LIV SPL BM BR LIV SPL CON
-
NEG
CON
Fig 2. Kinetics of human leukemic cell engraftment in SClD mice.
DNA specimens from various organswere analyzed for the presence
of human p-globin gene sequences by a previously published PCR
m e t h ~ d . ' ~(A)
' ' ~Lack of engraftment on days 1,2, or 5 after inoculation of NALM-6 leukemia cells. (B) Presence of human DNA in BM
and brain of SClD mice electively killed after becoming paralyzed 33
(SCID mouse 47921,or 36 days
days(SCIDmouse47911,37days
(SCID mouse 4796) after inoculationwith NALM-6 leukemia cells. (C
through E) Presence of human DNA in BM, brain, and liver of SClD
miceelectively killed 8 to 20 days after inoculation with NALM-6
leukemia cells. The time of elective killing (ie, days after inoculation)
is indicated under each SClD mouse number. BR, brain; LW, liver;
SPL, spleen; CH, spinal cord.
Colonization of the endosteum occurred earlier than colonization of the dura mater, and colonization of the dura mater
was alwaysaccompanied by involvementoftheadjacent
endosteum. (4) Colonization of the subarachnoid space was
always accompanied by involvement of the adiacent dura.
whereas thereverse was not always true. ( 5 ) Streams of
leukemiccellscouldbe
seenenteringthe
CNSfrom the
BM by direct penetration through the cortex. along vascular
channels or along nerve roots (Fig 3).
lntrrrtheccrl R43 (anti-CD19)-PAP therap! is cflectiw
agrrinst .renogrqfted human CNS leukemia in SClD mice.
Occult leukemia of the CNS was shown by PCR analysis of
brains at the time of initiation of treatments (14 days after
inoculation of I X IOh NALM-6 cells) in four of four SCID
mice that were randomly selected at the time of assignment
to treatmentgroups.Histologic
examination of the CNS
showed leukemic cells in one of six mice at this timepoint.
All 23 saline-injectedcontrolmicebecameparalyzedand
developed histologically demonstrableCNS and systemic
leukemia. Intrathecal administration of methotrexate did not
significantly increase survival of the SClD mice, and all I O
micethatreceivedintrathecalmethotrexatetreatmentdied
with extensive CNS leukemia at a median of 49 days (Table
2. Fig 4). Three mice treated with intrathecal injections of
unconjugated B43 antibody (2.5 pg per dose) died of leukemia at 46. 52, and S9 days. respectively (data not shown).
Similarly, all mice treated with weekly intraperitoneal B43PAP injections died of disseminated leukemia at a median
of 44 days (Table 2). The presence of xenografted NALM6 cells in BM andor CNSof SCID mice treated with normal
saline, intraperitoneal B43-PAP, or intrathecal methotrexate
was confirmed by PCR detection of human DNA (Fig SA
andB).Intrathecaladministration
of B43-PAP prolonged
survival of SClD mice inoculated with NALM-6 leukemia
cells(Table 2, Fig 4). The mediansurvivaltime
of mice
administered weekly intrathecal doses of 2.5 pg B43-PAP
(ie. 68 days) was significantly greater than the median survival time of the saline control group (ie. 41 days) or the
methotrexate-treated group (ie, 49 days: P < .OOOl: Table
2). Two mice in the 2.5 pg intrathecal B43-PAP group were
in good health at the end of the I I 1 day observation period
and had no histologic evidence of leukemia in the CNS or
BM. The human DNA contamination in the BM or brain of
these mice appeared to be less than 0.001% (Fig SA. C. and
B43-PAP treatment
D). For each ofthethreeintrathecal
groups receiving B43-PAP at 1.0. 2.0, or 2.5 pg per dose,
four mice that died during the 120-day observation period
were evaluated histologically. All 12 mice had disseminated
leukemia with CNS involvement.
Systemic administration of B43-PAP does not provide an
alternative to intrathecal B43-PAP therapy for xenografted
CNS leukemia,as B43-PAP immunotoxin molecules are too
large to cross the blood-brain barrier. As shown in Table 2,
intraperitoneal administration of 2.5 pg B43-PAP, given at
the same dose scheduleas the intrathecal B43-PAP regimen.
did not increase survival, indicating that intrathecal administration was the key factor in increasing survival time. Intraperitoneal administration of 20 pg B43-PAP in three equally
divided consecutive daily doses on days 14, 1S , and 16 after
inoculation with NALM-6 cells significantly prolonged survival (median survival, 65 days; P < .02), but all mice in
this group died with CNS leukemia (Table 2).
There was no histologic evidence of neurotoxicity in the
mice that received intrathecal B43-PAP doses up to 2.5 pg.
However. all 10 mice that received intrathecal injections of
IO pg B43-PAP developed rear leg paralysis shortly after
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
BIOTHERAPY OF CNS LEUKEMIA
2541
Fig 3. Penetration of human
leukemia cells from BM into
CNS.Caudalspinal
column of
a paralyzed SClD mouse showing streams of leukemia cells
(arrows) penetrating from the
vertebral BM space (BM) to the
epiduralspace IESI. H&E stain;
OM: (A), x 50; (B), x 200.
the onset of treatment (Table 2). All but one of these mice
only developed signs of neurotoxicity after two or more
injections of B43-PAP. All six histologically examined mice
in this group had severe damage of the terminal sacral spinal
cord (site of injection), characterized by severe necrosis of
grayandwhitematter,
as well as degeneration of tracts
within the ventral nerve roots of this region (Fig 6). There
were no morphologic lesions of the blood vessels.Neurotoxicity wasthe cause of paralysis in these mice. as no histologic
or PCR evidence of CNS leukemia was found.
DISCUSSION
We report that the SClD mouse microenvironment permits
the development of a rapidly progressive and invariably fatal
CNS leukemia after inoculation with human NALM-6 preB ALL cells via a tail vein. Also provided is experimental
evidence thatintrathecalB43
(anti-CD I9)-PAP immunotoxin therapy at a nontoxic dose level significantly improves
survival of SClD mice with xenografted CNS leukemia.
The first histologic evidence of CNSleukemia in this
modelisoftenfound
in the distal spinal cord.preceding
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
2542
GUNTHER ET AL
Table 2. Antileukemic Efficacy of Intrathecal B43 (anti-CDlS)-PAP
Cumulative Proportion Surviving
Event-Free (%)
Treatment Regimen
NS it qwk
x
No. of Mice
4 wk
0.5 p g 643-PAP it qwk x 4 wk
1.0 p g 643-PAP it qwk x 4 wk
2.0 p g 643-PAP it qwk x 4 wk
2.5 p g 643-PAP it qwk x 4 wk
10.0pg 643-PAP it qwk x 4 wk
23
2.5 @g643-PAP ip qwk x 4 wk
20 p g B43-PAP ip divided qd x 3d
15 pg methotrexate it qwk x 4 wk
Median EFS (dl
30 d
60 d
0
0
10
20
60
0
9
10
10
41
42
36
40
10
68*
100
100
80
100
100
10
29
37
10
10
44
65
100
100
10
49
100
20
90
10
90 d
0
0
0
10
20
0
0
0
0
SClD mice were injected IV with 1 x lo6 NALM-6 cells. Administration of it B43-PAP or methotrexate was done weekly for 4 weeks, starting
on day 14 after inoculation of leukemic cells. Controls were treated with normal saline (NS). Mice were monitored for paralysis and survival
and data analysis was performed as described in Materials and Methods.
Abbreviations: it, intrathecal; ip, intraperitoneal; qwk, every week.
* P < .0001.
involvement of the brain. The reason for this predilection is
not known, butmaybe
related to inoculation withthe
NALM-6 cells into the tail vein. There are anastomoses between abdominal veins and vertebral veins in other species,
suggesting that these maybe avenues of colonization of
the CNS.*' Of interest isthe finding that one avenue of
colonization of the meninges of the brain and spinal cord
appears to be direct extension from adjacent BM, as has been
suggested by other^.^^,^^ This may not be the consequence of
specific tropism of these cells for the CNS, however, as
invasion of the adjacent skeletal muscle from the BM also
commonly occurred.
PCR amplification detected human DNA in the CNS of
"p-?
1.o
b
.
----O"- NS ( b 2 3 )
..
$
&
J
0.6
* .
--+-MTX (N-10)
h !
T !
: !
Fig 4. Efficacy
of
intrathecal B43-PAP immunotoxin therapy
aaainst
xenoarafted
human CNS leukemia in SClD mice. SCIDmice
"
"
were injected IV with 1 x lo6 NALM-6 cells. W - P A P (2.5p g per
dose)or methotrexate l15 @gperdose) was administered intrathecally weakly for 4 waeks starting on
day 14 after inoculation with
leukemia cells. Controls receivednormal saline. The cumulative proportion of mice surviving event-freeare shown accordingtothe numdays
ber
of
after inoculation of leukemia cells.
the SCID mice before histologically demonstrable proliferation of NALM-6 cells in the adjacent BM, suggesting that
colonization of the CNS may also occur by a means other
than direct extension from the marrow. These leukemic cells
in the CNS may account for the eventual disseminated leukemiain mice treated only by intraperitoneal administration
of B43-PAP, as seen in this and a previous study."
Intrathecal therapy for occult CNS leukemia with B43PAP significantly prolonged survival of SCID mice inoculated 14 days previously with 1 X lo6 NALM-6 pre-B ALL
cells and resulted in the apparent cure of several mice. Intraperitoneal therapy with B43-PAP (20 pg total dose administered on 3 consecutive days) was not as efficacious as intrathecal therapy, as would be expected from therapy that
does not pass the blood-brain barrier. However, it increased
survival of SCID mice, suggesting that it eliminated many,
if not all, accessible leukemic cells. PCR assay showed that
human leukemic cells were present in the CNS of the SCID
mice at the initiation of intraperitoneal treatment, suggesting
that these cells may be the progenitors of the systemic leukemia found at the time of death in these mice.
Interestingly, mice treated intrathecally or intraperitoneally with 22.0 pg B43-PAP and that died with leukemia
early (less than 45 days) in the course of the study tended
to have only leukemic involvement of the spinal cord and
adjacent vertebral BM, whereas mice that died later (greater
than SO days) had disseminated disease with leukemic
involvement of the spinal cord, BM, and viscera. Therefore,
we speculate that (1) CNS leukemia causes paralysis that
necessitates euthanasia before severe morbidity due to
involvement of BM or viscera occurs, and (2) the CNS serves
as a source of cells that seed extraneural sites. According to
this hypothesis, incomplete elimination of leukemia cells in
the CNS
have resulted inParalysis from CNS leukemia
before dissemination of the disease, whereas late deaths from
disseminated leukemia could have occurred if a small number of
cells in the CNS survived therapy, lea&ng
to Slow colonization of the CNS and migration to extraneural
tissues before CNS involvement resulted in paralysis. This
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
2543
BIOTHERAW OF CNS LEUKEMIA
‘
MTX
B43PAP
B43 PAP (i.p)
NS
4003
4000 “ 4 7 9477 9 9
d56
d54 d41d41
(it)
”
4044
d53
(i.t)
”
4943
dl11
‘
- n n - 1 - 1 -
BM BR BM BR BM BR BM BR BM BR BM BR
- -
NEG
CON
POS
CON
A
MTX fi.t)
4041
d43
405Q‘
405Q‘
d61
r-lPOS
BR BM BR BM BR BM BR ‘BM
’BM BR’ BM
CON
4042
d40
4045
4046
d49
d44
“
m
n
n
BM BR
BM
4047
d40
4040
d40
-
NEG
CON
B
Fig 5. Absence of occult CNS leukemia after intrathecal (i.t.1B43-PAP immunotoxin therapy. (A, B)
PCR detection of human DNA in BM andlor brain
from normalsaline (NSI-, i.t. methotrexate-, or intraperitoneal (i.p.1 B43-PAP-treated SClD mouse
i.t. B43controls. SClD mouse 4943 was treated with
PAP (2.5 p g per dose every week x 4 weeks). (C, D)
BM and brain
specimens from normalSClD mice (not
inoculated with any human cells) were mixed with
various amounts of NALM-6 DNA t o determine the
sensitivity of the PCR assay. Less than 0.001% human DNA contamination was observed in the brain
or BM from t w o SClD mice (SCID mice 4943 and
4985) treated with B43-PAP immunotoxin (2.5 p g per
dose every week x 4 weeks). SClD mouse 5053 was
i.t. B43-PAP (2.5p g per
a control mouse treated with
dose every week x 4 weeks) starting 1 week after
inoculation with NALM-6 leukemia cells. Human
DNA was observed in the BMs as well as brains of
SClD mice treatedwith i.t. methotrexate (15 p g per
dose every week with the intent of
a 4-week therapy;
SClD mice 4848 and 4847). Both mouse 4848 and
mouse 4847 were electively killed on day 48 after
becoming paralyzed secondary t o CNS leukemia.
% NALM-6 DNA
C
r
B43PAP (i.t)
+
% NALM-6 DNA
I 100 50
10 1 0.1 0.01
0.001
’
MTX (1.t)
+‘
9
4’ 9
8 8 8 08 48
9
9
b
0
@
$g?
D
hypothesis is supported by the fact that in the described SCID
model, massive medullary and visceralleukemic burdens did
not cause morbidity; neurologic signs of CNS disease were
the first clinical evidence for onset of human leukemia.
There was no clinical or morphologic evidence of toxicity
in mice given four weekly intrathecal treatments with 2.5 pg
B43-PAP, a dose level that produced significantly prolonged
survival and apparent cure in some SCID mice. However,
intrathecal administration of B43-PAP at dose levels 210
pg per weekly single dose caused major damage to the spinal
cord. Interestingly, neurotoxicity of intrathecal B43-PAP appeared to be cumulative. Only one mouse became paralyzed
after a single intrathecal injection of I O pg, whereas paralysis
developed in the others only after multiple weekly injections
of 10 pg. Damage only occurred near the site of injection,
suggesting that neurotoxicity is caused by high local concentrations of B43-PAP, a situation that might be prevented by
slow infusion of a dilute solution of the immunotoxin. This
hypothesis can be tested in larger species.
The SCID mousemodelofhumanCNS
leukemia described in this report provides a basis for future comparative
studies of the toxicity and efficacy of novel CNS therapy
regimens in B-lineage ALL. These experiments provide a
paradigm for the critical evaluation of a new CNS therapy
regimen in B-lineage ALL. This report also extends previous
studies that showed the potential therapeutic benefit of in-
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
GUNTHER ET AL
2544
Fig 6. Toxicity of intrathecal
643-PAP immunotoxin therapy.
Caudalspinalcord of an SClD
mouse 5 days after the third
weekly intratheeal injection of
10 p g 643-PAP immunotoxin.
There is massive necrosis ofthe
parenchyma of the spinal cord,
as well as vacuolization of ventral nerve roots. H&E stain; OM:
(AI, x 50; (B),x 100.
trathecal immunotoxin therapy against leptomeningeal carcinomatosis."."
REFERENCES
1. Pinkel D. Woo S: Prevention and treatment of meningeal leuke-
mia in children. Blood 84:355-366, 1994
2. Tubergen DG. Gilchrist G S . O'Brien RT. Coccia PF, Sather
HN, Waskerwitz MJ,Hammond CD: Prevention of CNS disease
in intermediate-risk acute lymphoblastic leukemia: Comparison of
cranial radiation and intrathecal methotrexate and the importance of
systemic therapy: A Childrens Cancer Group report. J Clin Oncol
I 1520, 1993
3. Pullen J, Boyett J. Shuster J, Crist W, Land V, Frankel L, Iyer
R, Backstrom L, van Eys J, Hams M, Ravindranath Y,Sullivan M:
Extended triple intrathecal chemotherapy trial for prevention of CNS
relapse in good-risk and poor-risk patients with B-progenitor acute
lymphoblastic leukemia: A Pediatric Oncology Group study. J Clin
Oncol 5:839, 1993
4. Ortega JA, Nesbit ME, Sather HN. Robison LL, D'Angio GJ,
Hammond GD: Long-term evaluation of a CNS prophylaxis trialtreatment comparison and outcome after CNS relapse in childhood
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
BIOTHERAPY OF CNS LEUKEMIA
ALL: A report from the Children’s Cancer Study Group. J Clin
Oncol 5:1646, 1987
5. Kim TH, Ramsay NK, Steeves RA, Nesbit ME: Intermittent
central nervous system irradiation and intrathecal chemotherapy for
central nervous system leukemia in children. Int J Radiat Oncol Biol
Phys 13:1451, 1987
6. Mandell LR, Steinherz P, Fuks Z Delayed central nervous
system (CNS) radiation in childhood CNS acute lymphoblastic leukemia: Results of a pilot trial. Cancer 3:447, 1990
7. Mahmoud HH, Rivera GK, Hancock ML, Krance RA, Kun
LE, Behm FG, Ribeiro RC, Sandlund JT, Crist W M , h i CH: Low
leukocyte counts with blast cells in cerebrospinal fluid of children
with newly diagnosed acute lymphoblastic leukemia. N Engl J Med
5:314, 1993
8. Bleyer WA, Coccia PF, Sather HN, Level C, Lukens J, Neibrugge DJ, Siege1 S, Littman PS, Leikin SL, Miller DR, Chard RL,
Hammond GD, the Childrens Cancer Group: Reduction in central
nervous system leukemia with a pharmacokinetically derived intrathecal methotrexate dosage regimen. J Clin Oncol 1:317, 1983
9. Nesbit M E , Sather H, Robison LL, Donaldson M, Littman P,
Ortega JA, Hammond GD: Sanctuary therapy: A randomized trial
of 724 children with previously untreated acute lymphoblastic leukemia. Cancer Res 42:674, 1982
IO. Land VJ, Thomas PR, Boyett JM, Glicksman AS, Culbert S,
Castleberry RP, Berry DH, Vats T, Humphrey GB: Comparison
of maintenance treatment regimens for first central nervous system
relapse in children with acute lymphoblastic leukemia. Cancer
81:112, 1985
11. Winick NJ, Smith SD, Shuster J, Lauer S, Wharam MD, Land
V, Buchanan G, Gaston R Treatment of CNS relapse in children
with acute lymphoblastic leukemia: A Pediatric Oncology Group
study. J Clin Oncol 11:271, 1993
12. Uckun FM, Downing JR. Chelstrom LM, Gunther R, Ryan
M, Simon J, Carroll A J , Tuel-Ahlgren L, Crist WM: Human
t(4; 1l)(q21;q23) acute lymphoblastic leukemia in mice with severe
combined immunodeficiency. Blood 84:859, 1994
13. Chelstrom LM, Gunther R, Simon J, Raimondi SC, Krance
R, Crist WM, Uckun FM: Childhood acute myeloid leukemia in
mice with severe combined immunodeficiency. Blood 84:20, 1994
14. Kamel-Reid S, Letarte M, Sirard C, Doedens M, Grunberger
T, Fulop G, Freedman M, Phillips R, Dick J: A model of humanacute
lymphoblastic leukemia in immune-deficient SCID mice. Science
246:1597, 1989
IS. Jansen B, Kersey JH, Jaszcz WB, Gunther R, Nguyen DP,
Chelstrom LM, Tuel-Ahlgren L,Uckun FM: Effective immunochemotherapy of human t(4; 11) leukemia in mice with severe combined immunodeficiency (SCID) using B43 (anti-CD19)-pokeweed
antiviral protein immunotoxin plus cyclophosphamide. Leukemia
7:290, 1993
16. Gunther R, Chelstrom LM, Finnegan D, Tuel-Ahlgren L, Irvin
2545
JD, Myers DE, Uckun FM: In vivo anti-leukemic efficacy of antiCD7-pokeweed antiviral protein immunotoxin against hyman T-lineage acute lymphoblastic leukemia/lymphoma in mice with severe
combined immunodeficiency. Leukemia 7:298, 1993
17. Uckun FM, Manivel C, Arthur D, Chelstrom LM, Finnegan
D, Tuel-Ahlgren L, Irvin JD, Myers DE, Gunther R: In vivo efficacy
of B43 (anti-CD19)-pokeweed antiviral protein against human preB cell acute lymphoblastic leukemia in mice with severe combined
immunodeficiency. Blood 79:2201, 1992
18. Uckun FM, Downing JR, Gunther R, Chelstrom LM, Finnegan D, Land VJ, Borowithz MJ, Carroll A J , Crist WM: Human
t(1; 19)(q23;p13) pre-B acute lymphoblastic leukemia in mice with
severe combined immunodeficiency. Blood 81 :3052, 1993
19. Myers DE, Irvin JD, Smith RS, Kuebelbeck VM, Uckun FM:
Production of a pokeweed antiviral protein (PAP)-containing immunotoxin, B43-PAP, directed against the CD19 human B-lineage
lymphoid differentiation antigen in highly purified form for human
clinical trials. J Immunol Methods 136:221, 1991
20. Hylden JLK, Wilcox GL: Intrathecal substance P elicits a
caudally-directed biting and scratching behavior in mice. Brain Res
217:212, 1981
21. Batson OV: The function of the vertebral veins and their role
in the spread of metastases. Ann Surg 112:138, 1940
22. Cavallo F, Forni M, Riccardi C, Soleti A, Di Pierro F, Forni
G: Growth and spread of human malignant T-lymphoblasts in immunosuppressed nude mice: A model for meningeal leukemia. Blood
80:1279, 1992
23. Thomas L: Pathology of leukemia in the brain and meninges:
Postmortem studies of patients with acute leukemia andof mice
given inoculations of L1210 leukemia. Cancer Res 25:1555, 1961
24. Zovickian J, Youle R: Efficacyof intrathecal immunotoxin
therapy in an animal model of leptomeningeal neoplasia. J Neurosurg
68:767, 1988
25. Hertler A, Schlossman D, Borowitx M, Poplack D, Frankel
A: An immunotoxin for the treatment of T-acute lymphoblastic leukemic meningitis: Studies in rhesus monkeys. Cancer Immunol Immunother 28:59, 1989
26. Urch C, George A, Stevenson G, Bolognesi A, Stirpe F, Weller R, Glennie M: Intrathecal treatment of leptomeningeal lymphoma
with immunotoxin. Int J Cancer 47:909, 1991
27. Johnson V, Wrobel C, Wilson D, Zovickian J, Greenfield L,
Oldfield E, Youle R Improved tumor-specific immunotoxins in the
treatment of CNS and leptomeningeal neoplasia. J Neurosurg 70:240,
1989
28. Recht L, Torres C, Smith T, Raso V, Griffin T: Transfemn
receptor in normal and neoplastic brain tissues: Implications for the
brain-tumor immunotherapy. J Neurosurg 72:941, 1990
29. Muraszko K, Sung C, Walbridge S, Greenfield L, Dedrick
RL, Oldfield EH, Youle RJ: Pharmacokinetics and toxicology of
immunotoxins administered into the subarachnoid space in nonhuman primates and rodents. Cancer Res 53:3752, 1993