.JPG?ad=18481&adview=true)
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Cancer Research Campaign Department of Epithelial Biology, Paterson Institute, Christie Hospital National Health Service Trust, Manchester, M20 9BX [D. M. P., L. B., C. S. P.]; Cancer Research Campaign Molecular and Cellular Pharmacology Group, School of Biological Sciences, University of Manchester, Manchester M13 9PT [D. M. P., L. B., J. A. H.]; and Cancer Research Campaign Centre for Cancer Therapeutics, Institute of Cancer Research, Sutton, SM2 5NG [L. B., A. L. J.], United Kingdom
ABSTRACT
The thymidylate synthase inhibitor raltitrexed (ZD1694, Tomudex) induces greater intestinal toxicity, manifested as diarrhea and weight loss, in BALB/c than in DBA/2 mice. No convincing pharmacokinetic or pharmacodynamic reason for this strain difference has been established. We have investigated whether this strain difference in response to raltitrexed is related to differential susceptibilities of intestinal mucosae to undergo apoptosis and also whether p53 expression, a critical factor in 5-fluorouracil-induced intestinal apoptosis and toxicity, modulates this response. Ten mg/kg or 100 mg/kg raltitrexed were administered as single or double i.p. injections 24 h apart to BALB/c, DBA/2, and p53-/- mice. Apoptosis, mitosis, and tissue damage were assessed in intestinal epithelium, and animal weight was recorded. BALB/c mice developed diarrhea and weight loss following 100 mg/kg x2 raltitrexed, whereas DBA/2 mice did not. BALB/c mice were more sensitive than DBA/2 to induction of small-intestinal and colonic apoptosis 24 h following 100 mg/kg raltitrexed. Inhibition of mitosis was equivalent in both strains. Both strains showed histopathological damage to the small intestine after 100 mg/kg x2 raltitrexed, but only BALB/c mice demonstrated colonic damage. p53-null mice showed the same level of small intestinal apoptosis as their wild-type counterparts 24 h after 100 mg/kg x1 raltitrexed and also the same levels of intestinal toxicity 3, 5, and 7 days after 100 mg/kg x2 raltitrexed. Thus, BALB/c mice were more susceptible to induction of intestinal apoptosis by raltitrexed than DBA/2 mice and also demonstrated more histopathological damage in the colon correlating with the induction of diarrhea and weight loss. In contrast to 5-fluorouracil, the intestinal apoptosis and toxicity induced by raltitrexed were p53-independent.
INTRODUCTION
Raltitrexed (ZD1694, Tomudex) is a novel quinazoline pure
TS4
-inhibitor that has been assessed in Phase III trials in
advanced colorectal cancer (1
, 2)
. Raltitrexed has similar
efficacy in this disease to 5-FU modulated with leucovorin but
demonstrates a different profile of side effects. One particularly
troublesome side effect is diarrhea. Grade III/IV diarrhea occurs in
14% of patients given this drug and if they have coexisting
neutropenia, it may be life-threatening (2)
. A murine
model has been established previously to identify determinants of the
gastrointestinal toxicity of raltitrexed (3
, 4)
. One
inbred mouse strain (BALB/c) is much more sensitive to sequential
administration of raltitrexed than another (DBA/2), with a maximum
tolerated dose in BALB/c mice of 5–10 mg/kg/d x5, compared with >500
mg/kg/d x5 in DBA/2. This difference between the strains is primarily
attributable to different susceptibilities to gastrointestinal
toxicity. We have demonstrated differences between the two strains in
histopathological gut damage and in induction of diarrhea and weight
loss following raltitrexed treatment. However, to date, no
convincing pharmacodynamic or pharmacokinetic explanation for this
difference between the two strains has been established
(3)
. The intestinal toxicity caused by raltitrexed
administration can be reversed by leucovorin (4)
.
In addition to so-called "upstream" factors, which are directly related to the mechanism of action of a chemotherapeutic drug, the toxic effects of a cytotoxic agent are also influenced by "downstream" factors. These include the genes that modulate the thresholds at which particular cells undergo cell death (5 , 6) . Previous experiments have not demonstrated any differences in the baseline levels of thymidine, deoxyuridine, folate, and TS between the two strains (3) . Also, the concentrations of raltitrexed in gut scrapes of intestinal mucosa were the same in BALB/c and DBA/2 mice at early time points after drug administration (3) . We have therefore investigated whether the different susceptibilities of BALB/c and DBA/2 mice to intestinal toxicity following raltitrexed treatment are related to differential capabilities of intestinal epithelia from these animals to undergo apoptosis.
Mouse intestinal epithelium provides a well-characterized model for the
study of damage-induced apoptosis, and mice in which key genes
controlling cell death have been knocked out have been used to
demonstrate the effects of a number of genes that regulate this process
(7)
. Moreover, different mouse strains have been
shown previously to have differential susceptibilities to the acute
induction of apoptosis following 
-radiation (8)
. This
system has been used previously to investigate the genetic regulation
of the acute apoptosis and intestinal toxicity induced in intestinal
epithelia by the TS-inhibitor 5-FU (9
, 10)
. These
experiments demonstrated that acute induction of apoptosis was a factor
in determining the toxicity of 5-FU to intestinal epithelia.
Suppression of proliferation induced by this drug was also an important
factor determining toxicity. Both the acute apoptosis and
intestinal toxicity induced by 5-FU were very greatly reduced in
p53-null mice compared with their wild-type counterparts,
indicating that the pathological response to this drug in
vivo is strongly p53-dependent (10)
.
We have therefore investigated whether intestinal epithelia from BALB/c and DBA/2 mice show different susceptibilities to the induction of apoptosis after raltitrexed administration congruent with the different levels of gastrointestinal toxicity demonstrated in the two strains. We have also used p53-null mice to examine whether, as with 5-FU, the apoptosis and intestinal toxicity induced by raltitrexed are influenced by p53 expression.
MATERIALS AND METHODS
Materials.
Raltitrexed was a gift from Zeneca Pharmaceuticals (Macclesfield,
United Kingdom) and was dissolved in 0.05 M sodium bicarbonate and the
pH altered to 8.5–9.0.
Animals.
Male BALB/c and DBA/2 mice and p53 wild-type (+/+) and null
(-/-) mice on a mixed genetic background (11)
of either
sex were used. Mice were aged 10–12 weeks, and there were four mice in
each experimental group. They were housed under conventional animal
house conditions with food and water ad libitum and a
12 h dark/light cycle (lights on at 6 a.m. and off at 6 p.m.).
Treatment and Sampling.
Raltitrexed was administered by bolus i.p. injection to 38 BALB/c and
38 DBA/2 mice on four schedules: (a) 10 mg/kg at 9 a.m.
on day 0; (b) 100 mg/kg at 9 a.m. on day 0;
(c) 10 mg/kg at 9 a.m. on day 0 and again at 9 a.m. on day 1; and (d) 100 mg/kg at 9 a.m. on day 0 and
again at 9 a.m. on day 1. One group of 10 mice for each treatment
schedule was weighed daily at 9 a.m., and records of state of
health, viability, and nature of feces were kept. Four mice in
each group were sacrificed by cervical dislocation at times ranging
from 12 h to 10 days after administration of the last dose of
raltitrexed, and their intestines were dissected and fixed in Carnoy’s
fixative. Three-µm transverse sections of paraffin-embedded mid-small
intestine and middle third of colon were stained with H&E.
Groups of four p53 wild-type and null mice received either a single injection of 100 mg/kg raltitrexed at 9 a.m. or two injections of 100 mg/kg raltitrexed at 9 a.m. on consecutive days and were sacrificed at time intervals ranging from 24 h to 7 days. Tissues were processed as for BALB/c and DBA/2 mice.
Scoring.
Apoptosis was scored by a single observer (L. B.) who was blinded to
the treatment administered. A proportion of the slides (5%) were
also scored in a blinded manner by a second observer (D. M. P.) to
ensure reproducibility of the scoring procedure. Apoptosis and mitosis
were scored on a cell-positional basis, analyzing 50 half-crypts/mouse
using a light microscope as described in detail previously
(12)
. Each cell position was scored as showing either a
normal cell, apoptotic event, or mitotic event, with cell position
1 being at the crypt base (12)
. An apoptotic event was
defined as the presence of one or more apoptotic bodies at a particular
cell position. Apoptotic bodies are characterized by condensed
chromatin and are easy to distinguish in sections of intestinal
epithelium fixed in Carnoy’s fixative (12)
. Data are
presented as mean and SD of the apoptotic/mitotic index (calculated as
the percentage of counted cells that were apoptotic/mitotic) for a
group of four mice and as plots smoothed over three cell positions
(12)
of apoptotic/mitotic cell index percentage against
cell position along the crypt.
The number of cells per hemicrypt and per hemivillus were used as indicators of tissue damage induced by the drug. These parameters were measured in 10 small intestinal crypts, 10 small intestinal villi (where possible, crypts and villi in continuity were used because these were most likely to represent a true longitudinal section), and 10 midcolonic crypts per mouse as described previously (10) . Data are presented as means and standard errors in a group of four mice.
Clonogenic crypt survival was assessed 96 h following 2 doses of 100 mg/kg raltitrexed as described previously (13) . Crypt width correction was applied (13) .
Statistical analysis was by Student’s t test assuming unequal variance of the groups and by the modified median test described in detail previously (14 , 15) . A significance level of P < 0.05 was used for both tests.
RESULTS
Whole Animal Toxicity of Raltitrexed.
None of the doses of raltitrexed used in current experiments were
lethal to BALB/c or DBA/2 mice. Single or double doses of 10 mg/kg
raltitrexed and a single dose of 100 mg/kg raltitrexed did not cause
significant weight loss in either strain (Fig. 1)
. However, two doses of 100 mg/kg
raltitrexed caused significant weight loss, maximal at 4 days in BALB/c
mice (Fig. 1
a), but no significant weight loss in DBA/2 mice
(Fig. 1
b), in a similar fashion to the experiments described
in Ref. 3
. The difference between the two strains at the
time point of maximal weight loss (4 days) was highly statistically
significant following two doses of 100 mg/kg raltitrexed
(P < 0.0001). Additionally, BALB/c mice that received
two doses of 100 mg/kg raltitrexed developed diarrhea between days 3
and 7–8, manifested as brown fecal staining around the anal margin,
whereas DBA/2 mice produced formed fecal pellets throughout the
experiment.
|
a and b
demonstrate that for 72 h following a single injection of 100
mg/kg raltitrexed, BALB/c mice developed significantly more apoptosis
in both small intestinal and midcolonic epithelium than DBA/2 mice.
This difference was most marked at 24 h, the time point of peak
apoptosis (P = 0.001 in small intestine and
P = 0.011 in midcolon by Student’s t test
analysis). The zero time point represents the level of baseline
apoptosis in untreated control mice. There were no statistically
significant differences between the two strains in the levels of
baseline small intestinal or midcolonic apoptosis. Previous experiments
that analyzed small intestinal apoptosis at 3-hour intervals
from 3–48 h after administration of raltitrexed in BDF1 mice indicated
that 24 h was the time point of peak apoptosis (data not shown).
Analysis of the cell positional distribution of apoptosis in small
intestinal epithelium 24 h following 100 mg/kg raltitrexed (Fig. 3
a) demonstrated that cell
positions 8–12 showed peak levels of apoptosis in the BALB/c strain,
whereas cells higher up the crypt hierarchy (cell positions 14–16)
developed peak levels of apoptosis in DBA/2 mice. Modified median test
analysis confirmed significant differences between the two strains at
cell positions 4–12. In midcolonic epithelium, apoptosis was induced
at cell positions near the crypt base (cell positions 1–8) in BALB/c
mice, whereas no apoptosis above control levels was induced by 100
mg/kg raltitrexed in DBA/2 mice (Fig. 3
b). The modified
median test confirmed statistically significant differences between the
midcolons of the two strains between cell positions 1 and 9.
|
|
. As the
cytotoxic effect of raltitrexed depends upon cells undergoing active
cell division, the effects of the second injection are thus difficult
to interpret.
One hundred mg/kg raltitrexed caused a similar suppression of
proliferation (mitosis) in small intestinal (Fig. 2
c) and
midcolonic (Fig. 2
d) epithelia of BALB/c and DBA/2 mice. The
elevated mitotic index in BALB/c midcolon at 72 h suggests that
the epithelium is attempting to repair the apoptotic damage manifested
at 24 h; this increase in proliferation was absent in DBA/2 mice.
Changes in Crypt and Villus Cellularity after Raltitrexed.
Two doses of 100 mg/kg raltitrexed caused changes in the cellularity of
small intestinal and midcolonic crypts and villi (Fig. 4)
. There were minor differences in
hemicrypt and hemivillus cell numbers in control animals [small
intestinal crypt 20.0 (BALB/c) versus 18.4 (DBA/2),
P = 0.09; small intestinal villus 38.5 (BALB/c)
versus 45.0 (DBA/2), P = 0.08; and
midcolonic crypt 30.8 (BALB/c) versus 24.5 (DBA/2),
P = 0.0004], hence data are presented as the
percentage of change from control. BALB/c mice demonstrated a greater
loss of cellularity of small intestinal (Fig. 4
a;
P = 0.012 at 72 h) and midcolonic crypts (Fig. 4
c; P = 0.007 at 72 h) than DBA/2 mice,
but no differences were seen in the cellularity of small intestinal
villi (Fig. 4
b; P = 0.059 at 72 h).
Examination of crypt histology (Fig. 5)
confirmed this statistical analysis. Fig. 5
g demonstrates
particularly that changes in colonic crypt architecture were seen
72 h after two doses of 100 mg/kg raltitrexed in BALB/c mice,
whereas the histology of DBA/2 colonic crypts 72 h after the same
treatment was normal (Fig. 5
h).
|
|
Apoptosis and Crypt/Villus Cellularity in p53-Null
Mice.
We have demonstrated previously that p53 expression played a major role
in controlling the amount of apoptosis and intestinal toxicity induced
in murine intestinal epithelium by another cytotoxic drug that is an
inhibitor of TS, 5-FU, although this drug was acting via an inhibition
of RNA metabolism (9
, 10)
. We therefore investigated
whether p53 expression played any role in the differences observed in
apoptosis and intestinal toxicity between BALB/c and DBA/2 mice
following administration of raltitrexed. There was no statistically
significant difference in small intestinal apoptotic yield by
Student’s t test or by modified median test analysis
between p53 wild-type and null mice 24 h after 10 mg/kg
or 100 mg/kg raltitrexed (data not shown). One hundred mg/kg x1
raltitrexed caused no apoptosis above control levels in the midcolonic
crypts of p53 wild-type and null mice (data not shown).
There was no significant difference in small intestinal crypt or villus
cellularity between p53 wild-type and null mice at various
time points after two injections 24 h apart of 100 mg/kg
raltitrexed (Fig. 6)
. No changes in
colonic crypt cell number were observed in either p53
wild-type or null mice following this dose regimen of raltitrexed (data
not shown). The apoptosis and changes in crypt and villus cellularity
induced in intestinal epithelia by raltitrexed are therefore
independent of p53 expression.
|
BALB/c mice developed more diarrhea and lost more weight than
DBA/2 mice after sequential administration of the TS-inhibitor,
raltitrexed (Fig. 1)
, as we described previously (3
, 4)
.
This difference in overall toxicity between two mouse strains has been
shown to result primarily from the effects of the drug in the
gastrointestinal tract rather than in other organs such as the bone
marrow (3
, 4) . However, previous experiments have not
identified any upstream factors (directly related to the mode of
drug action) which were responsible for this strain difference in
toxicity (3)
. In particular, there were no differences
between the strains in the pretreatment levels of thymidine,
deoxyuridine, folate or TS and, at early times after drug
administration (up to day 3), there were no significant differences in
the tissue levels of raltitrexed and its polyglutamates in intestinal
mucosa. We have therefore investigated whether any downstream factors
(not directly related to the mechanism of drug action), particularly
the engagement of apoptosis, were important determinants of this
differential response. Indeed, BALB/c mice developed significantly more
small-intestinal and colonic apoptosis than DBA/2 mice 24 h after
a single injection of 100 mg/kg raltitrexed (Fig. 2)
. Interestingly,
raltitrexed induced apoptosis preferentially in cells at different
positions in the small intestinal crypts of the two mouse strains (Fig. 3)
. Cells nearer the small-intestinal crypt base underwent more
apoptosis in BALB/c than DBA/2 mice. The cells located nearer the crypt
base are more rapidly dividing and represent cells that are earlier in
the cell lineage than those more differentiated cells near the
crypt-villus junction. It is possible that the higher levels of
apoptosis in the proliferative compartment of BALB/c small intestine
compared with DBA/2 might result in more disruption of crypt function,
and one would predict that this might have more profound consequences
upon the overall architecture of the epithelium, especially as
proliferation is suppressed by this drug. This is supported by the
clonogenic crypt survival assay, which showed significantly less crypt
survival in BALB/c compared with DBA/2 mice 96 h following two
doses of 100 mg/kg raltitrexed (see "Results"). Why, on a
cell-by-cell basis, there is a difference in positional apoptosis in
the two strains is difficult to answer. This may have a pharmacodynamic
explanation, but is difficult to test on a cell-by-cell basis.
In our study of the p53-dependent pathology induced by 5-FU in the
murine intestine in vivo (10)
, we concluded
that toxicity was dependent upon apoptosis occurring against a
background of inhibition of proliferation. Scoring apoptosis alone
showed that the amount of apoptosis was not dependent on the dose of
5-FU past a certain threshold, with toxicity emerging above that
threshold (10)
. Instead, inhibition of proliferation was
5-FU dose-dependent so that cell loss, by apoptosis, could not be
replenished once proliferation was also inhibited. Raltitrexed also had
profound effects upon cellular proliferation within the intestinal
crypt compartment (Fig. 2)
, and so the increased apoptosis observed in
the BALB/c strain occurred against a background of inhibited
proliferation, increasing overall toxicity.
Is the strain difference in susceptibility to induction of intestinal
apoptosis described here specific to raltitrexed, or does it also occur
with other damaging stimuli? Previous experiments have shown
differences between the strains after the alternative stimulus of
-radiation (8)
. These differences between the same
strains were less marked than with raltitrexed. Small-intestinal
apoptotic index was 8.6% in BALB/c compared with 6.7% in DBA/2, and
midcolonic apoptosis was 3.3% in BALB/c compared with 4.1% in DBA/2
at 4.5 h after 1 Gy -radiation (8)
. As discussed
above, the final outcome, in terms of tissue integrity, depends on
apoptosis and the extent of suppression of mitosis. In the DBA/2 and
BALB/c mice, the suppression of mitosis was 3.1% and 1.6%,
respectively, in small intestine and 1.2% and 0.7%, respectively, in
midcolon 4.5 h after 1 Gy -radiation (8)
. This
suggests that, whereas there may be an inherent difference between the
mouse strains in susceptibility to apoptosis and suppression of
proliferation, part of the response described here may also be specific
to raltitrexed. Direct comparison of these two stimuli is not
completely valid because -radiation-induced apoptosis occurs in
different cell positions from that induced by raltitrexed. This
could influence the rate of cell replenishment after a wave of cell
death, for example, if many of the cells in the region necessary for
cell number amplification, higher up the crypt from the stem cells, the
so-called "transit cells," were lost. In addition, the effects of
-radiation are completely p53-dependent (16)
,
whereas raltitrexed-induced apoptosis is p53-independent (Fig. 6)
.
Damage to the small intestine and colon may not contribute equally to
the diarrhea induced by raltitrexed, and again, there are differences
in these two sites between the strains studied here. The changes in
crypt cell number (Fig. 4)
and particularly changes in gross crypt
histology (Fig. 5)
resulting from two injections of 100 mg/kg
raltitrexed were more marked in the colon than in the small intestine.
The colon is a vital organ for the reabsorption of water. In humans the
capacity for reabsorption can increase from 1.5 liters/day to 4–5
liters/day under stressful conditions (17)
. We hypothesize
that the reduction of damage to the colonic epithelium of DBA/2 mice
compensates for the increased water-load resulting from small
intestinal damage by reabsorbing water. Hence, no diarrhea or weight
loss results in the DBA/2 strain. By contrast, the damaged colonic
epithelium of BALB/c mice has reduced functional reserve and is not
able to reabsorb water fully. This results in diarrhea and weight
loss. Although differences in both small intestinal and colonic
apoptosis were observed after 100 mg/kg raltitrexed (Fig. 1)
, we
propose that the colonic component may be more important. Importantly,
the colonic apoptosis that was observed in BALB/c mice occurred
maximally at cell positions near the crypt base where the stem cells
are believed to be located (Fig. 3
b), whereas no midcolonic
apoptosis occurred above baseline levels in DBA/2 mice after
raltitrexed administration. Moreover, DBA/2 mice did not display any
late (>48 h) increase in mitotic index in the colon, which was
indicative of an attempt to repair damaged epithelium after cell loss
by apoptosis (Fig. 2)
.
The p53-independence of raltitrexed-induced apoptosis and
intestinal toxicity (Fig. 6)
was surprising in view of the drug’s
mechanism of action as a TS-inhibitor leading to strand breaks in DNA
and in view of the p53-dependent apoptosis caused by 5-FU (9
, 10)
. However, 5-FU and raltitrexed appear to have different
mechanisms of action in this context. Whereas the apoptosis induced by
raltitrexed can be prevented by coadministration of thymidine
(9)
, in keeping with a mechanism of action as a
TS-inhibitor, 5-FU-induced apoptosis was not affected by thymidine
administration, but was partially rescued by uridine administration,
suggesting a cytotoxic action by RNA damage (9)
. We and
others have also confirmed these findings in vitro using
p53-wild-type HCT116 human colon carcinoma cells
(18)
.5
The cytotoxicity of raltitrexed to this cell line was reduced by
thymidine administration and was maintained in clones of HCT116 cells
with reduced levels of functional p53. By contrast, the cytotoxicity of
5-FU to HCT116 was unaffected by thymidine administration, was
partially rescued by uridine administration, and was significantly
reduced in cells with reduced levels of p53 (18)
.5
Thus the mechanistic basis of
strain-dependent differences in apoptosis remains to be determined and
is open to genetic investigations.
In summary, we have demonstrated that BALB/c and DBA/2 mice show different thresholds for the induction of apoptosis after raltitrexed administration. We suggest that the colonic component of this apoptosis, coupled with an inhibition of proliferation, is an important determinant of the intestinal toxicity induced by raltitrexed. These findings may have important implications regarding the severe diarrhea caused by raltitrexed chemotherapy in some human patients. For example, it may be possible to predict which patients are most at risk of developing severe diarrhea by measuring apoptotic and proliferation indices in rectal biopsies after drug administration. However, the methodology associated with such an approach is fraught with problems (19) , and a genetic basis for the difference is being sought. Also, we would predict that therapeutic interventions aimed at either limiting the degree of colonic damage caused by raltitrexed or at increasing the colon’s capacity to reabsorb water might be useful for reducing the severity of diarrhea. The difference between 5-FU and raltitrexed in terms of the p53-dependence of intestinal apoptosis and toxicity also raises interesting possibilities for targeting chemotherapy based on the p53 status of colonic carcinomas and of considering combination chemotherapy with both agents in certain settings.
FOOTNOTES
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 in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 We thank the Medical Research Council (United
Kingdom), the Cancer Research Campaign (United Kingdom), and the
Institute of Cancer Research for funding. L. B. fulfilled these
studies as part of his B.Sc. degree in Pharmacology at Manchester
University. 
2 To whom requests for reprints should be
addressed, at University Clinical Department of Medicine, The Duncan
Building, Daulby Street, Liverpool, United Kingdom, L69 3GA. Phone:
0151-706-4074; Fax: 0151-706-5802; E-mail: dmpritch{at}liverpool.ac.uk
3 Present address: Institut de Recherches Servier,
11 rue des Moulineaux, 92150 Suresnes, Paris, France.
4 The abbreviations used are: TS, thymidylate
synthase; 5-FU, 5-fluorouracil.
5 D. M. Pritchard, L. Bower, S. Welsh, A. L.
Jackman, and J. A. Hickman. In vitro evidence for an RNA
induced, p53-associated apoptosis by 5-fluorouracil, and for a
p53-independent mode of action for the pure thymidylate synthase
inhibitor Raltitrexed (Tomudex, ZD1694), manuscript in preparation.
Received 5/26/00; revised 8/18/00; accepted 8/18/00.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  C. Muller, M. Bruhlmeier, P. A. Schubiger, and R. Schibli Effects of Antifolate Drugs on the Cellular Uptake of Radiofolates In Vitro and In Vivo J. Nucl. Med., December 1, 2006; 47(12): 2057 - 2064. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
)
or two (
) injections of 10 mg/kg and one (
) or two
(x) injections of 100 mg/kg raltitrexed. Error bars
represent standard deviations; 10 animals in each group.
