| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics, Preclinical Pharmacology |
The CRC Centre for Cancer Therapeutics at the Institute of Cancer Research, Surrey SM2 5NG, United Kingdom
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSEffect of RTX on...Plasma dThd, dUrd, and...Plasma and Tissue RTX...DISCUSSIONREFERENCES |
|---|
1 µM), and folate (40 ng/ml) were
similar in both mouse strains. A single injection of radiolabeled RTX
(5 mg/kg/day) did not lead to any marked difference 24 h later in
the total drug concentration and distribution of polyglutamates
(comprising 70–80% of drug extracted) in the liver, kidney, and
intestinal epithelium (large and small intestine) between the two mouse
strains. Further studies used a RIA to measure RTX polyglutamate
formation in tissues at various times and drug doses. This led to the
conclusion that, although there was a higher accumulation of RTX in
BALB/c small intestinal epithelium (days 4–6), it may be an effect
secondary to another undetermined cause of increased drug sensitivity.
This model represents a vehicle by which the etiology and treatment of
severe clinical toxicity induced by RTX may be evaluated. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSEffect of RTX on...Plasma dThd, dUrd, and...Plasma and Tissue RTX...DISCUSSIONREFERENCES |
|---|
14% of patients treated with RTX, which may be
life-threatening when there is coexisting neutropenia (8)
.
This has been graphically demonstrated by a recent United Kingdom
Medical Research Council study that compared RTX with two forms of
infusional 5-FU. Although the overall survival difference was not
statistically significant between the three arms, there was a 4% toxic
death rate in the RTX arm, thought to be again principally attributable
to the occurrence of severe diarrhea complicated by neutropenia
(10)
. Although in the hands of clinicians experienced with
using RTX such toxicities are generally manageable, these infrequent,
but serious side effects are obviously important issues for the
generalized use of this compound.
Preclinical studies had predicted that RTX would be active when given
infrequently because it is an excellent intracellular substrate
(Km, 1 µM)
for FPGS (1
, 11)
. This enzyme catalyzes the sequential
addition of extra glutamates (up to five) to the 
-carboxyl of the
glutamate ligand of the drug. RTX polyglutamates of high chain length
(mainly tri-pentaglutamates) constitute the majority of intracellular
drug (80–98%) in preclinical in vitro and in
vivo models (1
, 12
, 13)
. These polyglutamates are:
(a) 70-fold more potent TS inhibitors compared with the
parent drug; and (b) not readily effluxed from the cell.
Thus, retention of the drug in tumors (and some normal tissues) may
result in TS inhibition continuing after the majority of the drug has
been cleared from the plasma. In mice bearing the L5178Y lymphoma (i.m.
implanted), high tumor:plasma RTX ratios have been measured
several days after a single bolus injection of the drug
(14)
. Similarly in the ascitic L1210:ICR tumor, TS was
inhibited for at least 24 h after a single i.v. injection of RTX
(1)
. High levels of polyglutamates are also formed in the
liver and normal proliferating tissues, such as the intestinal
epithelium (2)
.
The therapeutic efficacy of TS inhibitors is believed to be dependent on a number of factors, which include the relative inhibition of TS in tumor versus normal proliferating tissues and the relative time for induction of DNA damage and/or irreversible commitment to cell death. Adequate inhibition of TS by RTX to give a cytotoxic response, in tumor and normal proliferating tissues, is dependent on the level of expression of TS, and the enzymes responsible for formation and breakdown of the polyglutamates (the reduced-folate carrier, FPGS, and folylpolyglutamate hydrolase). TS inhibition may be circumvented altogether by an active thymidine (dThd) salvage pathway. Inhibition of TS, in the absence of significant dThd salvage, results in depletion of TTP and accumulation of cells in the early S-phase of the cell cycle (15, 16, 17, 18) . The highly cytotoxic nature of TS inhibitors, such as RTX, is believed to be attributable to, at least in part, the concomitant elevation in the level of dUTP that occurs when TTP is decreased (19) . The large rise in the dUTP:TTP ratio leads to the misincorporation of uracil into DNA and damage during either synthesis or repair, or both (19, 20, 21) . Cells eventually become committed to the apoptotic cell death pathway. Thus numerous factors, including those downstream of TS inhibition, may determine response to RTX (22, 23, 24) .
A wide program of research has been initiated in this laboratory to
identify clinical determinants of response and toxicity to RTX. As part
of this research, we have investigated the interesting observation that
RTX has a significantly lower MTD in BALB/c (10 mg/kg/day x 5
days) compared with DBA2 mice (>500 mg/kg/day x 5 days). This
has now been ascribed to increased sensitivity of the gastrointestinal
tract to RTX-induced damage, although the genetic factors responsible
for this predisposition remain to be elucidated. Further investigation
of this model system should contribute to an understanding of the
underlying reason(s) for a small minority of patients experiencing
severe, RTX-related gastrointestinal toxicity and provide a vehicle by
which rescue procedures may be evaluated.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSEffect of RTX on...Plasma dThd, dUrd, and...Plasma and Tissue RTX...DISCUSSIONREFERENCES |
|---|
9 by dropwise addition of 1 M NaOH.
Solutions were sterile filtered by passing through a Ministart,
0.2-µm-pore filter size (Sartorius AG, Epsom, Surrey, United
Kingdom), frozen, and protected from light. Mice were injected i.p.
once a day for up to 5 days. Drug concentrations were calculated so
that mice received 0.1 ml solution per 10 g of body weight.
Control mice received solvent alone. Male BALB/c and DBA2 mice were purchased when 5–6 weeks of age from either the Oxford Laboratory Animal Colonies (Oxford, United Kingdom) or the Medical Research Council animal laboratories (Mill Hill, London, United Kingdom) and maintained on a standard diet (RM1E; Special Diets Services, Witham, Essex, United Kingdom) and water ad libitum. They were randomly grouped into cages of five mice and used when 7–8 weeks of age. Daily weights of individual mice were recorded, and mice inspected for signs of distress. Mice were sacrificed if they could not freely access food and water and/or had >30% body weight loss.
Histopathology of the Small and Large Intestine.
Mice were injected with 10, 100, or 500 mg/kg of RTX for up to 5 days
(days 1–5), and five mice from each group were killed by cervical
dislocation 24 h after each injection (days 2–6). In some
experiments, mice given 10 mg/kg/day x 5 days were killed 48 and
72 h after the fifth dose (days 7 and 8). The whole length of the
intestine was removed and flushed with Tris-HCl buffer (pH 10), using a
5-ml syringe and a pipette tip inserted through the proximal cut end of
the intestine (junction with stomach). The specimen was placed in
modified Methacarn consisting of 60% methanol, 30% Inhibisol, and
10% acetic acid, and left for at least 24 h. Tissues were fixed
and embedded in paraffin, and sections stained with H&E were prepared.
A graded scoring system was devised on a number of parameters. Villus
atrophy was graded as follows: 0, normal; 1, clubbing; 2, atrophy +; 3,
atrophy ++; 4, atrophy +++; and 5, tip necrosis. Edema of lamina
propri, inflammatory cells, and crypt abscesses (necrotic cells in
dilated crypts) were graded as either present or absent. The number of
goblet cells were recorded as either normal or increased. The mitotic
index was measured as the number of mitosis per five high-powered
fields.
The architecture of the small intestinal crypts was included as an additional measure of cytotoxic effect (23) . The number of cells/half crypt was measured by counting the number of cells from the crypt base to the crypt-villus junction down one side of an adequately sectioned crypt. Twelve crypts/mouse (60 per treatment group) were counted, two from each of six intestinal sections/mouse. More severe crypt damage was associated with cell loss and lower counts. In severely damaged sections, it was often difficult to find adequate crypts for cell counting; hence, this score was complemented by recording the number of "normal-looking crypts" per bowel section.
Hematology.
Groups of five mice received injections of RTX: BALB/c mice (5 mg/kg)
and DBA2 (5 and 100 mg/kg/day x 5 days). Control mice received
0.05 M NaHCO3. Blood for full blood
count and differential white cell count was collected by open cardiac
puncture under oxygen/halothane anesthesia (without recovery) into a
1-ml syringe and transferred to a 1.5-ml microcentrifuge tube
containing 2% EDTA to give a dilution of 10 µl of anticoagulant per
100 µl of blood. Analysis was performed at the Royal Marsden Hospital
Trust Clinical Hematology Laboratory.
Plasma dUrd and dThd Measurements.
Extraction and analysis of mouse plasma samples were by HPLC using a
modification of a method described previously (25
, 26) .
Briefly, this involves a two-stage (separated by peak collection,
lyophilization, and reconstitution in a small volume) HPLC isocratic
method with UV detection. Briefly, 250 µl of plasma samples were
treated with two volumes of acetonitrile, vortexed for 1 min, and
placed on ice for 5 min, prior to centrifugation at 11,000 x
g for 10 min. The supernatants were lyophilized (Savant SVC
200H refrigerated speed vacuum pump-IEC) and later reconstituted in 125
µl of the HPLC running buffer and centrifuged again for 5 min at
6,000 x g to remove particulate material prior to HPLC
injection of 100 µl. An Apex C-18 column was used in the second HPLC
run because this resulted in improved peak shape and separation as
compared with the Apex ODS Symm 5 µm packing (Jones Chromatography,
Ltd.) reported previously.
Plasma Folate Estimation.
Blood samples were collected in pediatric (3 ml) Vacutainer bottles
each containing 60 µl of EDTA and microcentrifuged, and the plasma
was removed and stored at -20°C until analyzed. Red cell folate and
plasma folate were measured using a Folate Radioassay kit
[125I] obtained from Becton Dickinson, Ltd.
(Oxford, United Kingdom).
Measurement of [5-3H]RTX and Its Polyglutamate Forms
in Mouse Tissues.
Radiolabeled RTX (19 Ci/mmol) was supplied by Zeneca Pharmaceuticals
(Alderley Park, Macclesfield, Cheshire, United Kingdom). This was
purified 24 h before use by a method described previously
(13)
. Three mice were injected with 5 mg/kg RTX (1.1
Ci/mmol), and the tissues were removed 24 h later and prepared as
described previously (2)
. Analysis was by HPLC (ion
pairing), using synthetic polyglutamate standards to identify the
peaks, and scintillation counting of collected fractions
(13)
.
Plasma and Tissue RTX (Polyglutamate) Levels Measured by RIA.
A polyclonal antibody raised to RTX was provided by Zeneca
Pharmaceuticals. This was used to measure plasma RTX as described
previously (14)
. This antibody cross-reacts equally with
the polyglutamate forms of RTX and was therefore used to measure total
tissue drug levels after injection of RTX to mice by a method described
previously (14)
.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSEffect of RTX on...Plasma dThd, dUrd, and...Plasma and Tissue RTX...DISCUSSIONREFERENCES |
|---|
).
However, over the next 24 h, the BALB/c, but not DBA2, mice lost a
significant amount of weight (day 4; 93% of starting weight).
Weight loss continued so that 24 h after completion of the course
of five injections of 10 mg/kg (day 6), BALB/c mice weighed 79% of
their initial weight. In contrast, DBA2 mice were 95% of their
initial body weight by this time. BALB/c mice continued to lose further
weight up to a nadir of day 8 (74% of initial weight). About 10% of
BALB/c mice receiving this 10-mg/kg/day dose became moribund and were
culled (days 7–9). In addition, the majority of BALB/c, but not DBA2
mice, displayed evidence of diarrhea for 1–3 days (commencing on days
4–5). However, the rapidity of weight loss over the first 7 days was
similar in those BALB/c mice that did not get diarrhea (data not
shown).
|
. For example, the lower dose of 5 mg/kg/day x 5 days to
BALB/c mice induced slightly less body weight loss (nadir day 7, rather
than day 8), and all mice recovered. Increasing the dose to 50
mg/kg/day x 5 days induced severe weight loss such that the
experiment was terminated on day 8. This and higher doses of RTX were
tolerated by DBA2 mice, although the weight loss seen for the first
4–5 days was similar at all doses between 10 and 500 mg/kg/day x
5 days (and interestingly, less than in BALB/c mice seen at 10
mg/kg/day). After this time, DBA2 mice that had received the higher
doses (100 and 500 mg/kg) of RTX lost increasingly more body weight
with a nadir at day 9 (77% of initial weight). Comparison of these
higher dose data for DBA2 mice with that of BALB/c mice injected with
the lower 10 mg/kg/day x 5 days dose demonstrates similar weight
loss, at their respective nadirs. However, the nadir was actually 1 day
later for the DBA2 mice, and the rate of weight recovery was slower
(starting weight regained by days 14–16 compared with day 12 for
BALB/c mice). In fact, the length of time that DBA2 mice (100 or 500
mg/kg) weighed less than their starting weight was 2/3 days longer than
for BALB/c mice (10 mg/kg). The MTD (defined here as the dose from
which all mice recover) is 5 and >500 mg/kg/day x 5 days for
the BALB/c and DBA2 mice, respectively. Diarrhea was not observed in
DBA2 mice at any dose or time. Coadministration of thymidine (dThd; 500 mg/kg three times a day for 8 days) with RTX (5 mg/kg day x 5 days) to BALB/c mice almost completely prevented the drug-induced weight loss (8% compared with 28% for RTX alone; data not shown).
Effect of RTX on Mouse Intestinal Histopathology
10 mg/kg/day.
Comparative histopathological studies on the small and large intestine
of both strains of mice were performed during and after administration
of 10 mg/kg/day RTX x 5 days in both strains of mice. The small
intestine was much more affected than the colon by the toxic effects of
RTX. However, some evidence of minor damage (characterized by damage
and shortening of the crypts) was seen in the colon of BALB/c, but not
DBA2, mice (measured on day 6). There was also a reduction in the
number of colonic crypt mitoses on days 3–7 in both mouse strains
(nadir day 6; 20% of day 1 mitoses; data not shown). These had
recovered to above pretreatment levels by day 8 (72 h after the fifth
injection). Minimal damage to the small intestine of both mouse strains
was observed on day 2 (characterized by a small but statistically
insignificant reduction in the number of crypt mitoses, some villus
blunting, and minimal edema), which became more marked from day 3 (Fig. 2
and data not shown). Additionally, an
inflammatory cell infiltrate in the lamina propria and crypt abscesses
were observed from this time, with 100% of BALB/c and DBA2 mice
showing evidence of the latter on days 3 and 5, respectively. Close
examination of the number of "normal-looking" crypts/crypt
circumference and the mean number of cells per hemicrypt revealed a
significant reduction in the level of both parameters on days 3–6
(later times not measured) that was greater in the small intestine of
BALB/c mice (data for day 6 given in Table 1
). This damage was also characterized by
a general villus atrophy that was apparent 24 h after the first
injection of RTX and became progressively worse over the next 2 days
(Fig. 2)
. This villus atrophy was greater in BALB/c compared with DBA2
mice from day 5 (statistical significance was only attained on days 5,
7, and 8). Sample histological sections taken from mice on day 7 are
shown in Fig. 3
. The difference between
the two mouse strains was greatest on day 8 (72 h after the fifth
injection) because of a significant improvement in the architecture of
the villi of the small intestines of DBA2 mice at this time (and no
evidence of crypt abscesses). Although the condition of the villi had
not markedly improved in BALB/c mice, the number of BALB/c mice with
inflammatory cells or crypt abscesses was reduced (50%), and the
crypt mitotic index had risen to above control levels (Fig. 2)
. Taken
together, these data suggest that 10 mg/kg/day x 5 days induces
more histopathological consequences of damage to the small intestine of
BALB/c compared with DBA2 mice, and that on day 7, and more
particularly on day 8, recovery was significantly more marked in the
DBA2 mice. This recovery is coincident with some increase in body
weight in this mouse strain, consistent with weight loss (and possibly
diarrhea) being a result of gut damage.
|
|
|
). Nevertheless, the score at 100 mg/kg was equal to that
of the BALB/c mice at both the 10- and 100-mg/kg doses. Similar results
were seen when the number of crypt cells/half crypt were counted. Thus,
100 mg/kg/day x 5 days of RTX given to DBA2 mice produced
histological damage to the small intestine similar to that seen at 10
and 100 mg/kg in BALB/c mice (day 6). However, DBA2 mice still lost
less weight than BALB/c mice at this time. Although the small intestine
of DBA2 mice showed some signs of recovery by day 8 (some improvement
in villus atrophy score and a normal mitotic index), improvement was
less than that observed in DBA2 mice that had received the lower dose
of 10 mg/kg daily (data not shown). However, at the higher dose,
histological consequences of damage (villus atrophy) on day 8 were
still less in DBA2 compared with BALB/c mice at 10 mg/kg (similar body
weight loss). Despite this improvement, DBA2 mice started to recover
body weight later and less rapidly.
DBA2 mice tolerated (as measured by recovery of body weight and absence
of moribundity) 500 mg/kg/day RTX, despite much greater small
intestinal crypt "damage" and reduction in number of cells per
crypt (day 6) than seen at the 100-mg/kg/day dose (Fig. 1
and Table 1
).
This damage was also greater than that seen in BALB/c mice after daily
10 or 100 mg/kg/day of RTX. However, villus atrophy was not increased
at the 500-mg/kg dose, and there was significantly less weight loss in
DBA2 mice over the first 4–5 days compared with BALB/c mice given 10
mg/kg. Furthermore, DBA2 mice did not get diarrhea at any dose of RTX.
|
Effect of RTX on Peripheral Blood Elements |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSEffect of RTX on...Plasma dThd, dUrd, and...Plasma and Tissue RTX...DISCUSSIONREFERENCES |
|---|
50 and 20% reduction in the number
of lymphocytes and platelets, respectively, and a 2-fold increase in
the number of neutrophils. By day 8, the only blood element remaining
different from controls was the platelets (25% reduction in both
strains; Table 2
). None of these changes
are large and are considered unlikely to have any biological
consequence. Because DBA2 mice could tolerate 100 mg/kg/day x 5
days, the effect of this dose was measured in this strain. RTX induced
significant neutropenia (95% reduction in neutrophils) and a fall in
lymphocytes (37% reduction) and platelets (55%) on day 8 (Table 2)
.
Recovery was evident by day 12, although the neutrophils were raised
nearly 6-fold compared with controls, consistent with a rebound
phenomenon (data not shown).
|
|
Plasma dThd, dUrd, and Folate Levels |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSEffect of RTX on...Plasma dThd, dUrd, and...Plasma and Tissue RTX...DISCUSSIONREFERENCES |
|---|
1 µM in
the plasma of both mouse strains. Twenty-four h after administration of
5 mg/kg of RTX, there was a 50% fall in the level of plasma dThd in
both strains (Fig. 4)
. However, 24 h
after completion of the 5-day course of RTX (day 6), the level of dThd
had actually returned to control level in BALB/c mice but remained 50%
reduced in the DBA2 mice (P < 0.05). Interestingly,
this had not recovered to pretreatment levels 7 days after treatment
(day 12), and because of a small second fall in dThd in BALB/c mice,
both strains did not have significantly different levels at this time.
In both mouse strains, plasma dUrd increased 3- and 5-fold on days 2
and 6, respectively, consistent with TS inhibition, and had recovered
to pretreatment levels by day 12.
|
|
Plasma and Tissue RTX Levels |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSEffect of RTX on...Plasma dThd, dUrd, and...Plasma and Tissue RTX...DISCUSSIONREFERENCES |
|---|
10 nM) 24 h
after treatment (Fig. 5)
. Highest levels
were seen in the liver (0.7–0.9 nmol/g), and although there was no
statistical difference, there was a suggestion of lower levels of the
longer chain-length polyglutamate forms in DBA2 liver. Overall,
polyglutamates (di-hexa) accounted for 70% of the total drug
extracted. Small intestinal epithelium accumulated less drug (0.3–0.4
nmol/g), and 80% was recovered as polyglutamates (mainly tetra and
penta). Similar results were observed in all sections of the intestine,
including the colon (data not shown). A significantly higher
(P < 0.05) concentration of total drug was measured in
DBA2 kidney (0.51 compared with 0.34 nmol/g), attributable to a higher
level of each polyglutamated form. A similar experiment was carried out
on day 4 (24 h after three injections of 5 mg/kg; data not shown).
There was a 3-fold increase in total drug present compared with that
on day 2 in the liver and small intestinal epithelium. Although there
was a trend toward higher drug levels in the liver and gut of BALB/c
compared with DBA2 mice, no statistical significance could be attached
to this because of the small number of mice examined (two/group).
Similarly, the distribution of polyglutamates could not be considered
different.
|
30- and 200-fold higher,
respectively (0.3 and 2.0 nmol/g; Fig. 6
). However, no significant difference in
these levels was apparent between strains. By day 3 (24 h after two
injections), levels in the plasma, liver, kidney, and small intestinal
epithelium of DBA2 mice had risen slightly (2-fold higher in the
epithelium compared with day 2). However, a similar increase was not
seen in BALB/c mice, leading to a statistically significantly higher
drug level (2-fold) in most of the tissues of DBA2 mice on day 3.
However, by day 4, because of an increased accumulation of drug in the
tissues of BALB/c but not DBA2 mice, the level in the tissues was
similar in both mouse strains with the exception of BALB/c intestinal
epithelium, which had a statistically significantly higher level
(2-fold) at this time. Further accumulation occurred in this tissue
of BALB/c mice, leading to an 4-fold higher level compared with DBA2
epithelium on days 5 and 6. In addition, a statistically significantly
higher concentration of RTX was measured in the plasma of BALB/c mice
on days 4–6. Coincident with the peak level seen in the plasma (day
6), there was a significantly higher drug concentration (2-fold) in the
kidney of BALB/c mice. No significant difference was seen in the liver
drug levels at any time. Two days after completion of the 5-day course
of injections (day 7), RTX still persisted in all tissues, although
some small reductions in drug levels were observed. However the
reduction in the plasma, small intestine, and kidneys was relatively
greater in BALB/c compared with DBA2 mice, which had the effect of
reducing the difference in drug levels between the two strains.
|
After administration of a 10-fold higher dose of RTX (100 mg/kg), the
plasma concentration increased by 5–10-fold in both mouse strains.
Liver and small intestinal epithelia drug levels rose 1–2-fold and
2–8-fold, respectively (Fig. 7
and data
not shown) compared with those seen after 10 mg/kg. Otherwise, the time
course and pattern of changes were similar to those seen at 10 mg/kg.
Comparison of the drug accumulation data for the small intestinal
epithelia at 10 mg/kg/day x 5 in BALB/c mice with that at 100
mg/kg/day in DBA2 mice (Fig. 7)
demonstrated that the RTX level in DBA2
epithelia was 2–7-fold higher on days 2 and 3 and comparable on
days 4 and 5 (although 2-fold lower on day 6). Furthermore, DBA2
given 500 mg/kg/day x 5 days had small intestinal drug levels
(measured on days 4–6) 5-fold higher than those found in the small
intestine of BALB/c mice after 10 mg/kg (Fig. 7)
. This does not
necessarily reflect the distribution of polyglutamate forms.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSEffect of RTX on...Plasma dThd, dUrd, and...Plasma and Tissue RTX...DISCUSSIONREFERENCES |
|---|
Of interest has been the observation that BALB/c mice are more
susceptible than DBA2 mice to the toxic effects of RTX with MTDs of
5 and >500 mg/kg, respectively, when given in a daily x 5
regimen. This very large difference appears to be attributable to an
increased sensitivity of BALB/c mouse intestine to RTX. At a dose of 10
mg/kg/day x 5, greater weight loss, evidence of diarrhea, and low
tolerance (moribundity) was associated with more marked
histopathological changes to the small intestine of BALB/c mice
compared with DBA2 mice. The effects observed are related to inhibition
of TS, as shown by the fact that coadministration of dThd largely
prevented the majority of the weight loss induced by RTX in BALB/c
mice. Increasing the dose of RTX administered to DBA2 mice to 100
mg/kg/day x 5 led to weight loss and histological consequences of
damage (inhibition of proliferation and/or tissue degeneration
attributable to cell death) that was comparable with that seen in
BALB/c mice at the lower dose. However, the weight loss was less rapid,
had a later nadir (day 9 rather than day 8), recovered more slowly, and
was not associated with diarrhea. Furthermore, there was evidence of a
greater improvement in the villus architecture of DBA2 compared with
BALB/c mice by day 8. This suggests that toxicity to another organ may
be contributing to the slow recovery of weight in DBA2 mice at 100
mg/kg. Similarly, the reason for the higher 100-mg/kg dose not being
tolerated in BALB/c mice (small intestinal histology similar at both
doses) may relate to the combined effects of a higher susceptibility of
these mice to intestinal damage and the toxic effects of RTX to another
organ. This toxicity could be, for example, to another part of the
gastrointestinal tract, giving rise to malabsorption or anorexia, or to
the bone marrow. Indeed, at 100 mg/kg daily x 5, severe
neutropenia was seen in DBA2 mice (day 8).
Thus, the relevant features of this mouse strain toxicity difference to be explained is the earlier weight loss, the presence of diarrhea, and the greater histopathological evidence of damage at an equivalent 10 mg/kg/day dose in BALB/c mice. Interestingly, DBA2 mice given 500 mg/kg/day x 5 had greater crypt changes and equivalent villus atrophy (day 6) in the small intestine compared with BALB/c mice given 10 mg/kg/day x 5. Nevertheless, this high dose was tolerated and not associated with early weight loss or diarrhea, suggesting that: (a) small intestinal damage, as measured by histopathological changes, in BALB/c mice is accompanied by, but not directly linked to, severe functional damage to this organ in this mouse strain at the cellular level; and/or (b) the small intestine is not the only area of the intestinal tract to be more sensitive to damage induced by RTX in BALB/c mice. Consistent with the latter is the observation that in BALB/c mice given 10 mg/kg/day x 5, mild histopathological changes to the colon were observed that were not seen at any dose in DBA2 mice. Thus, it may be these types of effects, which when combined with the small intestinal effects common to both mouse strains, that produces the severe nontolerated consequences in BALB/c mice.
The fact that the concentration of peripheral blood elements was not
suppressed more in BALB/c than DBA2 mice at the 10-mg/kg/day dose of
RTX is consistent with the difference in drug sensitivity between the
two strains being confined to the gastrointestinal tract. Furthermore,
it seems unlikely that systemic factors such as baseline plasma dThd or
folate levels are directly responsible, because measurement of
pretreatment levels of dThd or folate levels demonstrated no
difference. dThd fell by 50% (and dUrd increased 3-fold) after the
first treatment in both strains, suggesting that TS inhibition was
occurring (not necessarily to an equal degree). This reduction in
salvageable dThd is believed to be attributable to the inhibition of TS
in proliferating tissues, which both reduces the amount of dThd
nucleotides/nucleosides produced and increases the salvage of
dThd from the plasma. In DBA2 mice, the dThd level remained at this
reduced level after the fifth injection of RTX (day 6). However, the
dThd level in RTX-treated BALB/c mice had returned to the pretreatment
level. These data are contrary to what might have been predicted from
the increased toxicity seen in BALB/c mice, but it could be argued that
increased intestinal tissue damage was contributing toward increased
plasma dThd in the BALB/c strain. Alternatively, it could indicate a
higher rate of dThd salvage in the DBA2 mouse intestinal epithelium,
which in turn could be responsible for the difference in tolerance to
RTX. The reduced dThd level in both strains at day 12, when both body
weight and dUrd had recovered, could be attributable to increased
proliferation and hence demand for DNA synthesis during this period.
RTX is rapidly eliminated from the plasma of mice, although a prolonged terminal phase of elimination is evident (14) . However, polyglutamate retention in tissues such as liver, kidney, and intestinal epithelium leads to high tissue:plasma ratios that persist for several days (2 , 14) . This accounts for the delay in the recovery from the toxic effects of RTX after completion of 5 days of injections. Potentially, differences in drug uptake and/or polyglutamation in the intestinal epithelium between mouse strains could explain the differences observed in drug sensitivity. Indeed, increased weight loss and more marked changes to histology in BALB/c compared with DBA2 mice at the 10-mg/kg daily dose was associated with (but not necessarily caused by) increased drug levels in the small intestinal epithelium of BALB/c mice between days 4 and 7. However, it is highly relevant to note that on day 3, the gut epithelium from BALB/c mice did not have higher drug levels than from DBA2 mice (in fact, the reverse was true). Nevertheless, greater histopathological changes were observed on day 3 and more notably on day 4, consistent with greater damage in BALB/c small intestine. This was associated with greater weight loss in BALB/c mice by day 4. We cannot, therefore, eliminate the possibility that the higher drug levels measured at later times are an "effect" rather than a "cause" of greater damage. For example, reduced proliferation/more damage to the small intestine of BALB/c mice may result in higher drug levels per gram of tissue. Alternatively, drug-induced functional changes that may lead to diarrhea, coupled with dehydration and weight loss, may compromise renal function, leading to retention of the drug in the circulation and further uptake into the intestinal tissue.
Relevant is the fact that over the first 3 days of treatment, the
drug levels were found to be higher in the small intestinal epithelium
of DBA2 mice given 100 mg/kg/day x 5 than in BALB/c mice given 10
mg/kg/day x 5, but this was not coupled with histopathological
differences (measured on days 2, 5, and 6). Perhaps the most convincing
evidence that higher drug accumulation in the small intestine is not
linked to diarrhea/moribundity/early weight loss in BALB/c mice comes
from the observation that DBA2 mice given 500 mg/kg/day RTX not only
accumulated significantly higher (5-fold) small intestinal
epithelial drug levels (measured on days 4–6 only) compared with
BALB/c mice given 10 mg/kg but had comparatively more "crypt
damage" to the small intestine (although there was no evidence of
increased villus atrophy). Nevertheless, there was less rapid weight
loss and no evidence of diarrhea.
In summary, it appears that the gastrointestinal tract of BALB/c mice is susceptible to an "early phase" toxicity induced by RTX that may not be related to increased drug accumulation. Instead, this may be a colon/small intestinal dysfunction/damage, leading to early weight loss and diarrhea which, when combined with the later effects induced by higher doses, causes an intolerable amount of damage and a low MTD. The causative factor may be attributable to metabolic differences, such as a reduced dThd salvage capability in BALB/c mouse intestinal epithelium, attributable to, for example, reduced thymidine kinase expression. Studies outside of the scope of this report did not show a significant difference in TS or FPGS mRNA expression or TS protein levels. These studies will be expanded to activity measurements and determining the relative contribution of the de novo and dThd salvage pathways to thymidylate synthesis in these two mouse strains. An alternative explanation and continuing area of investigation is that BALB/c mouse intestinal epithelial cells (small intestinal and/or colonic) may sense drug-induced cellular damage more readily, leading to more rapid engagement of cell death pathways. Indeed, ongoing studies suggest increased susceptibility of BALB/c small intestinal epithelium to induction of apoptosis, not only 24 h after treatment with RTX (30) , but also 4.5 h after ionizing radiation (31) . Regardless of the underlying mechanisms of this strain difference, a BALB/c mouse model has been described that appears to be highly relevant to the study of RTX-induced gastrointestinal toxicity in humans. Indeed, it has proved to be a useful model for studying the effects of potential rescue agents such as dThd and LV (2) .
|
FOOTNOTES |
|---|
2 Present address: Sydney Cancer Centre,
Camperdown, New South Wales, Australia 2050.
3 Present address: CRC Cellular and Molecular
Pharmacology Group, School of Biological Sciences, University of
Manchester, Manchester M139PT, United Kingdom.
4 Present address: The Royal Marsden Hospital
Trust, Down’s Road, Sutton, Surrey, SM2 5NG United Kingdom.
5 To whom requests for reprints should be
addressed, at The CRC Centre for Cancer Therapeutics at the Institute
for Cancer Research, 15 Cotswold Road, Sutton, Surrey SM2 5NG, United
Kingdom. Phone: 44-181-643-8901; Fax: 44-181-643-6940; E-mail: annj{at}icr.ac.uk
6 The abbreviations used are: RTX, raltitrexed;
TS, thymidylate synthase; 5-FU, 5-fluorouracil; LV, leucovorin; MTD,
maximum tolerated dose; FPGS, folylpolyglutamate synthetase; HPLC,
high-performance liquid chromatography.
Received 8/25/99; revised 10/12/99; accepted 10/13/99.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSEffect of RTX on...Plasma dThd, dUrd, and...Plasma and Tissue RTX...DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. D. Gibbs, D. S. Theti, N. Wood, M. Green, F. Raynaud, M. Valenti, M. D. Forster, F. Mitchell, V. Bavetsias, E. Henderson, et al. BGC 945, a Novel Tumor-Selective Thymidylate Synthase Inhibitor Targeted to {alpha}-Folate Receptor-Overexpressing Tumors Cancer Res., December 15, 2005; 65(24): 11721 - 11728. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. W. Aherne, A. Hardcastle, E. Ward, D. Dobinson, T. Crompton, M. Valenti, L. Brunton, and A. L. Jackman Pharmacokinetic/Pharmacodynamic Study of ZD9331, a Nonpolyglutamatable Inhibitor of Thymidylate Synthase, in a Murine Model Following Two Curative Administration Schedules Clin. Cancer Res., September 1, 2001; 7(9): 2923 - 2930. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. van der Wilt, H. H. J. Backus, K. Smid, L. Comijn, G. Veerman, D. Wouters, D. A. Voorn, D. G. Priest, M. A. Bunni, F. Mitchell, et al. Modulation of Both Endogenous Folates and Thymidine Enhance the Therapeutic Efficacy of Thymidylate Synthase Inhibitors Cancer Res., May 1, 2001; 61(9): 3675 - 3681. [Abstract] [Full Text]
|
|
|
|
|
|
D. M. Pritchard, L. Bower, C. S. Potten, A. L. Jackman, and J. A. Hickman The Importance of p53-Independent Apoptosis in the Intestinal Toxicity Induced by Raltitrexed (ZD1694, Tomudex): Genetic Differences between BALB/c and DBA/2 Mice Clin. Cancer Res., November 1, 2000; 6(11): 4389 - 4395. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. C. Farrugia, G. W. Aherne, L. Brunton, S. J. Clarke, and A. L. Jackman Leucovorin Rescue from Raltitrexed (Tomudex)-induced Antiproliferative Effects: In Vitro Cell Line and in Vivo Mouse Studies Clin. Cancer Res., September 1, 2000; 6(9): 3646 - 3656. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||
, control; 
, 10 mg/kg/day
(also shown by a dotted line); 
, 50 mg/kg/day; 
,
100 mg/kg/day; •, 500 mg/kg/day. B, BALB/c mice. 
, day 5; 
, day 6. Bars, SD.
*, statistically significant difference between mouse strains
(P < 0.05).