[CANCER RESEARCH 32, 483-490,
March 1972]
Blood Flow in Irradiated Mouse Sarcoma as Determined by the
Clearance of Xenon-1331
Robert F. Kallman,2 Gerald L. DeNardo,3 and Mary J. Stasch
Department of Radiology, Stanford University School of Medicine, Stanford, California 94305
SUMMARY
133Xe dissolved in 0.9% NaCl solution was injected directly
into KHT sarcomas in C3H mice in order to study tumor
blood flow. The rate of blood flow in ml/100 g/min was
calculated from the clearance half-time and the partition
coefficient. The mean rate of blood flow in the tumors of
unanesthetized control mice was 21.2 ±0.8 ml/100 g/min,
which was significantly faster than the blood flow in the
tumors of pentobarbital-anesthetized
control mice. ' 33Xe was
cleared exponentially in virtually all tumors, and rate of
clearance decreased with increasing tumor volume in tumors
larger than approximately 500 to 600 cu mm; in smaller
tumors there was no conclusive evidence of dependence of
clearance rate on size. Clearance rate was unaffected by
multiple daily injections of 133Xe. At 3 hr after localized
tumor irradiation (250 kV; half-value layer, 5.9 mm Al) with
doses greater than 1,000 rads, blood flow tended to decrease
with increasing radiation dose. At 3 to 4 days postirradiation,
rate of blood flow was significantly increased in tumors
receiving 1,000 rads; blood flow was also increased in tumors
exposed to 2,000 and 4,000 rads, but not until approximately
7 days. Rate of blood flow decreased at 3 hr and 3 to 4 days in
tumors
irradiated
with 8,000 to 16,000 rads. The
postirradiation increases in blood flow were not correlated
simply with gross tumor shrinkage, although tumor growth
was slowed proportionally to radiation dose. Thus, irradiation
brought about changes in rate of blood flow in the KHT
sarcoma, and these changes were dependent on both radiation
dose and postirradiation time. The observed time course of
radiation-induced blood flow changes may partially account
for the tumor reoxygenation shown in our previous studies
and those of others.
supplied with oxygen, irradiation preferentially kills those cells
that contain oxygen, while anoxic or severely hypoxic tumor
cells are spared. The events that supervene during the 1st 1 to
2 days after moderate radiation doses have been termed
"reoxygenation,"
and this is characterized by a return of the
tumor to the net oxygénationstatus approximating that which
it exhibited prior to irradiation. Although "reoxygenation" is
a designation which cannot rigorously be supported by direct
measurements, it is difficult to conceive of any other
phenomenon that would account for the observed data.
Angiographie studies (18) have established that irradiation is
followed by alterations in the intravascular volume of certain
experimental tumors, and it has been inferred that such
alterations "may be translated into a relative increase of tumor
tissue oxygénation"(19).
The use of radioactive inert gases '3 3Xe and 8 5Kr for
studying circulatory physiology is now well established (4,
12-14). Tumor blood flow has been studied by the rate of
clearance of 85Kr from tumors in rabbits; here, the isotope
had entered the tumor from the lungs via the arterial
circulation. With the introduction of the direct injection
technique (13), the placement of dissolved '3 3Xe directly into
tumors has been used to measure blood flow in both human
(7) and transplanted (15-17) tumors. We used this technique
to determine whether the reoxygenation which occurs in
irradiated tumors might be related to the effects of irradiation
on tumor blood flow.
MATERIALS AND METHODS
Tumors and Irradiation. The KHT sarcoma in syngeneic
C3H mice was used for these experiments. Tumors were
transplanted from single cell suspensions prepared as described
previously (11): 10s viable tumor cells were inoculated
INTRODUCTION
intradermally in a volume of 0.05 ml. These inocula reached
sizes appropriate for clearance studies (about 100 to 200 cu
In a prior series of radiobiological experiments, we have
mm) after approximately 8 to 20 days. For irradiation, the
studied functional cellular changes that occur in tumors after
mice were anesthetized with i.p. sodium pentobarbital in a
irradiation (9-11, 24). As most tumors are heterogeneously
dosage of 0.06 mg/g body weight. The tumors, together with
'This investigation was supported by USPHS Research Grant CA overlying skin, were retracted gently and sandwiched between
the halves of special Plexiglas localizers (8) from which air was
03353 from the National Cancer Institute.
'Recipient of USPHS research career program award ÇAK3-3576 excluded by the use of a tissue-equivalent bolus made of rice
from the National Cancer Institute.
3Present address: Division of Nuclear Medicine, Department of paste. The localizers were fitted over a brass tube that is part
of a 6-place cone; this permits the simultaneous irradiation of
Radiology, University of California School of Medicine at Davis, Davis,
6 tumors in localizers. Thus assembled, the tumors were fixed
Calif. 95616.
at a position 35 cm from the target of a 250-kV constant
Received April 5, 1971 ; accepted November 22, 1971.
MARCH 1972
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1972 American Association for Cancer Research.
483
R. F. Kallman, G. L. DeNardo, and M. J. Stasch
potential X-ray machine, operated at 15 ma, without added
filtration and with a half-value layer of 5.9 mm Al. The
radiation dose rate was 301 rads/min. Each irradiated tumor
received a single dose ranging from 1,000 to 16,000 rads.
Radioisotope Techniques. '3 3Xe was dissolved in 0.9%
NaCl solution with a minimum concentration of 2 mCi/ml.
For injection into tumors, a 250-ßlsyringe was filled with the
isotope solution from a rubber-stoppered
injection vial;
precautions were taken to keep the solution free of air in both
the vial and the syringe. The procedure for tumor injection
was usually as follows: the point of the needle (30-gauge;
outside diameter, 0.3 mm) was made to enter through the skin
and into the tumor near its base; the needle was advanced
halfway across the diameter of the tumor, and 10 ¿/Iof
radioactive solution were deposited. The needle was then
withdrawn along the same track, but not removed from the
tumor, and then angled about 45°to the right or left; the
needle was advanced about halfway to the opposite margin of
the tumor and another 10 fi\ were deposited; this procedure
was repeated after the needle had been angled approximately
45° to the opposite direction. The total volume of 0.9%
NaCl solution injected was 30 /nl, which contained 60 to 150
juCi and resulted in maximum counting rates of 100,000 to
300,000/min.
Immediately
after
injection
of
the
radioisotope, the mouse was placed in a plastic box of inside
dimensions 3 x 3 x 7.4 cm; this was fitted with a thin Mylar lid
that was perforated for ventilation, as were the 2 end panels of
the box. The box was sufficiently large so that a mouse could
assume a normal rest position and could move a few mm in
any direction. The longest dimension of the box was
approximately the same as the diameter of the collimator of
the scintillation detector. The box was placed on a table top
with its lid about 2 cm from the end of the collimator.
For experiments with anesthetized mice, the animals were
placed on a 0.25-inch-thick lead diaphragm with a central 1-cm
diameter hole through which the isotope-injected tumor was
allowed to hang pendantly over the scintillation detector. This
lead diaphragm prevented counting of tissues other than the
tumor and overlying skin. Only a limited number of mice was
anesthetized for radioisotope clearance trials. All clearances
performed in irradiated tumors were done without subjecting
the mice to subsequent anesthesia after they had recovered
from their irradiation anesthetic. As the anesthetic was
effective for 0.5 to 1 hr and complete recovery required up to
1.5 hr, the shortest postirradiation time used was 3 hr.
Radioactivity was detected by a 1.5-inch Nal crystal with a
cylindrical collimator. The output of the detector was
processed with a rate meter, with a time constant of 3 sec, and
traced continuously on a linear recorder (Chart 1). The
radioactivity from each animal was recorded for at least 8 min,
which represented at least 1, but usually several, half-times.
These data were replotted on semilogarithmic graph paper and
fitted with straight lines for facilitation of computation of the
half-times of radioisotope clearance. With very few exceptions,
clearance was a single exponential function for as long as 30
min after injection. The rare instances in which more complex
functions occurred could be traced to faulty injection
technique; these data were not used in any analyses. The
half-times could be converted to rates of blood flow by the
relationship: blood flow (in ml/100 g/min) = 100 X (In 2/r,/2).
The partition coefficient, X, was determined for samples of
the tumor under investigation with a slight modification of the
IO¡O
0.0•corZ3
0.6
O
0.50
Chart 1. Strip chart recordings of
count rate (ordinate, CPM X 10"s) as a
function of time after interstitial in
jection of '3 3Xe in 0.9% NaCl solution
0.3
Ci
02ì_
^0.1is.\
0¿73</*U*
2TT
into 2 tumors. Upper, tumor of mouse
246, f|¿= 2.3 min, rate of blood flow =
26.8 ml/100 g/min; lower, tumor of
mouse 277, f^ = 7.0 min, rate of blood
flow = 8.8 ml/100 g/min.
6
5
i,
Time in minutes
484
CANCER
RESEARCH
VOL. 32
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1972 American Association for Cancer Research.
Blood Flow in Irradiated Mouse Tumors
method described by Veall and Mallet (25). Homogenates of
KHT tumor were prepared by grinding weighed fresh tissue in
a glass homogenizer with a close-fitting Teflon pestle. A
measured amount of distilled water was added during
homogenization so that the final mixture would have the
consistency of heavy cream. Fresh mouse blood was drawn
from the infraorbital sinus and was hemolyzed by rapid
freezing and thawing. Samples of tumor homogenate,
hemolyzed blood, and 0.9% NaCl solution were introduced
into sample tubes, 7 cm long, which were made from soft glass
tubing with an outside diameter of 7 mm and an inside
diameter of 5 mm. The tubes were half-filled with sample, and
to this was added approximately 0.1 ml of air containing 30 to
50 /¿Ciof 133Xe; immediately afterwards the tubes were
counts made from hemolyzed blood relative to counts in the
gaseous phase in blood sample tubes.
Blood Flow in Tumors of Anesthetized Mice. At the outset
of this study, it was intended that measurements of
radioisotope clearance from tumors be made by use of
anesthetized animals as described above. It became obvious
that the state of anesthesia itself significantly perturbed the
rate of tumor blood flow; there were no differences in the
shape of the '3 3Xe disappearance curves-only the slope was
water bath. Measurements in the gaseous and liquid phases of
each sample tube were made by simply sliding the sample tube
under its clamp, thereby placing the volume to be counted
directly in front of the 5- x 9-mm rectangular aperture in the
lead diaphragm.
at daily intervals. For this analysis it was necessary to use
blood flow data obtained with unirradiated tumors because
irradiation per se might be expected to change the blood flow
(see below); but it was possible separately to examine selected
data from irradiated tumors, groups of which had been
injected with 133Xe for a 1st or 2nd injection at the same
affected. Because of these findings and the difficulties in
achieving a reproducible level of anesthesia in these animals, all
subsequent measurements were made on unanesthetized mice.
The frequency distributions of blood flows in the tumors of
anesthetized and unanesthetized mice are shown in Chart 2.
sealed with an oxygen torch. These tubes were then fixed in The 36 determinations made with anesthetized mice showed a
the horizontal position to a shaker that was immersed in a mean blood flow of 13.8 ±1.2 ml/100 g/min, and the 133
water bath at 37°.They were shaken continuously for 2, 4,
tumors of unanesthetized
mice showed a significantly (p
and 24 hr before being removed for radioactivity
< 0.001) greater mean blood flow, 21.2 ±0.8 ml/100 g/min.
Single versus Multiple Injections. As the principal goal of
measurements. Immediately before measurement, each tube
was vigorously shaken by hand and then briefly centrifuged in this study was to measure changes in blood flow as a function
50-ml centrifuge buckets filled with water at 37°. This
of time after tumor irradiation, it was necessary to determine
procedure effectively separated the gaseous and liquid phases whether the ability of a tumor to clear a given injectate was
in the sample tubes. For radioactivity measurements, sample influenced or in any way compromised by the same tumor
tubes were clamped to a lead diaphragm that was attached to having been injected previously. The data listed in Table 1
the end of the collimator of a waterproofed scintillation
provide no evidence of change in blood flow with multiple
detector; the scintillation detector was immersed in a 37° injections of ' 33Xe. These 1 to 4 injections were administered
time after a given dose of radiation. These data also gave no
RESULTS
Partition Coefficient. Incubation of the sample tubes for 2
hr was sufficient to achieve equilibration of the ' 33Xe in all of
30
the samples tested.
The partition coefficient, XT, was obtained from the
solubility of ' 33Xe in tumor, ST, relative to the solubility in
20
Anesthetized Controls
n-36
BFM3.ôml/100q/min
mouse hemolyzed blood, SB :
10
As ST was determined to be 0.1288 and SB was 0.1446,
the partition coefficient was 0.891. This value was used to
convert all clearance half-times to rates of tumor blood flow in
ml/ 100 g/min.
ST was calculated from the relationship:
ST=[CT(WT+Vnì)-Vnì(SMl)]l[WT]
where CT is taken from the count rate of the tumor
homogenate and expressed as a fraction of the count rate of its
gaseous phase, WT is the volume of tumor tissue in its
homogenate (weight of tumor/tumor-specific gravity), Fsa, is
the volume of 0.9% NaCl solution added to the fresh tumor in
preparing the homogenate, and 5sal is the solubility of the
isotope in 0.9% NaCl solution, i.e., counts in NaCl solution
relative to counts in the gaseous phase. SB was obtained from
a- °
Unoncathctizcd Controla
30
n = 113
DF = 21.2 ml/100q/min
20
<TOF=6.3
crgfO.6
C = «Tßp/gp'0.39
10
O
10
20
30
4.0
60
Rate of blood flow (ml/100g/min)
Chart 2. Percentage frequency distributions of rates of blood flow
for tumors of anesthetized (upper) and unanesthetized (lower) mice.
MARCH 1972
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1972 American Association for Cancer Research.
485
R. F. Kallman, G. L. DeNardo, and M. J. Staseli
Table 3
Rate of blood flow0 in small and large tumors determined from the
rate of clearance of different volumes of '3 3Xe injected
evidence for changes in blood flow as a result of prior
injection.
Radioisotope Clearance in Tumors of Different Sizes and
with Different Injection Volumes. For tumors covering a wide
range of volumes and receiving multiple-locus injections, there
was no statistically significant difference in rate of blood flow
until the volume of approximately 560 cu mm was exceeded
(Table 2). Mean blood flow was slower in tumors greater than
this size. The variability in blood flow rates calculated from
133Xe clearances and the tendency for flow rate to be slower
Injection volume
(cumm)20
40
60Small
All (20-60)
tumorsb26.0
tumors020.6
tumors
(small
large)23.3
+
±3.1
±1.1
23.6 ±4.0
16.5 ±5.6
33.5 ±0.9Large23.6 ±8.9dAll
±1.8
20.1 ±3.5
29.2 ±3.9
27.7 ±2.0
24.0 ±1.9
19.9 ±3.0
in larger tumors are illustrated by the data in Table 3. In this
0 Shown as mean blood flow, HI-',in ml/100 g/min ±OBF.
experiment, the injected material was deposited in a single
6 Tumors with mean volume of 97 cu mm. Range is 59 to 126 cu
locus, i.e., at only 1 point rather than the usual 3 loci as mm; a y^j is 7 cu mm.
c Tumors with mean volume of 310 cu mm. Range is 207 to 498 cu
described above. Although for the 3 different injection
volumes the mean blood flows of the smaller tumors were mm; oy^j is 26 cu mm.
d n = 3 tumors. All other groups contained 4 tumors. The pooled
always greater than the larger tumors, none of these
differences was statistically significant, owing presumably to data in Column 4 represent 8, 8 and 7 tumors; the pooled data in
the small sample size and the relatively great variability of Line 4 represent 12, 11, and 23 tumors.
these data. Pooling these 3 sets of data for the 3 injection
volumes (Table 3, Line 4) gave a significantly (p < 0.05)
100
greater mean blood flow in the small tumors than in the large
00
ones. Within the relatively narrow range of injection volumes
60
tested, therefore, this experiment provided no clear-cut
evidence of dependence of radioisotope clearance rate on the
40
amount of 0.9% Nad solution injected.
c
For the collected data of Table 2, the best-fitting
polynomial curve was of the form Y = 0.525 + 0.869 X —¿ 1 ìZO
0.234 X*, where X is log tumor volume and Y is log blood
flow. This curve (Chart 3) shows that, within the range of
volumes studied and on the basis of clearances following the
standard multiple-locus
radioisotope
injection technique,
blood flow did not change appreciably up to volumes of about
500 cu mm and then became increasingly slower as tumor size
1,0
Iô
^
6
tu 4. Comparison
Table 1
of tumor Mood flow with frequency
OBP(ml/
±
ofinjection1234BF°
Day
g/min)17.7
100
of injection of '3 3Xe
±0^1(cu
2 -
mm)127
1.220.8±
17.6163
±
2.017.7
±
1.918.9
±
±1.9Volb
21.5182
±
17.7234
±
±22.8
a Mean rate of tumor blood flow (BF) for 14 tumors,
k Mean tumor volume for the same 14 tumors.
¿O 60 00 100
200
Tumor
400
600 SOU1000
2000
volume (cumm)
Chart 3. Rate of tumor blood flow (log scale) as a function of tumor
volume (log scale) for 140 determinations.
increased further. Because of this finding, no irradiation
experiments were performed with tumors larger than 500 cu
mm at the time of irradiation; also, the data obtained from
tumors that reached these higher volumes during the course of
postirradiation
growth were not used in the calculations
Tumor volume
g/min)22.1
100
mm)35-7071-140141-280281-560561-11201121-2240n"11463130193BF"±agp(ml/
(cu
presented below.
Effect of Radiation. The blood flows listed in Table 4 may
1.822.8
±
be examined against a mean blood flow of 21.2 ±0.8 ml/100
1.019.0
±
g/min in all 113 control mice, with the use of unirradiated
1.319.9
±
2.013.7
±
tumors in unanesthetized
mice. As it was necessary to
1.56.7
±
anesthetize the mice for localized irradiation and no clearance
±0.9
studies were to be done in animals under anesthesia, it was
0 The number of tumors tested.
impossible to determine
blood flows promptly
after
b Mean rate of tumor blood flow (BF).
irradiation, e.g., within 1 to 2 hr. While animals were usually
Table 2
Rate of blood flow in tumors of different sizes
486
CANCER RESEARCH VOL. 32
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1972 American Association for Cancer Research.
Blood Flow in Irradiated Mouse Tumors
Table 4
Tumor blood flow at different times after localized irradiation
3hr
3 to 4 days
a§p(ml/100g/min)20.3
±
±ay^,
Dose
(rads)1,0002.0004,0008,00012,00016,000n22121413129BF°
(cu
(cumin)165
mm)
H
1.819.0
±
0.715.±
1.817.4
7 ±
1.315.4
±
2.210.9
±
±1.1Volb
58197 16
31179 31
27207 21
21195 33
47291 20
31
11
5
52
10
9
7
7 days
BFa ±ogp
Volb i oy^,
(ml/100 g/mm)
(cu mm)
/;
(ml/100 g/min)
138
291
294
278
343
310
6
12
24
22.1 ±4.2
25.2 ±3.0
34.6 ±3.5
30.2 ±3.2
23.6 ±1.5
19.0 ±1.1
13.8+2.5
11.4 + 2.1
12.2 ±1.7
11
48
11
BF°±agj:
Voi6 ±oy^
305 +
135 +
156 ±
34
Mean rate of tumor blood flow (BF).
Mean tumor volume.
awake within 1 hr after irradiation, the anesthetic effect was volume at the time of irradiation. Tumor volume data shown
not completely
absent for approximately
another hr; for 1,000, 2,000, and 4,000 rads in Table 4 were drawn from
therefore, 3 hr was chosen as the earliest time at which the different experiments performed with Day 0 irradiation
tumors could be studied without complications arising from volumes which, although uniform within experiments, differed
anesthesia. At this time there was no clinical evidence of considerably from one experiment to another. Thus, these
anesthetic effect: respiration, heart rate, and behavior were data may not be used for the construction of postirradiation
entirely normal. Table 4 shows that, whereas there was no 3-hr growth curves. For the higher doses of 8,000 to 16,000 rads,
effect after 1,000 rads, blood flows at this early time tended however, the data for the 2 postirradiation times are from the
to be significantly and increasingly lower with higher radiation same experiments; «is lower at the 3- to 4-day interval only
doses.
because of deletion of individual blood flow data on the basis
The gross shrinkage that occurs in this tumor, especially at 4 of tumor volumes exceeding 500 cu mm or for purely
days and later and after doses greater than 4,000 rads, technical reasons. The data in Table 4 illustrate that, even at
precluded the investigation of blood flow changes at longer these dose levels that would be expected ultimately to bring
postirradiation intervals for the 3 higher doses. As summarized about virtually 100% regression, the average volume at 3 to 4
in Table 4, the irradiated tumors behaved in either of 2 ways days was greater than at the time of irradiation.
at the 3- to 4-day interval as contrasted with their acute
response at 3 hr: with 1,000 to 4,000 rads, the blood flow was
greater at the longer time, but only 1,000 rads increased the DISCUSSION
blood flow significantly above that of unirradiated controls at
the 3- to 4-day interval. Tumors that received doses greater
To the extent to which the clearance of interstitially
than 4,000 rads all showed blood flows at 3 to 4 days that deposited 133Xe is a valid measure of the state of the
were either lower than or approximately the same as (16,000
circulation, irradiation induced changes in the rate of blood
rads) the rates at 3 hr. None of these differences was flow in the tumors studied. It must be recognized, however,
statistically significant, but all were statistically significantly
that data provided by radioisotope clearance methods
lower than the blood flows in unirradiated tumors. At 7 days, constitute only indirect indicators of blood flow. The changes
the rate of blood flow in tumors irradiated with 1,000 rads that we observed were dependent upon radiation dose and
appeared to decline, but the difference was not statistically
postirradiation
time. Within the first few hours after
significant, owing primarily to the relatively great variability irradiation, high doses appear to cause a slowing of blood flow,
associated with the 7-day mean. After 2,000 rads, blood flow which is more pronounced as the dose is increased. The effects
increased progressively and was maximal at 7 days. After of single doses of 8,000 to 16,000 rads must be regarded as of
4,000 rads, there was a reversal in the earlier decline in rate of only theoretical interest, rather than of practical importance,
flow so that the highest rate, 34.6, was achieved at 7 days.
as this kind of treatment regimen is not used in modern
Table 4 also shows the course of tumor volume changes clinical radiotherapy. Also, it is unlikely that fractionated
associated with the different radiation doses. In almost every regimens reaching total doses of 6,000 rads, for example,
experiment, the mean tumor volume at 3 to 4 days was greater would elicit the same vascular changes as single doses of this
than at the time of irradiation, although tumor growth rate magnitude. The early decreased blood flow could result from
was reduced during this 3- to 4-day postirradiation period. In the release of vasoconstrictor substances resulting from local
addition to decreases in growth rate, the irradiated tumors also tissue radiation injury. Particularly in the case of low to
showed some shrinkage for all the doses used. Shrinkage never moderate radiation doses, such effects would be expected to
started earlier than 1 day postirradiation, and usually was not be short-lived and of relatively little consequence in curative
sufficiently advanced at the time of any of these '3 3Xe cancer radiotherapy. The longer term reductions in blood flow
clearance studies to have reduced tumor size to below the brought on by the very high doses may indeed be provoked by
MARCH 1972
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1972 American Association for Cancer Research.
487
R. F. Kallman, G. L. DeNardo, and M. J. Staseli
the same mechanism. An alternative explanation for the
observed reductions in blood flow is the fragmentation in
capillary architecture suggested by Rubin and Casarett (19).
However, the absence of relevant, specific evidence to
document these explanations identifies a need for further
investigation.
The improved blood flow found to occur up to 7 days after
the lower doses is of greater practical importance. Increased
blood flow tended generally to occur at later postirradiation
times with higher doses. In view of the variability reflected by
the standard errors of the mean blood flows, it is difficult to
detect unequivocal evidence of dose dependence.
The
maximum mean blood flow was 34.6 ±3.5 ml/10 g/min; this
was observed at 7 days after 4,000 rads. The next-highest rate
was 30.2 ±3.2ml/100 g/min; this occurred at 3 to4 days after
1,000 rads. Contrasted with the mean control rate of blood
now of 21.2 ±0.8 ml/100 g/min, the blood flow rate of
irradiated tumors appears to have been capable of being
increased by approximately 50%. Additional experiments in
the future are needed to establish, for several X-ray doses, the
exact time when blood flow is maximal and the height which
is attained at that time.
Provisionally, it is proposed that a mechanism that may
underlie the later increases in blood flow involves an altered
balance between tumor cell parenchyma and vascular stroma.
Thus, with increasing radiation dose, cell death and breakdown
may occur over a longer period, and this could account for
finding maximal flow rates at a slightly earlier time in the case
of the lowest dose tested.
Increased blood flow also may result from the relief of a
type of circulatory obstruction which is typical of rapidly
growing tumors (21, 22). Owing to its well-documented effect
of reducing parenchymal growth rate, irradiation should allow
the circulation to utilize such preexisting and relatively
radioresistant vascular channels more efficiently. Without
additional information, such as that afforded by careful
histological analyses, it is impossible to invoke any mechanism
with confidence.
Enhancement of tumor blood flow rates by irradiation
would be expected to improve tumor oxygénationby making
oxygen available to sites otherwise deprived of it, presumably
because of the growth momentum of the tumor itself. Whether
improved oxygénationis in fact achieved in these irradiated
tumors and whether this accounts for the kind of tumor
reoxygenation determined radiobiologjcally can be established
only by other experimental approaches (9, 11, 24). Although
presently available methods are not capable of providing
answers to these all-important questions, the likelihood that
increased blood flow plays an important causative part in
improving oxygénation of tumors is strengthened
by
polarographic measurements of tumor oxygénation (2, 3);
namely, tumor pO2 tended to be higher after radiotherapy
than before.
It seems reasonable that changes in blood flow as reported
herein could account for postirradiation reoxygenation. The
diffusion radius around blocked or collapsed capillaries might
be imagined as substantially less than 100 Aim, the critical
maximum diffusion radius suggested originally by the data and
extrapolations of Thomlinson and Gray (23). Opening all
488
capillary channels would at least ensure that all pericapillary
tumor cells would be subject to the same amount of oxygen to
a distance of approximately 100 jum. Although this reasoning
is attractive, the time scale of the blood flow changes makes
these findings inconsistent with separate observations on the
reoxygenation of this same tumor. The present experiments
give no evidence that significant improvements in blood flow
occur before approximately 3 days after irradiation, whereas
the radiobiological experiments suggest the reoxygenation
process (see "Introduction")
to be more than half-completed
by 3 hr and fully completed in this same tumor by
approximately 12 to 24 hr (9-11, 24). Although the present
report does not include observations between 3 hr and 3 days,
our incomplete data at intermediate times such as 1 day do
not indicate significant increases in blood flow before
approximately 3 days.
The present findings are at variance with those of Song and
Levitt (20), who worked with Walker Carcinoma 256 in rats,
but the different results might be entirely attributable to
differences in tumor, species, and experimental method
(s ! Cr-labeled red blood cells). It is more difficult to reconcile
our findings with those of Robert et al. (16), who used 133Xe
in a manner similar to ours and reported marked reductions in
blood flow rate (débitvasculaire spécifique)between 1 and 9
days after irradiation with 2,000 R in a rhabdomyosarcoma
transplanted and studied syngeneically in C3H mice. They
emphasized
that
this effect
was obtained
only in
well-vascularized tumors, Le., with blood flows of at least 5
ml/100 g/min; however, their more completely documented
publication (17) shows that they never found blood flows of
greater than double this rate. Considering that our mean flow
rate was 21.2 ml/100 g/min (see Charts 2 and 3 for
distributions of flow rates in our KHT sarcoma), it seems
likely that the discrepancy is related to those differences in
blood flow that are uniquely a function of differences in
histological structure of their rhabdomyosarcoma
and our
undifferentiated anaplastic sarcoma. Should this be the case,
one can readily appreciate
the need for caution in
extrapolating either set of findings to other tumors, either
animal or human.
The xenon clearance method used in these experiments
constitutes a useful and satisfactory means of studying the
circulation of experimental tumors, provided that the rate of
clearance of this radioisotope is sufficiently independent of
tumor size so that it can be used over a range of volumes
which are palpable, measurable, and of practical importance.
As there was little or no systematic change in clearance rate
over a wide and usable range of tumor sizes, application of this
method to radiobiological questions is amply justified.
Especially at the upper portion of our size range, there was an
inverse relationship between blood flow and tumor size; this is
similar to the findings of Cullino and Grantham (5). Their
study, however, was confined to larger tumors, usually greater
than 2 g, wet weight. The disagreement between the present
results and those of Cataland et al. (1) and Robert et al. (17) is
probably more illusory than real, as most of the tumors that
they used were larger than the ones that we used; furthermore,
their tumors were characterized by considerable necrosis and
were of different histological types.
CANCER RESEARCH VOL. 32
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1972 American Association for Cancer Research.
Blood Flow in Irradiated Mouse Tumors
KHT mouse sarcoma (0.0126 ml/hr/mg of tumor), after
allowance is made for the expected nitrogen content of wet
tumor tissue (approximately 10 to 20%). The report of slightly
lower flow rates for 3 other mouse tumors (15-17) may easily
be reconciled with the present findings: (a) a lower partition
coefficient was assumed by these other groups, namely, 0.72,
the value found for dog liver; (b) radioisotope clearance
measurements in 1 study (15) were made with anesthetized
animals; and (c) larger tumors were used, as noted above.
Although the direct introduction of a quantity of 0.9%
NaCl solution into these tumors must certainly cause some
disruption in tumor architecture, this disruption must be
minimal and easily reversible, as there were no obvious changes
in tumor growth or biological behavior. This is consistent with
the close agreement in flow rates between 1st and subsequent
injections of the radioisotope (Table 1; Ref. 17). As detailed
by earlier workers who have developed methods of
radioactively labeled inert gas clearance, the validity of this
approach rests on the assumption that dissolved gas molecules
diffuse through tissue spaces to be picked up in vascular
channels; then these dissolved molecules are carried into
venous drainage and then to the lungs, where they are expired
10
L,
6
and not recirculated. This assumption is made all the more
Days
reasonable when the method is applied in tissues in which the
Chart 4. Growth curve of KHT sarcoma in syngeneic C3H mice. components (tissue cells, interstitial fluid, and blood) partition
Symbols, mean tumor volumes in 2 separate experiments. », the radioisotope approximately equally. That approximately
Experiment 1, 6 tumors; o, Experiment 2, 6 tumors.
equal partitioning is achieved in the present experiments is
strengthened by finding the partition coefficient to be 0.891, a
A lower range of tumor volume acceptable for this kind of value slightly higher than that associated with the liver, in
experiment is established by a size which is conveniently and which circulation is routinely studied by this method.
reliably measurable (by calipers) in 3 dimensions. In addition,
this minimum tumor volume should be at least the 2nd or 3rd
such measurement on a given tumor, in order that the tumor REFERENCES
might be measured at least twice prior to treatment, thereby
ensuring the availability of a useful estimate of its
1. Catalana, S., Cohen, C., and Sapirstein, L. A. Relationship between
pretreatment
growth rate. These requirements
set the
Size and Perfusion Rate of Transplanted Tumors. J. Nati. Cancer
Inst., 29: 389-394, 1962.
minimum usable size of a tumor at a mean diameter of
approximately 5 mm (mean volume, 65 cu mm). The KHT
2. Cater, D. B., and Silver, I. A. Quantitative Measurements of
tumor grows nearly exponentially, but with continuously
Oxygen Tension in Normal Tissues and in Tumours of Patients
before and after Radiotherapy. Acta Radiol., 53: 233-256, 1960.
decreasing slope from the lower limit just mentioned to a
3.
Evans, N. T. S., and Naylor, P. F. D. The Effect of Oxygen
volume of approximately 1000 cu mm (Chart 4). At least until
Breathing and Radiotherapy upon the Tissue Oxygen Tension of
it reaches volumes approaching those close to the end of the
Some Human Tumours. Brit. J. Radiol., 36: 418-425, 1963.
period of near-exponential growth, this KHT tumor does not
4. Gelin, L.-E., Lewis, D. H., and Nilsson, L. Liver Blood Flow in
exhibit a simple gross distribution of necrosis that is anything
Man during Abdominal Surgery. I. Description of a Method
like the usual "orange" analogy, i.e., a central necrotic core
Utilizing Intrahepatic Injections of Radioactive Xenon (133Xe).
surrounded by a relatively thin, firm, and viable "rind."
Rather, necrotic foci which are frequently grossly apparent are
randomly distributed throughout
the tumor mass. It is
difficult to identify readily the appreciable volumes of necrosis
in tumors of approximately 9 mm mean diameter (volume,
382 cu mm).
Despite great differences in species, size of different
animals, tissue studied, and experimental techniques, the
blood flow of the KHT mouse sarcoma is similar to blood
flows found in other tumors. For example, in rabbit tumors
the blood flow rate was reported as 29 ml/100 g/min (6). The
average flow rates in rat and mouse tumors reported by
Cullino and Grantham (5), 0.14 to 0.17 ml/hr/mg of nitrogen,
are qualitatively similar to the average value established for the
5.
6.
7.
8.
Normal Values and Effect of Temporary Occlusion. Acta
Hepato-Splenol., 15: 13-20, 1968.
Cullino, P. M., and Grantham, F. H. Studies on the Exchange of
Fluids between Host and Tumor. II. The Blood Flow of Hepatomas
and Other Tumors in Rats and Mice. J. Nati. Cancer Inst., 27:
1465-1491, 1961.
Gump, F. E., and White, R. L. Determination of Regional Tumor
Blood Flow by Krypton-85. Cancer, 21: 871-875, 1968.
Johnson, R. J. R. A Comparison of the Effect of Hyperbaric
Oxygen and Oxygen Plus 5% CO2 on Tissue Circulation and
Oxygénation.Radiology, 98: 177-181, 1971.
Kallman, R. F. Some Aspects of the Time-Dose Relationship in
Spontaneous and Transplanted Mouse Tumors Irradiated in Vivo.
In: R. F. Kallman (ed.), Research in Radiotherapy, Nuclear Science
Series, Rept. 35, National Academy of Sciences-National Research
MARCH 1972
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1972 American Association for Cancer Research.
489
R. F. Kallman, G. L. DeNardo, and M. J. Staseli
9.
10.
11.
12.
13.
14.
15.
16.
17.
490
Council, Publication 888, pp. 30-48. Washington, D.C.: National
Academy of Sciences-National Research Council, 1961.
Kallman, R. F. Discussion. In: V. P. Bond, H. D. Suit, and V.
Marcial (eds.), Conference on Time and Dose Relationships in
Radiation Biology as Applied to Radiotherapy. BNL 50203(C-57),
p. 248. Springfield, Va.: Clearinghouse for Federal Scientific and
Technical Information, 1970.
Kallman, R. F. The Phenomenon of Reoxygenation and Its
Implications for Fractionated Radiotherapy. Radiology, in press.
Kallman, R. F., Silini, G., and Van Putten, L. M. Factors
Influencing the Quantitative Estimation of the in Vivo Survival of
Cells from Solid Tumors. J. Nati. Cancer Inst., 39: 539-549, 1967.
Kety, S. S. Theory and Application of the Exchange of Inert Gas at
the Lungs and Tissues. Pharmacol. Rev., 3: 1-41, 1951.
Lassen, N. A., Lindbjerg, J., and Munck, O. Measurement of
Blood-Flow Through Skeletal Muscle by Intramuscular Injection of
Xenon-133. Lancet, /. 686-689, 1964.
Lassen, N. A., and Trap-Jensen, J. Theoretical Considerations on
Measurement of Capillary Diffusion Capacity in Skeletal Muscle by
the Local Clearance Method. Scand. J. Clin. Lab. Invest., 21:
108-115,1968.
Peterson, H.-I., Appelgren, K. L., Rudenstam, C.-M., and Lewis,
D. H. Studies on the Circulation of Experimental Tumours. I.
Effect of Induced Fibrinolysis and Antifibrinolysis on Capillary
Blood Flow in the Capillary Transport Function of Two
Experimental Tumours in the Mouse. European J. Cancer, 5:
91-97, 1969.
Robert, J., Martin, J., and Burg, C. Action des Rayons X sur la
Vascularization Spécifiquede Tumeurs Isologues de la Souris et du
Rat. Compt. Rend. Soc. Biol., 161: 867-871, 1967.
Robert, J., Martin, J., and Burg, C. Evolution de la Vascularization
18.
19.
20.
21.
22.
23.
24.
25.
d'une Tumeur Isologue Solide de la Souris au Cours de Sa
Croissance. Strahlentherapie, 133: 621-630, 1967.
Rubin, P., and Casare», G. W. Microcirculation of Tumors. I.
Anatomy, Function, and Necrosis. Clin. Radio!., 17: 220-229,
1966.
Rubin, P., and Casarett, G. W. Microcirculation of Tumors. II. The
Supervascularized State of Irradiated Regressing Tumors. Clin.
Radiol., / 7: 346-355, 1966.
Song, C. W., and Levitt, S. H. Vascular Changes in Walker 256
Carcinoma of Rats Following X Irradiation. Radiology, 700.'
397-407,1971.
Thomlinson, R. H. Oxygen Therapy-Biological Considerations. In:
T. J. Deeley and C. A. P. Wood (eds.), Modern Trends in
Radiotherapy, pp. 52-72. New York: Appleton-Century-Crofts,
1967.
Thomlinson, R. H. Reoxygenation as a Function of Tumor Size
and Histopathological Type. In: V. P. Bond, H. D. Suit, and V.
Marcial (eds.), Conference on Time and Dose Relationships in
Radiation Biology as Applied to Radiotherapy. BNL 50203 (C-57),
pp. 242-247. Springfield, Va.: Clearinghouse for Federal Scientific
and Technical Information, 1970.
Thomlinson, R. H., and Gray, L. H. The Histológica!Structure of
Some Human Lung Cancers and the Possible Implications for
Radiotherapy. Brit. J. Cancer, 9: 539-549, 1955.
Van Putten, L. M., and Kallman, R. F. OxygénationStatus of a
Transplantable Tumor during Fractionated Radiation Therapy. J.
Nati. Cancer Inst., 40: 441-451, 1968.
Veall, N., and Malie«,B. L. The Partition of Trace Amounts of
Xenon between Human Blood and Brain Tissues at 37°C.Phys.
Med. Biol., 10: 375-380, 1965.
CANCER RESEARCH VOL. 32
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1972 American Association for Cancer Research.
Blood Flow in Irradiated Mouse Sarcoma as Determined by the
Clearance of Xenon-133
Robert F. Kallman, Gerald L. DeNardo and Mary J. Stasch
Cancer Res 1972;32:483-490.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/32/3/483
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1972 American Association for Cancer Research.