In/ J Radrarron Oncol,~~~~ BIO/ Phr.\ Vol Printed I” the U.S.A. All rights reserved
IX. pp.
S-602 Copyright
0360.3016190 $3.00 + .OO (c, I990 Pergamon Press plc
l Original Contribution
THERMAL
ENHANCEMENT
OF RADIATION-INDUCED HELEN
Radiation
Oncology
Research University
B. STONE,
LEG CONTRACTURE
PH.D.
Laboratory, CED-200, Department of Radiation of California, San Francisco. CA 94143
Oncology,
Early and late damage in the normal tissues of the legs of mice was compared following treatment with radiation alone or radiation followed immediately by hyperthermia. Hyperthermia was given by immersing the hind leg in a water bath at 43.0”, 43.3”, or 43S”C for 1 hr. Damage was assayed by measuring leg contracture at various intervals from 5 to 365 days after treatment. At 5 days after treatment, only hyperthermia-induced contracture was observed. At 10 and 20 days, contracture increased with radiation dose in heated legs, hut little contracture had developed in mice treated with radiation alone. By 45 through 365 days, however, contracture correlated with radiation dose both in mice treated with radiation alone as well as in those treated with radiation and hyperthermia. The greatest differential in the slopes of the dose response curves, suggesting hyperthermic radiosensitization, was seen 20 days after treatment. Nevertheless, at 365 days, contracture was still significantly greater in the mice treated with radiation and hyperthermia (43.5” bath) than in the irradiated controls. Thermal enhancement ratios (TERs) were calculated from LCD50 values (LCD50 = radiation dose that would give a stated level of leg contracture in 50% of the mice). For r3 mm contracture, TERs were 4.1 to 7.9 at 30 days, depending on bath temperature, but only 1 .l to 1.5 at 365 days. For an isoeffect of 27 mm contracture, TERs were 1.9 to 5.3 at 30 days, and 0.8 to 1.8 at 365 days. Thus, contracture was enhanced more at 20 to 30 days after treatment with radiation and hyperthermia than at 120 through 365 days. Radiation damage not only appeared earlier in mice treated with hyperthermia than in those treated with radiation alone, but after the highest temperature tested (43.5” bath), contracture was greater from 5 through 365 days after treatment than in controls treated with radiation alone. Hyperthermia,
Radiation,
Normal tissues,
Early responses,
Late responses,
Thermal
enhancement
tracture develops at the time of the acute skin reactions, at about 18 to 25 days, and may continue to develop for more than a year (19, 34). In patients, the damage may progress for several years after radiotherapy. The purpose of this study was to determine the effect of hyperthermia on the relationship between early and late injury, using leg contracture in mice as an endpoint. Hyperthermia treatments were started immediately after irradiation. For this paper and this assay, we arbitrarily define early damage as that which appears during the first 50 days after treatment, and late damage as that which is present between 120 and 365 days after irradiation, regardless of when that damage developed. Thus, late damage could represent early damage that failed to heal. progressive atrophy and fibrosis or necrosis, or a combination of these. Damage may increase, decrease, or remain constant during the early or late periods. These definitions of early
INTRODUCTION In conventional radiotherapy, normal tissues adjacent to the tumor are included in the treatment field to ensure that all microscopic disease is treated and because of the physical distribution of the radiation dose with external beam or implanted radiation sources. The probability that normal tissue complications will develop months to years after treatment limits the radiation dose that can be administered to a given patient. When radiotherapy is combined with hyperthermia therapy, it is likely that normal tissues will be exposed to both treatments. Tissue injury is manifested earlier after hyperthermia than after irradiation (cf. 4, 5, 13, 14, 20, 27). Leg contracture in mice, for example, peaks at about 5 days after mild hyperthermia, and then heals, leaving little residual contracture after 50 days (35). After irradiation, leg con-
Acknow~ledgements--I am grateful to Judy Sneider, Hiep Nguyen, Virginia Scialanca, and Randolph Cribbs for excellent technical assistance, to Dr. William C. Dewey and Rick P. Harding for helpful discussions, and to Dr. Kathleen Lamborn and Dr. David Heilbron for consultation on statistics. This work was supported by Grants #CA 20344 and #CA 33599 awarded
by the National Cancer Institute, DHEW, and by Grant #MSCOS from the Research Evaluation and Allocation Committee of the University of California, San Francisco, and by Grant PDT 283 awarded by the American Cancer Society. Accepted for publication 9 August 1989.
595
I. J. Radiation
596
Oncology
0 Biology 0 Physics
March
1990. Volume
18, Number
3
and late damage may not be appropriate for other normal tissue endpoints, or for patients, and are not intended to imply mechanisms of development of damage.
METHODS
AND
MATERIALS
C3H/HeJ female mice were 15- to 30-weeks old at the time of treatment. They were given acidified water (pH 2.5) and mouse chow ad lib. There were 10 to 21 mice in each group.
lrrudiation The mice were irradiated without anesthesia in a 13’Cs irradiator at a dose rate of 9.1 to 9.0 Gy/min, as measured by lithium fluoride dosimetry. The field size was 3 X 18 cm. The mice were held in a jig similar to that shown in reference 33, with the right hind leg extending across the beam. The foot and body were shielded.
Mice were held in a modified 50 cc plastic centrifuge tube, with the right hind leg extending through a hole in the tube and into the water bath. The toes were fastened to a bracket on the tube with cyanoacrylate glue (35). The glue was removed with acetone at the end of treatment. The legs were immersed in the water bath within 3 min after irradiation. Water bath temperatures were 43.0”, 43.3”, or 43.5”C (+-0.05”).
Water bath temperatures were monitored with a mercury-in-glass thermometer calibrated against a standard thermometer from the National Bureau of Standards. using the freezing point method. Temperatures were measured in the legs of 14 mice not used for evaluation of response to treatment. Three 24-gauge triple-sensor thermocouple probes were used in each mouse, positioned as shown in Figure 1. Water bath temperatures were 42.5”, 43.3”, and 43.5”. Coolest temperatures were found in the sensors in the deep muscle of the upper leg, and averaged 0.6 to 2.1” below the temperature of the water bath 10 min after immersion. and 0.6 to 1.0” below at 30 min. The largest differentials occurred in the 43.5” bath. Temperatures in the lower leg averaged 0.1 to 0.5” below bath temperatures at 10 min and 0.2 to 0.3” below at 30 min.
This assay has been described in detail previously (34). Briefly. contracture was measured at intervals of 5 to 365 days after treatment by placing the mouse in a jig and extending each hind leg posteriorly against a ruler embedded in the jig. The measurement from the treated leg was subtracted from that of the control leg. Dutu
unalysis
Means and standard deviations were calculated for each treatment group and time point. and compared using
Fig. I. Tracing from a radiograph showing the locations of thermocouples during thermometry. The vertical probe was located in the subcutaneous tissue, and the horizontal probes penetrated the deep muscle.
Student’s t test. Leg contracture at 20 days was compared with that at 120 days, and that at 120 days was compared with that at 365 days, using paired difference t-tests. Data from the linear portions of the dose response curves (i.e., not including data from the shoulder regions or tail regions) were fitted using the least squares method. LCD50 values (LCD50 = the radiation dose that would give the stated amount of leg contracture in 50% of the subjects) were calculated for ~3 mm and 27 mm leg contracture using the logit method ( 19,33,4 1). Thermal enhancement ratios (TERs) were determined from the ratios of LCDSOs in mice treated with radiation alone and those treated with radiation and hyperthermia. Statistical comparisons of TERs were made with z-tests, using pooled estimates of the variances.
RESULTS Leg contracture from radiation alone started to develop rapidly at about 20 days after treatment. when the acute skin reactions were developing (Fig. 2A). After about IO0 days, contracture reached a plateau or continued to progress slowly through 1 year. In contrast, hyperthermia alone resulted in leg contracture that was maximal at about 5 days after treatment. and then healed, leaving little contracture after 45 days (Fig. 2B, bottom curve). When radiation was given immediately before hyperthermia in a 43.5”C bath, the early damage was more severe. healed more slowly, and reached plateau levels above those from hyperthermia alone (Fig. 2B). After 30 and 40 Gy with hyperthermia, contracture developed to >15 mm within 20 days, leading to necrosis in the feet of 30% and 60% of the mice. respectively, by 90 days. Mice developing necrosis were euthanized. Hyperthermia
Thermal
enhancement
still appeared to be displaced to the left. There were insufficient data to determine whether the curves from the R + 43.3” and R + 43.0” groups were different from that from radiation alone. The ratios of the slopes of the dose response curves [(R + H)/(R only)] are shown as a function of time after treatment in Figure 4A and in Table 2. The ratios peaked at
RADIATION + 43.5%
RADIATION ONLY
,*$GY
+H
$%G,+H ,SO
GY
b
rrhnllt U”“UC
I
0
I
I
I
1
1
I
I
1
1
I
I
0 100 200 100 200 300 400 DAYS AFTER TREATMENT
I
1
300
I
400
I
Fig. 2. Leg contracture as a function of time after radiation alone (A), hyperthermia alone (H ONLY in B) or radiation followed immediately by hyperthermia (43.5”C water bath, 1 hr. B). Each point is the mean of a group of 10 to 21 mice. Error bars were omitted for clarity here, but are shown in Figure 3.
treatments at 43.3” and 43.0” resulted in less damage than at 43.5”, both alone and with irradiation. These data were not plotted as in Figure 2B, but are included in subsequent figures. Comparisons of the mean leg contracture at 20, 120, and 365 days in individual treatment groups indicated whether the damage progressed, regressed, or was constant during the interval (Table 1). For example, 30 Gy produced leg contracture that increased from 20 to 120 to 365 days, whereas after hyperthermia alone at all _____r__-_-___, -.. three _... -- water ..____ hath _____ tpmneratllre
20 > 20 < 20
(120 = 365) 20= 120=365 20< 120 20 > (120 = 365) 20> 120 (120 = 365)
* The symbols > and < indicate a significant difference (a I 0.05) and the symbol = indicates the difference was not significant (p > 0.05).
I.
598
J.
Radiation
Oncology
0 Biology
0 Physics
March
1990.
Volume
18. Number
10 DAYS
5 DAYS
3
20 DAYS
20 z d
15
2 0 a
10
0’ 1 /
r
R + 43.3’
1
I
!z 55 !i 0
R ONLY
R ONLY
I
I
I
I
1
I
I
I
I
/
I
I
I
I
I
0 1020304050607080
0 1020304050607060
45 PAYS
120 DAYS
I
I
I
I
I
I
I
I
I
I
0 1020304050607000
365
DAYS
20 : $ E a
I
A + 43.6O
+ 43.3O
15
II+ 43.0”
A
R ONLY
10 -
E
R ONLY
R + 43.3’
55
P
R + 43.0’
s 0
I
I
1
I
I
I
0 1020304050607060
I
/
I
I
1
I
I
I
1
0 1020304050607060
I
1
I
1
I
I
I
I
I
1
I
I
0 10 20304050607060
DOSE, GY was given Fig. 3. Leg contracture as a function of radk iticIn dose at 5 to 365 days af ‘ter treatment. Hyperthermia immediately after irradiation. Water bath temperatures are indicated. Each pomt IS the mean ot a group of 10 to 21 mice. Error bars are standard deviations. Curves on this figure were fitted to the data points by the least squares method or by spline smoothing interpolation using a graphics program (Tellagraf, Integrated Software Systems Corp., San Diego, CA).
from hyperthermia alone, from hyperthermic radiosensitization. itatively identical to radiation of sensitization of a process
radiation alone, and from The latter may be qual-
damage, or may be the result of secondary importance in
cell killing by radiation alone (26). Although mechanisms of thermal radiosensitization are not understood at the molecular level, a number of factors are known to affect the response of cells and tissues to the combined treatments. In cultured cells, the relative contribution of each modality to the total damage depends on the sequence and timing of irradiation and hyperthermia, on the cellular environment (for example, pH, oxygen tension, and nutritional status), on intrinsic cellular factors (for example, thermotolerance), and on the severity of each of the treatments in the combination (1, 3, 6, 8, 29). These factors also influence the response of tumors and normal tissues in viva, where blood flow or vascular collapse and other
physiological factors may also affect heating patterns and tissue response (2, 12, 17, 22, 30, 3 1, 40). Damage from hyperthermia alone correlated with treatment temperature in these studies and was transient, as we have reported previously (Figs. 2B and 3, and ref. 35). Damage from radiation alone depended on dose (Figs. 2A and 3, and ref. 34). Transient hyperthermic radiosensitization, peaking at about 20 days after treatment, was suggested by the steeper slopes of the dose response curves from groups treated with hyperthermia than those treated with radiation alone (Figs. 3 and 4A). This is analogous to the steeper slopes seen in the survival curves of cultured cells treated with radiation and hyperthermia (9). The differences in the slopes were apparently a consequence of the earlier appearance of radiation damage in mice treated with hyperthermia than in those treated with radiation alone, as the slopes from heated groups reached a constant
Thermal
II,,,1 0
enhancement
~ljj~~L/,/I 100 DAYS
200 AFTER
300
400
TREATMENT
Fig. 4. (A) Ratios of the slopes of the dose response curves of groups treated with radiation and hyperthermia to those from groups treated with radiation alone. The dose ranges over which the slopes were calculated are shown in Table 2. (B) Shoulder widths of the curves of Figure 3, estimated by extrapolating the “linear” portions of the curves to the level of contracture in unirradiated mice in the respective groups.
value sooner than those from groups treated with radiation alone (Fig. 3). Although the mechanisms through which this occurs are not known, the radiation probably inhibits healing of the damage caused by hyperthermia, and radiation damage develops as cells attempt to divide to repair the hyperthermia damage. In the present studies, the tissues of the leg were not uniformly heated. Cooling by blood flow and the low heat conductivity of tissue resulted in underheating of the deep tissues, especially in the upper leg. Therefore, the treatments cannot be characterized by a single temperature, unlike the studies of tissue injury in the ears, tail, and feet of mice (15, 17,2 1, 23, 32, 37). The interaction of damage from radiation and hyperthermia probably occurred only within a few millimeters of the surface of the skin, and deeper tissues probably had only radiation-induced damage. Furthermore, the gradual heating of the tissues reduced the amount of hyperthermic radiosensitization from that which might have occurred if the two treatments had been given simultaneously, or if tissues had been brought to treatment temperature more quickly. Clinical practice, however, will probably not involve simultaneous treatments. Relutive enhancement of’earl)’ and late damage Our finding that early normal tissue damage is enhanced by hyperthermia more than is late damage confirms the report of Urano et al. (37). They scored skin reactions in the feet of mice after treatment with radiation
0 H. B.
STONE
599
alone or in combination with hyperthermia (43.5“ water bath, 45 min). When radiation and hyperthermia were given 20 min apart, in either sequence, TERs were 1.3 to 1.6 times greater for damage present at 14 to 35 days than for that at 650 days. When they compared late responses in groups matched for early reactions, they found that late reactions were greater in mice that had been treated with radiation alone than in those treated with radiation and hyperthermia. In all groups, however, late damage was greater than early damage. Isodose comparisons showed that damage was greater in mice treated with hyperthermia and radiation than with radiation alone both for the early and the late endpoints. In a related study (38). they reported that the latency for carcinogenesis was shorter and the incidence increased in mice treated with radiation and hyperthermia than with radiation only. Peck and Gibbs, in contrast, reported greater enhancement of late than early damage for fibrosis in the jejunum of mice, as measured by elastic stiffness (25). They found that fibrosis increased gradually from 1 to 70 weeks in groups treated with and without hyperthermia (44”, 1.5 min, starting 10 min after irradiation). The mild hyperthermia treatment induced no fibrosis by itself. but it hastened the development of radiation induced fibrosis and reduced the shoulders of the dose response curves. TERs were determined at two isoeffect levels of damage, from the widths of the shoulders of the dose response curves, and from the slopes of the dose response curves calculated with a common intercept. With the exception ofthe TERs from the higher isoeffect level, TERs at 2 to 4 weeks after treatment were lower than those at 35 to 70 weeks. The statistical significance of the differences was not reported. Goffinet et uf. (7) scored myelitis in mice following treatment with radiation alone or with hyperthermia given immediately before or after irradiation. They found that hyperthermia shortened the latent period for development of paralysis. TERs calculated from the reported MCDSO doses (MCDSO = radiation dose to produce severe myelitis in 50% of the mice) at 6 months were similar to those at 12 months. The apparent discrepancy in the findings of Goffinet et ~1. and Peck and Gibbs from those of Urano et al. and our results may be due to the absence of an acute phase in the development of jejunal fibrosis and of myelitis: both represent damage to late-responding tissues. Acute skin reactions and the early phase of leg contracture probably develop through depletion of clonogenic cells in the basal layer of the epidermis followed by failure of the normal replacement of the cells lost from the more superficial layers (28). The development of late effects such as fibrosis and myelitis is less well understood, but is probably a consequence of complex interactions between stroma and parenchyma, and involves slowly proliferating cells or nonproliferating cells (10, 39, 43). Law and Ahier (16) found nearly equal enhancement of acute skin reactions and deformity in the ears of mice
600
I. J. Radiation Table
2. Ratios
Oncology
0 Biology 0 Physics
1990. Volume
18,Number 3
of slopes of dose response curves (X + H/X only) and estimates of shoulder width
Dose range Day
March
Gy
Slope ratio
9% conf. lim.
Shoulder
Signif.*
R only
5 IO 15 20 30 45 120
250 365 R + 43.0” 5 IO 15 20 30 45 120 250 365 R + 43.3” 5 IO 15 20 30 45 120 250 365 R + 43.5” 5 10 15 20 30 45 120 250 365
O-60 O-60 O-60 O-60 O-60 30-60 30-52 20-30 1O-30
-
-
-3.7 -3.4 -1.9 -0.8 +21.7 +15.7 +16.5 +I 1.1
-
(-0.4-+ I .6) (1.8-3.2) (3.2-5.2) (4.6-7.2) (2.3-3.1) (0.8-1.6) (1.0-1.6)
O-60 O-60 O-60 20-60 O-60 15-60 15-40 15-30 1O-20
-
O-40 O-40 O-40 15-40 O-40 15-40 I O-20
I.0 2.8 4.7 6.7 4.8 2.2 2.0
(-0.3-$2.9) ( 1.6-4.0) (3.4-6.0) (4.6-8.8) (4.0-5.6) ( I .3-3. I ) (I .3-2.7) -
NS HS HS HS HS HS HS
4.8 5.9 8.1 10.9 7.3 3.9 1.3 0.9 I.1
(I .8-7.X) (3.6-8.2) (5.9-10.3) (8.6-13.2) (6.2-8.4) (2.7-5.1) (0.9-I .7) (0.6-I .2) (0.7-I .5)
HS HS HS HS HS HS NS NS NS
0.6 2.5 4.2 5.9 2.7 1.3 1.3
GY
NS HS HS HS HS NS S -
-1.2
+3.3 +10.6 +4. I +12.7 +13.1
-0.6 +1.7 f3.6 +2.3 +4.5 +9. I
O-40 15-40 O-40 O-40 O-40 1O-30 5-20 5-20 5-20
-10.1
-3.8 -1.1 +1.2 $5.2 i-4.4 +4.7 +3.8
* Significance: HS = p 5 0.01; S = p i 0.05; NS = 11> 0.05.
6 months after treatment with radiation and hyperthermia in either order, whether or not they had been exposed to 19 Gy or 10 X 3.8 Gy 10 months before the combined treatment. Hopewell (1 1) has suggested that deformity may be a consequence of failure of early reactions to heal. in which case equal TERs would be expected.
The probability of late complications in normal tissues is generally dose-limiting for patients treated with radiation alone, although severe mucosal reactions may necessitate temporary interruption of treatment in some patients (42). When radiation is combined with hyperther-
mia treatments. early complications could occur more frequently than with radiation alone. In our experiments, some of the mice in the groups treated with 30 and 40 Gy plus hyperthermia had to be sacrificed by 100 days because of severe damage of the feet (Fig. 2B). With radiation alone, foot loss occurs later, and after higher doses (34). With the more severe hyperthermia treatments, however, both early and late damage were greater than with radiation alone (Figs. 3. 4. and 5). An important question for clinical trials is how well the early reactions predict late reactions. For radiation alone, there is a poor correlation between the two in individual patients (3) and between groups of patients
Thermal
12
T
6
enhancement
AT ISOEFFECT = 7.0 MM
AT ISOEFFECT = 3.0 MM
7 6 $5 4 3 2 1 I
0 0
100
I
200
I
I
300
I
400
0
100
200
300
400
500
DAYS AFTER TREATMENT
Fig. 5. TER as a function of time after treatment. TERs were calculated from LCD50 values for 3 mm and 7 mm contracture.
The earliest points plotted in both panels were from data obtained at 30 days, because of the great uncertainty in estimates of LCD50 at 520 days in mice treated with radiation bars indicate 95%1 confidence limits.
alone. Error
0 H. B.
STONE
601
treated with different schedules (44) or with different types of radiation (45). We would expect a poor correlation with combined radiation and hyperthermia as well. Our data suggest that when hyperthermia treatments are severe, early reactions may become dose-limiting in tissues that show an early response to radiation alone, and late reactions may also be enhanced, but not as much as the early reactions. However, in tissues that show only a late response, such as spinal cord, radiation injury may be expressed earlier when hyperthermia is given, and total damage may be greater than with radiation alone (e.g., 7). Therefore, even though early reactions are absent or are not enhanced, it would be risky to conclude that late complications will not be enhanced by hyperthermia. Normal tissue complications have been reported in clinical trials of hyperthermia, especially when overheating has occurred (e.g., 18, 24, 36). Continuing problems for the use of hyperthermia in the clinic are avoiding cool spots in the tumor and hot spots in the normal tissue, and measuring temperature distributions in both.
REFERENCES 1. Chu, F. C. H.; Glicksman,
2.
3.
4.
5.
6.
7.
8.
9
10
11
A. S.; Nickson, J. J. Late consequences of early skin reactions. Radiol. 94:669-672; 1970. Dewey, W. C.; Freeman. M. L.: Raaphorst, G. P.: Clark, E. P.; Wong, R. S. L.; Highfield. D. P.: Spiro, I. J.; Tomasovic, S. P.: Denman, D. L.: Goss, R. A. Cell biology of hyperthermia and radiation. In: Meyn. R. E., Withers. H. R., eds. Radiation biology in cancer research. New York: Raven Press; 1980:589-62 1. Dudar, T. E.: Jain, R. Differential response of normal and tumor microcirculation to hyperthermia. Cancer Res. 44: 605-612: 1984. Elkon, D.; Fechner, R. E.: Homzie, M. J.; Baker, D. G.: Constable, W. C. Response of mouse kidney to hyperthermia. Arch. Pathol. Lab. Med. 104: 153-l 58; 1980. Fowler, J. F.; Kragt, K.: Ellis, R. E.; Lindop, P. J.: Berry, R. J. The effect of divided doses of 15 MeV electrons on the skin response of mice. Int. J. Radiat. Biol. 9:241-252; 1965. Gerwick, L.; Rottinger, E. Enhancement of mammalian cell sensitivity to hyperthermia by pH alteration. Radiat. Res. 67:508-5 11; 1976. Goffinet, D. R.; Choi. K. Y.: Brown, J. M. The combined effects of hyperthermia and ionizing radiation on the adult mouse spinal cord. Radiat. Res. 72:238-245; 1977. Hahn, G. M. Metabolic aspects of the role of hyperthermia in mammalian cell inactivation and their possible relevance to cancer treatment. Cancer Res. 34:3 117-3 123; 1974. Holahan, E. V.: Highfield. D. P.; Holahan, P. K.; Dewey, W. C. Hyperthermic killing and hyperthermic radiosensitization in Chinese hamster ovary cells: effects of pH and thermal tolerance. Radiat. Res. 97: 108-I 3 1; 1984. Hopewell. J. W. The importance of vascular damage in the development of late radiation effects in normal tissues. In: Meyn. R. E., Withers, H. R., eds. Radiation biology in cancer research. New York: Raven Press; 1980:449-459. Hopewell, J. W. Persistent and late occurring lesions in irradiated feet of rats: their clinical relevance. Brit. J. Radio]. 55:574-578; 1982.
sensitization of 12. Hume, S. P.; Field, S. B. Hyperthermic mouse intestine to damage by X-rays: The effect of sequence and temporal separation of the two treatments. Brit. J. Radiol. 5 1:302-307; 1978. 13. Hume, S. P.; Marigold, J. C. L.; Field, S. B. The effect of local hyperthermia on the small intestine of the mouse. Brit. J. Radio]. 52:657-662; 1979. 14. Hume, S. P.; Myers, R. An unexpected effect of hyperthermia on the expression of X-ray damage in mouse skin. Radiat. Res. 97:186-199; 1984. on the response 15. Law, M. P. Some effects of fractionation ofthe mouse ear to combined heat and X-rays. Radiat. Res. 80:360-368; 1979. 16. Law, M. P.; Ahier, R. G. A long-term effect of prior irradiation on the thermal enhancement of radiation damage in the mouse ear. Int. J. Hyperther. 3:167-175; 1987. 17. Law, M. P.; Ahier, R. G.; Field, S. B. The response of mouse skin to combined hyperthermia and X-rays. Int. J. Radiat. Biol. 32:153-163; 1977. 18. Lindholm, C. E.; Kjellan, E.; Nilsson. P.; Hertzman, S. Microwave-induced hyperthermia and radiotherapy in human superficial tumours: clinical results with a comparative study of combined treatment versus radiotherapy alone. Int. J. Hyperther. 3:393-411; 1987. 19. Masuda, K.: Hunter, N.; Stone, H. B.; Withers, H. R. Leg contracture in mice after single and multifractionated 13’Cs exposure. Int. J. Radiat. Oncol. Biol. Phys. 13: 1209- I2 15; 1987. 20 Milligan, A. J.; Metz, J. A.; Leeper, D. B. Effect ofintestinal hyperthermia in the Chinese hamster. Int. J. Radiat. Oncol. Biol. Phys. 10:259-263: 1984. 21. Morris, C. C.; Meyers, R.; Field, S. B. The response of the rat tail to hyperthermia. Brit. J. Radiol. 50:576-580; 1977. 22. Myers, R.; Field. S. B. The response of the rat tail to combined heat and X-rays. Brit. J. Radio]. 50:581-586; 1977. 23. Myers, R.: Robinson, J. E.; Field, S. B. The relationship between heating time and temperature for inhibition of
602
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
I. J. Radiation
Oncology
0 Biology 0 Physics
growth in baby rat cartilage by combined hyperthermia and x-rays. Int. J. Radiat. Biol. 38:373-382; 1980. Oleson. J. R.; Sim. D. A.; Manning, M. R. Analysis of prognostic variables in hyperthermia treatment of I6 I patients. Int. J. Radiat. Oncol. Biol. Phys. lo:223 l-2239; 1984. Peck. J. W.: Gibbs, F. A. Assay of premorbid murine jejunal fibrosis based on mechanical changes after X irradiation and hyperthermia. Radiat. Res. 112:535-543; 1987. Peck, J. W.; Gibbs. F. A.. Jr.: Dethlefsen, L. A. Localized hyperthermia and X-irradiation of murine jejunum in .silrl. a new method. Int. J. Hyperther. 2:277-298; 1986. Phillips, T. L.: Benak. S.; Ross. G. Ultrastructural and cellular effects of ionizing radiation. Front. Radiat. Ther. Oncol. 6:21-43: 1972. Potten. C. S. The cell kinetic mechanism for radiation-induced cellular depletion of epithelial tissue based on hierarchical differences in radiosensitivity. Int. J. Radiat. Biol. 40:217-225; I98I. Sapareto, S. A.; Hopwood, L. E.; Dewey. W. C.; Raju. M. R.; Gray, J. W. EA‘ects of hyperthermia on survival and progression of Chinese hamster ovary cells. Cancer Res. 38: 393-400: 1978. Song, C. W.; Kang, M. S.: Rhee, J. G.: Levitt, S. H. Vascular damage and delayed cell death in tumours after hyperthermia. Brit. J. Cancer 4 I :309-3 12; 1980. Song, C. W.: Lokshina, A.; Rhee, J. G.: Patten, M.; Levitt. S. Implications of blood flow in hyperthermic treatment of tumors. IEEE Transactions on Boomed. Eng. BME-3 I :916: 1984. Stewart. F. A.: Denekamp. J. Sensitization of mouse skin to X-irradiation by moderate heating. Radiology 123: I95200: 1977. Stone. H. B. Radiotherapy of a mouse mammary carcinoma following treatment with 5-iodo-2’-deoxyuridine. Radiology ll2:719-725; 1974. Stone. H. B. Leg contracture in mice: An assay of normal tissue response. Int. J. Radiat. Oncol. Biol. Phys. IO: lO531061; 1984. Stone, H. B.; Harding, R. P. Reversible injury after mild hyperthermia. Int. J. Radiat. Oncol. Biol. Phys. 12:823827: 1986. Storm. F. K.; Baker, H. W.; Scanlon. E. F.: Plenk, H. P.:
March
37.
38.
39.
40.
41.
42.
43.
44.
45.
1990. Volume
18. Number
3
Meadows. P. M.: Cohen, S. C.; Olson, C. E.: Thompson, J. W.; Khandekar, J. D.: Roe, D.; Nizze, A.: Morton, D. L. Magnetic-induction hyperthermia: results of a 5-year multiinstitutional national cooperative trial in advanced cancer patients. Cancer 55:2677-2688: 1985. Urano, M.: Kenton, L.: Kahn, J. The effect of hyperthermia on the early and late appearing mouse foot reactions and on the radiation carcinogenesis: effect on the early and late appearing reactions. Int. J. Radiat. Oncol. Biol. Phys. 15: 159-166: 1988. Urano. M.; Kenton, L.: Kahn. J. The effect of hyperthermia on the early- and late-appearing mouse foot reactions and on radiation carcinogenesis: Part II. Effect on radiation carcinogenesis (thermal enhancement and oxygen enhancement). Int. J. Radiat. Oncol. Biol. Phys. 16:437-442: 1989. Van der Kogel, A. J. Mechanisms of late radiation injury in the spinal cord. In: Meyn, R. E.. Withers. H. R.. eds. Radiation biology in cancer research. New York: Raven Press: I980:46 l-470. Vaupel. P.; Muller-Kliescr. W.: Gabbert, H. Experimental evidence for a hyperthermia-induced breakdown of tumor blood flow, during normoglycemia. J. Cancer Res. Clin. Oncol. 105:303-304: 1983. Walker. A. M.: Suit, H. D. Assessment of local tumor control using censored tumor response data. Int. J. Radiat. Oncol. Biol. Phys. 9:383-386: 1983. Wang, C. C.; Blitzer, P. H.: Suit. H. D. Twice-a-day radiation therapy for cancer of the head and neck. Cancer 55:2 IOO2104: 1985. Withers. H. R.: Peters. L. J.; Kogelnik. H. D. The patholobiology of late effects of irradiation. In: Meyn. R. E.. Withers. H. R.. eds. Radiation biology in cancer research. New York: Raven Press: 1980:439-448. Withers, H. R.: Thames. H. D., Jr.: Flow. B. L.; Mason, K. A.; Hussey. D. H. The relationship of acute to late skin injury in 2 and 5 fraction/week x-ray therapy. Int. J. Radiat. Oncol. Biol. Phys. 4:595-601: 1978. Withers, H. R.; Thames. H. D.. Jr.; Hussey, D. H.: Flow, B. L.: Mason, K. A. Relative biological effectiveness (RBE) of 50 MV (Be) neutrons for acute and late skin injury. Int. J. Radiat. Oncol. Biol. Phys. 4:603-608; 1978.
IX. pp.
S-602 Copyright
0360.3016190 $3.00 + .OO (c, I990 Pergamon Press plc
l Original Contribution
THERMAL
ENHANCEMENT
OF RADIATION-INDUCED HELEN
Radiation
Oncology
Research University
B. STONE,
LEG CONTRACTURE
PH.D.
Laboratory, CED-200, Department of Radiation of California, San Francisco. CA 94143
Oncology,
Early and late damage in the normal tissues of the legs of mice was compared following treatment with radiation alone or radiation followed immediately by hyperthermia. Hyperthermia was given by immersing the hind leg in a water bath at 43.0”, 43.3”, or 43S”C for 1 hr. Damage was assayed by measuring leg contracture at various intervals from 5 to 365 days after treatment. At 5 days after treatment, only hyperthermia-induced contracture was observed. At 10 and 20 days, contracture increased with radiation dose in heated legs, hut little contracture had developed in mice treated with radiation alone. By 45 through 365 days, however, contracture correlated with radiation dose both in mice treated with radiation alone as well as in those treated with radiation and hyperthermia. The greatest differential in the slopes of the dose response curves, suggesting hyperthermic radiosensitization, was seen 20 days after treatment. Nevertheless, at 365 days, contracture was still significantly greater in the mice treated with radiation and hyperthermia (43.5” bath) than in the irradiated controls. Thermal enhancement ratios (TERs) were calculated from LCD50 values (LCD50 = radiation dose that would give a stated level of leg contracture in 50% of the mice). For r3 mm contracture, TERs were 4.1 to 7.9 at 30 days, depending on bath temperature, but only 1 .l to 1.5 at 365 days. For an isoeffect of 27 mm contracture, TERs were 1.9 to 5.3 at 30 days, and 0.8 to 1.8 at 365 days. Thus, contracture was enhanced more at 20 to 30 days after treatment with radiation and hyperthermia than at 120 through 365 days. Radiation damage not only appeared earlier in mice treated with hyperthermia than in those treated with radiation alone, but after the highest temperature tested (43.5” bath), contracture was greater from 5 through 365 days after treatment than in controls treated with radiation alone. Hyperthermia,
Radiation,
Normal tissues,
Early responses,
Late responses,
Thermal
enhancement
tracture develops at the time of the acute skin reactions, at about 18 to 25 days, and may continue to develop for more than a year (19, 34). In patients, the damage may progress for several years after radiotherapy. The purpose of this study was to determine the effect of hyperthermia on the relationship between early and late injury, using leg contracture in mice as an endpoint. Hyperthermia treatments were started immediately after irradiation. For this paper and this assay, we arbitrarily define early damage as that which appears during the first 50 days after treatment, and late damage as that which is present between 120 and 365 days after irradiation, regardless of when that damage developed. Thus, late damage could represent early damage that failed to heal. progressive atrophy and fibrosis or necrosis, or a combination of these. Damage may increase, decrease, or remain constant during the early or late periods. These definitions of early
INTRODUCTION In conventional radiotherapy, normal tissues adjacent to the tumor are included in the treatment field to ensure that all microscopic disease is treated and because of the physical distribution of the radiation dose with external beam or implanted radiation sources. The probability that normal tissue complications will develop months to years after treatment limits the radiation dose that can be administered to a given patient. When radiotherapy is combined with hyperthermia therapy, it is likely that normal tissues will be exposed to both treatments. Tissue injury is manifested earlier after hyperthermia than after irradiation (cf. 4, 5, 13, 14, 20, 27). Leg contracture in mice, for example, peaks at about 5 days after mild hyperthermia, and then heals, leaving little residual contracture after 50 days (35). After irradiation, leg con-
Acknow~ledgements--I am grateful to Judy Sneider, Hiep Nguyen, Virginia Scialanca, and Randolph Cribbs for excellent technical assistance, to Dr. William C. Dewey and Rick P. Harding for helpful discussions, and to Dr. Kathleen Lamborn and Dr. David Heilbron for consultation on statistics. This work was supported by Grants #CA 20344 and #CA 33599 awarded
by the National Cancer Institute, DHEW, and by Grant #MSCOS from the Research Evaluation and Allocation Committee of the University of California, San Francisco, and by Grant PDT 283 awarded by the American Cancer Society. Accepted for publication 9 August 1989.
595
I. J. Radiation
596
Oncology
0 Biology 0 Physics
March
1990. Volume
18, Number
3
and late damage may not be appropriate for other normal tissue endpoints, or for patients, and are not intended to imply mechanisms of development of damage.
METHODS
AND
MATERIALS
C3H/HeJ female mice were 15- to 30-weeks old at the time of treatment. They were given acidified water (pH 2.5) and mouse chow ad lib. There were 10 to 21 mice in each group.
lrrudiation The mice were irradiated without anesthesia in a 13’Cs irradiator at a dose rate of 9.1 to 9.0 Gy/min, as measured by lithium fluoride dosimetry. The field size was 3 X 18 cm. The mice were held in a jig similar to that shown in reference 33, with the right hind leg extending across the beam. The foot and body were shielded.
Mice were held in a modified 50 cc plastic centrifuge tube, with the right hind leg extending through a hole in the tube and into the water bath. The toes were fastened to a bracket on the tube with cyanoacrylate glue (35). The glue was removed with acetone at the end of treatment. The legs were immersed in the water bath within 3 min after irradiation. Water bath temperatures were 43.0”, 43.3”, or 43.5”C (+-0.05”).
Water bath temperatures were monitored with a mercury-in-glass thermometer calibrated against a standard thermometer from the National Bureau of Standards. using the freezing point method. Temperatures were measured in the legs of 14 mice not used for evaluation of response to treatment. Three 24-gauge triple-sensor thermocouple probes were used in each mouse, positioned as shown in Figure 1. Water bath temperatures were 42.5”, 43.3”, and 43.5”. Coolest temperatures were found in the sensors in the deep muscle of the upper leg, and averaged 0.6 to 2.1” below the temperature of the water bath 10 min after immersion. and 0.6 to 1.0” below at 30 min. The largest differentials occurred in the 43.5” bath. Temperatures in the lower leg averaged 0.1 to 0.5” below bath temperatures at 10 min and 0.2 to 0.3” below at 30 min.
This assay has been described in detail previously (34). Briefly. contracture was measured at intervals of 5 to 365 days after treatment by placing the mouse in a jig and extending each hind leg posteriorly against a ruler embedded in the jig. The measurement from the treated leg was subtracted from that of the control leg. Dutu
unalysis
Means and standard deviations were calculated for each treatment group and time point. and compared using
Fig. I. Tracing from a radiograph showing the locations of thermocouples during thermometry. The vertical probe was located in the subcutaneous tissue, and the horizontal probes penetrated the deep muscle.
Student’s t test. Leg contracture at 20 days was compared with that at 120 days, and that at 120 days was compared with that at 365 days, using paired difference t-tests. Data from the linear portions of the dose response curves (i.e., not including data from the shoulder regions or tail regions) were fitted using the least squares method. LCD50 values (LCD50 = the radiation dose that would give the stated amount of leg contracture in 50% of the subjects) were calculated for ~3 mm and 27 mm leg contracture using the logit method ( 19,33,4 1). Thermal enhancement ratios (TERs) were determined from the ratios of LCDSOs in mice treated with radiation alone and those treated with radiation and hyperthermia. Statistical comparisons of TERs were made with z-tests, using pooled estimates of the variances.
RESULTS Leg contracture from radiation alone started to develop rapidly at about 20 days after treatment. when the acute skin reactions were developing (Fig. 2A). After about IO0 days, contracture reached a plateau or continued to progress slowly through 1 year. In contrast, hyperthermia alone resulted in leg contracture that was maximal at about 5 days after treatment. and then healed, leaving little contracture after 45 days (Fig. 2B, bottom curve). When radiation was given immediately before hyperthermia in a 43.5”C bath, the early damage was more severe. healed more slowly, and reached plateau levels above those from hyperthermia alone (Fig. 2B). After 30 and 40 Gy with hyperthermia, contracture developed to >15 mm within 20 days, leading to necrosis in the feet of 30% and 60% of the mice. respectively, by 90 days. Mice developing necrosis were euthanized. Hyperthermia
Thermal
enhancement
still appeared to be displaced to the left. There were insufficient data to determine whether the curves from the R + 43.3” and R + 43.0” groups were different from that from radiation alone. The ratios of the slopes of the dose response curves [(R + H)/(R only)] are shown as a function of time after treatment in Figure 4A and in Table 2. The ratios peaked at
RADIATION + 43.5%
RADIATION ONLY
,*$GY
+H
$%G,+H ,SO
GY
b
rrhnllt U”“UC
I
0
I
I
I
1
1
I
I
1
1
I
I
0 100 200 100 200 300 400 DAYS AFTER TREATMENT
I
1
300
I
400
I
Fig. 2. Leg contracture as a function of time after radiation alone (A), hyperthermia alone (H ONLY in B) or radiation followed immediately by hyperthermia (43.5”C water bath, 1 hr. B). Each point is the mean of a group of 10 to 21 mice. Error bars were omitted for clarity here, but are shown in Figure 3.
treatments at 43.3” and 43.0” resulted in less damage than at 43.5”, both alone and with irradiation. These data were not plotted as in Figure 2B, but are included in subsequent figures. Comparisons of the mean leg contracture at 20, 120, and 365 days in individual treatment groups indicated whether the damage progressed, regressed, or was constant during the interval (Table 1). For example, 30 Gy produced leg contracture that increased from 20 to 120 to 365 days, whereas after hyperthermia alone at all _____r__-_-___, -.. three _... -- water ..____ hath _____ tpmneratllre
20 > 20 < 20
(120 = 365) 20= 120=365 20< 120 20 > (120 = 365) 20> 120 (120 = 365)
* The symbols > and < indicate a significant difference (a I 0.05) and the symbol = indicates the difference was not significant (p > 0.05).
I.
598
J.
Radiation
Oncology
0 Biology
0 Physics
March
1990.
Volume
18. Number
10 DAYS
5 DAYS
3
20 DAYS
20 z d
15
2 0 a
10
0’ 1 /
r
R + 43.3’
1
I
!z 55 !i 0
R ONLY
R ONLY
I
I
I
I
1
I
I
I
I
/
I
I
I
I
I
0 1020304050607080
0 1020304050607060
45 PAYS
120 DAYS
I
I
I
I
I
I
I
I
I
I
0 1020304050607000
365
DAYS
20 : $ E a
I
A + 43.6O
+ 43.3O
15
II+ 43.0”
A
R ONLY
10 -
E
R ONLY
R + 43.3’
55
P
R + 43.0’
s 0
I
I
1
I
I
I
0 1020304050607060
I
/
I
I
1
I
I
I
1
0 1020304050607060
I
1
I
1
I
I
I
I
I
1
I
I
0 10 20304050607060
DOSE, GY was given Fig. 3. Leg contracture as a function of radk iticIn dose at 5 to 365 days af ‘ter treatment. Hyperthermia immediately after irradiation. Water bath temperatures are indicated. Each pomt IS the mean ot a group of 10 to 21 mice. Error bars are standard deviations. Curves on this figure were fitted to the data points by the least squares method or by spline smoothing interpolation using a graphics program (Tellagraf, Integrated Software Systems Corp., San Diego, CA).
from hyperthermia alone, from hyperthermic radiosensitization. itatively identical to radiation of sensitization of a process
radiation alone, and from The latter may be qual-
damage, or may be the result of secondary importance in
cell killing by radiation alone (26). Although mechanisms of thermal radiosensitization are not understood at the molecular level, a number of factors are known to affect the response of cells and tissues to the combined treatments. In cultured cells, the relative contribution of each modality to the total damage depends on the sequence and timing of irradiation and hyperthermia, on the cellular environment (for example, pH, oxygen tension, and nutritional status), on intrinsic cellular factors (for example, thermotolerance), and on the severity of each of the treatments in the combination (1, 3, 6, 8, 29). These factors also influence the response of tumors and normal tissues in viva, where blood flow or vascular collapse and other
physiological factors may also affect heating patterns and tissue response (2, 12, 17, 22, 30, 3 1, 40). Damage from hyperthermia alone correlated with treatment temperature in these studies and was transient, as we have reported previously (Figs. 2B and 3, and ref. 35). Damage from radiation alone depended on dose (Figs. 2A and 3, and ref. 34). Transient hyperthermic radiosensitization, peaking at about 20 days after treatment, was suggested by the steeper slopes of the dose response curves from groups treated with hyperthermia than those treated with radiation alone (Figs. 3 and 4A). This is analogous to the steeper slopes seen in the survival curves of cultured cells treated with radiation and hyperthermia (9). The differences in the slopes were apparently a consequence of the earlier appearance of radiation damage in mice treated with hyperthermia than in those treated with radiation alone, as the slopes from heated groups reached a constant
Thermal
II,,,1 0
enhancement
~ljj~~L/,/I 100 DAYS
200 AFTER
300
400
TREATMENT
Fig. 4. (A) Ratios of the slopes of the dose response curves of groups treated with radiation and hyperthermia to those from groups treated with radiation alone. The dose ranges over which the slopes were calculated are shown in Table 2. (B) Shoulder widths of the curves of Figure 3, estimated by extrapolating the “linear” portions of the curves to the level of contracture in unirradiated mice in the respective groups.
value sooner than those from groups treated with radiation alone (Fig. 3). Although the mechanisms through which this occurs are not known, the radiation probably inhibits healing of the damage caused by hyperthermia, and radiation damage develops as cells attempt to divide to repair the hyperthermia damage. In the present studies, the tissues of the leg were not uniformly heated. Cooling by blood flow and the low heat conductivity of tissue resulted in underheating of the deep tissues, especially in the upper leg. Therefore, the treatments cannot be characterized by a single temperature, unlike the studies of tissue injury in the ears, tail, and feet of mice (15, 17,2 1, 23, 32, 37). The interaction of damage from radiation and hyperthermia probably occurred only within a few millimeters of the surface of the skin, and deeper tissues probably had only radiation-induced damage. Furthermore, the gradual heating of the tissues reduced the amount of hyperthermic radiosensitization from that which might have occurred if the two treatments had been given simultaneously, or if tissues had been brought to treatment temperature more quickly. Clinical practice, however, will probably not involve simultaneous treatments. Relutive enhancement of’earl)’ and late damage Our finding that early normal tissue damage is enhanced by hyperthermia more than is late damage confirms the report of Urano et al. (37). They scored skin reactions in the feet of mice after treatment with radiation
0 H. B.
STONE
599
alone or in combination with hyperthermia (43.5“ water bath, 45 min). When radiation and hyperthermia were given 20 min apart, in either sequence, TERs were 1.3 to 1.6 times greater for damage present at 14 to 35 days than for that at 650 days. When they compared late responses in groups matched for early reactions, they found that late reactions were greater in mice that had been treated with radiation alone than in those treated with radiation and hyperthermia. In all groups, however, late damage was greater than early damage. Isodose comparisons showed that damage was greater in mice treated with hyperthermia and radiation than with radiation alone both for the early and the late endpoints. In a related study (38). they reported that the latency for carcinogenesis was shorter and the incidence increased in mice treated with radiation and hyperthermia than with radiation only. Peck and Gibbs, in contrast, reported greater enhancement of late than early damage for fibrosis in the jejunum of mice, as measured by elastic stiffness (25). They found that fibrosis increased gradually from 1 to 70 weeks in groups treated with and without hyperthermia (44”, 1.5 min, starting 10 min after irradiation). The mild hyperthermia treatment induced no fibrosis by itself. but it hastened the development of radiation induced fibrosis and reduced the shoulders of the dose response curves. TERs were determined at two isoeffect levels of damage, from the widths of the shoulders of the dose response curves, and from the slopes of the dose response curves calculated with a common intercept. With the exception ofthe TERs from the higher isoeffect level, TERs at 2 to 4 weeks after treatment were lower than those at 35 to 70 weeks. The statistical significance of the differences was not reported. Goffinet et uf. (7) scored myelitis in mice following treatment with radiation alone or with hyperthermia given immediately before or after irradiation. They found that hyperthermia shortened the latent period for development of paralysis. TERs calculated from the reported MCDSO doses (MCDSO = radiation dose to produce severe myelitis in 50% of the mice) at 6 months were similar to those at 12 months. The apparent discrepancy in the findings of Goffinet et ~1. and Peck and Gibbs from those of Urano et al. and our results may be due to the absence of an acute phase in the development of jejunal fibrosis and of myelitis: both represent damage to late-responding tissues. Acute skin reactions and the early phase of leg contracture probably develop through depletion of clonogenic cells in the basal layer of the epidermis followed by failure of the normal replacement of the cells lost from the more superficial layers (28). The development of late effects such as fibrosis and myelitis is less well understood, but is probably a consequence of complex interactions between stroma and parenchyma, and involves slowly proliferating cells or nonproliferating cells (10, 39, 43). Law and Ahier (16) found nearly equal enhancement of acute skin reactions and deformity in the ears of mice
600
I. J. Radiation Table
2. Ratios
Oncology
0 Biology 0 Physics
1990. Volume
18,Number 3
of slopes of dose response curves (X + H/X only) and estimates of shoulder width
Dose range Day
March
Gy
Slope ratio
9% conf. lim.
Shoulder
Signif.*
R only
5 IO 15 20 30 45 120
250 365 R + 43.0” 5 IO 15 20 30 45 120 250 365 R + 43.3” 5 IO 15 20 30 45 120 250 365 R + 43.5” 5 10 15 20 30 45 120 250 365
O-60 O-60 O-60 O-60 O-60 30-60 30-52 20-30 1O-30
-
-
-3.7 -3.4 -1.9 -0.8 +21.7 +15.7 +16.5 +I 1.1
-
(-0.4-+ I .6) (1.8-3.2) (3.2-5.2) (4.6-7.2) (2.3-3.1) (0.8-1.6) (1.0-1.6)
O-60 O-60 O-60 20-60 O-60 15-60 15-40 15-30 1O-20
-
O-40 O-40 O-40 15-40 O-40 15-40 I O-20
I.0 2.8 4.7 6.7 4.8 2.2 2.0
(-0.3-$2.9) ( 1.6-4.0) (3.4-6.0) (4.6-8.8) (4.0-5.6) ( I .3-3. I ) (I .3-2.7) -
NS HS HS HS HS HS HS
4.8 5.9 8.1 10.9 7.3 3.9 1.3 0.9 I.1
(I .8-7.X) (3.6-8.2) (5.9-10.3) (8.6-13.2) (6.2-8.4) (2.7-5.1) (0.9-I .7) (0.6-I .2) (0.7-I .5)
HS HS HS HS HS HS NS NS NS
0.6 2.5 4.2 5.9 2.7 1.3 1.3
GY
NS HS HS HS HS NS S -
-1.2
+3.3 +10.6 +4. I +12.7 +13.1
-0.6 +1.7 f3.6 +2.3 +4.5 +9. I
O-40 15-40 O-40 O-40 O-40 1O-30 5-20 5-20 5-20
-10.1
-3.8 -1.1 +1.2 $5.2 i-4.4 +4.7 +3.8
* Significance: HS = p 5 0.01; S = p i 0.05; NS = 11> 0.05.
6 months after treatment with radiation and hyperthermia in either order, whether or not they had been exposed to 19 Gy or 10 X 3.8 Gy 10 months before the combined treatment. Hopewell (1 1) has suggested that deformity may be a consequence of failure of early reactions to heal. in which case equal TERs would be expected.
The probability of late complications in normal tissues is generally dose-limiting for patients treated with radiation alone, although severe mucosal reactions may necessitate temporary interruption of treatment in some patients (42). When radiation is combined with hyperther-
mia treatments. early complications could occur more frequently than with radiation alone. In our experiments, some of the mice in the groups treated with 30 and 40 Gy plus hyperthermia had to be sacrificed by 100 days because of severe damage of the feet (Fig. 2B). With radiation alone, foot loss occurs later, and after higher doses (34). With the more severe hyperthermia treatments, however, both early and late damage were greater than with radiation alone (Figs. 3. 4. and 5). An important question for clinical trials is how well the early reactions predict late reactions. For radiation alone, there is a poor correlation between the two in individual patients (3) and between groups of patients
Thermal
12
T
6
enhancement
AT ISOEFFECT = 7.0 MM
AT ISOEFFECT = 3.0 MM
7 6 $5 4 3 2 1 I
0 0
100
I
200
I
I
300
I
400
0
100
200
300
400
500
DAYS AFTER TREATMENT
Fig. 5. TER as a function of time after treatment. TERs were calculated from LCD50 values for 3 mm and 7 mm contracture.
The earliest points plotted in both panels were from data obtained at 30 days, because of the great uncertainty in estimates of LCD50 at 520 days in mice treated with radiation bars indicate 95%1 confidence limits.
alone. Error
0 H. B.
STONE
601
treated with different schedules (44) or with different types of radiation (45). We would expect a poor correlation with combined radiation and hyperthermia as well. Our data suggest that when hyperthermia treatments are severe, early reactions may become dose-limiting in tissues that show an early response to radiation alone, and late reactions may also be enhanced, but not as much as the early reactions. However, in tissues that show only a late response, such as spinal cord, radiation injury may be expressed earlier when hyperthermia is given, and total damage may be greater than with radiation alone (e.g., 7). Therefore, even though early reactions are absent or are not enhanced, it would be risky to conclude that late complications will not be enhanced by hyperthermia. Normal tissue complications have been reported in clinical trials of hyperthermia, especially when overheating has occurred (e.g., 18, 24, 36). Continuing problems for the use of hyperthermia in the clinic are avoiding cool spots in the tumor and hot spots in the normal tissue, and measuring temperature distributions in both.
REFERENCES 1. Chu, F. C. H.; Glicksman,
2.
3.
4.
5.
6.
7.
8.
9
10
11
A. S.; Nickson, J. J. Late consequences of early skin reactions. Radiol. 94:669-672; 1970. Dewey, W. C.; Freeman. M. L.: Raaphorst, G. P.: Clark, E. P.; Wong, R. S. L.; Highfield. D. P.: Spiro, I. J.; Tomasovic, S. P.: Denman, D. L.: Goss, R. A. Cell biology of hyperthermia and radiation. In: Meyn. R. E., Withers. H. R., eds. Radiation biology in cancer research. New York: Raven Press; 1980:589-62 1. Dudar, T. E.: Jain, R. Differential response of normal and tumor microcirculation to hyperthermia. Cancer Res. 44: 605-612: 1984. Elkon, D.; Fechner, R. E.: Homzie, M. J.; Baker, D. G.: Constable, W. C. Response of mouse kidney to hyperthermia. Arch. Pathol. Lab. Med. 104: 153-l 58; 1980. Fowler, J. F.; Kragt, K.: Ellis, R. E.; Lindop, P. J.: Berry, R. J. The effect of divided doses of 15 MeV electrons on the skin response of mice. Int. J. Radiat. Biol. 9:241-252; 1965. Gerwick, L.; Rottinger, E. Enhancement of mammalian cell sensitivity to hyperthermia by pH alteration. Radiat. Res. 67:508-5 11; 1976. Goffinet, D. R.; Choi. K. Y.: Brown, J. M. The combined effects of hyperthermia and ionizing radiation on the adult mouse spinal cord. Radiat. Res. 72:238-245; 1977. Hahn, G. M. Metabolic aspects of the role of hyperthermia in mammalian cell inactivation and their possible relevance to cancer treatment. Cancer Res. 34:3 117-3 123; 1974. Holahan, E. V.: Highfield. D. P.; Holahan, P. K.; Dewey, W. C. Hyperthermic killing and hyperthermic radiosensitization in Chinese hamster ovary cells: effects of pH and thermal tolerance. Radiat. Res. 97: 108-I 3 1; 1984. Hopewell. J. W. The importance of vascular damage in the development of late radiation effects in normal tissues. In: Meyn. R. E., Withers, H. R., eds. Radiation biology in cancer research. New York: Raven Press; 1980:449-459. Hopewell, J. W. Persistent and late occurring lesions in irradiated feet of rats: their clinical relevance. Brit. J. Radio]. 55:574-578; 1982.
sensitization of 12. Hume, S. P.; Field, S. B. Hyperthermic mouse intestine to damage by X-rays: The effect of sequence and temporal separation of the two treatments. Brit. J. Radiol. 5 1:302-307; 1978. 13. Hume, S. P.; Marigold, J. C. L.; Field, S. B. The effect of local hyperthermia on the small intestine of the mouse. Brit. J. Radio]. 52:657-662; 1979. 14. Hume, S. P.; Myers, R. An unexpected effect of hyperthermia on the expression of X-ray damage in mouse skin. Radiat. Res. 97:186-199; 1984. on the response 15. Law, M. P. Some effects of fractionation ofthe mouse ear to combined heat and X-rays. Radiat. Res. 80:360-368; 1979. 16. Law, M. P.; Ahier, R. G. A long-term effect of prior irradiation on the thermal enhancement of radiation damage in the mouse ear. Int. J. Hyperther. 3:167-175; 1987. 17. Law, M. P.; Ahier, R. G.; Field, S. B. The response of mouse skin to combined hyperthermia and X-rays. Int. J. Radiat. Biol. 32:153-163; 1977. 18. Lindholm, C. E.; Kjellan, E.; Nilsson. P.; Hertzman, S. Microwave-induced hyperthermia and radiotherapy in human superficial tumours: clinical results with a comparative study of combined treatment versus radiotherapy alone. Int. J. Hyperther. 3:393-411; 1987. 19. Masuda, K.: Hunter, N.; Stone, H. B.; Withers, H. R. Leg contracture in mice after single and multifractionated 13’Cs exposure. Int. J. Radiat. Oncol. Biol. Phys. 13: 1209- I2 15; 1987. 20 Milligan, A. J.; Metz, J. A.; Leeper, D. B. Effect ofintestinal hyperthermia in the Chinese hamster. Int. J. Radiat. Oncol. Biol. Phys. 10:259-263: 1984. 21. Morris, C. C.; Meyers, R.; Field, S. B. The response of the rat tail to hyperthermia. Brit. J. Radiol. 50:576-580; 1977. 22. Myers, R.; Field. S. B. The response of the rat tail to combined heat and X-rays. Brit. J. Radio]. 50:581-586; 1977. 23. Myers, R.: Robinson, J. E.; Field, S. B. The relationship between heating time and temperature for inhibition of
602
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
I. J. Radiation
Oncology
0 Biology 0 Physics
growth in baby rat cartilage by combined hyperthermia and x-rays. Int. J. Radiat. Biol. 38:373-382; 1980. Oleson. J. R.; Sim. D. A.; Manning, M. R. Analysis of prognostic variables in hyperthermia treatment of I6 I patients. Int. J. Radiat. Oncol. Biol. Phys. lo:223 l-2239; 1984. Peck. J. W.: Gibbs, F. A. Assay of premorbid murine jejunal fibrosis based on mechanical changes after X irradiation and hyperthermia. Radiat. Res. 112:535-543; 1987. Peck, J. W.; Gibbs. F. A.. Jr.: Dethlefsen, L. A. Localized hyperthermia and X-irradiation of murine jejunum in .silrl. a new method. Int. J. Hyperther. 2:277-298; 1986. Phillips, T. L.: Benak. S.; Ross. G. Ultrastructural and cellular effects of ionizing radiation. Front. Radiat. Ther. Oncol. 6:21-43: 1972. Potten. C. S. The cell kinetic mechanism for radiation-induced cellular depletion of epithelial tissue based on hierarchical differences in radiosensitivity. Int. J. Radiat. Biol. 40:217-225; I98I. Sapareto, S. A.; Hopwood, L. E.; Dewey. W. C.; Raju. M. R.; Gray, J. W. EA‘ects of hyperthermia on survival and progression of Chinese hamster ovary cells. Cancer Res. 38: 393-400: 1978. Song, C. W.; Kang, M. S.: Rhee, J. G.: Levitt, S. H. Vascular damage and delayed cell death in tumours after hyperthermia. Brit. J. Cancer 4 I :309-3 12; 1980. Song, C. W.: Lokshina, A.; Rhee, J. G.: Patten, M.; Levitt. S. Implications of blood flow in hyperthermic treatment of tumors. IEEE Transactions on Boomed. Eng. BME-3 I :916: 1984. Stewart. F. A.: Denekamp. J. Sensitization of mouse skin to X-irradiation by moderate heating. Radiology 123: I95200: 1977. Stone. H. B. Radiotherapy of a mouse mammary carcinoma following treatment with 5-iodo-2’-deoxyuridine. Radiology ll2:719-725; 1974. Stone. H. B. Leg contracture in mice: An assay of normal tissue response. Int. J. Radiat. Oncol. Biol. Phys. IO: lO531061; 1984. Stone, H. B.; Harding, R. P. Reversible injury after mild hyperthermia. Int. J. Radiat. Oncol. Biol. Phys. 12:823827: 1986. Storm. F. K.; Baker, H. W.; Scanlon. E. F.: Plenk, H. P.:
March
37.
38.
39.
40.
41.
42.
43.
44.
45.
1990. Volume
18. Number
3
Meadows. P. M.: Cohen, S. C.; Olson, C. E.: Thompson, J. W.; Khandekar, J. D.: Roe, D.; Nizze, A.: Morton, D. L. Magnetic-induction hyperthermia: results of a 5-year multiinstitutional national cooperative trial in advanced cancer patients. Cancer 55:2677-2688: 1985. Urano, M.: Kenton, L.: Kahn, J. The effect of hyperthermia on the early and late appearing mouse foot reactions and on the radiation carcinogenesis: effect on the early and late appearing reactions. Int. J. Radiat. Oncol. Biol. Phys. 15: 159-166: 1988. Urano. M.; Kenton, L.: Kahn. J. The effect of hyperthermia on the early- and late-appearing mouse foot reactions and on radiation carcinogenesis: Part II. Effect on radiation carcinogenesis (thermal enhancement and oxygen enhancement). Int. J. Radiat. Oncol. Biol. Phys. 16:437-442: 1989. Van der Kogel, A. J. Mechanisms of late radiation injury in the spinal cord. In: Meyn, R. E.. Withers. H. R.. eds. Radiation biology in cancer research. New York: Raven Press: I980:46 l-470. Vaupel. P.; Muller-Kliescr. W.: Gabbert, H. Experimental evidence for a hyperthermia-induced breakdown of tumor blood flow, during normoglycemia. J. Cancer Res. Clin. Oncol. 105:303-304: 1983. Walker. A. M.: Suit, H. D. Assessment of local tumor control using censored tumor response data. Int. J. Radiat. Oncol. Biol. Phys. 9:383-386: 1983. Wang, C. C.; Blitzer, P. H.: Suit. H. D. Twice-a-day radiation therapy for cancer of the head and neck. Cancer 55:2 IOO2104: 1985. Withers. H. R.: Peters. L. J.; Kogelnik. H. D. The patholobiology of late effects of irradiation. In: Meyn. R. E.. Withers. H. R.. eds. Radiation biology in cancer research. New York: Raven Press: 1980:439-448. Withers, H. R.: Thames. H. D., Jr.: Flow. B. L.; Mason, K. A.; Hussey. D. H. The relationship of acute to late skin injury in 2 and 5 fraction/week x-ray therapy. Int. J. Radiat. Oncol. Biol. Phys. 4:595-601: 1978. Withers, H. R.; Thames. H. D.. Jr.; Hussey, D. H.: Flow, B. L.: Mason, K. A. Relative biological effectiveness (RBE) of 50 MV (Be) neutrons for acute and late skin injury. Int. J. Radiat. Oncol. Biol. Phys. 4:603-608; 1978.
