Radiotherapy and Oncology, 21 (1991) 171-178
© 1991 Elsevier Science Publishers B.V. 0167-8140/91/$03.50
171
RADION 00857
Fractionated thermochemotherapy in vivo of a C3H mouse mammary carcinoma O d d R. M o n g e 1,2 a n d E i n a r K . R o f s t a d 2 Departments of ~Medical Oncology and Radiotherapy, and 2Biophysics, Institute for Cancer Research, The Norwegian Radium Hospital, and The Norwegian Cancer Society, Oslo, Norway
(Received 24 August 1990, revision received 26 March 1991, accepted 16 April 1991)
Key words: Hyperthermia; Thermotolerance; Drug resistance; Fractionated thermochemotherapy; Cyclophosphamide
Summary The interaction between fractionated heat treatment and fractionated drug treatment with cyclophosphamide (CTX) was investigated in a transplantable C3H mouse mammary carcinoma inoculated into the hind leg of C3D2F 1/Born mice. A tumour core temperature of 43.5 _+ 0.1 °C was achieved by immersing the tumour-bearing leg into a water bath thermostatically maintained at 43.7 + 0.1 ° C. CTX was administered i.p. using the maximum tolerated dose (MTD) (LD 1 ~o) for single fraction treatment (100 mg/kg) as the maximum fraction dose. For combined treatment CTX was given 15 min prior to heating. The endpoint was the time to reach a tumour volume of 5 times the volume at first treatment. Specific growth delay was used as effect parameter. In dose-effect experiments using total treatment time at 43.5 °C as dose parameter, drug enhancement ratio (DER) was determined as the ratio of the slope of the dose-effect curve for MTD of CTX plus heat to the slope of the curve for heat alone. In dose-effect experiments using total CTX dose as dose parameter, thermal enhancement ratio (TER) was determined as the ratio of the slope of the dose-effect curve for CTX plus 43.5 °C for 30 rain to the slope of the curve for CTX alone. The regimens investigated were single fraction treatment and 3 and 5 fractions with time intervals of 3 and 5 days. For single fraction treatment DER was 1.4 + 0.1 and TER 2.3 + 0.2. The drug sensitization of the effect of heat treatment tended to increase with increasing number of fractions. The thermal sensitization of the effect of drug treatment was about the same for fractionated and single fraction treatment. Both the drug sensitization and the thermal sensitization were larger for a treatment interval of 3 days than of 5 days, i.e. largest with the fraction interval at which the thermotolerance is known to be largest in this turnout. The data indicate a potential for improved therapeutic ratio by fractionation of thermochemotherapy.
Introduction Thermotolerance, i.e. heat-induced transient resistance to subsequent heat exposure, is one major obstacle for successful treatment of malignant tumours by fractionated hyperthermia [ 2,12,14,16,18 ]. Likewise, intrinsic and acquired tumour drug resistance are major obstacles for the success o f drug treatment o f malignant disease [3,11,15,20]. Thermotolerance and induced drug resistance are possibly interrelated phenomena, both representing biological adaptation to external environmental stress [2,4,8,18 ]. There are some indications that the problems of thermotolerance and drug
resistance can be reduced by the combined use of hyperthermia and c h e m o t h e r a p y - t h e r m o c h e m o t h e r a p y [ 1,2,4,8,9,17,18]. However, there are no reports on the drug sensitization o f the effect o f heat treatment or on the thermal sensitization o f the effect of drug treatment for fractionated thermochemotherapy of malignant tumours. We have previously reported that the drug sensitization o f the effect of heat treatment at 43.5 °C was 3 - 4 times larger in thermotolerant tumours than in previously untreated tumours both for cyclophosphamide (CTX) and mitomycin C in a C 3 H mouse m a m m a r y carcinoma, i.e. the expression of thermotolerance was to a large extent reduced, but not com-
Address for correspondence: Odd R. Monge, M.D., The Norwegian Radium Hospital, Department of Medical Oncology and Radiotherapy,
N-0310 Oslo 3, Norway
172 pletely overcome, by the thermochemotherapy [ 9]. Also the thermal sensitization of the effect of CTX tended to be larger in thermotolerant than in previously untreated tumours [9]. The effect of fractionated thermochemotherapy of the same tumour with CTX is reported here. The aim of the study was to determine drug enhancement ratio (DER) and thermal enhancement ratio (TER) as functions of the number of fractions and the time between fractions. Materials and methods Animal tumour system
Female C3D2F1/Bom mice weighing between 19 and 24 g were inoculated subcutaneously on the dorsal side of the right hind foot with a spontaneously arisen C3H mouse mammary carcinoma as previously described [ 10 ]. Tumour volume was determined using the formula n/6 x D 1 x D2 x D3. The three orthogonal diameters (D1-3) were measured with a slide gauge. Tumours
were randomly allocated to treatment when a volume between 130 and 200 mm 3 was reached. Treatment
During heat treatment the unanaesthetized mice were placed in lucite plastic jigs with the tumour-bearing leg loosely fixed with tape proximal to the tumour without impairing the blood flow. Local hyperthermia with a tumour core temperature of 43.5 + 0.1 °C was given by immersion of the tumour-bearing legs into a circulating water bath thermostatically maintained at 43.7 + 0.1 °C [ 10]. The body temperature of the mice was below 39 ° C. Cyclophosphamide (L~i~tke Farmos, Farmitalia Carlo Erba) was dissolved in sterile water and administered i.p. with 0.02 ml/g body weight. The maximum dose per fraction was 100mg/kg corresponding to maximum tolerated dose (MTD), i.e. the dose that kills approximately 1 ~o of the mice within 150 days after single fraction treatment [10]. Mice with tumours that did not regrow, were likewise observed at
TABLE I Heat alone at 43.5 °C: results after single fraction and fractionated treatment. Heating time per fraction (min at 43.5 ° C)
No. of fractions/ fraction interval (days)
No. of cured/no, of mice
D T + SE (days)
T G T + SE (days)
0
0*
0/95
1.8 _+ 0.1
4.0 + 0.1
10 30 50 70 90
1 1 I 1 1
0/32 0/54 0/46 0/44 0/42
1.8 1.9 2.1 2.1 2.1
+ + ± + ±
25 30 40 50 60 70
3/3 3/3 3/3 3/3 3/3 3/3
0/15 0/16 0/12 0/15 0/5 0/15
2.4 2.3 2,5 2,4 2.3 2.6
± 0.1 +_ 0.1 ± 0.1 + 0.I ± 0.1 ± 0.2
0.1 0.1 0.I 0.1 0.1
4.7 7.9 11.1 14.4 17.6
S G D ± SE
+ 0.2 _+ 0.3 +_ 0.3 + 0.5 + 0.9
0.4 2.1 3.3 5.2 6.4
+ + + + +
9.8 + 0.8
2.6 3.4 5.6 6.1 6.6 8.3
+ 0.4 + 0.3 + 0.6 _+ 0.5 ± 1.0 + 0.8
11.6 + 0.4
17.4 18.1 18.6 24.6
+ + + +
1.0 1.0 1.9 1.4
0.1 0,2 0,2 0.3 0.4
30
5/3
0/15
40 50
5/3 5/3
0/16 0/17
3.0 + 0.2 2.7 ± 0.2 2.8 ± 0.2
14.8 + 1.0 20.6 + 0.8 26.1 + 1.1
3.9 + 0.5 6.5 + 0.5 8.4 + 0.6 5,5 + 0.7 8.1 + 0.7 9.8 + 0.8
30
3/5
0/12
2.5 ± 0.1
40 50
3/5 3/5
0/16 0/10
2.3 + 0.4 2.6 ± 0.2
17.8 + 1.8 22.3 + 1.3 28.1 _+ 1.6
35 37.5 40 42.5
5/5 5/5 5/5 5/5
0/15 0/11 0/17 0/18
2.7 2.3 2.5 2.5
29.7 31.7 32.7 35.0
* Untreated controls.
+ ± + ±
0.1 0.2 0.1 0.2
+ + + ±
0.8 2.4 1.0 2.1
9.7 12.9 12.3 13.1
_+ 0.7 + 1.4 + 0.8 ± 1.1
173 least 150 days after the first treatment before they were considered cured of their tumour. For combined treatment CTX was given 15 rain prior to heating, i.e. the sequencing that yielded the largest tumour response after single fraction treatment [10].
Evaluation of response After treatment, tumour volume was measured 3 to 4 times a week. The time to reach a tumour volume of 5 times the volume at first treatment (day zero) was used as endpoint and recorded as tumour growth time (TGT). The growth rate during regrowth was determined as tumour volume doubling time (DT) for each tumour. Due to differences in mean DT between treatment groups, the data were analyzed using specific growth delay (SGD) as effect parameter: SGD
= TGTtrCat~d - T G T u n t r ¢ ~ t e d DTtreated
The investigated fractionation regimens were single fraction treatment and 3 and 5 fractions with time intervals of 3 and 5 days. For each fractionation regimen, dose-effect experiments were performed using two
different dose parameters, i.e. total heating time at 43.5 °C and total dose of CTX. For determination of DER, dose-effect curves were constructed for heat alone and for M T D of CTX given 15 rain prior to heat treatment. For determination of TER, dose-effect curves were constructed for CTX alone and for CTX given 15 rain prior to 30 rain heat treatment at 43.5 °C. Enhancement ratios were calculated as the ratios of slopes of dose-effect curves. For each fractionation regimen, the total dose range possible to investigate with the present animal tumour system was exploited. Regrowth to the endpoint before administration of the last treatment fraction prevented the use of lower doses than the lowest dose levels used here. The upper dose levels were limited by either treatment toxicity or a high frequency of controlled tumours. Although the data presented in Tables I - I V and Figs. 1-3 are mean values, the linear regression analysis of dose-effect curves was based on the individual SGDs, i.e. the individual DTs and TGTs, of all contributing tumours, not the mean values. Significance levels of enhancement ratios were determined by a two-tailed t-test and indicate at which level the ratios were different from 1. The experimental protocol was approved by the Experimental Animal Board.
TABLE I! Cyclophosphamide 100 mg/kg 15 min prior to each heating at 43.5 °C: results after single fraction and fractionated treatment. Heating time per fraction (min at 43.5 ° C)
No. of fractions/ fraction interval (days)
10 30 50 70 90
1
1 1 1 1
10 20 30
DT ± SE (days)
T G T i SE (days)
S G D ± SE
0/16 0/16 0/16 0/16 0/14
2.9 1.9 2.0 2.1 2.3
10.9 20.4 23.8 26.4 28.4
2.4 8.7 10.0 10.8 10.4
3/3 3/3 3/3
6/16 3/14 5/13
2.7 + 0.1 2.5 ± 0.1 2.5 + 0.2
38.1 ± 2.6 47.6 _+ 4.0 47.2 _+ 2.0
13.2 _+ 1.5 17.5 ± 1.0 17.6 ± 1.4
10 20 30
5/3* 5/3* 5/3*
1/16 3/15 1/16
2.9 ± 0.1 2.6 ± 0.2 2.5 ± 0.1
29.2 ± 1.2 42.7 ± 2.6 47.1 ± 2.5
8.9 ± 0.5 15.5 + 1.2 17.8 + 1.2
10 20 30
3/5 3/5 3/5
1/15 4/16 0/13
2.4 ± 0.1 2.5 ± 0.1 2.5 +_ 0.1
32.2 ± 1.6 44.7 ± 1.3 51.8 ± 1.7
11.7 ± 0.5 16.2 ± 0.7 19.2 _+ 0.9
5 7.5 10 15
5/5 5/5 5/5 5/5
0/16 4/16 8/15 10/17
3.3 2.6 2.7 2.8
31.8 38.2 48.9 56.4
8.6 13.7 16.8 18.7
* CTX 60 mg/kg.
No. of cured/no, of mice
+_ 0.1 ± 0.1 _+ 0.1 ± 0.1 + 0.2
+ + ± +
0.2 0.1 0.1 0.1
_+ 0.4 + 0.5 _+ 0.6 _+ 0.9 ± 1.2
± ± + ±
0.8 0.9 3.4 3.0
+ 0.2 i 0.4 ± 0.5 _+ 0.6 ± 0.6
± ± ± ±
0.4 0.7 1.4 1.5
174 T A B L E III Cyclophosphamide (CTX) alone: results after single fraction and fractionated treatment. CTX dose per fraction (mg/kg)
No. of fractions/ fraction interval (days)
No. of cured/no, of mice
D T _+ SE (days)
T G T + SE (days)
S G D + SE
33 67 100
1 1 1
0/16 0/16 0/16
2.3 +_ 0.1 2.5 + 0.2 2.8 + 0.2
5.3 + 0.2 9.4 + 0.6 11.6 + 0.7
0.6 +_ 0.1 2.2 + 0.3 2.9 + 0.4
50 75 100
3/3 3/3 3/3
0/16 1/16 0/19
3.6 +_ 0.2 2.7 + 0.1 2.7 +_ 0.2
10.8 + 0.6 17.0 + 0.6 25.2 + 1.8
2.0 + 0.2 4.9 + 0.4 8.5 + 0.8
60 80 100
5/3 5/3 5/3
1/16 1/16 7/25
3.7 + 0.2 3.4 + 0.2 3.2 +_ 0.2
19.3 + 0.5 25.2 + 1.5 30.3 + 1.9
4.3 + 0.3 6.7 + 0.7 8.8 + 0.8
67 83 100
3/5 3/5 3/5
0/15 0/15 0/15
3.1 + 0.3 3.3 + 0.2 2.5 + 0.1
16.9 + 0.6 19.4 + 0.5 26.8 ± 0.9
4.6 + 0.4 4.9 + 0.4 9.3 + 0.4
67 83 100
5/5 5/5 5/5
0/14 0/15 0/15
4.5 +_ 0.2 3.9 +_ 0.5 3.3 + 0.2
25.6 + 0.7 29.2 + 1.5 31.5 + 1.0
5.0 + 0.4 7.8 + 1.0 8.9 + 0.8
TABLE IV Cyclophosphamide 15 min prior to each 30 min heating at 43.5 °C: results after single fraction and fractionated treatment. CTX dose per fraction (mg/kg)
No. of fractions/ fraction interval (days)
No. of cured/no, of mice
D T + SE (days)
T G T + SE (days)
S G D +_ SE
33 67 100
1 1 1
0/16 0/16 0/16
2.1 + 0.1 1.9 + 0.1 1.9 +_ 0.1
11.0 + 0.6 16.0 + 0.5 20.4 _+ 0.5
3.4 + 0.4 6.5 + 0.3 8.7 + 0.4
33 67 100
3/3 3/3 3/3
0/16 2/16 5/13
2.0 + 0.1 2.0 + 0.1 2.5 + 0.2
20.6 + 0.7 37.0 + 1.7 47.2 + 2.0
8.4 + 0.5 16.7 + 0.9 17.6 + 1.4
33 50 60
5/3 5/3 5/3
0/16 1/16 1/16
2.3 + 0.1 2.3 + 0.1 2.5 + 0.1
29.6 + 0.7 37.5 +_ 0.9 47.1 + 2.5
11.1 + 0.4 14.9 + 0.7 17.8 + 1.2
33 67 100
3/5 3/5 3/5
0/15 1/16 0/13
2.3 + 0.1 2.5 +_ 0.1 2.5 + 0.1
27.7 + 1.0 38.0 + 1.0 51.8 + 1.7
10.5 + 0.4 13.8 + 0.6 19.2 _+ 0.9
17 33 50 67
5/5 5/5 5/5 5/5
0/ 5 0/16 0/15 0/16
3.4 2.6 3.0 3.0
29.5 39.6 49.8 59.2
8.1 14.2 15.5 18.5
+ 0.4 + 0.1 + 0.1 +_ 0.1
+ + + +
2.0 1.4 2.0 1.8
+ + + +
1.3 0.9 0.6 0.8
175
l
'
'
i
!
I
I
[
r
B
r
B c T x
/
'
i
i
~mll~ ÷~ot
•
CTX÷t,3S°C {30n',,n]
5/3
'0
50
i 100
i 150
i
i
200
250
i 50 D~
0
5O foe 15o Treotn~n( tn~' at 43.5"C (rain)
300 CTX ~
I00
i I 1~10 200
i | 1 CTXI~r~/l~ ÷ h*ot
I 250 b
(mg/kg)
Fig. 1. Dose-effect curves for single fraction treatment. (A) Treatment time at 43.5 °C as dose parameter: heat alone ( O ) and M T D of CTX 15 min prior to heating (O). ( B ) C T X dose as dose parameter: CTX alone ( A ) and CTX 15 min prior to 43.5 °C for 30 rain (&). Points and bars indicate mean S G D + SE.
15
Results
~5
The primary results are presented in Tables I-IV. Lack of tumour regrowth within 150 days, i.e. mice cured of their tumour, was observed mainly after combined TABLE V Slopes and standard deviations of dose-effect curves for single fraction and fractionated treatment with either heat alone at 43.5 ° C or cyclophosphamide (CTX) 100 mg/kg 15 rain prior to each heating at 43.5 °C. Drug enhancement ratios (DER) were calculated as ratios of slopes of dose effect curves. No. of fractions/ fraction interval (days)
Treatment
Slope _+ SD
DER + SE
1
CTX + heat Heat
0.100 _+ 0.008 0.071 + 0.002
1.4 -+ 0.1 b
3/3
CTX + heat Heat
0.115 + 0.017 0.040 + 0.002
2.9 + 0.5 b
5/3
CTX + heat a Heat
0.094 _+ 0.008 0,032 _+ 0.002
2.9 _+ 0.38
3/5
CTX + heat Heat
0.109 _+ 0.009 0,065 +_ 0.004
1.7 _+ 0.28
5/5
CTX + heat Heat
0,147 _+ 0.018 0,062 +_ 0.003
2.4 _+ 0.38
a CTX dose 60 mg/kg. b p ,~ 0.001. All slopes are significantly different from zero.
0
I
50
100
150
i
200 250 50 100 Total trgatrnent time at 43,5=C (rain)
I
150
I 200
I 250
Fig. 2. Dose-effect curves for fractionated treatment using total treatment time at 43.5 °C as dose parameter: beat alone (open symbols), M T D of CTX 15 min prior to heating (closed symbols). Upper panel: 3 days fraction interval. (A)3 fractions; (B)5 fractions (60 mg/kg of CTX). Lower panel: 5 days fraction interval. (C) 3 fractions; ( D ) 5 fractions. Points and bars indicate mean S G D + SE.
fractionated treatment. After 5 fractions/3 days interval (Table II), M T D of CTX and heating times of 10 min and more resulted in controlled tumours in the vast majority of the mice (data not shown). For this reason, a lower dose of CTX was used for combined treatment with that fractionation regimen (Table II). Controlled tumours were omitted from the calculation of mean SGD. The two sets of dose-effect curves for single fraction treatment are shown in Fig. 1 and the two sets of doseeffect curves for fractionated treatment are shown in Figs. 2 and 3. The slopes of the dose-effect curves in Fig. 1A and Fig. 2 are tabulated in Table V together with the resulting DERs. For fractionated heat treatment alone a dose range near zero was not possible to investigate. The regression analyses showed that none of the intercepts of the dose-effect curves for fractionated treat-
176 20
i
20
i
i
,
T A B L E VI Slopes and standard deviations of dose-effect curves for single fraction and fractionated treatment with either cyclophosphamide (CTX) alone or CTX 15 rain prior to each 30 min heat treatment at 43.5 ° C. Thermal enhancement ratios (TER) were calculated as the ratios of slopes of dose-effect curves.
c1x
rg3
313
!
!
,,oo ~o
100 i
C
CTX+/,3.5"C 130rain)
t
300
600
i
!
D
i
,~ CTX + ~3.5"C
No. of fractions/ fraction interval (days)
Treatment
Slope + SD
TER +_ SE
1
CTX + heat CTX
0.065 + 0.003 0.029 + 0.001
2.3 + 0.2 a
3/3
CTX + heat CTX
0.054 _+ 0.004 0.025 + 0.002
2.2 + 0.2 a
5/3
CTX + heat CTX
0.046 + 0.003 0.017 + 0.001
2.7 + 0.3 a
3/5
CTX + heat CTX
0.044 _+ 0.003 0.026 + 0.001
1.7 + 0.2 a
5/5
CTX + heat CTX
0.034 + 0.005 0.018 _+ 0.001
1.9 + 0.3 b
130mml
i
1st.
sk
3/s
1 100
i 20O
i 300
,
I
01
400
500
0
a p ~ 0.001. ~p < 0.01. All slopes are significantly different from zero. I 100
I 200
i
i
[
30O
/.00
500
4
Fig. 3. Dose-effect curves for fractionated treatment using total dose of CTX as dose parameter. CTX alone (open symbols), CTX 15 rain prior to 43.5 °C for 30 rain (closed symbols). Upper panel: 3 days fraction interval. (A) 3 fractions; (B) 5 fractions. Lower panel: 5 days fraction interval. (C) 3 fractions; (D) 5 fractions. Points and bars indicate mean S G D + SE.
ment with heat alone were significantly different from zero. Consequently, dose-effect curves through the origin were fitted to the S G D data for heat alone. It is evident from Table V that D E R was significantly larger than 1 both for single fraction and fractionated treatment. Thus, the administered CTX dose provided a significant sensitization of the effect of heat treatment at 43.5 °C. The slopes of the dose-effect curves in Figs. 1B and 3 are shown in Table VI together with the resulting TERs. For fractionated CTX alone, as for fractionated heat alone, a dose range near zero was not possible to investigate. The regression analyses of the dose-effect curves for fractionated CTX alone showed that the intercepts of the dose-effect curves for 5 fractions were not significantly different from zero, whereas both doseeffect curves for 3 fractions intercepted slightly below zero. Dose-effect curves through the origin were, however, fitted to all the S G D data for CTX alone. It is
I
I
A
Total CTX dose (mg/kg)
3
t'
t 3dO~ int~vol"
I 3
I S
~2 0
I
I
0
t,
B
3
-
~
__
- ~ S days inler~lt'
l
!
0
[
t
Number o! fractions
Fig. 4. D E R (Table V) and TER (Table VI) as functions of the number of treatment fractions for 3 and 5 days fraction interval. Bars indicate SE. ( A ) D E R ; (B)TER.
177 evident from Table VI that T E R was significantly larger than 1 both for single fraction and fractionated treatment. Thus, the administered heat dose of 43.5 °C for 30 min provided a significant sensitization of the effect of CTX. D E R and T E R are plotted as functions of the number of treatment fractions in Fig. 4. The D E R and TER levels were larger with 3 days fraction interval than with 5 days fraction interval. The DERs for fractionated treatment with 3 days interval were significantly larger than D E R for single fraction treatment. D E R for 5 fractions with 3 days interval was probably underestimated due to the use of a reduced dose of CTX. There was a tendency for D E R to increase with increasing number of fractions. There were no major differences between TER for fractionated treatment and single fraction treatment. Discussion
The present analysis was based on three assumptions that were not completely met. However, the lack of full compliance with these assumptions did not affect the main conclusions. Firstly, interpretation of tumour growth delay data must be cautious when some tumours are controlled by treatment. In the present study, cured mice were observed mainly after combinedfractionatedtreatment in a dose-dependent pattern (Tables I-IV). Thus, the omitting of controlled tumours from the analysis resulted in underestimation, not overestimation, of some enhancement ratios for fractionated treatment. The only exception to this was TER for 5 fractions/3 days interval (Fig. 3B). T E R for this fractionation was the only TER-value for fractionated treatment that tended to be larger than T E R for single fraction treatment. Due to the observation of a larger proportion of cured mice after CTX alone than after combined treatment for this fractionation regimen, this TER-value should be interpreted very cautiously. Secondly, linear dose-effect relationships through the origin were as sumed for fractionated single modality treatment. When using S G D as effect parameter, the dose-effect curves for single modality treatment start according to definition in origo. The absolute values of the enhancement ratios were not significantly affected by forcing these dose-effect curves through the origin. However, the SD of the slopes of the curves were reduced by using this curve-fitting procedure. The correlation coefficients were generally higher for linear regression than for quadratic regression. For some curves they were at the same level [9]. Thirdly, S G D was assumed to be a more appropriate
effect parameter than TGT. However, a similar analysis using T G T as effect parameter was also performed. The dose-effect curves most extensively influenced by the variation in DT, were the curves for CTX alone. The dose-effect curves for fractionated CTX alone tended to be relatively steeper with S G D as effect parameter than with TGT. The TERs for fractionated treatment were thus somewhat smaller when using S G D than when using TGT as effect parameter. The experimental design eliminated the possible pitfall represented by variation in overall treatment time between the fractionation regimens [6]: every single enhancement ratio was determined from two doseeffect curves representing the same fractionation and overall treatment time. Thus, direct comparison of the regrowth delay level of different regimens, was avoided. The enhancement ratios for thermochemotherapy were influenced by the treatment fractionation. Neither the drug sensitization of the effect of heat treatment nor the thermal sensitization of the effect of drug treatment were lost after 3 or 5 fractions (Fig. 4). The drug sensitization was largest for the shortest fractionation interval and tended to increase with increasing number of fractions (Fig. 4A). The thermal sensitization was less extensively influenced by the number of fractions, but was largest for the shortest fractionation interval (Fig. 4B). The data indicate a potential for improving the therapeutic gain by fractionation of thermochemotherapy. There are no previous reports on determination of D E R or TER on the basis of dose-effect curves for fractionated thermochemotherapy of malignant tumours with any cancer chemotherapeutic agent. However, Hazan etal. [5] and Urano etal. [19] studied the growth response of murine tumours as functions of CTX dose both for CTX alone and CTX plus heat. Dose-effect curves for single fraction treatment were compared with curves for fractionated treatment using the same CTX dose per fraction. Although TERs for the fractionation regimens could not be calculated, the data indicate a thermal enhancement of the effect of CTX for the investigated fraetionation regimens. This is in agreement with the present observations. The effect of different schedules of fractionated thermochemotherapy is likely to depend on the kinetics of thermotolerance in the tumour. Th~: magnitude and kinetics of heat-induced thermotolerance depend largely on the magnitude of the first heat treatment [2,7,12-14,16,18]. For the C3H mouse mammary carcinoma investigated here, maximum thermotolerance occurs about 8-30 h after priming heat treatments of 15-45 rain at 43.5 °C and disappears between 3 and 6
178 days after the priming treatment [7,12-14]. For multiple fractions, the development of thermotolerance depends also on the time interval between the fractions [ 14]. Significant thermotolerance developed when the C3H mouse mammary carcinoma was given multiple treatments of 42.5 °C for 60 min with 1-3 days interval, but no significant thermotolerance developed when the intervals were prolonged to 5 days [ 14]. Similar observations were made with multiple treatments of 43.5 °C for 30 min [ 14]. Thus, in the present experiments it may be assumed that a significant level of thermotolerance was present when the treatment interval was 3 days, but not when the interval was 5 days. D E R and TER were found to be larger for a treatment interval of 3 days than of 5 days (Fig. 4). This observation probably reflects the difference in thermotolerance level for the two treatment intervals. This interpretation is in accordance with the previous report of larger D E R and a tendency for larger TER in thermotolerant than in previously untreated tumours [10]. Thermotolerance is therefore probably of less importance for the outcome of fractionated thermochemotherapy with CTX than for the outcome of fractionated treatment with heat alone.
In conclusion, the present thermochemotherapy experiments with a malignant murine tumour show that: (1) For single fraction treatment, D E R was 1.4 + 0.1 and TER 2.3 + 0.2. (2) The drug sensitization of the effect of heat treatment tended to increase with the number of treatment fractions. (3) The thermal sensitization of the effect of drug treatment was about the same for fractionated as for single fraction treatment. (4) The drug sensitization and the thermal sensitization were larger for a treatment interval of 3 days than of 5 days, i.e. largest with the fraction interval at which the thermotolerance was largest. (5) There might be a potential for improved therapeutic ratio by fractionation of thermochemotherapy.
Acknowledgement Financial support from the Norwegian Cancer Society is gratefully acknowledged.
References 1 Dahl, O. Hyperthermia and drugs. In: Hyperthermia, pp. 121-153. Editors: D. J. Watmough and W. M. Ross. Blackie & Son Ltd., Glasgow, 1985. 2 Gerweck, L.E. Hyperthermia in cancer therapy: the biological basis and unresolved questions. Cancer Res. 45: 3408-3414, 1985. 3 Goldie, J. H. and Coldman, A.J. The genetic origin of drug resistance in neoplasms: implications for systemic therapy. Cancer Res. 44: 3643-3653, 1984. 4 Hahn, G.M. Thermotolerance and drugs. In: Thermotolerance, Vol. 1, pp. 97-112. Editor: K. J. Henle. CRC Press Inc., Boca Raton, FI., 1987. 5 Hazan, G., Ben-Hur, E. and Yerushalmi, A. Synergism between hyperthermia and cyclophosphamide in vivo: the effect of dose fractionation. Eur. J. Cancer 17: 681-684, 1981. 6 Kallman, R.F., Brown, J.M., Denekamp, J,, Hill, R.P., Kummermehr, J. and Trott, K.-R. The use of rodent tumors in experimental cancer therapy. Cancer Res. 45: 6541-6545, 1985. 7 Kamura, T., Nielsen, O.S., Overgaard, J. and Andersen, A.H. Development of thermotolerance during fractionated hyperthermia in a solid tumour in vivo. Cancer Res. 42: 1744-1748, 1982. 8 Li, G.C. Heat shock proteins: role in thermotolerance, drug resistance, and relationship to DNA topoisomerases. Natl. Cancer Inst. Monogr. 4: 99-103, 1987. 9 Monge, O. and Rofstad, E.K. Thermoehemotherapy in vivo of a C3H mouse mammary carcinoma: thermotolerant tumours, Int. J. Hyperthermia 5: 579-587, 1989. 10 Monge, O., Rofstad, E.K. and Kaalhus, O. Thermochemotherapy in vivo of a C3H mouse mammary carcinoma: single fraction heat and drug treatment. Eur. J. Cancer Clin. Oncol. 24: 1661-1669, 1988.
11 Myers, C., Cowan, K., Sinha, B. and Chabner, B. The phenomenon of pleiotropie drug resistance. In: Important Advances in Oncology, pp. 27-38. Editors: V.T. de Vita, S. Hellman and S. A. Rosenberg. Lippincott, Philadelphia, 1987. 12 Nielsen, O.S. Fractionated hyperthermia and thermotolerance. Dan. Med. Bull. 31: 376-390, 1984. 13 Nielsen, O. S. andOvergaard, J. Importance ofpreheatingtemperature and time for the induction ofthermotolerance in a solid tumour in vivo. Br. J. Cancer 46: 894-903, 1982. 14 Overgaard, J. and Nielsen, O, S. The importance of thermotolerance for the clinical treatment with hyperthermia. Radiother. Oncol. 1: 167-178, 1983. 15 Pastan, I. H. and Gottesman, M.M. Molecular biology ofmultidrug resistance in human ceils. In: Important Advances in Oncology, pp. 3-16. Editors: V. T. de Vita, S. Hellman and S. A. Rosenberg. Lippincott, Philadelphia, 1988. 16 Rofstad, E.K. Influence of cellular, microenvironmental, and growth parameters on thermotolerance kinetics in vivo in human melanoma xenografts. Cancer Res. 49: 5027-5032, 1989. 17 Hyperthermic Oneology 1988. Proceedings of the 5th International Symposium on Hyperthermic Oncology, Vol. 1-2. Editors:T. Sugahara and M. Saito. Taylor & Francis, London, 1989. 18 Urano, U. Kinetics of thermotolerance in normal and tumour tissues: a review. Cancer Res. 46: 474-482, 1986. 19 Urano, M., Kahn, J. and Kenton, L.A. Thermochemotherapyinduced resistance to cyclophosphamide. Br. J. Cancer 57: 295-297, 1988. 20 Young, R.C. Mechanisms to improve chemotherapy effectiveness. Cancer 65: 815-822, 1990.
© 1991 Elsevier Science Publishers B.V. 0167-8140/91/$03.50
171
RADION 00857
Fractionated thermochemotherapy in vivo of a C3H mouse mammary carcinoma O d d R. M o n g e 1,2 a n d E i n a r K . R o f s t a d 2 Departments of ~Medical Oncology and Radiotherapy, and 2Biophysics, Institute for Cancer Research, The Norwegian Radium Hospital, and The Norwegian Cancer Society, Oslo, Norway
(Received 24 August 1990, revision received 26 March 1991, accepted 16 April 1991)
Key words: Hyperthermia; Thermotolerance; Drug resistance; Fractionated thermochemotherapy; Cyclophosphamide
Summary The interaction between fractionated heat treatment and fractionated drug treatment with cyclophosphamide (CTX) was investigated in a transplantable C3H mouse mammary carcinoma inoculated into the hind leg of C3D2F 1/Born mice. A tumour core temperature of 43.5 _+ 0.1 °C was achieved by immersing the tumour-bearing leg into a water bath thermostatically maintained at 43.7 + 0.1 ° C. CTX was administered i.p. using the maximum tolerated dose (MTD) (LD 1 ~o) for single fraction treatment (100 mg/kg) as the maximum fraction dose. For combined treatment CTX was given 15 min prior to heating. The endpoint was the time to reach a tumour volume of 5 times the volume at first treatment. Specific growth delay was used as effect parameter. In dose-effect experiments using total treatment time at 43.5 °C as dose parameter, drug enhancement ratio (DER) was determined as the ratio of the slope of the dose-effect curve for MTD of CTX plus heat to the slope of the curve for heat alone. In dose-effect experiments using total CTX dose as dose parameter, thermal enhancement ratio (TER) was determined as the ratio of the slope of the dose-effect curve for CTX plus 43.5 °C for 30 rain to the slope of the curve for CTX alone. The regimens investigated were single fraction treatment and 3 and 5 fractions with time intervals of 3 and 5 days. For single fraction treatment DER was 1.4 + 0.1 and TER 2.3 + 0.2. The drug sensitization of the effect of heat treatment tended to increase with increasing number of fractions. The thermal sensitization of the effect of drug treatment was about the same for fractionated and single fraction treatment. Both the drug sensitization and the thermal sensitization were larger for a treatment interval of 3 days than of 5 days, i.e. largest with the fraction interval at which the thermotolerance is known to be largest in this turnout. The data indicate a potential for improved therapeutic ratio by fractionation of thermochemotherapy.
Introduction Thermotolerance, i.e. heat-induced transient resistance to subsequent heat exposure, is one major obstacle for successful treatment of malignant tumours by fractionated hyperthermia [ 2,12,14,16,18 ]. Likewise, intrinsic and acquired tumour drug resistance are major obstacles for the success o f drug treatment o f malignant disease [3,11,15,20]. Thermotolerance and induced drug resistance are possibly interrelated phenomena, both representing biological adaptation to external environmental stress [2,4,8,18 ]. There are some indications that the problems of thermotolerance and drug
resistance can be reduced by the combined use of hyperthermia and c h e m o t h e r a p y - t h e r m o c h e m o t h e r a p y [ 1,2,4,8,9,17,18]. However, there are no reports on the drug sensitization o f the effect o f heat treatment or on the thermal sensitization o f the effect of drug treatment for fractionated thermochemotherapy of malignant tumours. We have previously reported that the drug sensitization o f the effect of heat treatment at 43.5 °C was 3 - 4 times larger in thermotolerant tumours than in previously untreated tumours both for cyclophosphamide (CTX) and mitomycin C in a C 3 H mouse m a m m a r y carcinoma, i.e. the expression of thermotolerance was to a large extent reduced, but not com-
Address for correspondence: Odd R. Monge, M.D., The Norwegian Radium Hospital, Department of Medical Oncology and Radiotherapy,
N-0310 Oslo 3, Norway
172 pletely overcome, by the thermochemotherapy [ 9]. Also the thermal sensitization of the effect of CTX tended to be larger in thermotolerant than in previously untreated tumours [9]. The effect of fractionated thermochemotherapy of the same tumour with CTX is reported here. The aim of the study was to determine drug enhancement ratio (DER) and thermal enhancement ratio (TER) as functions of the number of fractions and the time between fractions. Materials and methods Animal tumour system
Female C3D2F1/Bom mice weighing between 19 and 24 g were inoculated subcutaneously on the dorsal side of the right hind foot with a spontaneously arisen C3H mouse mammary carcinoma as previously described [ 10 ]. Tumour volume was determined using the formula n/6 x D 1 x D2 x D3. The three orthogonal diameters (D1-3) were measured with a slide gauge. Tumours
were randomly allocated to treatment when a volume between 130 and 200 mm 3 was reached. Treatment
During heat treatment the unanaesthetized mice were placed in lucite plastic jigs with the tumour-bearing leg loosely fixed with tape proximal to the tumour without impairing the blood flow. Local hyperthermia with a tumour core temperature of 43.5 + 0.1 °C was given by immersion of the tumour-bearing legs into a circulating water bath thermostatically maintained at 43.7 + 0.1 °C [ 10]. The body temperature of the mice was below 39 ° C. Cyclophosphamide (L~i~tke Farmos, Farmitalia Carlo Erba) was dissolved in sterile water and administered i.p. with 0.02 ml/g body weight. The maximum dose per fraction was 100mg/kg corresponding to maximum tolerated dose (MTD), i.e. the dose that kills approximately 1 ~o of the mice within 150 days after single fraction treatment [10]. Mice with tumours that did not regrow, were likewise observed at
TABLE I Heat alone at 43.5 °C: results after single fraction and fractionated treatment. Heating time per fraction (min at 43.5 ° C)
No. of fractions/ fraction interval (days)
No. of cured/no, of mice
D T + SE (days)
T G T + SE (days)
0
0*
0/95
1.8 _+ 0.1
4.0 + 0.1
10 30 50 70 90
1 1 I 1 1
0/32 0/54 0/46 0/44 0/42
1.8 1.9 2.1 2.1 2.1
+ + ± + ±
25 30 40 50 60 70
3/3 3/3 3/3 3/3 3/3 3/3
0/15 0/16 0/12 0/15 0/5 0/15
2.4 2.3 2,5 2,4 2.3 2.6
± 0.1 +_ 0.1 ± 0.1 + 0.I ± 0.1 ± 0.2
0.1 0.1 0.I 0.1 0.1
4.7 7.9 11.1 14.4 17.6
S G D ± SE
+ 0.2 _+ 0.3 +_ 0.3 + 0.5 + 0.9
0.4 2.1 3.3 5.2 6.4
+ + + + +
9.8 + 0.8
2.6 3.4 5.6 6.1 6.6 8.3
+ 0.4 + 0.3 + 0.6 _+ 0.5 ± 1.0 + 0.8
11.6 + 0.4
17.4 18.1 18.6 24.6
+ + + +
1.0 1.0 1.9 1.4
0.1 0,2 0,2 0.3 0.4
30
5/3
0/15
40 50
5/3 5/3
0/16 0/17
3.0 + 0.2 2.7 ± 0.2 2.8 ± 0.2
14.8 + 1.0 20.6 + 0.8 26.1 + 1.1
3.9 + 0.5 6.5 + 0.5 8.4 + 0.6 5,5 + 0.7 8.1 + 0.7 9.8 + 0.8
30
3/5
0/12
2.5 ± 0.1
40 50
3/5 3/5
0/16 0/10
2.3 + 0.4 2.6 ± 0.2
17.8 + 1.8 22.3 + 1.3 28.1 _+ 1.6
35 37.5 40 42.5
5/5 5/5 5/5 5/5
0/15 0/11 0/17 0/18
2.7 2.3 2.5 2.5
29.7 31.7 32.7 35.0
* Untreated controls.
+ ± + ±
0.1 0.2 0.1 0.2
+ + + ±
0.8 2.4 1.0 2.1
9.7 12.9 12.3 13.1
_+ 0.7 + 1.4 + 0.8 ± 1.1
173 least 150 days after the first treatment before they were considered cured of their tumour. For combined treatment CTX was given 15 rain prior to heating, i.e. the sequencing that yielded the largest tumour response after single fraction treatment [10].
Evaluation of response After treatment, tumour volume was measured 3 to 4 times a week. The time to reach a tumour volume of 5 times the volume at first treatment (day zero) was used as endpoint and recorded as tumour growth time (TGT). The growth rate during regrowth was determined as tumour volume doubling time (DT) for each tumour. Due to differences in mean DT between treatment groups, the data were analyzed using specific growth delay (SGD) as effect parameter: SGD
= TGTtrCat~d - T G T u n t r ¢ ~ t e d DTtreated
The investigated fractionation regimens were single fraction treatment and 3 and 5 fractions with time intervals of 3 and 5 days. For each fractionation regimen, dose-effect experiments were performed using two
different dose parameters, i.e. total heating time at 43.5 °C and total dose of CTX. For determination of DER, dose-effect curves were constructed for heat alone and for M T D of CTX given 15 rain prior to heat treatment. For determination of TER, dose-effect curves were constructed for CTX alone and for CTX given 15 rain prior to 30 rain heat treatment at 43.5 °C. Enhancement ratios were calculated as the ratios of slopes of dose-effect curves. For each fractionation regimen, the total dose range possible to investigate with the present animal tumour system was exploited. Regrowth to the endpoint before administration of the last treatment fraction prevented the use of lower doses than the lowest dose levels used here. The upper dose levels were limited by either treatment toxicity or a high frequency of controlled tumours. Although the data presented in Tables I - I V and Figs. 1-3 are mean values, the linear regression analysis of dose-effect curves was based on the individual SGDs, i.e. the individual DTs and TGTs, of all contributing tumours, not the mean values. Significance levels of enhancement ratios were determined by a two-tailed t-test and indicate at which level the ratios were different from 1. The experimental protocol was approved by the Experimental Animal Board.
TABLE I! Cyclophosphamide 100 mg/kg 15 min prior to each heating at 43.5 °C: results after single fraction and fractionated treatment. Heating time per fraction (min at 43.5 ° C)
No. of fractions/ fraction interval (days)
10 30 50 70 90
1
1 1 1 1
10 20 30
DT ± SE (days)
T G T i SE (days)
S G D ± SE
0/16 0/16 0/16 0/16 0/14
2.9 1.9 2.0 2.1 2.3
10.9 20.4 23.8 26.4 28.4
2.4 8.7 10.0 10.8 10.4
3/3 3/3 3/3
6/16 3/14 5/13
2.7 + 0.1 2.5 ± 0.1 2.5 + 0.2
38.1 ± 2.6 47.6 _+ 4.0 47.2 _+ 2.0
13.2 _+ 1.5 17.5 ± 1.0 17.6 ± 1.4
10 20 30
5/3* 5/3* 5/3*
1/16 3/15 1/16
2.9 ± 0.1 2.6 ± 0.2 2.5 ± 0.1
29.2 ± 1.2 42.7 ± 2.6 47.1 ± 2.5
8.9 ± 0.5 15.5 + 1.2 17.8 + 1.2
10 20 30
3/5 3/5 3/5
1/15 4/16 0/13
2.4 ± 0.1 2.5 ± 0.1 2.5 +_ 0.1
32.2 ± 1.6 44.7 ± 1.3 51.8 ± 1.7
11.7 ± 0.5 16.2 ± 0.7 19.2 _+ 0.9
5 7.5 10 15
5/5 5/5 5/5 5/5
0/16 4/16 8/15 10/17
3.3 2.6 2.7 2.8
31.8 38.2 48.9 56.4
8.6 13.7 16.8 18.7
* CTX 60 mg/kg.
No. of cured/no, of mice
+_ 0.1 ± 0.1 _+ 0.1 ± 0.1 + 0.2
+ + ± +
0.2 0.1 0.1 0.1
_+ 0.4 + 0.5 _+ 0.6 _+ 0.9 ± 1.2
± ± + ±
0.8 0.9 3.4 3.0
+ 0.2 i 0.4 ± 0.5 _+ 0.6 ± 0.6
± ± ± ±
0.4 0.7 1.4 1.5
174 T A B L E III Cyclophosphamide (CTX) alone: results after single fraction and fractionated treatment. CTX dose per fraction (mg/kg)
No. of fractions/ fraction interval (days)
No. of cured/no, of mice
D T _+ SE (days)
T G T + SE (days)
S G D + SE
33 67 100
1 1 1
0/16 0/16 0/16
2.3 +_ 0.1 2.5 + 0.2 2.8 + 0.2
5.3 + 0.2 9.4 + 0.6 11.6 + 0.7
0.6 +_ 0.1 2.2 + 0.3 2.9 + 0.4
50 75 100
3/3 3/3 3/3
0/16 1/16 0/19
3.6 +_ 0.2 2.7 + 0.1 2.7 +_ 0.2
10.8 + 0.6 17.0 + 0.6 25.2 + 1.8
2.0 + 0.2 4.9 + 0.4 8.5 + 0.8
60 80 100
5/3 5/3 5/3
1/16 1/16 7/25
3.7 + 0.2 3.4 + 0.2 3.2 +_ 0.2
19.3 + 0.5 25.2 + 1.5 30.3 + 1.9
4.3 + 0.3 6.7 + 0.7 8.8 + 0.8
67 83 100
3/5 3/5 3/5
0/15 0/15 0/15
3.1 + 0.3 3.3 + 0.2 2.5 + 0.1
16.9 + 0.6 19.4 + 0.5 26.8 ± 0.9
4.6 + 0.4 4.9 + 0.4 9.3 + 0.4
67 83 100
5/5 5/5 5/5
0/14 0/15 0/15
4.5 +_ 0.2 3.9 +_ 0.5 3.3 + 0.2
25.6 + 0.7 29.2 + 1.5 31.5 + 1.0
5.0 + 0.4 7.8 + 1.0 8.9 + 0.8
TABLE IV Cyclophosphamide 15 min prior to each 30 min heating at 43.5 °C: results after single fraction and fractionated treatment. CTX dose per fraction (mg/kg)
No. of fractions/ fraction interval (days)
No. of cured/no, of mice
D T + SE (days)
T G T + SE (days)
S G D +_ SE
33 67 100
1 1 1
0/16 0/16 0/16
2.1 + 0.1 1.9 + 0.1 1.9 +_ 0.1
11.0 + 0.6 16.0 + 0.5 20.4 _+ 0.5
3.4 + 0.4 6.5 + 0.3 8.7 + 0.4
33 67 100
3/3 3/3 3/3
0/16 2/16 5/13
2.0 + 0.1 2.0 + 0.1 2.5 + 0.2
20.6 + 0.7 37.0 + 1.7 47.2 + 2.0
8.4 + 0.5 16.7 + 0.9 17.6 + 1.4
33 50 60
5/3 5/3 5/3
0/16 1/16 1/16
2.3 + 0.1 2.3 + 0.1 2.5 + 0.1
29.6 + 0.7 37.5 +_ 0.9 47.1 + 2.5
11.1 + 0.4 14.9 + 0.7 17.8 + 1.2
33 67 100
3/5 3/5 3/5
0/15 1/16 0/13
2.3 + 0.1 2.5 +_ 0.1 2.5 + 0.1
27.7 + 1.0 38.0 + 1.0 51.8 + 1.7
10.5 + 0.4 13.8 + 0.6 19.2 _+ 0.9
17 33 50 67
5/5 5/5 5/5 5/5
0/ 5 0/16 0/15 0/16
3.4 2.6 3.0 3.0
29.5 39.6 49.8 59.2
8.1 14.2 15.5 18.5
+ 0.4 + 0.1 + 0.1 +_ 0.1
+ + + +
2.0 1.4 2.0 1.8
+ + + +
1.3 0.9 0.6 0.8
175
l
'
'
i
!
I
I
[
r
B
r
B c T x
/
'
i
i
~mll~ ÷~ot
•
CTX÷t,3S°C {30n',,n]
5/3
'0
50
i 100
i 150
i
i
200
250
i 50 D~
0
5O foe 15o Treotn~n( tn~' at 43.5"C (rain)
300 CTX ~
I00
i I 1~10 200
i | 1 CTXI~r~/l~ ÷ h*ot
I 250 b
(mg/kg)
Fig. 1. Dose-effect curves for single fraction treatment. (A) Treatment time at 43.5 °C as dose parameter: heat alone ( O ) and M T D of CTX 15 min prior to heating (O). ( B ) C T X dose as dose parameter: CTX alone ( A ) and CTX 15 min prior to 43.5 °C for 30 rain (&). Points and bars indicate mean S G D + SE.
15
Results
~5
The primary results are presented in Tables I-IV. Lack of tumour regrowth within 150 days, i.e. mice cured of their tumour, was observed mainly after combined TABLE V Slopes and standard deviations of dose-effect curves for single fraction and fractionated treatment with either heat alone at 43.5 ° C or cyclophosphamide (CTX) 100 mg/kg 15 rain prior to each heating at 43.5 °C. Drug enhancement ratios (DER) were calculated as ratios of slopes of dose effect curves. No. of fractions/ fraction interval (days)
Treatment
Slope _+ SD
DER + SE
1
CTX + heat Heat
0.100 _+ 0.008 0.071 + 0.002
1.4 -+ 0.1 b
3/3
CTX + heat Heat
0.115 + 0.017 0.040 + 0.002
2.9 + 0.5 b
5/3
CTX + heat a Heat
0.094 _+ 0.008 0,032 _+ 0.002
2.9 _+ 0.38
3/5
CTX + heat Heat
0.109 _+ 0.009 0,065 +_ 0.004
1.7 _+ 0.28
5/5
CTX + heat Heat
0,147 _+ 0.018 0,062 +_ 0.003
2.4 _+ 0.38
a CTX dose 60 mg/kg. b p ,~ 0.001. All slopes are significantly different from zero.
0
I
50
100
150
i
200 250 50 100 Total trgatrnent time at 43,5=C (rain)
I
150
I 200
I 250
Fig. 2. Dose-effect curves for fractionated treatment using total treatment time at 43.5 °C as dose parameter: beat alone (open symbols), M T D of CTX 15 min prior to heating (closed symbols). Upper panel: 3 days fraction interval. (A)3 fractions; (B)5 fractions (60 mg/kg of CTX). Lower panel: 5 days fraction interval. (C) 3 fractions; ( D ) 5 fractions. Points and bars indicate mean S G D + SE.
fractionated treatment. After 5 fractions/3 days interval (Table II), M T D of CTX and heating times of 10 min and more resulted in controlled tumours in the vast majority of the mice (data not shown). For this reason, a lower dose of CTX was used for combined treatment with that fractionation regimen (Table II). Controlled tumours were omitted from the calculation of mean SGD. The two sets of dose-effect curves for single fraction treatment are shown in Fig. 1 and the two sets of doseeffect curves for fractionated treatment are shown in Figs. 2 and 3. The slopes of the dose-effect curves in Fig. 1A and Fig. 2 are tabulated in Table V together with the resulting DERs. For fractionated heat treatment alone a dose range near zero was not possible to investigate. The regression analyses showed that none of the intercepts of the dose-effect curves for fractionated treat-
176 20
i
20
i
i
,
T A B L E VI Slopes and standard deviations of dose-effect curves for single fraction and fractionated treatment with either cyclophosphamide (CTX) alone or CTX 15 rain prior to each 30 min heat treatment at 43.5 ° C. Thermal enhancement ratios (TER) were calculated as the ratios of slopes of dose-effect curves.
c1x
rg3
313
!
!
,,oo ~o
100 i
C
CTX+/,3.5"C 130rain)
t
300
600
i
!
D
i
,~ CTX + ~3.5"C
No. of fractions/ fraction interval (days)
Treatment
Slope + SD
TER +_ SE
1
CTX + heat CTX
0.065 + 0.003 0.029 + 0.001
2.3 + 0.2 a
3/3
CTX + heat CTX
0.054 _+ 0.004 0.025 + 0.002
2.2 + 0.2 a
5/3
CTX + heat CTX
0.046 + 0.003 0.017 + 0.001
2.7 + 0.3 a
3/5
CTX + heat CTX
0.044 _+ 0.003 0.026 + 0.001
1.7 + 0.2 a
5/5
CTX + heat CTX
0.034 + 0.005 0.018 _+ 0.001
1.9 + 0.3 b
130mml
i
1st.
sk
3/s
1 100
i 20O
i 300
,
I
01
400
500
0
a p ~ 0.001. ~p < 0.01. All slopes are significantly different from zero. I 100
I 200
i
i
[
30O
/.00
500
4
Fig. 3. Dose-effect curves for fractionated treatment using total dose of CTX as dose parameter. CTX alone (open symbols), CTX 15 rain prior to 43.5 °C for 30 rain (closed symbols). Upper panel: 3 days fraction interval. (A) 3 fractions; (B) 5 fractions. Lower panel: 5 days fraction interval. (C) 3 fractions; (D) 5 fractions. Points and bars indicate mean S G D + SE.
ment with heat alone were significantly different from zero. Consequently, dose-effect curves through the origin were fitted to the S G D data for heat alone. It is evident from Table V that D E R was significantly larger than 1 both for single fraction and fractionated treatment. Thus, the administered CTX dose provided a significant sensitization of the effect of heat treatment at 43.5 °C. The slopes of the dose-effect curves in Figs. 1B and 3 are shown in Table VI together with the resulting TERs. For fractionated CTX alone, as for fractionated heat alone, a dose range near zero was not possible to investigate. The regression analyses of the dose-effect curves for fractionated CTX alone showed that the intercepts of the dose-effect curves for 5 fractions were not significantly different from zero, whereas both doseeffect curves for 3 fractions intercepted slightly below zero. Dose-effect curves through the origin were, however, fitted to all the S G D data for CTX alone. It is
I
I
A
Total CTX dose (mg/kg)
3
t'
t 3dO~ int~vol"
I 3
I S
~2 0
I
I
0
t,
B
3
-
~
__
- ~ S days inler~lt'
l
!
0
[
t
Number o! fractions
Fig. 4. D E R (Table V) and TER (Table VI) as functions of the number of treatment fractions for 3 and 5 days fraction interval. Bars indicate SE. ( A ) D E R ; (B)TER.
177 evident from Table VI that T E R was significantly larger than 1 both for single fraction and fractionated treatment. Thus, the administered heat dose of 43.5 °C for 30 min provided a significant sensitization of the effect of CTX. D E R and T E R are plotted as functions of the number of treatment fractions in Fig. 4. The D E R and TER levels were larger with 3 days fraction interval than with 5 days fraction interval. The DERs for fractionated treatment with 3 days interval were significantly larger than D E R for single fraction treatment. D E R for 5 fractions with 3 days interval was probably underestimated due to the use of a reduced dose of CTX. There was a tendency for D E R to increase with increasing number of fractions. There were no major differences between TER for fractionated treatment and single fraction treatment. Discussion
The present analysis was based on three assumptions that were not completely met. However, the lack of full compliance with these assumptions did not affect the main conclusions. Firstly, interpretation of tumour growth delay data must be cautious when some tumours are controlled by treatment. In the present study, cured mice were observed mainly after combinedfractionatedtreatment in a dose-dependent pattern (Tables I-IV). Thus, the omitting of controlled tumours from the analysis resulted in underestimation, not overestimation, of some enhancement ratios for fractionated treatment. The only exception to this was TER for 5 fractions/3 days interval (Fig. 3B). T E R for this fractionation was the only TER-value for fractionated treatment that tended to be larger than T E R for single fraction treatment. Due to the observation of a larger proportion of cured mice after CTX alone than after combined treatment for this fractionation regimen, this TER-value should be interpreted very cautiously. Secondly, linear dose-effect relationships through the origin were as sumed for fractionated single modality treatment. When using S G D as effect parameter, the dose-effect curves for single modality treatment start according to definition in origo. The absolute values of the enhancement ratios were not significantly affected by forcing these dose-effect curves through the origin. However, the SD of the slopes of the curves were reduced by using this curve-fitting procedure. The correlation coefficients were generally higher for linear regression than for quadratic regression. For some curves they were at the same level [9]. Thirdly, S G D was assumed to be a more appropriate
effect parameter than TGT. However, a similar analysis using T G T as effect parameter was also performed. The dose-effect curves most extensively influenced by the variation in DT, were the curves for CTX alone. The dose-effect curves for fractionated CTX alone tended to be relatively steeper with S G D as effect parameter than with TGT. The TERs for fractionated treatment were thus somewhat smaller when using S G D than when using TGT as effect parameter. The experimental design eliminated the possible pitfall represented by variation in overall treatment time between the fractionation regimens [6]: every single enhancement ratio was determined from two doseeffect curves representing the same fractionation and overall treatment time. Thus, direct comparison of the regrowth delay level of different regimens, was avoided. The enhancement ratios for thermochemotherapy were influenced by the treatment fractionation. Neither the drug sensitization of the effect of heat treatment nor the thermal sensitization of the effect of drug treatment were lost after 3 or 5 fractions (Fig. 4). The drug sensitization was largest for the shortest fractionation interval and tended to increase with increasing number of fractions (Fig. 4A). The thermal sensitization was less extensively influenced by the number of fractions, but was largest for the shortest fractionation interval (Fig. 4B). The data indicate a potential for improving the therapeutic gain by fractionation of thermochemotherapy. There are no previous reports on determination of D E R or TER on the basis of dose-effect curves for fractionated thermochemotherapy of malignant tumours with any cancer chemotherapeutic agent. However, Hazan etal. [5] and Urano etal. [19] studied the growth response of murine tumours as functions of CTX dose both for CTX alone and CTX plus heat. Dose-effect curves for single fraction treatment were compared with curves for fractionated treatment using the same CTX dose per fraction. Although TERs for the fractionation regimens could not be calculated, the data indicate a thermal enhancement of the effect of CTX for the investigated fraetionation regimens. This is in agreement with the present observations. The effect of different schedules of fractionated thermochemotherapy is likely to depend on the kinetics of thermotolerance in the tumour. Th~: magnitude and kinetics of heat-induced thermotolerance depend largely on the magnitude of the first heat treatment [2,7,12-14,16,18]. For the C3H mouse mammary carcinoma investigated here, maximum thermotolerance occurs about 8-30 h after priming heat treatments of 15-45 rain at 43.5 °C and disappears between 3 and 6
178 days after the priming treatment [7,12-14]. For multiple fractions, the development of thermotolerance depends also on the time interval between the fractions [ 14]. Significant thermotolerance developed when the C3H mouse mammary carcinoma was given multiple treatments of 42.5 °C for 60 min with 1-3 days interval, but no significant thermotolerance developed when the intervals were prolonged to 5 days [ 14]. Similar observations were made with multiple treatments of 43.5 °C for 30 min [ 14]. Thus, in the present experiments it may be assumed that a significant level of thermotolerance was present when the treatment interval was 3 days, but not when the interval was 5 days. D E R and TER were found to be larger for a treatment interval of 3 days than of 5 days (Fig. 4). This observation probably reflects the difference in thermotolerance level for the two treatment intervals. This interpretation is in accordance with the previous report of larger D E R and a tendency for larger TER in thermotolerant than in previously untreated tumours [10]. Thermotolerance is therefore probably of less importance for the outcome of fractionated thermochemotherapy with CTX than for the outcome of fractionated treatment with heat alone.
In conclusion, the present thermochemotherapy experiments with a malignant murine tumour show that: (1) For single fraction treatment, D E R was 1.4 + 0.1 and TER 2.3 + 0.2. (2) The drug sensitization of the effect of heat treatment tended to increase with the number of treatment fractions. (3) The thermal sensitization of the effect of drug treatment was about the same for fractionated as for single fraction treatment. (4) The drug sensitization and the thermal sensitization were larger for a treatment interval of 3 days than of 5 days, i.e. largest with the fraction interval at which the thermotolerance was largest. (5) There might be a potential for improved therapeutic ratio by fractionation of thermochemotherapy.
Acknowledgement Financial support from the Norwegian Cancer Society is gratefully acknowledged.
References 1 Dahl, O. Hyperthermia and drugs. In: Hyperthermia, pp. 121-153. Editors: D. J. Watmough and W. M. Ross. Blackie & Son Ltd., Glasgow, 1985. 2 Gerweck, L.E. Hyperthermia in cancer therapy: the biological basis and unresolved questions. Cancer Res. 45: 3408-3414, 1985. 3 Goldie, J. H. and Coldman, A.J. The genetic origin of drug resistance in neoplasms: implications for systemic therapy. Cancer Res. 44: 3643-3653, 1984. 4 Hahn, G.M. Thermotolerance and drugs. In: Thermotolerance, Vol. 1, pp. 97-112. Editor: K. J. Henle. CRC Press Inc., Boca Raton, FI., 1987. 5 Hazan, G., Ben-Hur, E. and Yerushalmi, A. Synergism between hyperthermia and cyclophosphamide in vivo: the effect of dose fractionation. Eur. J. Cancer 17: 681-684, 1981. 6 Kallman, R.F., Brown, J.M., Denekamp, J,, Hill, R.P., Kummermehr, J. and Trott, K.-R. The use of rodent tumors in experimental cancer therapy. Cancer Res. 45: 6541-6545, 1985. 7 Kamura, T., Nielsen, O.S., Overgaard, J. and Andersen, A.H. Development of thermotolerance during fractionated hyperthermia in a solid tumour in vivo. Cancer Res. 42: 1744-1748, 1982. 8 Li, G.C. Heat shock proteins: role in thermotolerance, drug resistance, and relationship to DNA topoisomerases. Natl. Cancer Inst. Monogr. 4: 99-103, 1987. 9 Monge, O. and Rofstad, E.K. Thermoehemotherapy in vivo of a C3H mouse mammary carcinoma: thermotolerant tumours, Int. J. Hyperthermia 5: 579-587, 1989. 10 Monge, O., Rofstad, E.K. and Kaalhus, O. Thermochemotherapy in vivo of a C3H mouse mammary carcinoma: single fraction heat and drug treatment. Eur. J. Cancer Clin. Oncol. 24: 1661-1669, 1988.
11 Myers, C., Cowan, K., Sinha, B. and Chabner, B. The phenomenon of pleiotropie drug resistance. In: Important Advances in Oncology, pp. 27-38. Editors: V.T. de Vita, S. Hellman and S. A. Rosenberg. Lippincott, Philadelphia, 1987. 12 Nielsen, O.S. Fractionated hyperthermia and thermotolerance. Dan. Med. Bull. 31: 376-390, 1984. 13 Nielsen, O. S. andOvergaard, J. Importance ofpreheatingtemperature and time for the induction ofthermotolerance in a solid tumour in vivo. Br. J. Cancer 46: 894-903, 1982. 14 Overgaard, J. and Nielsen, O, S. The importance of thermotolerance for the clinical treatment with hyperthermia. Radiother. Oncol. 1: 167-178, 1983. 15 Pastan, I. H. and Gottesman, M.M. Molecular biology ofmultidrug resistance in human ceils. In: Important Advances in Oncology, pp. 3-16. Editors: V. T. de Vita, S. Hellman and S. A. Rosenberg. Lippincott, Philadelphia, 1988. 16 Rofstad, E.K. Influence of cellular, microenvironmental, and growth parameters on thermotolerance kinetics in vivo in human melanoma xenografts. Cancer Res. 49: 5027-5032, 1989. 17 Hyperthermic Oneology 1988. Proceedings of the 5th International Symposium on Hyperthermic Oncology, Vol. 1-2. Editors:T. Sugahara and M. Saito. Taylor & Francis, London, 1989. 18 Urano, U. Kinetics of thermotolerance in normal and tumour tissues: a review. Cancer Res. 46: 474-482, 1986. 19 Urano, M., Kahn, J. and Kenton, L.A. Thermochemotherapyinduced resistance to cyclophosphamide. Br. J. Cancer 57: 295-297, 1988. 20 Young, R.C. Mechanisms to improve chemotherapy effectiveness. Cancer 65: 815-822, 1990.
