PII S0360-3016(98)00565-3

BIOLOGY CONTRIBUTION

COMPARISON OF RADIOSENSITIZATION BY 41°C HYPERTHERMIADURING LOW DOSE RATE IRRADIATION AND DURING PULSED

SIMULATED LOW DOSE RATE IRRADIATION IN HUMAN GLIOMA CELLS

G. PETER RAAPHORST, PH.D., CHENG E. NG, PH.D., AND BILAL SHAHINE, M.SC.

Ottawa Regional Cancer Centre, Ottawa, Ontario, Canada

Purpose: Long duration mild hyperthermia has been shown to be an effective radiosensitizer when givenconcurrently with low dose rate irradiation. Pulsed simulated low dose rate (PSLDR) is now being used clinically,and we have set out to determine whether concurrent mild hyperthermia can be an effective radiosensitizer forthe PSLDR protocol.Materials and Methods: Human glioma cells (U-87MG) were grown to plateau phase and treated in plateau phasein order to minimize cell cycle redistribution during protracted treatments. Low dose rate (LDR) irradiation and41°C hyperthermia were delivered by having a radium irradiator inside a temperature-controlled incubator.PSLDR was given using a 150 kVp X-ray unit and maintaining the cells at 41°C between irradiations. Theduration of irradiation and concurrent heating depended on total dose and extended up to 48 h.Results: When 41°C hyperthermia was given currently with LDR or PSLDR, the thermal enhancement ratios(TER) were about the same if the average dose rate for PSLDR was the same as for LDR. At higher average doserates for PSLDR the TERs became less.Conclusions: Our data show that concurrent mild hyperthermia can be an effective sensitizer for PSLDR. Thissensitization can be as effective as for LDR if the same average dose rate is used and the TER increases withdecreasing dose rate. Thus mild hyperthermia combined with PSLDR may be an effective clinical protocol.© 1999 Elsevier Science Inc.

Human glioma, Mild hyperthermia, Low dose rate irradiation, Pulsed simulated low dose rate irradiation.

INTRODUCTION

A number of studies have shown that one of the mecha-nisms of thermoradiosensitization in mammalian cells isthrough the inhibition of recovery of potentially lethal dam-age repair (PLDR) and sublethal damage repair (SLDR)(1–5). It was further shown that thermoradiosensitizationcould be enhanced when mild hyperthermia treatments weregiven during low dose rate (LDR) irradiation (6–11). Theseresults support the concept that hyperthermia inhibitedSLDR under LDR conditions. Studies showing that hyper-thermia can inhibit repair of both DNA double- and single-strand breaks (12) and can inhibit DNA polymerase in amanner correlated to cell killing (13–17) further support themodel that at least in part thermoradiosensitization isthrough repair inhibition.

A study by Wanget al. (18) showed that in rat 9L cellsradiation sparing (SLDR) increased with decreased radi-ation dose rates, but this effect could be completelyeliminated by heating at 41°C during irradiation andresulted in an escalating thermal enhancement ratio(TER) as the dose rate decreased. These data indicated

that radiation dose rate and heating time and temperatureare important parameters that need to be further exam-ined in order to provide guidance to the optimal use ofthis approach clinically. Already, two reports show en-couraging clinical results for the simultaneous combina-tion of mild hyperthermia and brachytherapy (19, 20).Another clinical approach to brachytherapy is the use ofpulsed simulated low dose rate (PSLDR) irradiation.There is extensive investigation into the determination ofwhich pulse intervals and durations best simulate contin-uous LDR irradiation (21–23). Since fractionated irradi-ation results in SLDR it is possible that hyperthermiaduring PSLDR may be effective in radiosensitizationthrough the inhibition of SLDR. One study has alreadyshown that pulses of hyperthermia as well as continuousmild hyperthermia can cause radiosensitization in a ro-dent cell line (24).

In this study we have set out to evaluate the effect of mildhyperthermia (41°C) on pulsed radiation treatment in hu-man glioma cells. In addition, we have compared the effectof 41°C for pulsed irradiation PSLDR and true continuousLDR.

Reprint requests to: Dr. G. Peter Raaphorst, Medical PhysicsDepartment, Ottawa Regional Cancer Centre, 501 Smyth Rd.,

Ottawa, K1H 8L6, Canada.Accepted for publication 11 December 1998.

Int. J. Radiation Oncology Biol. Phys., Vol. 44, No. 1, pp. 185–188, 1999Copyright © 1999 Elsevier Science Inc.Printed in the USA. All rights reserved

0360-3016/99/$–see front matter

185

METHODS AND MATERIALS

The cell lines U-87MG, derived from a human malignantglioblastoma and obtained from the American Type CultureCollection (ATCC) cell repository, are described in theATCC catalogue. Cells were grown in the medium combi-nation Dulbecco’s modified Eagle medium/F-12 (DMEM/F-12) (1:1) with 15% fetal bovine serum/newborn calf se-rum (FBS:NCS) (1:1), 1 mM Eagle’s minimum essentialmedium (MEM) nonessential amino acids, 10 mM sodiumbicarbonate, and 20 mM Hepes. All cell cultures wereincubated at 37°C in a humidified atmosphere of 2% CO2

and 98% air. Plating efficiency of this cell line under theseconditions was about 25–35%.

For all experiments 43 103 cells per cm2 were plated in25-cm2 flasks for high dose rate (HDR) irradiation and in1.3-cm2 glass vials for LDR irradiation. Experiments com-menced 3 days after the last medium change as the cellsentered a plateau growth phase and as verified by growthcurves and cell cycle distributions (87% G0/G1). Cell den-sity at confluency was 1–1.53 105 per cm2 and the mediumpH at the start of experiments was about 7.0 and did notchange.0.1 during the experiments. This was the equilib-rium pH for the plateau phase cell populations. Cell cyclemeasurements using flow cytometry showed that cell cycleprogression and redistribution did not occur during theexperiments.

For HDR irradiation, cells were irradiated on ice withX-rays from a 150 kVp tube having a half-value layer(HVL) of 2.75 cm aluminum at a dose rate of 1126 3cGy/min. Cells were taken from the 37°C or 41°C incuba-tor, placed on ice, irradiated, and then warmed in a 37°C or41°C water bath before returning to the incubator. The doserate was determined by Fricke dosimetry. For LDR irradi-ation, cells were irradiated by an array of226Ra sources(total activity 800 mCi) contained in a temperature-con-trolled incubator. Two low dose rates were used, LDR1 at0.88 cGy/min, and LDR2 at 0.41 cGy/min. These dose rateswere determined by thermoluminescent dosimeter (TLD)relative to a60Co reference protocol.

Cell cycle distributions were measured by standard flowcytometry techniques using an Ortho cytofluorograf andpropidium iodide staining of the nuclear DNA.

Hyperthermia for LDR was achieved by incubating thecells at the desired temperature in the LDR irradiator whichwas inside a temperature- and atmospheric-controlled incu-bator. For the fractionation experiments the flasks contain-ing the cells were held in temperature-controlled waterbaths between irradiations. In both the temperature-con-trolled incubators and water baths the temperature wasmaintained to60.05°C.

All data points on all graphs represent the average of atleast three independent experiments with error bars signi-fying the standard error of the mean. Radiation survivalcurves were fitted using the linear quadratic survival curvemodel.

RESULTS

The hyperthermia response for the U-87MG cell line isshown in Fig. 1. The data show that heating to 41°C for 24 hresults in very little cell killing whereas for 43°C the effectshows increased killing with a survival curve shoulder.These data also show that long duration hyperthermia at43°C is not possible due to excessive cell killing effect.

The results for combining hyperthermia at 41°C withLDR irradiation are shown in Fig. 2. The data show thatLDR irradiation at 0.81 and 0.41 cGy/min give progres-sively greater survival compared to HDR. When 41°C wascombined with LDR irradiation, radiosensitization occurredand the results were between the HDR and LDR curves. Thesummary in Table 1 shows that the greatest TER wasachieved at the lowest dose rate, possibly implicating agreater effect on SLDR inhibition.

Figure 3 compares the effects of 41°C hyperthermiacombined with fractionated irradiation to HDR irradiationand fractionation without hyperthermia. As described inMethods and Materials, cells were incubated at either 37°Cor 41°C during the intervals between radiation fractions.When 1-Gy doses were given every 1 or 2 hours there wasincreased survival compared to HDR and survival levelswere about the same as for LDR irradiation in Fig. 2. Whenhyperthermia was given there was extensive radiosensitiza-tion and the curves approached the HDR curve. The TERvalues are given in Table 1.

Similar experiments as shown in Fig. 3 for the 1-Gy dosefractions were repeated for dose fractions of 1.5 and 2 Gy.Incubation at 37 and 41°C was done between the fractions.The data at the 0.1 survival level was used to calculate theTERs which are shown in Table 1. The TERs are alsoshown in Figure 4 to demonstrate the relationship between

Fig. 1. The survival of glioma cells exposed to 41°C and 43°Cheating is shown.

186 I. J. Radiation Oncology● Biology ● Physics Volume 44, Number 1, 1999

TER and average dose rate in all experiments including theLDR and the PSLDR experimental data. These data show atrend that TER increases as the average dose rate decreases.

DISCUSSION

For the continuous LDR the dose rate of LDR1 is approx-imately 0.5 Gy per hour. For the PSLDR this would besimulated by 1 Gy/2 h, 1.5 Gy/3 h and 2 Gy/4 h. From Table1 it can be seen that 41°C continuous heating during LDR1

gives a TER of 1.51. For the PSLDR protocols simulating0.5 Gy/h the TERs range from 1.41–1.58. Thus, these datashow that the effect of 41°C during irradiation is equal forLDR and PSLDR. The data also show that as higher dose

Fig. 2. The response of glioma cells is shown for HDR and twoLDR irradiations as well as for mild hyperthermia combined withLDR. Note: LDR1 5 0.81 cGy/min and LDR2 5 0.41 cGy/min.

Fig. 3. The responses of glioma cells are shown for HDR irradi-ation, and two 1-Gy fractionation regimens with and withoutconcurrent hyperthermia at 41°C.

Fig. 4. TER versus dose rate are shown. The data were taken fromTable 1.

Table 1. Thermal enhancement ratios 10%

Treatment1 TER2(SEM)

Sparing effect

At 37°C3 41°C4

LDR1 1.516 .009 1.83 1.22LDR2 2.016 0.11 2.33 1.161 Gy/1 h 1.456 0.10 1.59 1.091 Gy/2 h 1.586 0.11 1.86 1.181.5 Gy/0.5 h 1.056 0.11 1.24 1.131.5 Gy/1 h 1.176 0.11 1.32 1.131.5 Gy/2 h 1.156 0.08 1.35 1.171.5 Gy/3 h 1.416 0.12 1.66 1.172 Gy/0.5 h 1.056 0.09 1.27 1.042 Gy/1 h 1.216 0.08 1.32 1.092 Gy/2 h 1.326 0.11 1.48 1.112 Gy/3 h 1.276 0.10 1.25 1.162 Gy/4 h 1.516 0.10 1.51 1.16

1. LDR1 and LDR2 are continuous LDR irradiation at 0.88 and0.41 cGy/min respectively.

2. TER5dose for radiation alone

dose for hyperthermia1 irradiationat the 10% survival level

3. Sparing effect at 37°C

5dose for LDR or fractionated irradiation

dose for HDR

at the 10% survival level

4. Sparing effect at 41°C5

dose for LDR or fractionatedirradiation plus hyperthermia

dose for HDRat the 10% survival level

187Concurrent mild hyperthermia for LDR irradiation● G. P. RAAPHORST et al.

rates are simulated for PSLDR, the TERs decline. Forexample in Table 1 TERs range from 1.1–1.41 for 1.5Gy/0.5 h–1.5 Gy/3 h and from 1.21–1.51 for 2 Gy/0.5 h–2Gy/4 h. This trend is further emphasized when all the dataare combined in Fig. 4. There is a continuous increase inTER as the dose rate declines from 4 Gy/h to 0.25 Gy/h.

Earlier studies by Armouret al. (24) showed that in 9Lgliosarcoma cells, 41°C heating during the entire time toadminister 25 Gy in 5 fractions was the most effectiveradiosensitizing protocol and eliminated any fractionationeffects. Shorter heating times before each fraction resultedin reduced radiosensitization and also some thermotoleranceeffects. Our data using human tumor cells support the dataof Armour et al. (24). We have shown that radiosensitiza-tion is high when 41°C hyperthermia is applied continu-ously during the entire course of pulsed radiation. In addi-

tion, our results show that if the average dose rate duringpulsed irradiation is equivalent to the continuous LDR,thermoradiosensitization is about the same. These resultsmay indicate that in both protocols hyperthermia reduces(inhibits) repair of SLDR to the same extent. Table 1 showsabout the same sparing effect at 37°C for LDR1, 1 Gy/2 h,1.5 Gy/3 h, and 2 Gy/4 h which ranges from 1.51–1.86.When 41°C was given concurrently with these protocolssparing was reduced to a range of 1.16–1.22.

Thus our data show that in human glioma cells mildhyperthermia during pulsed HDR protocols can be as effec-tive as during LDR. This result suggests that for PSLDRclinical treatments concurrent mild hyperthermia may be aneffective adjuvant treatment and that when combined withhyperthermia the protocols using the lowest average doserate may be more effective.

REFERENCES

1. Li GC, Evans RG, Hahn GM. Modification and inhibition ofrepair of potentially lethal X-ray damage by hyperthermia.Radiat Res1976;67:491–501.

2. Murthy AK, Harris JR, Belli JA. Hyperthermia and radiationresponse of plateau phase cells.Radiat Res1977;70:241–241.

3. Raaphorst GP, Azzam EI, Feeley MM. Potentially lethal ra-diation damage repair and its inhibition by hyperthermia innormal hamster cells, mouse cells and transformed mousecells.Radiat Res1988;113:171–182.

4. Raaphorst GP, Freeman ML, Dewey WC. Radiosensitivityand recovery from radiation damage in cultured CHO cellsexposed to hyperthermia at 42.5°C or 45.5°C.Radiat Res1979;79:390–402.

5. Raaphorst GP. Recovery of sublethal radiation damage and itsinhibition by hyperthermia in normal and transformed mousecells. Int J Radiat Oncol1992;20:1035–1041.

6. Ben-Hur E, Elkind E, Bronk BV. Thermally enhanced radio-response of cultured Chinese hamster cells. Inhibition of re-pair of sublethal damage and enhancement of lethal damage.Radiat Res1974;58:38–51.

7. Raaphorst GP, Mao JP, Ng CE. Thermotolerance in humanglioma cells.Int. J. Hypertherm1995;11:523–529.

8. Armour EP, Wang Z, Corry PM,et al. Sensitization of rat 9Lgliosarcoma cells to low dose rate irradiation by long duration41°C hyperthermia.Cancer Res1991;51:3088–3095.

9. Spiro IJ, McPherson S, Cook JA,et al. Sensitization of lowdose-rate irradiation by nonlethal hyperthermia.Radiat Res1991;127:111–114.

10. Ling CC, Robinson E. Moderate hyperthermia and low doserate irradiation.Radiat Res1988;114:379–384.

11. Raaphorst GP, Heller D, Bussey A,et al. Thermal radiosen-sitization by 41°C hyperthermia during low dose-rate irradia-tion in human normal and tumour cell lines.Int J Hypertherm1993;10:263–270.

12. Raaphorst GP. Thermal radiosensitization in vitro. In: UranoM, Douple E, editors. Hyperthermia and oncology. Vol. 2.VSP Utrecht; 1989. p. 17–51.

13. Spiro IJ, Denman DL, Dewey WC. Effect of hyperthermia onCHO DNA polymerasesa and b. Radiat Res1982;89:134–149.

14. Mivechi N, Dewey WC. Effect of glycerol and low pH on

heat-induced cell killing and loss of cellular DNA polymeraseactivities in Chinese hamster ovary cells.Radiat Res1984;99:352–362.

15. Dikomey E, Jung H. Correlation between polymeraseb ac-tivity and thermal radiosensitization in Chinese hamster ovarycells.Recent Results Cancer Res1988;109:35–41.

16. Mivechi NF, Miyachi H, Scanlon KJ. Heat radiosensitizationand the level of DNA polymerasesa andb of human colony-forming macrophage and myeloid leukemias sensitive andresistant to chemotherapeutic agents.Cancer Res1990;50:2044–2048.

17. Raaphorst GP, Feeley MM, Chu GL,et al. A comparison ofhyperthermia enhancement of radiation sensitivity and DNApolymerase inactivation in human glioma cells.Radiat Res1993;134:331–336.

18. Wang Z, Armour EP, Corry PM,et al. Elimination of dose-rate effects by mild hyperthermia.Int J Radiat Oncol1992;24:965–973.

19. Martinez A, Gersten D, Leslie J,et al. Interstitial thermo-brachytherapy with continuous mild hyperthermia for thetreatment of locally advanced or recurring pelvic malignan-cies. Proc 41st Cong Radiat Res; 1993. S-12-4, 122.

20. Nussbaum GH, Garcia DM, Fathman AE,et al.Hyperthermictemperature distributions produced in treatment of prostaticand intracranial malignancies with conductive interstitial hy-perthermia. Proc 41st Cong Radiat Res; 1993. P-04-14, 24.

21. Swift PS, Purser P, Roberts LW,et al. Pulsed low dose ratebrachytherapy for pelvic malignancies.Int J Radiat Oncol1997;37:811–817.

22. Brenner DJ, Hall EJ, Pehrson GR,et al.Quantitative compar-isons of continuous and pulsed low dose rate regiments in amodel late-effect system.Int J Radiat Oncol1996;34:905–910.

23. Visser AG, van den Aardweg GJMJ, Levendag PC. Pulseddose rate and fractionated high dose rate brachytherapy:Choice of brachytherapy schedules to replace low dose ratetreatments.Int J Radiat Oncol1996;34:497–505.

24. Armour EP, Wang W, Corry PM,et al. Hyperthermic en-hancement of high dose-rate irradiation in 9L gliosarcomacells. Int J. Radiat Oncol1993;36:171–177.

188 I. J. Radiation Oncology● Biology ● Physics Volume 44, Number 1, 1999