Inr J Hudrarwn Onw/o~y BI,I/ Phn Vol. Pnnted I” the IJ.S.A. All nghts reserved.
19. pp. 941-943 Copyright
0360.3016190 $3.00 + .oO @ 1990 Pergamon Press plc
0 Original Contribution RADIATION SARCOMA V.
RESPONSE IN 10 HIGH-GRADE HUMAN SOFT TISSUE XENOGRAFTS TO PHOTONS AND FAST NEUTRONS
BUDACH,
M.D., M. M.
MOLLS,
Dept. of Radiooncology.
STUSCHKE, M.D.
AND
M.D., H.
W.
SACK,
University of Essen, Hufelandstr.
BUDACH,
M.D.,
M.D.
55. 4300 Essen- I, FRG
From a panel of 48 human soft tissue sarcomas growing as permanent xenografts, 10 tumor lines (five leiomyosarcomas, three malignant fibrous histiocytomas, two neurofibrosarcomas) have been selected to determine the radiation response to photons and fast neutrons. Using the specific growth delay (SGD) as an end-point, considerable variability of inherent radiosensitivity was observed. Isoeffective radiation doses varied by a factor of 27 for photons and of 9.4 for neutrons at a specific growth delay level of 0.5. The heterogeneity of the relative biological effectiveness (RBE) at this specific growth delay-level differed by a factor of 8. Relative biological effectiveness values for clamped tumors exceeded those of the normal tissues (RBE x 3) in 6 out of 10 tumor lines. Assuming a ratio of 0.5 for oxygen enhancement ratio-values of neutrons and photons, a therapeutic gain for neutrons existed in 4 out of 10 tumor lines under oxic conditions. No correlation between volume doubling times and relative biological effectiveness was seen. Soft tissue sarcomas,
Xenografts, Radiation response to photons and neutrons, Relative biological effectiveness.
INTRODUCTION
from biopsy was aneuploid in eight and diploid in two cases. The mice underwent a whole-body gamma-irradiation with 5 Gy “Co to suppress residual immunity. thus preventing host versus graft reactions for immunogenic tumors. Groups of lo-12 male mice were stratified according to tumor size (minimum tumor volume: 150 ~1) and randomly allocated to treatment and control groups. Single doses in the range of 6-40 Gy (60Co) and 1.5-10 Gy (5.8 MeV neutrons) were used to generate dose-response curves. The tumors were clamped 10 min prior to and during irradiation to minimize the influence of the oxygenation status of the tumors. Anesthetised mice were fixed in specially developed radiation devices for gammaand n-irradiation, which allowed a homogeneous dose distribution within the tumors (? 10%). The dose rate of the Co-source was 2.9 Gy/min and for the cyclotron-generated d(14) + Be neutrons (mean energy 5.8 MV), 0.5 Gy/min. The endpoint of these studies was the specific growth delay (SGD), which was calculated from the growth times to twice the treatment tumor volume and the doubling times of the tumor lines in the controls. Doses required for SGD-levels of 0.5, 1.0, and 2.0 were obtained from the dose-response curves by linear-quadratic regression. The follow-up period was 3 months. The SGD endpoint was used to account for the different doubling times of the tumor lines (2.8- 13.1 days).
Radiobiological parameters such as radiosensitivity, repair capacity, proliferation, oxygenation status and stromal, and vascular components of a tumor have an important impact on the radio-responsiveness of human tumors. Since actual tumor therapy has reached the limits of the use of standardized treatment schedules in many respects, new approaches with more individualized therapeutic strategies according to well-defined prognostic criteria are required. A certain variability of these parameters is a prerequisite for individualization. To estimate the degree of intertumoral heterogeneity of the radiation response to photons and fast neutrons, a panel of human xenografted soft tissue sarcomas was used.
METHODS
AND
MATERIALS
From a pane1 of 48 established soft tissue sarcomas in nude mice (NMRI-strain), 10 were chosen for the experiments. Previous tumor characterization by means of histomorphology, volume doubling times, flow cytometry, and (iso-)enzyme analysis confirmed the human origin of the tumor lines. The median volume doubling times (DT) were in the range of 2.8 to 13.1 days: the DNA-index
Accepted for publication
Reprint requests to: Volker Budach. M.D., This work was supported by Grant Deutsche Forschungsgemeinschaft.
B2, SFB
102 of the 941
26 April 1990.
942
I. J. Radiation
Oncology 0 Biology 0 Physics
October 1990. Volume 19, Number 4
RESULTS IO-
The SGD of the different tumor lines differed considerably. For a SGD of 0.5 and 1.O, single 6oCo doses between 1.0-27.0 Gy and 1.8-31.4 Gy, respectively, were needed for the different tumor lines. For fast neutrons, 0.7-6.6 Gy and 1.3-7.2 Gy, respectively, were necessary. The ratio between the largest and smallest isoeffective doses at a SGD of 0.5 and 1.O for tumors of this panel was taken as a measure for heterogeneity in radioresponsiveness. Values of 27 versus 17.4for gamma-irradiation and 9.4 versus 5.5 for fast neutrons were obtained. The relative biological effectiveness (RBE) values at SGD-levels of 0.5, 1.O, and 2.0 are shown in Table 1. They decreased with increasing effect level in 6/ 10 cases, increased in three cases, and remained stable in one case. Highest RBE-values were obtained at an SGD of 0.5. For a comparison of these values with oxic values from the literature, the RBE-values can be corrected using a ratio OER,,,l,./ 30, 1979). No correlation GERrhot. = 0.5 (ICRU-Report was observed between tumor doubling time and the corresponding RBE values at a SGD of 1.0 (r, = 0.2, not significant) [Fig. 11. A comparison of the SGD after 24 Gy of Cobalt gamma-irradiation and 8.0 Gy of fast neutrons showed a rank correlation coefficient of rs = 0.87 (p < 0.005) [Fig. 21. The fact that the ranking ofthe tumor lines according to photon- or neutron sensitivity is not significantly different shows the predominance of the great variation in radioresponsiveness over RBE-values.
:
u 5.
Fig. I. Scatter plot of RBE-values f SEM of a SGD-level of I .O under hypoxic conditions versus tumor doubling times f SEM (DT) of the IO soft tissue sarcoma lines. No significant rank correlation coefficient (v, = 0.2) was observed.
Gy of “Co as an endpoint. From these data it seems useful to look for predictive assays to identify the more sensitive tumors. which will profit from radiotherapy. The adhesive tumor cell culture system and the MTT-assay are promising (1, 4, 5). The greater variation in doses needed for an SGD of 0.5 and 1.0 observed for photons compared with neutrons can be explained by the influence of the pterms, which are responsible for the curviness of the un-
DISCUSSION This panel of human soft tissue sarcoma xenografts revealed large differences in radioresponsiveness, which might be the most important factor for the clinical outcome. Nevertheless 9 of the 10 tumor lines were more resistant than the two squamous cell xenograft lines described by Lindenberger et al. (6) using a SGD after 24
Table 1. RBE-values of 10 human soft tissue sarcoma neutrons at SGD-levels of 0.5, I .O. and 2.0
lines for
Effect levels [specific growth delay] Tumor
line
EL5 EL6 EL7 EL8 EL9 EF8 EF9 EFIO EN3 EN1 * Extrapolated
0.5 6.0 6.6 12.7 I .6 3.3 2.8 10.0 2.4 2.2 10.6 value.
f * -t + + + + + + +
I.0 2.3 1.4 7.0 0.3 3.3 0.9 6.5 0.8 0.5 3.7
5.0 5.0 6.3 1.8 2.1 3.0 6.8 2.8 2.0 6.6
f f -t * f f ? f t f
1.9 0.9 4.1 0.3 1.3 0.6 4.5* 0.2 0.4 2.3
2.0 4.4 f 1.4 4.0 + 0.6 ti 2.0 t 0.3 2.0 f 0.8 3.3 f 0.6 (1 3.2 t- 0.4 2.3 +- 0.4 4.7 f 0.4
0 epec.
I.0
2.0
c,,-owth
delay
3,O
4.0
af+er24Gv=co
Fig. 2. Plot of mean SGD-values + SEM after 8 Gy of neutrons versus SGD-values after 24 Gy of “?Yo for 9 out of 10 tumor lines. The remaining tumor line. EL9. had a SGD of 18.4 ? 4.6 and 13.7 -t I .5 after 8 Gy of neutrons and 24 Gy of “‘Co, which is outside the scale of the plot. A highly significant positive rank correlation coefficient (r, = 0.87. p < 0.005) was observed.
Radiation
response
in 10 high grade soft tissue : sarcoma
derlying cellular survival curves according to the linearquadratic model. In the analysis of our data it was assumed that isoeffective doses of neutrons and photons produce the same level of cell kill in the xenografts. However, survival curve parameters as the LY-and p-values are not measurable directly by the regrowth delay assay, because cell loss and accelerated proliferation after irradiation might cause a non-linear relation between 1n-survival and induced regrowth delay. RBE-values greater than three in
xenografts
0 V. BUDACH
et al.
943
4 out of 10 tumor lines indicate a potential benefit of fast neutron therapy for at least a subgroup of soft tissue sarcomas. Similar results were observed for experimental and in situ tumors. RBE-values for fast neutrons ranged from 1.5-6.0 (2, 3). No correlation between tumor volume doubling times and RBE-values was found. Therefore, growth kinetics in unirradiated xenografts may not be a good predictor for the therapeutic gain of neutrons in comparison to photons.
REFERENCES Baker, F. L.; Spitzer, G.; Ajani. J. A.: Brock. W. A.; Lukeman, J.; Pathak, S.: Tomasovic, B.; Thielvoldt, D.: Williams. M.; Vines, C.: Tofilon. P. Drug and radiation sensitivity measurements of successful primary monolayer culturing of human tumor cells using cell-adhesive matrix and supplemented medium. Cancer Res. 46: 1263- 1274; 1986. Barendsen. G. W.: Broerse, J. J. Differences in radiosensitivity of cells from various types of experimental tumors in relation to the RBE of 15 MeV neutrons. Int. J. Radiat. Oncol. Biol. Phys. 32 11-2 14; 1977. Batterman, J. J.; Breur, K.: Hart, A. A. M.: Perpezeel, H. A. Observations on pulmonary metastases in patients after single doses and multiple fractions of fast neutrons
and Cobalt-60 gamma rays. Eur. J. Cancer 17:3033-3036; 1981. 4. Cole, S. P. C. Rapid chemosensitivity testing of human lung tumour cells using the MTT assay. Cancer Chemother. Pharmacol. 17:259-264; 1986. 5. Denizot. F.; Lang. R. Rapid calorimetric assay for cell growth and survival. Modifications to the tetrazolium dye procedure giving improved sensitivity and reliability. J. Immunol. Methods 89:27 l-275; 1986. 6. Lindenberger, J. H.; Hermeking, H.; Kummermehr, J.: Denekamp, J. Response of human tumour xenografts to fractionated X-irradiation. Radiother. Oncol. 6: 15-28: 1986.
19. pp. 941-943 Copyright
0360.3016190 $3.00 + .oO @ 1990 Pergamon Press plc
0 Original Contribution RADIATION SARCOMA V.
RESPONSE IN 10 HIGH-GRADE HUMAN SOFT TISSUE XENOGRAFTS TO PHOTONS AND FAST NEUTRONS
BUDACH,
M.D., M. M.
MOLLS,
Dept. of Radiooncology.
STUSCHKE, M.D.
AND
M.D., H.
W.
SACK,
University of Essen, Hufelandstr.
BUDACH,
M.D.,
M.D.
55. 4300 Essen- I, FRG
From a panel of 48 human soft tissue sarcomas growing as permanent xenografts, 10 tumor lines (five leiomyosarcomas, three malignant fibrous histiocytomas, two neurofibrosarcomas) have been selected to determine the radiation response to photons and fast neutrons. Using the specific growth delay (SGD) as an end-point, considerable variability of inherent radiosensitivity was observed. Isoeffective radiation doses varied by a factor of 27 for photons and of 9.4 for neutrons at a specific growth delay level of 0.5. The heterogeneity of the relative biological effectiveness (RBE) at this specific growth delay-level differed by a factor of 8. Relative biological effectiveness values for clamped tumors exceeded those of the normal tissues (RBE x 3) in 6 out of 10 tumor lines. Assuming a ratio of 0.5 for oxygen enhancement ratio-values of neutrons and photons, a therapeutic gain for neutrons existed in 4 out of 10 tumor lines under oxic conditions. No correlation between volume doubling times and relative biological effectiveness was seen. Soft tissue sarcomas,
Xenografts, Radiation response to photons and neutrons, Relative biological effectiveness.
INTRODUCTION
from biopsy was aneuploid in eight and diploid in two cases. The mice underwent a whole-body gamma-irradiation with 5 Gy “Co to suppress residual immunity. thus preventing host versus graft reactions for immunogenic tumors. Groups of lo-12 male mice were stratified according to tumor size (minimum tumor volume: 150 ~1) and randomly allocated to treatment and control groups. Single doses in the range of 6-40 Gy (60Co) and 1.5-10 Gy (5.8 MeV neutrons) were used to generate dose-response curves. The tumors were clamped 10 min prior to and during irradiation to minimize the influence of the oxygenation status of the tumors. Anesthetised mice were fixed in specially developed radiation devices for gammaand n-irradiation, which allowed a homogeneous dose distribution within the tumors (? 10%). The dose rate of the Co-source was 2.9 Gy/min and for the cyclotron-generated d(14) + Be neutrons (mean energy 5.8 MV), 0.5 Gy/min. The endpoint of these studies was the specific growth delay (SGD), which was calculated from the growth times to twice the treatment tumor volume and the doubling times of the tumor lines in the controls. Doses required for SGD-levels of 0.5, 1.0, and 2.0 were obtained from the dose-response curves by linear-quadratic regression. The follow-up period was 3 months. The SGD endpoint was used to account for the different doubling times of the tumor lines (2.8- 13.1 days).
Radiobiological parameters such as radiosensitivity, repair capacity, proliferation, oxygenation status and stromal, and vascular components of a tumor have an important impact on the radio-responsiveness of human tumors. Since actual tumor therapy has reached the limits of the use of standardized treatment schedules in many respects, new approaches with more individualized therapeutic strategies according to well-defined prognostic criteria are required. A certain variability of these parameters is a prerequisite for individualization. To estimate the degree of intertumoral heterogeneity of the radiation response to photons and fast neutrons, a panel of human xenografted soft tissue sarcomas was used.
METHODS
AND
MATERIALS
From a pane1 of 48 established soft tissue sarcomas in nude mice (NMRI-strain), 10 were chosen for the experiments. Previous tumor characterization by means of histomorphology, volume doubling times, flow cytometry, and (iso-)enzyme analysis confirmed the human origin of the tumor lines. The median volume doubling times (DT) were in the range of 2.8 to 13.1 days: the DNA-index
Accepted for publication
Reprint requests to: Volker Budach. M.D., This work was supported by Grant Deutsche Forschungsgemeinschaft.
B2, SFB
102 of the 941
26 April 1990.
942
I. J. Radiation
Oncology 0 Biology 0 Physics
October 1990. Volume 19, Number 4
RESULTS IO-
The SGD of the different tumor lines differed considerably. For a SGD of 0.5 and 1.O, single 6oCo doses between 1.0-27.0 Gy and 1.8-31.4 Gy, respectively, were needed for the different tumor lines. For fast neutrons, 0.7-6.6 Gy and 1.3-7.2 Gy, respectively, were necessary. The ratio between the largest and smallest isoeffective doses at a SGD of 0.5 and 1.O for tumors of this panel was taken as a measure for heterogeneity in radioresponsiveness. Values of 27 versus 17.4for gamma-irradiation and 9.4 versus 5.5 for fast neutrons were obtained. The relative biological effectiveness (RBE) values at SGD-levels of 0.5, 1.O, and 2.0 are shown in Table 1. They decreased with increasing effect level in 6/ 10 cases, increased in three cases, and remained stable in one case. Highest RBE-values were obtained at an SGD of 0.5. For a comparison of these values with oxic values from the literature, the RBE-values can be corrected using a ratio OER,,,l,./ 30, 1979). No correlation GERrhot. = 0.5 (ICRU-Report was observed between tumor doubling time and the corresponding RBE values at a SGD of 1.0 (r, = 0.2, not significant) [Fig. 11. A comparison of the SGD after 24 Gy of Cobalt gamma-irradiation and 8.0 Gy of fast neutrons showed a rank correlation coefficient of rs = 0.87 (p < 0.005) [Fig. 21. The fact that the ranking ofthe tumor lines according to photon- or neutron sensitivity is not significantly different shows the predominance of the great variation in radioresponsiveness over RBE-values.
:
u 5.
Fig. I. Scatter plot of RBE-values f SEM of a SGD-level of I .O under hypoxic conditions versus tumor doubling times f SEM (DT) of the IO soft tissue sarcoma lines. No significant rank correlation coefficient (v, = 0.2) was observed.
Gy of “Co as an endpoint. From these data it seems useful to look for predictive assays to identify the more sensitive tumors. which will profit from radiotherapy. The adhesive tumor cell culture system and the MTT-assay are promising (1, 4, 5). The greater variation in doses needed for an SGD of 0.5 and 1.0 observed for photons compared with neutrons can be explained by the influence of the pterms, which are responsible for the curviness of the un-
DISCUSSION This panel of human soft tissue sarcoma xenografts revealed large differences in radioresponsiveness, which might be the most important factor for the clinical outcome. Nevertheless 9 of the 10 tumor lines were more resistant than the two squamous cell xenograft lines described by Lindenberger et al. (6) using a SGD after 24
Table 1. RBE-values of 10 human soft tissue sarcoma neutrons at SGD-levels of 0.5, I .O. and 2.0
lines for
Effect levels [specific growth delay] Tumor
line
EL5 EL6 EL7 EL8 EL9 EF8 EF9 EFIO EN3 EN1 * Extrapolated
0.5 6.0 6.6 12.7 I .6 3.3 2.8 10.0 2.4 2.2 10.6 value.
f * -t + + + + + + +
I.0 2.3 1.4 7.0 0.3 3.3 0.9 6.5 0.8 0.5 3.7
5.0 5.0 6.3 1.8 2.1 3.0 6.8 2.8 2.0 6.6
f f -t * f f ? f t f
1.9 0.9 4.1 0.3 1.3 0.6 4.5* 0.2 0.4 2.3
2.0 4.4 f 1.4 4.0 + 0.6 ti 2.0 t 0.3 2.0 f 0.8 3.3 f 0.6 (1 3.2 t- 0.4 2.3 +- 0.4 4.7 f 0.4
0 epec.
I.0
2.0
c,,-owth
delay
3,O
4.0
af+er24Gv=co
Fig. 2. Plot of mean SGD-values + SEM after 8 Gy of neutrons versus SGD-values after 24 Gy of “?Yo for 9 out of 10 tumor lines. The remaining tumor line. EL9. had a SGD of 18.4 ? 4.6 and 13.7 -t I .5 after 8 Gy of neutrons and 24 Gy of “‘Co, which is outside the scale of the plot. A highly significant positive rank correlation coefficient (r, = 0.87. p < 0.005) was observed.
Radiation
response
in 10 high grade soft tissue : sarcoma
derlying cellular survival curves according to the linearquadratic model. In the analysis of our data it was assumed that isoeffective doses of neutrons and photons produce the same level of cell kill in the xenografts. However, survival curve parameters as the LY-and p-values are not measurable directly by the regrowth delay assay, because cell loss and accelerated proliferation after irradiation might cause a non-linear relation between 1n-survival and induced regrowth delay. RBE-values greater than three in
xenografts
0 V. BUDACH
et al.
943
4 out of 10 tumor lines indicate a potential benefit of fast neutron therapy for at least a subgroup of soft tissue sarcomas. Similar results were observed for experimental and in situ tumors. RBE-values for fast neutrons ranged from 1.5-6.0 (2, 3). No correlation between tumor volume doubling times and RBE-values was found. Therefore, growth kinetics in unirradiated xenografts may not be a good predictor for the therapeutic gain of neutrons in comparison to photons.
REFERENCES Baker, F. L.; Spitzer, G.; Ajani. J. A.: Brock. W. A.; Lukeman, J.; Pathak, S.: Tomasovic, B.; Thielvoldt, D.: Williams. M.; Vines, C.: Tofilon. P. Drug and radiation sensitivity measurements of successful primary monolayer culturing of human tumor cells using cell-adhesive matrix and supplemented medium. Cancer Res. 46: 1263- 1274; 1986. Barendsen. G. W.: Broerse, J. J. Differences in radiosensitivity of cells from various types of experimental tumors in relation to the RBE of 15 MeV neutrons. Int. J. Radiat. Oncol. Biol. Phys. 32 11-2 14; 1977. Batterman, J. J.; Breur, K.: Hart, A. A. M.: Perpezeel, H. A. Observations on pulmonary metastases in patients after single doses and multiple fractions of fast neutrons
and Cobalt-60 gamma rays. Eur. J. Cancer 17:3033-3036; 1981. 4. Cole, S. P. C. Rapid chemosensitivity testing of human lung tumour cells using the MTT assay. Cancer Chemother. Pharmacol. 17:259-264; 1986. 5. Denizot. F.; Lang. R. Rapid calorimetric assay for cell growth and survival. Modifications to the tetrazolium dye procedure giving improved sensitivity and reliability. J. Immunol. Methods 89:27 l-275; 1986. 6. Lindenberger, J. H.; Hermeking, H.; Kummermehr, J.: Denekamp, J. Response of human tumour xenografts to fractionated X-irradiation. Radiother. Oncol. 6: 15-28: 1986.
