- Email: [email protected]
Decreased bone growth arrest in weanling rats with multiple radiation fractions per day
Inl. .I. Radiation Oncology Biol. Phys.. Vol. 15, pp. 141-145 Printed in the U.S.A. All rights reserved.
Copyright
0360-3016/88 $3.00 + .oO 0 1988 Pergamon Press plc
??Original Contribution DECREASED
BONE
GROWTH ARREST IN WEANLING RATS WITH MULTIPLE RADIATION FRACTIONS PER DAY PATRICIA
Department
J. EIFEL,
M.D.
of Radiation Therapy, The University of Texas Medical Branch, Galveston, TX 77550
The relative growth arrest caused by fractionated irradiation delivered in single or multiple daily fractions was studied in weanling rats. Twenty-two day old male rats were treated to a total dose of either 20 or 25 Gy in five consecutive days to the distal femoral and proximal tibia1 epiphyses of the right and left leg. For each dose three treatment groups were followed for longitudinal tibia1 growth as measured on serial radiographs: (a) no treatment, (b) 5 fractions in 5 days or (c) 10 fractions in 5 days. Tibia1 length was significantly greater in the legs treated with twice-daily fractions (TDF) as compared with single daily fractions (SDF) with 23% and 27% sparing of growth arrest (at 200 days) in legs treated to total doses of 20 and 25 Gy respectively (p < 0.001). This appeared to result from a continuously greater rate of growth during the first 40-50 days following TDF irradiation as compared with SDF. These data suggest that hyperftactionation may provide a means of reducing growth deficits in children when skeletal growth centers must be included in the irradiated volume. Radiation, Rat, Bone growth, Fractionation, Hyperftactionation.
have confirmed that daily doses of 2-2.5 Gy can be hyperfractionated with little loss in the probability of tumor control and with improved sparing of normal tissue, specifically skin and mucosa.‘~3916S’9~23 The aim of work described in this report was to explore the relative effects of SDF and TDF on growth arrest in weanling rats. The results indicate that significant sparing of radiation-induced epiphysial growth arrest may be achieved through the use of multiple daily fractions.
INTRODUCKION
The damaging effect of therapeutic radiation on growing bone has been an important source of morbidity and a major dose-limiting factor in the radiotherapeutic management of pediatric malignancies.“~‘7~20~2’Retrospective clinical reviews have qualitatively related the degree of growth arrest to dose, daily fraction size, and age at the time of treatment. In young children, significant growth arrest may be incurred with fractionated doses of 15 Gy and above and, in children under 1 year of age, with doses as low as 10 Gy. Concern among clinicians regarding late effects has resulted in attempts to maximize the risk-benefit ratio, often leading to therapeutic compromises. The use of “hyperfractionation”, the division of a daily dose into two or three small fractions, to improve normal tissue tolerance has been the subject of intense interest, leading to clinical and laboratory investigations. Studies with various tumor models indicate that hyperfractionation of daily doses of 2-2.5 Gy should result in little or no decrease in tumor response.2*9 In addition, the separation of these smaller fractions by an adequate period of time (4-6 hr.) may allow repair of sublethal damage to normal cells. 5-7,‘oA number of clinical studies
METHODS
AND
MATERIALS
Animals and anesthesia Weanling (22 days old), male, Sprague Dawley (Harlan Sprague Dawley) rats were used in all experiments. For irradiations and subsequent measurements, animals were anesthetized with pentobarbital sodium solution (Abbott Laboratories) at a dose of 25-30 mg/kg. As experience was gained, very few animals were lost to anesthesia. After the first day of treatment, they appeared to develop a tolerance and were usually alert within f hour of treatment. Any effect of anesthesia upon oxygenation (and possibly radiation sensitivity) was controlled by making primary comparisons between differentially treated legs in the same animal.
Supported by Grant 5P30-CA 1770 1, National Cancer Institute, National Institutes of Health, U.S.A. Presented at the Annual Meeting of the American Society of Therapeutic Radiologists; November 3-7, 1986; Los Angeles, CA.
Reprint requests to: Patricia J. Eifel, M.D., Department of Clinical Radiotherapy, M.D. Anderson Hospital and Tumor Institute, 15 15 Holcombe-97, Houston, TX 77030. Accepted for publication 1 February 1988. 141
142
I. J. Radiation Oncology 0 Biology 0 Physics
Fig. 1. Anesthetized animals were treated in lead jigs with the right or left leg and tail protruding and taped in position. Additional pieces of lead were placed to protect the tail, distal tibia and foot. Correct exposure of the knee joint was verified by palpation prior to each treatment.
Animals to be dissected for direct measurements were sacrificed with pentobarbital sodium solution ( 125 mg/kg). Irradiation procedures Anesthetized male weanling rats were irradiated in lead jigs designed to shield all but the right or left knee (distal femur and proximal tibia) (Fig. 1). Irradiation conditions were: 250 kVp X rays with 0.4 Thoreus filtration, 15 mA, at a focus-skin distance 50 cm yielding a half-value layer of 2.7 mm Cu at a dose-rate of 52 cGy/ min. Dose-rate calculations were verified with thermoluminescent dosimeters (TLD) embedded in tissue-equivalent material 4 cm thick (the approximate thickness of the weanling rat leg). Total body dose was less than 1% of the prescribed dose as measured by TLD. In fractionated experiments, animals were treated 5 sequential days. In all experiments animals were 22 days old on the first radiation day. Twice-daily fractions were given 6 hours apart. Unless otherwise specified, animals were treated in groups of eight. In most experiments, animals received the same total dose of radiation to the right and left knee with SDF delivered to one leg and TDF to the other. Some animals had only one leg treated which was compared with the opposite untreated leg to rule out any possible abscopal effects. An additional group of untreated animals were followed to establish normal growth curves. Growth curves for treated and untreated legs were found to be independent of the management of the opposite leg.
July 1988, Volume 15, Number 1
Measurements Periodic measurements of tibia1 and femoral length were made by taking diagnostic radiographs of anesthetized animals and measuring images of the irradiated bones to the nearest 0.5 mm. Although both femoral and tibia1 length were recorded and yielded qualitatively similar results in the experiments described below, tibia1 measurements were found to yield more consistent results with smaller standard deviations within groups. With careful positioning of the knee and heel against the cassette the tibia can be placed in the desirable position with excellent reproducibility. In a group of eight animals removed and repositioned on a film cassette, the variation in mean tibia1 measurements was less than kO.5 mm. The three groups of animals used to generate data shown in Figures 2 and 3 were sacrificed at the conclusion of the follow-up period (200 days of age). The legs were disarticulated at the hip, and muscles and patella dissected away from the femur and tibia. Direct measurements of tibia1 and femoral length were made with a calipers to the nearest 0.2 mm. A total of 46 legs were dissected and the mean tibia1 lengths of comparably treated legs (in most cases eight per group) were compared with radiographic measurements made just prior to sacrifice. RESULTS
The X ray dose-response relationship for single dose and fractionated radiation (5 Gy/day) using the ratio of 45 40
E
E 35 s= 30
c
,o 3 ._ I-
25
20 I
0
2s
I
60
,
I
1
76
100 126
Age
(days)
1
160
f
I
176 200
Fig. 2. Tibia1 growth curves for right or left legs treated with 25 Gy in SDF (5 Gy/day) (m) or TDF (2.5 Gy/lO fractions/5 days) (A). The solid line represents the normal growth curve for untreated legs in a separate group of eight animals. Each point represents the mean tibia1 length of right or left legs in eight animals + standard deviation (S.D.).
143
Bone growth arrest 0 P. J. EIFEL
0
25
50
75
100
12s
150
17s
200
Age (days) Fig. 3. Tibia1 growth curves for right or left legs treated with 20 Gy in SDF (4 Gy/day) (H) or TDF (2.0 Gy/lO fractions/5 days) (A) and normal growth curve (0). Data points represent the mean tibia1 lengths in eight animals (+-S.D.) except for the last two points on the growth curves of irradiated animals. These represent the mean lengths in seven animals (one animal succumbed to anesthesia on day 99).
observed tibia1 growth and normal growth at 200 days as an endpoint is shown in Figure 2. Each data point represents a single animal. In addition, a single data point represents the mean (*SD.) growth arrest from a course of hyperfractionated irradiation (2.5 Gy twice daily/5 days) in a group of eight animals. The slopes in Figure 4 are indicated only for the purposes of comparison. No attempt was made in these experiments to determine precisely the proportionate growth of the proximal and distal tibia1 epiphyses. As explained above, the distal tibia1 epiphysis was carefully shielded in all experiments allowing normal growth distally. Consequently the measured tibia1 lengths and the calculated “fraction normal growth” in Figure 4 all include a presumably constant growth fraction from the untreated distal end. In fact, the highest doses used in these experiments (20 Gy single fraction and 40 Gy in 5 Gy fractions) caused very severe, probably near complete growth arrest. The doses used in subsequent experiments were chosen to cause growth arrest of intermediate severity. Significant sparing of radiation-induced bone growth arrest was observed when a 5 day course of fractionated irradiation (5.0 Gy/day) was delivered with TDF of 2.5 Gy (Total Dose (TD) = 25 Gy) as compared with SDF (Fig. 2). At 200 days of age the mean difference in tibia1 length as measured from radiographs was 1.9 mm, representing a sparing of bone growth arrest of 2 1%. Similar results were seen with a 5 day course of 20 Gy (Fig. 3). Animals were treated in groups of 8 with right and left legs receiving the same total dose with different fraction-
ation schedules. In this way, any confounding variables which could possibly result from variations in anesthesia, nutrition, etc. should have had an identical influence upon right and left leg and, therefore, upon both treatment groups. The differences in tibia1 length of legs treated with SDF vs. those treated with TDF were highly statistically significant (p < 0.001). These data were confirmed in two additional similar experiments. The differential effect on linear bone growth observed with SDF and TDF is manifested within the first 6-8 weeks following treatment, corresponding to the rapid growth period and pubescence of the male rat (Figs. 2, 3). Following this, the growth curves appear to be parallel but never achieve the growth rate of untreated legs. The primary endpoint being studied in these experiments was linear bone growth. As expected, other acute and subacute effects were also decreased with TDF. Legs treated with multiple daily fractions consistently demonstrated less severe epilation and more complete regrowth of fur at 100 days. After 200 days of age, animals from the experiment depicted in Figures 2 and 3 were sacrificed and direct measurements were made of tibia1 and femoral length. These results confirmed those obtained from radiographic measurements, with comparable, highly significant, differences in mean tibia1 length between legs treated with SDF and TDF. (Data not shown.) In all
10
! 0
1
I
I
1
10
20
30
40
DOSE (GRAY) Fig. 4. Percent normal growth (Alength/Alength,,,,,) from treatment to age 200 days as a function of total dose delivered. (A) Single fraction of 12-20 Gy. Each point (0) represents one treated animal. (B) Fractionated irradiation (5 Gy/day). Each point represents the mean tibia1 length in four animals +_S.D. (C) A single point (mean of six animals f S.D.) representing growth arrest with TDF (2.5 Gy twice daily, total dose = 25 Gy). Do = 19.5,60 and 80 for A, B and C respectively.
144
1. J. Radiation Oncology 0 Biology 0 Physics
cases, the mean tibia1 lengths measured in these two ways (directly or from radiographs) were comparable.
DISCUSSION These data show that division of a daily dose of 4-5 Gy into two equal fractions results in dramatic improvement of linear bone growth in weanling rats. The results demonstrate conclusively that a portion of the epiphysial damage caused by ionizing radiation is recoverable during a 6 hr period. A number of clinical studies have confirmed that daily doses of 2-2.5 Gy can be hyperfractionated with little loss in the probability of tumor control and with improved sparing of normal tissue, specifically skin and mucosa.‘933’63’9*23 Norin et a1.,18Sarrazin et al.” and Ziegle? have reported the use of hyperfractionated whole-abdominal irradiation in the treatment of a small number of children with Wilms’ tumor and Burkitt’s lymphoma. These children had less acute gastrointestinal toxicity than would have been expected with standard fractionation. There has been some recent enthusiasm for the use of hyperfractionation in treatment of pediatric astrocytomas.24 However no clinical studies have been performed to evaluate the late effects, specifically bone growth effects, of hyperfractionated irradiation in children. This may reflect a reluctance to change relatively successful therapeutic approaches and to address the logistical problems of immobilizing children two or three times a day without compelling laboratory evidence that such an approach would improve therapeutic index. This study provides the first evidence of this kind. Our knowledge about the growth-sparing effect of fractionated irradiation derives from anecdotal observations and qualitative retrospective clinical reviews. These studies have suggested that the use of relatively large daily fractions results in a less satisfactory therapeutic index and has led to the use of increasingly small daily fractions in the treatment of children. Most laboratory investigations of radiation induced growth arrest have involved single or two dose experiments.438”2-15 The most extensive series of experiments was done by Hinkel.‘2-‘4 He observed the effects of single doses of irradiation on growing long bones in rats and determined that the degree of stunting was related to dose and age at the time of treatment. Histological abnormalities were appreciated at doses lower than those required for minimal growth arrest. He noted “dropout” of chondrocytes and disorganization of their normal columnar arrangement. Augmentation of mineral deposition in the matrix caused growth arrest lines. Higher doses also caused changes in the supporting structure and blood vessels;
July 1988, Volume 15, Number 1
the latter became narrow and irregular and ended short of the cartilage zone. Kember” published the results of the only study which measured the effect of fractionated doses on growing cartilage. Using an in vivo assay in which clones of recovering cartilage cells were counted, he described a survival curve with a Do of 165 rad. When 2200 rad was fractionated into 1600 rad and 600 rad fractions, a maximal recovery ratio of 6 at 6 hours was found. This suggested an extrapolation number of 6. Although in viva tissue interactions in the complex, multicomponent structure of cartilage may have influenced the observed survival curve parameters for cartilage cells, this work showed that two-dose fractionation resulted in decreased damage to this cellular compartment. Since the work of Hinke112-‘4 and Kember,” little has been added to our understanding of the mechanism(s) of radiation-induced growth arrest. The data presented demonstrate that a portion of the damage caused by a dose of ionizing radiation to actively growing epiphysial bone is repaired during the 6 hour period between fractions. This implies that at least one of the several important cellular compartments of this complex biological system sustains some component of sublethal injury. The slopes and extrapolation numbers of dose response curves shown in Figure 4 can not be used to draw any conclusion about the inherent radiosensitivity or repair capacity of specific tissue compartments (i.e., endothelial, chondrocyte, osteoblast); however, they do demonstrate substantial recovery of the growth plate with fractionated and twice-daily fractionated irradiation. In fact, very little is known about the mechanisms of expression of radiation damage in growing bone. Histopathological observations described by Hinkel14 suggest that important damage is sustained by vascular, chondrocyte and matrix compartments. However, his work was confined to experiments with large single fractions the effects of which are probably qualitatively as well as quantitatively different than those seen with fractionated irradiation. Work is currently in progress to explore the comparative histopathology in epiphysial bone exposed to single dose, fractionated and hyperfractionated irradiation. The observation that hyperfractionation of therapeutic radiation results in less damage to growing bone than single daily fractions has obvious important implications for the radiation therapy of children with neoplastic disease. Although the fraction sizes used in these initial experiments were relatively large, it is likely that recovery of sublethal injury and similar tissue sparing will be observed with hyperfractionation of more standard daily doses. If so, the radiotherapeutic management of children with cancer might be achieved with substantially less late morbidity than is currently experienced.
REFERENCES 1.
Ang, ILK., Van der Schueren, E., Notter, G., Horiot, J.C., Chenal, Fauchon, F., Raps, J., Peperzeel, H., Goffin, J.C.,
M., Glabbeke, M.V.: Split course multiple daily fractionated radiotherapy schedule combined with mis-
Vessiere,
Bone growth arrest 0 P J. EIFEL
onidazole for the management ofgrade III and IV ghomas. Int. J. Radiat. Oncol. Biol. Phys. 8: 1657-1664, 1982. 2.Arcangeli, G., Mauro, R., Nervi, C.: Multiple daily fractionation in radiotherapy: Biological rationale and preliminary clinical experiences. Eur. J. Cancer 15: 1077-1083, 1979. 3.Barker, J.L., Montague, E.D., Peters, L.J.: Clinical experience with irradiation of inflammatory carcinoma of the breast with and without elective chemotherapy. C’ancer45: 625-629,198O. 4.Blackburn, B.A., Wells, B.A.: Radiation damage to growing bone: the effect of X-ray doses of 100 to 1,000 r on mouse tibia and knee joint. Br. J. Radiol. 36: 505-513, 1963. 5.Collis, C.H., Down, J.D.: The effect ofdose rate and multiple fractions per day on radiation-induced lung damage in mice. Br. J. Radiol. 57: 1037-1039, 1984. 6.Douglas, B.G.: Superfractionation: Its rationale and anticipated benefits. Int. J. Radiat. Oncol. Biol. Phys. 8: 11431 153, 1982. 7.Elkind, M.M., Sutton, H.: Radiation response of mammalian cells grown in culture. 1. Repair of X-ray damage in surviving Chinese hamster cells. Radiat. Rex 13: 556-593, 1960. 8.Engstrom, H., Magnusson, B.C., Turesson, I.: Effect of 50 kv irradiation on enzyme activities of growing rat bone. Acta Radiol. Oncol. 22: 65-70, 1983. 9.Fowler, J.F., Sheldon, P.W., Harris, S.R., Hill, S.A., Ayres, SE.: Relative effectiveness of 12-hourly fractionation and a non-uniform x-ray schedule in the optimum fractionation of C3H mouse mammary tumours. Br. J. Radiol. 48: 581-589, 1975. 10. Giri, P.G.S., Kimler, B.F., Giri, U.P., Cox, G.G., Reddy, E.K.: Comparison of single, fractionated and hyperfractionated irradiation on the develoDment of normal tissue damage in rat lung. Int. J. Radial Oncol. Biol. Phys. 11: 527-534. 1985. 11. Gonzalez, D.G., Breur, K.: Clinical data from irradiated growing long bones in children. Int. J. Radiat. Oncol. Biol. Phys. 9: 841-846, 1983. 12. Hinkel, C.L.: The effect of roentgen rays upon the growing long bones of albino rats. I. Quantitative studies of the growth limitation following irradiation. Am. J. Roentgenol. 47: 439-457, 1942.
145
13. Hinkel, CL.: The effect of irradiation upon the composition and vascular&y of growing rat bones. Am. J. Roentgenol. 50: 5 16-526, 1943. 14. Hinkel, C.L.: The effect of roentgen rays upon the growing long bones of albino rats. II. Histopathological changes involving endochondral growth centers. Am. J. Roentgenol. 49: 321-348, 1943. 15. Kember, N.F.: Cell survival and radiation damage in growth cartilage. Br. J. Radiol. 40: 496-505, 1967. 16. Kotalik, J.F.: Multiple daily fractions in radiotherapy. Ca. Treat. Rev. 8: 127-146, 1981. 17. Neuhauser, E.B.D., Wittenborg, M.H., Berman, C.Z., Cohen, J.: Irradiation effects of roentgen therapy on growing spine. Radiol. 59: 637-650, 1952. 18. Norin, T., Clifford, P., Einhorn, J., Einhorn, N., Johansson, B., Klein, G., Onyango, J., DeSchryver, A., Walstam, R.: Conventional and Superfractionated radiation therapy in Burkitt’s lymphoma. Acta Radiol. 10:545-557, 197 1. 19. Parsons, J.T., Cassisi, N.J., Million, R.R.: Results oftwicea-day irradiation of squamous cell carcinomas of the head and neck. Int. J. Radiat. Oncol. Biol. Phys. 10:204 l-205 1, 1984. 20. Probert, J.C., Parker, B.R.: The effects of radiation therapy on bonegrowth. Radiol. 114: 155-162, 1975. 21. Rubin, P., Duthie, R.B., Young, L.W.: The significance of scoliosis in post-irradiated Wilms’ tumor and neuroblastoma. Radiol. 79: 539-558, 1962. 22. Sarrazin, D.: Hyperfractionated irradiation of the whole abdomen in five children (Abstr.). Proceedings of the 14th Annual SIOP Meeting. Berne, Switzerland, International Society of Pediatric Oncology. 1982, p. A33. 23. Shukovsky, L.J., Fletcher, G.H., Montague, E.D., Withers, H.R.: Experience with twice-a-day fractionation in clinical radiotherapy. Am. J. Roentgenol. 126:155-162, 1976. 24. Wara, W.M., Edwards, M.S.B., Levin, V.A., Heidman, R. Urtasun, R., Diaz, R.E., Silver, P., Sposto, R.: A new treatment regimen for brainstem glioma: A pilot study of the brain tumor research center and children’s cancer study group (Abstr.). Int. J. Radiat. Oncol. Biol. Phys. 12: 143144,1986. 25. Ziegler, J.L.: Management of Burkitt’s lymphoma: An update. Ca. Treat. Rev. 6: 95-105, 1979.
Copyright
0360-3016/88 $3.00 + .oO 0 1988 Pergamon Press plc
??Original Contribution DECREASED
BONE
GROWTH ARREST IN WEANLING RATS WITH MULTIPLE RADIATION FRACTIONS PER DAY PATRICIA
Department
J. EIFEL,
M.D.
of Radiation Therapy, The University of Texas Medical Branch, Galveston, TX 77550
The relative growth arrest caused by fractionated irradiation delivered in single or multiple daily fractions was studied in weanling rats. Twenty-two day old male rats were treated to a total dose of either 20 or 25 Gy in five consecutive days to the distal femoral and proximal tibia1 epiphyses of the right and left leg. For each dose three treatment groups were followed for longitudinal tibia1 growth as measured on serial radiographs: (a) no treatment, (b) 5 fractions in 5 days or (c) 10 fractions in 5 days. Tibia1 length was significantly greater in the legs treated with twice-daily fractions (TDF) as compared with single daily fractions (SDF) with 23% and 27% sparing of growth arrest (at 200 days) in legs treated to total doses of 20 and 25 Gy respectively (p < 0.001). This appeared to result from a continuously greater rate of growth during the first 40-50 days following TDF irradiation as compared with SDF. These data suggest that hyperftactionation may provide a means of reducing growth deficits in children when skeletal growth centers must be included in the irradiated volume. Radiation, Rat, Bone growth, Fractionation, Hyperftactionation.
have confirmed that daily doses of 2-2.5 Gy can be hyperfractionated with little loss in the probability of tumor control and with improved sparing of normal tissue, specifically skin and mucosa.‘~3916S’9~23 The aim of work described in this report was to explore the relative effects of SDF and TDF on growth arrest in weanling rats. The results indicate that significant sparing of radiation-induced epiphysial growth arrest may be achieved through the use of multiple daily fractions.
INTRODUCKION
The damaging effect of therapeutic radiation on growing bone has been an important source of morbidity and a major dose-limiting factor in the radiotherapeutic management of pediatric malignancies.“~‘7~20~2’Retrospective clinical reviews have qualitatively related the degree of growth arrest to dose, daily fraction size, and age at the time of treatment. In young children, significant growth arrest may be incurred with fractionated doses of 15 Gy and above and, in children under 1 year of age, with doses as low as 10 Gy. Concern among clinicians regarding late effects has resulted in attempts to maximize the risk-benefit ratio, often leading to therapeutic compromises. The use of “hyperfractionation”, the division of a daily dose into two or three small fractions, to improve normal tissue tolerance has been the subject of intense interest, leading to clinical and laboratory investigations. Studies with various tumor models indicate that hyperfractionation of daily doses of 2-2.5 Gy should result in little or no decrease in tumor response.2*9 In addition, the separation of these smaller fractions by an adequate period of time (4-6 hr.) may allow repair of sublethal damage to normal cells. 5-7,‘oA number of clinical studies
METHODS
AND
MATERIALS
Animals and anesthesia Weanling (22 days old), male, Sprague Dawley (Harlan Sprague Dawley) rats were used in all experiments. For irradiations and subsequent measurements, animals were anesthetized with pentobarbital sodium solution (Abbott Laboratories) at a dose of 25-30 mg/kg. As experience was gained, very few animals were lost to anesthesia. After the first day of treatment, they appeared to develop a tolerance and were usually alert within f hour of treatment. Any effect of anesthesia upon oxygenation (and possibly radiation sensitivity) was controlled by making primary comparisons between differentially treated legs in the same animal.
Supported by Grant 5P30-CA 1770 1, National Cancer Institute, National Institutes of Health, U.S.A. Presented at the Annual Meeting of the American Society of Therapeutic Radiologists; November 3-7, 1986; Los Angeles, CA.
Reprint requests to: Patricia J. Eifel, M.D., Department of Clinical Radiotherapy, M.D. Anderson Hospital and Tumor Institute, 15 15 Holcombe-97, Houston, TX 77030. Accepted for publication 1 February 1988. 141
142
I. J. Radiation Oncology 0 Biology 0 Physics
Fig. 1. Anesthetized animals were treated in lead jigs with the right or left leg and tail protruding and taped in position. Additional pieces of lead were placed to protect the tail, distal tibia and foot. Correct exposure of the knee joint was verified by palpation prior to each treatment.
Animals to be dissected for direct measurements were sacrificed with pentobarbital sodium solution ( 125 mg/kg). Irradiation procedures Anesthetized male weanling rats were irradiated in lead jigs designed to shield all but the right or left knee (distal femur and proximal tibia) (Fig. 1). Irradiation conditions were: 250 kVp X rays with 0.4 Thoreus filtration, 15 mA, at a focus-skin distance 50 cm yielding a half-value layer of 2.7 mm Cu at a dose-rate of 52 cGy/ min. Dose-rate calculations were verified with thermoluminescent dosimeters (TLD) embedded in tissue-equivalent material 4 cm thick (the approximate thickness of the weanling rat leg). Total body dose was less than 1% of the prescribed dose as measured by TLD. In fractionated experiments, animals were treated 5 sequential days. In all experiments animals were 22 days old on the first radiation day. Twice-daily fractions were given 6 hours apart. Unless otherwise specified, animals were treated in groups of eight. In most experiments, animals received the same total dose of radiation to the right and left knee with SDF delivered to one leg and TDF to the other. Some animals had only one leg treated which was compared with the opposite untreated leg to rule out any possible abscopal effects. An additional group of untreated animals were followed to establish normal growth curves. Growth curves for treated and untreated legs were found to be independent of the management of the opposite leg.
July 1988, Volume 15, Number 1
Measurements Periodic measurements of tibia1 and femoral length were made by taking diagnostic radiographs of anesthetized animals and measuring images of the irradiated bones to the nearest 0.5 mm. Although both femoral and tibia1 length were recorded and yielded qualitatively similar results in the experiments described below, tibia1 measurements were found to yield more consistent results with smaller standard deviations within groups. With careful positioning of the knee and heel against the cassette the tibia can be placed in the desirable position with excellent reproducibility. In a group of eight animals removed and repositioned on a film cassette, the variation in mean tibia1 measurements was less than kO.5 mm. The three groups of animals used to generate data shown in Figures 2 and 3 were sacrificed at the conclusion of the follow-up period (200 days of age). The legs were disarticulated at the hip, and muscles and patella dissected away from the femur and tibia. Direct measurements of tibia1 and femoral length were made with a calipers to the nearest 0.2 mm. A total of 46 legs were dissected and the mean tibia1 lengths of comparably treated legs (in most cases eight per group) were compared with radiographic measurements made just prior to sacrifice. RESULTS
The X ray dose-response relationship for single dose and fractionated radiation (5 Gy/day) using the ratio of 45 40
E
E 35 s= 30
c
,o 3 ._ I-
25
20 I
0
2s
I
60
,
I
1
76
100 126
Age
(days)
1
160
f
I
176 200
Fig. 2. Tibia1 growth curves for right or left legs treated with 25 Gy in SDF (5 Gy/day) (m) or TDF (2.5 Gy/lO fractions/5 days) (A). The solid line represents the normal growth curve for untreated legs in a separate group of eight animals. Each point represents the mean tibia1 length of right or left legs in eight animals + standard deviation (S.D.).
143
Bone growth arrest 0 P. J. EIFEL
0
25
50
75
100
12s
150
17s
200
Age (days) Fig. 3. Tibia1 growth curves for right or left legs treated with 20 Gy in SDF (4 Gy/day) (H) or TDF (2.0 Gy/lO fractions/5 days) (A) and normal growth curve (0). Data points represent the mean tibia1 lengths in eight animals (+-S.D.) except for the last two points on the growth curves of irradiated animals. These represent the mean lengths in seven animals (one animal succumbed to anesthesia on day 99).
observed tibia1 growth and normal growth at 200 days as an endpoint is shown in Figure 2. Each data point represents a single animal. In addition, a single data point represents the mean (*SD.) growth arrest from a course of hyperfractionated irradiation (2.5 Gy twice daily/5 days) in a group of eight animals. The slopes in Figure 4 are indicated only for the purposes of comparison. No attempt was made in these experiments to determine precisely the proportionate growth of the proximal and distal tibia1 epiphyses. As explained above, the distal tibia1 epiphysis was carefully shielded in all experiments allowing normal growth distally. Consequently the measured tibia1 lengths and the calculated “fraction normal growth” in Figure 4 all include a presumably constant growth fraction from the untreated distal end. In fact, the highest doses used in these experiments (20 Gy single fraction and 40 Gy in 5 Gy fractions) caused very severe, probably near complete growth arrest. The doses used in subsequent experiments were chosen to cause growth arrest of intermediate severity. Significant sparing of radiation-induced bone growth arrest was observed when a 5 day course of fractionated irradiation (5.0 Gy/day) was delivered with TDF of 2.5 Gy (Total Dose (TD) = 25 Gy) as compared with SDF (Fig. 2). At 200 days of age the mean difference in tibia1 length as measured from radiographs was 1.9 mm, representing a sparing of bone growth arrest of 2 1%. Similar results were seen with a 5 day course of 20 Gy (Fig. 3). Animals were treated in groups of 8 with right and left legs receiving the same total dose with different fraction-
ation schedules. In this way, any confounding variables which could possibly result from variations in anesthesia, nutrition, etc. should have had an identical influence upon right and left leg and, therefore, upon both treatment groups. The differences in tibia1 length of legs treated with SDF vs. those treated with TDF were highly statistically significant (p < 0.001). These data were confirmed in two additional similar experiments. The differential effect on linear bone growth observed with SDF and TDF is manifested within the first 6-8 weeks following treatment, corresponding to the rapid growth period and pubescence of the male rat (Figs. 2, 3). Following this, the growth curves appear to be parallel but never achieve the growth rate of untreated legs. The primary endpoint being studied in these experiments was linear bone growth. As expected, other acute and subacute effects were also decreased with TDF. Legs treated with multiple daily fractions consistently demonstrated less severe epilation and more complete regrowth of fur at 100 days. After 200 days of age, animals from the experiment depicted in Figures 2 and 3 were sacrificed and direct measurements were made of tibia1 and femoral length. These results confirmed those obtained from radiographic measurements, with comparable, highly significant, differences in mean tibia1 length between legs treated with SDF and TDF. (Data not shown.) In all
10
! 0
1
I
I
1
10
20
30
40
DOSE (GRAY) Fig. 4. Percent normal growth (Alength/Alength,,,,,) from treatment to age 200 days as a function of total dose delivered. (A) Single fraction of 12-20 Gy. Each point (0) represents one treated animal. (B) Fractionated irradiation (5 Gy/day). Each point represents the mean tibia1 length in four animals +_S.D. (C) A single point (mean of six animals f S.D.) representing growth arrest with TDF (2.5 Gy twice daily, total dose = 25 Gy). Do = 19.5,60 and 80 for A, B and C respectively.
144
1. J. Radiation Oncology 0 Biology 0 Physics
cases, the mean tibia1 lengths measured in these two ways (directly or from radiographs) were comparable.
DISCUSSION These data show that division of a daily dose of 4-5 Gy into two equal fractions results in dramatic improvement of linear bone growth in weanling rats. The results demonstrate conclusively that a portion of the epiphysial damage caused by ionizing radiation is recoverable during a 6 hr period. A number of clinical studies have confirmed that daily doses of 2-2.5 Gy can be hyperfractionated with little loss in the probability of tumor control and with improved sparing of normal tissue, specifically skin and mucosa.‘933’63’9*23 Norin et a1.,18Sarrazin et al.” and Ziegle? have reported the use of hyperfractionated whole-abdominal irradiation in the treatment of a small number of children with Wilms’ tumor and Burkitt’s lymphoma. These children had less acute gastrointestinal toxicity than would have been expected with standard fractionation. There has been some recent enthusiasm for the use of hyperfractionation in treatment of pediatric astrocytomas.24 However no clinical studies have been performed to evaluate the late effects, specifically bone growth effects, of hyperfractionated irradiation in children. This may reflect a reluctance to change relatively successful therapeutic approaches and to address the logistical problems of immobilizing children two or three times a day without compelling laboratory evidence that such an approach would improve therapeutic index. This study provides the first evidence of this kind. Our knowledge about the growth-sparing effect of fractionated irradiation derives from anecdotal observations and qualitative retrospective clinical reviews. These studies have suggested that the use of relatively large daily fractions results in a less satisfactory therapeutic index and has led to the use of increasingly small daily fractions in the treatment of children. Most laboratory investigations of radiation induced growth arrest have involved single or two dose experiments.438”2-15 The most extensive series of experiments was done by Hinkel.‘2-‘4 He observed the effects of single doses of irradiation on growing long bones in rats and determined that the degree of stunting was related to dose and age at the time of treatment. Histological abnormalities were appreciated at doses lower than those required for minimal growth arrest. He noted “dropout” of chondrocytes and disorganization of their normal columnar arrangement. Augmentation of mineral deposition in the matrix caused growth arrest lines. Higher doses also caused changes in the supporting structure and blood vessels;
July 1988, Volume 15, Number 1
the latter became narrow and irregular and ended short of the cartilage zone. Kember” published the results of the only study which measured the effect of fractionated doses on growing cartilage. Using an in vivo assay in which clones of recovering cartilage cells were counted, he described a survival curve with a Do of 165 rad. When 2200 rad was fractionated into 1600 rad and 600 rad fractions, a maximal recovery ratio of 6 at 6 hours was found. This suggested an extrapolation number of 6. Although in viva tissue interactions in the complex, multicomponent structure of cartilage may have influenced the observed survival curve parameters for cartilage cells, this work showed that two-dose fractionation resulted in decreased damage to this cellular compartment. Since the work of Hinke112-‘4 and Kember,” little has been added to our understanding of the mechanism(s) of radiation-induced growth arrest. The data presented demonstrate that a portion of the damage caused by a dose of ionizing radiation to actively growing epiphysial bone is repaired during the 6 hour period between fractions. This implies that at least one of the several important cellular compartments of this complex biological system sustains some component of sublethal injury. The slopes and extrapolation numbers of dose response curves shown in Figure 4 can not be used to draw any conclusion about the inherent radiosensitivity or repair capacity of specific tissue compartments (i.e., endothelial, chondrocyte, osteoblast); however, they do demonstrate substantial recovery of the growth plate with fractionated and twice-daily fractionated irradiation. In fact, very little is known about the mechanisms of expression of radiation damage in growing bone. Histopathological observations described by Hinkel14 suggest that important damage is sustained by vascular, chondrocyte and matrix compartments. However, his work was confined to experiments with large single fractions the effects of which are probably qualitatively as well as quantitatively different than those seen with fractionated irradiation. Work is currently in progress to explore the comparative histopathology in epiphysial bone exposed to single dose, fractionated and hyperfractionated irradiation. The observation that hyperfractionation of therapeutic radiation results in less damage to growing bone than single daily fractions has obvious important implications for the radiation therapy of children with neoplastic disease. Although the fraction sizes used in these initial experiments were relatively large, it is likely that recovery of sublethal injury and similar tissue sparing will be observed with hyperfractionation of more standard daily doses. If so, the radiotherapeutic management of children with cancer might be achieved with substantially less late morbidity than is currently experienced.
REFERENCES 1.
Ang, ILK., Van der Schueren, E., Notter, G., Horiot, J.C., Chenal, Fauchon, F., Raps, J., Peperzeel, H., Goffin, J.C.,
M., Glabbeke, M.V.: Split course multiple daily fractionated radiotherapy schedule combined with mis-
Vessiere,
Bone growth arrest 0 P J. EIFEL
onidazole for the management ofgrade III and IV ghomas. Int. J. Radiat. Oncol. Biol. Phys. 8: 1657-1664, 1982. 2.Arcangeli, G., Mauro, R., Nervi, C.: Multiple daily fractionation in radiotherapy: Biological rationale and preliminary clinical experiences. Eur. J. Cancer 15: 1077-1083, 1979. 3.Barker, J.L., Montague, E.D., Peters, L.J.: Clinical experience with irradiation of inflammatory carcinoma of the breast with and without elective chemotherapy. C’ancer45: 625-629,198O. 4.Blackburn, B.A., Wells, B.A.: Radiation damage to growing bone: the effect of X-ray doses of 100 to 1,000 r on mouse tibia and knee joint. Br. J. Radiol. 36: 505-513, 1963. 5.Collis, C.H., Down, J.D.: The effect ofdose rate and multiple fractions per day on radiation-induced lung damage in mice. Br. J. Radiol. 57: 1037-1039, 1984. 6.Douglas, B.G.: Superfractionation: Its rationale and anticipated benefits. Int. J. Radiat. Oncol. Biol. Phys. 8: 11431 153, 1982. 7.Elkind, M.M., Sutton, H.: Radiation response of mammalian cells grown in culture. 1. Repair of X-ray damage in surviving Chinese hamster cells. Radiat. Rex 13: 556-593, 1960. 8.Engstrom, H., Magnusson, B.C., Turesson, I.: Effect of 50 kv irradiation on enzyme activities of growing rat bone. Acta Radiol. Oncol. 22: 65-70, 1983. 9.Fowler, J.F., Sheldon, P.W., Harris, S.R., Hill, S.A., Ayres, SE.: Relative effectiveness of 12-hourly fractionation and a non-uniform x-ray schedule in the optimum fractionation of C3H mouse mammary tumours. Br. J. Radiol. 48: 581-589, 1975. 10. Giri, P.G.S., Kimler, B.F., Giri, U.P., Cox, G.G., Reddy, E.K.: Comparison of single, fractionated and hyperfractionated irradiation on the develoDment of normal tissue damage in rat lung. Int. J. Radial Oncol. Biol. Phys. 11: 527-534. 1985. 11. Gonzalez, D.G., Breur, K.: Clinical data from irradiated growing long bones in children. Int. J. Radiat. Oncol. Biol. Phys. 9: 841-846, 1983. 12. Hinkel, C.L.: The effect of roentgen rays upon the growing long bones of albino rats. I. Quantitative studies of the growth limitation following irradiation. Am. J. Roentgenol. 47: 439-457, 1942.
145
13. Hinkel, CL.: The effect of irradiation upon the composition and vascular&y of growing rat bones. Am. J. Roentgenol. 50: 5 16-526, 1943. 14. Hinkel, C.L.: The effect of roentgen rays upon the growing long bones of albino rats. II. Histopathological changes involving endochondral growth centers. Am. J. Roentgenol. 49: 321-348, 1943. 15. Kember, N.F.: Cell survival and radiation damage in growth cartilage. Br. J. Radiol. 40: 496-505, 1967. 16. Kotalik, J.F.: Multiple daily fractions in radiotherapy. Ca. Treat. Rev. 8: 127-146, 1981. 17. Neuhauser, E.B.D., Wittenborg, M.H., Berman, C.Z., Cohen, J.: Irradiation effects of roentgen therapy on growing spine. Radiol. 59: 637-650, 1952. 18. Norin, T., Clifford, P., Einhorn, J., Einhorn, N., Johansson, B., Klein, G., Onyango, J., DeSchryver, A., Walstam, R.: Conventional and Superfractionated radiation therapy in Burkitt’s lymphoma. Acta Radiol. 10:545-557, 197 1. 19. Parsons, J.T., Cassisi, N.J., Million, R.R.: Results oftwicea-day irradiation of squamous cell carcinomas of the head and neck. Int. J. Radiat. Oncol. Biol. Phys. 10:204 l-205 1, 1984. 20. Probert, J.C., Parker, B.R.: The effects of radiation therapy on bonegrowth. Radiol. 114: 155-162, 1975. 21. Rubin, P., Duthie, R.B., Young, L.W.: The significance of scoliosis in post-irradiated Wilms’ tumor and neuroblastoma. Radiol. 79: 539-558, 1962. 22. Sarrazin, D.: Hyperfractionated irradiation of the whole abdomen in five children (Abstr.). Proceedings of the 14th Annual SIOP Meeting. Berne, Switzerland, International Society of Pediatric Oncology. 1982, p. A33. 23. Shukovsky, L.J., Fletcher, G.H., Montague, E.D., Withers, H.R.: Experience with twice-a-day fractionation in clinical radiotherapy. Am. J. Roentgenol. 126:155-162, 1976. 24. Wara, W.M., Edwards, M.S.B., Levin, V.A., Heidman, R. Urtasun, R., Diaz, R.E., Silver, P., Sposto, R.: A new treatment regimen for brainstem glioma: A pilot study of the brain tumor research center and children’s cancer study group (Abstr.). Int. J. Radiat. Oncol. Biol. Phys. 12: 143144,1986. 25. Ziegler, J.L.: Management of Burkitt’s lymphoma: An update. Ca. Treat. Rev. 6: 95-105, 1979.