Radiotherapy and Oncology, 17 (1990) 7-15 Elsevier RADION 00667
Preclinical evaluation of pions in vivo" experience at TRIUMF D. J. Chaplin 1, B. G. Douglas 1, T. Saito ~, L. D. Skarsgard J. D e n e k a m p 2
1,
G. K. Y. L a m i and
~Medical Biophysics Unit, B.C. Cancer Research Centre, Vancouver, Canada, and 2Gray Laboratory, Mount Vernon Hospital, Northwood, Middlesex HA6 2RN, U.K.
(Received 11 October 1988, revision received 24 July 1989, accepted 13 September 1989)
Key words: Pions; RBE, early and late effects; X-ray dose
Summary This paper describes the results obtained from in vivo studies of the pion beam at the TRI University Meson Facility (TRIUMF). The studies encompass work (from 1978 to 1986), designed to evaluate the RBE for early and late effects and to assess the importance of X-ray dose rate and treatment volume on these values. Results with early responding tissues, i.e. mouse and pig skin and mouse intestine indicate a pion RBE of about 1.5 in the clinically relevant dose per fraction range of 2-3 Gy. At these doses, RBE appears to be independent of the reference X-ray dose rate. However, at high doses per fraction, the RBE values become increasingly X-ray dose rate dependent. The induction of late effects by pions has been assessed by monitoring the late dermal response of pig skin; late fibrosis was not assessed in this study. The values obtained using the chosen endpoint indicate that the RBE is not significantly higher than that seen in any of the early responding tissues for pion doses as low as 2-3 Gy per fraction. The effect of increasing the treatment volume for pion therapy has been assessed using mouse intestine. The results show that for a constant field size, RBE decreases with increasing peak width. However, if peak width is held constant and field size increased, there is evidence for an increased RBE. Thus, for pion volumes used in clinical treatments where both field size and peak width are increased, the effective RBE may not be that much different from values obtained in animal studies where narrow peak widths and small field sizes are commonly used. The in vivo studies undertaken at T R I U M F have provided a basis for choosing the pion doses to be used in the clinical studies. Indeed the clinical data obtained to date indicate that the RBE value of 1.5 obtained from the preclinical in vivo studies is suitable for skin, brain, colorectum and vagina.
Address for correspondence: D. J. Chaplin, Medical Biophysics Unit, B.C. Cancer Research Centre, 601 West 10 Ave., Vancouver, B.C. V5Z 1L3, Canada.
0167-8140/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
Introduction
It was Fowler and Perkins in 1961 who first formally suggested the potential advantages of pion radiotherapy [ 5 ]. These advantages arise because of the physical properties of negative pi-mesons (pions), which provide improved depth dose characteristics and increased RBE as compared to conventional photon therapy. As pions pass through tissue, they slow down as a result of the normal coulomb interactions, and eventually stop and are captured by atomic nuclei of the material. The destabilised nuclei then disintegrate, releasing short range alpha particles and other high LET fragments. The resulting events are known as pion stars. As a consequence of these pion absorption processes, a radiation dose containing a significant proportion of high LET events can be deposited at a given depth by manipulating the initial pion momentum. Furthermore by allowing an appropriate range and distribution ofpion momenta in the incident beam, this peak dose can be spread to encompass the chosen treatment volume. The TRI University Meson Facility (TRIUMF) in Vancouver is one of three meson factories where the biological effects ofpions have been extensively evaluated. The other installations are SIN (Schweitzerisches Institut for Nuklearforschung) in Villigen, Switzerland and LAMPF (Los Alamos Meson Physics Facility) in Los Alamos, New Mexico, U.S.A. The in vivo
evaluation of pions at T R I U M F was initiated in 1978 and has primarily involved the use of mouse skin and intestine to evaluate early damage and pig skin to evaluate both early and late effects. These in vivo studies have been described in part in previous publications [3,4,15-17]. In the present paper, we will summarise the in vivo data obtained to date from the studies at T R I U M F and discuss them with reference to those obtained at SIN and LAMPF.
Materials and methods
Irradiation procedures Pions The biomedical beam line at T R I U M F provides a horizontal beam of pions of momentum 200 + 14 MeV/c. For our studies, the majority of animals were immobilised in a static position during irradiation and irradiated using small field sizes with narrow peak widths (4 cm at 90~o isodose contour). To investigate the effects of increasing field size (i.e. volume), we have utilised the patient transport couch to provide a beamscanning arrangement. The field sizes and dose rates for the treatment protocols are shown in Table I. For each study sufficient bolus was used to ensure that the pion-stopping region encompassed the target tissue.
TABLE I Details of the irradiation fields and dose rates used for in vivo studies with the pion beam at TRIUMF. System
Scanning mode
Field size at 90% isodose contour
Dose rate (cGy/min)
Mouse foot Pig skin
Static Static
15 cGy/min 10 cGy/min
Mouse intestine
Scanned
3.5 × 4 cm Circular diameter 3 cm 3 × 15 cm (rectangular) 10.8 x 13.3 cm (hexagonal)
Scanned
33 cGy/min 5 cGy/min
X-rays The X-ray studies were carried out for mouse skin, jejunum and pig skin using irradiation systems described in detail in previous publications [3,4,15]. The X-rays were generated at 270 kVp, 20 mA providing dose rates of 80 cGy/min (for the pig skin studies), 15 and 150 cGy/min (for the mouse foot studies), and 5, 20 and 33 cGy/min (for the mouse jejunum studies).
Biological systems Animals Mice.
Male B6C3F 1 mice were used for the mouse skin and jejunum assays. All mice were housed in plastic cages on sawdust bedding with six animals per cage and kept under filter lids. Food and water were freely available at all times.
two discernable waves of damage after irradiation, an early epidermal reaction 1 to 2 months after exposure and a later dermal reaction which occurs 3 to 5 months later. Pig skin also exhibits late fibrosis but this was not measured in our studies. Pigs were irradiated unanaesthetised while immobilised in a sling. For the present study, schedules of 4, 7, 9 and 10 fractions were used with interfraction intervals of approximately 24 h. For each pig, four areas were irradiated: one with pions, the other three with graded X-ray doses chosen to encompass the expected RBE. The relative position of the pion and X-ray fields were varied from pig to pig to avoid the possible influence of systemic positional variation in radiosensitivity. The animals were scored for radiation damage using a scale which increases from 0 to 18 in line with increasing severity of skin reaction, as described previously [3].
Pigs. Immature female and castrated male pigs weighing 35-45 kg were used for the experiments. All pigs were housed in communal pens containing 3-6 pigs. Food (16~o protein hog food) and water were available each day.
Mouse skin In order to study the RBE over a wide range of doses per fraction, schedules of 1, 2, 4, 10 and 20 fractions were used, with interfraction intervals of 24 h for the 2, 4 and 10 fraction schedule and 12 h for the 20 fraction schedule. The data were analysed using the mean skin reaction for a 7-day period which encompassed the highest reaction. This analysis has been described in detail elsewhere [2]. RBE values were obtained by comparing the X-ray and pion doses needed to produce the same level of skin damage [4]. This comparison can be made at any level, but for the subsequent summaries they have been estimated at an average reaction of 13, 17 and 22 U.
Pig skin Pig skin studies were undertaken because of the fact that the skin structure of pigs better reflects that of man than does rodent skin. Pig skin shows
Mouse jejunum Mouse jejunum studies have been used to evaluate the effects of different pion-stopping volumes. For these studies, single doses were used for the most part but, in addition, a limited number of 2 and 4 fraction studies have been performed. The response of mouse jejunum was assessed 84-96 h after the mid-point of irradiation using the crypt cell assay described by Withers and Elkind[23]. RBE values were obtained by comparing the doses required for given X-ray and pion treatments to produce the same number of surviving crypts [ 15].
Results
Figure 1 shows a comparison of the X-ray and pion reactions observed in the 4 and 10 fraction mouse foot studies. RBE values can be obtained from the ratio of X-ray and pion doses required to produce a given level of skin damage. A summary of the complete set of data obtained from 1, 2, 4, 10 and 20 fraction mouse foot experiments performed at T R I U M F over the period Septem-
10 tU
o u'J
28
--
4 FRACTIONS
26
--
o
24
-
22
-
20
-
18
-
W n-
16
-
Z
14
-
12
-
10
-
8
-
z
o I-
uJ 10
-
10 FRACTIONS
,/? I
I
I
I
20
30
40
50
i
I
I
60
10
20
I
I
30
40
I
I
50
60
I 70
I
I
80
90
TOTAL DOSE ( G y . )
Fig. 1. Dose-response curves for 4 and 10 fraction pions (stars), 150 cGy/min X-rays ( • ) and 15 cGy/min X-rays (o). RBE values can be obtained from the ratio of X-ray to pion doses to achieve a given reaction score. Error bars are + S.E.M.
ber, 1978 until September, 1983 has been published [4]. A summary of the RBEs obtained in this study which involved a total of 4269 animals is shown in Fig. 2. The figure shows the RBE values derived when X-rays at either 15 or 150 cGy/min are used as the reference. It can be
seen that RB E increases with decreasing dose per fraction when either 150cGy/min X-rays, or 15 cGy/min X-rays are used as reference. The results shown indicate that although the RBE values at high doses per fraction are dependent on the dose rate of the reference X-rays, this
1.8
I
W
Z 1.7
4,0
I
_
o
1.6 3.5
1.5 3.0 o
1.4 '~"
1.3 ~
__o
1.2
2.5
2.0
i
°
I-
,
Preclinical evaluation of pions in vivo" experience at TRIUMF D. J. Chaplin 1, B. G. Douglas 1, T. Saito ~, L. D. Skarsgard J. D e n e k a m p 2
1,
G. K. Y. L a m i and
~Medical Biophysics Unit, B.C. Cancer Research Centre, Vancouver, Canada, and 2Gray Laboratory, Mount Vernon Hospital, Northwood, Middlesex HA6 2RN, U.K.
(Received 11 October 1988, revision received 24 July 1989, accepted 13 September 1989)
Key words: Pions; RBE, early and late effects; X-ray dose
Summary This paper describes the results obtained from in vivo studies of the pion beam at the TRI University Meson Facility (TRIUMF). The studies encompass work (from 1978 to 1986), designed to evaluate the RBE for early and late effects and to assess the importance of X-ray dose rate and treatment volume on these values. Results with early responding tissues, i.e. mouse and pig skin and mouse intestine indicate a pion RBE of about 1.5 in the clinically relevant dose per fraction range of 2-3 Gy. At these doses, RBE appears to be independent of the reference X-ray dose rate. However, at high doses per fraction, the RBE values become increasingly X-ray dose rate dependent. The induction of late effects by pions has been assessed by monitoring the late dermal response of pig skin; late fibrosis was not assessed in this study. The values obtained using the chosen endpoint indicate that the RBE is not significantly higher than that seen in any of the early responding tissues for pion doses as low as 2-3 Gy per fraction. The effect of increasing the treatment volume for pion therapy has been assessed using mouse intestine. The results show that for a constant field size, RBE decreases with increasing peak width. However, if peak width is held constant and field size increased, there is evidence for an increased RBE. Thus, for pion volumes used in clinical treatments where both field size and peak width are increased, the effective RBE may not be that much different from values obtained in animal studies where narrow peak widths and small field sizes are commonly used. The in vivo studies undertaken at T R I U M F have provided a basis for choosing the pion doses to be used in the clinical studies. Indeed the clinical data obtained to date indicate that the RBE value of 1.5 obtained from the preclinical in vivo studies is suitable for skin, brain, colorectum and vagina.
Address for correspondence: D. J. Chaplin, Medical Biophysics Unit, B.C. Cancer Research Centre, 601 West 10 Ave., Vancouver, B.C. V5Z 1L3, Canada.
0167-8140/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
Introduction
It was Fowler and Perkins in 1961 who first formally suggested the potential advantages of pion radiotherapy [ 5 ]. These advantages arise because of the physical properties of negative pi-mesons (pions), which provide improved depth dose characteristics and increased RBE as compared to conventional photon therapy. As pions pass through tissue, they slow down as a result of the normal coulomb interactions, and eventually stop and are captured by atomic nuclei of the material. The destabilised nuclei then disintegrate, releasing short range alpha particles and other high LET fragments. The resulting events are known as pion stars. As a consequence of these pion absorption processes, a radiation dose containing a significant proportion of high LET events can be deposited at a given depth by manipulating the initial pion momentum. Furthermore by allowing an appropriate range and distribution ofpion momenta in the incident beam, this peak dose can be spread to encompass the chosen treatment volume. The TRI University Meson Facility (TRIUMF) in Vancouver is one of three meson factories where the biological effects ofpions have been extensively evaluated. The other installations are SIN (Schweitzerisches Institut for Nuklearforschung) in Villigen, Switzerland and LAMPF (Los Alamos Meson Physics Facility) in Los Alamos, New Mexico, U.S.A. The in vivo
evaluation of pions at T R I U M F was initiated in 1978 and has primarily involved the use of mouse skin and intestine to evaluate early damage and pig skin to evaluate both early and late effects. These in vivo studies have been described in part in previous publications [3,4,15-17]. In the present paper, we will summarise the in vivo data obtained to date from the studies at T R I U M F and discuss them with reference to those obtained at SIN and LAMPF.
Materials and methods
Irradiation procedures Pions The biomedical beam line at T R I U M F provides a horizontal beam of pions of momentum 200 + 14 MeV/c. For our studies, the majority of animals were immobilised in a static position during irradiation and irradiated using small field sizes with narrow peak widths (4 cm at 90~o isodose contour). To investigate the effects of increasing field size (i.e. volume), we have utilised the patient transport couch to provide a beamscanning arrangement. The field sizes and dose rates for the treatment protocols are shown in Table I. For each study sufficient bolus was used to ensure that the pion-stopping region encompassed the target tissue.
TABLE I Details of the irradiation fields and dose rates used for in vivo studies with the pion beam at TRIUMF. System
Scanning mode
Field size at 90% isodose contour
Dose rate (cGy/min)
Mouse foot Pig skin
Static Static
15 cGy/min 10 cGy/min
Mouse intestine
Scanned
3.5 × 4 cm Circular diameter 3 cm 3 × 15 cm (rectangular) 10.8 x 13.3 cm (hexagonal)
Scanned
33 cGy/min 5 cGy/min
X-rays The X-ray studies were carried out for mouse skin, jejunum and pig skin using irradiation systems described in detail in previous publications [3,4,15]. The X-rays were generated at 270 kVp, 20 mA providing dose rates of 80 cGy/min (for the pig skin studies), 15 and 150 cGy/min (for the mouse foot studies), and 5, 20 and 33 cGy/min (for the mouse jejunum studies).
Biological systems Animals Mice.
Male B6C3F 1 mice were used for the mouse skin and jejunum assays. All mice were housed in plastic cages on sawdust bedding with six animals per cage and kept under filter lids. Food and water were freely available at all times.
two discernable waves of damage after irradiation, an early epidermal reaction 1 to 2 months after exposure and a later dermal reaction which occurs 3 to 5 months later. Pig skin also exhibits late fibrosis but this was not measured in our studies. Pigs were irradiated unanaesthetised while immobilised in a sling. For the present study, schedules of 4, 7, 9 and 10 fractions were used with interfraction intervals of approximately 24 h. For each pig, four areas were irradiated: one with pions, the other three with graded X-ray doses chosen to encompass the expected RBE. The relative position of the pion and X-ray fields were varied from pig to pig to avoid the possible influence of systemic positional variation in radiosensitivity. The animals were scored for radiation damage using a scale which increases from 0 to 18 in line with increasing severity of skin reaction, as described previously [3].
Pigs. Immature female and castrated male pigs weighing 35-45 kg were used for the experiments. All pigs were housed in communal pens containing 3-6 pigs. Food (16~o protein hog food) and water were available each day.
Mouse skin In order to study the RBE over a wide range of doses per fraction, schedules of 1, 2, 4, 10 and 20 fractions were used, with interfraction intervals of 24 h for the 2, 4 and 10 fraction schedule and 12 h for the 20 fraction schedule. The data were analysed using the mean skin reaction for a 7-day period which encompassed the highest reaction. This analysis has been described in detail elsewhere [2]. RBE values were obtained by comparing the X-ray and pion doses needed to produce the same level of skin damage [4]. This comparison can be made at any level, but for the subsequent summaries they have been estimated at an average reaction of 13, 17 and 22 U.
Pig skin Pig skin studies were undertaken because of the fact that the skin structure of pigs better reflects that of man than does rodent skin. Pig skin shows
Mouse jejunum Mouse jejunum studies have been used to evaluate the effects of different pion-stopping volumes. For these studies, single doses were used for the most part but, in addition, a limited number of 2 and 4 fraction studies have been performed. The response of mouse jejunum was assessed 84-96 h after the mid-point of irradiation using the crypt cell assay described by Withers and Elkind[23]. RBE values were obtained by comparing the doses required for given X-ray and pion treatments to produce the same number of surviving crypts [ 15].
Results
Figure 1 shows a comparison of the X-ray and pion reactions observed in the 4 and 10 fraction mouse foot studies. RBE values can be obtained from the ratio of X-ray and pion doses required to produce a given level of skin damage. A summary of the complete set of data obtained from 1, 2, 4, 10 and 20 fraction mouse foot experiments performed at T R I U M F over the period Septem-
10 tU
o u'J
28
--
4 FRACTIONS
26
--
o
24
-
22
-
20
-
18
-
W n-
16
-
Z
14
-
12
-
10
-
8
-
z
o I-
uJ 10
-
10 FRACTIONS
,/? I
I
I
I
20
30
40
50
i
I
I
60
10
20
I
I
30
40
I
I
50
60
I 70
I
I
80
90
TOTAL DOSE ( G y . )
Fig. 1. Dose-response curves for 4 and 10 fraction pions (stars), 150 cGy/min X-rays ( • ) and 15 cGy/min X-rays (o). RBE values can be obtained from the ratio of X-ray to pion doses to achieve a given reaction score. Error bars are + S.E.M.
ber, 1978 until September, 1983 has been published [4]. A summary of the RBEs obtained in this study which involved a total of 4269 animals is shown in Fig. 2. The figure shows the RBE values derived when X-rays at either 15 or 150 cGy/min are used as the reference. It can be
seen that RB E increases with decreasing dose per fraction when either 150cGy/min X-rays, or 15 cGy/min X-rays are used as reference. The results shown indicate that although the RBE values at high doses per fraction are dependent on the dose rate of the reference X-rays, this
1.8
I
W
Z 1.7
4,0
I
_
o
1.6 3.5
1.5 3.0 o
1.4 '~"
1.3 ~
__o
1.2
2.5
2.0
i
°
I-
,
