By G. M. Morris, J. W. Hopewell and A. D. Morris. CRC Normal Tissue Radiobiology Research Group, Research Institute (University of Oxford), Churchill.
1992, The British Journal of Radiology, 65, 152-156
The influence of methotrexate on radiation-induced damage to different lengths of the rat spinal cord By G. M. Morris, J . W. Hopewell and A. D. Morris CRC Normal Tissue Radiobiology Research Group, Research Institute (University of Oxford), Churchill Hospital, Oxford 0X3 7LJ, UK
{Received 11 February 1991 and in revised form 6 September 1991, accepted 20 September 1991) Keywords: Rat spinal cord, Radiation myelopathy, Intrathecal methotrexate, Volume effect
Abstract. An experimental model in the rat has been used to assess the possible enhancement of damage to the spinal cord when radiation is given in the presence of methotrexate (MTX). The dose of MTX used, 4 mg/kg, was the maximum dose that could be given to the rat, when administered into the cerebral spinal fluid circulation, without risk of serious neurological effects. Lengths of 4, 8 and 16 mm of the cervical spine were irradiated with single doses of X rays. For animals that developed paralysis within 30 weeks, caused predominantly by the presence of white matter necrosis, there was no evidence to indicate that MTX enhanced the radiation response of the rat spinal cord, at least at a more clinically relevant level of effect i.e. a low incidence of paralysis. However, for the doses associated with the 50% level of effect (ED50) to an 8 mm long field a significant (p < 0.005) enhancement of the response was seen, suggesting a dose modification factor of 1.19 + 0.07. This was interpreted in terms of an hypothesis to explain the well documented volume effect for very small fields in the rat spinal cord which is based on the migration of viable cells in the periphery of the irradiated site. The apparent smaller effect seen when only 4 mm of the spine was irradiated might be related to the nature of the lesion induced at doses required to produce paralysis in these small fields; the lesions were not restricted to white matter but were more severe and also involved the grey matter and nerve roots after a slightly shorter latent period.
The prophylactic irradiation of the central nervous system (CNS) in combination with intrathecally administered methotrexate (MTX) has proved to be highly effective in the treatment of leukaemia; preventing CNS relapse and prolonging complete remission (Pinkel et al, 1977; Schweinle & Alperin, 1980). However, neurological complications involving the brain and spinal cord have been reported (Luddy & Gilman, 1973; Gagliano & Costanzi, 1976). Usually these complications are transient, but cases of permanent late sequelae have been reported including sensory paraparesis and death. The histological picture of necrosis in the white matter or vascular damage is indicative of radiation injury. However, there have been reports of severe myelopathy after intrathecal MTX alone (Clark et al, 1982) and hence the role played by either agent is uncertain. In the present communication a well established rat model (van der Kogel, 1979) has been used to assess further the possible modification of the radiosensitivity of the spinal cord by MTX. In an earlier study it has been demonstrated that the maximally tolerated dose of MTX had no effect on the radiosensitivity of the rat spinal cord for either the early delayed or the late delayed lesion (van der Kogel & Sissingh, 1985). However, these studies were carried out after the irradiation of a relatively long length of spinal cord in the rat (15 mm). In a subsequent communication (Hopewell et al, 1987), the results of the effects of radiation of the cervical spine of the rat had indicated that the radio152
sensitivity of the cord, at least for the early delayed white matter lesion, was highly dependent on the length irradiated. The ED50 for the paralysis resulting from this lesion increased from about 21.5 Gy to about 50 Gy when the length of cord exposed was changed from 16 mm to 4 mm. It has been argued that the volume effects for small lengths of spinal cord might be related to the migration of viable cells from the edges of the irradiated volume (Hopewell & van der Kogel, 1988). The purpose of the present study was to examine the effects of MTX on the known volume effect for the rat spinal cord after single doses of irradiation. Materials and methods
Female rats of the Sprague Dawley strain, 12-14 weeks of age, were used in these investigations. Three different lengths of the cervical spine, 4, 8 and 16 mm, were locally irradiated with a range of single doses of 250 kV X rays, as described previously (Hopewell et al, 1987). Irradiations were carried out with the rats under chloral hydrate anaesthesia (300 mg/kg). Groups of 5-6 animals were irradiated at each dose level for each field size. At an interval of between 30 and 45 min before irradiation MTX (4 mg/kg-32 mg/m2) was introduced into the cerebral spinal fluid circulation. This dose of MTX was the maximum tolerated by the rats without acute neurotoxic effects. In a previous pharmacokinetic study (van der Kogel & Sissingh, 1985) peak concentrations of the drug were measured in the rat spinal cord at The British Journal of Radiology, February 1992
The influence of methotrexate on radiation-induced damage to different lengths of the rat spinal cord
30 min after injection. MTX was administered, via the left lateral ventricle, by first "thinning" a small area of the left parietal bone using a dental burr, 2 mm from the midline and 1 mm behind the left coronal suture. The delivery needle was inserted to a depth of 5.5 mm into the left lateral ventricle and MTX (Lederle Laboratories) at a concentration of 25 mg/ml was delivered slowly over approximately 30 s. The initial incision of the overlying skin was sutured. Following irradiation the animals were housed in cages in groups of 5-6 and fed on a standard laboratory diet (4IB) and water ad libitum. They were examined at least three times per week to determine the first appearance of neurological signs that might progress to paralysis. Animals that developed paralysis within 30 weeks of irradiation were perfusion fixed under light ether anaesthesia using a mixture of 10% formal saline with 1% acetic acid. Following fixation the cervical and upper thoracic spine containing the irradiated section was decalcified in 8% formic acid for 10 days. 2 mm thick transverse sections of the irradiated length were then taken, they were dehydrated and embedded in paraffin wax. Sections were cut at 5 fim and these were stained with either Mayer's haematoxylin and eosin or Luxol Fast Blue PAS. All data are expressed as the mean of the individual values, plus or minus one standard error ( + SE). Student's /-test was used to calculate probability (p) values. In all tests a value of p < 0.05 was regarded as statistically significant. Dose effect curves were fitted to the data using probit analysis. Results
The dose-related changes in the incidence of paralysis within 30 weeks in animals irradiated in combination with MTX to 4, 8 and 16 mm of the cervical spinal column are shown in Fig. 1. These results are compared with those for irradiation alone in experiments carried 100
n
16mm
20
8mm
30
Field size (mm)
Treatment schedule
ED50 + SE (Gy)
16
X rays MTX + X rays
21.5 + 0.3 21.5 + 0.4
8
X rays MTX + X rays
30.1 + 1.5 25.3 + 0.7
4
X rays MTX + X rays
51.0 + 2.3 47.5 + 2.1
out over the same time period and for which information has already been published (Hopewell et al, 1987). The resulting ED50 values (+ SE) for both sets of data are listed in Table I. In common with the results for X rays alone the ED50 values for paralysis after X irradiation combined with MTX increased as the field size decreased. For the largest 16 mm field the dose-effect curves for the incidence of paralysis were very similar with identical ED50 values of 21.5 Gy. When an 8 mm length of the cervical spine was irradiated the dose-effect for the group receiving MTX was shifted to the left and was significantly steeper (p < 0.001) than that for X rays alone. The resulting ED50 values (±SE) of 25.3 + 0.7 Gy and 30.1 +1.5 Gy were significantly different from each other (p < 0.005), suggesting a modification factor of 1.19 + 0.07. There was also a small shift in the dose-response relationship for paralysis in animals irradiated to 4 mm of the spinal cord in combination with MTX, the dose-effect curve having a similar slope to that for X rays alone. The dose modification factor of 1.07 + 0.07 was not significant (p > 0.2). The latency times for the development of paralysis within 30 weeks for animals irradiated in combination
4mm
40
50
Dose (Gy) Vol. 65, No. 770
Table I. Variation in ED50 values (±SE) for animals developing paralysis with 30 weeks of X irradiation alone or a combination of MTX and X rays to the spinal cord of rats
60
70
Figure 1. Dose-related changes in the proportion of animals developing paralysis within 30 weeks of the irradiation of 4 ( • , • ) , 8 ((},•) and 16 mm (A, A) lengths of the spinal cord of rats with single doses of X rays. Previous data (Hopewell et al, 1987) for irradiation given alone (D,O,A) is compared with that for irradiation in combination with MTX (B,#,A)- Error bars indicate + SE. 153
G. M. Morris, J. W. Hopewell and A. D. Morris
with MTX as compared with X rays alone are shown in Fig. 2. Values for the latency time are only shown in dose groups where at least two animals developed paralysis. Some animals treated with X rays in combination with MTX did not develop paralysis within 30 weeks. Some of these animals, in line with the X ray alone groups, did show neurological damage after latent periods > 45 weeks. However, external unrelated factors prevented a systematic evaluation of animals surviving for > 30 weeks. There were no major differences in the latency for paralysis between animals receiving local cord irradiation and those irradiated in combination with MTX. Although there was not a clear distinction, the latent period for paralysis in animals receiving irradiation to 4 mm of spinal cord tended to be slightly shorter than that for the two larger fields; doses of at least 40 Gy were required to induce a response. A notable exception was the group of five rats irradiated to 8 mm of cord with a single dose of 40 Gy alone. All five animals developed paralysis after an average latency of
20.9 + 0.6 weeks. The dose associated with the same incidence of animals developing paralysis in the combined treatment was 30 Gy where the mean latency was significantly longer at 25.5 + 0.7 weeks (p < 0.001). After the irradiation of 4 mm of spinal cord with doses necessary to produce paralysis, degeneration of grey matter and demyelination of the nerve roots were seen in addition to the more characteristic white matter lesions. This applied equally to the groups receiving irradiation alone or the combined treatment. Selective white matter necrosis was usually characteristic of animals irradiated to 8 and 16 mm of the spinal cord. Discussion
MTX is a folic acid antagonist and as such it has an inhibitory effect on DNA synthesis. Therefore, populations of cells that are cycling tend to accumulate in the G,-S transition phase of the cell cycle after MTX administration (Chabner & Young, 1973). This could explain the observed enhancement of the radiation response by a similar dose of 4 mg/kg of MTX, given
30-
\l 25-
-f-.
CD CD
O CD Q_
20-
CD
15-
Figure 2. Dose-related changes in the mean latency for the development of paralysis within 30 weeks of the irradiation of 4, 8 or 16 mm of the spinal cord of rats with single doses of X rays. Data points are only plotted for doses at which 2 animals in the group developed paralysis (for keys to symbols see Fig. 1). 154
10 20
30
i
i
40
50
60
70
Dose (Gy) The British Journal of Radiology, February 1992
The influence of methotrexate on radiation-induced damage to different lengths of the rat spinal cord
into the lateral ventricles, on the neuroglia stem cell population in the subependymal region of the rat brain (Morris & Hopewell, 1983). However, the possibility of a simple additivity of the cytotoxic effects of the two agents could not be excluded. In the spinal cord of the adult rat there is no relatively rapid proliferating neuroglial stem cell population and the dose of MTX used in the present study was, on the basis of a pilot investigation, the maximum that could be given to the rat without evidence of acute neurological damage. This dose produced no apparent longterm effects. The similarity of the dose-effect curves for the incidence of paralysis for irradiation alone or in combination with MTX to 16 mm of the spinal cord, supports the view that in this slowly proliferating cell system MTX at the dose used produced no enhancement or evidence for additive cytotoxicity in the spinal cord. This supports earlier findings (van der Kogel & Sissingh, 1985) in the rat when similar field sizes were used. However, it should be remembered that very high doses of MTX, when administered alone to patients, have resulted in a demonstrable cytotoxic effect on the CNS (Skullerud & Halvorsen, 1978; Clark et al, 1982). It cannot be ruled out that lower doses of drug may produce generalized subclinical damage in the CNS. Given that there was no obvious modification in the radiation response when 16 mm of the spinal cord was irradiated, it was somewhat of a surprise to see a marked modification in the effect when only 8 mm of the cord was exposed. The resulting dose-effect curve for the combined treatment was steeper than for radiation alone. The implications of this change are a large dose modification factor (DMF) at a high level of incidence of paralysis (DMF > 1.2) declining to insignificantly small values at more clinically relevant levels of effect. (< ED5). This would suggest that MTX is unlikely to enhance the radiation-induced response in the spinal cord even in small high dose regions where treatment volumes overlap. The most likely explanation of these findings is that MTX does indeed produce subclinical injury to the CNS and that the migration of cells from the periphery of the 8 mm long irradiated site is in some way inhibited. This results in an enhancement of the radiation effect in that volume. The response of longer lengths of the spinal cord ( ^ 16 mm) would not be affected in this way, since other studies (van der Kogel, 1991) have suggested that ~ 10 mm is the maximum length over which cell migration might exert an effect in modifying the late radiation response of the spinal cord. The possible effects of MTX in inhibiting or reducing cell migration are supported by more recent irradiation studies (van der Kogel, unpublished). In these studies a small length of the spinal cord (4 mm) was irradiated within a larger area (18 mm) that had received a subthreshold dose. There was a significant reduction in the ED50 for paralysis within 30 weeks, from 45 Gy to 27.5 Gy, indicating that the sparing of damage usually associated with small volumes had largely disappeared. Although both of these studies provide support for the view that the migration of viable cells from the edges of Vol. 65, No. 770
the irradiated length provides an explanation for the observed volume effect in small fields for the spinal cord, the exact explanation of the observed volume effect still remains uncertain. The migration of viable epithelial cells from the edges of the unirradiated site would appear to explain area effect from acute reactions in pig skin (Hopewell, 1990). Given the effect of combined treatment with MTX in modifying the response of 8 mm of the spinal cord to radiation, and its possible explanation, a similar if not more marked effect might have been anticipated after the irradiation of 4 mm of the spinal cord. Although the magnitude of the reduction in the ED50 for paralysis for the 4 mm and 8 mm fields was similar, 3.5 Gy as compared with 4.8 Gy, the dose-modification factor for the 4 mm fields was not significant and there was no change in the slope of the dose-effect curve. Although this might appear to be a contradictory finding, it may be related to the different nature of the lesion induced at doses required to produce paralysis when only a 4 mm length of the spinal cord is irradiated. The lesions seen were not restricted to the white matter, but are more extensive and also involve the grey matter and the nerve roots. Details of the different histological lesions associated with 4 mm as compared with larger lengths of irradiated cord have been discussed in detail previously (Hopewell et al, 1987). The latency associated with such lesions is also slightly shorter, both for radiation alone and the combined treatment group when compared with that for effects in larger fields. Here again an analogy can be drawn with effects seen in the skin, for fields of < 5 mm diameter; the migration of epithelial cells from the periphery of these sites is such that moist desquamation is avoided and any lesion seen is related to effects in other cell populations, the death of which results in a more severe lesion in a shorter latent interval (Hopewell, 1990). Thus, in summary, at clinically relevant levels of effect, i.e. a low incidence of radiation-induced paralysis, there was no evidence to indicate that MTX enhanced the radiation response of the rat spinal cord. For more severe levels of damage, changes were seen that could be interpreted as indicating that cell migration from the periphery of fields of no more than 8 mm length was a key factor in explaining tissue volume related effects, rather than any model based on the specific arrangement of any theoretical functional subunits (Withers et al, 1988), the biological identity of which has still to be validated. Acknowledgments The authors would like to thank Miss K. Lane for her day to day care of the animals and Mr B. Hopkins for his technical assistance. This work was partly supported by the Leukaemia Research Fund. The Research Group at the Churchill Hospital is supported by a grant from the Cancer Research Campaign. References CHABNER,
B.
A.
&
YOUNG,
R.
C,
1973.
Threshold
methotrexate concentration for in vivo inhibition of DNA 155
G. M. Morris, J. W. Hopewell and A. D. Morris synthesis in normal and tumorous target tissues. Journal of Clinical Investigation, 52, 1804-1811. CLARK, A. W., STEVEN, R. C , NISSENBLATT, M. J. & WILSON,
S. K., 1982. Paraplegia following intrathecal methotrexate. Neurologic findings and elevation of myelin basic protein. Cancer, 50, 42-47.
S. A. Leibel and G. E. Sheline (Raven Press, New York), 91-111. VAN DER KOGEL, A. J. & SISSINGH, H. A. S., 1985. Effects of
intrathecal methotrexate and cytosine arabinoside on the rat spinal cord. Radiotherapy and Oncology, 4, 229-251. LUDDY, R. E. & GILMAN, P. A., 1973. Paraplegia following
Paraplegia
intrathecal methotrexate. Journal of Pediatrics, 83, 988-992.
following intrathecal methotrexate. Report of a case and review of the literature. Cancer, 37, 1663-1668. HOPEWELL, J. W., 1990. The skin, its structure and response to ionising radiation. International Journal of Radiation Research, 57, 751-773.
MORRIS, A. D. & HOPEWELL, J. W., 1983. Combined effects of
GAGLIANO,
R.
G.
& CONSTANZI,
J.
J.,
1976.
HOPEWELL, J. W. & VAN DER KOGEL, A. J., 1988. Volume effect
in the spinal cord. British Journal of Radiology, 61, 973-975. HOPEWELL, J. W., MORRIS, A. D. & DIXON-BROWN, A., 1987.
The influence of field size on the late tolerance of the rat spinal cord to single doses of X rays. British Journal of Radiology, 61, 1099-1108. VAN DER KOGEL, A. J., 1979. Late Effects of Radiation in the Spinal Cord. PhD Thesis (University of Amsterdam, The Netherlands). , 1991. Central nervous system injury in small animals. In: Radiation Injury To The Nervous System, ed by P. K. Gutin,
156
radiation and methotrexate on the cells of the subependymal plate. Journal of the Royal Society of Medicine, 76, 848-852. PINKEL, D., HUSTU, H. O., AUR, R. J. A., SMITH, K., BORELLA,
L. D. & SIMONE, J., 1977. Radiotherapy in leukemia and lymphoma of children. Cancer, 39, 817-824. SCHWEINLE, J. E. & ALPERIN, J. B., 1980. Central nervous
system recurrence ten years after remission of acute lymphoblastic leukemia. Cancer, 45, 16-18. SKULLERUD,
D.
&
HALVORSEN,
K.,
1978.
Encephalomyelopathy following intrathecal methotrexate treatment in a child with acute leukemia. Cancer, 42, 1211-1215. WITHERS, H. R., TAYLOR, J. M. G. & MACIEJEWSKI, B., 1988.
Treatment volume and tissue tolerance. International Journal of Radiation Oncology, Biology, Physics, 14, 751-760.
The British Journal of Radiology, February 1992
