- Email: [email protected]
Spinal Nerve Tolerance to Single-Session Stereotactic Ablative Radiation Therapy
International Journal of
Radiation Oncology biology
physics
www.redjournal.org
Clinical Investigation
Spinal Nerve Tolerance to Single-Session Stereotactic Ablative Radiation Therapy Brian Hrycushko, PhD,* Albert J. van der Kogel, PhD,y Lauren Phillips, MD,z Michael R. Folkert, MD, PhD,* James W. Sayre, PhD,x Steven Vernino, MD, PhD,z Nima Hassan-Rezaeian, PhD,* Ryan D. Foster, PhD,k Yoshiya Yamada, MD,{ Robert Timmerman, MD,* and Paul M. Medin, PhD* *Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas; y Department of Human Oncology, University of Wisconsin, Madison, Wisconsin; zDepartment of Neurology and Neurotherapeutics, University of Texas Southwestern Medical Center, Dallas, Texas; x Departments of Biostatistics and Radiology, University of California Los Angeles, California; kLevine Cancer Institute, Atrium Health, Charlotte, North Carolina; {Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, New York Received Oct 18, 2018. Accepted for publication Mar 25, 2019.
Summary Stereotactic ablative radiation therapy is increasingly used for the treatment of spinal lesions. Data regarding the dose tolerance of peripheral nerves in the setting of stereotactic ablative radiation therapy are sparse, and there are no consensus guidelines. This
Purpose: This study was performed to determine the dose-related incidence of neuropathy from single-session irradiation of the C6-C8 spinal nerves using a pig model and to test the hypothesis that the spinal nerves and spinal cord have the same tolerance to full cross-sectional irradiation. Methods and Materials: Twenty-five Yucatan minipigs received study treatment. Each animal underwent computed tomography and magnetic resonance imaging for treatment planning, followed by single-session stereotactic ablative radiation therapy. A 1.5-cm length of the left-sided C6, C7, and C8 spinal nerves was targeted. Pigs were distributed into 5 groups with prescription doses of 16 (n Z 7), 18 (5), 20 (5), 22 (5), or 24 (3) Gy with corresponding maximum nerve doses of 17.3, 19.5, 21.6, 24.1, and 26.2 Gy. The neurologic status of all animals was followed for approximately 52 weeks by serial electrodiagnostic examination and daily observation of gait. Histopathologic
Reprint requests to: Paul M. Medin, PhD, Department of Radiation Oncology, UT Southwestern Medical Center, 2280 Inwood Rd, Dallas, TX 75390-9303. Tel: (214) 645-8591; E-mail: paul.medin@utsouthwestern. edu This project was funded by the Cancer Prevention & Research Institute of Texas. Disclosures: P.M., B.H., L.P., S.V., and R.F. were investigators and received salary support for this project funded by a research grant paid to their institution (University of Texas Southwestern Medical Center) from the Cancer Prevention & Research Institute of Texas. M.F. has received Int J Radiation Oncol Biol Phys, Vol. 104, No. 4, pp. 845e851, 2019 0360-3016/$ - see front matter Ó 2019 Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.ijrobp.2019.03.044
travel funding from Varian Medical Systems. Y.Y. has received payment for speaking engagements from Varian Medical Systems, Brainlab, Vision RT, and the Institute for Medical Education. Y.Y. is also on the Chordoma Foundation medical advisory board. R.T. is principal investigator of clinical research grants paid to his institution (University of Texas Southwestern Medical Center) from Varian Medical Systems, Elekta Oncology, and Accuray, Inc. P.M. teaches in radiosurgery courses sponsored by Brainlab. Supplementary material for this article can be found at https://doi.org/ 10.1016/j.ijrobp.2019.03.044.
846
International Journal of Radiation Oncology Biology Physics
Hrycushko et al.
work aims to determine the dose response of spinal nerves to single-session irradiation in a swine model. The dose tolerance is observed to be similar to that of the spinal cord in the same animal model.
examination of paraffin-embedded sections with Luxol fast blue/periodic acid-Schiff staining was performed on bilateral spinal nerves and the spinal cord. Results: Marked gait change was observed in 15 of the 25 irradiated pigs. Affected animals presented with a limp in their left front limb, and electromyography demonstrated evidence of denervation in C6 and C7 innervated muscles. Probit analysis showed the ED50 for gait change after irradiation of the spinal nerves to be 19.7 Gy (95% confidence interval, 18.5-21.1). The latency for all responding pigs was 9 to 15 weeks after irradiation. All symptomatic pigs had demyelination and fibrosis in their irradiated nerves, whereas contralateral nerves and spinal cord were normal. Conclusions: The ED50 for symptomatic neuropathy after full cross-sectional irradiation of the spinal nerves was found to be 19.7 Gy. The dose response of the C6-C8 spinal nerves is not significantly different from that of full cross-sectional irradiation of the spinal cord as observed in companion studies. Ó 2019 Elsevier Inc. All rights reserved.
Introduction Potent, few-fraction (oligofractionated) radiation therapy with modern image guided radiation therapy technologies has demonstrated efficacy and safety in multiple reports.1 Dose is often escalated to a target limited only by the perceived tolerance of surrounding normal tissues. The radiation tolerance of the central nervous system is of particular concern, considering the potential morbidity, functional impairment, and impact on quality of life from radiation-related injury. In recent years there has been a dramatic increase in the use of stereotactic ablative radiation therapy (SAbR) for spinal lesions, both primary and metastatic. Mostly singleinstitution analyses have demonstrated the effectiveness of single-fraction dose escalation to 16 to 24 Gy with approximately 85% local control and significant symptomatic pain relief.2-8 The spinal cord dose tolerance has been well studied in different animal models,9-13 and few spinal cord injuries have been reported in humans receiving SAbR.14,15 Although limiting the dose received by the spinal cord is appropriately prioritized, the adjacent peripheral nerves are also at risk for radiation injury. The Radiation Therapy Oncology Group has consistently specified dose constraints for peripheral nerves such as the brachial plexus to be greater than those for the spinal cord (eg, Radiation Therapy Oncology Group 0631 and 0915), but there are no consensus guidelines for spinal nerves. The spinal nerves are often encompassed within the planning target volume (PTV) when treating spinal lesions, making dose reduction to the nerves a challenge. The clinical significance of this was demonstrated by Stubblefield et al in their recent report of 14 peripheral nerve injuries affecting the nerve root or plexus after single-fraction spinal SAbR to doses of 18 to 26 Gy.16 Their study determined the percentage of patients developing neuropathies, but the dose distribution to the spinal nerves was not analyzed. All injuries occurred in
patients with tumors in the cervical or lumbosacral spine receiving prescription doses 24 Gy. Injury to the spinal nerves may result in symptoms of peripheral neuropathy. Much of the clinical and animal data on radiationinduced peripheral neuropathy are derived from intraoperative radiation therapy (IORT) of nerve segments and plexuses distal from the spinal nerves.17-22 Translation of peripheral nerve tolerances from IORT to SAbR applications may be problematic because of the unknown impact of surgical manipulation on the radiosensitivity of affected tissues. For example, the tolerance of peripheral nerves may be altered by devascularization if nerves are skeletonized. While evaluating vertebral bone tolerance to single-session spinal irradiation in a pig model, our group observed gait changes in all animals receiving maximum point doses 24.3 0.8 Gy to bilateral C5-C8 cervical spinal nerves.23 After re-evaluation of spinal cord and spinal nerve doses, 2 additional animals were irradiated to only the left-sided C5-C8 spinal nerves. Both of these pigs experienced gait changes at an average maximum point dose of 24.9 Gy. No pig experienced gait changes when the maximum point dose to the spinal nerve was 19.0 Gy, whereas all pigs with a maximum point dose 24.1 Gy experienced changes in gait. Observations of dose response in the pig study were consistent with the latency and response for the spinal cord in the same animal model.13 Radiation-induced peripheral neuropathy can present with a wide range of symptoms and can result in significant morbidity, causing functional impairment and affecting quality of life and emotional wellbeing. The peripheral nervous system innervates all organs; thus, gains in knowledge of regional tolerances to ablative radiation, a long-term goal of ours, will enable clinicians to balance maximal treatment effect and prevention of normal tissue injury for a variety of disease sites. This work focuses on the spinal nerves and is designed to determine the ED50 for gait change from single-session irradiation of the C6-C8 spinal nerves in a pig model in the absence of surgical manipulation. Based on the observations from animal and
Volume 104 Number 4 2019
human data, we hypothesize that spinal nerves have a single-fraction dose response similar to that of the spinal cord.
Methods and Materials This study conformed to all national and local regulations regarding the use of animals for research and was approved by the Institutional Animal Care and Use Committee. For all procedures, animals were anesthetized with a mixture of Telazol and Xylazine and maintained on isoflurane. A total of 25 female Yucatan minipigs weighing 21 to 43 kg and aged 41 to 49 weeks were enrolled. All pigs received targeted unilateral irradiation to a 1.5-cm length of each of the left-sided C6-C8 spinal nerves. Right-sided nerves of the same vertebral levels served as controls. All animals received a treatment planning computed tomography scan (Brilliance CT Big Bore 16 slice, Philips Healthcare) with a 0.1-cm slice thickness and 34- to 40ecm field of view. T2weighted spectral attenuated inversion recovery, T1weighted turbo spin echo, and 3D short TI inversion recovery composite images were acquired on a 3 T magnetic resonance imaging unit (Achieva, Philips Medical Systems, Best, Netherlands) to visualize the spinal nerves for treatment planning. Animals were positioned supine in molded full-body vacuum immobilization cushions for each computed tomography scan, magnetic resonance imaging scan, and treatment procedure. Treatment planning was performed using either iPlan v4.5.1 (Brainlab, Feldkirchen, Germany) or Eclipse v13.7 (Varian Medical Systems, Palo Alto, CA) software. Magnetic resonance images were fused with the computed tomography image set to identify and contour a 1.5-cm length
Fig. 1. Spinal nerve with superimposed dose distribution. (A) and (B) Nerve segments cut for longitudinal and transverse histologic sections.
Spinal nerve tolerance to SAbR
847
of each of the C6-C8 spinal nerves beginning approximately 0.4 cm from the edge of the spinal cord (Fig. 1). A 0.1-cm circumferential PTV expansion was applied to each nerve, but the contoured length remained unchanged. Organs at risk including the spinal cord, the C5-C7 vertebral bodies, and the contralateral C6-C8 spinal nerves were contoured to analyze the dose received. Treatment plans consisted of four 6 MV, non-coplanar conformal arcs defined with a multileaf collimator to ensure the full crosssection of a 1.5-cm length of each targeted nerve was covered by at least 95% of the prescription dose and no more than 110% of the prescription dose (see Table 1 for dose metrics and Fig. E1, available online at https://doi.org/ 10.1016/j.ijrobp.2019.03.044, for sample dose distribution from the treatment planning system). Single-session treatment was delivered using either a Vero (Brainlab) or TrueBeam (Varian Medical Systems) image-guided linear accelerator. Cone beam computed tomography was used for target localization, and the 6-degrees-of-freedom treatment couch was rotated and translated until the actual position and required treatment position differed by less than 0.1 cm Table 1 Dose and latency period until initial presentation of motor deficits for individual irradiated pigs Spinal Spinal Spinal Spinal Rx nerve nerve Contralateral cord Cord dose, Dmax, D1.5 cm, spinal nerve Dmax, V10Gy, Latency, Dmax, Gy Gy Gy Gy cm3 wk* Gy 16 16 16 16 16 16 16 18 18 18 18 18 20 20 20 20 20 22 22 22 22 22 24 24 24
17.5 17.4 17.3 17.3 17.2 17.3 17.2 19.6 19.6 19.3 19.4 19.4 21.9 21.6 21.6 21.5 21.6 24.2 24.0 23.7 23.9 23.7 26.2 26.1 26.0
15.6 15.7 16.5 15.9 16.1 15.9 16.0 18.6 17.9 18.2 18.2 17.7 19.9 20.0 20.3 20.4 20.0 21.9 21.9 21.9 22.3 22.4 24.0 23.9 24.0
4.8 3.6 3.4 4.0 3.1 2.6 3.1 4.5 4.3 5.0 4.7 4.3 4.7 6.3 4.9 4.6 4.8 4.0 5.6 5.6 5.1 5.6 5.7 5.5 5.5
10.1 10.6 9.0 4.9 5.4 5.0 8.4 11.8 9.7 11.1 12.0 5.9 8.7 10.9 12.0 11.6 5.0 9.4 12.6 10.2 8.9 9.5 11.1 12.8 9.6
0.046 0.056 0.005 0 0 0 0 0.112 0.022 0.105 0.225 0 0 0.081 0.166 0.132 0 0.011 0.2 0.04 0.007 0.023 0.09 0.391 0.001
NA NA NA NA NA NA NA NA 15.0 NA NA 9.4 10.7 12.0 12.6 13.9 12.3 11.7 11.0 9.3 10.3 11.0 11.9 13.3 10.4
Abbreviations: D1.5cm Z minimum dose to 1.5-cm length; Dmax Z minimum dose to 0.035 cm3 volume; V10Gy Z volume receiving greater than 10 Gy. * NA Z not applicable; no neurologic deficits during stated followup period.
Hrycushko et al.
and 1 degree in each of the 3 primary planes. A prescription dose of 16 to 24 Gy was delivered in approximately 6 to 10 minutes at a rate of 2 to 4 Gy/min to the nerve. Animals were followed for a minimum of 52 weeks or until a change in gait was observed. The general health of animals was observed daily by animal husbandry personnel and approximately weekly by the study principal investigator with attention toward asymmetric gait, unusual restlessness, vocalization, loss of mobility, guarding of a painful area, unkempt appearance, open sores, loss of appetite, and weight loss. Gait was observed with the animals walking freely in a large space. Endpoint response was defined as any study-related change in gait. Electrodiagnostic examination was performed on all pigs at approximately e1, 2, 10, 20, 30, and 50 weeks (or until reaching the study endpoint) to evaluate for the presence of nerve dysfunction. Nerve conduction studies were performed to assess for potential loss of axons in motor and sensory nerves indicative of peripheral nerve disorder. Needle electromyography (EMG) was performed to assess for increased insertional activity and abnormal spontaneous activity. Animals recognized to have a change in gait were evaluated by a veterinarian for symptoms indicative of pain and were humanely killed. The bilateral C6-C8 spinal nerves and the spinal cord were harvested for histologic processing. All tissues were fixed in formalin before being sectioned and processed for embedding in paraffin. Spinal nerve segments were cut in longitudinal and transverse sections (inset of Fig. 1). Sections of spinal nerve and spinal cord were stained with a Luxol fast blue/ periodic acid-Schiff combination for light microscopy. The dose-related incidence of gait change was used as a clinically relevant endpoint to obtain quantal data for probit analysis using IBM SPSS Statistics v24 software (IBM, Armonk, NY). The analysis of dose response was performed using the maximum dose (0.035 cm3 volume) to the targeted spinal nerve, which was calculated as the mean of the maximum doses to nerves C6-C8 for each pig. The full dose-response curve and 95% confidence bands were generated. The dose associated with a 50% incidence of motor deficit (ED50) was determined and compared with the ED50 for spinal cord myelopathy from Medin et al in the same animal model.13
Results Dosimetric and response data for individual pigs are presented in Table 1 and include (1) prescription dose, (2) maximum spinal nerve dose, (3) minimum dose to the 1.5cm length of targeted nerve, (4) maximum dose to the contralateral spinal nerve, (5) maximum dose to the spinal cord, (6) volume receiving at least 10 Gy, and (7) latency to response. Fifteen of 25 pigs developed front limb motor changes on the irradiated (left) side. No deficits were observed on the unirradiated (right) side of any pig. Motor deficits presented as a mild limp with occasional dragging
International Journal of Radiation Oncology Biology Physics
of the foot and progressed to limb weakness. None of the pigs showed behavior indicative of pain. The latency period until initial presentation of motor deficits ranged from 9 to 15 weeks. The dose-response curve and 95% confidence intervals for the spinal nerve are presented in Figure 2 (P Z .74) with superimposed dose-response points from a previous spinal cord tolerance study.13 A test for parallelism shows no significant difference between the spinal nerve and spinal cord data (P Z .584). No gait changes were observed when the maximum point dose to the spinal nerves was <19.4 Gy, and all pigs had gait changes when the maximum point dose was 21.5 Gy. EMG results are summarized in Table 2. Evidence of denervation in the C6 and C7 innervated forelimb muscles (deltoid, triceps, and forelimb extensor muscles) was observed in 76% of the animals, but this was not predictive of gait changes. Denervation in the form of increased insertional activity or abnormal spontaneous activity was observed in 50% of asymptomatic pigs but more commonly in symptomatic pigs (93%). The median latency for the appearance of EMG changes was 21.9 weeks postirradiation. In some instances, denervation changes were not observed immediately after motor deficit but rather several weeks later, which in practice is not unexpected as a neuropathic lesion evolves. Nerve conduction studies assessing bilateral median and ulnar motor nerves and bilateral ulnar sensory nerves showed no significant change in compound motor action potential amplitude or sensory nerve action potential amplitude between baseline and postirradiation measurements. As a whole, the pattern of electrodiagnostic findings in affected pigs was most consistent with polyradiculopathy with motor axonal loss. Given the limitations of testing in this model, electrophysiologic evidence of demyelination or conduction block could not be accurately demonstrated. Histolopathologic findings were in agreement with the gait-change data. Left-sided unilateral degeneration of (5/5)
1
(5/5)
(3/3)
0.9 0.8 0.7 Probability
848
0.6 0.5 (2/5)
0.4 0.3 0.2 0.1 0 12
(0/7)
14
16
18 20 24 22 Maximum Dose (Gy)
26
28
Fig. 2. Dose-response curve (solid line) for motor neurologic deficit with 95% confidence intervals (dashed line). (O) are spinal nerve data points, and (x) are spinal cord data points from Medin et al.13
Volume 104 Number 4 2019 Table 2
Spinal nerve tolerance to SAbR
849
Summary of electrodiagnostic results
Rx Dose, Gy
Symptomatic (Motor Deficit), n
16 (n Z 7) 18 (n Z 5) 20 (n Z 5) 22 (n Z 5) 24 (n Z 3) Entire cohort (n Z 25)
0 2 5 5 3 15
Abnormal EMG (signs of denervation, n 4 (all asymptomatic) 3 (1 asymptomatic) 4 5 3 Total: 19 Asymptomatic: 5 Symptomatic: 14
Appearance of EMG changes, median (range), wk 37.5 26.6 15.5 12.8 17 21.9
(30-50) (10-50) (2-30) (2-30) (2-30) (2-50)
Abbreviation: EMG Z electromyography.
spinal nerves including loss of myelin, axonal swelling, and various degrees of fibrosis was observed in all pigs with motor deficits. These changes do not exclude microvascular injury occurring at an early stage of pathogenesis, but at the time the tissues were collected specific vascular injury was not a dominating feature. The replacement of myelinated fibers by fibrotic tissue was clearly visible in the irradiated (left) nerves, whereas the unirradiated (right) side was normal (Fig. 3). No histologic damage was observed in the white or gray matter of the spinal cord tissue of any animal in this study. Some spinal cord sections showed 1-sided scattered fibrosis of nerve roots. In those cases, the leftsided longitudinal spinal nerve sections also showed fibrosis and scattered demyelination in the ganglia, indicating retrograde degeneration.
Discussion Only limited information is available to characterize the dose response of the peripheral nervous system to SAbR, and there is a general perception that peripheral nerves are more tolerant than the spinal cord to irradiation, as evidenced by current tissue tolerance recommendations and clinical trial guidelines.24-26 This study was performed to
characterize the dose response of a specific set of peripheral nerves, the spinal nerves, to single-session external beam irradiation in an animal model of similar anatomic size to humans. The Yucatan minipig was a suitable animal model for this study because its anatomic dimensions allowed for a 1.5-cm length of spinal nerve to be visualized and targeted while maintaining a relatively low dose to the spinal cord. The Yucatan minipig has also previously been used to study the tolerance of the spinal cord to single-session irradiation.13 Because of a sparsity of data, the relationship between the tolerance of the nervous systems in pigs and humans is unknown, and caution is warranted when extrapolating from animal to human populations. The dose-response of the spinal cord has marked similarities across a variety of animal models.9-11,13 Data on peripheral nerve injury have a much broader range of observed tolerances, possibly a result of differences in animal models, irradiation technique, or anatomic site.17,19,27-34 Data from IORT studies targeting different peripheral nerves in dogs have resulted in a range of tolerance values. Kinsella et al surgically exposed and irradiated the lumbosacral plexus and sciatic nerves in an American Foxhound model to singlefraction doses of 20 to 75 Gy, and 19 of the 21 irradiated dogs showed clinically evident nerve injury.17 A follow-up study designed to determine the threshold dose for
Fig. 3. Spinal nerve (C7) 96 days after 22 Gy. (A) Left side irradiated spinal nerve showing degenerating nerve fibers (loss of blue-stained myelin) and replacement with fibrosis (red stain), eg, arrow. (B) Nontargeted right-side spinal nerve. Luxol fast blue/periodic acid-Schiff stain, 10 objective.
850
Hrycushko et al.
peripheral nerve injury from IORT showed 100% paresis in animals receiving 20 Gy, whereas no dog receiving 10 or 15 Gy showed signs of injury.18 In a tolerance study of lumbar spinal nerves and proximal femoral nerves of beagles, LeCouteur et al reported an ED50 of 19.5 Gy.19 In similar studies performed at the same institution, the tolerance of the sciatic nerve and mid-femoral nerves was investigated. One of these studies found that none of the 5 dogs receiving 20 Gy experienced neurologic deficit, and only 2 of 5 dogs experienced deficits after 28 Gy.28 In a second study, the ED50 for hind limb paresis was estimated to be 22 Gy.33 The tolerance of the lumbosacral plexus and sciatic nerve to IORT was investigated at the National Cancer Institute, and an ED50 of 17.2 Gy was reported.29 Lin et al delivered a maximal dose of 25 Gy using external beam irradiation to the sciatic nerves of 12 rabbits, and none developed leg weakness within the 7-month follow-up period.34 In contrast, we observed that all of 13 pigs receiving a maximal dose of 21.6 Gy experienced motor neurologic deficits within 13.9 weeks. The tolerance disagreement between the pig and rabbit studies may be explained by the existence of a length effect, although response differences from species variation or radiosensitivity variation between anatomic sites (spinal nerve vs. sciatic nerve) is also plausible. For the pig spinal nerves, a 1.5-cm length received the prescription dose and the hot spot within the targeted nerve was approximately 8% greater than the prescription dose, whereas the rabbit sciatic nerves were irradiated using a 0.77-cm diameter stereotactic cone and the maximum dose was reported. Although not specifically stated, irradiation with a circular collimator typically results in a dose of approximately 50% of the maximum extending the length of the diameter. The existence of a dose-length effect has not been studied in the peripheral nervous system but is well established in the spinal cord. Two independent laboratories have observed that the dose tolerance of the rat spinal cord increases dramatically as the irradiated length is decreased below a threshold of approximately 1 cm for single-session irradiation.9,35-38 Stubblefield et al recently reported 14 peripheral nerve injuries from a series of 447 patients (557 treatments) who were treated with single-session radiosurgery for spinal tumors.16 For these treatments, nerve roots were included in the PTV and received up to 26 Gy. The lengths of nerve irradiated in this series of patients are not reported, but a dose-length effect may explain why the rate of neuropathy was very low despite the large doses delivered. This study was performed to determine the doseresponse curve for symptomatic neuropathy after singlesession irradiation of the C6-C8 spinal nerves. We observed that the tolerance to full cross-sectional irradiation of the spinal nerves is not significantly different from that of the spinal cord.13 It is important to recognize that doseresponse data previously reported for the spinal cord using this same pig model included spinal nerves in the irradiated volume; thus, the observed motor deficits may
International Journal of Radiation Oncology Biology Physics
have been influenced by the response of the spinal nerves, even though neurologic deficits and histologic change (demyelination, necrosis) of the spinal cord were highly correlated in the spinal cord studies. Exact dose-length data for the spinal nerves included in our spinal cord tolerance studies are not available, nor were tissue samples of the spinal nerves collected. A similar dose tolerance in the spinal nerves and spinal cord may indicate a common pathogenesis, most likely beginning with microvascular injury and progressing to late vasculopathies. It is also important to recognize that almost all historic spinal cord tolerance studies using animals included nerve rootlets or spinal nerves along with the spinal cord in their irradiated tissues. To our knowledge, only 1 spinal cord tolerance study has specifically excluded nerve rootlets and spinal nerves. Bijl et al irradiated the midline of the spinal cord with a narrow proton beam having a penumbra of 0.8 mm and observed that the central white matter and grey matter of the spinal cord are much more radioresistant than the lateral white matter.39 As part of our long-term goal to characterize the dose-response of the peripheral nervous system to ablative radiation therapy, future studies aim to investigate the existence of a dose-length effect for the spinal nerves and to determine if radiation tolerance changes as the peripheral nerves travel from the vertebral column to the level of the divisions of the brachial plexus.
Conclusions The ED50 for symptomatic neuropathy following full crosssectional irradiation of the spinal nerves was found to be 19.7 Gy. The dose-response of the C6-C8 spinal nerves is not significantly different from that of full cross-sectional irradiation of the spinal cord as observed in companion studies.
References 1. Folkert MR, Timmerman RD. Stereotactic ablative body radiosurgery (SABR) or stereotactic body radiation therapy (SBRT). Adv Drug Deliv Rev 2017;109:3-14. 2. Yamada Y, Bilsky MH, Lovelock DM, et al. High-dose, single-fraction image-guided intensity-modulated radiotherapy for metastatic spinal lesions. Int J Radiat Oncol Biol Phys 2008;71:484-490. 3. Garg AK, Shiu AS, Yang J, et al. Phase 1/2 trial of single-session stereotactic body radiotherapy for previously unirradiated spinal metastases. Cancer 2012;118:5069-5077. 4. Sellin JN, Suki D, Harsh V, et al. Factors affecting survival in 43 consecutive patients after surgery for spinal metastases from thyroid carcinoma. J Neurosurg Spine 2015;23:419-428. 5. Bate BG, Khan NR, Kimball BY, Gabrick K, Weaver J. Stereotactic radiosurgery for spinal metastases with or without separation surgery. J Neurosurg Spine 2015;22:409-415. 6. Chang EL, Shiu AS, Mendel E, et al. Phase I/II study of stereotactic body radiotherapy for spinal metastasis and its pattern of failure. J Neurosurg Spine 2007;7:151-160. 7. Gerszten PC, Burton SA, Welch WC, et al. Single-fraction radiosurgery for the treatment of spinal breast metastases. Cancer 2005; 104:2244-2254.
Volume 104 Number 4 2019 8. Staehler M, Haseke N, Nuhn P, et al. Simultaneous anti-angiogenic therapy and single-fraction radiosurgery in clinically relevant metastases from renal cell carcinoma. BJU Int 2011;108:673-678. 9. Hopewell JW, Morris AD, Dixon-Brown A. The influence of field size on the late tolerance of the rat spinal cord to single doses of X rays. Br J Radiol 1987;60:1099-1108. 10. Lo YC, McBride WH, Withers HR. The effect of single doses of radiation on mouse spinal cord. Int J Radiat Oncol Biol Phys 1992;22: 57-63. 11. Knowles JF. The effects of single dose X-irradiation on the guinea-pig spinal cord. Int J Radiat Biol Relat Stud Phys Chem Med 1981;40:265275. 12. Powers BE, Thames HD, Gillette SM, Smith C, Beck ER, Gillette EL. Volume effects in the irradiated canine spinal cord: Do they exist when the probability of injury is low? Radiother Oncol 1998;46:297-306. 13. Medin PM, Foster RD, van der Kogel AJ, Sayre JW, McBride WH, Solberg TD. Spinal cord tolerance to single-session uniform irradiation in pigs: Implications for a dose-volume effect. Radiother Oncol 2013;106:101-105. 14. Ryu S, Jin JY, Jin R, et al. Partial volume tolerance of the spinal cord and complications of single-dose radiosurgery. Cancer 2007;109:628636. 15. Gibbs IC, Patil C, Gerszten PC, Adler JR Jr., Burton SA. Delayed radiation-induced myelopathy after spinal radiosurgery. Neurosurgery 2009;64:A67-A72. 16. Stubblefield MD, Ibanez K, Riedel ER, et al. Peripheral nervous system injury after high-dose single-fraction image-guided stereotactic radiosurgery for spine tumors. Neurosurg Focus 2017;42:E12. 17. Kinsella TJ, Sindelar WF, DeLuca AM, et al. Tolerance of peripheral nerve to intraoperative radiotherapy (IORT): Clinical and experimental studies. Int J Radiat Oncol Biol Phys 1985;11:1579-1585. 18. Kinsella TJ, DeLuca AM, Barnes M, et al. Threshold dose for peripheral neuropathy following intraoperative radiotherapy (IORT) in a large animal model. Int J Radiat Oncol Biol Phys 1991;20:697-701. 19. LeCouteur RA, Gillette EL, Powers BE, Child G, McChesney SL, Ingram JT. Peripheral neuropathies following experimental intraoperative radiation therapy (IORT). Int J Radiat Oncol Biol Phys 1989;17:583-590. 20. Sindelar WF, Kinsella TJ, Chen PW, et al. Intraoperative radiotherapy in retroperitoneal sarcomas. Final results of a prospective, randomized, clinical trial. Arch Surg 1993;128:402-410. 21. Azinovic I, Calvo FA, Puebla F, Aristu J, Martı´nez-Monge R. Longterm normal tissue effects of intraoperative electron radiation therapy (IOERT): Late sequelae, tumor recurrence, and second malignancies. Int J Radiat Oncol Biol Phys 2001;49:597-604. 22. Petersen IA, Haddock MG, Donohue JH, et al. Use of intraoperative electron beam radiotherapy in the management of retroperitoneal soft tissue sarcomas. Int J Radiat Oncol Biol Phys 2002;52:469-475. 23. Medin PM, Foster RD, van der Kogel AJ, et al. Paralysis following stereotactic spinal irradiation in pigs suggests a tolerance constraint for single-session irradiation of the spinal nerve. Radiother Oncol 2013;109:107-111.
Spinal nerve tolerance to SAbR
851
24. Ryu S, Pugh SL, Gerszten PC, et al. RTOG 0631 phase 2/3 study of image guided stereotactic radiosurgery for localized (1-3) spine metastases: Phase 2 results. Pract Radiat Oncol 2014;4:76-81. 25. Benedict SH, Yenice KM, Followill D, et al. Stereotactic body radiation therapy: The report of AAPM Task Group 101. Med Phys 2010; 37:4078-4101. 26. Emami B. Tolerance of normal tissue to therapeutic radiation. Radiother Oncol 2013;1:35-48. 27. Hassler O. Cellular kinetics of the peripheral nerve and striated muscle after a single dose of x-rays. A histological and autoradiographical study, using 3H-thymidine. Z Zellforsch Mikrosk Anat 1968; 85:62-66. 28. Vujaskovic Z, Gillette SM, Powers BE, et al. Intraoperative radiation (IORT) injury to sciatic nerve in a large animal model. Radiother Oncol 1994;30:133-139. 29. Johnstone PA, DeLuca AM, Bacher JD, et al. Clinical toxicity of peripheral nerve to intraoperative radiotherapy in a canine model. Int J Radiat Oncol Biol Phys 1995;32:1031-1034. 30. Gillette EL, Mahler PA, Powers BE, Gillette SM, Vujaskovic Z. Late radiation injury to muscle and peripheral nerves. Int J Radiat Oncol Biol Phys 1995;31:1309-1318. 31. Powell S, Cooke J, Parsons C. Radiation-induced brachial plexus injury: Follow-up of two different fractionation schedules. Radiother Oncol 1990;18:213-220. 32. Sindelar WF, Hoekstra H, Restrepo C, Kinsella TJ. Pathological tissue changes following intraoperative radiotherapy. Am J Clin Oncol 1986; 9:504-509. 33. Vujaskovic Z, Gillette SM, Powers BE, et al. Effects of intraoperative irradiation and intraoperative hyperthermia on canine sciatic nerve: Neurologic and electrophysiologic study. Int J Radiat Oncol Biol Phys 1996;34:125-131. 34. Lin Z, Wu VW, Ju W, Yamada Y, Chen L. Radiation-induced changes in peripheral nerve by stereotactic radiosurgery: A study on the sciatic nerve of rabbit. J Neurooncol 2011;102:179-185. 35. Bijl HP, van Luijk P, Coppes RP, Schippers JM, Konings AW, van der Kogel AJ. Dose-volume effects in the rat cervical spinal cord after proton irradiation. Int J Radiat Oncol Biol Phys 2002;52: 205-211. 36. Bijl HP, van Luijk P, Coppes RP, Schippers JM, Konings AW, van der Kogel AJ. Unexpected changes of rat cervical spinal cord tolerance caused by inhomogeneous dose distributions. Int J Radiat Oncol Biol Phys 2003;57:274-281. 37. Philippens ME, Pop LA, Visser AG, Schellekens SA, van der Kogel AJ. Dose-volume effects in rat thoracolumbar spinal cord: An evaluation of NTCP models. Int J Radiat Oncol Biol Phys 2004;60: 578-590. 38. van der Kogel AJ. Dose-volume effects in the spinal cord. Radiother Oncol 1993;29:105-109. 39. Bijl HP, van Luijk P, Coppes RP, Schippers JM, Konings AW, van Der Kogel AJ. Regional differences in radiosensitivity across the rat cervical spinal cord. Int J Radiat Oncol Biol Phys 2005;61: 543-551.
Radiation Oncology biology
physics
www.redjournal.org
Clinical Investigation
Spinal Nerve Tolerance to Single-Session Stereotactic Ablative Radiation Therapy Brian Hrycushko, PhD,* Albert J. van der Kogel, PhD,y Lauren Phillips, MD,z Michael R. Folkert, MD, PhD,* James W. Sayre, PhD,x Steven Vernino, MD, PhD,z Nima Hassan-Rezaeian, PhD,* Ryan D. Foster, PhD,k Yoshiya Yamada, MD,{ Robert Timmerman, MD,* and Paul M. Medin, PhD* *Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas; y Department of Human Oncology, University of Wisconsin, Madison, Wisconsin; zDepartment of Neurology and Neurotherapeutics, University of Texas Southwestern Medical Center, Dallas, Texas; x Departments of Biostatistics and Radiology, University of California Los Angeles, California; kLevine Cancer Institute, Atrium Health, Charlotte, North Carolina; {Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, New York Received Oct 18, 2018. Accepted for publication Mar 25, 2019.
Summary Stereotactic ablative radiation therapy is increasingly used for the treatment of spinal lesions. Data regarding the dose tolerance of peripheral nerves in the setting of stereotactic ablative radiation therapy are sparse, and there are no consensus guidelines. This
Purpose: This study was performed to determine the dose-related incidence of neuropathy from single-session irradiation of the C6-C8 spinal nerves using a pig model and to test the hypothesis that the spinal nerves and spinal cord have the same tolerance to full cross-sectional irradiation. Methods and Materials: Twenty-five Yucatan minipigs received study treatment. Each animal underwent computed tomography and magnetic resonance imaging for treatment planning, followed by single-session stereotactic ablative radiation therapy. A 1.5-cm length of the left-sided C6, C7, and C8 spinal nerves was targeted. Pigs were distributed into 5 groups with prescription doses of 16 (n Z 7), 18 (5), 20 (5), 22 (5), or 24 (3) Gy with corresponding maximum nerve doses of 17.3, 19.5, 21.6, 24.1, and 26.2 Gy. The neurologic status of all animals was followed for approximately 52 weeks by serial electrodiagnostic examination and daily observation of gait. Histopathologic
Reprint requests to: Paul M. Medin, PhD, Department of Radiation Oncology, UT Southwestern Medical Center, 2280 Inwood Rd, Dallas, TX 75390-9303. Tel: (214) 645-8591; E-mail: paul.medin@utsouthwestern. edu This project was funded by the Cancer Prevention & Research Institute of Texas. Disclosures: P.M., B.H., L.P., S.V., and R.F. were investigators and received salary support for this project funded by a research grant paid to their institution (University of Texas Southwestern Medical Center) from the Cancer Prevention & Research Institute of Texas. M.F. has received Int J Radiation Oncol Biol Phys, Vol. 104, No. 4, pp. 845e851, 2019 0360-3016/$ - see front matter Ó 2019 Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.ijrobp.2019.03.044
travel funding from Varian Medical Systems. Y.Y. has received payment for speaking engagements from Varian Medical Systems, Brainlab, Vision RT, and the Institute for Medical Education. Y.Y. is also on the Chordoma Foundation medical advisory board. R.T. is principal investigator of clinical research grants paid to his institution (University of Texas Southwestern Medical Center) from Varian Medical Systems, Elekta Oncology, and Accuray, Inc. P.M. teaches in radiosurgery courses sponsored by Brainlab. Supplementary material for this article can be found at https://doi.org/ 10.1016/j.ijrobp.2019.03.044.
846
International Journal of Radiation Oncology Biology Physics
Hrycushko et al.
work aims to determine the dose response of spinal nerves to single-session irradiation in a swine model. The dose tolerance is observed to be similar to that of the spinal cord in the same animal model.
examination of paraffin-embedded sections with Luxol fast blue/periodic acid-Schiff staining was performed on bilateral spinal nerves and the spinal cord. Results: Marked gait change was observed in 15 of the 25 irradiated pigs. Affected animals presented with a limp in their left front limb, and electromyography demonstrated evidence of denervation in C6 and C7 innervated muscles. Probit analysis showed the ED50 for gait change after irradiation of the spinal nerves to be 19.7 Gy (95% confidence interval, 18.5-21.1). The latency for all responding pigs was 9 to 15 weeks after irradiation. All symptomatic pigs had demyelination and fibrosis in their irradiated nerves, whereas contralateral nerves and spinal cord were normal. Conclusions: The ED50 for symptomatic neuropathy after full cross-sectional irradiation of the spinal nerves was found to be 19.7 Gy. The dose response of the C6-C8 spinal nerves is not significantly different from that of full cross-sectional irradiation of the spinal cord as observed in companion studies. Ó 2019 Elsevier Inc. All rights reserved.
Introduction Potent, few-fraction (oligofractionated) radiation therapy with modern image guided radiation therapy technologies has demonstrated efficacy and safety in multiple reports.1 Dose is often escalated to a target limited only by the perceived tolerance of surrounding normal tissues. The radiation tolerance of the central nervous system is of particular concern, considering the potential morbidity, functional impairment, and impact on quality of life from radiation-related injury. In recent years there has been a dramatic increase in the use of stereotactic ablative radiation therapy (SAbR) for spinal lesions, both primary and metastatic. Mostly singleinstitution analyses have demonstrated the effectiveness of single-fraction dose escalation to 16 to 24 Gy with approximately 85% local control and significant symptomatic pain relief.2-8 The spinal cord dose tolerance has been well studied in different animal models,9-13 and few spinal cord injuries have been reported in humans receiving SAbR.14,15 Although limiting the dose received by the spinal cord is appropriately prioritized, the adjacent peripheral nerves are also at risk for radiation injury. The Radiation Therapy Oncology Group has consistently specified dose constraints for peripheral nerves such as the brachial plexus to be greater than those for the spinal cord (eg, Radiation Therapy Oncology Group 0631 and 0915), but there are no consensus guidelines for spinal nerves. The spinal nerves are often encompassed within the planning target volume (PTV) when treating spinal lesions, making dose reduction to the nerves a challenge. The clinical significance of this was demonstrated by Stubblefield et al in their recent report of 14 peripheral nerve injuries affecting the nerve root or plexus after single-fraction spinal SAbR to doses of 18 to 26 Gy.16 Their study determined the percentage of patients developing neuropathies, but the dose distribution to the spinal nerves was not analyzed. All injuries occurred in
patients with tumors in the cervical or lumbosacral spine receiving prescription doses 24 Gy. Injury to the spinal nerves may result in symptoms of peripheral neuropathy. Much of the clinical and animal data on radiationinduced peripheral neuropathy are derived from intraoperative radiation therapy (IORT) of nerve segments and plexuses distal from the spinal nerves.17-22 Translation of peripheral nerve tolerances from IORT to SAbR applications may be problematic because of the unknown impact of surgical manipulation on the radiosensitivity of affected tissues. For example, the tolerance of peripheral nerves may be altered by devascularization if nerves are skeletonized. While evaluating vertebral bone tolerance to single-session spinal irradiation in a pig model, our group observed gait changes in all animals receiving maximum point doses 24.3 0.8 Gy to bilateral C5-C8 cervical spinal nerves.23 After re-evaluation of spinal cord and spinal nerve doses, 2 additional animals were irradiated to only the left-sided C5-C8 spinal nerves. Both of these pigs experienced gait changes at an average maximum point dose of 24.9 Gy. No pig experienced gait changes when the maximum point dose to the spinal nerve was 19.0 Gy, whereas all pigs with a maximum point dose 24.1 Gy experienced changes in gait. Observations of dose response in the pig study were consistent with the latency and response for the spinal cord in the same animal model.13 Radiation-induced peripheral neuropathy can present with a wide range of symptoms and can result in significant morbidity, causing functional impairment and affecting quality of life and emotional wellbeing. The peripheral nervous system innervates all organs; thus, gains in knowledge of regional tolerances to ablative radiation, a long-term goal of ours, will enable clinicians to balance maximal treatment effect and prevention of normal tissue injury for a variety of disease sites. This work focuses on the spinal nerves and is designed to determine the ED50 for gait change from single-session irradiation of the C6-C8 spinal nerves in a pig model in the absence of surgical manipulation. Based on the observations from animal and
Volume 104 Number 4 2019
human data, we hypothesize that spinal nerves have a single-fraction dose response similar to that of the spinal cord.
Methods and Materials This study conformed to all national and local regulations regarding the use of animals for research and was approved by the Institutional Animal Care and Use Committee. For all procedures, animals were anesthetized with a mixture of Telazol and Xylazine and maintained on isoflurane. A total of 25 female Yucatan minipigs weighing 21 to 43 kg and aged 41 to 49 weeks were enrolled. All pigs received targeted unilateral irradiation to a 1.5-cm length of each of the left-sided C6-C8 spinal nerves. Right-sided nerves of the same vertebral levels served as controls. All animals received a treatment planning computed tomography scan (Brilliance CT Big Bore 16 slice, Philips Healthcare) with a 0.1-cm slice thickness and 34- to 40ecm field of view. T2weighted spectral attenuated inversion recovery, T1weighted turbo spin echo, and 3D short TI inversion recovery composite images were acquired on a 3 T magnetic resonance imaging unit (Achieva, Philips Medical Systems, Best, Netherlands) to visualize the spinal nerves for treatment planning. Animals were positioned supine in molded full-body vacuum immobilization cushions for each computed tomography scan, magnetic resonance imaging scan, and treatment procedure. Treatment planning was performed using either iPlan v4.5.1 (Brainlab, Feldkirchen, Germany) or Eclipse v13.7 (Varian Medical Systems, Palo Alto, CA) software. Magnetic resonance images were fused with the computed tomography image set to identify and contour a 1.5-cm length
Fig. 1. Spinal nerve with superimposed dose distribution. (A) and (B) Nerve segments cut for longitudinal and transverse histologic sections.
Spinal nerve tolerance to SAbR
847
of each of the C6-C8 spinal nerves beginning approximately 0.4 cm from the edge of the spinal cord (Fig. 1). A 0.1-cm circumferential PTV expansion was applied to each nerve, but the contoured length remained unchanged. Organs at risk including the spinal cord, the C5-C7 vertebral bodies, and the contralateral C6-C8 spinal nerves were contoured to analyze the dose received. Treatment plans consisted of four 6 MV, non-coplanar conformal arcs defined with a multileaf collimator to ensure the full crosssection of a 1.5-cm length of each targeted nerve was covered by at least 95% of the prescription dose and no more than 110% of the prescription dose (see Table 1 for dose metrics and Fig. E1, available online at https://doi.org/ 10.1016/j.ijrobp.2019.03.044, for sample dose distribution from the treatment planning system). Single-session treatment was delivered using either a Vero (Brainlab) or TrueBeam (Varian Medical Systems) image-guided linear accelerator. Cone beam computed tomography was used for target localization, and the 6-degrees-of-freedom treatment couch was rotated and translated until the actual position and required treatment position differed by less than 0.1 cm Table 1 Dose and latency period until initial presentation of motor deficits for individual irradiated pigs Spinal Spinal Spinal Spinal Rx nerve nerve Contralateral cord Cord dose, Dmax, D1.5 cm, spinal nerve Dmax, V10Gy, Latency, Dmax, Gy Gy Gy Gy cm3 wk* Gy 16 16 16 16 16 16 16 18 18 18 18 18 20 20 20 20 20 22 22 22 22 22 24 24 24
17.5 17.4 17.3 17.3 17.2 17.3 17.2 19.6 19.6 19.3 19.4 19.4 21.9 21.6 21.6 21.5 21.6 24.2 24.0 23.7 23.9 23.7 26.2 26.1 26.0
15.6 15.7 16.5 15.9 16.1 15.9 16.0 18.6 17.9 18.2 18.2 17.7 19.9 20.0 20.3 20.4 20.0 21.9 21.9 21.9 22.3 22.4 24.0 23.9 24.0
4.8 3.6 3.4 4.0 3.1 2.6 3.1 4.5 4.3 5.0 4.7 4.3 4.7 6.3 4.9 4.6 4.8 4.0 5.6 5.6 5.1 5.6 5.7 5.5 5.5
10.1 10.6 9.0 4.9 5.4 5.0 8.4 11.8 9.7 11.1 12.0 5.9 8.7 10.9 12.0 11.6 5.0 9.4 12.6 10.2 8.9 9.5 11.1 12.8 9.6
0.046 0.056 0.005 0 0 0 0 0.112 0.022 0.105 0.225 0 0 0.081 0.166 0.132 0 0.011 0.2 0.04 0.007 0.023 0.09 0.391 0.001
NA NA NA NA NA NA NA NA 15.0 NA NA 9.4 10.7 12.0 12.6 13.9 12.3 11.7 11.0 9.3 10.3 11.0 11.9 13.3 10.4
Abbreviations: D1.5cm Z minimum dose to 1.5-cm length; Dmax Z minimum dose to 0.035 cm3 volume; V10Gy Z volume receiving greater than 10 Gy. * NA Z not applicable; no neurologic deficits during stated followup period.
Hrycushko et al.
and 1 degree in each of the 3 primary planes. A prescription dose of 16 to 24 Gy was delivered in approximately 6 to 10 minutes at a rate of 2 to 4 Gy/min to the nerve. Animals were followed for a minimum of 52 weeks or until a change in gait was observed. The general health of animals was observed daily by animal husbandry personnel and approximately weekly by the study principal investigator with attention toward asymmetric gait, unusual restlessness, vocalization, loss of mobility, guarding of a painful area, unkempt appearance, open sores, loss of appetite, and weight loss. Gait was observed with the animals walking freely in a large space. Endpoint response was defined as any study-related change in gait. Electrodiagnostic examination was performed on all pigs at approximately e1, 2, 10, 20, 30, and 50 weeks (or until reaching the study endpoint) to evaluate for the presence of nerve dysfunction. Nerve conduction studies were performed to assess for potential loss of axons in motor and sensory nerves indicative of peripheral nerve disorder. Needle electromyography (EMG) was performed to assess for increased insertional activity and abnormal spontaneous activity. Animals recognized to have a change in gait were evaluated by a veterinarian for symptoms indicative of pain and were humanely killed. The bilateral C6-C8 spinal nerves and the spinal cord were harvested for histologic processing. All tissues were fixed in formalin before being sectioned and processed for embedding in paraffin. Spinal nerve segments were cut in longitudinal and transverse sections (inset of Fig. 1). Sections of spinal nerve and spinal cord were stained with a Luxol fast blue/ periodic acid-Schiff combination for light microscopy. The dose-related incidence of gait change was used as a clinically relevant endpoint to obtain quantal data for probit analysis using IBM SPSS Statistics v24 software (IBM, Armonk, NY). The analysis of dose response was performed using the maximum dose (0.035 cm3 volume) to the targeted spinal nerve, which was calculated as the mean of the maximum doses to nerves C6-C8 for each pig. The full dose-response curve and 95% confidence bands were generated. The dose associated with a 50% incidence of motor deficit (ED50) was determined and compared with the ED50 for spinal cord myelopathy from Medin et al in the same animal model.13
Results Dosimetric and response data for individual pigs are presented in Table 1 and include (1) prescription dose, (2) maximum spinal nerve dose, (3) minimum dose to the 1.5cm length of targeted nerve, (4) maximum dose to the contralateral spinal nerve, (5) maximum dose to the spinal cord, (6) volume receiving at least 10 Gy, and (7) latency to response. Fifteen of 25 pigs developed front limb motor changes on the irradiated (left) side. No deficits were observed on the unirradiated (right) side of any pig. Motor deficits presented as a mild limp with occasional dragging
International Journal of Radiation Oncology Biology Physics
of the foot and progressed to limb weakness. None of the pigs showed behavior indicative of pain. The latency period until initial presentation of motor deficits ranged from 9 to 15 weeks. The dose-response curve and 95% confidence intervals for the spinal nerve are presented in Figure 2 (P Z .74) with superimposed dose-response points from a previous spinal cord tolerance study.13 A test for parallelism shows no significant difference between the spinal nerve and spinal cord data (P Z .584). No gait changes were observed when the maximum point dose to the spinal nerves was <19.4 Gy, and all pigs had gait changes when the maximum point dose was 21.5 Gy. EMG results are summarized in Table 2. Evidence of denervation in the C6 and C7 innervated forelimb muscles (deltoid, triceps, and forelimb extensor muscles) was observed in 76% of the animals, but this was not predictive of gait changes. Denervation in the form of increased insertional activity or abnormal spontaneous activity was observed in 50% of asymptomatic pigs but more commonly in symptomatic pigs (93%). The median latency for the appearance of EMG changes was 21.9 weeks postirradiation. In some instances, denervation changes were not observed immediately after motor deficit but rather several weeks later, which in practice is not unexpected as a neuropathic lesion evolves. Nerve conduction studies assessing bilateral median and ulnar motor nerves and bilateral ulnar sensory nerves showed no significant change in compound motor action potential amplitude or sensory nerve action potential amplitude between baseline and postirradiation measurements. As a whole, the pattern of electrodiagnostic findings in affected pigs was most consistent with polyradiculopathy with motor axonal loss. Given the limitations of testing in this model, electrophysiologic evidence of demyelination or conduction block could not be accurately demonstrated. Histolopathologic findings were in agreement with the gait-change data. Left-sided unilateral degeneration of (5/5)
1
(5/5)
(3/3)
0.9 0.8 0.7 Probability
848
0.6 0.5 (2/5)
0.4 0.3 0.2 0.1 0 12
(0/7)
14
16
18 20 24 22 Maximum Dose (Gy)
26
28
Fig. 2. Dose-response curve (solid line) for motor neurologic deficit with 95% confidence intervals (dashed line). (O) are spinal nerve data points, and (x) are spinal cord data points from Medin et al.13
Volume 104 Number 4 2019 Table 2
Spinal nerve tolerance to SAbR
849
Summary of electrodiagnostic results
Rx Dose, Gy
Symptomatic (Motor Deficit), n
16 (n Z 7) 18 (n Z 5) 20 (n Z 5) 22 (n Z 5) 24 (n Z 3) Entire cohort (n Z 25)
0 2 5 5 3 15
Abnormal EMG (signs of denervation, n 4 (all asymptomatic) 3 (1 asymptomatic) 4 5 3 Total: 19 Asymptomatic: 5 Symptomatic: 14
Appearance of EMG changes, median (range), wk 37.5 26.6 15.5 12.8 17 21.9
(30-50) (10-50) (2-30) (2-30) (2-30) (2-50)
Abbreviation: EMG Z electromyography.
spinal nerves including loss of myelin, axonal swelling, and various degrees of fibrosis was observed in all pigs with motor deficits. These changes do not exclude microvascular injury occurring at an early stage of pathogenesis, but at the time the tissues were collected specific vascular injury was not a dominating feature. The replacement of myelinated fibers by fibrotic tissue was clearly visible in the irradiated (left) nerves, whereas the unirradiated (right) side was normal (Fig. 3). No histologic damage was observed in the white or gray matter of the spinal cord tissue of any animal in this study. Some spinal cord sections showed 1-sided scattered fibrosis of nerve roots. In those cases, the leftsided longitudinal spinal nerve sections also showed fibrosis and scattered demyelination in the ganglia, indicating retrograde degeneration.
Discussion Only limited information is available to characterize the dose response of the peripheral nervous system to SAbR, and there is a general perception that peripheral nerves are more tolerant than the spinal cord to irradiation, as evidenced by current tissue tolerance recommendations and clinical trial guidelines.24-26 This study was performed to
characterize the dose response of a specific set of peripheral nerves, the spinal nerves, to single-session external beam irradiation in an animal model of similar anatomic size to humans. The Yucatan minipig was a suitable animal model for this study because its anatomic dimensions allowed for a 1.5-cm length of spinal nerve to be visualized and targeted while maintaining a relatively low dose to the spinal cord. The Yucatan minipig has also previously been used to study the tolerance of the spinal cord to single-session irradiation.13 Because of a sparsity of data, the relationship between the tolerance of the nervous systems in pigs and humans is unknown, and caution is warranted when extrapolating from animal to human populations. The dose-response of the spinal cord has marked similarities across a variety of animal models.9-11,13 Data on peripheral nerve injury have a much broader range of observed tolerances, possibly a result of differences in animal models, irradiation technique, or anatomic site.17,19,27-34 Data from IORT studies targeting different peripheral nerves in dogs have resulted in a range of tolerance values. Kinsella et al surgically exposed and irradiated the lumbosacral plexus and sciatic nerves in an American Foxhound model to singlefraction doses of 20 to 75 Gy, and 19 of the 21 irradiated dogs showed clinically evident nerve injury.17 A follow-up study designed to determine the threshold dose for
Fig. 3. Spinal nerve (C7) 96 days after 22 Gy. (A) Left side irradiated spinal nerve showing degenerating nerve fibers (loss of blue-stained myelin) and replacement with fibrosis (red stain), eg, arrow. (B) Nontargeted right-side spinal nerve. Luxol fast blue/periodic acid-Schiff stain, 10 objective.
850
Hrycushko et al.
peripheral nerve injury from IORT showed 100% paresis in animals receiving 20 Gy, whereas no dog receiving 10 or 15 Gy showed signs of injury.18 In a tolerance study of lumbar spinal nerves and proximal femoral nerves of beagles, LeCouteur et al reported an ED50 of 19.5 Gy.19 In similar studies performed at the same institution, the tolerance of the sciatic nerve and mid-femoral nerves was investigated. One of these studies found that none of the 5 dogs receiving 20 Gy experienced neurologic deficit, and only 2 of 5 dogs experienced deficits after 28 Gy.28 In a second study, the ED50 for hind limb paresis was estimated to be 22 Gy.33 The tolerance of the lumbosacral plexus and sciatic nerve to IORT was investigated at the National Cancer Institute, and an ED50 of 17.2 Gy was reported.29 Lin et al delivered a maximal dose of 25 Gy using external beam irradiation to the sciatic nerves of 12 rabbits, and none developed leg weakness within the 7-month follow-up period.34 In contrast, we observed that all of 13 pigs receiving a maximal dose of 21.6 Gy experienced motor neurologic deficits within 13.9 weeks. The tolerance disagreement between the pig and rabbit studies may be explained by the existence of a length effect, although response differences from species variation or radiosensitivity variation between anatomic sites (spinal nerve vs. sciatic nerve) is also plausible. For the pig spinal nerves, a 1.5-cm length received the prescription dose and the hot spot within the targeted nerve was approximately 8% greater than the prescription dose, whereas the rabbit sciatic nerves were irradiated using a 0.77-cm diameter stereotactic cone and the maximum dose was reported. Although not specifically stated, irradiation with a circular collimator typically results in a dose of approximately 50% of the maximum extending the length of the diameter. The existence of a dose-length effect has not been studied in the peripheral nervous system but is well established in the spinal cord. Two independent laboratories have observed that the dose tolerance of the rat spinal cord increases dramatically as the irradiated length is decreased below a threshold of approximately 1 cm for single-session irradiation.9,35-38 Stubblefield et al recently reported 14 peripheral nerve injuries from a series of 447 patients (557 treatments) who were treated with single-session radiosurgery for spinal tumors.16 For these treatments, nerve roots were included in the PTV and received up to 26 Gy. The lengths of nerve irradiated in this series of patients are not reported, but a dose-length effect may explain why the rate of neuropathy was very low despite the large doses delivered. This study was performed to determine the doseresponse curve for symptomatic neuropathy after singlesession irradiation of the C6-C8 spinal nerves. We observed that the tolerance to full cross-sectional irradiation of the spinal nerves is not significantly different from that of the spinal cord.13 It is important to recognize that doseresponse data previously reported for the spinal cord using this same pig model included spinal nerves in the irradiated volume; thus, the observed motor deficits may
International Journal of Radiation Oncology Biology Physics
have been influenced by the response of the spinal nerves, even though neurologic deficits and histologic change (demyelination, necrosis) of the spinal cord were highly correlated in the spinal cord studies. Exact dose-length data for the spinal nerves included in our spinal cord tolerance studies are not available, nor were tissue samples of the spinal nerves collected. A similar dose tolerance in the spinal nerves and spinal cord may indicate a common pathogenesis, most likely beginning with microvascular injury and progressing to late vasculopathies. It is also important to recognize that almost all historic spinal cord tolerance studies using animals included nerve rootlets or spinal nerves along with the spinal cord in their irradiated tissues. To our knowledge, only 1 spinal cord tolerance study has specifically excluded nerve rootlets and spinal nerves. Bijl et al irradiated the midline of the spinal cord with a narrow proton beam having a penumbra of 0.8 mm and observed that the central white matter and grey matter of the spinal cord are much more radioresistant than the lateral white matter.39 As part of our long-term goal to characterize the dose-response of the peripheral nervous system to ablative radiation therapy, future studies aim to investigate the existence of a dose-length effect for the spinal nerves and to determine if radiation tolerance changes as the peripheral nerves travel from the vertebral column to the level of the divisions of the brachial plexus.
Conclusions The ED50 for symptomatic neuropathy following full crosssectional irradiation of the spinal nerves was found to be 19.7 Gy. The dose-response of the C6-C8 spinal nerves is not significantly different from that of full cross-sectional irradiation of the spinal cord as observed in companion studies.
References 1. Folkert MR, Timmerman RD. Stereotactic ablative body radiosurgery (SABR) or stereotactic body radiation therapy (SBRT). Adv Drug Deliv Rev 2017;109:3-14. 2. Yamada Y, Bilsky MH, Lovelock DM, et al. High-dose, single-fraction image-guided intensity-modulated radiotherapy for metastatic spinal lesions. Int J Radiat Oncol Biol Phys 2008;71:484-490. 3. Garg AK, Shiu AS, Yang J, et al. Phase 1/2 trial of single-session stereotactic body radiotherapy for previously unirradiated spinal metastases. Cancer 2012;118:5069-5077. 4. Sellin JN, Suki D, Harsh V, et al. Factors affecting survival in 43 consecutive patients after surgery for spinal metastases from thyroid carcinoma. J Neurosurg Spine 2015;23:419-428. 5. Bate BG, Khan NR, Kimball BY, Gabrick K, Weaver J. Stereotactic radiosurgery for spinal metastases with or without separation surgery. J Neurosurg Spine 2015;22:409-415. 6. Chang EL, Shiu AS, Mendel E, et al. Phase I/II study of stereotactic body radiotherapy for spinal metastasis and its pattern of failure. J Neurosurg Spine 2007;7:151-160. 7. Gerszten PC, Burton SA, Welch WC, et al. Single-fraction radiosurgery for the treatment of spinal breast metastases. Cancer 2005; 104:2244-2254.
Volume 104 Number 4 2019 8. Staehler M, Haseke N, Nuhn P, et al. Simultaneous anti-angiogenic therapy and single-fraction radiosurgery in clinically relevant metastases from renal cell carcinoma. BJU Int 2011;108:673-678. 9. Hopewell JW, Morris AD, Dixon-Brown A. The influence of field size on the late tolerance of the rat spinal cord to single doses of X rays. Br J Radiol 1987;60:1099-1108. 10. Lo YC, McBride WH, Withers HR. The effect of single doses of radiation on mouse spinal cord. Int J Radiat Oncol Biol Phys 1992;22: 57-63. 11. Knowles JF. The effects of single dose X-irradiation on the guinea-pig spinal cord. Int J Radiat Biol Relat Stud Phys Chem Med 1981;40:265275. 12. Powers BE, Thames HD, Gillette SM, Smith C, Beck ER, Gillette EL. Volume effects in the irradiated canine spinal cord: Do they exist when the probability of injury is low? Radiother Oncol 1998;46:297-306. 13. Medin PM, Foster RD, van der Kogel AJ, Sayre JW, McBride WH, Solberg TD. Spinal cord tolerance to single-session uniform irradiation in pigs: Implications for a dose-volume effect. Radiother Oncol 2013;106:101-105. 14. Ryu S, Jin JY, Jin R, et al. Partial volume tolerance of the spinal cord and complications of single-dose radiosurgery. Cancer 2007;109:628636. 15. Gibbs IC, Patil C, Gerszten PC, Adler JR Jr., Burton SA. Delayed radiation-induced myelopathy after spinal radiosurgery. Neurosurgery 2009;64:A67-A72. 16. Stubblefield MD, Ibanez K, Riedel ER, et al. Peripheral nervous system injury after high-dose single-fraction image-guided stereotactic radiosurgery for spine tumors. Neurosurg Focus 2017;42:E12. 17. Kinsella TJ, Sindelar WF, DeLuca AM, et al. Tolerance of peripheral nerve to intraoperative radiotherapy (IORT): Clinical and experimental studies. Int J Radiat Oncol Biol Phys 1985;11:1579-1585. 18. Kinsella TJ, DeLuca AM, Barnes M, et al. Threshold dose for peripheral neuropathy following intraoperative radiotherapy (IORT) in a large animal model. Int J Radiat Oncol Biol Phys 1991;20:697-701. 19. LeCouteur RA, Gillette EL, Powers BE, Child G, McChesney SL, Ingram JT. Peripheral neuropathies following experimental intraoperative radiation therapy (IORT). Int J Radiat Oncol Biol Phys 1989;17:583-590. 20. Sindelar WF, Kinsella TJ, Chen PW, et al. Intraoperative radiotherapy in retroperitoneal sarcomas. Final results of a prospective, randomized, clinical trial. Arch Surg 1993;128:402-410. 21. Azinovic I, Calvo FA, Puebla F, Aristu J, Martı´nez-Monge R. Longterm normal tissue effects of intraoperative electron radiation therapy (IOERT): Late sequelae, tumor recurrence, and second malignancies. Int J Radiat Oncol Biol Phys 2001;49:597-604. 22. Petersen IA, Haddock MG, Donohue JH, et al. Use of intraoperative electron beam radiotherapy in the management of retroperitoneal soft tissue sarcomas. Int J Radiat Oncol Biol Phys 2002;52:469-475. 23. Medin PM, Foster RD, van der Kogel AJ, et al. Paralysis following stereotactic spinal irradiation in pigs suggests a tolerance constraint for single-session irradiation of the spinal nerve. Radiother Oncol 2013;109:107-111.
Spinal nerve tolerance to SAbR
851
24. Ryu S, Pugh SL, Gerszten PC, et al. RTOG 0631 phase 2/3 study of image guided stereotactic radiosurgery for localized (1-3) spine metastases: Phase 2 results. Pract Radiat Oncol 2014;4:76-81. 25. Benedict SH, Yenice KM, Followill D, et al. Stereotactic body radiation therapy: The report of AAPM Task Group 101. Med Phys 2010; 37:4078-4101. 26. Emami B. Tolerance of normal tissue to therapeutic radiation. Radiother Oncol 2013;1:35-48. 27. Hassler O. Cellular kinetics of the peripheral nerve and striated muscle after a single dose of x-rays. A histological and autoradiographical study, using 3H-thymidine. Z Zellforsch Mikrosk Anat 1968; 85:62-66. 28. Vujaskovic Z, Gillette SM, Powers BE, et al. Intraoperative radiation (IORT) injury to sciatic nerve in a large animal model. Radiother Oncol 1994;30:133-139. 29. Johnstone PA, DeLuca AM, Bacher JD, et al. Clinical toxicity of peripheral nerve to intraoperative radiotherapy in a canine model. Int J Radiat Oncol Biol Phys 1995;32:1031-1034. 30. Gillette EL, Mahler PA, Powers BE, Gillette SM, Vujaskovic Z. Late radiation injury to muscle and peripheral nerves. Int J Radiat Oncol Biol Phys 1995;31:1309-1318. 31. Powell S, Cooke J, Parsons C. Radiation-induced brachial plexus injury: Follow-up of two different fractionation schedules. Radiother Oncol 1990;18:213-220. 32. Sindelar WF, Hoekstra H, Restrepo C, Kinsella TJ. Pathological tissue changes following intraoperative radiotherapy. Am J Clin Oncol 1986; 9:504-509. 33. Vujaskovic Z, Gillette SM, Powers BE, et al. Effects of intraoperative irradiation and intraoperative hyperthermia on canine sciatic nerve: Neurologic and electrophysiologic study. Int J Radiat Oncol Biol Phys 1996;34:125-131. 34. Lin Z, Wu VW, Ju W, Yamada Y, Chen L. Radiation-induced changes in peripheral nerve by stereotactic radiosurgery: A study on the sciatic nerve of rabbit. J Neurooncol 2011;102:179-185. 35. Bijl HP, van Luijk P, Coppes RP, Schippers JM, Konings AW, van der Kogel AJ. Dose-volume effects in the rat cervical spinal cord after proton irradiation. Int J Radiat Oncol Biol Phys 2002;52: 205-211. 36. Bijl HP, van Luijk P, Coppes RP, Schippers JM, Konings AW, van der Kogel AJ. Unexpected changes of rat cervical spinal cord tolerance caused by inhomogeneous dose distributions. Int J Radiat Oncol Biol Phys 2003;57:274-281. 37. Philippens ME, Pop LA, Visser AG, Schellekens SA, van der Kogel AJ. Dose-volume effects in rat thoracolumbar spinal cord: An evaluation of NTCP models. Int J Radiat Oncol Biol Phys 2004;60: 578-590. 38. van der Kogel AJ. Dose-volume effects in the spinal cord. Radiother Oncol 1993;29:105-109. 39. Bijl HP, van Luijk P, Coppes RP, Schippers JM, Konings AW, van Der Kogel AJ. Regional differences in radiosensitivity across the rat cervical spinal cord. Int J Radiat Oncol Biol Phys 2005;61: 543-551.