J Neurosurg Spine 21:711–718, 2014 ©AANS, 2014
Spine stereotactic body radiotherapy for renal cell cancer spinal metastases: analysis of outcomes and risk of vertebral compression fracture Clinical article Isabelle Thibault, M.D.,1 Ameen Al-Omair, M.D.,1,2 Giuseppina Laura Masucci, M.D., 2 Laurence Masson-Côté, M.D., 2 Fiona Lochray, M.R.T.T.,1 Renée Korol, Ph.D.,1 Lu Cheng, M.Math., 3 Wei Xu, Ph.D., 3 Albert Yee, M.D., 4 Michael G. Fehlings, M.D., Ph.D., 5 Georg A. Bjarnason, M.D., 6 and Arjun Sahgal, M.D.1,2 Department of Radiation Oncology, Sunnybrook Odette Cancer Centre; Departments of 2Radiation Oncology and 3Biostatistics, Princess Margaret Cancer Centre; 4Department of Orthopaedic Surgery, Sunnybrook Health Sciences Centre; 5Division of Neurosurgery and Spinal Program, Toronto Western Hospital; and 6Department of Medical Oncology, Sunnybrook Odette Cancer Centre, University of Toronto, Ontario, Canada 1
Object. The aim of this study was to evaluate local control (LC) and the risk of vertebral compression fracture (VCF) after stereotactic body radiotherapy (SBRT) in patients with renal cell cancer spinal metastases. Methods. Prospectively collected data on 71 spinal segments treated with SBRT in 37 patients were reviewed. The median follow-up was 12.3 months (range 1.2–55.4 months). The LC rate was assessed based on each spinal segment treated and overall survival (OS) according to each patient treated. Sixty of 71 segments (85%) were radiation naive, 11 of 71 (15%) were previously irradiated, and 10 of 71 (14%) were treated with postoperative SBRT. The median SBRT total dose and number of fractions were 24 Gy and 2, respectively. The VCF analysis also included evaluation of the Spinal Instability Neoplastic Score criteria. Results. The 1-year OS and LC rates were 64% and 83%, respectively. Multivariate analysis identified oligometastatic disease (13 of 37 patients) as a positive prognostic factor (p = 0.018) for OS. Of 61 non-postoperative spinal segments treated, 10 (16%) developed VCFs; 3 of 10 were de novo VCFs and 7 of 10 occurred as progression of an existing VCF. The 1-year VCF-free probability rate was 82%. Multivariate analysis identified single-fraction SBRT and baseline VCF as significant predictors of SBRT-induced VCF (p = 0.028 and p = 0.012, respectively). Conclusions. Spine SBRT yields high rates of local tumor control in patients with renal cell cancer. Baseline VCF and 18–24 Gy delivered in a single fraction were predictive of further collapse. Patients with oligometastatic disease may benefit most from such aggressive local therapy, given the prolonged survival observed. (http://thejns.org/doi/abs/10.3171/2014.7.SPINE13895)
R
Key Words • renal cell cancer • spine metastases • oncology • stereotactic body radiotherapy • stereotactic radiosurgery
cell cancer (RCC) metastases have been traditionally considered to be radioresistant, with poor response rates to conventional palliative radiotherapy. However, high dose per fraction radiation delivered using stereotactic body radiotherapy (SBRT) has the potential to maximize local control (LC) safely and to enal
Abbreviations used in this paper: ASIA = American Spinal Injury Association; CBCT = cone-beam CT; CTV = clinical target volume; HR = hazard ratio; LC = local control; OS = overall survival; PRV = planning organ at risk volume; PTV = planning target volume; RCC = renal cell cancer; SBRT = stereotactic body radiotherapy; SINS = Spinal Instability Neoplastic Score; VCF = vertebral compression fracture.
J Neurosurg: Spine / Volume 21 / November 2014
treat patients with locally “curative” intent rather than locally “palliative” intent.20 The SBRT technique has been recently defined by the Canadian Association of Radiation Oncologists as “The precise delivery of highly conformal and image-guided hypofractionated external-beam radiotherapy, delivered in a single or few fraction(s), to an extra-cranial body target with doses at least biologically equivalent to a radical course when given over a protracted conventionally fractionated (1.8–3.0 Gy/fraction) schedule.”20 This article contains some figures that are displayed in color online but in black-and-white in the print edition.
711
I. Thibault et al. Spinal metastases from RCC represent an important site for the application of SBRT, to optimize LC and prevent malignant epidural spinal cord compression. Although surgery can be performed up front for specific indications such as malignant epidural spinal cord compression or as salvage on progression after radiation, surgery can be associated with major risks including intraoperative hemorrhage. Surgery also has associated delays in its initiation with respect to performing preoperative embolization, and results in delays with respect to starting antiangiogenic targeted therapies for RCC postoperatively given the risk of wound complications. The latter may compromise patient outcomes with respect to survival. It is important to note that although targeted therapy for RCC has significant activity, bone metastases often do not respond, and their presence is an independent predictor of worse outcome.4 Hence, aggressive local management to spinal metastases is justified even in the era of targeted systemic therapies for patients with metastatic RCC. The field of spine SBRT is rapidly emerging into mainstream practice; however, it is still considered investigational and not without risk—serious toxicities such as radiation-induced myelopathy and vertebral compression fracture (VCF) have been reported.16,18,19,21,22 Our spine SBRT technique has been evaluated for its precision, safe spinal cord dose tolerances for SBRT have been established, and the risk of VCF for a mixed cohort of patients has been reported.9,12,19,21 The aim of this investigation was to report our outcomes and toxicity in patients with RCC who had spinal metastases, with a focus on LC and the incidence of VCF. In addition, we applied the recently developed Spinal Instability Neoplastic Score (SINS) criteria10 to determine its predictive value for SBRT-induced VCF.
Methods
We identified 71 spinal segments in 37 patients with RCC who had been treated with spine SBRT between October 2007 and August 2012; data were obtained from our prospective spine SBRT database. This retrospective review of clinical and dosimetric data was approved by our ethics review board. Our spine SBRT technique has been previously described in detail. In brief, patients were immobilized in the BodyFIX system (Elekta AB) for spinal segments below the T-3 spinal level. For cervical and upper thoracic spinal segments (up to the T-2 spinal level), patients were immobilized in a thermoplastic head and shoulder mask. The treatment planning CT scan was performed with a slice thickness of 1 mm, and 1.25to 1.5-mm axial T1 and T2 volumetric MRI sequences, comprising at least 1 spinal level above and below the target volume, were fused to the CT. For patients with contraindications to MRI, or in postoperative patients with sufficient MR-induced artifacts that compromised image quality, a treatment planning CT myelogram was performed. The clinical target volume (CTV) was based on an anatomical concept consistent with recently published guidelines.8 We applied a 2-mm planning target volume (PTV), and for the spinal cord a 1.5-mm planning organ 712
at risk volume (PRV) was applied. For the cauda equina, the thecal sac alone was contoured. We applied published dose limits to the critical neural structure including the cord PRV and/or thecal sac.19,21 Treatment was delivered using the Elekta Axesse (Elekta AB, Stockholm, Sweden) equipped with a 4-mm multileaf collimator, Hexapod robotic couch, and cone-beam CT (CBCT) image guidance system. Our delivery system and CBCT approach has been evaluated and its precision reported on. Our image guidance protocol has also been previously reported.12 In brief, the initial patient positioning is verified on the basis of a second CBCT postshift (verification CBCT preSBRT) with a tolerance for repositioning of 1 mm and 1°. A midfraction and a posttreatment CBCT were also performed to determine intrafraction motion. For singlefraction SBRT, we perform 2 intratreatment CBCT scans due to the long treatment times (typically 60–70 minutes). Local control was based on each spinal segment treated (a total of 71), and overall survival (OS) was calculated according to each patient treated (a total of 37). An oligometastatic state was defined as fewer than 5 sites of metastatic disease at the time of spine SBRT. Each patient was followed according to our institutional policy of a complete MRI sequence of the spine at 2- to 3-month intervals. All patients were treated and followed meticulously by a single radiation oncologist (A.S.). Local progression was determined based on the radiologist’s interpretation of the MR images. In cases in which it was unclear if MR signal changes within the bone were secondary to radiation or disease progression, 2 successive MRI evaluations showing enlarged dimensions of the T1 signal abnormality were required for progression. Increased epidural disease or enlargement of paraspinal disease was documented as progression. All MR images were retrospectively reviewed to evaluate the exact pattern of local recurrence and to classify the epidural tumor extent according to the grading scale proposed by Bilsky et al.5 Both the initial treatment planning CT and MRI were reviewed to score patients according to SINS.10 In brief, SINS breaks down radiological and clinical characteristics according to 6 criteria: 1) location; 2) pain; 3) type of bone lesion (lytic, mixed, or blastic); 4) spinal alignment; 5) presence of a VCF at baseline; and 6) involvement of the posterior spinal elements (spinous process, lamina, pedicle).10 A final score classifies the patient as stable, potentially unstable, and unstable. Our interest was in the predictive value of each of the criteria. Data for 61 sites that did not receive operative intervention were used to assess the predictive values of the SINS criteria for outcome. Statistical Analysis
Summary statistics are provided, with frequency and proportion for categorical factors; median and range are given for continuous risk factors. Group-specific LC, OS, and the fracture-free probabilities and confidence intervals were calculated from the start of SBRT, using the Kaplan-Meier method. Univariate analyses were performed using the Cox proportional hazard regression model, and hazard ratios (HRs) and corresponding confidence limits were calculated. Two-sided tests were used, J Neurosurg: Spine / Volume 21 / November 2014
Stereotactic body radiotherapy for RCC spinal metastases with the significance level set at 0.05. Multivariate models were conducted using the stepwise selection procedure. Statistical analyses were performed using version 9.3 of the SAS system and also the R software version 2.14.0.
Results
Baseline Patient, Tumor, and Treatment Characteristics
Among the 37 patients with RCC, 34 (92%) had a prior nephrectomy, and the median interval between the initial diagnosis of RCC and spine SBRT treatment was 26.6 months (range 0.4–189 months). The median age was 63 years (range 33–82 years). Twenty-five patients (68%) were male, 12 (32%) were female, and 13 patients (35%) had oligometastatic disease. The Eastern Cooperative Oncology Group performance status was 0 in 1 (3%), 1 in 25 (67%), and 2 in 11 (30%) patients. Neurological status according to the American Spinal Injury Association (ASIA) scale was normal (ASIA E), except for 1 patient with partial motor loss of function below the neurological level (ASIA D), and another patient had sensory dysfunction secondary to bilateral neural foramina invasion at the level of L-1 but no motor compromise. Other baseline tumor and treatment characteristics for the 71 spinal metastatic segments are presented in Table 1. Eleven of 71 segments (15%) were previously irradiated with 1 prior course of palliative radiotherapy (8 Gy in 1 fraction, n = 1; 20 Gy in 5 fractions, n = 3; and 30 Gy in 10 fractions, n = 7). The median interval between the first radiotherapy course and SBRT was 16 months (range 5.5–56.4 months). Of the 7 of 71 segments (10%) with epidural disease compressing the spinal cord (Bilsky Grade 2), 4 underwent surgery, but 3 of the 7 were not surgical candidates due to medical comorbidities or disease that was too extensive. These 3 segments were treated with SBRT alone. Of the 17 of 71 segments (24%) with a baseline VCF, 3 segments with > 33% loss of vertebral body height were surgically stabilized before SBRT; 1 of the 14 segments that had ≤ 33% loss of height also had a stabilization procedure pre-SBRT, but the majority of segments (13 of 14) that had ≤ 33% collapse were treated with SBRT alone. For the 10 postoperative SBRT cases, surgery consisted of stabilization alone in 2 of 10 segments, decompression without stabilization in 4 of 10 segments, and decompression with instrumented stabilization in 4 of 10. The SINS baseline characteristics of the 61 spinal segments treated with SBRT alone are summarized in Table 2. The median total dose and number of fractions were 24 Gy (range 18–30 Gy) and 2 (range 1–5), respectively. The median volumes of the CTVs receiving 80% and 90% of the prescribed dose were 91.8% (range 64.2%–99.8%) and 87% (range 56.2%–99.5%), respectively. The median volumes of the PTVs receiving 80% and 90% of the prescribed dose were 88.8% (range 57%–99.7%) and 84.1% (range 47.8%–98.8%), respectively. The median spinal cord PRV and thecal sac maximal point doses, normalized in biologically equivalent doses in 2-Gy fractions (nBED or also called EQD2 when using an a/b ratio of 2), were 28.41 Gy2 (range 0.02–59.58 Gy2; n = 53 spinal cord segments) and 37.76 Gy2 (range 1.65–201.22 Gy2; n = 71 J Neurosurg: Spine / Volume 21 / November 2014
thecal sac segments), respectively. The median follow-up after SBRT was 12.3 months (range 1.2–55.4 months). Overall Survival and Local Control
The 1- and 2-year rates and median OS duration were 64.1%, 45.6%, and 1.5 years, respectively. Multivariate analysis identified oligometastatic disease (13 of 37 patients [35%]) versus widespread metastatic disease as the only significant prognostic factor for OS (HR 0.25, 95% CI 0.08–0.79, p = 0.018; Fig. 1). Local progression was observed in 12 of 71 spinal segments. The 1- and 2-year LC rates were 83.4% and 66.2%, respectively (Fig. 2). Among the 12 segments with recurrence, the median time to local recurrence was 11.5 months (range 5–41 months). Patterns of local failure included progression within the vertebral bone alone (2 of 12 [17%]), paraspinal tissue (2 of 12 [17%]), and epidural progression (8 of 12 [66%]; in isolation, 5 of 8 [63%]). Management consisted of decompression and stabilization surgery alone (1 of 12, 8%), decompressive surgery followed by repeat SBRT (1 of 12, 8%), repeat SBRT and no surgery (3 of 12, 25%), systemic therapy alone (4 of 12, 33%), or palliative care (3 of 12, 25%).
Adverse Events
In total, 10 VCFs were observed post-SBRT in the 61 (16%) nonsurgical spinal segments; 3 of 10 were de novo VCFs and 7 of 10 were progression of an existing VCF. The 1-year rate for freedom from any VCF was 82%, and the median time to SBRT-induced VCF among these 10 cases was 47.5 days (range 1–237 days). Ultimately, 6 of 10 remained asymptomatic and were observed, and 4 of 10 required a stabilization surgery. Table 2 presents a summary of those with and without a VCF according to each of the SINS criteria at baseline. Based on SINS, 3 were unstable, 21 were potentially unstable, and 37 were stable. Analysis of baseline patient, tumor, and treatment factors (as summarized in Tables 1 and 2) identified on the multivariate model determined that single-fraction SBRT (HR 5.03, 95% CI 1.19–21.28, p = 0.028) and presence of a baseline VCF (HR 9.25, 95% CI 1.64–52.31, p = 0.012) were the 2 significant predictors of subsequent VCF post-SBRT. At 1 year post-SBRT, the risk of developing a VCF was 9% for multiple fractions (20–30 Gy in 2–5 fractions) versus 25% for single-fraction SBRT (18–24 Gy in 1 fraction), and 7% versus 59% for intact baseline segments compared with segments with a preexisting VCF (Fig. 3). Figure 4 shows an example of fracture progression. With respect to non-VCF events, no acute or late Grade 3–5 toxicities were observed.
Residual Setup Error and Intrafraction Motion
We observed intrafraction motion of 1.25 mm and 0.56° for single-fraction SBRT cases (25 segments) and 0.91 mm and 0.66° for multiple-fraction SBRT cases (34 segments), with 95% confidence. When comparing single- versus multiple-fraction SBRT residual errors, there were no significant differences. In addition, there was no significant difference between the verification, intrascan, and postscan residual setup errors. 713
I. Thibault et al. TABLE 1: Spinal tumor and treatment characteristics in 37 patients with spinal metastases from RCC* Characteristic no. of segments histological finding clear cell RCC papillary RCC spinal level cervical thoracic lumbar sacral paraspinal extension yes no epidural disease grade† 0 1a 1b 1c 2 prior EBRT yes no systemic therapy prior targeted therapy + chemo‡ prior targeted therapy, no chemo‡ bisphosphonate therapy CTV single vertebral segment multiple adj segments w/in 1 CTV total dose/no. of Fxs 18–24 Gy/1 Fx 20–24 Gy/2 Fxs 24–27 Gy/3 Fxs 30 Gy/4–5 Fxs
Overall No. (%)
SBRT Alone
Postop SBRT
71
61
10
69 (97) 2 (3)
59 2
10 0
9 (13) 33 (46) 21 (30) 8 (11)
5 32 16 8
4 1 5 0
30 (42) 41 (58)
22 39
8 2
36 (51) 6 (8) 19 (27) 3 (4) 7 (10)
35 5 15 3 3
1 1 4 0 4
11 (15) 60 (85)
10 51
1 9
30 (42) 16 (23) 13 (18)
16 29 12
0 1 1
39 (55) 32 (45)
36 25
3 7
30 (42) 26 (37) 11 (15) 4 (6)
29 21 8 3
1 5 3 1
* adj = adjacent; chemo = chemotherapy; EBRT = external-beam radiotherapy; Fxs = fractions. † Bilsky grades: 0 = tumor confined within the bone; 1a = epidural impingement, but tumor doesn’t compress the thecal sac; 1b = tumor compresses the thecal sac, no spinal cord abutment; 1c = tumor compresses dura and abuts the cord without displacing it; and 2 = spinal cord compression, but CSF is visible around the cord. ‡ Targeted therapies used for RCC include agents targeting the vascular endothelial growth factor (VEGF) and the mammalian target of rapamycin, such as sunitinib and everolimus, respectively.
Discussion
In this paper we report our favorable SBRT experience for spinal metastases specific to RCC. Metastases from RCC have long been considered radioresistant, and strategies to escalate doses in spinal lesions have been limited by both delivery and treatment planning technology. However, our data demonstrate that SBRT is an advanced radiotherapeutic technique that has matured beyond lung and liver indications to the treatment of both primary and metastatic spinal tumors.20 714
Our results conclude a 1-year actuarial LC rate of 83.4%, and this is consistent with the limited spine-specific SBRT RCC literature as summarized in Table 3.3,11,15 With respect to predictors of LC, none of the baseline clinical and dosimetric predictors investigated (Table 1) were predictive on univariate analysis. This may be due to the limited sample sizes of individual series and the low event rate. Although minimum target dose and degree of epidural disease have been reported as predictors in various series, there has been no consistent predictive factor to guide treatment planning.1,7,13 J Neurosurg: Spine / Volume 21 / November 2014
Stereotactic body radiotherapy for RCC spinal metastases TABLE 2: Baseline SINS classification according to VCF status for 61 spinal segments with no prior surgery SINS Factor location junctional (Oc–C2, C7–T2, T11–L1, L5–S1) mobile spine (C3–6, L2–4) semirigid (T3–10) rigid (S2–5) pain mechanical occasional & nonmechanical pain free bone lesion lytic mixed blastic radiographic spinal alignment subluxation or translation kyphosis or scoliosis normal vertebral body collapse ≥50% 50% body involved none of the above posterolat involvement of spinal elements bilat unilat none of the above SINS classification* unstable indeterminate stability (potentially unstable) stable
VCF (n = 10)
No VCF (n = 51)
Overall Fracture (%)
6 3 1 0
18 9 19 5
6/24 (25) 3/12 (25) 1/20 (5) 0/5 (0)
4 3 3
4 29 18
4/8 (50) 3/32 (9) 3/21 (14)
10 0 0
48 3 0
10/58 (17) 0/3 (0) 0
0 0 10
0 6 45
0 0/6 (0) 10/55 (18)
0 7 3 0
0 6 8 37
0 7/13 (54) 3/11 (27) 0/37 (0)
1 5 4
8 24 19
1/9 (11) 5/29 (17) 4/23 (17)
0 8 2
3 13 35
0/3 (0) 8/21 (38) 2/37 (5)
* A vertebral segment is considered stable if the total SINS score ranges from 0 to 6, indeterminate or potentially unstable at a total score of 7–12, and unstable at scores from 13 to 18.
With respect to OS, the majority of patients died within the follow-up period (21 of 37, 57%). Our analysis identified a significantly prolonged survival in patients with oligometastatic disease compared with patients with more widely disseminated disease (Fig. 1). The 1-year OS rate was 83.9% in patients with oligometastatic disease versus 52.5% for patients with nonoligometastatic disease. This suggests that patients with oligometastatic disease will probably live long enough to benefit from the high rates of prolonged LC resulting from SBRT. In addition, this result supports the recommendation that patients with oligometastatic disease are candidates for spine SBRT as per the American Society for Radiation Oncology evidence-based guideline for bone metastases.14 It is important to note that this therapy represents a major undertaking for radiotherapy departments20 and is associated with major risks of pain flare,6 radiation myelopathy,19,21 and VCF.9,16,22 To date, there are no randomized controlled trials comparing spine SBRT to standard conventional palliative radiotherapy. J Neurosurg: Spine / Volume 21 / November 2014
We observed no case of radiation myelopathy. We attribute this to compliance with published guidelines for spinal cord tolerance for both the radiation-naive patients and the ones treated with repeat irradiation.19,21 However, our 1-year VCF-free rate was 82%, with a median time to VCF of 1.6 months. These data are consistent with previous reports22 and with the RCC spine SBRT series reported by Balagamwala et al.3 (Table 3). We performed a detailed analysis to determine predictors of VCF based on clinical and dosimetric factors as summarized in Table 1 and the SINS-specific factors as summarized in Table 2. We identified single-fraction treatment with 18–24 Gy and the presence of a baseline VCF as significant predictors of subsequent VCF. The presence of a preexisting fracture as a predictor of subsequent fracture makes biomechanical sense,22 as confirmed in a recent large multiinstitutional study evaluating VCF and spine SBRT reported by Sahgal et al.16 (the 61 non-postoperative patients in this paper were part of 715
I. Thibault et al.
Fig. 1. Kaplan-Meier chart showing OS curves for patients with oligometastatic and nonoligometastatic RCC (p = 0.013).
that analysis). Our data showing that the number of SBRT fractions affects the risk of VCF are also in agreement with the study by Sahgal et al. In that study, vertebrae treated with > 19 Gy/fraction had a greater VCF risk than those treated with ≤ 19 Gy/fraction (HR 4.91 for 20–23 Gy/fraction and HR 5.25 for ≥ 24 Gy/fraction relative to patients treated with ≤ 19 Gy/fraction). This is consistent with our result in which 18–24 Gy delivered in a single fraction yielded higher rates of VCF. Radiation necrosis is thought to be a factor in the pathomechanism as reported by Al-Omair et al.,2 and it is well known that the normal tissues are sensitive to dose per fraction. With respect to lytic disease, we observed that almost all tumors (95%) were lytic in our study population and that all VCF occurred in these patients. There were only 3 mixed tumors in our cohort, no purely blastic tumors, and they did not fracture (Table 2). It has been shown in the study by Sahgal et al.16 that lytic disease is indeed a powerful predictor of fracture. That study also concluded that spinal misalignment is predictive; the majority of pa-
Fig. 2. Kaplan-Meier chart showing local control after SBRT in RCC spinal metastases.
716
Fig. 3. Kaplan-Meier chart showing the risk of SBRT-induced VCF in 48 tumors with no baseline VCF compared with 13 tumors with a baseline VCF (p = 0.012).
tients had a kyphotic deformity. In our analysis, because the majority of tumors were lytic (95%), we didn’t have the variability in tumor type to show significance. Furthermore, our patient population predominantly consisted of those with normally aligned spines (90%). Therefore, our finding that baseline fracture and high dose per fraction SBRT are independent predictors of fracture could be considered specific to patients with lytic tumors and normal spinal alignment. The major strengths of this study are the rigorous imaging and clinical follow-up, identification of consecutive patients treated from a prospective database created since the program’s inception, and all patients treated by a single dedicated radiation oncologist (A.S.). The major limitation is the lack of pain outcome assessments performed using a validated instrument. However, when evaluating the 3 published series on RCC spinal metastases treated with spine SBRT (Table 3), high rates of pain relief have been reported. The first report, by Gerszten et al.,11 focused mainly on pain control and reported pain relief in 34 of 38 patients (89%) treated primarily for pain. Their crude LC rate was 88%. Actuarial 1-year rates of radiographic LC have since been reported by the Cleveland Clinic and M. D. Anderson Cancer Center at 71.2% (crude LC in 68 of 88 total tumors [77.3%]) and 82.1% (43 of 55 tumors [78.2%]), respectively. With respect to pain control, both the Cleveland Clinic and M. D. Anderson series used the Brief Pain Inventory to document pain and reported complete response rates for pain in approximately 40%–50% of patients at 6 months. Our clinical observation has been similar, and such high rates of complete response may be the major advantage for spine SBRT, given that rates of complete response with conventional radiation for spine-specific metastases are largely disappointing. With respect to our pattern of failure analysis, we report epidural disease progression as the most common presentation of tumor progression post-SBRT. This is consistent with the literature17 and a result of inherently aggressive tumor biology, relative underdosing to respect J Neurosurg: Spine / Volume 21 / November 2014
Stereotactic body radiotherapy for RCC spinal metastases
Fig. 4. Example of a baseline pathological VCF and subsequent fracture progression observed 6 weeks after spine SBRT. A: Pretreatment sagittal T1-weighted MRI study. B: Pretreatment sagittal CT showing the dose distribution and representative isodose lines. C: Posttreatment sagittal T1-weighted MRI study showing fracture development.
spinal cord tolerance, the dose prescribed, or lack of coverage within the epidural space within the CTV/PTV. In our series neither the absolute spinal cord and thecal sac dose nor their respective biologically effective doses predicted for LC, and none of the dosimetric parameters were predictive. Furthermore, Al-Omair et al. reported that 24 Gy in 1–2 fractions predicted for better LC in the postoperative patient, and that was our most common approach.1 With respect to coverage by the target volume, on radiological review of each case of failure, only 1 case of epidural disease progression could be attributed to marginal miss. Therefore, at this time we surmise that the biology of the disease is a major factor governing failure, and aggressive management to debulk epidural disease with surgery should be considered on a case-by-case basis. This is best described in a recent analysis, in which the presence of high-grade epidural disease predicted for LC, and surgical downgrading by removing the epidural tumor improved LC post-SBRT.1
Conclusions
We report favorable rates of local tumor control with SBRT, specifically in RCC spinal metastases, with a median dose of 24 Gy in 2 fractions. Patients with oligometastatic disease were most likely to benefit given their likelihood of longer-term OS, and this can aid in patient selection for this emerging technique. Although our rate of VCF was clinically acceptable at 16%, in patients with a baseline VCF and/or treated with 18–24 Gy in a single fraction, careful clinical and imaging follow-up is warranted for fracture progression. Disclosure Dr. Arjun Sahgal has received honoraria for previous educational seminars for Medtronic Kyphoplasty division and Elekta; otherwise, there are no conflicts of interest to disclose. Author contributions to the study and manuscript preparation include the following. Conception and design: Sahgal, Fehlings. Acquisition of data: Sahgal, Thibault, Al-Omair, Masucci, Masson-
TABLE 3: Literature review of SBRT for RCC spinal metastases*
Authors & Year
No. of No. of % w/ Pts Spinal Mets Prior RT
SBRT Total Dose Range (Gy)/No. % w/ 1-Yr of Fxs OS
Gerszten et al., 2005
48
60
70
17.5–25/1
NR
Nguyen et al., 2010
48
55
58
24–30/1–5
72
Balagamwala et al., 2012
57
88
22
8–16/1
49
present study
37
71
15
18–30/1–5
64
LC
VCF
imaging-based LC: 88% NR pain improvement: 89% imaging-based LC: 78.2% NR 1-yr spinal tumor PFS: 82.1% pain-free rate at 1 yr: 52% imaging-based LC: 77.3% 14% 1-yr radiographic PFS: 71.2% complete pain relief at 6 & 9 mos: 41% & 67% 1-yr imaging-based LC: 83% 16%; 1-yr VCF free: 82%
Follow-Up (mos) 37 13
5.4
12
* mets = metastases; NR = not reported; PFS = progression-free survival; pts = patients; RT = radiotherapy.
J Neurosurg: Spine / Volume 21 / November 2014
717
I. Thibault et al. Côté, Lochray, Korol, Yee, Fehlings. Analysis and interpretation of data: Sahgal, Thibault, Al-Omair, Masucci, Masson-Côté, Korol, Cheng, Yee, Fehlings, Bjarnason. Drafting the article: Sahgal, Thibault, Masson-Côté, Korol, Cheng, Xu, Yee, Fehlings, Bjarnason. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Sahgal. Statistical analysis: Sahgal, Thibault, Cheng, Xu, Fehlings, Bjarnason. Study supervision: Sahgal. References 1. Al-Omair A, Masucci L, Masson-Cote L, Campbell M, Atenafu EG, Parent A, et al: Surgical resection of epidural disease improves local control following postoperative spine stereotactic body radiotherapy. Neuro Oncol 15:1413–1419, 2013 2. Al-Omair A, Smith R, Kiehl TR, Lao L, Yu E, Massicotte EM, et al: Radiation-induced vertebral compression fracture following spine stereotactic radiosurgery: clinicopathological correlation. Report of 2 cases. J Neurosurg Spine 18:430–435, 2013 3. Balagamwala EH, Angelov L, Koyfman SA, Suh JH, Reddy CA, Djemil T, et al: Single-fraction stereotactic body radiotherapy for spinal metastases from renal cell carcinoma. Clinical article. J Neurosurg Spine 17:556–564, 2012 4. Beuselinck B, Oudard S, Rixe O, Wolter P, Blesius A, Ayllon J, et al: Negative impact of bone metastasis on outcome in clearcell renal cell carcinoma treated with sunitinib. Ann Oncol 22:794–800, 2011 5. Bilsky MH, Laufer I, Fourney DR, Groff M, Schmidt MH, Varga PP, et al: Reliability analysis of the epidural spinal cord compression scale. Clinical article. J Neurosurg Spine 13:324–328, 2010 6. Chiang A, Zeng L, Zhang L, Lochray F, Korol R, Loblaw A, et al: Pain flare is a common adverse event in steroid-naïve patients after spine stereotactic body radiation therapy: a prospective clinical trial. Int J Radiat Oncol Biol Phys 86:638– 642, 2013 7. Choi CY, Adler JR, Gibbs IC, Chang SD, Jackson PS, Minn AY, et al: Stereotactic radiosurgery for treatment of spinal metastases recurring in close proximity to previously irradiated spinal cord. Int J Radiat Oncol Biol Phys 78:499–506, 2010 8. Cox BW, Spratt DE, Lovelock M, Bilsky MH, Lis E, Ryu S, et al: International Spine Radiosurgery Consortium consensus guidelines for target volume definition in spinal stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 83:e597–e605, 2012 9. Cunha MV, Al-Omair A, Atenafu EG, Masucci GL, Letourneau D, Korol R, et al: Vertebral compression fracture (VCF) after spine stereotactic body radiation therapy (SBRT): analysis of predictive factors. Int J Radiat Oncol Biol Phys 84:e343–e349, 2012 10. Fisher CG, DiPaola CP, Ryken TC, Bilsky MH, Shaffrey CI, Berven SH, et al: A novel classification system for spinal instability in neoplastic disease: an evidence-based approach and expert consensus from the Spine Oncology Study Group. Spine (Phila Pa 1976) 35:E1221–E1229, 2010 11. Gerszten PC, Burton SA, Ozhasoglu C, Vogel WJ, Welch WC, Baar J, et al: Stereotactic radiosurgery for spinal metastases from renal cell carcinoma. J Neurosurg Spine 3:288–295, 2005
718
12. Hyde D, Lochray F, Korol R, Davidson M, Wong CS, Ma L, et al: Spine stereotactic body radiotherapy utilizing cone-beam CT image-guidance with a robotic couch: intrafraction motion analysis accounting for all six degrees of freedom. Int J Radiat Oncol Biol Phys 82:e555–e562, 2012 13. Lovelock DM, Zhang Z, Jackson A, Keam J, Bekelman J, Bilsky M, et al: Correlation of local failure with measures of dose insufficiency in the high-dose single-fraction treatment of bony metastases. Int J Radiat Oncol Biol Phys 77:1282– 1287, 2010 14. Lutz S, Berk L, Chang E, Chow E, Hahn C, Hoskin P, et al: Palliative radiotherapy for bone metastases: an ASTRO evidencebased guideline. Int J Radiat Oncol Biol Phys 79:965–976, 2011 15. Nguyen QN, Shiu AS, Rhines LD, Wang H, Allen PK, Wang XS, et al: Management of spinal metastases from renal cell carcinoma using stereotactic body radiotherapy. Int J Radiat Oncol Biol Phys 76:1185–1192, 2010 16. Sahgal A, Atenafu EG, Chao S, Al-Omair A, Boehling N, Balagamwala EH, et al: Vertebral compression fracture after spine stereotactic body radiotherapy: a multi-institutional analysis with a focus on radiation dose and the spinal instability neoplastic score. J Clin Oncol 31:3426–3431, 2013 17. Sahgal A, Bilsky M, Chang EL, Ma L, Yamada Y, Rhines LD, et al: Stereotactic body radiotherapy for spinal metastases: current status, with a focus on its application in the postoperative patient. A review. J Neurosurg Spine 14:151–166, 2011 18. Sahgal A, Larson DA, Chang EL: Stereotactic body radiosurgery for spinal metastases: a critical review. Int J Radiat Oncol Biol Phys 71:652–665, 2008 19. Sahgal A, Ma L, Weinberg V, Gibbs IC, Chao S, Chang UK, et al: Reirradiation human spinal cord tolerance for stereotactic body radiotherapy. Int J Radiat Oncol Biol Phys 82:107–116, 2012 20. Sahgal A, Roberge D, Schellenberg D, Purdie TG, Swaminath A, Pantarotto J, et al: The Canadian Association of Radiation Oncology scope of practice guidelines for lung, liver and spine stereotactic body radiotherapy. Clin Oncol (R Coll Radiol) 24:629–639, 2012 21. Sahgal A, Weinberg V, Ma L, Chang E, Chao S, Muacevic A, et al: Probabilities of radiation myelopathy specific to stereotactic body radiation therapy to guide safe practice. Int J Radiat Oncol Biol Phys 85:341–347, 2013 22. Sahgal A, Whyne CM, Ma L, Larson DA, Fehlings MG: Vertebral compression fracture after stereotactic body radiotherapy for spinal metastases. Lancet Oncol 14:e310–e320, 2013 Manuscript submitted October 7, 2013. Accepted July 11, 2014. This work was presented in part, in abstract form, at the In ter national Stereotactic Radiosurgery Society 2013 Congress in Toronto, Canada, and at the American Society for Radiation On cology 2013 annual meeting in Atlanta, GA. Please include this information when citing this paper: published online August 29, 2014; DOI: 10.3171/2014.7.SPINE13895. Address correspondence to: Arjun Sahgal, M.D., Department of Radiation Oncology, Sunnybrook Health Sciences Centre and the Princess Margaret Cancer Centre, University of Toronto, 2075 Bayview Ave., Toronto, ON M4N 3M5, Canada. email: arjun.sahgal @sunnybrook.ca.
J Neurosurg: Spine / Volume 21 / November 2014
Spine stereotactic body radiotherapy for renal cell cancer spinal metastases: analysis of outcomes and risk of vertebral compression fracture Clinical article Isabelle Thibault, M.D.,1 Ameen Al-Omair, M.D.,1,2 Giuseppina Laura Masucci, M.D., 2 Laurence Masson-Côté, M.D., 2 Fiona Lochray, M.R.T.T.,1 Renée Korol, Ph.D.,1 Lu Cheng, M.Math., 3 Wei Xu, Ph.D., 3 Albert Yee, M.D., 4 Michael G. Fehlings, M.D., Ph.D., 5 Georg A. Bjarnason, M.D., 6 and Arjun Sahgal, M.D.1,2 Department of Radiation Oncology, Sunnybrook Odette Cancer Centre; Departments of 2Radiation Oncology and 3Biostatistics, Princess Margaret Cancer Centre; 4Department of Orthopaedic Surgery, Sunnybrook Health Sciences Centre; 5Division of Neurosurgery and Spinal Program, Toronto Western Hospital; and 6Department of Medical Oncology, Sunnybrook Odette Cancer Centre, University of Toronto, Ontario, Canada 1
Object. The aim of this study was to evaluate local control (LC) and the risk of vertebral compression fracture (VCF) after stereotactic body radiotherapy (SBRT) in patients with renal cell cancer spinal metastases. Methods. Prospectively collected data on 71 spinal segments treated with SBRT in 37 patients were reviewed. The median follow-up was 12.3 months (range 1.2–55.4 months). The LC rate was assessed based on each spinal segment treated and overall survival (OS) according to each patient treated. Sixty of 71 segments (85%) were radiation naive, 11 of 71 (15%) were previously irradiated, and 10 of 71 (14%) were treated with postoperative SBRT. The median SBRT total dose and number of fractions were 24 Gy and 2, respectively. The VCF analysis also included evaluation of the Spinal Instability Neoplastic Score criteria. Results. The 1-year OS and LC rates were 64% and 83%, respectively. Multivariate analysis identified oligometastatic disease (13 of 37 patients) as a positive prognostic factor (p = 0.018) for OS. Of 61 non-postoperative spinal segments treated, 10 (16%) developed VCFs; 3 of 10 were de novo VCFs and 7 of 10 occurred as progression of an existing VCF. The 1-year VCF-free probability rate was 82%. Multivariate analysis identified single-fraction SBRT and baseline VCF as significant predictors of SBRT-induced VCF (p = 0.028 and p = 0.012, respectively). Conclusions. Spine SBRT yields high rates of local tumor control in patients with renal cell cancer. Baseline VCF and 18–24 Gy delivered in a single fraction were predictive of further collapse. Patients with oligometastatic disease may benefit most from such aggressive local therapy, given the prolonged survival observed. (http://thejns.org/doi/abs/10.3171/2014.7.SPINE13895)
R
Key Words • renal cell cancer • spine metastases • oncology • stereotactic body radiotherapy • stereotactic radiosurgery
cell cancer (RCC) metastases have been traditionally considered to be radioresistant, with poor response rates to conventional palliative radiotherapy. However, high dose per fraction radiation delivered using stereotactic body radiotherapy (SBRT) has the potential to maximize local control (LC) safely and to enal
Abbreviations used in this paper: ASIA = American Spinal Injury Association; CBCT = cone-beam CT; CTV = clinical target volume; HR = hazard ratio; LC = local control; OS = overall survival; PRV = planning organ at risk volume; PTV = planning target volume; RCC = renal cell cancer; SBRT = stereotactic body radiotherapy; SINS = Spinal Instability Neoplastic Score; VCF = vertebral compression fracture.
J Neurosurg: Spine / Volume 21 / November 2014
treat patients with locally “curative” intent rather than locally “palliative” intent.20 The SBRT technique has been recently defined by the Canadian Association of Radiation Oncologists as “The precise delivery of highly conformal and image-guided hypofractionated external-beam radiotherapy, delivered in a single or few fraction(s), to an extra-cranial body target with doses at least biologically equivalent to a radical course when given over a protracted conventionally fractionated (1.8–3.0 Gy/fraction) schedule.”20 This article contains some figures that are displayed in color online but in black-and-white in the print edition.
711
I. Thibault et al. Spinal metastases from RCC represent an important site for the application of SBRT, to optimize LC and prevent malignant epidural spinal cord compression. Although surgery can be performed up front for specific indications such as malignant epidural spinal cord compression or as salvage on progression after radiation, surgery can be associated with major risks including intraoperative hemorrhage. Surgery also has associated delays in its initiation with respect to performing preoperative embolization, and results in delays with respect to starting antiangiogenic targeted therapies for RCC postoperatively given the risk of wound complications. The latter may compromise patient outcomes with respect to survival. It is important to note that although targeted therapy for RCC has significant activity, bone metastases often do not respond, and their presence is an independent predictor of worse outcome.4 Hence, aggressive local management to spinal metastases is justified even in the era of targeted systemic therapies for patients with metastatic RCC. The field of spine SBRT is rapidly emerging into mainstream practice; however, it is still considered investigational and not without risk—serious toxicities such as radiation-induced myelopathy and vertebral compression fracture (VCF) have been reported.16,18,19,21,22 Our spine SBRT technique has been evaluated for its precision, safe spinal cord dose tolerances for SBRT have been established, and the risk of VCF for a mixed cohort of patients has been reported.9,12,19,21 The aim of this investigation was to report our outcomes and toxicity in patients with RCC who had spinal metastases, with a focus on LC and the incidence of VCF. In addition, we applied the recently developed Spinal Instability Neoplastic Score (SINS) criteria10 to determine its predictive value for SBRT-induced VCF.
Methods
We identified 71 spinal segments in 37 patients with RCC who had been treated with spine SBRT between October 2007 and August 2012; data were obtained from our prospective spine SBRT database. This retrospective review of clinical and dosimetric data was approved by our ethics review board. Our spine SBRT technique has been previously described in detail. In brief, patients were immobilized in the BodyFIX system (Elekta AB) for spinal segments below the T-3 spinal level. For cervical and upper thoracic spinal segments (up to the T-2 spinal level), patients were immobilized in a thermoplastic head and shoulder mask. The treatment planning CT scan was performed with a slice thickness of 1 mm, and 1.25to 1.5-mm axial T1 and T2 volumetric MRI sequences, comprising at least 1 spinal level above and below the target volume, were fused to the CT. For patients with contraindications to MRI, or in postoperative patients with sufficient MR-induced artifacts that compromised image quality, a treatment planning CT myelogram was performed. The clinical target volume (CTV) was based on an anatomical concept consistent with recently published guidelines.8 We applied a 2-mm planning target volume (PTV), and for the spinal cord a 1.5-mm planning organ 712
at risk volume (PRV) was applied. For the cauda equina, the thecal sac alone was contoured. We applied published dose limits to the critical neural structure including the cord PRV and/or thecal sac.19,21 Treatment was delivered using the Elekta Axesse (Elekta AB, Stockholm, Sweden) equipped with a 4-mm multileaf collimator, Hexapod robotic couch, and cone-beam CT (CBCT) image guidance system. Our delivery system and CBCT approach has been evaluated and its precision reported on. Our image guidance protocol has also been previously reported.12 In brief, the initial patient positioning is verified on the basis of a second CBCT postshift (verification CBCT preSBRT) with a tolerance for repositioning of 1 mm and 1°. A midfraction and a posttreatment CBCT were also performed to determine intrafraction motion. For singlefraction SBRT, we perform 2 intratreatment CBCT scans due to the long treatment times (typically 60–70 minutes). Local control was based on each spinal segment treated (a total of 71), and overall survival (OS) was calculated according to each patient treated (a total of 37). An oligometastatic state was defined as fewer than 5 sites of metastatic disease at the time of spine SBRT. Each patient was followed according to our institutional policy of a complete MRI sequence of the spine at 2- to 3-month intervals. All patients were treated and followed meticulously by a single radiation oncologist (A.S.). Local progression was determined based on the radiologist’s interpretation of the MR images. In cases in which it was unclear if MR signal changes within the bone were secondary to radiation or disease progression, 2 successive MRI evaluations showing enlarged dimensions of the T1 signal abnormality were required for progression. Increased epidural disease or enlargement of paraspinal disease was documented as progression. All MR images were retrospectively reviewed to evaluate the exact pattern of local recurrence and to classify the epidural tumor extent according to the grading scale proposed by Bilsky et al.5 Both the initial treatment planning CT and MRI were reviewed to score patients according to SINS.10 In brief, SINS breaks down radiological and clinical characteristics according to 6 criteria: 1) location; 2) pain; 3) type of bone lesion (lytic, mixed, or blastic); 4) spinal alignment; 5) presence of a VCF at baseline; and 6) involvement of the posterior spinal elements (spinous process, lamina, pedicle).10 A final score classifies the patient as stable, potentially unstable, and unstable. Our interest was in the predictive value of each of the criteria. Data for 61 sites that did not receive operative intervention were used to assess the predictive values of the SINS criteria for outcome. Statistical Analysis
Summary statistics are provided, with frequency and proportion for categorical factors; median and range are given for continuous risk factors. Group-specific LC, OS, and the fracture-free probabilities and confidence intervals were calculated from the start of SBRT, using the Kaplan-Meier method. Univariate analyses were performed using the Cox proportional hazard regression model, and hazard ratios (HRs) and corresponding confidence limits were calculated. Two-sided tests were used, J Neurosurg: Spine / Volume 21 / November 2014
Stereotactic body radiotherapy for RCC spinal metastases with the significance level set at 0.05. Multivariate models were conducted using the stepwise selection procedure. Statistical analyses were performed using version 9.3 of the SAS system and also the R software version 2.14.0.
Results
Baseline Patient, Tumor, and Treatment Characteristics
Among the 37 patients with RCC, 34 (92%) had a prior nephrectomy, and the median interval between the initial diagnosis of RCC and spine SBRT treatment was 26.6 months (range 0.4–189 months). The median age was 63 years (range 33–82 years). Twenty-five patients (68%) were male, 12 (32%) were female, and 13 patients (35%) had oligometastatic disease. The Eastern Cooperative Oncology Group performance status was 0 in 1 (3%), 1 in 25 (67%), and 2 in 11 (30%) patients. Neurological status according to the American Spinal Injury Association (ASIA) scale was normal (ASIA E), except for 1 patient with partial motor loss of function below the neurological level (ASIA D), and another patient had sensory dysfunction secondary to bilateral neural foramina invasion at the level of L-1 but no motor compromise. Other baseline tumor and treatment characteristics for the 71 spinal metastatic segments are presented in Table 1. Eleven of 71 segments (15%) were previously irradiated with 1 prior course of palliative radiotherapy (8 Gy in 1 fraction, n = 1; 20 Gy in 5 fractions, n = 3; and 30 Gy in 10 fractions, n = 7). The median interval between the first radiotherapy course and SBRT was 16 months (range 5.5–56.4 months). Of the 7 of 71 segments (10%) with epidural disease compressing the spinal cord (Bilsky Grade 2), 4 underwent surgery, but 3 of the 7 were not surgical candidates due to medical comorbidities or disease that was too extensive. These 3 segments were treated with SBRT alone. Of the 17 of 71 segments (24%) with a baseline VCF, 3 segments with > 33% loss of vertebral body height were surgically stabilized before SBRT; 1 of the 14 segments that had ≤ 33% loss of height also had a stabilization procedure pre-SBRT, but the majority of segments (13 of 14) that had ≤ 33% collapse were treated with SBRT alone. For the 10 postoperative SBRT cases, surgery consisted of stabilization alone in 2 of 10 segments, decompression without stabilization in 4 of 10 segments, and decompression with instrumented stabilization in 4 of 10. The SINS baseline characteristics of the 61 spinal segments treated with SBRT alone are summarized in Table 2. The median total dose and number of fractions were 24 Gy (range 18–30 Gy) and 2 (range 1–5), respectively. The median volumes of the CTVs receiving 80% and 90% of the prescribed dose were 91.8% (range 64.2%–99.8%) and 87% (range 56.2%–99.5%), respectively. The median volumes of the PTVs receiving 80% and 90% of the prescribed dose were 88.8% (range 57%–99.7%) and 84.1% (range 47.8%–98.8%), respectively. The median spinal cord PRV and thecal sac maximal point doses, normalized in biologically equivalent doses in 2-Gy fractions (nBED or also called EQD2 when using an a/b ratio of 2), were 28.41 Gy2 (range 0.02–59.58 Gy2; n = 53 spinal cord segments) and 37.76 Gy2 (range 1.65–201.22 Gy2; n = 71 J Neurosurg: Spine / Volume 21 / November 2014
thecal sac segments), respectively. The median follow-up after SBRT was 12.3 months (range 1.2–55.4 months). Overall Survival and Local Control
The 1- and 2-year rates and median OS duration were 64.1%, 45.6%, and 1.5 years, respectively. Multivariate analysis identified oligometastatic disease (13 of 37 patients [35%]) versus widespread metastatic disease as the only significant prognostic factor for OS (HR 0.25, 95% CI 0.08–0.79, p = 0.018; Fig. 1). Local progression was observed in 12 of 71 spinal segments. The 1- and 2-year LC rates were 83.4% and 66.2%, respectively (Fig. 2). Among the 12 segments with recurrence, the median time to local recurrence was 11.5 months (range 5–41 months). Patterns of local failure included progression within the vertebral bone alone (2 of 12 [17%]), paraspinal tissue (2 of 12 [17%]), and epidural progression (8 of 12 [66%]; in isolation, 5 of 8 [63%]). Management consisted of decompression and stabilization surgery alone (1 of 12, 8%), decompressive surgery followed by repeat SBRT (1 of 12, 8%), repeat SBRT and no surgery (3 of 12, 25%), systemic therapy alone (4 of 12, 33%), or palliative care (3 of 12, 25%).
Adverse Events
In total, 10 VCFs were observed post-SBRT in the 61 (16%) nonsurgical spinal segments; 3 of 10 were de novo VCFs and 7 of 10 were progression of an existing VCF. The 1-year rate for freedom from any VCF was 82%, and the median time to SBRT-induced VCF among these 10 cases was 47.5 days (range 1–237 days). Ultimately, 6 of 10 remained asymptomatic and were observed, and 4 of 10 required a stabilization surgery. Table 2 presents a summary of those with and without a VCF according to each of the SINS criteria at baseline. Based on SINS, 3 were unstable, 21 were potentially unstable, and 37 were stable. Analysis of baseline patient, tumor, and treatment factors (as summarized in Tables 1 and 2) identified on the multivariate model determined that single-fraction SBRT (HR 5.03, 95% CI 1.19–21.28, p = 0.028) and presence of a baseline VCF (HR 9.25, 95% CI 1.64–52.31, p = 0.012) were the 2 significant predictors of subsequent VCF post-SBRT. At 1 year post-SBRT, the risk of developing a VCF was 9% for multiple fractions (20–30 Gy in 2–5 fractions) versus 25% for single-fraction SBRT (18–24 Gy in 1 fraction), and 7% versus 59% for intact baseline segments compared with segments with a preexisting VCF (Fig. 3). Figure 4 shows an example of fracture progression. With respect to non-VCF events, no acute or late Grade 3–5 toxicities were observed.
Residual Setup Error and Intrafraction Motion
We observed intrafraction motion of 1.25 mm and 0.56° for single-fraction SBRT cases (25 segments) and 0.91 mm and 0.66° for multiple-fraction SBRT cases (34 segments), with 95% confidence. When comparing single- versus multiple-fraction SBRT residual errors, there were no significant differences. In addition, there was no significant difference between the verification, intrascan, and postscan residual setup errors. 713
I. Thibault et al. TABLE 1: Spinal tumor and treatment characteristics in 37 patients with spinal metastases from RCC* Characteristic no. of segments histological finding clear cell RCC papillary RCC spinal level cervical thoracic lumbar sacral paraspinal extension yes no epidural disease grade† 0 1a 1b 1c 2 prior EBRT yes no systemic therapy prior targeted therapy + chemo‡ prior targeted therapy, no chemo‡ bisphosphonate therapy CTV single vertebral segment multiple adj segments w/in 1 CTV total dose/no. of Fxs 18–24 Gy/1 Fx 20–24 Gy/2 Fxs 24–27 Gy/3 Fxs 30 Gy/4–5 Fxs
Overall No. (%)
SBRT Alone
Postop SBRT
71
61
10
69 (97) 2 (3)
59 2
10 0
9 (13) 33 (46) 21 (30) 8 (11)
5 32 16 8
4 1 5 0
30 (42) 41 (58)
22 39
8 2
36 (51) 6 (8) 19 (27) 3 (4) 7 (10)
35 5 15 3 3
1 1 4 0 4
11 (15) 60 (85)
10 51
1 9
30 (42) 16 (23) 13 (18)
16 29 12
0 1 1
39 (55) 32 (45)
36 25
3 7
30 (42) 26 (37) 11 (15) 4 (6)
29 21 8 3
1 5 3 1
* adj = adjacent; chemo = chemotherapy; EBRT = external-beam radiotherapy; Fxs = fractions. † Bilsky grades: 0 = tumor confined within the bone; 1a = epidural impingement, but tumor doesn’t compress the thecal sac; 1b = tumor compresses the thecal sac, no spinal cord abutment; 1c = tumor compresses dura and abuts the cord without displacing it; and 2 = spinal cord compression, but CSF is visible around the cord. ‡ Targeted therapies used for RCC include agents targeting the vascular endothelial growth factor (VEGF) and the mammalian target of rapamycin, such as sunitinib and everolimus, respectively.
Discussion
In this paper we report our favorable SBRT experience for spinal metastases specific to RCC. Metastases from RCC have long been considered radioresistant, and strategies to escalate doses in spinal lesions have been limited by both delivery and treatment planning technology. However, our data demonstrate that SBRT is an advanced radiotherapeutic technique that has matured beyond lung and liver indications to the treatment of both primary and metastatic spinal tumors.20 714
Our results conclude a 1-year actuarial LC rate of 83.4%, and this is consistent with the limited spine-specific SBRT RCC literature as summarized in Table 3.3,11,15 With respect to predictors of LC, none of the baseline clinical and dosimetric predictors investigated (Table 1) were predictive on univariate analysis. This may be due to the limited sample sizes of individual series and the low event rate. Although minimum target dose and degree of epidural disease have been reported as predictors in various series, there has been no consistent predictive factor to guide treatment planning.1,7,13 J Neurosurg: Spine / Volume 21 / November 2014
Stereotactic body radiotherapy for RCC spinal metastases TABLE 2: Baseline SINS classification according to VCF status for 61 spinal segments with no prior surgery SINS Factor location junctional (Oc–C2, C7–T2, T11–L1, L5–S1) mobile spine (C3–6, L2–4) semirigid (T3–10) rigid (S2–5) pain mechanical occasional & nonmechanical pain free bone lesion lytic mixed blastic radiographic spinal alignment subluxation or translation kyphosis or scoliosis normal vertebral body collapse ≥50% 50% body involved none of the above posterolat involvement of spinal elements bilat unilat none of the above SINS classification* unstable indeterminate stability (potentially unstable) stable
VCF (n = 10)
No VCF (n = 51)
Overall Fracture (%)
6 3 1 0
18 9 19 5
6/24 (25) 3/12 (25) 1/20 (5) 0/5 (0)
4 3 3
4 29 18
4/8 (50) 3/32 (9) 3/21 (14)
10 0 0
48 3 0
10/58 (17) 0/3 (0) 0
0 0 10
0 6 45
0 0/6 (0) 10/55 (18)
0 7 3 0
0 6 8 37
0 7/13 (54) 3/11 (27) 0/37 (0)
1 5 4
8 24 19
1/9 (11) 5/29 (17) 4/23 (17)
0 8 2
3 13 35
0/3 (0) 8/21 (38) 2/37 (5)
* A vertebral segment is considered stable if the total SINS score ranges from 0 to 6, indeterminate or potentially unstable at a total score of 7–12, and unstable at scores from 13 to 18.
With respect to OS, the majority of patients died within the follow-up period (21 of 37, 57%). Our analysis identified a significantly prolonged survival in patients with oligometastatic disease compared with patients with more widely disseminated disease (Fig. 1). The 1-year OS rate was 83.9% in patients with oligometastatic disease versus 52.5% for patients with nonoligometastatic disease. This suggests that patients with oligometastatic disease will probably live long enough to benefit from the high rates of prolonged LC resulting from SBRT. In addition, this result supports the recommendation that patients with oligometastatic disease are candidates for spine SBRT as per the American Society for Radiation Oncology evidence-based guideline for bone metastases.14 It is important to note that this therapy represents a major undertaking for radiotherapy departments20 and is associated with major risks of pain flare,6 radiation myelopathy,19,21 and VCF.9,16,22 To date, there are no randomized controlled trials comparing spine SBRT to standard conventional palliative radiotherapy. J Neurosurg: Spine / Volume 21 / November 2014
We observed no case of radiation myelopathy. We attribute this to compliance with published guidelines for spinal cord tolerance for both the radiation-naive patients and the ones treated with repeat irradiation.19,21 However, our 1-year VCF-free rate was 82%, with a median time to VCF of 1.6 months. These data are consistent with previous reports22 and with the RCC spine SBRT series reported by Balagamwala et al.3 (Table 3). We performed a detailed analysis to determine predictors of VCF based on clinical and dosimetric factors as summarized in Table 1 and the SINS-specific factors as summarized in Table 2. We identified single-fraction treatment with 18–24 Gy and the presence of a baseline VCF as significant predictors of subsequent VCF. The presence of a preexisting fracture as a predictor of subsequent fracture makes biomechanical sense,22 as confirmed in a recent large multiinstitutional study evaluating VCF and spine SBRT reported by Sahgal et al.16 (the 61 non-postoperative patients in this paper were part of 715
I. Thibault et al.
Fig. 1. Kaplan-Meier chart showing OS curves for patients with oligometastatic and nonoligometastatic RCC (p = 0.013).
that analysis). Our data showing that the number of SBRT fractions affects the risk of VCF are also in agreement with the study by Sahgal et al. In that study, vertebrae treated with > 19 Gy/fraction had a greater VCF risk than those treated with ≤ 19 Gy/fraction (HR 4.91 for 20–23 Gy/fraction and HR 5.25 for ≥ 24 Gy/fraction relative to patients treated with ≤ 19 Gy/fraction). This is consistent with our result in which 18–24 Gy delivered in a single fraction yielded higher rates of VCF. Radiation necrosis is thought to be a factor in the pathomechanism as reported by Al-Omair et al.,2 and it is well known that the normal tissues are sensitive to dose per fraction. With respect to lytic disease, we observed that almost all tumors (95%) were lytic in our study population and that all VCF occurred in these patients. There were only 3 mixed tumors in our cohort, no purely blastic tumors, and they did not fracture (Table 2). It has been shown in the study by Sahgal et al.16 that lytic disease is indeed a powerful predictor of fracture. That study also concluded that spinal misalignment is predictive; the majority of pa-
Fig. 2. Kaplan-Meier chart showing local control after SBRT in RCC spinal metastases.
716
Fig. 3. Kaplan-Meier chart showing the risk of SBRT-induced VCF in 48 tumors with no baseline VCF compared with 13 tumors with a baseline VCF (p = 0.012).
tients had a kyphotic deformity. In our analysis, because the majority of tumors were lytic (95%), we didn’t have the variability in tumor type to show significance. Furthermore, our patient population predominantly consisted of those with normally aligned spines (90%). Therefore, our finding that baseline fracture and high dose per fraction SBRT are independent predictors of fracture could be considered specific to patients with lytic tumors and normal spinal alignment. The major strengths of this study are the rigorous imaging and clinical follow-up, identification of consecutive patients treated from a prospective database created since the program’s inception, and all patients treated by a single dedicated radiation oncologist (A.S.). The major limitation is the lack of pain outcome assessments performed using a validated instrument. However, when evaluating the 3 published series on RCC spinal metastases treated with spine SBRT (Table 3), high rates of pain relief have been reported. The first report, by Gerszten et al.,11 focused mainly on pain control and reported pain relief in 34 of 38 patients (89%) treated primarily for pain. Their crude LC rate was 88%. Actuarial 1-year rates of radiographic LC have since been reported by the Cleveland Clinic and M. D. Anderson Cancer Center at 71.2% (crude LC in 68 of 88 total tumors [77.3%]) and 82.1% (43 of 55 tumors [78.2%]), respectively. With respect to pain control, both the Cleveland Clinic and M. D. Anderson series used the Brief Pain Inventory to document pain and reported complete response rates for pain in approximately 40%–50% of patients at 6 months. Our clinical observation has been similar, and such high rates of complete response may be the major advantage for spine SBRT, given that rates of complete response with conventional radiation for spine-specific metastases are largely disappointing. With respect to our pattern of failure analysis, we report epidural disease progression as the most common presentation of tumor progression post-SBRT. This is consistent with the literature17 and a result of inherently aggressive tumor biology, relative underdosing to respect J Neurosurg: Spine / Volume 21 / November 2014
Stereotactic body radiotherapy for RCC spinal metastases
Fig. 4. Example of a baseline pathological VCF and subsequent fracture progression observed 6 weeks after spine SBRT. A: Pretreatment sagittal T1-weighted MRI study. B: Pretreatment sagittal CT showing the dose distribution and representative isodose lines. C: Posttreatment sagittal T1-weighted MRI study showing fracture development.
spinal cord tolerance, the dose prescribed, or lack of coverage within the epidural space within the CTV/PTV. In our series neither the absolute spinal cord and thecal sac dose nor their respective biologically effective doses predicted for LC, and none of the dosimetric parameters were predictive. Furthermore, Al-Omair et al. reported that 24 Gy in 1–2 fractions predicted for better LC in the postoperative patient, and that was our most common approach.1 With respect to coverage by the target volume, on radiological review of each case of failure, only 1 case of epidural disease progression could be attributed to marginal miss. Therefore, at this time we surmise that the biology of the disease is a major factor governing failure, and aggressive management to debulk epidural disease with surgery should be considered on a case-by-case basis. This is best described in a recent analysis, in which the presence of high-grade epidural disease predicted for LC, and surgical downgrading by removing the epidural tumor improved LC post-SBRT.1
Conclusions
We report favorable rates of local tumor control with SBRT, specifically in RCC spinal metastases, with a median dose of 24 Gy in 2 fractions. Patients with oligometastatic disease were most likely to benefit given their likelihood of longer-term OS, and this can aid in patient selection for this emerging technique. Although our rate of VCF was clinically acceptable at 16%, in patients with a baseline VCF and/or treated with 18–24 Gy in a single fraction, careful clinical and imaging follow-up is warranted for fracture progression. Disclosure Dr. Arjun Sahgal has received honoraria for previous educational seminars for Medtronic Kyphoplasty division and Elekta; otherwise, there are no conflicts of interest to disclose. Author contributions to the study and manuscript preparation include the following. Conception and design: Sahgal, Fehlings. Acquisition of data: Sahgal, Thibault, Al-Omair, Masucci, Masson-
TABLE 3: Literature review of SBRT for RCC spinal metastases*
Authors & Year
No. of No. of % w/ Pts Spinal Mets Prior RT
SBRT Total Dose Range (Gy)/No. % w/ 1-Yr of Fxs OS
Gerszten et al., 2005
48
60
70
17.5–25/1
NR
Nguyen et al., 2010
48
55
58
24–30/1–5
72
Balagamwala et al., 2012
57
88
22
8–16/1
49
present study
37
71
15
18–30/1–5
64
LC
VCF
imaging-based LC: 88% NR pain improvement: 89% imaging-based LC: 78.2% NR 1-yr spinal tumor PFS: 82.1% pain-free rate at 1 yr: 52% imaging-based LC: 77.3% 14% 1-yr radiographic PFS: 71.2% complete pain relief at 6 & 9 mos: 41% & 67% 1-yr imaging-based LC: 83% 16%; 1-yr VCF free: 82%
Follow-Up (mos) 37 13
5.4
12
* mets = metastases; NR = not reported; PFS = progression-free survival; pts = patients; RT = radiotherapy.
J Neurosurg: Spine / Volume 21 / November 2014
717
I. Thibault et al. Côté, Lochray, Korol, Yee, Fehlings. Analysis and interpretation of data: Sahgal, Thibault, Al-Omair, Masucci, Masson-Côté, Korol, Cheng, Yee, Fehlings, Bjarnason. Drafting the article: Sahgal, Thibault, Masson-Côté, Korol, Cheng, Xu, Yee, Fehlings, Bjarnason. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Sahgal. Statistical analysis: Sahgal, Thibault, Cheng, Xu, Fehlings, Bjarnason. Study supervision: Sahgal. References 1. Al-Omair A, Masucci L, Masson-Cote L, Campbell M, Atenafu EG, Parent A, et al: Surgical resection of epidural disease improves local control following postoperative spine stereotactic body radiotherapy. Neuro Oncol 15:1413–1419, 2013 2. Al-Omair A, Smith R, Kiehl TR, Lao L, Yu E, Massicotte EM, et al: Radiation-induced vertebral compression fracture following spine stereotactic radiosurgery: clinicopathological correlation. Report of 2 cases. J Neurosurg Spine 18:430–435, 2013 3. Balagamwala EH, Angelov L, Koyfman SA, Suh JH, Reddy CA, Djemil T, et al: Single-fraction stereotactic body radiotherapy for spinal metastases from renal cell carcinoma. Clinical article. J Neurosurg Spine 17:556–564, 2012 4. Beuselinck B, Oudard S, Rixe O, Wolter P, Blesius A, Ayllon J, et al: Negative impact of bone metastasis on outcome in clearcell renal cell carcinoma treated with sunitinib. Ann Oncol 22:794–800, 2011 5. Bilsky MH, Laufer I, Fourney DR, Groff M, Schmidt MH, Varga PP, et al: Reliability analysis of the epidural spinal cord compression scale. Clinical article. J Neurosurg Spine 13:324–328, 2010 6. Chiang A, Zeng L, Zhang L, Lochray F, Korol R, Loblaw A, et al: Pain flare is a common adverse event in steroid-naïve patients after spine stereotactic body radiation therapy: a prospective clinical trial. Int J Radiat Oncol Biol Phys 86:638– 642, 2013 7. Choi CY, Adler JR, Gibbs IC, Chang SD, Jackson PS, Minn AY, et al: Stereotactic radiosurgery for treatment of spinal metastases recurring in close proximity to previously irradiated spinal cord. Int J Radiat Oncol Biol Phys 78:499–506, 2010 8. Cox BW, Spratt DE, Lovelock M, Bilsky MH, Lis E, Ryu S, et al: International Spine Radiosurgery Consortium consensus guidelines for target volume definition in spinal stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 83:e597–e605, 2012 9. Cunha MV, Al-Omair A, Atenafu EG, Masucci GL, Letourneau D, Korol R, et al: Vertebral compression fracture (VCF) after spine stereotactic body radiation therapy (SBRT): analysis of predictive factors. Int J Radiat Oncol Biol Phys 84:e343–e349, 2012 10. Fisher CG, DiPaola CP, Ryken TC, Bilsky MH, Shaffrey CI, Berven SH, et al: A novel classification system for spinal instability in neoplastic disease: an evidence-based approach and expert consensus from the Spine Oncology Study Group. Spine (Phila Pa 1976) 35:E1221–E1229, 2010 11. Gerszten PC, Burton SA, Ozhasoglu C, Vogel WJ, Welch WC, Baar J, et al: Stereotactic radiosurgery for spinal metastases from renal cell carcinoma. J Neurosurg Spine 3:288–295, 2005
718
12. Hyde D, Lochray F, Korol R, Davidson M, Wong CS, Ma L, et al: Spine stereotactic body radiotherapy utilizing cone-beam CT image-guidance with a robotic couch: intrafraction motion analysis accounting for all six degrees of freedom. Int J Radiat Oncol Biol Phys 82:e555–e562, 2012 13. Lovelock DM, Zhang Z, Jackson A, Keam J, Bekelman J, Bilsky M, et al: Correlation of local failure with measures of dose insufficiency in the high-dose single-fraction treatment of bony metastases. Int J Radiat Oncol Biol Phys 77:1282– 1287, 2010 14. Lutz S, Berk L, Chang E, Chow E, Hahn C, Hoskin P, et al: Palliative radiotherapy for bone metastases: an ASTRO evidencebased guideline. Int J Radiat Oncol Biol Phys 79:965–976, 2011 15. Nguyen QN, Shiu AS, Rhines LD, Wang H, Allen PK, Wang XS, et al: Management of spinal metastases from renal cell carcinoma using stereotactic body radiotherapy. Int J Radiat Oncol Biol Phys 76:1185–1192, 2010 16. Sahgal A, Atenafu EG, Chao S, Al-Omair A, Boehling N, Balagamwala EH, et al: Vertebral compression fracture after spine stereotactic body radiotherapy: a multi-institutional analysis with a focus on radiation dose and the spinal instability neoplastic score. J Clin Oncol 31:3426–3431, 2013 17. Sahgal A, Bilsky M, Chang EL, Ma L, Yamada Y, Rhines LD, et al: Stereotactic body radiotherapy for spinal metastases: current status, with a focus on its application in the postoperative patient. A review. J Neurosurg Spine 14:151–166, 2011 18. Sahgal A, Larson DA, Chang EL: Stereotactic body radiosurgery for spinal metastases: a critical review. Int J Radiat Oncol Biol Phys 71:652–665, 2008 19. Sahgal A, Ma L, Weinberg V, Gibbs IC, Chao S, Chang UK, et al: Reirradiation human spinal cord tolerance for stereotactic body radiotherapy. Int J Radiat Oncol Biol Phys 82:107–116, 2012 20. Sahgal A, Roberge D, Schellenberg D, Purdie TG, Swaminath A, Pantarotto J, et al: The Canadian Association of Radiation Oncology scope of practice guidelines for lung, liver and spine stereotactic body radiotherapy. Clin Oncol (R Coll Radiol) 24:629–639, 2012 21. Sahgal A, Weinberg V, Ma L, Chang E, Chao S, Muacevic A, et al: Probabilities of radiation myelopathy specific to stereotactic body radiation therapy to guide safe practice. Int J Radiat Oncol Biol Phys 85:341–347, 2013 22. Sahgal A, Whyne CM, Ma L, Larson DA, Fehlings MG: Vertebral compression fracture after stereotactic body radiotherapy for spinal metastases. Lancet Oncol 14:e310–e320, 2013 Manuscript submitted October 7, 2013. Accepted July 11, 2014. This work was presented in part, in abstract form, at the In ter national Stereotactic Radiosurgery Society 2013 Congress in Toronto, Canada, and at the American Society for Radiation On cology 2013 annual meeting in Atlanta, GA. Please include this information when citing this paper: published online August 29, 2014; DOI: 10.3171/2014.7.SPINE13895. Address correspondence to: Arjun Sahgal, M.D., Department of Radiation Oncology, Sunnybrook Health Sciences Centre and the Princess Margaret Cancer Centre, University of Toronto, 2075 Bayview Ave., Toronto, ON M4N 3M5, Canada. email: arjun.sahgal @sunnybrook.ca.
J Neurosurg: Spine / Volume 21 / November 2014
