Spine Stereotactic Body Radiotherapy to Three or More Contiguous Vertebral Levels

Background With survival improving in many metastatic malignancies, spine metastases have increasingly become a source of significant morbidity; achieving durable local control (LC) is critical. Stereotactic body radiotherapy (SBRT) may offer improved LC and/or symptom palliation. However, due to setup concerns, SBRT is infrequently offered to patients with ≥3 contiguous involved levels. Because data are limited, we sought to evaluate the feasibility, toxicity, and cancer control outcomes of spine SBRT delivered to ≥3 contiguous levels. Methods We retrospectively identified all SBRT courses delivered between 2013 and 2019 at a tertiary care institution for postoperative or intact spine metastases. Radiotherapy was delivered to 14–35 Gy in 1–5 fractions. Patients were stratified by whether they received SBRT to 1–2 or ≥3 contiguous levels. The primary endpoint was 1-year LC and was compared between groups. Factors associated with increased likelihood of local failure (LF) were explored. Acute and chronic toxicity was assessed. In-depth dosimetric data were collected. Results Overall, 165 patients with 194 SBRT courses were identified [54% were men, median age was 61 years, 93% had Karnofsky Performance Status (KPS) ≥70, and median follow-up was 15 months]. One hundred thirteen patients (68%) received treatment to 1–2 and 52 to 3–7 (32%) levels. The 1-year LC was 88% (89% for 1–2 levels vs. 84% for ≥3 levels, p = 0.747). On multivariate analysis, uncontrolled systemic disease was associated with inferior LC for patients with ≥3 treated levels. No other demographic, disease, treatment, or dosimetric variables achieved significance. Rates of new/progressive fracture were equivalent (8% vs. 9.5%, p = 0.839). There were no radiation-induced myelopathy or grade 3+ acute or late toxicities in either group. Coverage of ≥95% of the planning target volume with ≥95% prescription dose was similar between groups (96% 1–2 levels vs. 89% ≥3 levels, p = 0.078). Conclusions For patients with ≥3 contiguous involved levels, spine SBRT is feasible and may offer excellent LC without significant toxicity. Prospective evaluation is warranted.


INTRODUCTION
Spine metastases are a common source of morbidity for patients with cancer, as resultant pain and neurologic dysfunction can significantly diminish quality of life (1). Thus, radiation therapy (RT) ± surgery is frequently indicated for improved local control (LC) and/or symptom palliation (1). Historically, conventional external beam radiation therapy (EBRT) was considered the standard of care for patients requiring palliative RT. However, stereotactic body radiotherapy (SBRT) has recently been shown to offer significantly improved LC and motor function, particularly for patients with radioresistant tumor histologies (2)(3)(4)(5)(6). While the prospective phase III trial RTOG 0631 demonstrated that SBRT can be safely and accurately delivered to 2 contiguous vertebral levels using image-guided radiotherapy (IGRT) (7), there are limited data evaluating the feasibility of SBRT to ≥3 spinal levels. One concern is that small setup errors (particularly rotational) could be magnified when treating ≥3 levels resulting in tumor undercoverage or increased dose to critical neural structures such as the spinal cord and cauda equina (8). However, modern treatment techniques such as 6-degree-offreedom (6-DOF) couch tops, full-body immobilization, and cone-beam computed tomography (CBCT) may allow reproducible and accurate SBRT delivery within 1 mm even when targeting ≥3 contiguous vertebral levels (9). If so, the LC, symptom palliation, and decreased toxicity benefits of SBRT could be extended to appropriately selected patients with ≥3 contiguous involved levels who currently are frequently offered conventional EBRT. To demonstrate the feasibility of multilevel spine SBRT, we compared LC, dosimetric, and toxicity data of patients who received treatment to ≥3 contiguous spinal levels to those receiving SBRT to just 1-2 levels.

Study Design
We retrospectively evaluated patients with spine metastases treated with SBRT at a single tertiary care institution from 2013 to 2019. Patients who underwent surgical resection or stabilization before SBRT were eligible for inclusion. Patients with cervical, thoracic, lumbar, or sacral metastases were included. Pertinent patient demographic and disease characteristics were collected including, but not limited to, age, gender, performance status defined using Karnofsky Performance Status (KPS), tumor histology, systemic disease control, receipt of concurrent systemic therapy, site of spine metastases, Spinal Instability Neoplastic Score (SINS), and Bilsky grade. Concurrent systemic therapy was defined as systemic therapy given within 2 weeks of RT. For patients undergoing surgery, the type of surgical intervention was documented. Pertinent information regarding RT was collected including, but not limited to, prescribed dose and fractionation and the size of the planning target volume (PTV). Biologically effective doses (BEDs) were calculated using a/b ratios of 10 for the tumor and 2 for the spinal cord and cauda equina. Using these BED values, dosimetric data were collected including, but not limited to, the minimum and maximum doses to the PTV, dose received by 95% of the PTV (D95%) and D90%, and max point doses to the spinal cord and cauda equina, where applicable. This study was approved by our institutional review board.

Treatment
All patients underwent spine magnetic resonance imaging (MRI) prior to surgery or spine SBRT (General Healthcare 1.5 Tesla MR, Chicago, IL, USA, or Siemens Healthcare 1.5T or 3T MR, Erlangen, Germany). Patients who underwent surgery completed additional postoperative MRI and received a computed tomography (CT) myelogram on the same day as simulation for radiation planning. Patients were simulated in the supine position without intravenous contrast (GE Discovery CT590RT, Chicago, IL, USA). A thermoplastic frameless mask was used for immobilization of the cervical spine. A stereotactic body frame and vac loc bag were used to immobilize patients for the treatment of the thoracic, lumbar, and/or sacral spine. For accurate delineation of target volumes and organs at risk, spine MRI sequences, including but not limited to, axial and sagittal T1 postcontrast and T2, were fused to the CT simulation. For patients receiving postoperative SBRT, both preoperative imaging and postoperative imaging were fused as was the CT myelogram.
The clinical target volume (CTV) included gross osseous and extraosseous disease plus bony anatomy at risk of harboring microscopic disease as per consensus contouring guidelines for both postoperative and definitive spine SBRT (10,11). Regardless of the number of treated levels, the PTV was equivalent to the CTV (no expansion). The spinal cord was delineated on the postsurgery CT myelogram or the MRI T2 axial series for nonoperative patients with a 2-mm circumferential expansion to create a planning risk volume (PRV) avoidance structure. SBRT was prescribed at a dose of 14-35 Gy in 1-5 fractions and normalized with the goal that at least 95% of the PTV received the prescription dose. Spinal cord constraints were based on the The American Association of Physicists in Medicine (AAPM) TG 101 recommendations appropriate for the chosen treatment fractionation (12). Volumetric modulated arc therapy with coplanar arcs utilizing 6 or 10 megavoltage beams was predominantly used for treatment planning (Eclipse version 16, Varian Medical Systems, Palo Alto, CA, USA). Treatments were delivered on a linear accelerator (Varian TrueBeam, Palo Alto, CA, USA) with 6-DOF and daily CBCT to minimize setup uncertainty.

Study Endpoints
The primary study endpoint was the rate of LC at 1 year. LC was assessed using follow-up MRIs of the treated spine that occurred every 3 months post-SBRT. Patients were also followed with clinic appointments at that interval with a detailed history and physical exam. Radiographic tumor response was assessed as per SPIne response assessment in Neuro-Oncology recommendations (13). Serial MRI concerning radiographic evidence of progression was retrospectively reviewed for confirmation of local failure (LF) by two independent neuroradiologists. Serial imaging was routinely used to differentiate between progression and pseudoprogression based upon evolution over time. Where available, perfusion sequences were also utilized. Secondary endpoints included progression-free survival (PFS), overall survival (OS), and rates of acute and chronic toxicity. PFS was defined as the time from the start of SBRT to the time of progression, death, or last follow-up. OS was defined as the time from the start of SBRT to the time of death or last follow-up. Acute and chronic toxicities were defined as per the National Cancer Institute Common Terminology Criteria for Adverse Events v5.0. The proportion of plans where 95% of the PTV was covered by at least 95% of the prescription volume was also assessed to investigate the feasibility of multilevel spine SBRT.

Statistical Analysis
Patients were stratified into two groups. Group A consisted of patients who received SBRT to 1-2 contiguous levels, while group B included patients with ≥3 contiguous treated levels. Continuous and discrete variables were compared between groups using the Mann-Whitney U and Pearson's chi-square tests, respectively (14). Median follow-up was calculated using the reverse Kaplan-Meier (KM) method. KM curves were used to assess the endpoints of OS, PFS, and LC. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated with a twosided significance threshold of 0.05. For analysis, dosimetric variables were dichotomized by the median. Univariate analysis (UVA) was performed using a proportional hazards model to assess for a significant relationship between clinically relevant variables and improved LC. Cox proportional hazards modeling multivariate analyses (MVAs) were then performed using variables that achieved significance on UVA and/or clinically relevant factors to examine for an association with increased likelihood of LF. All statistical analyses were performed using SPSS Statistics version 27 (IBM, Armonk, NY, USA).
With respect to patients with 1-2 treated levels, patients with ≥3 irradiated levels were more likely to have received a surgical intervention (61% vs. 21%, p = 0.001), have received fractionated SBRT rather than single-fraction radiosurgery (94% vs. 80%, p = 0.007), have a larger PTV (150 vs. 52 cc, p = 0.001), have progressive systemic disease (30% vs. 18%, p = 0.001), and have an SINS score greater than 7 (50% vs. 25%, p = 0.001) and a higher Bilsky grade (p = 0.001). Patients with ≥3 irradiated levels were more likely to have breast, prostate, head and neck, or sarcoma histologies and less likely to have NSCLC or RCC (p = 0.04). There was no significant difference between groups by gender, age, receipt of systemic therapy, or performance status.

Local Control and Survival Outcomes
The cumulative incidence of LF for the whole cohort at 6, 12, and 18 months was 6%, 12%, and 18%, respectively ( Figure 1). When stratified by the number of contiguous levels treated, there was no difference in 1-year rates of LF (11% for 1-2 levels vs. 16% for ≥3 levels, p = 0.747, Figure 2). In patients who experienced LF, time to failure did not differ between groups (9.8 for 1-2 levels vs. 7.6 months for ≥3, p = 0.873). On UVA, no demographic or disease characteristics significantly predicted improved LC in patients with ≥3 treated levels (

Acute and Chronic Toxicities
There was no significant difference between the two groups in terms of acute toxicities (18% vs. 21%), which were limited to grade 1-2 fatigue, grade 1 dermatitis, grade 1 nausea, and grade 1 pain flare. There were no grade 3 or higher acute events. Pain flare incidence was 5.5% in the whole cohort in which patients required steroids ± opioids. Regarding chronic toxicity, of the 131 courses of definitive spine SBRT without surgical intervention, 11 cases (8.4%) presented with progressive/new vertebral compression fracture (VCF) over a median follow-up of 15 months. There was no difference between patients with 1-2 (8%) and ≥3 treated levels (10%) in terms of progressive/new VCF (OR 1.18, CI 0.236-5.902, p = 0.839). There were no events of radiation-induced myelopathy, radiculopathy, or plexopathy in either group, and there were no reported grade 3 and above late toxicities.

DISCUSSION
This study represents one of the first analyses to suggest that SBRT to ≥3 contiguous spinal levels may be feasible and similarly safe and effective to treatment to 1-2 levels. These findings have strong clinical relevance because a recent phase III trial demonstrated improved treatment response with SBRT with   respect to conventional EBRT (15), but patients with multilevel disease have been previously considered to be suboptimal candidates for SBRT due to feasibility concerns. Regarding feasibility, coverage of at least 95% of the PTV with at least 95% prescription dose was achievable in 89% of patients with ≥3 treated contiguous levels, no different from the rate seen for patients with 1-2 treated levels. While repositioning with reimaging was required for some fractions, particularly for patients with 4 or more involved levels, there were no issues with treatment delivery. Additionally, treatment for these complex patients was similarly effective to SBRT to 1-2 levels in both our patients and historical controls. Multiple studies have demonstrated that 1-year control rates with spine SBRT of 71%-97% may be superior to tumor control with conventional radiotherapy (2)(3)(4)(5)(16)(17)(18)(19). Our 1-year LF rate of 12% for the entire cohort (1-7 treated levels) was similar to these historical controls. Most importantly, our 1-year LF rate for those patients with ≥3 treated levels was just 16%, which was not statistically different from the rate of LF for patients receiving treatment to 1-2 levels. While patients with ≥3 treated levels were more likely to undergo surgery or have breast or prostate histologies, this imbalance likely does not explain the excellent rates of LC in this subcohort, as these variables did not significantly predict LC on UVA or MVA. While infrequent, the failures in this subgroup were primarily driven by the preponderance of patients with extremely radioresistant histologies such as sarcoma. These findings are similar to those from Beeler et al. (20) showing 1year LC rates for single-level and multilevel SBRT of 95% vs. 85% (p = 0.11). On MVA, progressive systemic disease at the time of SBRT was significantly associated with inferior LC. This relationship may be because poor systemic disease control portends a more treatment-resistant disease process. Additionally, female gender was significantly associated with inferior LC. Given the almost borderline p value for this variable, these results are likely attributable to chance.
Notably, SBRT to ≥3 levels did not result in increased acute or chronic toxicity with respect to our patients with fewer treated levels or to historical controls. While SBRT is known to increase the risk of VCF to about 15% and radiation-induced myelopathy can also be a concern (16,(21)(22)(23), the risk of progressive/new VCF in our patients with ≥3 levels was just 9.5% with a 0% rate of radiation-induced myelopathy. In our study, there was no statistically significant difference between groups in terms of acute or late toxicities and there were no high-grade toxicities regardless of the number of contiguous treated levels. With this excellent durable LC and low incidence of side effects, multilevel contiguous SBRT may be a good option for appropriately selected patients with spine metastases.  Some limitations of this approach and this analysis should be considered. Due to differences in group size, match-pair analysis was not feasible, and there may be unaccounted for differences between groups. However, besides systemic disease control and gender, no variables were significant predictors of LF on MVA, suggesting that the inability to conduct match-pair analysis may have a relatively minimal impact. Spine SBRT, particularly multilevel, poses unique barriers to reliable daily setup and accurate treatment delivery. A sharp dose falloff is required close to critical structures with the potential for around 2-Gy perturbation to the Dmax delivered to the spinal cord if there is even a 1-mm vertical misalignment (9). The utilization of a headrest for patient comfort with a thermoplastic head and neck mask for immobilization for cervical and upper thoracic spine SBRT can result in an air gap with cervical and upper thoracic spine curvature being difficult to reproduce on a daily basis (9). However, modern equipment such as 6-DOF couches can correct these translational and rotational errors (24,25). This capability, in combination with full-body immobilization and CBCT, has led to the achievement of submillimeter accuracy for treatment delivery in about 98% of SBRT cases, particularly for treatments delivered to T2-S5 (9). Thus, at our institution, we used a 0-mm expansion from CTV to PTV and 2-mm PRV expansion around the spinal cord/cauda equina for both postoperative and definitive spine SBRT patients. However, another retrospective study noted mean translational errors of 0.5 mm with a standard deviation of 0.5 mm for multilevel spine SBRT (26). Thus, although our rates of LF were excellent despite no PTV expansion, we acknowledge that PTV expansions should be considered depending upon institutional comfort with setup reproducibility. Additionally, the best candidates for SBRT with multilevel disease are likely to be well-performing patients eligible for debulking and with excellent systemic therapy options; otherwise, the potential benefit in improved durable LC with this approach, with respect to standard palliative techniques, may not be meaningful. Additional limitations primarily stem from the study's retrospective nature and include potential patient selection biases.
In summary, in appropriately selected patients, spine SBRT to 3-7 contiguous vertebral levels is feasible and may offer high rates of LC with minimal toxicity. As a result, the benefits of spine SBRT may be extendable to patients with multilevel disease who have been considered more appropriate for conventional EBRT. A prospective evaluation of this approach is warranted.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by The Ohio State University Institutional Review Board. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.