Stereotactic radiosurgery versus decompressive surgery followed by postoperative radiotherapy for metastatic spinal cord compression (STEREOCORD): Study protocol of a randomized non-inferiority trial

Abstract

Current treatment standard for patients with metastatic spinal cord compression (MSCC) is decompressive surgery followed by radiotherapy. Stereotactic radiosurgery (SRS) could be considered a treatment option for MSCC for patients with minor neurologic deficits. If SRS is safely and effectively delivered with equivalent functional outcome, the patients would avoid the risks associated with an invasive procedure. This paper presents the design of a non-inferiority clinical trial evaluating the safety, tolerability and feasibility of SRS vs. current standard treatment for patients with MSCC. Patients fulfilling inclusion criteria will be randomized 1:1 to each arm. The primary endpoint is ability to walk six weeks after treatment. Secondary endpoints are levels of pain, bladder control, quality of life, response rate, toxicity and number of treatment days. 65 patients in each arm are required for the power of 89% to detect a clinically relevant inferior outcome.

1. BACKGROUND

Metastatic spinal cord compression (MSCC) is an acute event that untreated leads to paralysis and severe pain [1]. It is a serious condition for the individual cancer patient and a challenge for the healthcare system [2]. MSCC or impending MSCC is usually treated with either de-compressive surgery followed by external radiation or external radiation alone. The majority of patients are treated with radiation alone whereas surgery is indicated for among 10-15% of patients [3]. At this time only one randomized clinical trial comparing surgery and radiotherapy versus radiotherapy alone has been completed and published. The study was stopped when interim analysis showed that patient allocated to surgery regained the ability to walk earlier and maintained gait function for a longer period of time [4]. Patchell et al. included 101 patients in years 1992 to 2002, signifying that accrual was very slow and that patient selection was a major challenge. Surgical candidates with a life expectancy of more than three months were considered for inclusion and only a single area of MSCC was allowed. Compression of the cord was not sufficient as an inclusion criterion. Displacement without the patients being paraplegic for more than 48 hours was required. This trial clearly favor decompression combined with radiotherapy but a greater non-neurological morbidity was seen in the radiotherapy group. Since this there has been a tremendous development in radiotherapy techniques. Therefore, a new randomized trial has been proposed to investigate the role of radiotherapy versus surgery among patients with MSCC [5].

The recent technological advances in radiotherapy have enabled the application of stereotactic body radiotherapy (SBRT) and stereotactic radiosurgery (SRS) to cancer patients. SRS has been demonstrated to allow for delivering a high radiation dose to the target volume for MSCC patients, with a steep dose fall-off towards the organs at risk (OAR) [13]. The very close proximity as well as the severe consequences of radiation-induced complications makes the spinal cord the OAR of primary concern in the delivery of SRS to MSCC [14]. No clinical evidence from randomized trials exists for the application of SRS and SBRT modalities for treating MSCC patients, nevertheless the techniques are used outside protocols in some institutions [6]. Presently, clinical trial data involving SRS/SBRT in spinal tumors consist primarily of single institutions series [12, 15, 16].

The primary indication for SBRT is painful metastasis to the spine as seen in the largest report by Gerszten et al [7], where a long-term tumor control of 90% and pain improvement in 86% was reported. Several other series report similar results with both satisfactory tumor control and pain relief [8–11]. The efficacy of SBRT compared to conventional palliative radiotherapy in patients with localized spinal metastasis is presently under evaluation in the RTOG study 0631[12].

The treatment option of radiosurgical decompression in MSCC has been explored by Ryu et al [17]. Sixty-two patients with MSCC confirmed by imaging and mild neurological symptoms were allocated to single fraction SRS of 14-20 Gy. Neurological function was improved in 81%. In a similar study twenty-four patients with MSCC, due to myeloma, were treated with SRS of 10-18 Gy. Seven patients suffered from neurological symptoms and five of these had complete neurological recovery [18]. In the treatment report by Gerszten et al. thirty-five patients had progressive neurological symptoms before SRS but only with compression of the spinal cord and/or cauda equina in one patient. Thirty of these patients had neurological improvement after treatment. All five patients without neurological improvement had received prior radiation to the spine [7]. SRS and SBRT of metastases in the spine, when delivered according to protocol, are considered safe with acceptably low risk of grade 3 toxicity, including nausea, vomiting, diarrhea, fatigue, trismus and pain [8, 9]. A risk of vertebral compression fracture also exists with the use of SBRT/SRS on spine tumors [19]. Risk factors of vertebral compression fractures has been described by Cunha et al. [20]

The optimal use of SRS and SBRT in the treatment of spinal metastasis and MSCC compared to surgery and fractionated radiotherapy has yet to be defined. Treatment series has shown benefit for patients post-operatively and for patients not fit for surgery [21]. Whether stereotactic radiosurgery treatment is beneficial can be tested in a randomized trial comparing this approach with the current standard of treatment. In the present work, we describe a randomized clinical protocol comparing SRS vs. surgery with adjuvant RT for MSCC patients.

2. OBJECTIVES

This is an open label randomized trial between two treatment arms. Patients will be randomized to either the conventional arm (1) consisting of de-compressive surgery followed by RT, or to the experimental arm (2) consisting of SRS as the only treatment. Blinding is not done due to use of surgery and different fractionation regimes.

The primary objective is to determine whether SRS of metastatic spinal cord compression is non-inferior to conventional therapy (arm 1) in terms of symptom control as measured by the ability to walk. The primary endpoint of the trial is the patients’ chance of maintaining ability to walk at 6 weeks from end of treatment. The endpoint will measured by patient reported outcome of dimension of mobility using the EQD5-5L questionnaire. Answers on this dimension have 5 levels: no problems, slight problems, moderate problems, severe problems and extreme problems.

Secondary objectives are

1. Self reported pain by EQD5-5L

2. Self reported bladder control

3. Quality of life measured by EQD5-5L questionnaire

4. Response rate determined by MRI at 6 weeks.

5. Relate toxicity and complications to treatment

6. Determine days from referral to last day of radiotherapy

3. ELIGIBILITY

3.1 Inclusion criteria

• –Confirmed malignancy without curative measures
• –Localized spine metastasis with involvement or compression of spinal cord on MRI
• –A maximum of two separate sites with a maximum of two vertebrae pr. site
• –Eligible for surgery
• –No contraindication for anesthesia
• –No former treatment of metastatic spinal cord compression
• –Patients with mild to moderate neurologic symptoms including radiculopathy, dermatomal sensory change, and muscle strength of involved extremity 4/5 on MRC scale are eligible
• –Age ≥18 years.
• –Eastern Cooperative Oncology (ECOG) performance status ≤ 2.
• –Life expectancy exceeding 3 months.
• –Ability to understand and the willingness to sign a written informed consent document.

3.2 Exclusion criteria

• –Histology of myeloma or lymphoma
• –Patients with any spine metastasis with other than protocol described treatment
• –Spine instability due to a compression fracture or impending vertebral compression fracture
• –Patients with rapid neurologic decline within 24 hours
• –Bony retropulsion causing neurologic abnormality
• –Prior radiation to the index spine;
• –Patients for whom an MRI of the spine is medically contraindicated
• –Patients allergic to contrast dye used in MRIs
• –Patients who are receiving any other investigational agents.
• –Patients with more than two known brain metastases
• –Patient with any serious neurologic condition other than MSCC that could confound the diagnosis and interpretation of radiation induced myelopathy.
• –Uncontrolled intercurrent illness

4. EVALUATION AND RANDOMIZATION

Initially, the patient’s medical history is reviewed and the attending oncologist performs a physical examination. A revised Tokuhashi score [22] along with base line toxicity by CTCAE 4.0 is also registered. Patients perceived control of micturition and use of mobility aid is also recorded.

Randomization will occur once metastatic spinal compression is diagnosed, the patient has fulfilled inclusion/exclusion criteria and the patient has signed the consent form. Randomization will be done by a centralized internet-based online randomization service. Patient ID and treatment will be announced afterwards. The patient ID will be the primary identification of the patient during the study. Randomization will be in a block of four and stratification will occur by primary diagnosis grouped from 0 to 5 as in Tokuhashi score.

5. THERAPY

All patients entering this study should be treated with a glucocorticoid or equivalent and proton pump inhibitor according to local guidelines. The treatment protocol is based on the RTOG 0631 protocol currently recruiting [12] in order to facilitate comparisons.

CT and MR simulation will be performed in the treatment positioning and with the preferred immobilization device. Image fusion between MRI (T1-weighted and T2-weighted images) and simulation CT is required for delineation of both the soft tissue tumor component and the spinal cord. All radiotherapy plans in both conventional and experimental arm must meet constraints for organs at risk (OAR). The treatment plan is to be evaluated regarding target and OAR dosimetry, and will be approved by the attending senior physician and medical physicist. Medical physics staff will confirm all treatment plans and dose delivery.

Use of 6 MV photon beams produced by a dedicated stereotactic linear accelerators with micro leaf-collimator (Novalis Tx or Novalis STx) is preferred (4-18 MV is allowed). Volumetric modulated arc therapy (VMAT) or intensity modulated radiation therapy (IMRT) is preferred but other techniques are allowed.

Before each treatment and in the treatment position, a cone-beam CT (OBI, Varian Medical Systems) will be acquired and used for volume reconstruction and registration onto the planning CT. The patient position will be adjusted according to the CBCT and planning CT registration prior to RT.

6. Conventional treatment arm

Patients randomized to the conventional treatment arm will undergo posterior decompression/laminectomy on relevant spinal levels depending on neurological symptoms. If there is a need of spinal stabilization after decompression, patients will undergo posterior instrumentation with pedicle screws and titanium rods. Instrumentation will be done two or three levels above and below each level with metastatic disease.

Patients randomized to the conventional arm will receive postoperative radiotherapy commencing between 10 to 21 days after de-compressive surgery. The radiation target includes the entire vertebral body and the vertebral arch at the operated level of the vertebral column and will be defined as the clinical target volume (CTV). A 5 mm isotropic uncertainty margin is added to the CTV to account for positioning uncertainty (PTV). Patients receiving postoperative radiotherapy will receive 30 Gy in 10 fractions (i.e. 3 Gy/fr.). The prescribed dose should cover at least 90% of the defined PTV. Positioning will be performed using (only) CBCT prior to RT on all fractions. An example delineation and plan is shown in Figure 1.

Figure 1.

Example of both conventional and experimental treatment plans in two different patients with metastatic compression of the cauda equina at L4 of the spine. In the experimental arm treatment is planned with co-registered axial T1 weighted MRI of the appropriate level of the spine. Absolute dose color wash is shown with lower cut off is set to 3 Gy.

7. Experimental treatment arm

Patients allocated to the experimental treatment arm are treated with stereotactic radiosurgery (SRS) and will receive a prescribed dose of 16 Gy in one fraction. Radiosurgery target volume in this protocol follows partly the target definition from RTOG 0631 protocol but defines a GTV as contouring guidelines by Cox et al. [23]. Note that the guidelines by RTOG and Cox et al. do not include patients with metastatic spinal cord compression that are included in this trial. The radiation target includes the involved part of the vertebral body (CTV) and/or the vertebral arch. The PTV will be defined as CTV with a 2 mm isotropic margin. The VMAT treatment plan will be optimized using three arc fields and should comply with the dose-volume constraints listed in Table 1. The spinal cord constraints are regarded as ‘hard constraints’ and will be complied with at the expense of the prescription dose-coverage of the PTV. We aim for as large coverage of the target as possible of the 16 Gy isodose contour (i.e. the prescription dose). Typically, 90% isodose line is used as prescription line, thus the target will receive a maximum of approximately 18 Gy. The PTV coverage of the 16 Gy isodose contour will be recorded as part of the study but will not be used as basis for exclusion. PTV coverage of >90% of the target volume is defined as “Good” coverage, coverage of < 90% of the target volume is an “Adequate” coverage, and any coverage of < 80% of the target volume is a partial coverage.

Table 1.

The following table lists maximum dose limits to a point or volume within several critical organs recommended for stereotactic body radiation therapy (SBRT). The recommended dose constraints are shown in volume and the maximum dose to the given volume for each organ[24].

Serial tissue	Volume (mL)	Volume Max (Gy)	Endpoint(≥ Grade 3)
Spinal cord	Less than or equal to 0.35	10 Gy	Myelitis
	AND
Spinal Cord	Less than or equal to 10% of the partial spinal cord	10 Gy	Myelitis
	AND
Spinal Cord	Less than or equal to 0.03	14 Gy	Myelitis
Cauda Equina	<0.03 < 5	16 Gy 14Gy	Neuritis
Sacral plexus	<0.03 <5	18 Gy 14.4Gy	Neuropathy
Esophagus	<0.03 <5	16 Gy 11.9 Gy	Stenosis/Fistula
Ipsilateral Brachial Plexus	<0.03 <3	17.5 Gy 14 Gy	Neuropathy
Heart/Pericardium	<0.03 <15	22 Gy 16 Gy	Pericariditis
Great vessels	<0.03 <10	37 Gy 31 Gy	Aneurysm
Trachea and larynx	<0.03 <4	20.2 Gy 10.5 Gy	stenosis/fistula
Skin	<0.03 <10	26 Gy 23 Gy	Ulceration
Stomach	<0.03 <10	16 Gy 11.2 Gy	ulceration/fistula
Duodenum	<0.03 <5	16 Gy 11.2 Gy	Ulceration
Jejunum/ilium	<0.03 <5	15.4 Gy 11.9 Gy	enteritis/obstruction
Colon	<0.03 <20	18.4 Gy 14.3 Gy	colitis/fistula
Rectum	<0.03 <20	18.4 Gy 14.3 Gy	proctitis/fistula
Renal hilum	<2/3 volume	10.6 Gy	malignant hypertension
Parallel tissue	Critical volume (mL)	Critical volume Dose max (Gy)	Endpoint (≤ grade 3)
Lung (right and left)	1000	7.4 Gy	Pneumonitis

Due to the shape of the target and the intersection and/or vicinity of primarily the spinal cord, there can be hot spots in the immediate vicinity outside of the target volume. The areas of high dose (“hot spot”) can e.g. be found in the paraspinal areas or within the paraspinal muscle, or the rib cage and including the intercostal muscles. The limits of the dose outside of the target include: the volume to greater than or equal to 105% of the prescription dose needs to be limited to less than or equal to 3 cm³ and should be within 1 cm of the edge of the target volume. No part of the patient should receive 110% of the prescription outside of the PTV.

The aim of the accuracy of localization should be less than 2 mm from simulation/planning to the end of treatment for the experimental arm. Before treatment a cone-beam CT is acquired and registered to the planning CT, and afterwards shifted into the correct position according to the registration. Subsequently, the patient position is verified using 2D stereoscopic imaging (ExacTrac, BrainLab) prior to the first, second and third arc field, as well as at the end of the treatment. The primary purpose of the CBCT is to ensure that the correct vertebrae are treated and that the patient’s anatomy is not significantly deformed compared to the planning CT. The purpose of the 2D stereoscopic imaging is to enable a quick and accurate positioning verification in order to minimize intra-fractional motion. An example of delineation and plan is shown in Figure 1.

8. EVALUATION AND FOLLOW-UP

At the time of study inclusion all patients will fulfill the EDQ5-5L questionnaire. Patients will receive a letter and be asked to complete and return the EQ5D-5L questionnaire at weeks 6, 12, 26 and 52 by letter. In the case of missing answers patients will be contacted by telephone by the investigators or designated research nurses. In the letter patients will furthermore be asked questions regarding self perceived control of micturition, use of urinary catheter, use of cane, walker, stick or other mobility aid. Imaging of treated lesions will be performed in all patients with MRI at 6 weeks after ended treatment to evaluate the response rate. Response will be classified according to RECIST 1.1. MRI will be performed with T1-weighted, T2-weight and axial imaging. Documentations of treatment related side effects and adverse events are defined by the Common Terminology Criteria for Adverse Events 4.0 (CTCAE).

Fig. 1.

Schedule of follow-up	Week 0	Week 6	Week 12	Week 26	Week 52
Screening for Eligibility	X
MRI	X	X
EDQ5	X	X	X	X	X
CTCAE 4.0	X	X

9. SAMPLE SIZE

This study is designed to examine whether that patients treated with stereotactic radiosurgery have a non-inferior outcome of the primary endpoint after 6 weeks. A non-inferiority design has been chosen because no evidence suggests that stereotactic radiosurgery is superior to decompressive surgery with adjuvant radiation therapy. It would be beneficial to evaluate whether this treatment option could provide improvement in safety, tolerability and feasibility of care compared to surgical decompression [30]. The power calculation is based on a departmental series of 54 comparable patients that have returned EDQ5-3L questionnaires. The hypothesis of inferiority is that the distribution of patients answers on the dimension of mobility deteriorates when answering ‘I have no problems walking about’ from 0.2 to 0.1; ‘I have some problems walking about’ from 0.67 to 0.62 and ‘I am confined to bed’ form 0.13 to 0.28. This reflects a 15 % rise of worst outcome as level of inferiority. We wish to be able to detect such deterioration with 89% probability as the primary design parameter. In order to limit the patient number while prioritizing the detection of a possible inferior outcome of the experimental arm, we choose to set the significance level to two-sided 0.2. I other words, a p-value of 0.1 in favor of the conventional arm is sufficient to draw the conclusion that the experimental arm is potentially inferior and should not be adopted without further study. This corresponds to a one-sided ~10% risk of rejecting the experimental arm as inferior even if the null hypothesis of no difference is true. If there is truly a difference of 15% between the standard arm and experimental arm, then 130 patients are required to be 89% sure that this difference is detected. In other words, we conservatively spend our statistical power on the ability to detect inferiority. Power calculation is obtained by simulation of 1000 trials assuming that the hypothesis of inferiority and equivalence is true and counting the proportion of false positives and false negatives in the simulated trials. The use of EQ-5D-5L in this trial as compared to the historical series is expected to increase power slightly. Patients will be stratified by primary cancer diagnosis grouped by Tokuhashi score.

10. DISCUSSION

Previous studies have suggested that stereotactic radiosurgery could be a non-inferior alternative to surgical decompression and subsequent radiotherapy with regards to the clinical outcome for patients presenting with metastatic spinal cord compression and minor neurologic deficits. Local tumor control could be equivalent with minimal risk in comparison to risk associated with decompression surgery and postoperative conventional radiotherapy. It would be beneficial for patients in palliative care to avoid invasive surgery and shorten time used for fractionated radiotherapy. To accept stereotactic radiosurgery as a first treatment option this treatment has to provide an improvement in safety, tolerability and feasibility of care. As a randomized trial in stereotactic radiotherapy this trial may guide us further in the use of this modality in cancer treatment.

11. ETHICAL AND LEGAL CONSIDERATIONS

No patient can enter this trial without oral information and a signed declaration of consent. Data obtained about patients are protected by the Act on processing of Personal Data and the Act on the Health Act. The Danish Data Protection Agency has approved this project. Approval has been obtained through the joint application procedure of Capital Region of Denmark for the Danish Data Protection Agency. This trial is approved by the ethical committee of Capital Region of Denmark ID number H-3-2014-050. This trial is further registered at clinicaltrial.gov identifier NCT02167633. Patient inclusion and follow-up will be monitored according to adherence to protocol by the Danish unit of Good Clinical Practice (GCP-unit). Data will be revised after 50 patients have reached 6 weeks of follow-up regarding differences in death, side effects and slow accrual despite rigorous efforts to raise accrual. This trial will be terminated if there is a statistical difference regarding spinal cord injury in either conventional or experimental treatment. This trial will also be terminated if any serious unexpected treatment related events occur during treatment. The trial will not commence until further investigations about the nature of the event and safety of enrolled patients have occurred. Serious adverse events are defined as CTCAE grade 4 or 5. If this trial is ended prematurely for safety reasons, all enrolled patients will be informed about this decision and underlying reasons.

12. SPONSORSHIP

The investigators at Department of Oncology, Section of Radiotherapy have initiated this project. The trial is funded by the Department of Oncology, with an unrestricted investigator initiated research grant by Varian Medical Systems. Involved departments cover remaining expenses regarding the use of treatment utilities. The primary investigator Morten Hiul Suppli has no financial attachment to any private enterprises or foundations with interest in the research project stated above. Patients in this trial will receive no remuneration.

13. ABBREVIATIONS

MSCC, Metastatic spinal cord compression; SBRT, Stereotactic body radiotherapy; SRS, Stereotactic radiosurgery; MRI, Magnetic resonance imaging.