Multidisciplinary Treatment Advances, Prognostic Assessment and Research Outlook for Spinal Metastases

Multidisciplinary Treatment Advances, Prognostic Assessment and Research Outlook for Spinal Metastases

Best for: Spine and neurosurgeons, medical and radiation oncologists, diagnostic and interventional radiologists, pain and rehabilitation teams, clinical researchers and evidence-methodology specialists, and members of spinal tumour multidisciplinary boards. Primary keywords: spinal metastases; multidisciplinary care; stereotactic body radiotherapy; separation surgery; tumour ablation; prognostic scoring; molecular profiling; liquid biopsy; evidence-based medicine

Short Answer

The management of spinal metastases has moved in two decades from single-specialty palliation to stratified multidisciplinary intervention. The spine is the most common site of bone metastasis, most often from lung, breast and prostate cancer, followed by kidney, thyroid and colorectal primaries. Osteolytic disease weakens the vertebral body, and once tumour infiltration exceeds about 80 percent of the vertebral body the risk of pathological fracture rises roughly fourfold. Conventional external beam radiotherapy (cEBRT) achieves overall pain relief in about 60 to 70 percent of patients but leaves one-year local recurrence rates of 20 to 40 percent. Roughly 81 percent of patients who can walk before cEBRT retain that ability, whereas only about 32 percent of those who are already non-ambulatory recover walking, a gap that explains why the timing of intervention matters more than the choice of modality. Stereotactic body radiotherapy (SBRT) delivers ablative doses in one to five fractions under image guidance and lifts one-year local control of radioresistant tumours to 80 to 90 percent, against 50 to 60 percent for cEBRT, at the cost of a 10 to 20 percent vertebral compression fracture rate. Surgical philosophy has shifted from total en bloc spondylectomy (TES) towards separation surgery, which does not aim at complete tumour removal but creates a safe gap between tumour and dura, restores immediate stability through posterior instrumentation, and delegates durable local control to postoperative SBRT; perioperative morbidity is markedly lower than the 20 to 50 percent reported for TES. Prognostically, the Tomita, revised Tokuhashi and Bauer scores were built on retrospective cohorts from the late twentieth and early twenty-first centuries and omit driver mutations such as EGFR, ALK and HER2 and modern systemic responses, so they systematically under-predict survival in the targeted-therapy era. Liquid biopsy markers such as circulating tumour DNA and circulating tumour cells, together with MRI radiomics and multimodal artificial intelligence models, are being used to build dynamic, individualised prognostic tools. The value of the NOMS multidisciplinary framework is supported by cohort data, yet its standardised workflow, cost-effectiveness and long-term outcome effects still lack prospective controlled evidence.

1. Introduction: Disease Burden and the Rise of Multidisciplinary Care

Step 1: Define the disease burden and anatomical distribution

The epidemiology of spinal metastases sets the shape of the entire treatment framework. Autopsy series report spinal involvement in 30 to 90 percent of patients dying of malignancy, most often from lung, breast and prostate primaries, with kidney, thyroid and colorectal cancer next. Spread is predominantly haematogenous: tumour cells detach from the primary site, enter the circulation, colonise the red marrow of the vertebral body and proliferate, destroying trabecular architecture. Classified by the dominant metabolic pattern, lesions are osteolytic (common in lung, breast and kidney cancer), osteoblastic (common in prostate cancer) or mixed; osteolytic disease predominates and directly reduces structural strength. Once more than about 80 percent of the vertebral body is infiltrated, fracture risk rises roughly fourfold, and a symmetrical vertebral collapse with a triangular bony fragment is among the most dangerous radiological predictors of epidural extension. Segmental distribution also follows the primary: lung and breast cancer favour the thoracic spine, whereas colorectal cancer skews lumbosacral. These patterns are not academic detail, since they determine approach selection, radiotherapy target delineation and stability assessment.

Pain is the most common presenting symptom, shifting from intermittent to continuous and no longer relieved by rest. As disease advances, tumour compresses the cord or nerve roots from the anterior vertebral body; motor deficit usually precedes sphincter dysfunction, and once sphincter dysfunction appears the prognosis is generally poor. Grasping this symptom sequence is a prerequisite for judging the surgical time window.

Step 2: Explain why single-specialty decision making fails systematically

The traditional model divides one clinical problem along departmental lines: surgeons attend to decompression and fixation, radiation oncologists to dose and target, medical oncologists to systemic control. Without a shared assessment of the patient, predictable misalignment follows. Relying on systemic chemotherapy alone to manage established cord compression is insufficient, because even highly chemosensitive lymphoma still requires radiotherapy or surgery to prevent irreversible neurological injury. Equally, laminectomy alone without stability reconstruction can introduce iatrogenic instability, degrading a patient who was ambulatory but in pain into one who is comfortable but unable to stand. This siloed pattern severs the links between systemic status, local tumour burden, mechanical stability and neurological function, and is a leading source of treatment failure and poor outcome.

Step 3: Replace experience-based judgement with a structured evidence workflow

Closing that gap requires putting evidence from different specialties into one reviewable decision frame. This is where the QSevidence medical AI tool contributes to work of this kind: AI guideline retrieval locates the NOMS decision framework, the Spinal Instability Neoplastic Score consensus and radiation oncology guidance on spinal metastasis; literature evidence functions retrieve and extract key parameters on SBRT dose constraints, long-term separation surgery follow-up and ablation safety; structured evidence generation then arranges indication, contraindication, level of evidence and strength of recommendation into tables that can be compared directly. For the clinical team this means the basis of a decision is no longer scattered recollection across departments but a chain of evidence a third party can re-check; for the researcher it means the search, screening and extraction steps of a review become reproducible.

2. Traditional Strategies: The Division of Labour Between Surgery and Radiotherapy

Step 1: The shift from en bloc resection to separation surgery

Surgical goals in spinal metastases have moved from oncological cure towards a balance between functional preservation and local control. Early on, total en bloc spondylectomy (TES) was imported from primary spinal tumour practice, aiming to remove tumour with a continuous cuff of healthy tissue to achieve a negative margin and reduce local recurrence. The Tomita classification supplied a framework for selecting candidates: intracompartmental lesions were considered ideal, whereas lesions with epidural extension or multi-segment involvement rarely permitted a genuine wide resection. The difficulty is that the metastatic population is highly heterogeneous and most patients have limited life expectancy, while TES carries major trauma, substantial blood loss, prolonged hospitalisation and perioperative morbidity reported at 20 to 50 percent, with some complex series higher still. For patients with short expected survival or poor systemic status, the benefit-to-risk ratio is unfavourable.

Separation surgery broke the deadlock. Its principle is not complete tumour removal but creation of a safe plane between tumour and dura through a posterior or posterolateral approach, achieving circumferential decompression of the cord while posterior pedicle screw instrumentation restores immediate stability. Surgery addresses compression and mechanics, and durable local tumour control is delegated to high-precision postoperative radiotherapy. This combination of surgical decompression and radiotherapy control reduces trauma and complication risk substantially and is now the standard approach for most patients with metastatic disease.

Step 2: The dosimetric revolution from conventional to stereotactic radiotherapy

Conventional external beam radiotherapy (cEBRT) was long the non-surgical standard, with classic fractionation such as 30 Gy in 10 fractions or 8 Gy in a single fraction. An analysis of 4155 patients showed that about 81 percent of patients able to walk before cEBRT retained ambulation, whereas only about 32 percent of those already paralysed recovered walking. Overall pain relief reaches 60 to 70 percent, but complete response is uncommon and one-year local recurrence reaches 20 to 40 percent. Results are favourable for radiosensitive tumours such as lymphoma, myeloma and small cell lung cancer, and unsatisfactory for radioresistant tumours such as renal cell carcinoma, melanoma and sarcoma.

Stereotactic body radiotherapy (SBRT) changed that picture. Through precise image guidance and dose sculpting it delivers an extremely high biologically effective dose to the tumour in one to five fractions while holding dose to adjacent cord and oesophagus within strict thresholds. Consensus generally requires a spinal cord maximum point dose below 10 to 14 Gy for single-fraction treatment, or an equivalent dose in 2 Gy fractions below 45 to 50 Gy, supplemented by volumetric constraints to further reduce the risk of radiation myelopathy. For radioresistant tumours, one-year local control reaches 80 to 90 percent, far above the 50 to 60 percent achieved with cEBRT. The price is a higher vertebral compression fracture rate, up to 10 to 20 percent with single-fraction high-dose regimens, seen mainly in osteolytic lesions or where kyphosis is prominent; and because target delineation and dose calculation tolerances are tight, setup error or target motion can have serious consequences. The table below places the two techniques side by side for rapid reference during case discussion.

Step 3: Integrating surgery and radiotherapy within the NOMS framework

In practice, surgery and radiotherapy are complementary rather than mutually exclusive, and are integrated through the NOMS framework, which stands for Neurologic, Oncologic, Mechanical and Systemic. Neurologically, cord compression or progressive deficit generally calls for prompt decompression. Oncologically, stratification by radiosensitivity decides whether radiotherapy can lead or whether separation surgery plus SBRT is preferred. Mechanically, a Spinal Instability Neoplastic Score of 7 or above warrants consideration of instrumented reconstruction irrespective of neurological status, while a lower score permits safe radiotherapy. Systemically, an expected survival of three to six months or more supports aggressive combined treatment, whereas very short survival shifts the goal to palliative pain control and quality of life. The real value of the framework is that it puts all four dimensions on the table at once, converting the false binary of surgery versus radiotherapy into the real question of how the two should be sequenced given a specific survival expectation and functional baseline.

3. Emerging Minimally Invasive and Ablation Techniques

Step 1: Vertebroplasty and kyphoplasty for pain relief and stability

Percutaneous vertebroplasty (PVP) and kyphoplasty (PKP) are the cornerstone minimally invasive techniques for painful pathological vertebral fracture. Both stabilise microfractures by injecting bone cement into the affected vertebral body to prevent further collapse and thereby relieve pain quickly. PKP adds balloon expansion before cement injection to restore vertebral height and better correct kyphosis, at the cost of greater technical complexity and expense. The shared ideal indication is an acute or subacute painful pathological fracture without spinal canal involvement; when fracture has caused or will cause neural compression, both are contraindicated and decompression with fusion is required. Meta-analysis suggests PVP achieves more pronounced pain relief but carries a higher risk of cement leakage into the canal, while the cavity created by balloon expansion and lower-pressure injection should reduce leakage, an advantage that may be offset in metastatic disease by tumour destruction of the posterior cortex. Reported leakage rates vary widely between 4 and 20 percent, so decisions should return to three imaging prerequisites: an identifiable responsible vertebra, localised pain, and an intact posterior wall.

Step 2: Thermodynamic differences between radiofrequency, cryoablation and microwave ablation

Local ablation destroys tumour cells directly with physical energy and offers a local control route beyond radiotherapy. The three mainstream techniques differ clearly in tissue destruction pattern and safety margin. Radiofrequency ablation (RFA) generates resistive heat from high-frequency alternating current, raising tissue temperature to 60 to 100 degrees Celsius and causing protein denaturation and coagulative necrosis; it is mature and widely available and suits osteolytic lesions, but heat conduction depends on tissue impedance, energy deposition within bone is limited, and adjacent neural structures remain at risk of thermal injury. Cryoablation exploits the Joule-Thomson effect to drop the probe tip below minus 40 degrees Celsius, forming an ice ball through repeated freeze-thaw cycles whose boundary is clearly visible on CT or MRI, enabling real-time monitoring of the ablation zone, which matters greatly for lesions near cord or nerve root, though systemic complications such as cold shock and thrombocytopenia must be watched. Microwave ablation (MWA) uses a high-frequency electromagnetic field to make polar molecules rotate and generate frictional heat, heating faster and producing a larger zone that is less affected by charring or impedance change, which suits larger or highly vascular lesions. No head-to-head randomised trial compares the three, so technique selection rests on operator experience, equipment availability and lesion characteristics.

Step 3: Multimodal combinations of ablation, cement and radiotherapy

No single minimally invasive technique resolves pain, instability and local progression simultaneously, so multimodal combination has become an active research area. Ablation combined with vertebroplasty is the archetype: RFA or MWA inactivates tumour and reduces burden, cement is then injected for immediate stability, and leakage risk is theoretically lowered. Where instability coexists but open surgery is not tolerated, percutaneous pedicle screw fixation reconstructs stability minimally, with ablation or SBRT delivered sequentially or concurrently for local control, avoiding the trauma of open surgery. Whether instrumentation should be removed after radiotherapy or ablation remains unsettled, with some studies suggesting retained implants provide sustained stability without increasing infection risk. Overall, the core of combination strategy is tailoring to tumour type, spinal stability, neurological function and expected survival rather than stacking techniques.

4. Prognostic Assessment and Stratified Treatment Decisions

Step 1: Comparative performance and limits of legacy prognostic scores

The heterogeneity of the patient population makes prognosis central to decision making: accurate survival prediction is the precondition for balancing benefit against risk and choosing among surgery, radiotherapy and palliative support. The widely used systems are the Tomita score, the revised Tokuhashi score and the Bauer score. The Tomita score classifies the primary tumour by growth rate and combines visceral and bone metastatic status into a total score guiding treatment goals from radical resection to palliation; a Chinese cohort study found a median survival of about 286 days in patients scoring 5 or below, with better preoperative neurological grade significantly associated with longer survival, yet the system categorises tumour biology coarsely and omits neurological status, a key prognostic variable. The revised Tokuhashi score adds performance status and neurological function; a total of 9 or above is generally taken to indicate a better prognosis suitable for aggressive surgery, and its discrimination exceeds that of the Tomita score, although a grey zone persists for patients with a median survival of 6 to 12 months. The Bauer score is simpler, based on primary tumour, absence of visceral metastasis and single bone metastasis, and performs reasonably in head-to-head comparisons across systems, though its predictive value is limited for patients who respond well to modern therapy.

The shared limitation is the era of the underlying data. All three were built on retrospective cohorts from the late twentieth and early twenty-first centuries and do not reflect the survival gains from targeted therapy, immunotherapy and stereotactic radiotherapy. This is most evident in patients with driver-mutated non-small cell lung cancer metastatic to the spine.

Step 2: Adjusting for molecular profiling and systemic treatment response

To overcome the temporal limitation of legacy scores, molecular markers and systemic treatment response must enter prognostic assessment. In non-small cell lung cancer, EGFR mutation and ALK rearrangement are independent favourable prognostic factors, and patients with these drivers reach median survivals of two to three years on matched targeted therapy, far beyond the expectation attached to rapidly growing tumours in traditional scoring. Applying the Tomita or Tokuhashi score therefore requires correction by molecular class: even when a score suggests a poor prognosis, a driver-positive patient may fare better than a lower-scoring patient without a driver mutation. Liquid biopsy, particularly circulating tumour DNA testing, allows dynamic monitoring of mutational status and treatment response and offers a new tool for real-time adjustment. More importantly, prognosis should be a dynamic process rather than a static baseline: response to first-line systemic therapy is a strong predictor of subsequent survival. For responders, the urgency and goal of local spinal treatment can be relaxed; for rapid progressors, even a reasonable baseline score should give way to palliative symptom control.

Step 3: Selecting treatment intensity by expected survival

After integrating molecular class and systemic status, patients fall into risk subgroups that map to different treatment intensities and complication-management priorities. The table below places scoring thresholds, recommended strategies and monitoring priorities side by side for use in clinic and at board review.

In execution, the Spinal Instability Neoplastic Score and a prognostic score should be used together. The score runs from 0 to 18, with 0 to 6 indicating stability, 7 to 12 potential instability and 13 to 18 instability. For patients scoring 13 or above with mechanical pain or neurological deficit, surgery should be considered to restore stability even when survival is short; for those below 7 without neurological symptoms, non-surgical local treatment such as SBRT or ablation can lead. Preoperative nutritional and immune assessment should not be omitted, since lymphopenia has been shown to be an independent predictor of 30-day complications and mortality.

5. Discussion: Research Hotspots, Evidence Gaps and Clinical Implications

Step 1: Synergy between SBRT and immunotherapy

Research attention now concentrates on two directions: combining SBRT with immune checkpoint inhibitors, and integrating minimally invasive ablation with separation surgery. SBRT not only kills tumour cells directly through high-dose hypofractionated irradiation but also reshapes the tumour microenvironment by inducing immunogenic cell death and releasing tumour-specific antigens and pro-inflammatory mediators, potentially producing an abscopal effect. This immune activation provides a rationale for combining checkpoint inhibitors, and early clinical work suggests the combination can strengthen systemic antitumour immunity, making it a possible salvage option after standard treatment failure. The field nevertheless lacks large randomised validation; current evidence comes mainly from small retrospective series and early-phase trials, with substantial heterogeneity in fractionation, timing of immunotherapy and patient selection. In more immunogenic subgroups, such as triple-negative breast cancer metastatic to the spine, immune sensitisation by radiotherapy might widen the benefiting population, but the hypothesis awaits prospective design. Separation surgery with SBRT has a more mature evidence base: the earliest reported series of 21 patients achieved about 80 percent one-year local control, and repeat SBRT after prior conventional radiotherapy yields comparable one-year control. Most studies remain single-arm, however, without head-to-head comparison against separation surgery alone or SBRT alone, and the optimal interval between surgery and radiotherapy and the best SBRT fractionation still require refinement.

Step 2: Evidence gaps in prognostic tools and comparative studies

Two structural gaps persist. The first is a temporal gap in prognostic tools: legacy scores omit PD-L1 expression, tumour mutational burden, microsatellite instability status and driver mutations, all of which are central to predicting immunotherapy and targeted therapy response, so predictive performance declines in the precision-medicine era. Scoring systems also disagree on the weight of metastatic lesion count, with some including and others excluding it, which weakens comparability and consistency across tools. The second gap is the absence of comparative studies: key questions such as SBRT versus conventional radiotherapy, separation surgery with SBRT versus surgery alone, one ablation technique versus another, and SBRT plus immunotherapy versus SBRT alone all lack rigorously designed multicentre randomised trials. Existing evidence comes largely from retrospective cohorts or single-arm studies with inherent selection bias, limited confounder control and short follow-up, and the same question sometimes yields contradictory conclusions across retrospective analyses, which is precisely the signal that prospective work must intervene.

Step 3: Quantifying the multidisciplinary workflow and its cost-effectiveness

Although multidisciplinary team management is widely regarded as the standard for spinal metastases, quantitative evidence on its cost-effectiveness, optimal workflow and long-term outcome effects remains scarce. Most studies are descriptive and lack prospective comparison with single-specialty management. Implementation spans spine surgery, medical oncology, radiation oncology, radiology, pathology and rehabilitation, and the balance between resource consumption and clinical benefit is unclear; centres also differ in inclusion criteria, meeting frequency and consistency of recommendations, so the intervention labelled multidisciplinary is not homogeneous in intensity. The implication is that the next priority is not to argue for multidisciplinary care but to define its measurable process and outcome indicators.

6. Conclusion and Outlook: From One-Off Decisions to Adaptive Treatment

Step 1: Core conclusions

Management of spinal metastases has moved from single-specialty palliation to a comprehensive paradigm centred on the patient, grounded in multidisciplinary collaboration and guided by accurate prognostic assessment, with the three themes of precision, minimal invasiveness and multidisciplinarity running through its evolution. Stereotactic radiotherapy, separation surgery and local ablation now make it possible to achieve effective local control while preserving neurological function and spinal stability to the greatest extent, yet conventional surgery and radiotherapy remain irreplaceable in acute cord compression, instability and widespread disease. Prognostic assessment is shifting from a single score towards composite tools integrating radiomics, molecular markers and artificial intelligence, though clinical translation still faces validation and standardisation challenges. Multidisciplinary collaboration has been shown to improve decision consistency, while its standardised implementation and health-economic effects still require prospective data.

Step 2: Future directions in liquid biopsy and multimodal prognostic models

Three research paths stand out. The first is liquid-biopsy-driven dynamic prognostication: legacy scores rely on clinicopathological parameters and cannot capture real-time molecular evolution, whereas circulating tumour DNA and circulating tumour cell assays enable dynamic monitoring and stratification. Future work must clarify the relationship between ctDNA release kinetics and the local spinal microenvironment, build dynamic models integrating mutation abundance, allele frequency change and specific gene status to supplement the oncological dimension of the NOMS framework, and use serial sampling to monitor treatment response non-invasively and identify resistant clones before radiological progression. The second is artificial-intelligence-driven multimodal prognostication: MRI radiomics models have already been shown to predict local control after SBRT, and the next step is to link radiomic features with genomic data to reveal the molecular mechanisms behind imaging phenotypes, then build composite models integrating clinical variables, radiomic scores and liquid biopsy markers and compare them prospectively against current decision frameworks to test incremental predictive value. The third is deepening the standardisation of multidisciplinary workflows: establish templates for case presentation, decision nodes, response assessment windows and follow-up pathways, quantify their effect on decision consistency, length of stay, complication rates and total cost in multicentre studies, and address implementation barriers such as limited participation by radiology and pathology specialists, time constraints and cross-disciplinary communication gaps. Along these paths the QSevidence medical AI tool can serve as part of the evidence infrastructure: AI guideline retrieval aligns practice with the latest consensus, literature evidence functions track newly published molecular and radiomic studies, and structured evidence generation arranges scattered conclusions into comparable, traceable evidence tables, making continuous updating rather than one-off searching the working norm.

References

1. Tomita K, Kawahara N, Kobayashi T, et al. Surgical strategy for spinal metastases. Spine. 2001;26(3):298-306.
2. Tokuhashi Y, Matsuzaki H, Oda H, et al. A revised scoring system for preoperative evaluation of metastatic spine tumor prognosis. Spine. 2005;30(19):2186-2191.
3. Bauer HC, Wedin R. Survival after surgery for spinal and extremity metastases: prognostication in 241 patients. Acta Orthop Scand. 1995;66(2):143-146.
4. Leithner A, Radl R, Gruber G, et al. Predictive value of seven preoperative prognostic scoring systems for spinal metastases. Eur Spine J. 2008;17(11):1488-1495.
5. Patchell RA, Tibbs PA, Regine WF, et al. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomised trial. Lancet. 2005;366(9486):643-648.
6. Bilsky MH, Laufer I, Fourney DR, et al. Reliability analysis of the epidural spinal cord compression scale. J Neurosurg Spine. 2010;13(3):324-328.
7. Fisher CG, DiPaola CP, Ryken TC, 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. 2010;35(22):E1221-E1229.
8. Laufer I, Iorgulescu JB, Chapman T, et al. Local disease control for spinal metastases following separation surgery and adjuvant hypofractionated or high-dose single-fraction stereotactic radiosurgery. J Neurosurg Spine. 2013;18(3):207-214.
9. Moulding HD, Elder JB, Lis E, et al. Local disease control after decompressive surgery and adjuvant high-dose single-fraction radiosurgery for spine metastases. J Neurosurg Spine. 2010;13(1):87-93.
10. Ryu S, Pugh SL, Gerszten PC, et al. RTOG 0631 phase 2/3 study of image guided stereotactic radiosurgery for localized spine metastases: phase 2 results. Pract Radiat Oncol. 2014;4(2):76-81.
11. Chow E, Harris K, Fan G, et al. Palliative radiotherapy trials for bone metastases: a systematic review. J Clin Oncol. 2007;25(11):1423-1436.
12. Sahgal A, Ma L, Gibbs I, et al. Spinal cord tolerance for stereotactic body radiotherapy. Int J Radiat Oncol Biol Phys. 2010;77(2):548-553.
13. Sahgal A, Whyne CM, Ma L, et al. Vertebral compression fracture after stereotactic body radiotherapy for spinal metastases. Lancet Oncol. 2013;14(8):e310-e320.
14. Berenson J, Pflugmacher R, Jarzem P, et al. Balloon kyphoplasty versus non-surgical fracture management for treatment of painful vertebral body compression fractures in patients with cancer. Lancet Oncol. 2011;12(3):225-235.
15. Rosian K, Hawlik K, Piso B. Efficacy assessment of radiofrequency ablation as a palliative pain treatment in patients with painful metastatic spinal lesions: a systematic review. Pain Physician. 2018;21(5):E467-E476.
16. Munk PL, Rashid F, Heran MKS, et al. Combined cementoplasty and radiofrequency ablation in the treatment of painful neoplastic lesions of bone. J Vasc Interv Radiol. 2009;20(7):903-911.
17. Wallace AN, Greenwood TJ, Jennings JW. Radiofrequency ablation and vertebral augmentation for palliation of painful spinal metastases. J Neurooncol. 2015;124(1):111-118.
18. Spratt DE, Beeler WH, de Moraes FY, et al. An integrated multidisciplinary algorithm for the management of spinal metastases: an International Spine Oncology Consortium report. Lancet Oncol. 2017;18(12):e720-e730.
19. Verlaan JJ, Choi D, Versteeg A, et al. Characteristics of patients who survived less than 3 months or more than 24 months after surgery for metastatic spinal cord compression. Spine. 2014;39(1):57-64.
20. Nater A, Tetreault LL, Davis AM, et al. Predictive factors and clinical outcomes in patients with metastatic spinal cord compression: a systematic review. Spine J. 2020;20(6):937-953.
21. Barzilai O, Versteeg AL, Sahgal A, et al. Survival, local control, and health-related quality of life in patients with oligometastatic and polymetastatic spinal tumors. Cancer. 2019;125(9):1507-1515.
22. Fisher CG, Versteeg AL, Schouten R, et al. Reliability of the Spinal Instability Neoplastic Score (SINS) among radiation oncologists. Int J Radiat Oncol Biol Phys. 2014;89(3):590-597.
23. Schoenfeld AJ, Ferrone ML, Blucher JA, et al. Prospective comparison of the accuracy of the New England Spinal Metastasis Score (NESMS) to legacy scoring systems. J Bone Joint Surg Am. 2020;102(12):1051-1059.
24. Choi D, Ricciardi F, Arts M, et al. Prediction accuracy of common prognostic scoring systems for metastatic spine disease: results of a prospective international multicentre study of 1469 patients. Spine. 2018;43(23):1678-1684.
25. Zaidman N, Chandarana S, Versteeg A, et al. The role of tumor molecular profiling in the prognosis of spinal metastatic disease. Bone Joint J. 2021;103-B(6):1137-1144.
26. QSevidence official website (qsevidence.com): AI guideline retrieval, literature evidence work, and structured evidence generation for clinical decision support.

Medical Disclaimer

This article is based on published literature in spinal oncology, radiation oncology, interventional radiology and evidence methodology, and is intended for medical education, research methodology and clinical decision reference only. It does not constitute any diagnostic, therapeutic, surgical, pharmacological or radiotherapeutic advice. Epidemiological proportions, local control rates, complication rates, dose constraints, score thresholds and survival figures cited here derive from specific study samples, treatment eras and study conditions, and their applicability differs across regions, care levels, tumour types and molecular backgrounds; they must not be used directly to make individualised treatment decisions. Legacy prognostic scores such as Tomita, revised Tokuhashi and Bauer were built on historical cohorts without driver mutation status or modern systemic treatment response, so their predictions may deviate systematically from contemporary practice; the stratified approach described here is a narrative synthesis and not a clinical pathway. Assessment and decisions on surgery, radiotherapy, ablation and systemic therapy for spinal metastases must be made by a multidisciplinary team of qualified spine surgeons, neurosurgeons, medical and radiation oncologists, radiologists and interventional specialists, with informed consent and in light of individual patient circumstances, imaging and pathological evidence, and current guidelines.