This prospective, open-label randomized trial evaluates a dual-simulation planning strategy that combines standard brachytherapy TPS with patient-specific biomechanical modeling for radioactive seed implantation in bone metastases. The approach aims to improve dose coverage while accounting for fracture risk, needle path stability, and seed migration. Eligible patients with painful and/or progressive bone metastases are randomized to dual-simulation planning versus conventional TPS. All undergo image-guided implantation with post-implant dosimetric verification and standardized follow-up. The primary endpoint is 3-month pain response (BPI/VAS, adjusted for analgesic use). Secondary endpoints include dosimetry (D90, V100, CI, HI), local control/progression, seed migration, skeletal-related events and fractures, SINS and functional status, quality of life, procedure-related complications (CTCAE v5.0), and procedure metrics. We hypothesize the dual-simulation strategy will enhance dosimetric quality and reduce biomechanics-related complications, improving pain and function.

This study aims to evaluate a "dual-simulation" optimization strategy that combines a brachytherapy treatment planning system (TPS) with biomechanical modeling to improve the feasibility, safety, and effectiveness of radioactive seed implantation for bone metastases (palliative/local control). Conventional TPS is primarily dose-centric and may not adequately account for the mechanical stability of metastatic bone lesions, feasibility of needle trajectories, or risk of seed migration, potentially leading to suboptimal dose distribution or increased post-procedural biomechanics-related adverse events. To address this gap, we integrate a patient-specific finite element biomechanical model into standard TPS to predict load-bearing behavior, fracture risk, needle path stability, and seed migration risk, enabling iterative, dose-mechanics constrained plan optimization.

This is a prospective, open-label, randomized controlled trial comparing "TPS + biomechanical dual-simulation" versus conventional TPS. Eligible participants are patients with bone metastases who meet indications for radioactive seed implantation and have pain and/or risk of local progression. In the experimental arm, preoperative imaging segmentation and individualized biomechanical modeling inform coupled optimization of needle trajectories and seed distributions; the control arm receives standard TPS-based planning. All patients undergo image-guided implantation, with post-implant dosimetric verification and standardized follow-up.

The primary endpoint is pain response at 3 months (per BPI or VAS, accounting for changes in analgesic use). Secondary endpoints include dosimetric parameters (e.g., D90, V100, conformity index [CI], homogeneity index [HI]), local control rate and time to progression, seed migration incidence, skeletal-related events (SREs) and pathologic fracture incidence/time, changes in SINS score and functional status (e.g., ECOG, TESS), quality of life (EORTC QLQ-C30), procedure-related complications (CTCAE v5.0), and procedure time/number of needle adjustments. Safety will be assessed at prespecified time points. Imaging will be performed at baseline, post-procedure, and during follow-up, with standardized post-implant dosimetric verification.

We hypothesize that the dual-simulation strategy will maintain or improve dose coverage and conformity while reducing seed migration and biomechanics-related complications, thereby improving pain relief and functional outcomes, and providing a more comprehensive, individualized optimization pathway for radioactive seed implantation in bone metastases.

Inclusion Criteria:

• Age ≥18 years.

Radiologically or histologically confirmed bone metastasis with a lesion suitable for percutaneous radioactive seed implantation (e.g., I-125), per multidisciplinary assessment.

Indication for local palliation/control: moderate-to-severe pain at lesion (e.g., BPI/VAS ≥4) and/or imaging evidence of progression or high-risk features warranting local therapy.

Measurable/evaluable target lesion on CT/MRI; target location accessible for needle placement per institutional practice.

ECOG performance status 0-2.

Estimated life expectancy ≥3 months.

Adequate hemostasis: platelets ≥80×10^9/L, INR ≤1.5 (or per protocol), and able to hold/bridge anticoagulation as clinically indicated.

Adequate organ function to undergo the procedure and anesthesia/sedation per site standards.

Able to undergo required imaging (CT; MRI if applicable).

Willing and able to provide written informed consent and comply with follow-up.

For women of childbearing potential and men with partners of childbearing potential: agreement to use effective contraception during and for the protocol-defined period after implantation.

Exclusion Criteria:

• Need for urgent surgical decompression or stabilization (e.g., acute/impending neurologic compromise, unstable pathologic fracture) that precludes percutaneous implantation at this time.

Uncorrected coagulopathy or ongoing antithrombotic therapy that cannot be safely managed periprocedurally.

Active systemic or local infection at/near the planned access route.

Diffuse marrow replacement or extensive cortical destruction where percutaneous implantation is unsafe or unlikely to achieve local control without stabilization, per MDT judgment.

Prior radiation or surgery to the index lesion that, in the investigator's opinion, makes additional seed implantation unsafe or non-beneficial; postoperative bed without a discrete target for seed placement.

Known hypersensitivity to materials/agents required for the procedure (e.g., contrast) not amenable to premedication or alternative imaging.

Uncontrolled medical conditions posing prohibitive procedural risk (e.g., severe cardiopulmonary disease, uncontrolled hypertension/arrhythmia).

Pregnant or breastfeeding.

Inability to lie still or contraindications to required imaging/sedation not correctable.

Concurrent participation in another interventional study that could confound efficacy/safety assessment at the treated site.

Any condition that, in the investigator's judgment, would interfere with protocol adherence, safety monitoring, or outcome assessment.

De-identified participant-level data including baseline demographics and clinical characteristics; lesion site and imaging features; planning inputs (e.g., TPS parameters, seed model), dosimetry metrics (D90, V100, CI, HI), biomechanical model outputs (stress/strain maps summarized, fracture/migration risk scores), procedure details (needle count/paths, procedure time, intraoperative deviations), safety data (AEs/SAEs per CTCAE v5.0), analgesic use, pain scores (BPI/VAS), functional status (ECOG/TESS), quality of life (EORTC QLQ-C30), and imaging-derived efficacy endpoints (local control/progression dates and assessments). Dates will be shifted or bucketed to protect privacy; direct identifiers will be removed.

IPD and supporting documents will be available starting 12 months after the primary completion date or upon publication of the primary results, whichever occurs first, and will remain available for 5 years thereafter (extensions considered upon justified request).

Qualified researchers with a methodologically sound proposal may request access. Required materials include a brief proposal and analysis plan, IRB/ethics approval or exemption (as applicable), and a signed Data Use Agreement. Requests are reviewed by the study's Data Access Committee within ~30 calendar days. Approved users will access a de-identified, analysis-ready dataset via a controlled-access secure environment; raw downloads are restricted. Available materials include: de-identified participant-level dataset, data dictionary/codebook, de-identification schema, redacted protocol, statistical analysis plan, blank CRFs, and analytic code used for primary endpoints (R/Python). Submit requests via the study data portal or email the IPD contact.

Masking: None (Open Label). Due to the procedural nature of seed implantation, participants and treating clinicians cannot be blinded. To reduce bias, key outcomes are assessed under blinded procedures: (1) centralized, de-identified imaging and dosimetry review by independent radiologists/physicists blinded to allocation; (2) pain and QoL collected via standardized instruments by staff not involved in planning, with pre-specified analgesic adjustment rules; (3) an independent endpoint adjudication committee and the statistical analysis team remain blinded to group assignment, using Arm A/B codes. Allocation is concealed until plan finalization; intraoperative teams are unblinded. Participants are instructed not to disclose perceived assignment to assessors.