The inclusion criteria were: (1). Patients must have pathologically proven malignancy and radiographic evidence of vertebral metastases, number ≤3; (2). The pain of vertebral metastases was not relieved after previous therapy and pain score was not less than 4, Which was measured using a 0 to10 numeric rating scale (NRS) (22); (3). Except for vertebral metastases, no other organs had metastases or metastatic lesions were controlled;(4). No dysfunction of important organs, including heart, lung, kidney, etc.; (5). Karnofsky performance status (KPS) ≥70, and expected survival≥3 months; (6). All patients were discussed by a combination of radiation oncologists, medical oncologists, spine surgeons, pain medicine specialists, interventional radiologists, psychiatrists, and palliative care professionals before deciding on a course of treatment; (7). An informed consent for ¹²⁵I seed implantation was signed by the patient or his legal guardian.

The exclusion criteria involved: (1). Patients with known Central Nervous System (CNS) metastases or a history of CNS metastases before treatment. For patients with clinically suspected CNS metastases, CT or MRI examination must be performed within 14 days before treatment to exclude CNS metastases; (2). Patients who had relapsed within 6 months after radiotherapy for vertebral metastases or had received radiotherapy within 6 months in adjacent vertebral sites; (3). Severe organ dysfunction;(4). Coagulation dysfunction, anticoagulant therapy should be stopped at least 5-7 days before implantation;(5). Poor general condition or cachexia;(6). No CT and other imaging data after ¹²⁵I seed implantation.

Preoperative plan was delineated by clinicians, radiation oncologists and physicians together. CT scan was performed within 1 week before the treatment with a slice thickness of 5mm. The patient was placed in a prone position, secured by a vacuum negative pressure pad, with the centerline of setup marked on the body surface. CT images were transmitted to computer-assisted treatment planning-system to evaluate the feasibility of treatment and to design preoperative planning. Brachytherapy treatment planning system (BTPS, Beijing University of Aeronautics and Astronautics and Beijing Astro Technology Co. Ltd) was used. The prescription dose of this study was 120 Gy.

The radiation oncologist delineated the target volume and organs at risk (spinal cord, great blood vessels and adjacent tissues), set the prescribed dose and particle activity, determined the distribution and depth of the insertion needle, calculated the number of ¹²⁵I seed and simulated the spatial distribution of particles. The 3D printing coplanar template (3DPCT) was made of corn resin and provided by Beijing Atomic Technology Co., Ltd, with the specifications of8 cm × 8 cm × 2 cm or 10 cm× 10 cm×2 cm. It was punched in accordance with 0.5cm spacing.

For ¹²⁵I seed implantation, a 64-row spiral CT scanner (Siemens, Germany) was used. The ¹²⁵I seed was provided by Beijing Atomic Technology Co., Ltd (China), which was 0.8 mm × 4.5 mm (diameter × length) with a radioactivity of 0.4-0.8mCi and radioactive half-life of 59.6d. Supporting device for 3D PCT was connected to the bed. The puncture trajectory was marked on the skin. After skin disinfection and local anesthesia with 2% lidocaine, the coplanar template was placed on the patient skin entry point, which was consistent with the preoperative plan. All 18-gauge needles were inserted step-by-step into the lesions through the holes on the coplanar template. The material of the inner needle and the outer needle of the puncture needle were SUS304 cold-pressed steel plate specified by JIS G4305. When acquiring the CT half-way through needle placement, the needles may cause artifacts. It can be dealt with by adjusting the window width and window position of CT scan image appropriately. When all needles were deemed in place,¹²⁵I seeds were implanted according to the preoperative plan. During the operation, radioactive particles should not be exposed in the air, to avoid causing unnecessary radiation to people around. After the operation, the instrument and the surrounding environment were detected by radiation monitors.

CT scan was performed 3 days after operation to reduce the error of tumor volume due to tissue edema. And images were transmitted to TPS for dose verification ( Figure 1 ). Dose parameters were calculated to evaluate the dose distribution, which include D90, D100, V90, V100, V150, V200, GTV volume, CI (23, 24) (Conformal index), EI (24) (External volume index) and HI (24) (Homogeneity Index).

CI is used to evaluate the degree of coverage to the target volume. CI =   V T , r e f V T   ×   V T , r e f V r e f .

EI is used to evaluate overdosage to the surrounding tissues. EI = V r e f   V T , r e f V T .

HI is used to evaluate dose homogeneity within the target volume. HI = V T , r e f   V T , 1.5 r e f V T , r e f .

V T , r e f is volume of target receiving a dose equal to or greater than the reference dose; V T is volume of target; V r e f is volume receiving a dose equal to or greater than the reference dose (treated volume); V T , 1.5 r e f is volume of target receiving 150% of the reference dose.

The primary outcome of our study was pain relief. NRS were used to evaluate the severity of patients’ pain, had been validated as an outcome measure (25–27). According to NRS for chronic pain, pain intensity at the treated vertebral level was evaluated and graded as follows:0, no pain; 1–3, mild pain; 4–6, moderate pain; and 7–10, severe pain. Patients completed NRS for the focal pain metastasis with the assistance of a trained visitor before surgery and 24 hours, 1 week, 4 weeks, 8 weeks, 12 weeks and 6 months after surgery.

Contrast-enhanced CT and MRI scans were performed 1 month after treatment and then every 3 months and compared with the post-procedural CT scans to identify local tumor progression. Local tumor progression was defined as an osteolytic defect of the tumor or growth of soft tissue components. Local control was defined as CR + PR + SD [LCR = (CR + PR + SD)/total], according to RECIST 1.1. Secondary outcomes were OS (time from the day of radioactive ¹²⁵I seed implantation to death from any cause), preoperative and postoperative dosimetry evaluation.

Treatment-related adverse events were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 3.0 and were coded and summarized according to the preferred terms in the Medical Dictionary for Regulatory Activities, version 15.0. American Spinal Injury Association (ASIA) International Standards for Classification of Spinal Cord Injury was used for neurological assessment (28).

SPSS 26.0 statistical software was used to analyze and compare all the data. The count data was analyzed by x² test and expressed by [n (%)]. The measurement data was analyzed by t-test and expressed by ( x ¯ ± s). When p< 0.05, the difference has statistically significance. A two-sided test with P< 0.05 was considered statistically significant.

58 patients (31males, 27females) were included in our analyses. The patients were aged from 28 to 82 years old, with an average age of (60 ± 13) years old. The primary tumor sites among the patients were lung cancer in26 (26/58, 44.9%), breast cancer in 15 (15/58, 25.9%), liver cancer in 7 (7/58, 12.1%), renal cancer in 4 (4/58, 6.9%), bladder cancer in 2(2/58, 3.4%), cervical cancer in 2(2/58, 3.4%), colon cancer in 1 (1/58, 1.7%), and esophageal cancer in 1 (1/58, 1.7%). A single lesion was treated in 45 (45/58, 77.6%) patients while two lesions were treated in 8 (8/58, 13.8%) patients and three lesions treated in 5 (5/58, 8.6%) patients, for a total of 76 lesions treated. The thoracic spine was the most common location for all vertebral metastatic lesions treated (36/76, 47.3%). There were 10 vertebral bodies with incomplete posterior margins (10/76, 13.2%). The patients’ characteristics were delineated in Table 1 .

The NRS score for worst pain was 6.1 ± 1.1 before ¹²⁵I seed implantation. The mean post-operative NRS scores decreased significantly at T_4w (3.5 ± 0.9, p<0.01), T_8w (2.1 ± 0.9, p<0.01), T_12w (1.5 ± 0.7, p< 0.01) and T_6m (1.2 ± 0.6, p< 0.01) respectively. There was no significant difference in scores among T₀, T_24h (P=0.10) and T_1w (P=0.09) ( Table 2 , Figure 2 ).

All patients were successfully performed implantation at the first time. Median number of ¹²⁵I seeds implanted was 55 (range, 10-96). The specific activity of seeds ranged from 0.5 to 0.8 mCi per seed, with a median of 0.7mCi/seed.

No patients were lost to follow-up. The patients were evaluated radiographically for all of the ¹²⁵I seed implantation procedures. The local control rates after 3, 6, 9 and 12 months were 100% (58/58), 93.1% (54/58), 87.9% (51/58), and 81% (47/58), respectively. The median overall survival time was 18.52months (95% CI, 16.24-20.8), and 1- and 2-year survival rates were 81% (47/58) and 34.5% (20/58), respectively.

There were 76 diseases in 58 patients. The dosimetric comparison before and after ¹²⁵I seed implantation is shown in Table 3 . There were no significant differences between preoperative and postoperative parameters, including D90,D100,V90,V100,V150,V200,GTV volume,CI,EI and HI, Which were compared by paired t-test (P > 0.05) ( Table 3 ).

All operations were completed successfully. Ten patients (17.2%, 10/58) appeared small volume subcutaneous hemorrhage, which may be associated with the injury of small subcutaneous vessels by puncture needle. All patients were relieved after pressing the puncture point to stop bleeding. Five patients (8.6%, 5/58) presented mild radiation dermatitis, but none developed radiation dermatitis above grade III. There were no deaths associated with ¹²⁵I seed implantation. There were no serious complications such as infection, radiation osteonecrosis, or radiation myelitis during the follow-up period. Follow-up imaging showed that no patient had radioactive particle displacement.