This study will evaluate whether starting denosumab early, together with a locked uniform PD-1-based systemic therapy, can reduce skeletal-related events and delay worsening bone pain compared with a delayed/rescue bone-modifying strategy in patients with hepatocellular carcinoma and bone metastases. Participants will be randomly assigned in a 1:1 ratio to receive either early denosumab plus the same PD-1-based systemic therapy or the same systemic therapy with no routine prophylactic bone-modifying agent at baseline and rescue treatment only when predefined triggers occur. The primary outcome is skeletal-related event-free survival. Secondary outcomes include time to first skeletal-related event, pain outcomes, quality of life, intrahepatic antitumor activity at Week 12, progression-free survival, overall survival, and safety.

This is a single-center, prospective, randomized, open-label, blinded-endpoint (PROBE) exploratory phase 2 strategy study in patients with hepatocellular carcinoma (HCC) and radiologically or pathologically confirmed bone metastases. A total of 50 participants will be randomized in a 1:1 ratio to either early denosumab plus a locked uniform PD-1-based systemic therapy or the same PD-1-based systemic therapy with a delayed/rescue bone-modifying strategy.

The study is designed to evaluate whether an early denosumab strategy can improve bone-related clinical outcomes in HCC with bone metastases. The key clinical question is not whether denosumab improves short-term intrahepatic radiographic response, but whether early RANKL inhibition can reduce or delay skeletal-related events, delay pain worsening, and improve quality of life when added to a uniform PD-1-based systemic therapy backbone.

The PD-1-based systemic therapy backbone must be locked before enrollment of the first participant and remain consistent throughout the study. Both study arms receive the same locked systemic regimen, standard supportive care, analgesic therapy, and clinically necessary local treatment for bone metastases. The only protocol-defined treatment difference between the two arms is whether denosumab is started at baseline.

Participants randomized to the experimental arm will receive denosumab 120 mg subcutaneously every 4 weeks, with correction of hypocalcemia before treatment initiation and calcium/vitamin D supplementation as indicated. Participants randomized to the control arm will not routinely receive prophylactic bone-modifying agents at baseline. Rescue bone-modifying therapy may be initiated when predefined triggers occur, such as impending or actual pathologic fracture, spinal cord compression or worsening spinal instability, worsening bone pain despite standard management, rapidly progressive bone destruction judged to substantially increase skeletal-related event risk, or malignant hypercalcemia.

The primary endpoint is skeletal-related event-free survival (SREFS), defined as the time from randomization to the first on-study skeletal-related event or death from any cause. Skeletal-related events include pathologic fracture, spinal cord compression, radiation therapy to a bone lesion, and orthopedic surgery to a bone lesion. Malignant hypercalcemia will be recorded separately as an extended bone event. A fixed-time key analysis is planned at Month 6, with continued follow-up through Month 24.

Secondary endpoints include time to first skeletal-related event; cumulative incidence of skeletal-related events at 6 and 12 months; bone pain outcomes measured by the Brief Pain Inventory; time to pain progression; time to analgesic escalation; quality-of-life outcomes using EORTC QLQ-C30 with optional QLQ-BM22 or QLQ-HCC18; intrahepatic objective response rate and intrahepatic disease control rate at Week 12; intrahepatic progression-free survival, overall progression-free survival, and overall survival; incidence and severity of adverse events; and the frequency and timing of rescue bone-modifying treatment in the control arm.

Baseline imaging must be completed within 14 days before randomization. Tumor imaging is recommended every 8 weeks during the first 6 months and every 12 weeks thereafter until disease progression, death, loss to follow-up, or study end. Pain assessments are performed at baseline, Week 4, Week 8, Month 3, and every 4 weeks through Month 12, then every 8 to 12 weeks during follow-up. The study includes independent blinded endpoint adjudication for skeletal-related events and a Data and Safety Monitoring Board for ongoing safety oversight.

Inclusion Criteria:

• Age 18 to 75 years
• Hepatocellular carcinoma confirmed by imaging and/or histology according to current institutional diagnostic standards
• Bone metastasis confirmed by CT, MRI, bone scintigraphy, PET-CT, or pathology
• ECOG performance status 0 to 1
• Child-Pugh class A or stable Child-Pugh B7 judged suitable for systemic treatment
• At least 1 measurable intrahepatic lesion at baseline and planned initiation of a PD-1 inhibitor-based systemic therapy
• Estimated life expectancy of at least 3 months
• Adequate organ function according to protocol-defined laboratory criteria
• Corrected serum calcium within the normal range, or corrected to protocol-defined range after calcium/vitamin D supplementation
• Completed baseline oral/dental risk assessment and willingness to avoid nonessential invasive dental procedures during the study
• Written informed consent and willingness to comply with study visits, questionnaires, sample collection, and follow-up

Exclusion Criteria:

• Prior treatment with PD-1, PD-L1, CTLA-4, or other immune checkpoint inhibitors
• Receipt of denosumab or intravenous/oral bisphosphonates for tumor-related bone disease within 6 months before randomization
• Uncorrected hypocalcemia or severe vitamin D deficiency that cannot be corrected promptly
• Current or prior osteonecrosis of the jaw, jaw osteomyelitis, or high-risk dental condition requiring near-term extraction, implantation, or other invasive dental procedures that cannot be deferred
• Active autoimmune disease or need for systemic immunosuppressive treatment, except for low-risk conditions judged acceptable by the investigator
• Life-threatening bone complication within 4 weeks before randomization that is not stabilized, including unresolved spinal cord compression or clinically significant spinal instability
• Active infection, including uncontrolled bacterial or fungal infection, active tuberculosis, or other condition judged unsafe for systemic treatment
• Symptomatic or uncontrolled central nervous system metastases
• Pregnancy or breastfeeding, or refusal to use effective contraception when applicable
• Known severe hypersensitivity to denosumab or any study-related treatment component
• Another active malignancy, poor compliance, or any condition judged by the investigator to make study participation unsuitable

This is an open-label study because denosumab administration and rescue strategy management cannot be practically masked. However, the primary bone-related endpoint, skeletal-related event adjudication, and key endpoint review will be performed by blinded independent outcome assessors.