The research, led by Tatsuki Ikeda of the Department of Pharmacy at Iwate Medical University Hospital and published in Supportive Care in Cancer, tackles a question that has lingered in oncology for years. Bone-modifying agents, which include bisphosphonates such as zoledronic acid and the RANKL inhibitor denosumab, are well established at reducing skeletal complications in patients whose solid tumors have seeded the skeleton. Randomized trials have repeatedly shown that these drugs delay first skeletal events compared with placebo. What has remained stubbornly unclear, however, is when therapy should begin after the diagnosis of bone metastasis itself, and whether the timing of that first dose changes the downstream risk.

Guidelines from major bodies such as the American Society of Clinical Oncology and the European Society for Medical Oncology recommend bone-modifying agents for patients with bone metastases from solid tumors, but they offer limited specificity about the ideal moment to start. In everyday clinical practice, initiation is shaped by a tangle of competing priorities: staging workups, dental clearance to reduce the risk of osteonecrosis of the jaw, chemotherapy planning, and the simple reality that some patients deteriorate quickly from their systemic disease. The result is wide variation in how quickly patients reach their first infusion or injection, and little evidence to say whether that variation matters.

To answer the question rigorously, the Iwate team turned to a single-institution cohort of patients with solid tumors who received bone-modifying agent therapy for bone metastasis between 2010 and 2024. The central methodological challenge in any study of treatment timing is a statistical trap known as immortal time bias. If researchers simply compare patients who started treatment early with those who started late, counting follow-up time from the diagnosis of bone metastasis, the early starters are guaranteed a period of survival before their treatment begins, because a patient cannot receive a drug before they are alive and under care. That guaranteed event-free window artificially favors the early-treatment group and can manufacture an apparent benefit out of nothing.

The investigators neutralized this bias with a landmark analysis, a technique that fixes a common reference point in time and compares only patients who are still alive, under follow-up, and free of skeletal events at that point. They chose a landmark of six months after the bone metastasis diagnosis. Patients who had already begun bone-modifying agent therapy by the landmark were classified as the Early group, while those who had not yet started were assigned to the Delayed group. Everyone included in the comparison had, by construction, survived at least six months without a skeletal event, ensuring that the groups started on equal statistical footing.

Equal footing, however, requires more than a shared starting line. Patients who begin bone-targeted therapy quickly may differ systematically from those who wait, in tumor type, disease burden, performance status, or the aggressiveness of their cancer. To address this, the researchers performed one-to-one propensity score matching, a method that estimates each patient’s probability of belonging to the early-initiation group based on their baseline characteristics and then pairs similar patients across the two groups. After matching, 134 patients were included in each arm, producing balanced cohorts in which the primary difference was the timing of therapy.

The results were striking. During follow-up after the six-month landmark, skeletal-related events occurred in 86 patients across the matched cohorts. When the researchers modeled the time from the landmark to the first skeletal event, the Delayed group showed a substantially higher hazard: the hazard ratio was 1.82, with a 95 percent confidence interval of 1.16 to 2.84 and a P value below 0.01. In practical terms, patients whose bone-modifying agent therapy had not started by six months after diagnosis faced roughly an 80 percent higher rate of subsequent skeletal complications than their matched counterparts who had already begun treatment, despite being comparable at baseline.

The finding aligns with a growing body of work on the biology of metastatic bone disease. Tumor cells in the skeleton disrupt the normal coupling between bone-resorbing osteoclasts and bone-forming osteoblasts, tipping the balance toward destruction. Bone-modifying agents suppress osteoclast activity, and the earlier that suppression begins, the less time the tumor has to erode cortical bone and compromise the structural integrity of the vertebrae and long bones. Once a vertebral body has collapsed or a fracture has occurred, the damage is largely irreversible, and the clinical focus shifts from prevention to palliation. A therapy that arrives after the skeleton has already failed cannot undo that failure, which may partly explain why delay in this analysis translated into higher event rates even among patients who were event-free at the landmark.

The study’s authors, who also include Satoru Nihei, Kazuki Saito, Junya Sato of Shonan University of Medical Sciences, and senior author Kenzo Kudo, are careful about the limits of the design. As a retrospective, single-institution analysis, the work cannot prove causation in the way a randomized trial would. Unmeasured confounding remains possible: clinicians may have delayed therapy in patients they judged too fragile, too close to death from systemic progression, or already harboring subclinical skeletal damage that later declared itself as an event. The propensity score matching balanced the characteristics that were recorded, but it cannot balance what was never written down. The authors also note that the requirement for patients to remain under follow-up and event-free at six months means the analysis speaks specifically to the subset of patients who live long enough for the timing question to matter, a clinically meaningful group but not the entire metastatic population.

Even with those caveats, the message for clinical practice is difficult to ignore. Prolonged delay in initiating bone-modifying agents after a diagnosis of bone metastasis appears to be associated with a higher subsequent risk of skeletal-related events, and the effect size is large enough to be clinically relevant, not merely statistically detectable. The findings add weight to the argument that once bone metastasis is confirmed, clinicians should move promptly toward initiating bone protection, coordinating the necessary dental evaluation and treatment planning so that protective therapy is not pushed further down the calendar. For patients living with metastatic breast, prostate, or lung cancer, where skeletal involvement is common and often prolonged, the difference between starting bone-targeted therapy within months and waiting half a year or more may shape whether they spend that time walking, or recovering from a fracture. The study does not settle the question of the perfect starting moment, but it draws a clear line: the longer the wait, the greater the risk, and the skeleton keeps score.

Subject of Research: The effect of timing of bone-modifying agent initiation on skeletal-related event risk in patients with bone metastases from solid tumors

Article Title: Timing of bone-modifying agent initiation and skeletal-related event risk in patients with bone metastases from solid tumors: a retrospective landmark and propensity score–matched analysis

Article References: Ikeda, T., Nihei, S., Saito, K., Sato, J., & Kudo, K. (2026). Timing of bone-modifying agent initiation and skeletal-related event risk in patients with bone metastases from solid tumors: a retrospective landmark and propensity score–matched analysis. Supportive Care in Cancer, 34(10), Article 925. https://doi.org/10.1007/s00520-026-11136-5

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11136-5

Keywords: bone metastases, bone-modifying agents, skeletal-related events, bisphosphonates, denosumab, landmark analysis, propensity score matching, immortal time bias, supportive care in cancer, oncology, zoledronic acid, fracture risk