KEY POINTS
- This retrospective multicenter study initiated by the DEGRO working group for radiosurgery and stereotactic radiotherapy analyzed 34 patients with 54 bone metastases from head and neck squamous cell carcinoma, treated at 11 European institutions between 2015 and 2024. Median age was 63 years, and the most common primary sites were the oropharynx (35.3%), nasopharynx (23.5%), and oral cavity (14.7%). Most patients had limited metastatic burden, with 91.2% having five or fewer metastases at the time of SBRT.
- Bone metastases predominantly involved the spine (34/54, 63.0%), followed by the pelvis (18.5%) and ribs (9.3%). Most lesions were osteolytic (68.5%), and only 13/54 (24.1%) were documented as symptomatic before treatment. All known metastatic sites were treated in 32/34 patients (94.1%). Systemic therapy was administered within four weeks of SBRT in 18 patients (52.9%), including immune checkpoint inhibitors in ten.
- Most lesions received five-fraction SBRT, with 40 Gy in five fractions (25.9%) and 35 Gy in five fractions (16.7%) representing the most common prescriptions. Other schedules included 30 Gy in three fractions, 30 Gy in five fractions, and 48.5 Gy in ten fractions. Median prescription BED10 was 60.0 Gy (IQR 56.0–72.0 Gy), with median GTV and PTV volumes of 10.9 and 28.6 mL, respectively.
- Local failure was relatively uncommon after SBRT. Seven local failures occurred among 54 treated lesions, yielding a two-year cumulative incidence of local failure of 14.1% (95% CI 3.0–27.8%), with death treated as a competing event. An analysis restricted to the first treated lesion per patient produced a similar two-year incidence of 12.7%. However, only seven lesions remained at risk at 24 months, making the long-term estimate imprecise.
- Systemic disease progression remained frequent despite favorable treated-site control. Median overall survival was 18.0 months, and median progression-free survival was only 3.0 months. Two-year OS was 41.2% (95% CI 24.3–57.3%), while two-year PFS was 8.8% (95% CI 2.3–21.1%). During follow-up, 26 deaths and 28 disease progression events were recorded among the 34 patients, highlighting the importance of systemic disease biology and treatment.
- Patients receiving SBRT for a single bone metastasis had more favorable survival than those treated for multiple bone lesions. Two-year OS was 50.0% versus 21.2% (p=0.027), while treatment of multiple bone metastases was associated with worse OS in univariable analysis (HR 2.58, 95% CI 1.08–6.15; p=0.033). Karnofsky performance status ≤70% was also associated with inferior survival (HR 3.70, 95% CI 1.31–10.46; p=0.013). These exploratory associations were not corrected for multiple testing.
- Severe toxicity and post-SBRT fractures were infrequently documented. One patient (2.9%) developed grade 3 acute pain, while no grade 4–5 adverse events were reported. A single vertebral fracture occurred 4.6 months after treatment with 24 Gy in two fractions, corresponding to a two-year fracture incidence of 1.9%. The absence of standardized pain-response assessment, central imaging review, and a non-SBRT comparator limits conclusions about clinical benefit and late safety.
CLINICAL TAKEAWAY
SBRT for bone metastases from head and neck squamous cell carcinoma was associated with relatively low local failure and a low observed incidence of severe toxicity in this selected multicenter cohort. However, short progression-free survival and heterogeneous systemic treatment indicate that local control alone does not reliably translate into durable disease control, and prospective studies are needed to define optimal patient selection and clinical benefit.