This retrospective multi-center analysis aimed to assess the clinical response and stabilizing effects of palliative radiotherapy (RT) for spinal bone metastases (SBM) in head and neck cancer (HNC), and to establish potential predictive factors for stability and overall survival (OS). Patients included in this analysis were treated at the University Hospitals of Mainz, Freiburg, and Heidelberg between 2001 and 2019. Clinical information was taken from the medical records. The stability of affected vertebral bodies was assessed according to the validated spine instability neoplastic score (SINS) based on CT-imaging before RT, as well as 3 and 6 months after RT. OS was quantified as the time between the start of palliative RT and death from any cause or last follow-up. Potential predictive factors for stability and OS were analyzed using generalized estimating equations and Cox regression for time-varying covariates to take into account multiple observations per patient. The mean follow-up time of 66 included patients after the first palliative RT was 8.1 months (range 0.3–85.0 months). The majority of patients (70%; n = 46) had squamous cell carcinomas (SCC) originating from the pharynx, larynx and oral cavity, while most of the remaining patients (26%; n = 17) suffered from salivary glands tumors. A total of 95 target volumes including 178 SBM were evaluated that received a total of 81 irradiation series. In patients with more than one metastasis per irradiated region, only the most critical bone metastasis was analyzed according to the SINS system. Prior to RT, pain and neurologic deficits were present in 76% (n = 72) and 22% (n = 21) of irradiated lesions, respectively, and 68% of the irradiated lesions (n = 65) were assessed as unstable or potentially unstable prior to RT. SBM-related pain symptoms and neurologic deficits responded to RT in 63% and 47% of the treated lesions, respectively. Among patients still alive at 3 and 6 months after RT with potentially unstable or unstable SBM, a shift to a better stability class according to the SINS was observed in 20% and 33% of the irradiated SBM, respectively. Pathological fractures of SBM were frequently detected before the start of irradiation (43%; n = 41), but after RT, new fractures or increasing vertebral body sintering within the irradiated region occurred rarely (8%; n = 8). A pathological fracture before RT was negatively associated with stabilization 6 months after RT (OR 0.1, 95% CI 0.02–0.49, p = 0.004), while a Karnofsky performance score (KPS) ≥ 70% was associated positively with a stabilization effect through irradiation (OR 6.09, 95% CI 1.68–22.05, p = 0.006). Mean OS following first palliative RT was 10.7 months, and the KPS (≥70% vs. <70%) was shown to be a strong predictive factor for OS after RT (HR 0.197, 95% CI 0.11–0.35, p < 0.001). There was no significant difference in OS between patients with SCC and non-SCC. Palliative RT in symptomatic SBM of HNC provides sufficient symptom relief in the majority of patients, while only about one third of initially unstable SBM show re-stabilization after RT. Since patients in our multi-center cohort exhibited very limited OS, fractionation schemes should be determined depending on the patients’ performance status.