KEY POINTS
- This retrospective single-institution study evaluated 36 patients with recurrent or metastatic head and neck squamous cell carcinoma treated with CyberKnife SBRT reirradiation at Boston Medical Center between 2009 and 2022, comprising 38 treatment plans. Median age was 64.5 years, 83.3% had stage III–IV disease, and the median gross tumor volume was 37.0 cm³.
- All patients had previously received radiotherapy, with a median prior dose of 69.96 Gy and a median interval of 745 days before CyberKnife treatment. Treatment used five fractions, with prescriptions ranging from 30 to 40 Gy, most commonly 30 Gy in five fractions, delivered over a median of nine days, generally on an every-other-day schedule.
- Treatment planning used noncoplanar robotic beam delivery, a 5-mm GTV-to-PTV expansion, and retrospective review of cumulative OAR doses. Across evaluable plans, median PTV coverage was 95.9% and GTV coverage 99.9%, with a median conformity index of 1.12 and prescription to a median 72.5% isodose line. Dose calculations used a ray-tracing algorithm rather than Monte Carlo.
- Among 29 patients evaluable for local control, 13 local failures occurred over a median local-control follow-up of only six months. Estimated local control was 67.7% at six months, 55.1% at one year, and 34.5% at two years, with a median time to recurrence of 14.1 months; long-term estimates should be interpreted cautiously because of limited follow-up and small numbers remaining at risk.
- Among 34 patients evaluable for overall survival, median survival was 9.6 months, with estimated one-year OS of 43.5% and two-year OS of 34.2%. Median follow-up was 8.5 months, and 20 deaths were recorded, reflecting the poor prognosis of this heavily pretreated population.
- Smoking exposure independently predicted outcomes. A history of more than 30 pack-years was associated with markedly worse local control (adjusted HR 5.68, 95% CI 1.42–22.70; p=0.014), while former or never smokers had longer median survival than current smokers (22.7 versus 6.8 months; adjusted HR 0.26, 95% CI 0.08–0.82; p=0.022).
- Larger tumor volume was also associated with poorer outcomes in univariable analyses. Patients with GTV >37 cm³ had shorter median time to local recurrence (5.3 vs 19.8 months; p=0.044) and shorter median overall survival (8.0 vs 66.5 months; p=0.003), although the survival association was no longer statistically significant after adjustment (p=0.060) and the long survival estimate was based on very few patients at risk. Severe toxicity was not systematically collected, preventing reliable conclusions about reirradiation safety.
CLINICAL TAKEAWAY
CyberKnife SBRT remains a potential salvage reirradiation approach for selected patients with recurrent head and neck squamous cell carcinoma, particularly when focal disease can be treated within acceptable cumulative OAR exposure. Smoking history and tumor volume may inform prognosis and treatment selection, but the retrospective design, limited follow-up, and lack of systematic toxicity reporting prevent firm conclusions about comparative efficacy or safety.