KEY POINTS
- This single-centre retrospective study included 8 patients with radiation-induced secondary sarcoma treated with definitive carbon-ion RT between 2018 and 2024. Median age was 72 years, median tumor diameter 7.7 cm, and seven tumors were considered unresectable; five were pelvic, two head and neck and one CNS.
- These were true long-latency second malignancies. Median prior RT dose was 44 Gy, and the median interval between initial irradiation and CIRT was 34.2 years; 7/8 patients had received their original RT more than 10 years earlier.
- Seven patients received 70.4 Gy(RBE) in 16 fractions, while one received 55.2 Gy(RBE) in 12 fractions because of cumulative skin and esophageal dose. Treatment was delivered four times weekly using fast-raster scanning carbon ions.
- At a median follow-up of 18.5 months, two patients developed local recurrence and three developed distant metastases. One-year local control was 71.4%, progression-free survival 37.5%, and overall survival 75.0%; median PFS and OS were 9.0 and 30.2 months, respectively.
- Severe toxicity occurred in 2/8 patients (25%). One pelvic case developed grade 3 pain, lymphedema and sacral fracture together with grade 2 cauda-equina myelitis; another head-and-neck patient developed a grade 3 cerebrospinal-fluid leak with meningitis four months after treatment.
- No grade ≥4 toxicity was observed during the available follow-up. However, median follow-up was less than two years, which is particularly important when evaluating late effects after high-dose re-irradiation.
- Initial RT dose distributions were unavailable. For cumulative OAR calculations, the investigators assumed approximately 50% recovery of prior normal-tissue dose after >6 years, adding substantial uncertainty to cumulative dose estimation and limiting transferability of the reported constraints.
CLINICAL TAKEAWAY
For radiation-induced sarcoma arising inside a previously irradiated field, treatment options can become extremely limited when surgery is impossible. CIRT produced meaningful short-term local control in this tiny series, but eight heterogeneous patients, uncertain historical dosimetry and 25% grade 3 toxicity make this an early feasibility signal rather than evidence for an established re-irradiation strategy.