Carbon ion therapy for pancreatic cancer with risk-adapted dose escalation: initial clinical experience

Background and purpose: This study reports the initial clinical outcomes of carbon ion radiotherapy (CIRT) using a risk-adapted dose-escalation protocol for pancreatic cancer at Taiwan's first heavy-ion therapy center. Materials and methods: Eighty-four patients, primarily with locally advanced or oligometastatic disease, received definitive CIRT in 12 fractions. Doses were escalated to 60–66 Gy(RBE) for PET-avid lesions and tumor-vessel interfaces, 55.2 Gy(RBE) for gross tumor volume, and 43.2 Gy(RBE) for subclinical regions. All patients underwent neoadjuvant chemotherapy (NAC) for a median of 4.27 months. Kaplan–Meier and Cox regression were used to assess outcomes and prognostic factors. Results: As of March 25, 2025, with a median follow-up of 12.7 months, 1-year rates for local tumor control (LTC), overall survival (OS), distant metastasis-free survival (DMFS), and progression-free survival (PFS) were 94.7 %, 75.3 %, 44.5 %, and 42.9 %, respectively. Among non-metastatic patients, median PFS was 13.6 months. Poorer OS, DMFS, and PFS were linked to metastatic disease, prolonged NAC (>9 months), and elevated CA19-9 (≥120 U/mL). Improved survival and local control were associated with CA19-9 kinetics, specifically a prolonged decline or rebound that stayed below baseline during NAC. The majority of acute events were restricted to Grade 1 toxicities, and there were very few Grade ≥3 toxicities. Conclusion: In pancreatic cancer, including anatomically difficult cases, CIRT with risk-adapted dose escalation produced excellent local control and low toxicity. Results could be further enhanced by CA19-9 dynamics and the best time to start CIRT. Integration with systemic therapies should be investigated in future trials.