Pancreatic SBRT achieves 82% one-year local control across 17 Italian centers

In 473 pancreatic cancer patients, SBRT achieved 82% one-year local control, with higher biologically effective dose associated with better response.

KEY POINTS

  • This retrospective multicenter study evaluated 473 patients with pancreatic adenocarcinoma treated at 17 Italian centers between January 2021 and December 2025. Patients were categorized as resectable (15.2%), borderline resectable (19.2%), unresectable locally advanced (50.3%), oligometastatic stage IV (6.6%), or locally recurrent (6.3%), with disease status missing in 2.3%. Median age was 69.6 years, and 78% received systemic therapy before SBRT, predominantly FOLFIRINOX or gemcitabine/nab-paclitaxel.
  • Most patients received five-fraction SBRT, with doses ranging from 24 to 67.5 Gy in three to six fractions. The most common prescription was 45 Gy in five fractions, corresponding to BED10 85.5 Gy. Conventional linac-based SBRT was used in 378 patients (79.9%), while 95 (20.1%) received MR-guided treatment. MR-guided workflows generally used 3–5 mm PTV margins, compared with 5–7 mm for conventional SBRT, and online adaptive replanning was available when clinically required.
  • Disease control was favorable within the limitations of the mixed clinical population. After median follow-up of 13.2 months, median overall survival was 18.4 months, median progression-free survival was 10.2 months, and 12-month local control was 82%. Estimated OS was 62% at one year and 28% at two years, while PFS was 41% and 19%, respectively. These outcomes combine patients treated for different disease stages and treatment intents rather than representing a uniform locally advanced pancreatic cancer population.
  • Radiological disease control was documented in most evaluable patients. Among 422 patients with RECIST 1.1 assessments, 21 (5.0%) achieved complete response, 117 (27.7%) partial response, 187 (44.3%) stable disease, and 97 (23.0%) progressive disease. The resulting disease control rate was 77.0% (325/422). Follow-up imaging protocols and timing varied across centers, and no central radiological review was performed.
  • Higher biologically effective dose was associated with better response and local control. Patients receiving BED10 ≥80 Gy had partial response rates of 38% versus 19% and progressive disease rates of 12% versus 23% compared with patients receiving lower BED10. Kaplan–Meier analyses also favored higher-dose treatment for local control, but treatment allocation, target anatomy, proximity to gastrointestinal structures, and systemic therapy were not randomized. The 80-Gy threshold therefore remains a study-defined dose category, not a validated individualized prescription rule.
  • Severe toxicity was uncommon in the available records. Grade ≥3 acute toxicity occurred in 17/455 evaluable patients (3.7%), while grade ≥3 late toxicity was documented in 12/405 (3.0%). No treatment-related duodenal perforation or gastrointestinal bleeding was reported. However, 18 patients lacked acute toxicity assessments and 68 lacked late toxicity assessments, while the relatively short follow-up limits confidence in the detection of delayed gastrointestinal complications.
  • MR-guided treatment showed favorable dosimetry but no demonstrated survival advantage. Duodenal V33Gy <0.5 cm³ was achieved in 91% of MR-guided versus 74% of conventional SBRT plans. Twelve-month local control was 86% versus 80%, respectively, without a statistically significant difference (p=0.09), and neither OS nor PFS differed significantly between modalities. MR-guided treatment was also associated with different planning margins and potentially different patient selection, preventing direct attribution of outcome differences to the technology.

CLINICAL TAKEAWAY

Contemporary pancreatic SBRT was associated with encouraging local control and low documented severe toxicity across a large Italian multicenter cohort. Higher BED10 correlated with better radiological response, while MR guidance facilitated smaller margins and greater compliance with selected gastrointestinal constraints. These observational findings support individualized ablative planning but cannot establish the superiority of higher doses or MR-guided treatment without prospective comparative validation.

SOURCE

Cancers

Browse more research Suggest a correction