SBRT added to chemotherapy cut early local recurrence in pancreatic cancer

Design

Phase 2, randomised (2:1), open-label, multicentre

Population

74 adults with high-risk localised pancreatic adenocarcinoma, Australia and New Zealand

Primary outcome

12-month locoregional recurrence

Effect

20.2% vs 52.0%; subdistribution HR 0.44 (95% CI 0.22-0.88)

MASTERPLAN, an open-label phase 2 trial at 14 hospitals in Australia and New Zealand, randomised 74 adults with high-risk localised pancreatic ductal adenocarcinoma, 2:1, to chemotherapy (mFOLFIRINOX or gemcitabine plus nab-paclitaxel) with or without stereotactic body radiotherapy (SBRT, 40 Gy in five fractions). High risk included tumours over 4 cm, node-positive, borderline resectable or locally advanced disease. Published in The Lancet Oncology on 23 October 2026.

Twelve-month locoregional recurrence was 20.2% with SBRT and 52.0% with chemotherapy alone (subdistribution HR 0.44, 95% CI 0.22 to 0.88). Grade 3 or worse adverse events were 70% versus 64%, not significantly different. Two serious events were attributed to SBRT: one small bowel obstruction and one pancreatitis.

Local recurrence is a common way pancreatic cancer fails after modern chemotherapy, and conventional chemoradiation trials have been disappointing. A short SBRT course is a practical way to add local control.

The trial closed early for slow accrual, recruited 74 of a planned 120, and did not report survival as its primary endpoint. A larger phase 3 trial is needed before this becomes routine.

• Discuss high-risk localised pancreatic cancer at a multidisciplinary meeting that includes radiation oncology
• Consider SBRT within a trial or protocol for borderline resectable or locally advanced disease
• Use local control as the endpoint this evidence supports; survival benefit is not yet shown
• Watch for bowel obstruction and pancreatitis after SBRT

Why it matters

Short-course SBRT may give back local control that conventional chemoradiation has failed to deliver.

Don't overread it

Phase 2, 74 patients, closed early, and no overall survival result as a primary endpoint.

The statistics, in plain English

A subdistribution hazard ratio of 0.44 means local recurrence occurred at less than half the rate with SBRT, accounting for deaths as a competing event. With only 74 patients the confidence interval is wide (0.22 to 0.88), so the true effect could be much smaller.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.