Hepatobiliary
One trial: RTOG 0848 asks whether modern adjuvant chemoRT still earns a role after gemcitabine in resected pancreatic cancer.
NRG Oncology/RTOG 0848
ForResected pancreatic head adenocarcinoma, progression-free after adjuvant chemo
HR 0.96
90% CI 0.79-1.18, 1-sided P=.38; primary EP not met
TL;DROS primary EP not met (HR 0.96, 90% CI 0.79-1.18) adding adjuvant CXRT to gemcitabine; node-negative subgroup OS benefit on interaction (P=.0063).
The RT read is the node-negative interaction: CXRT improved OS (P=.0063) and DFS (P=.014) in node-negative pts, though the subgroup HR/median isn't in the source text. Unlike ESPAC-1's older 5-FU CXRT, modern CXRT showed no OS harm. Moves selective adjuvant CXRT for node-negative resected head tumors, pending FOLFIRINOX-era confirmation.
This reopens selective adjuvant CXRT in node-negative resected head tumors (OS interaction P=.0063), where modern technique avoided the OS harm ESPAC-1 saw. Cost is doubled grade 3 toxicity (38% vs 19%); dose and target volume aren't in the source, and node-positive pts got no benefit.
Adjuvant gemcitabine-based chemo alone stays the reference: adding CXRT didn't improve OS (HR 0.96) or DFS (HR 0.82, P=.089) overall and raised grade 3 toxicity to 38% vs 19%. The node-negative interaction (P=.0063) would need retesting against FOLFIRINOX before altering adjuvant sequencing.
8 details 3 trials watching
Multicenter phase IIR/III, two-step RCT. Step 1: gemcitabine vs gemcitabine + erlotinib. Step 2 (reported here): random assignment to a sixth chemo cycle ยฑ CXRT after 5 progression-free cycles. N=354 (180 CXRT, 174 chemo).
Curative-intent resected pancreatic head adenocarcinoma, progression-free on adjuvant chemo. Median age 63, 55% male, 56% PS 1.
Fluoropyrimidine-sensitized CXRT added to the sixth cycle. Dose, fractionation, and target volume not stated in the source excerpt.
Primary: overall survival. Secondary: DFS. Powered for HR 0.76 (316 events, 80% power, 1-sided ฮฑ=.05).
Primary OS not met (HR 0.96). The one positive signal was the treatment-by-nodal-status interaction (OS P=.0063, DFS P=.014) favoring CXRT in node-negative pts; subgroup effect size not in source.
| Endpoint | Chemo + CXRT | Chemo alone |
|---|---|---|
| Median OS | 2.3 yr (2.0-2.6) | 2.6 yr (2.1-3.1) |
| 5-yr OS | 27.9% (22.2-33.6) | 23.1% (17.7-28.6) |
| OS HR (90% CI) | 0.96 (0.79-1.18), P=.38 | ref |
No increase in grade 4/5 toxicity. Grade 3 toxicity higher with CXRT: 38% vs 19% (P<.001).
ESPAC-1's older 5-FU-based CXRT suggested inferior outcomes; here modern CXRT shows no OS harm but no all-comer benefit. Backbone predates FOLFIRINOX-type adjuvant chemo, named in the source as current SOC.
Node-negative benefit rests on a subgroup interaction, not the prespecified primary. Trial underwent multiple mid-course modifications (LAP07, accrual, slow event rate), and no local-control or pattern-of-failure endpoint is reported despite RT's locoregional rationale.
Clean negative primary OS (HR 0.96) reinforces no routine adjuvant CXRT; node-negative benefit is a subgroup interaction needing confirmation; gemcitabine backbone predates FOLFIRINOX.
In resected node-negative pancreatic head adenocarcinoma completing adjuvant gemcitabine, this offers a subgroup signal favoring fluoropyrimidine-sensitized CXRT (OS interaction P=.0063); it does not extend to node-positive pts, where CXRT showed no OS or DFS benefit.
- Does node-negative CXRT benefit hold with mFOLFIRINOX backbone? n=770 ยท primary completion 2028-06 ยท phase 3 adjuvant CXRT vs chemo, resected high-risk
- Neoadjuvant vs adjuvant CXRT sequencing in node-negative resected pancreas recruiting Preoperative Stereotactic Body Radiation Therapy in Patients With Resectable Pancreatic Cancer Phase 2n=25 ยท primary completion 2026-09 ยท preop SBRT then resection in resectable pancreas
- Local-regional control and pattern of failure with modern CXRT n=30 ยท primary completion 2028-10 ยท freedom from local failure at 12mo, SBRT