Head & Neck
First ESTRO post-op CTV consensus for oral cavity SCC, volumes and EQD2 by risk tier, no outcome data behind it.
ESTRO OCSCC Post-op CTV Delineation Guidelines
ForResected oral cavity SCC proceeding to post-operative radiotherapy
TL;DRFirst ESTRO guideline for post-op CTV delineation in oral cavity SCC: GTV-P pre-op + 10 mm composited with surgical defect/flap + 5 mm.
The margin recipe is asymmetric and that is the operative detail: 10 mm around the re-created pre-op GTV-P but only 5 mm around the surgical defect or flap, composited rather than either alone. Fig 3.1/3.2 shows why, the re-created GTV-P extended superiorly beyond the defect into infratemporal fossa, a geographical miss if you contour the defect alone.
In resected OCSCC going to PORT, this supports re-creating the pre-op GTV-P from diagnostic MRI alongside the defect or flap rather than contouring the operative bed alone; it does not extend to R2 resections or other head and neck subsites.
The margin recipe is asymmetric: 10 mm around the re-created pre-op GTV-P but 5 mm around the surgical defect or flap, composited rather than either alone. Nodal margin is pENE-gated at 5 mm versus 10 mm on GTV-N pre-op. Fig 3.1/3.2 shows the pre-op GTV extending past the defect into infratemporal fossa, a geographical miss if you contour the bed alone.
11 details 5 trials watching
ESTRO-convened multi-disciplinary expert group developing delineation guidelines through discussion and review of current evidence and international practice. Drafts were reviewed by HNSCC experts from countries outside the authorship (Japan, Hong Kong, Australia, Brazil, Mexico, Canada, Denmark, France, Spain, Poland, Ireland, UK) and modified on their feedback. No efficacy endpoint, no patient cohort.
Patients with oral cavity squamous cell carcinoma requiring post-operative radiotherapy, regardless of margin status and other histological risk factors. R2 resection (macroscopic residual disease) is explicitly out of scope. Companion background manuscript from the same group covers indications for PORT.
Planning CT 2.0 mm slices (range 1-3 mm), skull base to below sterno-clavicular joint, IV contrast mandatory, rigid co-registration with pre-op contrast-enhanced CT and/or MRI matched to C1-C3 vertebral bodies or nearby bone, not to soft tissue. CTV-P is the composite of GTV-P pre-op + 10 mm and surgical defect/flap + 5 mm, edited for bone, fascia, air, teeth and any intra-oral prosthesis. Nodal margin is 5 mm on GTV-N pre-op without pENE, 10 mm with pENE.
None. The stated aim is consistency of delineation to enable multi-institutional audit, clinical trials and RTQA. Authors position prospective audits of practice and outcomes as the route to establishing these volumes as standard of care.
Extends the 2018 international CTV-P consensus for definitive HNSCC RT, whose 5+5 mm geometric expansion supplies the 10 mm used here around GTV-P pre-op. Cites the DAHANCA finding that geometric expansion is more conformal than anatomical margins, a post-hoc De-ESCALaTE analysis correlating the anatomical-to-geometric protocol change with lower late dysphagia, and non-randomised Dutch series where reducing the high-risk margin 10 mm to 6 mm cut salivary and constrictor dose. GORTEC's 2020 flap delineation guidance is named as an adjunct.
The 5 mm and 10 mm margins are imported from definitive-setting geometry and one surgical pathology series (>95 % of microscopic infiltration within 5 mm of GTV-P edge), not from post-operative recurrence mapping. The dose to dissected but uninvolved levels rests on a 1993 MD Anderson observation never tested prospectively, and is left to clinician discretion (EQD2 50-60 Gy). The whole method assumes accurate pre-op to planning CT co-registration, and the fallback where it fails is to treat the entire involved level, a larger volume.
The novel move is refusing to pick between the two available surrogates for a resected tumour: re-created pre-op GTV and operative bed are contoured independently and unioned, because each fails in a different direction. Fig 3.1 shows a GTV-P pre-op extending superiorly past the defect toward the infratemporal fossa, and Fig 8.1 a pectoralis major pedicled flap whose composite volume extends outside the oral cavity and is trimmed back. What is left unsettled is dose de-escalation to central flap tissue, where the guideline offers a flap avoidance structure for standardisation while stating there is a lack of data and consequently a lack of consensus.
| Volume | Indication | EQD2 dose |
|---|---|---|
| CTV-P post-op | PTV associated with post-op primary CTV | 60 Gy |
| CTV-P high-risk | Positive (<1 mm) margin, whole CTV-P or localised strip | over 60 Gy, e.g. 64-66 Gy |
| CTV-N1 | Involved nodal levels | 60 Gy |
| CTV-N2 | Undissected at-risk levels | 50 Gy |
| CTV-N2, dissected at-risk levels | Optimal dose unknown | 50 Gy to 60 Gy, clinician discretion |
| CTV-N high-risk | Pathological extranodal extension | over 60 Gy, e.g. 64-66 Gy |
ESTRO expert guideline, no efficacy endpoint. Fills a documented gap (no prior post-op HNSCC CTV consensus); authors themselves position prospective audit as the validation step.
- Optimal dose to dissected but uninvolved nodal levels recruiting Preservation of Swallowing in Respected Oral Cavity Squamous Cell Carcinoma: Examining Radiation Volume Effects (PRESERVE): A Randomized Trial Phase 2n=90 · primary completion 2026-09 · OCSCC RCT: omit RT to dissected pN0 hemi-neckn=50 · primary completion 2029-01 · compartment-based post-op volume reduction in HNSCCrecruiting De-Intensification of Postoperative Radiotherapy in Patients With Squamous Cell Carcinoma of the Head and Neck Phase 2/3n=508 · primary completion 2031-12 · omits elective neck RT after neck dissection
- Whether flap dose de-escalation is safe not yet Lipiodal Injection Technique for Free Flap Sparing Adjuvant RT Planning in HNC Phase Early 1n=10 · primary completion 2026-03 · lipiodol marks flap borders for sparing plansn=348 · primary completion 2031-03 · phase 3 RCT: flap-sparing vs standard post-op RT
- Whether 10 mm post-op primary margin can be reduced as in definitive setting