Head & Neck
2026-09-09
TORPEdO
ForLocally advanced oropharyngeal SCC, 96% p16+, cisplatin-eligible, no N3
18% vs 7%
adj OR 2.80 (97.5% CI 0.75–10.41), p=0.079, not met; UW-QoL p=0.56
TL;DR18% vs 7% 12-mo G-tube dependence/≥20% weight loss (p=0.079), UW-QoL 78.3 vs 77.1 (p=0.56): IMPT not superior to IMRT, locally advanced oropharyngeal SCC.
Lower OAR doses on IMPT did not become function: 12-mo UW-QoL 78.3 vs 77.1, and ≥20% weight loss ran higher on IMPT (20/110 vs 3/53), persisting at 24 mo in 12 of 95. Dutch NTCP-enriched pts also showed no clear gain (78.2 vs 75.8), undercutting model-based IMPT referral in locally advanced oropharyngeal SCC.
In cisplatin-eligible locally advanced oropharyngeal SCC, mostly p16-positive, planned for 70 Gy in 33 fractions with bilateral neck RT, this does not support IMPT over optimised IMRT for 12-mo swallowing, nutrition or QoL; it does not extend to N3 disease, induction chemo, or post-operative RT. The longer read
The RT read: IMPT against IMRT planned to identical 70 Gy/33 fx targets and OAR priorities gave no 12-mo functional gain (UW-QoL 78.3 vs 77.1), while ≥20% weight loss ran 20/110 vs 3/53. Dutch NTCP-enriched pts showed 78.2 vs 75.8, weakening model-based IMPT referral.
11 details 2 trials watching
Phase 3, multicentre, open-label RCT at 20 UK NHS hospitals, 2:1 by minimisation, N=205 (136 IMPT, 69 IMRT). IMPT delivered at two proton centres, IMRT at the local referring centre. Median follow-up 28.3 mo (IQR 26.5 to 39.3).
Newly diagnosed locally advanced oropharyngeal SCC suitable for concurrent CRT including bilateral neck treatment; 197/205 (96%) p16-positive, 99 (48%) T3 or T4. Excluded: N3, upfront neck dissection, induction chemo, prior H&N RT, or a feeding tube needed before treatment.
70 Gy to the therapeutic target and 56 Gy to lower-risk volumes in 33 daily fractions over 6.5 weeks, RBE 1.1 for IMPT, same OAR planning priorities in both arms. IMPT plans checked against 3-mm setup and 3.5% range uncertainty, with daily CBCT and a week-3 repeat planning CT in both arms.
Concurrent cisplatin 100 mg/m² every 3 weeks x2. 138 (69%) of 199 received both cisplatin cycles; 49 (25%) received cisplatin then carboplatin.
Co-primary at 12 mo: gastrostomy-tube dependence or severe (≥20%) weight loss, and UW-QoL physical composite score; either reaching significance counted as success (α=0.025 each). LRC and OS were exploratory.
Neither co-primary met. Composite events were driven by weight loss (20/110 vs 3/53); G-tube dependence was 2% in both arms.
| Endpoint | IMPT | IMRT | Effect |
|---|---|---|---|
| G-tube dependence or ≥20% weight loss, 12 mo (co-primary) | 21/119 (18%) | 4/59 (7%) | adj OR 2.80 (97.5% CI 0.75–10.41), p=0.079 |
| ≥20% weight loss, 12 mo | 20/110 (18%) | 3/53 (6%) | n/a |
| G-tube dependence, 12 mo | 2/119 (2%) | 1/59 (2%) | n/a |
| UW-QoL physical composite, 12 mo (co-primary) | 78.3 | 77.1 | diff 1.3 (97.5% CI –3.7 to 6.2), p=0.56 |
| UW-QoL physical composite, 3 mo | 70.8 | 66.8 | diff 4.0 (99% CI –1.4 to 9.5) |
| UW-QoL physical composite, 24 mo | 81.6 | 79.9 | diff 1.7 (99% CI –3.6 to 7.0) |
| 24-mo freedom from LRR | 94% (99% CI 86–98) | 97% (82–100) | HR 2.6 (99% CI 0.3–20.3), p=0.24 |
| 24-mo OS | 95% (86–98) | 95% (81–99) | HR 1.6 (99% CI 0.3–8.8), p=0.47 |
14 serious AEs in 12 pts, treatment-related in one IMPT vs four IMRT; no treatment-related deaths. Grade 3 weight loss persisted at 24 mo in 12 (13%) of 95 IMPT pts vs none on IMRT.
PARSPORT made parotid-sparing IMRT the photon benchmark. TORPEdO asks whether IMPT adds function over IMRT planned to the same OAR targets; at 12 mo it did not.
Feeding-tube policy was set per centre and prophylactic placement was lower on IMPT (26% vs 38%), confounding tube and weight endpoints. IMPT delivery was more disrupted (replanning 61% vs 30%, cyclotron-driven interruptions), so a mature proton service may perform differently.
The composite leaned against IMPT on weight loss (18% vs 6% at 12 mo), opposite to the dosimetric hypothesis, though p=0.079 does not establish harm. With OAR doses lower on IMPT for most structures and no functional gain, the gap between dosimetric advantage and patient-reported benefit is now the open question.
CONSORT flow
Randomised phase 3, both co-primaries null, source concludes IMRT remains SOC. Open-label, N=205, 2:1 allocation: wide CIs cannot exclude a meaningful difference either way.
- Does IMPT reduce late toxicity beyond 24 months? n=440 · primary completion 2031-12 · phase 3 IMPT vs IMRT, stage III-IVB OPSCC, side effects
- Why did severe weight loss run higher with IMPT?
- Can any NTCP threshold select pts who gain function from IMPT? n=400 · primary completion 2030-04 · phase 3 randomises NTCP-selected pts, protons vs photons
📚 Sources · 📄 1 paper
2026-07-08 ASTRO Annual Meeting 2025
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 longer read
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
📚 Sources · 📄 1 paper
2026-07-01
ReCOG HNSCC Reirradiation Consensus
TL;DR24 of 31 (77%) statements reached ≥85% agreement: no ENI, definitive IMRT ~66 Gy, postop ~60 Gy, SBRT ~40 Gy in five fractions.
RTOG 9610RTOG 9911GORTEC 2008–01GORTEC 98–03GORTEC-GETTEC
Subsite, not only volume, gates SBRT: in Vargo's 291-pt series severe toxicity reached up to 50% for laryngeal or hypopharyngeal recurrences vs 6-20% at other subsites, which is the basis for the 94% caution statement. Postop SBRT data are thin (n=28), which is why IMRT is preferred after salvage surgery.
For a recurrence 6-12 months after prior RT, this supports reirradiation only when disease is small, localised, and outside high-dose regions after MDT review; it does not extend to recurrence within 6 months outside urgent palliation, or to nasopharyngeal primaries. The longer read
Subsite, not only volume, gates SBRT: in Vargo's 291-pt series severe toxicity reached up to 50% for laryngeal or hypopharyngeal recurrences vs 6-20% at other subsites. Postop SBRT data are thin (n=28), so IMRT is preferred after salvage surgery.
8 details 1 trial watching
International expert consensus: a ReCOG core group drafted statements, then a 17-member radiation oncology panel voted once on 31 statements. Thresholds were high ≥85%, moderate 70-84%, low <70%, with no-opinion votes counted in the denominator. 24 (77%) reached high consensus.
Recurrent or second primary HNSCC in a previously irradiated field, covering definitive and postoperative reirradiation. Nasopharyngeal cancer and rare histologies excluded. Evidence levels adapted from Oxford CEBM, level 1 (randomised) to level 4 (expert consensus).
Definitive: CTV = GTV + 5 mm, no ENI, IMRT ~66 Gy for >25 cc, IMRT or SBRT for ≤25 cc, SBRT ~40 Gy in five fractions. Postoperative: primary-RT CTV rules, IMRT preferred, ~60 Gy. Cumulative doses in EQD2 with α/β 2-3 Gy, voxel-wise summation with DIR when available, with conservative manual point-dose addition acceptable otherwise.
Carotid: avoid cumulative EQD2 >120-125 Gy (α/β=2); encasement >180° and prior ulceration add risk. Mandible: ORN risk rises above 114-119 Gy EQD2. Cord/brainstem: D0.1 cc ≤50 and ≤55 Gy as targets, ~60 Gy at the surface case by case, recovery factors (10-25%) only with caution.
Split-course chemo-reRT in RTOG 9610 and RTOG 9911 gave 2-year OS of 15.2% and 25.9%, with grade 3-4 toxicity up to 34% and treatment-related mortality around 8%. The IMRT-era Ward cohort (n=412) reported median OS 16.5 mo with fatal toxicity in 1%. GORTEC-GETTEC (n=130), the randomised postop trial, improved locoregional control and RFS with no OS benefit.
Guidance for RPA class II and III rests largely on one retrospective RPA model, since most reirradiation outcome data come from RPA class I pts. The IMRT vs SBRT split and the dose statements come from non-randomised multi-institutional cohorts. The 10-25% recovery factor has very low evidence.
The seven moderate-consensus statements cluster on poor-prognosis subgroups, dose and fractionation, and SBRT, and the lowest agreement (70%) falls on cord and brainstem cumulative limits. The authors present dose thresholds and OAR constraints as pragmatic targets rather than validated limits. Sequencing reirradiation against first-line ICI has no randomised data.
| Statement | Consensus | Evidence level |
|---|---|---|
| Review prior RT plan: in-field vs marginal failure | High (100%) | 3 |
| Recurrence <6 mo: generally not advised, except urgent palliation | High (100%) | 2 |
| Recurrence 6-12 mo: only small, localised, outside high-dose regions, after MDT | High (88%) | 2 |
| RPA I with positive margins or ECE: consider postop reRT | High (100%) | 2 |
| R0 unlikely even without clearly positive margins: reRT can be considered | Moderate (82%) | 4 |
| RPA III: generally no curative-intent reRT | Moderate (76%) | 3 |
| Statement | Consensus | Evidence level |
|---|---|---|
| Postop: normofractionated IMRT preferred over SBRT | High (94%) | 3 |
| RPA II unresected ≤25 cc (cT1-T2): IMRT or SBRT similar | High (88%) | 3 |
| RPA II unresected >25 cc: IMRT over SBRT | High (88%) | 3 |
| SBRT with caution for unresected larynx or hypopharynx | High (94%) | 3 |
| Concomitant chemo case by case, large tumours in fit pts | High (88%) | 3 |
| Protons in selected cases to spare OARs | High (88%) | 3 |
| Brachy for <25 cc oral cavity or oropharynx, expert centres | High (88%) | 3 |
| Statement | Consensus | Evidence level |
|---|---|---|
| Definitive: CTV = GTV + 5 mm | High (94%) | 4 |
| Postop high-risk CTV: same rules as primary RT | Moderate (82%) | 4 |
| No elective nodal irradiation | High (100%) | 3 |
| Definitive IMRT: 66 Gy commonly used | Moderate (82%) | 4 |
| Hyperfractionated IMRT: an option, no clear benefit | Moderate (76%) | 3 |
| Postop: 60 Gy commonly used | High (94%) | 3 |
| SBRT: ~40 Gy in five fractions | Moderate (76%) | 3 |
| Statement | Consensus | Evidence level |
|---|---|---|
| Carotid: avoid cumulative EQD2 >120-125 Gy (α/β=2) | High (94%) | 3 |
| Carotid encasement >180° or prior ulceration raises blowout risk | High (100%) | 3 |
| Mandible: ORN risk rises above cumulative EQD2 114-119 Gy | High (100%) | 3 |
| Cord/brainstem recovery factors (10-25%): caution, document | High (88%) | 4 |
| Target cumulative EQD2 D0.1 cc ≤50 Gy cord, ≤55 Gy brainstem | Moderate (70%) | 4 |
| ~60 Gy at cord or brainstem surface, case by case | High (94%) | 4 |
| RPA class | Definition | n | 2-yr OS |
|---|---|---|---|
| I | Recurrence >2 yr after RT, salvage surgery | 91 | 61.9% |
| II | Unresected >2 yr, or <2 yr without organ dysfunction | 230 | 40.0% |
| III | <2 yr plus feeding tube or tracheostomy dependence | 81 | 16.8% |
| Dose | 2-yr OS |
|---|---|
| ≥66.0 Gy | 49.3% |
| 60.0-65.9 Gy | 34.2% |
| <60.0 Gy | 30.4% |
- Sequencing of reirradiation vs first-line immune checkpoint inhibitors recruiting Re-Radiochemotherapy and Pembrolizumab vs. Immuno(Chemo)Therapy for Locoregionally Recurrent PD-L1 Positive (CPS≥1) HNSCC Phase 3n=214 · primary completion 2032-12 · phase 3 re-RCT then pembro vs pembro ± chemo, OS
- Proton therapy vs IMRT outcomes in reirradiation
- Validated cumulative cord and brainstem constraints and recovery factors
📚 Sources · 📄 1 paper
2026-06-15
KEYNOTE-689 vs NIVOPOSTOP
TL;DREducational round-up contrasting perioperative pembro vs postoperative nivo in resectable LA-HNSCC; 3yr DFS ~63% vs 53% cited for NIVOPOSTOP.
KEYNOTE-689NIVOPOSTOP
RT is the constant in both frameworks, not the variable: cisplatin CRT stays the postoperative backbone and the question is only where IO is inserted around it. The source gives no dose, fractionation or target volume, and no in-field vs out-of-field failure data, so nothing here changes an RT plan.
Postoperative cisplatin CRT is unchanged in both designs; only the IO timing moves. With no dose, fractionation, target volume or locoregional failure data in the source, the radiation decision is untouched, though a neoadjuvant pembro window shifts the timing of adjuvant CRT initiation.
The two designs answer different sequencing questions: pembro started before surgery and continued through adjuvant CRT and maintenance, versus nivo added to postoperative CRT only in high-risk pathology. Only NIVOPOSTOP carries a number here, 3-year DFS approximately 63% vs 53%.
The perioperative design puts two cycles of pembrolizumab before resection, so the operation is delayed by a neoadjuvant window. The source reports no compromise in surgery completion but gives no numbers on delay, resectability, or margin status.
10 details 2 trials watching
- 🔍 Curator-shared teaching infographic (single-author X post), not a primary trial report or peer-reviewed review
- 🔍 Trial framing as presented in the graphic
Feature KEYNOTE-689 NIVOPOSTOP (GORTEC 2018-01) Drug Pembrolizumab Nivolumab Setting Perioperative (before + after surgery) Purely postoperative Population Resectable Stage III-IVA HNSCC High-risk resected HNSCC Control arm Surgery → RT/CRT Post-op CRT Primary endpoint Event-Free Survival (EFS) Disease-Free Survival (DFS) - 🔍 RT is fixed background in both: adjuvant cisplatin CRT in the KEYNOTE-689 experimental arm, postoperative cisplatin CRT in both NIVOPOSTOP arms
- 🔍 NIVOPOSTOP high-risk features listed
- Positive margins
- Extranodal extension (ENE)
- ≥4 involved nodes
- Extensive perineural invasion and other adverse pathological features
- 📊 NIVOPOSTOP: 3-year DFS approximately 63% vs 53% with standard CRT alone
- 📊 KEYNOTE-689: no effect size reported in source; graphic states only "significant improvement in EFS"
- ⚠️ No HR, CI, or p-value given for either trial in the source
- ⚠️ "3-year DFS approximately 63% vs 53%" is an approximation in the source, not a reported trial value
- ⚠️ No RT dose, fractionation, or target volume reported for either trial in source
- ⚠️ Graphic asserts higher PD-L1 CPS derives greater benefit with no supporting numbers
- Head-to-head of perioperative vs postoperative IO timing n=80 · primary completion 2029-02 · perioperative IO vs pembro + surgery + adjuvant RT
- Whether PD-L1 CPS should gate IO in resected HNSCC
- Locoregional failure patterns with IO added to postoperative CRT n=173 · primary completion 2025-11 · anti-PD-1 added to adjuvant CCRT, DFS primary
📚 Sources · 🐦 1 tweet
🧠 High-yield: KEYNOTE-689 vs NIVOPOSTOP
— Dr Rupam Manna MD (@DrRupamOncology) June 15, 2026
These two trials are redefining standards for resectable LA-HNSCC.
1/ KEYNOTE-689 (NEJM 2025)
Perioperative pembro → significant EFS benefit
First positive perioperative IO trial in >2 decades
2/ NIVOPOSTOP (ASCO 2025)
Post-op nivo +… pic.twitter.com/1mRQr9jmVX
2026-06-04 ASCO Annual Meeting 2026
AREST
ForResected pT1-2N0 OSCC, margins ≥5mm, ≥1 intermediate-risk feature
HR 0.52
95% CI 0.30-0.91, p=0.02 · 3-yr 89.2% vs 80.9% (ITT)
TL;DR3-yr LRFS 89.2% vs 80.9% (HR 0.52, p=0.02) with adjuvant RT vs observation in resected intermediate-risk pT1-2N0 OSCC; no significant DFS/OS difference.
Competing-risk LRF incidence was 10.6% vs 18.9% (HR 0.52) with 60Gy/30fx to tumor bed and at-risk neck. Controls had ≥5mm margins and a level I-III dissection with ≥16 nodes, so this is RT's increment over high-quality surgery alone. It moves adjuvant RT vs observation most for oral tongue, an exploratory subgroup.
In pT1-2N0 OSCC resected with ≥5mm margins and an adequate level I-III neck dissection but carrying PNI, LVE, poor differentiation or DOI ≥5 to ≤10mm, this supports discussing adjuvant RT for loco-regional control; it does not extend to close-margin, limited-neck or node-positive disease. The longer read
Competing-risk LRF incidence 10.6% vs 18.9% (HR 0.52) with 60Gy/30fx to tumor bed and at-risk neck, after surgery that already met ≥5mm margins and ≥16 nodes. It moves adjuvant RT vs observation in pT1-2N0 OSCC, strongest for oral tongue in an exploratory subgroup; smaller target volumes untested.
Benefit was measured against surgery with clear margins ≥5mm and an ipsilateral level I-III dissection yielding ≥16 nodes, so RT added loco-regional control even after high-quality resection. Pathology reporting of PNI, LVE, differentiation and DOI now gates the adjuvant referral; the data do not cover close margins or a limited neck.
9 details
Multicenter, open-label, phase III RCT from India, 1:1, stratified by oral cavity subsite, PNI/LVE and differentiation. N=392 (191 RT, 201 observation); median follow-up 47.2 mo (IQR 30-59.4). Powered at 80% to detect HR 0.6256, assuming 3-yr LRFS 70% with observation.
pT1-T2, pN0 OSCC after adequate surgery: clear margins ≥5mm and at least ipsilateral level I-III neck dissection with ≥16 nodes. Required ≥1 intermediate risk factor: DOI ≥5 to ≤10mm, PNI, LVE, or poor differentiation.
60Gy in 30 fractions over 6 weeks to the resected tumor bed and at-risk neck nodal region. Technique and nodal volume definitions not stated in source; no concurrent systemic therapy described.
Primary: loco-regional recurrence-free survival, from randomization to first local and/or regional recurrence. Also reported: cumulative incidence of loco-regional failure with death as competing event, DFS, OS.
LRFS benefit held on PP and competing-risk analyses; DFS and OS not significantly different, with no survival HRs in source.
| Endpoint | Adjuvant RT | Observation | HR (95% CI), p |
|---|---|---|---|
| 3-yr LRFS, ITT | 89.2% (84.3-93.3) | 80.9% (74.6-86.1) | 0.52 (0.30-0.91), p=0.02 |
| 3-yr LRFS, PP | 91.1% | 80.9% | 0.43 (0.23-0.80), p=0.01 |
| LRF cumulative incidence (death competing), ITT | 10.6% (6.1-15.1) | 18.9% (13.3-24.6) | 0.52 (0.30-0.91), p=0.021 |
| LRF cumulative incidence (death competing), PP | 8.7% (4.3-13.1) | 18.9% (13.3-24.6) | 0.43 (0.23-0.79), p=0.007 |
The abstract describes the evidence for adjuvant RT in this setting as largely retrospective; AREST is the randomised test of that practice. The RTOG and EORTC postoperative trials tested adding concurrent chemo to RT in high-risk resected disease and did not randomise RT itself against observation.
Control arm beat its design assumption (3-yr LRFS 80.9% vs 70% assumed), so the absolute gain sits against a lower-risk baseline than planned. Oral tongue vs buccal mucosa is an exploratory subgroup with no HR in source, and no result is given per risk factor. Salvage treatment at recurrence is not described, which bears on the OS read.
Loco-regional control improved without DFS or OS separation; one explanation is effective salvage of isolated recurrences in observed pts, which the abstract cannot confirm. The decision it moves is adjuvant RT vs observation after adequate resection, weighed against RT morbidity the abstract does not report.
CONSORT flow
Randomised phase III hit LRFS primary at 47.2 mo median f/u; DFS/OS not significantly different and no toxicity data in abstract, so supports rather than mandates adjuvant RT.
- Which intermediate-risk feature (PNI, LVE, DOI, grade) drives the benefit?
- Does the oral tongue benefit hold as a prespecified subgroup?
- Late RT toxicity and salvage outcomes in the observation arm
📚 Sources · 🐦 1 tweet · 📄 1 paper
#ASCO26
— Dr Rishabh Jain (@DrRishabhOnco) May 27, 2026
🗣️ The AREST trial tackles one of the biggest gray zones in oral cavity cancer.
After adequate surgery in pT1-2N0 OSCC with intermediate-risk features:
✅ Adjuvant RT improved loco-regional control
❌ No OS benefit observed
3-year LRFS:
🔹 89.2% vs 80.9%
🔹 HR 0.52… https://t.co/qsALPFX032 pic.twitter.com/F4XzTMETih
Abstract
2026-05-18 ESTRO Congress 2026
INRT-AIR & DARTBOARD pooled analysis
ForHNSCC oropharynx/larynx/hypopharynx, stage I-IVB, excluding T1-2N0 larynx
TL;DR5yr solitary elective nodal recurrence 0% with ENI omission in HNSCC; 5yr OS 87%, PFS 74%, n=117.
The number that matters is 0% solitary elective nodal recurrence at 5 yrs: elective volumes are where the parotid, constrictor and pharyngeal dose lives, and this is the first pooled long-term read that omitting them does not trade nodal control. MDADI 84.9 at 12 mo with no significant decline is the swallowing correlate. No dose or CTV detail in source, so the contouring approach cannot be replicated from this abstract.
In stage I-IVB oropharynx, larynx or hypopharynx SCC being planned for definitive chemoRT, this supports enrolling on an INRT protocol rather than adopting nodal omission off-trial; it does not extend to T1-2N0 larynx, which was excluded. The longer read
This is the elective-volume decision, the one de-escalation lever definitive chemoRT has never randomised. 0% solitary elective nodal recurrence at 5 yrs with MDADI 84.9 at 12 mo says the trade is not nodal control, but the nodal selection ran through an AI model on staging PET/CT, so the transferable piece may be the selection step, not the omission.
+1 more figure
11 details 2 trials watching
Patient-level pooled analysis of 2 prospective trials, INRT-AIR and DARTBOARD, both testing involved nodal radiotherapy. N=117, median follow-up 3.4 years. No randomised comparator arm receiving standard elective nodal irradiation.
HNSCC of oropharynx, larynx, and hypopharynx, stage I-IVB, explicitly excluding T1-2N0 larynx. Completed PET/CT and neck CT were required for eligibility, which is also the imaging substrate the nodal-selection step depends on.
Definitive chemoradiotherapy with omission of elective nodal irradiation (ENI), treating involved nodes only (INRT). Suspicious node identification was assisted by an artificial-intelligence model reading the staging PET/CT and neck CT. Dose, fractionation and margin expansions are not reported in the source.
5-yr solitary elective nodal recurrence 0%. 3-yr cumulative incidence: local 9.5%, regional 4.3%, distant 11%. 5-yr OS 87%, PFS 74%. Mean composite MDADI 84.9 at 12 months with no significant decline after treatment.
Pooling two protocols with different designs into one patient-level cohort assumes their INRT definitions were interchangeable, which the source does not establish. Median follow-up of 3.4 years supports a 5-year estimate on a shrinking risk set, and HPV status, which drives OS in an oropharynx-weighted cohort, is not reported.
The zero solitary elective nodal failures make the mechanistic case that undissected, PET-negative elective levels rarely harbour disease that only ENI would sterilise. What the pooled cohort cannot settle is whether the result survives outside two experienced centres using an AI-assisted nodal-selection step, which is precisely the component a general department would have to reproduce.
Pooled single-arm prospective cohorts, n=117, no randomised comparator against standard elective nodal RT. Presenters state randomised evidence needed before non-trial use.
- Randomised INRT vs elective nodal irradiation in definitive chemoRT recruiting Dose De-escalation and Sentinel LN Mapping Driven Radiotherapy of Contralateral Neck in Ipsilateral Node Positive HNSCC Phase NAn=147 · primary completion 2027-01 · sentinel node mapping tailors contralateral neck volumerecruiting Invert-Prospective Phase II Randomized Trial of Involved Nodal Versus Elective Neck RadioTherapy Phase 2n=80 · primary completion 2028-07 · randomised INRT vs ENI, solitary elective recurrence
- Does AI-assisted nodal selection outperform physician contouring
- Whether INRT toxicity benefit is measurable against standard elective volumes
📚 Sources · 🐦 1 tweet
Day FOUR of #ESTRO26 Coverage by OncoAlert 🚨
— OncoAlert (@OncoAlert) May 18, 2026
Omission of elective nodal irradiation in HNSCC: long-term results and patient-level pooled analysis from 2 prospective trials (INRT-AIR & DARTBOARD)
Presenter Sympascho Young 🇺🇸
A patient-level pooled analysis of 117 patients… pic.twitter.com/KaaT70nSNH
2026-05-17 ESTRO Congress 2026
DIREKHT
ForResected HNSCC referred for post-operative RT
TL;DRDe-escalated post-op HNSCC RT: contralateral neck sparing and primary CTV to 56 Gy in selected pts; no outcome numbers in source tweet.
The two levers named are the ones that gate post-op HNSCC RT morbidity: contralateral neck omission and a 56 Gy primary CTV rather than standard higher-dose coverage. Which pts qualified is the whole question, and the source text does not give the selection criteria or any control or toxicity numbers.
Both levers are RT-side decisions: whether to cover the contralateral neck electively, and whether 56 Gy suffices for the primary CTV in selected post-op pts. The source names the approach but gives no selection rule and no control or toxicity numbers, so it flags a trial to read rather than a volume or dose change to adopt.
6 details
Two de-escalation levers reported in source: contralateral neck sparing in a specified subgroup, and reduction of the primary CTV dose to 56 Gy. Fractionation, technique, and the dose to involved or high-risk volumes are not stated in the source text.
Described as a trial, but phase, randomisation, N, sites, and follow-up are not stated in the source tweet.
No primary endpoint, effect size, or toxicity result reported in source. The tweet is commentary on the trial's approach, not its results.
Everything that would let a reader act on this is missing from the source: the selection rule for contralateral neck omission, the comparator, and any locoregional control or late-toxicity figures. A de-escalation result is only interpretable against its non-inferiority margin, which is not given.
Source is a single commentary tweet with no design, N, endpoint, or effect size. Nothing to classify beyond the de-escalation concept.
- Which pts qualify for contralateral neck sparing post-operatively
- Whether 56 Gy primary CTV holds locoregional control vs standard dose
📚 Sources · 🐦 1 tweet
There are tremendous opportunities to improve post-operative radiotherapy in HNSCC. The DIREKHT trial is an excellent example of such work, in which they spared the contralateral neck in a specified group of patients and/or reduced the primary CTV dose to 56 Gy.
— David Sher (@DavidSherMD) May 16, 2026
The details… https://t.co/7W84LYIofR