Head & Neck
IO sequencing around surgery, with post-op cisplatin CRT unchanged in every arm.
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