Opinion|Videos|November 13, 2026

The NIVOPOSTOP Controversy: Open-Label Design, a Negative Central Read, and the KEYNOTE-689 Comparison

The primary outcome in this study, as defined, was disease-free survival as assessed by the treating investigator — the physician giving the nivolumab if the patient was randomized to that arm.

Sun: The primary outcome in this study, as defined, was disease-free survival as assessed by the treating investigator — the physician giving the nivolumab if the patient was randomized to that arm. The issue is that this was a cooperative group trial out of the French group, and in many ways it was a very practical, very well-conducted trial done across several countries in Europe. But it was open label. Physicians knew exactly what their patients were receiving. You can imagine how that could, in theory, bias providers and patients on outcome assessment, particularly in the post-operative setting for head and neck cancer, where imaging interpretation is already challenging because of post-operative and post-treatment changes.

Sun: So the primary outcome was positive. You might say, great, positive trial. But one of the secondary outcomes was disease-free survival by blinded independent central review — and that outcome was technically negative. The other issue is that overall survival in this trial is immature, and of the curves we see so far, they don’t appear to separate all that much. One of our longer-term questions will be what to do with this data. Is this really the best thing for patients? What if we’re able to salvage many of these patients at the time of recurrence with immunotherapy? How do you make the trade-off between delaying recurrence and potentially introducing more toxicity with nivolumab — which we know to be true — versus treating patients later?

Sun: The way I think about it: this is something that is being discussed in our guideline organizations as to whether to include it as part of our standard. It is not quite there yet, but I wouldn’t be surprised if it is very soon, and I think it is worth discussing with patients today. That said, I think it will likely be a temporary standby — we will have more data and more trials down the road that change this paradigm, and I don’t expect this to be our standard of care for very long.

Jahan: Have you started discussing this with your patients already?

Sun: Yes. We’ve been discussing it ever since the initial abstract report came out, because patients read about it. Part of the discussion is how to balance the earlier perioperative data with pembrolizumab against this adjuvant data with nivolumab. There is no direct head-to-head trial, and running such a trial would be very challenging — as we’ve seen in other diseases like lung cancer. Then there are our intrinsic biases about neoadjuvant therapy. Many of us believe that having the tumor in place at the time you give immunotherapy matters, in a way that’s different from only minimal residual or microscopic disease. So when I discuss NIVOPOSTOP, I’m also discussing KEYNOTE-689 at the same time. It’s important to recognize these are not the exact same patient population — if anything, they are quite different — but it becomes a much more complex interplay, not just between us and the patient, but also between us, the patient, the surgeon, and the radiation oncologist. It emphasizes how important tumor board discussions and multidisciplinary care are for these diseases.

Jahan: Yes — more and more, all of these advanced and early-stage settings across cancer types are becoming very multidisciplinary. Having a strong team is critical.


Related to this article