I am perfectly happy to stick with K/M. But the post I replied was making a point that these numbers were not K/M but "actual".
If we go to K/M to, then you need to also get the first half K/M. And that has not been made known. [And can not be calculated from the reversed engineered data set provided by Senti's and others work]
I would guess that a proper K/M calculation would come out more like 30% in H2 vs 26% in H1. That is better than the "actual" improvement from H1-h2. And I would say a better estimate than the "actual".
On the last 149 patients: 40% treatment number at 36 months would TKO 15% SOC number at 36 months.
We just need to wait the real numbers. I believe the number would not deviate so much. Probably with 1% more or less. It is hard to say otherwise since it is in the math. Imo.