While you may be right, I wonder what sort of surrogate endpoint would be necessary when all patients still alive in the trial have survived for at least 4 years, many much longer.
I would think that all current survivors would be evaluated either by the clinicians or their current doctors to classify their current status. I think they could be put into a few broad categories. One would certainly be those who appear to be a no current risk as no cancer can currently be found. Another would still have discernible cancer, but no progression observable. Others would be progressing, and some may be in hospital or hospice care and considered critical.
Perhaps if you had all this information you might be able to predict, for instance, where the median for the Top 100 would end up, but then you could probably look at the Top 50, and that would keep advancing. We really have little or no knowledge about the living, if most of them essentially are in complete remission, that median for the Top 100 could exceed a decade by a decade from now. I don't know that they'll look back at the trial after that much time, but it would be interesting to learn what they'd see if they did.
Other than the current status of all those still living, I don't know that they have much they can estimate. I wouldn't be surprised if many of the Top 100 lived to the point that it wasn't GBM that ended their lives.
I certainly don't know, but for discussion let's say that 50 of the top 100 are in complete remission, no cancer can be found. If that were the case, how would the anticipated life span of these people be figured. I don't believe the K-M plots have a provision for noting that certain patients seen on the plot are in complete remission and potentially could live decades longer. In reality, these people could be spread over the entire curve beyond 49 months from its origin.
Happy Holiday's all,
Gary
