NorthWest Biotherapeutics Inc (NWBO)

[akasidney86]

Related to the previous thread about use of median OS as an endpoint for immunotherapies when it's been shown that they don't act well on rapid progressors or early in a cycle, but do act very well later in the cycle, something I hadn't noticed the significane of before is the following from the PR link posted by Lunatick, the germane part of which is quoted below:

The current trial plan involves enrollment of 312 total patients in the trial, and counting 110 “events” of tumor recurrence or patient deaths from among these 312 patients to determine whether the primary endpoint of the trial is met. As announced separately today, the Company is modifying its trial plan so that it will enroll 348 total patients, and will count 248 “events” from among these 348 total patients to determine whether the primary endpoint of the trial is met. So, the number of “events” counted in the statistical analysis will increase from 110 to 248, but the total enrollment in the trial will only increase from 312 to 348.

Like I said, I hadn't noted the significance off the fact that events increased by 125%, while the total number of patients was only increased by 11%. Another way to understand the significance is to consider the fraction each represents of the total patients. Under the old regime, results of only 110 out of 312 patients would constitute the end point, or 35% of the total. Under the new protocol, the new total events number, 248, represents 71% of the total OS, GREATLY increasing the odds of success by PICKING UP A FAR GREATER NUMBER OF THE LATER, LONG TAILED eventers.

Sometimes I just don't see the forest for the trees.