Yeah, I think we've talked about this previously. So, this is - it's an interesting - the trial design, because it's such a - thromboembolic events with valve are very low, which is why we have a 1,000 patient trial. And the way this is set up as a non-inferiority to Coumadin. And so there really is no interim look. I mean you could get to a situation where you had the trials, and you're very familiar with how this works from a statistic standpoint, where you can have a trial stop early, but they would have to be a significant difference between Coumadin and Eliquis, which, again, we don't think is going to happen. This is powered for non-inferiority. So, I think, maybe the way I would answer it is we're assuming the trial enrolls in the second quarter next year. That means that you've got 1,000 valves, half on Eliquis, half on Coumadin. And we started this trial, as you know, in May of 2020. That means we will, at that point, have already had some of the first patients on Eliquis for two years. And I think the way to think about this trial, once this trial enrolls, every quarter that goes by, it's another quarter with 500 patients on Eliquis. So, in the end, we won't know the answer to the question until we get to our full 2-year enrollment. So, I think that's probably the best I can answer at this time.