InvestorsHub Logo
Followers 22
Posts 3527
Boards Moderated 0
Alias Born 10/15/2016

Re: jamesnhansen post# 35874

Monday, 02/22/2021 11:33:52 AM

Monday, February 22, 2021 11:33:52 AM

Post# of 43784
Nice...

Some notes

When I did it, I assumed all enrollment numbers for a month were at the very end of the month because there is no information reported other than monthly. Earlier when doing this sort of thing, that assumption was significant in pondering when 298 events will be reached.

Pretty cool breakdown by cancer stage. Comment more about those assumptions please since afaik we can't know that, how many in each stage when.

The three controlling assumptions I came up with are the OS assumptions, the dropout assumptions, and the % responding to injection assumptions.

For the risk side of things, our "sample size" is about 800 compared to SEER data sampling maybe 200,000. So I think we figured 5% slop was in order.

That led me to do a stochastic (monte carlo) but that was limited by available information to the assumption of simply IF there is a difference, WHEN does it appear most dramatic, as the data runs along. Basically it simply characterized the enrollment and drew two lines right on top of each other until Q2 to Q4 last year when there was a clear separation. Possibly otherwise known as "delayed clinical effect", but actually a function of "waves" of enrollment.

In the end I came up with a more conservative estimate of 11.25% differential measured at year 3 ( of the individual patient year 3 each, not the over all trial year 3 ) This was more of a "chart interpretation" than a direct math.





Volume:
Day Range:
Bid:
Ask:
Last Trade Time:
Total Trades:
  • 1D
  • 1M
  • 3M
  • 6M
  • 1Y
  • 5Y
Recent CVM News