Replies to post #635 on Biotech Values

[Biowatch]

>>This is a project I did for a statistics course. Thought I would share it on the board.<<

Tell us what grade you get, or what the teacher says... :-)

[DewDiligence]

Bishop: thanks for posting the findings of your statistics project. As you probably know, it is dangerous to compare statistics across clinical trials because differences in the patient pool are apt to mask some of the differences in the therapies themselves. With that caveat in mind, it’s interesting that you are able to conclude (with a p-value of .04) that Squalamine is more efficacious than Macugen (in terms of the % of patients gaining 3+ lines) by at least 12.5 percentage points.

--
From QLTI’s webcast last week, I was able to fill in more of the blanks in the Macugen phase-3 data and construct the following table (the Macugen data refer to the 0.3mg dose, which had the best efficacy of the three tested doses):

 
 Change in lines 
 visual acuity        % of Patients 
 at 1 year         MACUGEN    PLACEBO 
  
 >=+3 (“improved”)    6          2 
 >= 0                33         23 
 >=-3 (“stable”)     70         55 
 >=-6                90         78 
 

Note: more detail on the visual-acuity outcomes is available, but I condensed the data to simplify the presentation.

[DewDiligence]

>>
Test 2:
Squalamine (13/39) vs. Macugen (Phase II – 54/900), Hypothesized Difference = -0.125
<<

I assume that’s a typo and you actually input the correct 6/24 figure for the Macugen phase-2 data as detailed in your narrative (rather than the 54/900 phase-3 figure as shown above). The test result looks OK so a typo is the only explanation that makes sense.

Once again, thanks for sharing your work. Did you notice that Biowatch asked what grade you received? I’d like to know too. Regards, Dew