Bioequivalence and Bioavailability Forum

Hi everyone,

My sample size program has undergone a few renovations. Version 1.6 is available for download here:
http://individual.utoronto.ca/ddubins/FARTSSIE.xls

Please remember to enable macros, otherwise sample size estimation will not work.

The reason for the most recent version is an addition of a worksheet ("Scaled BE - HVD") which will calculate the scaled bioequivalence limits for each method detailed in the FDA proposal for Highly Variable Drugs, outlined here:
http://www.fda.gov/ohrms/dockets/ac/04/briefing/4034B1_07_Bioequivalence%20Requirments-Draft.pdf
If method C2 is to go forward, essentially highly variable drugs will get a big break. Provided the Test/Reference ratio is between 95-105%, there shouldn't be a need for more than 40 subjects to attain 80% power. In fact, as the intraCV of the reference increases, the number of subjects required actually decreases asymptotically to 33. The impact on the industry would be massive.

Here is a guide to what the BE limits would look like using method C2, scaled to the intraCV of the reference product:
ISV  SIGwr    Scaled BE     SS  SS(C2)
30%  0.2936  80.00-125.00   39   39
32%  0.3122  78.88-126.78   44   38
34%  0.3307  77.77-128.58   49   37
36%  0.3491  76.70-130.38   54   36
38%  0.3673  75.64-132.20   60   36
40%  0.3853  74.62-134.02   66   36
42%  0.4031  73.61-135.84   72   35
44%  0.4207  72.63-137.68   78   35
46%  0.4381  71.68-139.51   84   35
48%  0.4553  70.75-141.35   91   34
50%  0.4724  69.84-143.19   98   34
52%  0.4892  68.95-145.04  105   34
54%  0.5058  68.08-146.88  112   34
56%  0.5223  67.24-148.73  119   34
58%  0.5385  66.41-150.57  126   33
60%  0.5545  65.61-152.42  134   33
62%  0.5703  64.83-154.26  142   33
64%  0.5859  64.06-156.10  149   33
66%  0.6013  63.32-157.94  157   33
68%  0.6165  62.59-159.77  165   33
70%  0.6315  61.88-161.60  173   33
72%  0.6463  61.19-163.42  181   33
74%  0.6608  60.52-165.24  189   33
76%  0.6752  59.86-167.06  198   33
78%  0.6894  59.22-168.87  206   33
80%  0.7033  58.59-170.67  214   33

Column Headers:
ISV:       IntraCV of reference product (1^st column)
SIGwr:     Intrasubject standard deviation of reference product (2^nd column)
Scaled BE: Bioequivalence limits scaled using the proposed FDA method C2 (3^rd column)
SS:        Sample size using traditional 80-125% limits, given a T/R ratio between 95-105% at
           80% power (4^th column)
SS(C2):    Sample size using the scaled BE limits in the third column, given a T/R ratio
           between 95-105% at 80% power (5^th column)

Best,

David Dubins

--
David Dubins, Ph.D., B.Eng.
Global Bioequivalence Consulting
Assistant Professor, Leslie Dan Faculty of Pharmacy
University of Toronto
---