Bioequivalence and Bioavailability Forum

Messy numbers - PowerTOST V1.3-01 preview [Regulatives / Guidelines]

posted by d_labes  – Berlin, Germany, 2015-09-24 11:18 (3463 d 14:48 ago) – Posting: # 15465
Views: 16,824

Dear Helmut,

❝ ... On another note I don’t understand why the FDA came up with their “BE approach #4”. It would have been sufficient to state “apply the NTID-approach given in the Warfarin guidance”. Given the reported CVs [...] downscaling would be unlikely. The sample size would be driven by part of the requirement “must pass 80.00–125.00%”. Sample sizes (80% power, 4-period full rep­li­cate, CV_T=CV_R):

IMHO totally correct. Lets compare NTIDFDA to HVNTID via the upcoming PowerTOSTV1.3-01 new function sampleN.HVNTID():

          GMR= 0.90       GMR= 0.95
  CV   NTIDFDA HVNTID  NTIDFDA HVNTID
 ------------------------------------
 0.20    36     20       18     14
 0.25    32     30       18     18
 ------------------------------------
 0.30    40     40       22     22
 0.35    54     54       28     28
 0.40    68     68       34     34
 0.45    84     84       42     42
 0.50   102    102       50     50
 0.55   120    120       60     60
 0.60   140    140       68     68

As expected: Identical numbers for higher variabilities, nearly totally driven by the ABE requirement .
For instance GMR=0.95, n from table above, power.NTIDFDA(..., details=TRUE)

CV       p(BE)  p(BE-sABEc)    p(BE-ABE) p(BE-sratio)
0.3    0.82570      0.95213      0.83752      0.99070
0.4    0.80727      0.99704      0.80747      0.99974
0.5    0.80496      0.99993      0.80496      1.00000
0.6    0.80286      1.00000      0.80286      1.00000
p(BE-sABEc) = prob. of scaled ABE test alone
p(BE-ABE)   = prob. of ABE test alone
p(BE-sratio)= prob. of test of variabilities alone

Of course HVNTID is simpler. No need to deal with "linearized scaled ABE criterion" and it's 95% upper CI.

Complete thread:

• New D. Guid­ance – Dabigatran jag009 2015-09-21 23:19 [Regulatives / Guidelines] 
  • New D. Guid­ance – Dabigatran nobody 2015-09-22 11:04
    • The FDA’s fourth BE approach! Helmut 2015-09-22 11:43
      • The FDA’s fourth BE approach! nobody 2015-09-22 13:06
        • The FDA’s fourth BE approach! Helmut 2015-09-22 13:25
          • The FDA’s fourth BE approach! Yada Yada Yada jag009 2015-09-23 20:29
            • Yucky… Helmut 2015-09-23 22:06
      • The FDA’s fourth BE approach! d_labes 2015-09-22 14:08
        • New PowerTOST function(s) Helmut 2015-09-22 14:27
          • New PowerTOST function(s) d_labes 2015-09-22 14:59
            • Pradaxa = HVDP (dabi­ga­tran = HVD?) Helmut 2015-09-22 15:16
              • Pradaxa = HVDP (dabi­ga­tran = HVD?) Shuanghe 2015-09-22 16:53
                • HVNTID's d_labes 2015-09-23 08:34
                  • HVNTID's Shuanghe 2015-09-25 16:14
              • Highly variable NTID d_labes 2015-09-23 08:23
                • What a mess! Helmut 2015-09-23 13:20
                  • Messy numbers - PowerTOST V1.3-01 previewd_labes 2015-09-24 09:18
                  • What a mess! Shuanghe 2015-09-25 16:32
        • OT: power.NTIDFDA() Helmut 2015-09-23 00:10
          • OT: power.NTIDFDA() d_labes 2015-09-23 09:37