PharmPK Discussion - Allometry and Dedrick Plots

• On 16 Apr 2007 at 12:09:33, Tonika Bohnert (tonika.bohnert.aaa.biogenidec.com) sent the message
  

  
  Hello,
  
  I am new to the field of allometry and am having some troubles with
  one particular prediction that I am working on. For this one
  compound, we have IV CL(mL/min) and Vss(L/kg) respectively as
  follows: Mouse (0.0555, 0.778), Rat (0.65, 0.548), Dog (20.28, 1.1).
  When I plot log Cl vs log BW, I get very nice straight lines for both
  Cl and Vss with equations being : log CL = 0.9488 log BW + 0.3106 and
  log Vss = 1.069 log BW - 0.837 with R-squared values of 0.999 for
  both CL and Vss. However, after this for concentration projections,
  when I convert the concentrations and times to plot Complex Dedrick
  plots (I am using Complex Dedrick since exponent of Vss is greater
  than 1) , instead of overlapping plots , I get almost 3 parallel
  plots with C0 for Dog about 10-fold higher than rat and C0-rat being
  10 fold higher than C0-mouse (on a log scale). When the plots are
  overlapping, then concentration vs time projections calculations are
  straight forward but when the plots are parallel with about 10-fold
  difference in their C0 values , then how does one proceed to do these
  projections? Does this mean allometry fails to predict conc vs time
  (for a specified dose & BW of another species in question/
  prediction). Do we need to get data from another species to make any
  further predictions?
  
  For Dedrick plots I am plotting C/(Dose/BW d) vs Time/ BW  d-b where
  d = 1.069 and b = 0.9488 from above equations.
  
  For another compound in the same series we got v.similar data but for
  that we also had monkey data and surprisingly, the monkey and rat
  plots were exactly superimposible and the dog had 10-fold higher C0
  (but parallel plot) and mouse had 10-fold lower C0 (again parallel
  plot). So what does this tell us?
  
  Thanks so much for any advice/input on this. Shall be eagerly looking
  forward to it !
  
  Tonika Bohnert, Ph.D
  DMPK.
  Biogen Idec Inc.,
  15 Cambridge Center
  Cambridge, MA 02142
  
  

  Back to the Top

• On 17 Apr 2007 at 08:57:44, Nick Holford (n.holford.-a-.auckland.ac.nz) sent the message
  

  
  The following message was posted to: PharmPK
  
  Tonika,
  
  It is essentially impossible with such small amounts of data to
  estimate allometric model exponents with any hope they might be
  unbiased and precise. The theoretically correct values for the
  exponent of Clearance of 0.75 and for volume is 1.0.
  
  What are you trying to do with your allometric predictions?
  
  Nick
  
  --
  Nick Holford, Dept Pharmacology & Clinical Pharmacology
  University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New
  Zealand
  email:n.holford.at.auckland.ac.nz tel:+64(9)373-7599x86730 fax:373-7556
  http://www.health.auckland.ac.nz/pharmacology/staff/nholford/
  
  

  Back to the Top

• On 17 Apr 2007 at 10:16:18, "Ma Guangli" (guanglima.-a-.gmail.com) sent the message
  

  
  Dear Tonika,
       Although there are many allmetric methods, why not to examine
  the molecular properties of your compound first? The studied
  molecular descriptors are simple.
  Jolivette LJ, Ward KW 2005. Extrapolation of human pharmacokinetic
  parameters from rat, dog, and monkey data: Molecular properties
  associated with extrapolative success or failure. J Pharm Sci  94(7):
  1467-1483.
  Evans CA, Jolivette LJ, Nagilla R, Ward KW 2006. Extrapolation of
  preclinical pharmacokinetics and molecular feature analysis of
  "discovery-like" molecules to predict human pharmacokinetics. Drug
  Metab Dispos  34(7):1255-1265.
  Guangli
  
  

  Back to the Top

Copyright 1995-2011 David W. A. Bourne (david@boomer.org)