Dear All,Back to the Top
I have concentrations sampled upto 72.0 hrs, I got few concentrations
at 72.0 hrs, but I want to calculate or extrapolate the data upto 168.0
hrs. Whether is it possible to calculate or extrapolate concentrations
upto 168.0 hrs with the current data? If so, please help me how to
calculate or extrapolate the concentrations of 72.0 hrs to 168.0 hrs.
With Regards,
Radhakrishna
[Do you know (from other experiments) the shape of the line after 72
hr?. If not I don't believe you can extrapolate beyond 72 hr. The line
may curve up (unlikely) or down (linear or nonlinear). - db]
Back to the Top
You will have to fit a PK model to your data in order to obtain the
accurate estimate of 168 hr. The accuracy of the estimate depends on
how good the fit is.
Back to the Top
Dear Radhakrishna,
I agree with the comments by David Bourne and Vuong Trieu. Extrapolation
from 72 to 168 hours is rather speculative, unless one has a nice
log-linear
phase with at least 4 - 6 data points (and even then nobody knows what
happens after the last sampling point; this is always speculative, as
indicated by David Bourne). If so, one can calculate the confidence
interval
of the predicted concentration at 168 h. I am quite sure that this
interval
will be rather broad, as a result of the extrapolation over such a long
period (far exceeding the period of measurement). And in that case the
prediction is of limited value, of course. A predicted concentration
without
a confidence interval is meaningless, or better, misleading.
Best regards,
Hans Proost
Johannes H. Proost
Dept. of Pharmacokinetics and Drug Delivery
University Centre for Pharmacy
Antonius Deusinglaan 1
9713 AV Groningen, The Netherlands
tel. 31-50 363 3292
fax 31-50 363 3247
Email: j.h.proost.-a-.farm.rug.nl
Back to the Top
hi
As you are said that you have data upto 72 hrs and would like to simulate
the concentration i am imagining that ur drug's half life is extremely
long. what is the half life of ur drug?
Yes it is possible to calculate the extrapolated concentration upto 168
hrs using any software like winnonlin or kinetica using the simulation
technique.
In that case you should have very reliable estimates of the initial
input parameters (macro or micro constants) to simulate the latter
points. the simulated data could also cross verify if the observed time
points match with the simulated ones. the best way would be to obtain
these initial parameters from an intravenous or infusion study.
For simulations everything boils down to how reliable are your initial
estimates. This technique could give you a fair idea but wont be better
than the actual data (which might be time taking).
If believe that ur observed concentrations are still in the declining
phase of the curve and are still high enough, i would suggest that you
should increase the sensitivity of the assay and capture the complete
curve. If the ratio of AUC0-t/AUC0-inf is not greater than 90% then it
means that you have not captured most of the profile.
hope this helps
manish issar
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)