Back to the Top
Hello All,
What is the normal practice of reporting half-life values for a following drug profile:
* Exhibits a bi-exponential decline.
* Terminal phase concentrations were LLOQ for lower doses and hence half-life was calculated based
on alpha slope.
* Lower dose group half-life values are much lower than the higher dose half-life values (as
terminal phase concentrations were above LLOQ)
Questions:
* Should I just report one half-life?
* Should I report both half-life values?
* Or should I additionally report model predicted thalf values ?
Thanks and appreciate your help,
Soujanya
Back to the Top
Dear Soujanya:
If you have a biexponential decline, then you need to report the appropriate rate constants. As
to the issue of below LOQ, this is an illusion brought about by using CV% as a measure of
precision. You should, of course, get the SD of your assay over its working range, but then square
this to get the variance, and use the reciprocal of the assay variance at the measured
concentration as the correct measure of precision. Why the laboratory community does not do this I
simply do not know. The errors of assay measurements are Gaussian. The correct measure of the
precision of a measurement having a Gaussian distribution of its errors is the reciprocal of the
assay variance. Because if this, the whole issue of LOQ is seen to be an illusion brought about by
the custom of using CV%.
References are:
1. Jelliffe RW, Schumitzky A, Van Guilder M, Liu M, Hu L, Maire P, Gomis P, Barbaut X, and Tahani
B: Individualizing Drug Dosage Regimens: Roles of Population Pharmacokinetic and Dynamic Models,
Bayesian Fitting, and Adaptive Control. Therapeutic Drug Monitoring, 15: 380-393, 1993.
2. De Groot M: Probability and Statistics, 2nd ed. 1986, p. 403.
Hope this helps,
Roger Jelliffe
Back to the Top
No. Being less than the LLOQ is not just a matter of failed CV or precision. LLOQ is also a limit
to the accuracy of the method. Without adherence to these simple limits we may as well use a single
point or consult a psychic. Analytical methods are constructed and validated to give some measure
of certainty around the result. It sounds like the method validated was for the higher dose and was
apparently inappropriate for the lower dose. This is curiously similar to the earlier described
case where dilutions above the range were not validated and the tmax and cmax were lost.
As analysts we develop a methods based on discussions with pkists and what they want. We ought
perhaps to discuss more carefully about what they need.
Back to the Top
The solution-given there is sample remaining- would be to revalidate the method to adequately give
values in the terminal phase for each dose group and to capture tmax and cmax using validated
dilutions for the early phase for each group.
Otherwise there is a significant waste of company or taxpayers money, animals, and that most
valuable commodity, time!
Back to the Top
Soujanya,
I had the same experience with a phase I single ascending dose study. I
would advise that, if you can, work with the bioanalytical group to
validate the method for the lower dose concentrations and use backup
aliquots of sample to redo the analysis. This may not be possible for
various reasons:
1. Not possible to push the lower limit of the bioanalytical method any
lower for the compound
2. No backup samples
3. Development team or bioanalytical group is unwilling to expend the
time or resources to redo the analytical work. This no redo decision
may be rationalized as justifible because the lower doses will not be
studied further in the clinic.
When faced the same situation, I reported all calculable half-lives
whether by NCA or compartmental modeling. Then in the report caveat
emptored the heck out of it. Said that at the lower doses the terminal
phase may be "under water" and that the half-lives reported may not
represent the "true" terminal phase.
Dan
Back to the Top
In addition to the approaches suggested by Dan Combs, -if needed the specifications (allowed
precision and accuracy) at the lower range of the method could also be widened and validated.
-Tom
Thomas Tarnowski, Ph.D.
Director, Bioanalytiocal Chemistry
Gilead Sciences, Inc.
|
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@lists.ucdenver.edu with "Reporting half-life" as the subject |
Copyright 1995-2014 David W. A. Bourne (david@boomer.org)