Dear Livan,

(A) Do you want to know how to mathematically perform the deconvolution? How was your data collected? At fixed time intervals or not? Do you know your "unit input function " ie the system response to a "1 unit" input?

(B) Or you do want to convolve the drug product dissolution profile with some (known) UIF in order to predict the drug plasmatic profile (ie, the convolution outcome)?

From your question title I would say you mean (A) but taking into consideration the description I guess it's (B). Can you please clarify?.

BR Luis

Dear Dr. Gouveia,

Thank you very much for your answer, I fact, I would like to convolve the drug product by dissolution profile in order to predict the drug plasmatic profile in vivo.

Thanks in advance

Dear Livan, as you know the in-vitro drug product dissolution profile does not necessarily correlates with the in-vivo drug release and made available for the steps ahead: absorption, distribution, metabolization and excretion.

If you have experimental data on plasmatic drug concentration following the administration of similar tablets (same drug) and the corresponding in-vitro dissolution profiles you may try to predict (after making a "good" number of assumptions) the plasma concentration. You also need to write down/guess/estimate the underlying physiological/pharmacokinetic mechanistic model: what are the steps involving drug mass transport, what kind of kinetics is driving that transport (linear, non-linear, saturable or not) etc. Enough experimental data (some may be found in the literature) may allow to derive the model, to establish the critical/limiting steps and to "validate" via a series of consecutive convolutions the developed model.

If you succeed to achieve a suitable prediction accuracy you can then proceed with your own tablets data.

For what I've seen and dealt with so far it is not an easy task, although quite rewarding when one is able to compile all the data and create the model.

Obviously, you may address the problem via an empirical (qualitative or semi-quantitaive) approach but it depends a lot on the drug physicochemical characteristics (including but not limited to solubility and permeability).

Dear Dr. Gouveia,

Thank yoiu very much for your response and useful comments.

Best regards