Wendy writes:

"I'd like to use Mirapex in the graph if I can. Its TMAX is 2 hrs, THALF is 8-12 hrs. depending on age, and the conversion factor is 100. ... How do I figure out the CMAX?"
Inhaler for Levadopa

Unfortunately my program does not include Mirapex. But, given the interest shown I'll extend it to include this drug and a few other common ones too. The difficulty is not in the programming, but rather finding reliable sources of the pharmacokinetic parameters on which the model is based. This will take some time to do, so in the meantime a work-around is required. I will go into a lot of detail, because I think it important that anyone using this tool understands how it works.

The simplest, but least accurate way, is to take a similar drug and use that as a proxy. In this case, ropinirole is a possible alternative. The conversion factor for levodopa is 1, for ropinirole it is 20 and for Mirapex it is 100. To balance the books you can select ropinirole and give it a dose of 5 times more than the Mirapex dose. So, a 1mg dose of Mirapex is entered as a 5mg dose of ropinirole. If you enter these values into the grid and also enter a line showing a dose of 100mg levodopa, you will find that as expected the two doses have almost the same area under the curve (AUC). The problem is that the program has chosen a TMAX of 90 minutes (when you wanted 120) and a THALF of 360 minutes (when you wanted 480-720 minutes).

A more accurate way to work around the problem is to change the default values of TMAX and THALF to those that you want, but initially keep the default value of CMAX (5.59). Putting TMAX to 120 and THALF to 600, and pressing Calculate gives an AUC of 20602, which is wrong: it should equal 14686 (because 5mg of ropinirole is equivalent to 100mg of levodopa, and LED is based on equal AUCs). The final step is to by trial and error adjust the CMAX value to get the correct AUC: there is a linear relationship between CMAX and AUC). CMAX=4 gives AUC=14742, which is close enough.

Please remember that this is just a mathematical model, which makes numerous assumptions, which is based on data which is inexact, and which does not take into account endogenous dopamine production and storage. Also, it is one step removed from what we really want to know, which is the pharmacodynamic impact of our drug regimen: how, for instance, does our bradykinesia change during the day.

My efforts are going into "dynamic dosing": producing electronic tools that detect when you're nearing an "off" with enough notice for a dose taken now to kick-in before the "off" happens.

John