Log in
Register|
neoc ● 2007-12-18 14:13 (6351 d 08:46 ago) Posting: # 1391 Views: 7,827 |
|
|
I am a pharmacy student currently doing an internship and would like to ask a question regarding WinNonLin (currently nobody here has experience with winnonlin, so I'm kind of learning it myself). I want to do noncompartmental PK analysis of a drug (drug "A"). The goal is to see whether or not a second drug (drug "B") shows effect on PK parameters of drug A. The studydrug was administered twice as an IV bolus dose of 1 mg injected over 1-2 minutes, at t=0 min and t=40 min. The half life was estimated at 200 min. On two separate occasions subjects took either only drug A or a combination of drug A and B The question: What kind of modelsettings/dosing regimen is best to choose? The drug is dosed twice, but no steady state concentration is reached. - Should I use an NCA IV bolus model with single dosing (e.g. 2 mg total dose) with time of dosing at t=0. - Should I use the steady state option in the dosing regimen window, using something as: Dose: 1, Tau=40 min, Last dose=0. In this way I can define the multiple dosing, however, no steady state is reached, so I feel the first option is better. - Or something else? Thanks in advance for any tips! |
|
Helmut ★★★   Vienna, Austria, 2007-12-18 16:13 (6351 d 06:46 ago) @ neoc Posting: # 1392 Views: 6,717 |
|
|
Dear neoc! ❝ I want to do noncompartmental PK analysis of a drug (drug "A"). ❝ The goal is to see whether or not a second drug (drug "B") shows effect on ❝ PK parameters of drug A. ❝ The studydrug was administered twice as an IV bolus dose of 1 mg injected ❝ over 1-2 minutes, at t=0 min and t=40 min. The half life was estimated at ❝ 200 min. On two separate occasions subjects took either only drug A or a ❝ combination of drug A and B ❝ ❝ The question: What kind of modelsettings/dosing regimen is best to choose? ❝ ❝ The drug is dosed twice, but no steady state concentration is reached. Were there any particular reasons for this design? To perform a PK interaction study (which is by design and it's evaluation basically a BE study), we should look either at single dose or steady state, although according to the guidelines a steady state study is either mandatory (EU-EMEA) or should be considered (US-FDA Actually right now you are assessing just the beginning of the saturation phase - which my vary between subjects... ❝ - Should I use an NCA IV bolus model with single dosing (e.g. 2 mg total ❝ dose) with time of dosing at t=0. ❝ - Should I use the steady state option in the dosing regimen window, using ❝ something as: Dose: 1, Tau=40 min, Last dose=0. In this way I can define ❝ the multiple dosing, however, no steady state is reached, so I feel the ❝ first option is better. Actually in WinNonlin you have only one option, namely the wrong one. If you go with the SS option (which would allow you to input correct doses and tau), you will end up with AUCtau (in WinNonlin's output given as AUClast = AUC from t=tau to t=last) only. Even AUCinf will be wrong, because the first profile will be simply ignored in the calculations. In order to calculate the AUC over both profiles, you will have to uncheck the [x] Steady state option in the Dose regimen tab of NCA Model 201. You may input any value for dose (acting just as a dummy); in the Parameter names tab of the Model Options for NCA Model 201 change all parameters which will use the dose in calculations (Cmax_D, C0, AUCINF_D_obs, AUCINF_D_pred, Vz_pred, Cl_pred, Vss_obs, Vss_pred) in column 'Include in Workbook' from 'Yes' to 'No'. These parameters depend on a correct dose definition - which is not true. Attention: these settings apply only to the workbook, but are still given in the NCA Text file - don't use them! I'm afraid you have to struggle with the evaluation of a flawed design. The 'correct' evaluation would be assessment of the first profile only - but probably you have just a few data points there... Edit: Links corrected for FDA’s new site. [Helmut] — Dif-tor heh smusma 🖖🏼 Довге життя Україна! ![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png) Helmut Schütz ![[image]](https://static.bebac.at/img/CC by.png) The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
neoc ● 2007-12-20 12:03 (6349 d 10:56 ago) @ Helmut Posting: # 1402 Views: 6,416 |
|
|
Thanks for your reply!! ❝ Were there any particular reasons for this design? The only reason for this strange dosing regimen I could find was safety: in case too many side effects occurred in the first dosing, the second dosing would not be administered (it was an ACh inhibitor). I have still only little knowledge on trial design buy I think an IV infusion over 30-60 minutes would have been better. If side effects would become to strong, drug infusion could always be terminated. ❝ To perform a PK interaction study (which is by design and it's ❝ evaluation basically a BE study), we should look either at single ❝ dose or steady state, although according to the guidelines ❝ a steady state study is either mandatory (EU-EMEA) or should ❝ be considered (US-FDA 1999, 2006) Actually the primary goal was to asses the efficacy of additional administration of drug B on the already proven effect of drug A. For sake of simplicity I left this out. I think the NCA PK analysis can be seen as post-hoc, and therefore not explicitly mandatory. ❝ Actually in WinNonlin you have only one option, namely the wrong one. ❝ If you go with the SS option (which would allow you to input correct doses ❝ and tau), you will end up with AUCtau (in WinNonlin's output ❝ given as AUClast = AUC from t=tau to t=last) only. Even AUCinf will be ❝ wrong, because the first profile will be simply ignored in the ❝ calculations. ❝ In order to calculate the AUC over both profiles, you will have to uncheck ❝ the [x] Steady state option in the Dose regimen tab of NCA Model 201. You ❝ may input any value for dose (acting just as a dummy); Currently I choose to use the NCA 201 model without SS. For dosing I entered the total dose of 2 mg (2x1mg) with dosing at t=0. The main parameters I'll have to calculate are Thalf, Cmax and Clearance. Clearance is calculated from AUCinf by Cl=D/AUCinf, and Thalf is based on the curve fitting part (Lambda_Z). When choosing for NCA 201 *with* SS, WinNonLin gives me the same values for AUCinf (seems logical), and therefore actually there is no change in the final parameters I need. (If my method is correct off course!). However, someone else (which I cannot reach right now) wants to use my WinNonLin modeling data for doing some compartmental analysis in NONMEM. Does it matter which method (SS or no SS) I use? E.g. I don't know if the specific SS parameters such as Clss, AUC_Tau etc. would be needed for a compartmental analysis. I would guess some estimates of Lambda_Z and AUC_inf might come in handy for the NONMEM analysis, but I doubt parameters such as Clss will, because no actual SS is reached yet probably. I would appreciate it if you (HS) or someone else would be able to clarify this last hesitation!  Thank you! |
Ing. Helmut Schütz