cakhatri
★    

India,
2013-11-24 11:13
(4178 d 20:13 ago)

Posting: # 11951
Views: 15,974
 

 Simulation of data to predict Steady state [General Sta­tis­tics]

Post reply
Dear members,
We have completed a pilot study on a modified formulation to be marketed for EU submission. Is there any way of simulating the data to predict steady state from the pilot study data?
Regards
Chirag
 
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2013-11-24 15:30
(4178 d 15:56 ago)

@ cakhatri
Posting: # 11952
Views: 16,028
 

 Nonparametric superposition

Post reply
Hi Chirag,

❝ Is there any way of simulating the data to predict steady state from the pilot study data?


Two ways.
  1. PK modeling.
  2. Nonparametric superposition.
For both you need a reliable estimate of the slowest phase (elimination or absorption if ka < kel) and have to assume:
  • Independence of doses (no auto-induction or -inhibition).
  • identical rate and extent of absorption, and
  • identical clearances.
#1 is tricky. Therefore, I would suggest #2.
  • At the multiples of τ stack the single dose profile on top of remaining concentrations.
  • Add the estimated time course of the slowest phase of the preceding profiles.

    [image]

  • Don’t use mean values. In order to plan for the calibration range of the analytical method in the multiple dose study perform the superposition of the most “extreme” subjects of the pilot study. Allow for some “headroom” on both sides. In the example I would suggest LLOQ ≤20 and ULOQ ≥300.

    [image]

For an example in WinNonlin 5.x see this post. Easier in Phoenix.

[image]


BTW, look at the steady state profiles and the inserted plot. Compared to single dose tmax slightly shifts towards earlier times. Sometimes it makes sense not to blindly use the single dose’s sampling time schedule, but to adjust it accordingly.
—
Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
 
jag009
★★★

NJ,
2013-11-25 00:13
(4178 d 07:13 ago)

@ Helmut
Posting: # 11954
Views: 14,715
 

 Nonparametric superposition

Post reply
Hi Helmut!

Interesting topic! I once attempted to simulate carbamazepine (autoinduction) to steady state. Fun stuff...

John
 
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2013-11-25 03:56
(4178 d 03:30 ago)

@ jag009
Posting: # 11955
Views: 14,963
 

 Nonlinear nightmare

Post reply
Hi John!

❝ I once attempted to simulate carbamazepine (autoinduction) to steady state. Fun stuff...


One of László’s favorites. ;-)
Last June after his recent paper* I played around with different types of induction. That’s a 400 mg single dose (Vd 83.5 L, F 1, ka 0.472 h-1 (R), 0.94 h-1 (T1), 3.6 h-1 (T2), non-induced clearance 1.35 L/h, fully induced 5.84 L/h):

[image]

Nice shapes. Steady state was demanding and I put it on the to-do list…

  • Tóthfálusi L, Endrényi L. Approvable generic carbamazepine formulations may not be bioequivalent in target patient populations. Int J Clin Pharmacol Ther. 2013;51(6): 525–8. doi 10.5414/CP201845.
—
Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
 
jag009
★★★

NJ,
2013-11-25 18:43
(4177 d 12:43 ago)

@ Helmut
Posting: # 11963
Views: 14,742
 

 Nonlinear nightmare

Post reply
Hi Helmut!

Goodie. Have you goofed around with DDI before (drug interaction->inhibition)?

You can't run carbamazepine single dose BE due to autoinduction. We ran the BE program before and went with multiple-dose studies with no washout period like 8 yrs ago. I hope no one did...

I don't have Laszlo's paper so I don't know the details. i will ask him/them for one since both Ls are my former prof and postdoc. The Hungarian Dynmaic Duos :-)

John
 
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2013-11-26 15:32
(4176 d 15:54 ago)

@ jag009
Posting: # 11972
Views: 14,802
 

 Nonlinear nightmare

Post reply
Hi John,

❝ Have you goofed around with DDI before (drug interaction->inhibition)?


Yes, but always in steady state and not with CBZ.

❝ You can't run carbamazepine single dose BE due to autoinduction.


No? Single dose studies are recommended in all of FDA’s carbamazepine-guidances. I don’t think that full autoinduction is seen in the first dose. It will take the enzymes a while to tune up. That’s why I started to play around with time-dependent increase of clearance.

Regulators differ in their opinions whether CBZ is an NTID. Even if the conventional 80–125% are applicable to Cmax and AUC, pAUC would be a sensitive PK metric being able to discriminate between formulations with different absorption rates.

❝ We ran the BE program before and went with multiple-dose studies with no washout period like 8 yrs ago. I hope no one did...


Oh boy…

❝ I don't have Laszlo's paper so I don't know the details.


You’ve got mail. ;-)
—
Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
 
jag009
★★★

NJ,
2013-11-26 17:00
(4176 d 14:26 ago)

@ Helmut
Posting: # 11974
Views: 14,600
 

 Nonlinear nightmare

Post reply
Hi Helmut,

❝ No? Single dose studies are recommended...


Well we did it 8 yrs ago and there was no guidance on ER...

❝ ❝ We ran the BE program before and went with multiple-dose studies with no washout period like 8 yrs ago. I hope no one did...

❝

❝ Oh boy…


Yes, before guidance showed up...

❝ You’ve got mail. ;-)


Thanks :-)

By the way, which part of Austria are you located (Vienna, right?)? Tourist district? I might go visit Euro sometime during winter season so maybe I can meet the great swordsman (pirate?)

John
 
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2013-11-26 17:27
(4176 d 13:59 ago)

@ jag009
Posting: # 11975
Views: 14,787
 

 OT: Vienna

Post reply
Hi John,

❝ By the way, which part of Austria are you located (Vienna, right?)?


Correct.

❝ Tourist district?


Absolutely. Check it out.

❝ I might go visit Euro sometime during winter season…


You are into skiing, right?

❝ …so maybe I can meet the great swordsman (pirate?)


Sure, I can play the tour-guide – though visiting Vienna in Winter is not the best season (can be f**g cold; it started snowing last night).
—
Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
 
jag009
★★★

NJ,
2013-11-26 18:29
(4176 d 12:57 ago)

@ Helmut
Posting: # 11976
Views: 14,664
 

 OT: Vienna

Post reply
Hi Helmut!

❝ ❝ I might go visit Euro sometime during winter season…

❝

❝ You are into skiing, right?


Yup! Been to Zermatt and some other parts of Switzerland (late season skiing, cheaper).

❝ ❝ …so maybe I can meet the great swordsman (pirate?)

❝

❝ Sure, I can play the tour-guide – though visiting Vienna in Winter is not the best season (can be f**g cold; it started snowing last night).


I used to live in Toronto Canada (which has the so called great lake effect) so -xx degree Celsius is not a big issue for me :cool:

John
 
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2013-11-27 14:04
(4175 d 17:22 ago)

@ jag009
Posting: # 11979
Views: 14,618
 

 OT: Vienna

Post reply
Hi John!

❝ Yup! Been to Zermatt and some other parts of Switzerland (late season skiing, cheaper).


Don’t forget nicer weather!

❝ […] so -xx degree Celsius is not a big issue for me :cool:


Well, have mercy with your tour-guide!
—
Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
 
jag009
★★★

NJ,
2013-11-25 18:48
(4177 d 12:38 ago)

@ Helmut
Posting: # 11964
Views: 14,615
 

 Conc at dosing interval

Post reply
Hi Helmut,

I sometime use the following too:
Compare the concentration at 24 hours (or the dosing interval for the product) between the test and reference. Yes for BE the difference in general is not huge but it can affect the MD BE outcome with respect to Cmax.

John
 
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2013-11-26 16:01
(4176 d 15:25 ago)

@ jag009
Posting: # 11973
Views: 14,891
 

 Conc at dosing interval

Post reply
Hi John,

❝ Compare the concentration at 24 hours (or the dosing interval for the product) between the test and reference.


Yes, that’s a nice one! Currently for EMA BE of MR products has to be demonstrated both after SD and in steady state. Paixão et al.* showed that for drugs with linear PK Cτ is predictive of steady state. We will see whether the mighty oracle will accept this approach and drop the requirement of MD studies in the final guideline.

  • Paixão P, Gouveia LF, Morais JAG. An alternative single dose parameter to avoid the need for steady-state studies on oral extended-release drug products. Eur J Pharmaceut Biopharmaceut. 2012; 80(2): 410–7. doi:10.1016/j.ejpb.2011.11.001.

    In October 2012 the authors received the AAPS’
    “Outstanding Manuscript Award in Modeling and Simulation”.
    [image]
    (Paulo to the right)
—
Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
 
jag009
★★★

NJ,
2013-11-26 18:31
(4176 d 12:55 ago)

@ Helmut
Posting: # 11977
Views: 14,570
 

 Conc at dosing interval

Post reply
Hi Helmut,

❝ Yes, that’s a nice one! Currently for EMA BE of MR products has to be demonstrated both after SD and in steady state. Paixão et al.1 showed that for drugs with linear PK Cτ is predictive of steady state.2 We will see whether the mighty oracle will accept this approach and drop the requirement of MD studies in the final guideline.


I never agree to the cancelling of steady state study by FDA because there are some cases where SD study is not entirely sensitive...

John
 
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2013-11-27 14:10
(4175 d 17:16 ago)

@ jag009
Posting: # 11980
Views: 14,514
 

 Conc at dosing interval

Post reply
Hi John!

❝ I never agree to the cancelling of steady state study by FDA because there are some cases where SD study is not entirely sensitive...


Do you have any examples (except drugs with nonlinear PK) where SD passed and MD failed? See also this post.
—
Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
 
jag009
★★★

NJ,
2013-11-27 21:00
(4175 d 10:25 ago)

@ Helmut
Posting: # 11984
Views: 14,409
 

 Conc at dosing interval

Post reply
Hi Helmut,

❝ Do you have any examples (except drugs with nonlinear PK) where SD passed and MD failed? See also this post.


Or vice versa. Let me look around, the findings were from my previous company so I dont even know if I have them... It's not about nonlinear PK.

John
 
luvblooms
★★  

India,
2013-11-28 10:19
(4174 d 21:07 ago)

@ Helmut
Posting: # 11987
Views: 14,614
 

 Passing SD and Failing MD

Post reply
Hi HS

We have done a couple of studies for a molecule (BCS class II molecule, solubility lesser at higher pH, almost insoluble at pH 6 and above, MR pellets) .

In SD we got results like this
Cmax: 97% (87- 108)
AUC: 89 (82-96)

In MD, product failed in Cmin ss and AUCTss.

Exact values I do not remember, I need to look my old book for the same.
If you need details let me know, I will look for the same.

Regards
—
~A happy Soul~
 
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2013-11-29 18:29
(4173 d 12:57 ago)

@ luvblooms
Posting: # 11990
Views: 14,526
 

 Passing SD and Failing MD

Post reply
Hi Luv,

❝ We have done a couple of studies for a molecule (BCS class II molecule, solubility lesser at higher pH, almost insoluble at pH 6 and above, MR pellets) .

❝

❝ In MD, product failed in Cmin ss and AUCTss.


That’s an interesting story! Were the PEs in the MD study shifted to the same direction (i.e., were in MD the ratios also <100%)? How did the CVs “behave”?

The only case I know from the literature is quite old.* The authors were disappointed that the CVs did not substantially decrease in MD (at that time MD was an accepted method to assess BE of HVDPs). On the other hand their single dose studies were amazingly small (n=4 and n=5)…

  • Van Hoogdalem EJ, Terpstra IJ, Krauwinkel WJJ, Volkers-Kamermans NJ, Baven, ALM, Verschoor JSC. Multiple dose bioequivalence study with josamycin propionate, a drug with highly variable kinetics, in healthy volunteers. Int J Clin Pharmacol Ther. 1996;34(5):202–7.
—
Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
 
luvblooms
★★  

India,
2013-11-29 19:57
(4173 d 11:29 ago)

@ Helmut
Posting: # 11991
Views: 14,503
 

 Passing SD and Failing MD

Post reply
Hi HS,

❝ That’s an interesting story! Were the PEs in the MD study shifted to the same direction (i.e., were in MD the ratios also <100%)? How did the CVs “behave”?


Yes, the ratio's were on lower side.The T/R ratio was ~87 for Cmin (lower CI was 76.8) and Same was the case for AUCtss (~89 with range of 78-99).

CVs were almost similar to that of SD.

When we were trying to understand the reason of failure, one thing that draw our attention was the lower AUC values in SD. And this lower accumulations might have lead to the lower Cmin and AUCtss.

Further modified the formula in order to achieve AUC (by increasing its solubility in lower GIT) and dosed it for SD and BINGO!!! Cmax ~96% AUC= 95 (still was not able to reach 100). When the same was dosed for MD, it passed with flying colors ;-) (Cmin and AUCtss ratios were 93 and 94 respectively)

❝ The only case I know from the literature is quite old.*

❝

  • Van Hoogdalem EJ, Terpstra IJ, Krauwinkel WJJ, Volkers-Kamermans NJ, Baven, ALM, Verschoor JSC. Multiple dose bioequivalence study with josamycin propionate, a drug with highly variable kinetics, in healthy volunteers. Int J Clin Pharmacol Ther. 1996;34(5):202–7.

Read this article long back, need to refresh the memory :-D

Regards
—
~A happy Soul~
 
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2013-11-30 18:25
(4172 d 13:01 ago)

@ luvblooms
Posting: # 11993
Views: 14,701
 

 Cτ predictive of Cmin

Post reply
Hi Luv,

❝ Yes, the ratio's were on lower side.The T/R ratio was ~87 for Cmin (lower CI was 76.8) and Same was the case for AUCtss (~89 with range of 78-99).

❝

❝ When we were trying to understand the reason of failure, one thing that draw our attention was the lower AUC values in SD. And this lower accumulations might have lead to the lower Cmin and AUCtss.


I tried to simulate some stuff matching your SD results (AUC 89%, Cmax 97%) and MD (AUC 89%). The best I could get (assuming identical t1/2) was F 79.44% combined with faster absorption. Example:

[image]

    Test    Reference
F%  79.44   100.00   
ka   0.3466   0.4578

               %T/R   
             SD     MD
AUCt        90.1       
AUC∞        89.0       
AUCτ               89.1
Cmax        97.0   95.1
Cmax/AUC∞  108.9       
Cmax/AUCτ         107.7
Cτ          79.7       
Cmin               79.7

The faster absorption of the test counteracts its lower bioavailability (Cmax-ratio > AUC-ratio). Since Cmax is influenced by extent of absorption, Cmax/AUC tells a better story – although ka of the test is actually by 32% faster. Cτ is predictive of Cmin. However, if I want to get your 87% I would have to assume different half-lives. Were you close to flip-flop PK? Did you have any hints of nonlinear PK?
—
Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
 
UA Flag
Activity
 Admin contact
23,424 posts in 4,927 threads, 1,667 registered users;
83 visitors (0 registered, 83 guests [including 7 identified bots]).
Forum time: 08:26 CEST (Europe/Vienna)

Patients may recover in spite of drugs or because of them.    John Gaddum

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5