NONMEM Users Network Archive

Hosted by Cognigen

[NMusers] RE: Using MCP-MOD in dose finding for Phase 3

From: Steimer, Jean-Louis <jean-louis.steimer_at_novartis.com>
Date: Mon, 23 Mar 2015 12:54:04 +0000

Dear Nele, Dear all,
Below in red in Nele's e-mail, you will find the input of Bjoern Bornkamp, =
a statistician from the Novartis Stats/Methods group. I forwarded your mail=
 to him. Bjoern was involved in the qualification discussion with EMA toget=
her with Jose Pinheiro and Frank Bretz. He is one of the implementers of th=
e MCP-Mod methodology within Novartis, and applies it routinely in Phase 2 =
studies.
I am sure that Bjoern' answers will help.
Bye
Jean-Louis Steimer

+++++++++++++++++++++++
Dear all,

I am writing to you as we are currently discussing the implementation of th=
e MCP-MOD approach for dose finding based on Phase 2B results and would lik=
e to hear your opinion on this approach.
Original MCP-Mod is not intended to be used in Ph III, special adaptations =
are necessary (closed testing).
It would be good to get feedback from both statisticians and classical mode=
lers.
I have thought about the approach, and have a few problems about seeing the=
 advantage of the approach over complete population-PK/PD modeling.
These are two different approaches that complement each other. MCP-Mod is n=
ot intended to replace population-PK/PD modeling (the idea is to replace AN=
OVA-type models).
I can see benefits to do both a simple cross-sectional dose-response analys=
is (like MCP-Mod) and a complete dose-exposure-response characterization.
If results are consistent between both approaches one would have more confi=
dence overall in the analysis results than from either analysis alone.
If results are not consistent one needs to dig a bit "deeper", but this is =
also useful information.
From what I understood, I can see the following issues:
MCP-MOD

· Only uses trial endpoints, i.e. it ignores the time course of t=
he treatment effect. I have a problem with this because there might be nois=
e in the endpoint (e.g. if the effect has reached a plateau), which might p=
otentially lead to the selection of the wrong model structure. Including th=
e time-course like in PKPD modeling approaches would detect that the deviat=
ion is just noise, and thus probably be able to identify the right model st=
ructure despite this.
MCP-Mod can handle longitudinal data, see Pinheiro et al. (2014), Stat Med.=
 33,1646-61 for one example, which is also available in the DoseFinding R p=
ackage.

· Uses dose-response models instead of exposure-response models
Correct. Again, MCP-Mod is not intended to replace population-PK/PD modelin=
g. We have started thinking how to extend the key ideas of MCP-Mod to expos=
ure-response models and encourage the community to look into this.

· Pre-specifies the model structure. While I understand that for =
pivotal trials prespecification is crucial, I would assume that Phase 2 is =
performed to allow exploration of the data to come up with the best model g=
iven the data we have. What happens if the true model is not part of the te=
sted ones? What if we have new physiological insights that tell us about th=
e model structure after we have seen the data? Do we then ignore what we kn=
ow and fit all bad models, and if none gives a good description we do model=
 averaging of bad models?
Excellent questions. Candidate models for MCP-Mod should always be selected=
 based on entire teams input and operating characteristics should be evalua=
ted upfront. More specifically, our experience shows that MCP-Mod is relati=
vely robust if the true model is not part of the tested ones, see for examp=
le Pinheiro et al. (2006), J. Biopharm. Statist. 16,639-656. This is also s=
omething that can be evaluated to some extend upfront (at the design stage)=
 by simulations.
Among other things one advantage of pre-specification is that it makes the =
modelling more transparent/credible for externals (e.g. health authorities)=
, if one specifies before seeing the data what will be done. But of course =
there is a trade-off: Not sure if it is possible to pre-specify a full popu=
lation PK/PD analysis.

· If we include a model with many parameters in the prespecificat=
ion and only have a few dose strength, wouldn't the model with more paramet=
ers be more likely to give a good fit (e.g. when comparing Emax to logistic=
), with the consequence that a wrong dose might be selected?
Not sure whether I fully understand this question. Of course the model-sele=
ction/averaging step of MCP-Mod would take into account the model complexit=
y by using AIC/BIC (not only looking at model fit). Again, operating charac=
teristics need to be evaluated in advance, which include precision of targe=
t dose estimation and also possible convergence problems if the number of p=
arameters is to larger.

Colleagues from statistics recommend to cover all potential models with dif=
ferent shapes in the candidate set to avoid potential bias in dose selectio=
n, but they argue that post-hoc model fitting leads to data-dredging and ov=
er-fitting, does not account for model uncertainty and gives overly-optimis=
tic results. I am wondering however what the difference in the approach is =
if anyway ALL potential models are considered (which can lead to overfittin=
g as well)?
There is a penalty for using many models in MCP-Mod: In the MCP step the mu=
ltiplicity adjustment would get higher if there are more models included (i=
n particular if they are very different).
In the Mod step the variance of the dose-response curve would increase with=
 an increased number of models, so there one faces the usual variance/bias =
trade-off.
Might a good solution be to combine PKPD modeling with MCP-Mod?
Yes, see above

Your opinion will be highly appreciated, and I am looking forward to receiv=
ing comments both in favour and against the approach :-)

Best
Nele

From: owner-nmusers_at_globomaxnm.com [mailto:owner-nmusers_at_globomaxnm.com] On=
 Behalf Of Mueller-Plock, Nele
Sent: Friday, March 20, 2015 1:02 PM
To: nmusers_at_globomaxnm.com
Subject: [NMusers] Using MCP-MOD in dose finding for Phase 3

Dear all,

I am writing to you as we are currently discussing the implementation of th=
e MCP-MOD approach for dose finding based on Phase 2B results and would lik=
e to hear your opinion on this approach. It would be good to get feedback f=
rom both statisticians and classical modelers.
I have thought about the approach, and have a few problems about seeing the=
 advantage of the approach over complete population-PK/PD modeling. From wh=
at I understood, I can see the following issues:
MCP-MOD

· Only uses trial endpoints, i.e. it ignores the time course of t=
he treatment effect. I have a problem with this because there might be nois=
e in the endpoint (e.g. if the effect has reached a plateau), which might p=
otentially lead to the selection of the wrong model structure. Including th=
e time-course like in PKPD modeling approaches would detect that the deviat=
ion is just noise, and thus probably be able to identify the right model st=
ructure despite this.

· Uses dose-response models instead of exposure-response models

· Pre-specifies the model structure. While I understand that for =
pivotal trials prespecification is crucial, I would assume that Phase 2 is =
performed to allow exploration of the data to come up with the best model g=
iven the data we have. What happens if the true model is not part of the te=
sted ones? What if we have new physiological insights that tell us about th=
e model structure after we have seen the data? Do we then ignore what we kn=
ow and fit all bad models, and if none gives a good description we do model=
 averaging of bad models?

· If we include a model with many parameters in the prespecificat=
ion and only have a few dose strength, wouldn't the model with more paramet=
ers be more likely to give a good fit (e.g. when comparing Emax to logistic=
), with the consequence that a wrong dose might be selected?

Colleagues from statistics recommend to cover all potential models with dif=
ferent shapes in the candidate set to avoid potential bias in dose selectio=
n, but they argue that post-hoc model fitting leads to data-dredging and ov=
er-fitting, does not account for model uncertainty and gives overly-optimis=
tic results. I am wondering however what the difference in the approach is =
if anyway ALL potential models are considered (which can lead to overfittin=
g as well)?
Might a good solution be to combine PKPD modeling with MCP-Mod?

Your opinion will be highly appreciated, and I am looking forward to receiv=
ing comments both in favour and against the approach :-)

Best
Nele
______________________________________________________________

Dr. Nele Mueller-Plock, CAPM
Associate Scientific Director Pharmacometrics
Global Pharmacometrics
Translational Medicine

Takeda Pharmaceuticals International GmbH
Thurgauerstrasse 130
8152 Glattpark-Opfikon (Zürich)
Switzerland

Visitor address:
Alpenstrasse 3
8152 Glattpark-Opfikon (Zürich)
Switzerland

Phone: (+41) 44 / 55 51 404
Mobile: (+41) 79 / 654 33 99

mailto: nele.mueller-plock_at_takeda.com<mailto:nele.kaessner_at_nycomed.com>
http://www.takeda.com<http://www.takeda.com/>


--------------------------------------------------------------------



The content of this email and of any files transmitted may contain confiden=
tial, proprietary or legally privileged information and is intended solely =
for the use of the person/s or entity/ies to whom it is addressed. If you h=
ave received this email in error you have no permission whatsoever to use, =
copy, disclose or forward all or any of its contents. Please immediately no=
tify the sender and thereafter delete this email and any attachments.



--------------------------------------------------------------------



Received on Mon Mar 23 2015 - 08:54:04 EDT

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to: nmusers-request_at_iconplc.com. Once subscribed, you may contribute to the discussion by emailing: nmusers@globomaxnm.com.