Re: [NMusers] unbalanced data set

From: Denney, William S. <>
Date: Wed, 6 Jan 2016 13:33:27 +0000

Hi Zheng,

I'll take an intermediate view between Joachim and Nick.

The rich data from Phase 1 provides the ability to define the structural mo=
del and a few of the important covariates. The control of Phase 1 gives pr=
ecision that cannot be achieved in Phase 2 or 3 studies. But, there are us=
ually important differences between Phase 1 and later phase populations tha=
t makes the later phase separately important.

With later phase trials, the range of covariates is expanded [1]. On top o=
f the expanded covariate range, sometimes late-phase patient populations ar=
e categorically different than early phase [2].

In practice, this means that I fit a single model to all data. The model w=
ill allow for the dense data from Phase 1 with more inter-individual variab=
ility (IIV) terms (fix the IIV to 0 for sparse data) and the expanded covar=
iate range with a richer set of fixed effects as the model is expanded for =
later phase. Finally, due to typical differences in data quality, I will o=
ften include a different residual error structure for sparse data. This ap=
proach allows the complexity of the Phase 1 structural model to carry into =
the richness of the late phase covariate model.

[1] A specific example is that typically renal function is allowed to be lo=
wer especially when Phase 1 is in healthy subjects.
[2] My true belief is that there may be unobserved covariates causing what =
appears to be a categorical difference. The functional impact of that beli=
ef is semantic only. In practice, the model would include a categorical pa=



On Jan 6, 2016, at 4:09, "Joachim Grevel" <<mailto:jgr=>> wrote:

Dear Zheng,
This is indeed a fundamental and recurring problem in drug development. You=
 have rich data from Phase 1 studies (single ascending dose, multiple ascen=
ding dose, others e.g. QTc) and sparse data from Phase 3 studies. Should yo=
u mix them all in one large meta-analysis and derive the definitive popPK m=
odel for that drug/project?
After years of experience, I tend to not mix Phase 1 with Phase 3 data. Pha=
se 1 can be used to establish the first popPK model which may contain speci=
al features such as nonlinearities/saturation effects as a consequence of t=
he wide range of doses studied. This can be the starting point for the buil=
ding of a fit-for purpose model using Phase 3 data only. I have come to bel=
ieve that the specific patient population(s) of Phase 3 require their own p=
opPK model that predicts exposure without bias. This is then used in the ex=
posure-response (E-R) modelling that is important for market approval. Only=
 a dedicated Phase 3 popPK model, that does not carry unnecessary legacies =
of Phase 1 development, is fit for E-R modelling and can give the important=
 answers about the dose rate(s) to be put in the drug label.

I would be interested to hear some other opinions.

Good luck,

Joachim Grevel, PhD
Scientific Director
BAST Inc Limited
Science & Enterprise Park
Loughborough University
Loughborough, LE11 3AQ
United Kingdom

Tel: +44 (0)1509 222908<

From:<> [ma=] On Behalf Of Zheng Liu
Sent: 06 January 2016 02:03
Subject: [NMusers] unbalanced data set

Dear all,

I recently have a data set for pk parameters fitting. The issue is some pat=
ients have far more measurement points than others (i.e. a few patients hav=
e ~15 points, other patients have only 1 or 2). I speculate in the fitted p=
arameters, those patients with many points would contribute much more than =
those with less points. Then the population "average" values of fitted pk p=
arameters are not anymore average from all the patients, but more biased to=
 those patients with many points. This is not what I expect.

Of course I could take away some points from the patients with many points,=
 in order to be comparable to less-points patients. Then I will be forced =
to lose some information from the data set. I just wonder are there anyone =
who have better proposal to solve this problem? I appreciate your help very=

Best regards,


Received on Wed Jan 06 2016 - 08:33:27 EST

This archive was generated by hypermail 2.3.0 : Fri Sep 27 2019 - 16:48:17 EDT