Back to the Top
Dear David:
I often get the general impression that Clinical
Pharmacology is now usually regarded as a field in which one largely
is trained to work for the drug industry, and that the clinical part
of it is usually devoted to designing and managing clinical trials.
For example, subjects usually considered in Clinical
Pharmacology training often seem to cover clinical pharmacokinetic/
pharmacogenomic study design, methodology, data analysis and
presentation. One might also cover drug metabolism/transport
investigations in addition to developing and initiating individual
research projects. One is urged to gain expertise in pharmacokinetic
data analysis (using software programs such as WinNonlin; SAAM II
etc.), statistical analysis, scientific writing, and grantsmanship.
One is also urged to become skilled in drug assay procedures and
advanced coursework in pharmacokinetics, statistics, pharmacology, and
genetics.
Is this really all there is to Clinical Pharmacology? Is it
really only data analysis and parametric population modeling for the
drug industry? I usually do not see any consideration of real clinical
application of the PK/PD models, for example, only to developing
various clinical research projects such as clinical trials
I would like to ask all of you also to consider the
application of these efforts in Clinical Pharmacology to real patient
care, which is real Clinical Pharmacology in my view. I would like to
ask you to consider the following additional areas of training:
1. What are the strengths and weaknesses of parametric versus
nonparametric population modeling approaches? What are the relative
capabilities of the two approaches?
2. What approaches have consistency, the property that as you study
more patients, the parameter estimates get closer to the truth? What
approaches do not have this property, and why?
3. How do these two approaches lead to the ability to discover
previously unsuspected subpopulations?
4. How do you evaluate the expected precision with which a stated
dosage regimen will hit the desired therapeutic target selected by the
clinician?
5. What type of dosage design optimizes the precision of the initial
dosage regimen for an individual patient, which is usually based on
data from a genetically polymorphic population, and how is this done?
6. There are now 4 types of Bayesian analysis for obtaining updated
models for individual patients after beginning the initial regimen,
based on therapeutic drug monitoring.
I usually do not see any consideration of TDM or optimized
individualized patient care in clinical pharmacology any more. Where
will physicians, clinical pharmacologists and clinical pharmacists get
this training? For sure the drug industry doesn't want it!
The four Bayesian methods are
A. Conventional parametric MAP Bayesian analysis
B. Nonparametric Bayesian analysis using nonparametric
population models as the prior.
C. Hybrid Bayesian analysis combining these two approaches
D. Interacting Multiple Model (IMM) sequential Bayesian
analysis, in which the Bayesian posterior parameter values are not
fixed throughout the period of data analysis, but instead can change.
This method has now tracked the behavior of gentamicin and vancomycin
better that any other of the methods in unstable patients in the post
coronary surgery unit in Glasgow. A paper by Dr. Alison Thomson is in
press in Therapeutic Drug Monitoring. The IMM method was described in
Bayard D, and Jelliffe R: A Bayesian Approach to Tracking Patients
having Changing Pharmacokinetic Parameters. J. Pharmacokin.
Pharmacodyn. 31 (1): 75-107, 2004. Again, how do you optimize the
precision of new adjusted regimen?
Another relevant reference is Jelliffe R, Bayard D, Milman M,
Van Guilder M, Schumitzky A. Achieving target goals most precisely
using nonparametric compartmental models and "multiple model" design
of dosage regimens. Ther Drug Monit. 2000; 22(3):346-353.
7. What is the best way to weight the lab assay error? And the
remaining environmental error? How does one evaluate the relative
contributions of the noise due to the lab error, and that of the
environment?
I would suggest that these are useful things to consider as
part of training in Clinical Pharmacology, where individual patient
care is still the most important product we deliver. We thought so in
the past, but we seem to have forgotten about the real care of the
patient recently. What most "clinical pharmacologists" now call
individualized regimens are simply the initial regimen based on the
pop model. That is clearly not enough. The real job of a Clinical
Pharmacologist treating a patient just begins when the drug industry's
job ends.
Very best regards,
Roger Jelliffe
Roger W. Jelliffe, M.D. Professor of Medicine,
Division of Geriatric Medicine,
Laboratory of Applied Pharmacokinetics,
USC Keck School of Medicine
2250 Alcazar St, Los Angeles CA 90033, USA
Phone (323)442-1300, fax (323)442-1302, email= jelliffe.-at-.usc.edu
Our web site= http://www.lapk.org
[I've interacted with just a few 'Clinical Pharmacologists' (MD) over
the years and they seem to have another medical specialty (Cardiology,
Neurology, etc.) but a real interest in understanding PK and a
willingness to collaborate with Ph.D.'s to conducted some clinical/
basic research. Their primary jobs was the speciality though, not
Clinical Pharmacology. As a pharmacy educator I had hoped that the
'Pharm.D.' practitioner might provide the clinical PK science with TDM
services etc. but that doesn't seem to have happened... and I'd almost
given up until a recent out of state visit with some interesting
pharmaceutical care providers. Maybe it is happening... db]
Back to the Top
The following message was posted to: PharmPK Dear Roger:
Wise words, as always. I think that you are correct in your
assessment of a decreasing willingness of Pharma to support Phase II &
III studies on TDM. One can argue that Phase I PK and Pre-clinical
xenograft PD studies could be used to interpolate desired plasma and
tissue target concentrations and required doses, but my sense is that
this is not being supported by companies in their development plans.
Clearly, anything that complicates the use of a drug (such as TDM) is
a concern to Pharma.
We struggle to attract and train clinical pharmacologists to the
academic center. At the UWisc, we are developing a new graduate
program in translational (clinical) research through our new CTSA
(formerly GCRC), and hope to meld the Ph.D. into MD and PharmD
programs. We'll see if it works, but your suggestions below are
useful additions to the scope of their training.
Best wishes.
Paul
--
Paul R. Hutson, Pharm.D.
Associate Professor
UW School of Pharmacy
777 Highland Avenue
Madison WI 53705-2222
Back to the Top
Dear Roger and Forum,
I heartily agree that "What is Clinical Pharmacology" is indeed a
worthy question for debate and thankyou for raising it. During my 30
years as an industrial clinical pharmacologist I have often wondered
why there is so little contact between my side of the fence and
academic or clinical clinical pharmacology. The clinician's prime role
was brought home to me when one of the most prestigious hospitals in
London, the Hammersmith, nearly killed my mother by giving her digoxin
and clarithromycin together..she nearly died of digoxin-induced
vomiting. Arguably it is the prescriber's role to know about
interactions like these but with so many drugs in use now that may not
be a realistic attitude and large hospitals using unusual treatments
really do need a CP watchdog for them.
Then I think I have two examples of my own crossing the fence from
trials to clinical practice that might be relevant to this discussion
as they might show how an Industrial CP has a duty to be concerned
with clinical practice. One involves patients in pain who have stress-
related myocardial infarctions. I encountered it after my unit
developed a way of recording the mood changes that occur in chronic
pain and we found that an effective NSAID analgesic reduces stress
factors such as anger, fatigue, irritability, anxiety and sadness or
depression ( Kilminster SG & Mould GP 1999; Clin Drug Invest, 5:
345-54) So when I read an academic paper in the British Journal of CP
in 2006 ( Hawkey et al 2006; 61: 730-7) that claimed to have detected
an excess of cardiac events when patients took NSAID analgesics... I
wondered if these stress factors had been controlled. Sure enough
their control population was not in pain so they did not have these
additional stress-related emotions. So then I checked on Google for
references to anger and myocardial infarction and found over 90,000
references amongst which Mittelman et al, 1995 (Circulation; 92:
1720-5) reported a relative risk ratio of 2.3 cardiac events (95%
Confidence limits 1.7 - 3.2) in the two hours following an episode of
anger and that considerably exceeded the risk reported by Hawkey et
al. Then when the infamous Vioxx study is considered in the light of
Mittelman's findings it is obvious that the cardiac events that killed
not only some 30 patients but also the drug could have been due to it
failing to provide adequate analgesia. If Vioxx patients were still in
pain due to an inadequate dose they would have suffered these stress
factors while those in the control group on the well-known drug
diclofenac had adequate doses, so they did not. That means the
cardiac events were not a toxic event at all but a fault in the study
design or dose selection. This is important now because recent
American guidelines have stated that NSAIDs should not be used in
patients who have already suffered a myocardial infarction .....but if
they have pain and are stressed that course may well represent a
bigger risk than having good pain relief even by an NSAID and
certainly the side effects of other stronger analgesics are worse....
Accordingly I have written to the guideline authorities and
Merck. ..but it is interesting that I have been completely ignored.
Could that possibly be because I do come from the industrial side of
the fence ?
I would argue that interfaces often produce new ideas and this
potential insight comes from thinking about trials as models of
clinical practice and treatment and anyone in my place could have
thought of it. Its one reason why I, as a clinician, enjoy this forum
in fact.
My second example is being involved in the new use of zolpidem, the
favourite hypnotic of Americans as Ambien. It has been found to
produce extraordinary recoveries from brain damage and this has been
the stuff of TV and newspaper headlines, several papers and many
anecdotes on the internet so understandably many patients relatives
want their brain-damaged child or stroke-victim spouse to try it. The
risks are so small with such a well characterised medication that I
have been astonished that so many doctors refuse to try it, when
nothing else does this and sometimes the recovery has been emphatic
(though it is usually subtle). Accordingly, my role as the industrial
physician concerned with measuring this effect and getting it properly
established, is to help patients and their relatives and their
physicians by giving them the information that justifies a trial. It
can have an enormous effect on individual patient's lives, not to
mention their carers. Nobody else is in a position to fulfil my role
as far as I know. In fact I hope that by reporting this to the forum
some other physicians the other side of the fence might contact me so
that we can collaborate and make Clinical Pharmacology more like the
brotherhood I feel it should be.
I am encouraged that this debate has occurred in this forum because I
have thought for years that this is a vibrant and even vital
interface. In general I think clinicians should keep to clinical
factors as their prime concerns and not, as Roger says, stray into the
details of PK analysis when they have colleagues with much more
expertise and understanding, but that does require PKists and CPists
will keep talking to each other!
Best wishes for the New Year to all
Andrew Sutton
Andrew Sutton, MD(London) FFA,
Consultant in Clinical Pharmacology
Back to the Top
Dear Paul:
Thanks so much for your thoughtful note. It seems that today,
Clinical Pharmacology is nothing more than training people for the
drug industry, and in my view, that simply is not real clinical
pharmacology, which should be devoted to optimal management of drug
therapy in real patients. This should not be at all related to big
Pharma, but to internal medicine, which, it seems, wants to die before
learning how to use drugs well, or to learn about it and teach it. It
should not relate to Phase anything studies. These should serve as
Bayesian priors in planning.monitoring, further individualizing, and
adjusting drug doses. People see the costs of TDM and this
individualization, but do not see the well demonstrated reduction of
complications, adverse reactions, hospital stay, and their now also
well demonstrated costs. Complications cost. Optimal therapy, by
reducing them, saves.
Thanks so much for thinking of me and these points.
Roger
Roger W. Jelliffe, M.D. Professor of Medicine,
Division of Geriatric Medicine,
Laboratory of Applied Pharmacokinetics,
USC Keck School of Medicine
2250 Alcazar St, Los Angeles CA 90033, USA
Phone (323)442-1300, fax (323)442-1302, email= jelliffe.aaa.usc.edu
Our web site= http://www.lapk.org
Back to the Top
Dear David:
Thanks so much for putting this out for discussion. You are
right about most of the old clinical pharmacologists. They were
usually MD's with some other specialty, just as you say. It actually
seems to me that just as most of the medical profession appears unable
to understand these PK modeling and control approaches to therapy, it
seems to me that the real thing that has killed Clinical Pharmacology
as a medical specialty has been the PK/PD approaches themselves. MD's,
and increasingly more Pharm.D's, usually don't want to deal with these
issues ("it's too hard, not worth the effort, etc, responses that I
have encountered essentially worldwide), and so they have simply
avoided them, crippled themselves as therapists, and then denied this.
This now has gone on for at least a generation with the curriculum
committees in medical schools essentially worldwide. Even Pharmacy
schools teach less and less PK/PD now. Certainly this seems true at
USC. True, notable exceptions exist, in several centers, but they are
still comparatively few, and essentially nobody is actually teaching
goal-oriented, model-based optimal techniques for individualized drug
regimens to medical students or most Pharmacy students. Such a dismal
outlook! Let's keep working and hoping!
All the best,
Roger
Roger W. Jelliffe, M.D. Professor of Medicine,
Division of Geriatric Medicine,
Laboratory of Applied Pharmacokinetics,
USC Keck School of Medicine
2250 Alcazar St, Los Angeles CA 90033, USA
email= jelliffe.at.usc.edu
Our web site= http://www.lapk.org
Back to the Top
The following message was posted to: PharmPK
Dear Dr. Sutton,
I'm sorry to hear that your mother suffered from such a well-known
drug interaction. In the hospital setting, the detection point for
this in the USA would have been with the pharmacist that dispensed the
medication(s)--I am assuming it is likely that digoxin was a chronic
medication and clarithromycin was prescribed as a secondary medication
for an acute infection.
Pharmacists in the USA are thoroughly trained to detect, monitor, and
advise physicians regarding drug interactions. I would suspect that
every community and hospital pharmacy deals with situations like this
on a daily basis, although I am not suggesting that each drug
interaction is handled optimally. Although I am not completely sure
about UK practice standards, I would be very surprised if the same
were not true in your country.
Larry Bauer
Professor
UW School of Pharmacy
Seattle, WA, USA
Back to the Top
Hi Larry,
Yes I think in general the pharmacist is in the front line here as
well and you are correct in thinking that my mother's digoxin was a
long term treatment and the clarithromycin was added as an emergency,
probably after the pharmacy had closed one evening. That hospital is
unusual in having a clinical pharmacology department so I wonder if
that has clouded the issue of responsibility for flagging
interactions. Even then it should have been detected much sooner than
it actually was.
All the best
Andrew
Andrew Sutton, MD(London) FFA,
Consultant in Clinical Pharmacology
Back to the Top
The following message was posted to: PharmPK
Andrew:
One major issue is the quality of the resources available to the
pharmacist. There is a large, large body of literature that
demonstrates the current drug-drug interaction resources for
pharmacists and physicians are not adequate.
The drug pair digoxin-clarithromycin particularly illustrates the
inadequacy of the current products. Co-administration of
clarithromycin and digoxin in healthy subjects results in a 27%
decrease in clearance and 64% increase in AUC of digoxin but no side
effects in clinical trials- so most resources recommend only that
digoxin levels be monitored closely during co-administration. However,
the effect of clarithromycin on digoxin is highly dependent on the
patient's renal function. In a case series of patients with renal
failure, digoxin levels increased up to 397% during co-administration.
Furthermore, digoxin levels remained high days after digoxin was
discontinued. Clearly, in patients with moderate to severe renal
failure, this pair of drugs should never be co-administered--yet you
would be hard pressed to find a resource that makes this distinction.
When we start talking about therapeutic areas where drugs with with
narrow therapeutic margins are frequently administered, then the data
regarding the quality of resources is far more concerning. For
example, available commercial resources miss up to 70% of clinically
significant drug-drug interactions in transplant patients.
I believe there is a huge need to convene panels of experts that
understand both the basic science of drug interactions and the
therapeutic use of drugs to discuss safe practices and make
recommendations. For example, both clarithromycin and its "cousin",
erythromycin are problematic drugs that probably should be relegated
to third tier therapeutics (both are probably overused due to their
cheap price and familiarity). They are mechanism-based inhibitors of
CYP3A4 and inhibitors of P-glycoprotein. They also prolong QTc
themselves. As antibiotics, they are just not that "special" and I can
think of few situations where clarithromycin is the first line
therapeutic choice.
Sincerely,
Carol Collins MD
Department of Pharmaceutics
University of Washington
[Important clinical example and an interesting teaching example... in
poor renal function patients metabolism is a higher percentage of
overall elimination therefore a drug interaction restricting
metabolism has a much greater influence. I wonder how common this is?
- db]
Back to the Top
The following message was posted to: PharmPK
David:
Impairment of multiple clearance pathways is of great interest to me
(Collins et al. Curr Drug Metab. 2006 Apr;7(3):295-9). We have a
population that is growing older, taking lots more drugs concurrently
and often has some degree of renal or liver impairment. Evidence is
showing many elderly are taking 8 or more drugs so I think the
situation you suggest is very common. But as critical as I think the
issue is, it is sure hard to find data to model impairment of multiple
pathways. So the clinical significance remains unrecognized.
Carol Collins
Back to the Top
The following message was posted to: PharmPK
Roger: At Hartford Hosp, on rounds with a clinical pharmacologist, we
reviewed a patient being treated for stress following cardiac bypass.
The drug was valium and paradoxically, the patient's perception of
stress actually worsened with additional doses. After some
investigation the clinical pharmacologist identified the issue as
deriving from the lactic acid in the formulation. Once a different
formulation was used, the valium did what it was supposed to do for the
patient.
--
Ed O'Connor, Ph.D.
Laboratory Director
Matrix BioAnalytical Laboratories
25 Science Park at Yale
New Haven, CT 06511
Web: www.matrixbioanalytical.com
Email: eoconnor.at.matrixbioanalytical.com
Back to the Top
Dear Ed:
Thanks for your note. Very interesting! Good for the Clinical
Pharmacologist!
All the best,
Roger
Back to the Top
An additional complication to renal impairment of a patient can be his/
her genotype of the metabolising enzyme. An ultra-rapid metabolizer of
CYP2D6 who takes both metoprolol and quinidine is at the highest risk
of drug-drug interaction; however a poor metabolizer would be at much
lower risk. The situation reverses if the same subjects concomitantly
take metoprolol and ketoconazole; addition of renal impairment again
adds to the complexity.
As it was mentioned our elderly population take 8 or more drugs; how
would it be possible to consider all different dug-drug interactions
with multiple inhibitors/inducers and/or mechanism-based inhibitors?
It is true that the available data to model impairment of multiple
pathways are limited; however there are vast amount of demographic,
physiological, genomic and in vitro biochemical data which if
integrated using physiologically-based mechanistic model can
significantly improve our ability to simulate and predict drug-drug
interactions in patient populations. Of course, nobody can claim that
modelling and simulation is the answer to all questions; though surely
these emerging approaches can be used to assist us in better design of
clinical studies and provide us with more insights.
- Simulation and Prediction of in vivo metabolic drug clearance from
in vitro data.
Nature Vol. 6 No. 2, 2007
http://delivery.fostereprints.com/?Action=DeliverPDF&DeliveryCode=15151
- 'In silico' simulations to assess the 'in vivo' consequences of
'in vitro' metabolic drug-drug interactions. Drug Discovery Today:
Technologies, Vol. 1, No. 4, 2004.
Masoud
Back to the Top
The following message was posted to: PharmPK
***Many thanks for relaying this account of a paradoxical adverse
reaction.
Manfred Hauben and I have been reviewing the published literature on
paradoxical adverse drug reactions and have suggested a
classification--see our
preliminary account in Drug Safety (Hauben M, Aronson JK. Paradoxical
reactions: under-recognized adverse effects of drugs. Drug Saf 2006:
29(10):
970).
This one would count as a pseudoparadoxical reaction in our
classification,
since it was not due to the drug itself but to an excipient.
Regards
Jeff Aronson
--
J K Aronson, University Department of Primary Health Care,
Rosemary Rue Building, Old Road Campus, Roosevelt Drive, Headington,
Oxford OX3 7LF
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "What is Clinical Pharmacology" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)