E-DRUG: Clinical trials vs pharmacology - thiazides, cox-2s, HRT, HPV vaccine
---------------------------------------------------------------------
Dear E-druggers,
In this message I will respond to 3 different threads: Thiazides and high blood pressure, FDA Testimony opposing COX-2 inhibitor Arcoxia, HPV vaccine and I will also mention HRT.
I thank Alberto Donzelli and Maria Font for supporting my position based on clinical trial evidence in favour of chlorthalidone rather than Leo Offerhaus' pharmacological arguments for hydrochlorothiaziade.
Leo suggests that this debate is not important.
I think it is important for 3 reasons:
1. We do not know if there is a clinically important difference in the efficacy or adverse effects at the optimal (low) dose between chlorthalidone and hydrochlorothiaziade. If there is then that will be important for the very large number of people who have hypertension and should be taking one of these 2 drugs.
2. Our uncertainty about what is the best drug for hypertension is another example showing that the current dominance of research by patent monopoly based funding is distorting priorities away from providing the information that clinicians and patients need most. We need to change the system.
See:
Mansfield P. Industry-Sponsored Research: A More Comprehensive Alternative. PLoS Med 2006;3(10): e463
http://medicine.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pmed.0030463
The Quality Adjusted Life Years (QALYs) gained per cost ratio from implementation of the results of a large chlorthalidone versus hydrochlorothiaziade trial may be much better than many other common treatments.
3. I think my debate with Leo is a subset of a wider debate about the place of pharmacology vs clinical trials in prescribing decision making. Leo's argument was based on pharmacology, my argument was base on clinical trials. Alberto and Maria integrated both.
I think that if pharmacology and clinical trial results conflict then the pharmacology is probably wrong or more likely incomplete. However I would like to balance that by giving some examples of how pharmacology (speculation about mechanisms) can still have an important role especially for hypothesis generation.
a. Hypotheses about pharmacological mechanisms are very important for new drug development.
b. Knowing which other drugs a drug is similar to can alert us to the possibility of similar effects.
For example, Sid Wolfe's FDA Testimony opposing COX-2 inhibitor Arcoxia noted evidence that diclofenac causes a high rate of cardiovascular events compared to other old NSAIDs. This evidence is less surprising and takes on a different meaning when people know that diclofenac is almost as COX-2 selective as celecoxib.
See:
Selective COX-2 inhibitors: Are they safer NSAIDs? Therapeutics Letter 39, January - February 2001
www.ti.ubc.ca/node/61
Consequently, if diclofenac is included in the comparators for any new COX-2 selective drug the cardiovascular harm caused by the new drug will be obscured.
Our understanding that diclofenac is a COX-2 selective drug is part of why we recommend that all COX-2-selective drugs (including diclofenac) should be removed from the market until they have been properly evaluated.
See:
Mansfield PR, Vitry AI, Wright JM. Withdraw all COX-2-selective drugs. Med J Aust 2005; 182 (4):197.
www.mja.com.au/public/issues/182_04_210205/matters_arising_210205_fm-1.html
c. Thinking about mechanisms can also help us when trying to extrapolate from clinical trials with short durations (rarely more than 5 years) to predict treatment effects over a lifetime.
For example HRT has a benefit / harm profile similar to cigarettes. HRT causes or more likely accelerates breast cancer.
We know that cancers take years to grow before they are detected. Consequently we can predict that he harm caused by taking HRT for 5 years then ceasing includes not just the breast cancers diagnosed during those 5 years but also breast cancers that are diagnosed many years later.
Another example is speculation many possible mechanisms that could reduce the effectiveness of the HPV vaccine for reducing cervical cancer death rates:
i. The targeted strains could become resistant (a la influenza)
ii. Other strains could become more prevalent once the competition from the targeted strains has been removed. This could include strains that are currently unknown and more dangerous.
iii. Women may have less cytology screening tests out of a false sense of security.
iv. The inverse care law. Those women who are most at risk of dying of cervical cancer are least likely to receive the vaccine.
regards,
Peter
Dr Peter Mansfield
GP
Director, Healthy Skepticism Inc
Countering misleading drug promotion.
www.healthyskepticism.org
Research Fellow, Discipline of General Practice, University of Adelaide
peter.mansfield@adelaide.edu.au
See publication list at:
www.adelaide.edu.au/directory/peter.mansfield