I’m getting my medical news from the front page of The Washington Post where Christopher Rowland discusses the possiblity that embrel, an anti arthritis drug, reduces the risk of Alzheimer’s dementia.
The issue is clearly incredibly important. The article raises interesting ethical, economic, statistical, and biological questions.
Better to click the link, but I will attempt a quick summary. There is evidence from anonymized insurance claims records that people who take Embrel for arthritis are less likely to develop Alzheimer’s. This hypothesis could be tested against the null of no effect with a huge long lasting extremely costly clinical trial. The patent on Embrel is about to expire. Obviously the Pfizer executives decided not to fund the trial. They also didn’t publish the retrospective study (it seems someone leaked it to Rowland).
I am most interested in the biology. Pfizer claims that they decided this, because of scientific evidence that Embrel wouldn’t work. In particular it does not cross the blood brain barrier. The argument is that it can’t help the brain without entering the brain. This is nonsense.
Embrel binds the appealingly named but nasty protein hormone tumor necrosis factor alpha (TNF-alpha). TNF-alpha promotes inflamation. It has a role in fighting tumors. It also causes cachexia wasting away of cancer patients and tuberculosis patients. It was discovered two ways. One group was purifying the wonderful TNF, the other the evil cause of cachexia which they called Cachexin. Both were surprised to discover they had purified the exact same molecule.
embrel binds and blocks TNF-alpha. It is a soluble form of the TNF-alpha receptor which competes with actual receptors which send a signal to cells when they bind TNF-alpha. It is a very important treatment for rheumatoid arthritis.
I say it doesn’t matter than embrel doesn’t enter the brain, because TNF-alpha does. If embrel binds TNF-alpha the dimer doesn’t enter the brain. If TNF-alpha in the brain causes Alzheimer’s then Embrel can reduce the risk by inactivating TNF-alpha which would enter the brain (also keeping it out of the brain but that doesn’t matter).
Pfizer’s argument is like saying that strangling someone can’t cause brain damage, because the strangler’s hands don’t enter the brain. They block oxygenation of hemoglobin which does enter the brain. The argument is silly (and effectively refuted by Rowland and scientists he interviewed).
I mean it’s obvious that inflammation spreads. That’s why cold viruses in your nose make you feel aches and pains all over (that one is interferon not TNF-alpha but the logic is the same).
There is fairly extensive evidence that inflammation has a key role in Alzheimer’s.
Of course the economics is also very clear. A huge investment in finding a new use for a molecule whose patent is about to expire is absolutely inconsistent with any duty to shareholders. This is clearly a case in which the government should fund the trial.
The statistical or methodological issue is that doctors will not trust non experimental evidence. The FDA has very strict rules for approval of a new pharmaceutical and for advertizing a new use of a pharmaceutical. They do not have to be applied to the decision of whether to prescribe a pharmaceutical off label (not for the use for which it was approved) but they are. The idea is that prescribing something with known and very minor side effects without knowing it will work is irresponsible. The logic is completely different from maximizing the expected welfare of the patient (in this case the healthy adult).
The idea that non experimental evidence is highly suspect and to be used only to propose experiments is alien to me. The rules which are called medical ethics seem to me to have very little relationship with the actions which are, in fact, morally right and morally wrong.
I think everyone should take Embrel (as soon as it is off patent and affordable). I also think there really should be a publicly funded clinical trial (which will be possible because no one will take the advice I gave in the past paragraph).
update: pulled back from comments. My knowledge of US pharmaceutical regulation is out of date (30 months out of date, but that’s out of date). I said non experimental data should be considered. So did Congress and Barack Obama and their signed act is law.
June 5, 2019 6:59 am
It became far easier to bring things to market under the recent passage of the 21st Century Cures Act.
“The 21st Century Cures Act modified the FDA Drug Approval process. It was intended to expedite the process by which new drugs and devices are approved by easing the requirements put on drug companies looking for FDA approval on new products or new indications on existing drugs. For instance, under certain conditions, the act allows companies to provide “data summaries” and “real world evidence” such as observational studies, insurance claims data, patient input, and anecdotal data rather than full clinical trial results.”
Pharma R&D is given the normal tax break also.
Someone is lying.
Pfizer might have decided not to publish the data back before December 13 2916. In any case, they don’t need FDA approval to publish data (1st amendment and all that). They need approval to advertise a new use of Embrel, but they can publish the data.
It seems clear that their problem is that Embrel is going off patent and will soon just be competition for their newer still on patent drugs. The news is out (as of yesterday on the front page of the Washington Post).