The Rest of the Story

Sub-subtitle: A study of the intersection between pharmaceutical science, marketing, and reimbursement

Dapoxetine  is under review by the US FDA for treatment of premature ejaculation, having completed phase III trials.  The company announced that makes it plans to market 30mg and 60mg doses.  Dapoxetine is a serotonin reuptake inhibitor.  Presumably, it acts by stimulating the 5HT-2 receptors, thereby diminishing the strength of perception of sexual stimulation.  This effect occurs in some, but not all people; the reasons for the variation in response are not at all clear.

The drug has an interesting history.  It was developed by Eli Lilly, the makers of fluoxetine, atomoxetine, and duloxetine, as a putative antidepressant.  In was licensed to a different company, PPD, for development.  PPD subsequently obtained patent rights, then licensed it to ALZA, which partnered with Ortho-McNeil Pharmaceutical, Inc. (a subsidiary of Johnson & Johnson), to bring it to market.  You'll have to ask someone else to explain all the financial wheeling and dealing.  I did learn that Forbes anticipates sales of dapoxetine to be about $500 million in the first year.  Perhaps that is too small a sum for Lilly to bother themselves with.

The curious reader will ask, no doubt: why in the world we need another serotonin reuptake inhibitor?  In the USA, we already have the selective serotonin reuptake inhibitors fluoxetine, paroxetine, sertraline, fluvoxamine, citalopram, and escitalopram (Prozac, Zoloft, Paxil, Luvox, Celexa, and Lexapro, respectively).  (Did I leave any out?)  If you include nonselective agents, add venlafaxine, duloxetine, and soon norvenlafaxine (Effexor, Cymbalta, and brand name pending, respectively).  Not to mention almost all of the tricyclics (almost all of the tricyclics, respectively).

Last year, Marcia Angell wrote a book (The Truth About the Drug Companies) criticizing the pharmaceutical industry for developing "me too" drugs, saying that research dollars would be better spent developing novel drugs.  I actually wrote a post criticizing that notion, saying that these drugs are not really interchangeable in clinical practice.  The development of dapoxetine might change my opinion on that.  How many of these things do we really need?

The devil is in the details.  In this particular situation, the devil is in the details of pharmacokinetics.  According to PPD (170 KB PDF file) dapoxetine has "favorable" pharmacokinetics.  I wasn't able to find the details; Medline gives only five hits for dapoxetine and none of them is very informative.  I assume, with a high degree of confidence, that the drug is absorbed quickly, metabolized (via N-dealkylation) quickly, and looses effect quickly.  14C-labeled dapoxetine is undetectable in rats within 72 hours, which implies a half-life (in rats) of about 7-14 hours.  (See how hard bloggers sometimes have to scrounge for tidbits of information.)

As an aside, pharmaceutical companies don't publish most of their basic science studies, which makes sense from a fiduciary standpoint, but is irritating anyway.  Even so, publicly available information can be pretty informative, perhaps more so than they realize. 

The PPD document that I linked to above indicates that the pharmacokinetic profile is suitable for PRN dosing.  PRN means "as needed."  You can take the drug whenever you feel like taking it.  Will there be adverse effects?  They don't say, but most likely nausea, headache, and insomnia will occur in 10-20% of people.  Some might get sleepy instead of having insomnia. 

Do I have any concerns about the drug?  Not really.  I wonder, though, if they really will get a half billion dollars a year out of it.  Assuming they get the FDA approval, and they probably will, it will be the only FDA-approved drug for premature ejaculation.  But all those other drugs (except maybe duloxetine) I mentioned probably will work just as well.  What this means is that dapoxetine will come to market with FDA approval, which is a marketing advantage, but probably will not have any significant medical advantage. 

I could be wrong about that, since Lilly/PPD/ALZA/Ortho-McNeil/J&J haven't release all the specifics yet, but I am fairly sure their scientists won't be able to come up with anything that makes the drug clearly preferable over the others.  It will be interesting to see what their marketing people will try to come up with.  I'm sure they'll have to use a turnip press to squeeze science out of a vegetable.  (I know that's a metsed mixaphore, but you know what I mean.)

Science and marketing always have an interesting relationship in the pharmaceutical industry.  The dapoxetine story is different, though, in the way that science, marketing, and reimbursement will converge once the product is on the market.

In order for a pharmaceutical product to be profitable, it is (or ought to be) safe and effective, AND be accepted by patients, AND be reimbursed for by insurance companies.  That latter point may be diminished somewhat in the case of dapoxetine, since people are more willing to pay out of pocket for something that improves their sex life, as compared to, say, something that treats cancer.  Still, the insurance reimbursement is a significant factor in the profitability of a drug.  If Forbes is correct, and dapoxetine results in $500 million in sales in the first year, that potentially means that the insurance industry will have to come up with an extra $500 million dollars next year, that they did not have to pay this year.  Actually, 500,000,000 minus unspecified discounts, minus copays and deductibles, minus what people pay out of pocket.  Even with all those offsets, it would be a significant expense. 

It is safe to assume that the insurance companies aren't going to want to part with that half-billion dollars, but do they have a choice? Perhaps: generic paroxetine has a half life of 16 hours (on average, in humans).  The retail price (the insurance company will pay much less) of generic paroxetine is $72 for 30 pills online at Walgreens.  The price is going to decline rapidly over the next year. Probably a half of one of those pills would do the job.  Therefore, even before the price drops, paroxetine can be gotten for $36 dollars for thirty doses, and probably the average person would use at most 15 doses in a month.  Let's say the retail cost for 15 doses is going to be $18/month, and the copay is $5, and the insurance company actually pays less than $10 per month.

We do not know what Ortho-McNeil is going to charge for dapoxetine.  However, nobody is going to go through the trouble of bringing a new pharmaceutical to market if the retial price is only going to be $18/patient-month.  I'll venture a guess that they are going to shoot for about $75/patient-month, with the insurance company paying ~$50/month.  Those are very rough estimates, but it illustrates the point. 

Given the enormous price differential, it would make sense for insurance companies to urge physicians to prescribe generic paroxetine instead of brand-name dapoxetine.  Will they do that? Probably?  Will they try to do it with an honest countenance?  Surely.  If they do this, will it expose them as damn liars?  Yes.

You see, insurance company often try to deny payment for perfectly legitimate prescriptions by arguing that the drug is "not approved" for the patient's diagnosis.  That's already an untruth, since the fact that the doctor wrote the Rx is prima facie evidence that the doctor approved it.  So what does the phrase "not approved" mean in this context?  What it might mean is that it was not FDA approved, although they never actually specify it that way.  If that is their intended meaning, then the reason they don't specify "FDA approved" is that there is not a shred of evidence that FDA approval is a sensible or justifiable way to make reimbursement decisions.  The artful use of the passive voice ("it is not approved" vs. "The FDA has not approved it") obscures the fact that the agency that not approving it has no legitimate jurisdiction over the approval process.  The other possible rationale for the use of the passive voice is that they do not want to come right out and say "we do not approve it."  If that is the case, then they are dodging their responsibility.  This may not seem obvious to the uninitiated, but take my word for it, it is true.

Now, the introduction of dapoxetine will put them in a difficult position.  Do they try to maintain the charade of insisting that FDA approval should be a precondition for reimbursement (if that is the meaning of the vague phrase "not approved"), or do they admit that it is the insurance company who is taking responsibility for not approving it; or, do they suggest that the doctor prescribe (inexpensive) generic paroxetine instead of (expensive) branded dapoxetine?  Do they stick to their principles, as misguided as they may be; or, do they try to save a few hundred million dollars?

This might turn out to be an interesting study in medical economics.  I wouldn't know about that.  However, I do think it is an interesting study of human behavior. 

Addendum: It occurs to me that there might be a rason to be concerned about unexpected adverse effects from dapoxetine.  Although the predicatable adverse effects will be minor, there will be a group of patients that could be susceptible to psychiatric disturbance.  Specifically, patients with bipolar disorder could be prone to have a manic episode if exposed to a serotonin reuptake inhibitor.  Admittedly, the probability of that is small for any given person, but if millions of people take it, it would be reasonable to expect several cases per year.  If the FDA is doing their job, they will require postmarketing survelliance for this potential problem.

I have noticed that it can be fun to read blogs.  Yesterday, I read 90% Crud, which is, I suppose, a good description of the Blogosphere itself, but which also happens to be the name of one particular blog, which is subtitled as follows: Sure, 90% of weblogs are crud. That's because 90% of everything is crud.  The post I noticed contains a proposal to combine TiVo-like video recording with an RSS-like aggregator to make news reports more useful as well as more palatable.  Personally, I think that is a great idea; figuring out how to do that and make it commercially viable is the next project. 

The author, George Hotelling, makes reference to another blog, Creating Passionate Users, authored by Kathy Sierra, who put up a picture of herself on a horse.  She tells us that she learned a lot about human psychology from her horse.  That probably is true, although I think baboons are much better teachers than horses. 


She points out that "You can't be afraid and rational at the same time. Pick one."  At first, it is not obvious how that applies to Mr. Hotelling's proposal, but it does.  Either trust me on that, or go read his post to see how.  There isn't any reason for me to explain it here because he explains it just fine himself, and it is not really the point I am trying to make. 

Ms. Sierra has picked up an an old idea in psychotherapy, so my point is to outline the history of that idea, and explain why it is so profound.  The idea was developed in the 1950's by a psychoanalyst turned behavioral psychiatrist, Joseph Wolpe.  His ideas strongly influenced one of my old supervisors, George Curtis, who also had been an analyst but later became a cognitive-behavioral specialist, and who was the director of the Anxiety Disorders clinic at the University of Michigan. 

The notion that you can't be afraid and rational at the same time is an example of reciprocal inhibition.  This psychological principle was developed as an analogy to a similar principle in neuromuscular physiology. 

Most muscles are arranged into opposing groups.  In order to move a body part, it is necessary to tighten that group of muscles that perform the movement.  At the same time, the opposing muscles must loosen.  The brain is wired such that this happens more or less automatically.  One does not consciously relax the triceps, so that the biceps can flex the arm at the elbow.  Rather, one consciously contracts the biceps, and the triceps seem to relax on their own. 

Similarly, in psychology, some emotions are mutually exclusive.  If one focuses enough attention on relaxation, then fear is reduced.  Also, if one becomes sufficiently bored, then anxiety does not occur.  Intellectual curiosity can serve the same function.  If one is sufficiently curious about what is going on, then fear does not occur. 

The concept of reciprocal inhibition is central to the clinical treatment of anxiety disorders.  The idea is simple.  Everyone knows that if you have a fear of heights, the only way to get over it is to spend a lot of time in high places.  If you go up high enough in a building to feel the fear, then wait for it to wane, and do this often enough, eventually the fear will not occur.  In actual practice, it usually takes at least 45 minutes of exposure several times per week.  Going to a top floor of a building for 45 minutes every day, and just sitting there, sooner or later you will start to get bored.  The more bored you get, the less fear you experience.  That is reciprocal inhibition in action.

Almost anyone can do this to overcome most simple phobias; it is extremely reliable.  Not all fear stems from simple phobias, though, so sometimes the therapy gets to be more complicated.  Sometimes the fear is triggered by situations that cannot be replicated at will. 

Sometimes the fear is so great that it seems insurmountable.  In such cases, patients feel guilty and inadequate, and often are stigmatized by others.  Such judgments may be premature, however, because there may be situations in which conscious control is not possible.

To explain, it is helpful to return to the neuromuscular analogy.  There are occasions when the muscles do not respond to conscious will.  For example, if you touch a hot stove, your arm will pull back.  If the heat is sufficiently great, no amount of willpower will enable you to overcome the reflex.  It does not matter how many episodes of Kung Fu you watch, you simply cannot exert executive control over the withdrawal reflex.  That is because the reflex occurs in the spinal cord.  The pain is felt, reaches a certain threshold, and the appropriate muscle movements are initiated, all before any signal reaches the brain.  It is impossible to exert conscious control, because your brain doesn't even know what is happening. 

The circuits inside the brain are not understood as well as those in the spinal cord, but it is likely that there are situations in which the drive to avoid the anxiogenic (anxiety-provoking) stimulus is so great, that the avoidance behavior is initiated without the frontal lobes having a chance to intervene.  Sometimes it is possible to learn to override this, but sometimes not.  We do not know what makes the difference.  It is likely that, in some cases, people can override the drive to avoid the situation by modifying their perception.  After all, it is not the feared object itself that creates fear; rather, it is our perception of the object that provokes fear.

Some theoreticians and clinicians believe that reciprocal inhibition is the primary means by which psychotherapy is helpful.  By helping he patient become curious about his or her own behavior, the fear associated with that behavior is diminished.  Although many people think that it is insight that results in change, it may be more accurate to say that it is curiosity that does the job.  Personally, I am highly skeptical of any simplistic model that is proposed as an all-encompassing explanation in any branch of psychology.  The brain is too complex for that to be true.  Even so, the concept of reciprocal inhibition stands as one of the cornerstones in our understanding of the human mind.