The Rest of the Story

I haven't written about neuroscience for a long time.  Since Tuesday, to be exact.  That has got to change.  Yesterday, I found some information about NRP104.  Due to clamoring demand from hordes of readers, all curious about NRP104, I'll tell all I know.

For decades, amphetamine has been one of the two main treatments for ADHD.  Amphetamine and methylphenidate are old molecular entities, so the only marketing action has been in repackaging them into various intermediate and long-acting forms. 

When I read that yet another company is coming up with yet another way to deliver amphetamine to the human brain, I thought there couldn't be much substance to it.  Perhaps I was wrong.  A company called New River Pharmaceuticals has developed what they call Carrierwave™ technology.  Basically, this is a method of modifying existing drugs to alter their pharmacokinetics: where they go in the body, how fast they get there, and how quickly they go away.  Of course we have to have a long Latin-sounding name for that.  (NEJM has a good, free, but somewhat technical review of the clinical implications of pharmacokinetics here.)

In the case of NRP104, they take a molecule of amphetamine and stick an amino acid on it.  That's all.  They take a perfectly good molecule and render it inactive.  That may sound like a bad idea, but it can make sense.  NRP104 is another one of those prodrugs.  It is not active until it undergoes a chemical reaction in the body to make it active again.  In this case, an enzyme sits and waits somewhere, usually in the liver, until the prodrug floats by.  It then snips the amino acid off the amphetamine. 

Other than finding a way to patent something again, after the original patent has expired, is there any benefit?  Perhaps.  The idea is that the body only has a certain amount of the enzyme that converts the prodrug into active drug.  That enzyme operates at a fixed rate.  Thus, the active drug is produced at a steady rate over the course of the day.  In cases where the active drug has a significant abuse potential, the fixed rate of conversion might make the substance less appealing to those who might want to abuse it.  It also might provide some degree of safety in cases of overdose.  This would be effective only if the rate of conversion is in the correct range.  If the prodrug is converted fairly quickly, it might only make the drug safer for those who abuse it, but not make it less appealing to abusers.  If the rate is too slow, the drug will not have the desired effect.  It might just make it hard to sleep at night, without actually improving daytime functioning.  We already have lots of drugs that do that.

New River just reported the results of a phase II study:

A total of 52 children aged 6-12 with Attention Deficit Hyperactivity Disorder (ADHD) were enrolled in a double-blind, placebo- and active- controlled, randomized, 3-treatment, 3-period crossover study that compared NRP104's and Adderall XR's efficacy, duration and incidence of adverse events to placebo.

The primary efficacy endpoint in this study was SKAMP-Deportment (Swanson, Kotkin, Agler, M.Flynn and Pelham rating scale). In the study, patients treated with NRP104 showed a statistically significant improvement on primary endpoint compared to placebo across all three doses (p values <0.0001). We believe that the studies also demonstrated that efficacy results of NRP104 when compared to placebo and Adderall XR when compared to placebo were similar in terms of primary and secondary endpoints and should support the filing for the inclusion of a dose conversion table in the label of NRP104.

The significant therapeutic effects of NRP104 continued throughout the last assessment time point (i.e., 12 hours post morning dose), compared to placebo, suggesting a 12-hour duration of drug action.

Ok, that sound promising.  No doubt they have phase III studies underway.  New River says that, so far, the DEA has not required that the drug be considered a controlled substance.  If that holds up, it will be a big advantage in the marketplace. 

The only problem that I might anticipate with this product is that there tends to be a lot of variability in the amount of enzyme activity from person to person, for any given enzyme.  Some people have a lot of the enzyme, or have a mutated version that works faster than expected (rapid metabolizers).  Some people have multiple copies of the same gene, and produce more enzyme that other people.  Others have less, or have relatively slow-acting version of the enzyme.  In either case, a prodrug might not have the desired effect.  Below is an illustration from the NEJM article, showing this effect for a different drug.  The illustration shows how great the individual differences can be:


Personally, I would like to see the company address this issue before they put the drug on the market.  Regarding the variation in rates of conversion from prodrug to active drug, both the rapid metabolizers and the slow metabolizers might be unusually susceptible to adverse effects.  Rapid metabolizers could develop unusually high peak drug levels after each dose.  Slow metabolizers could have the drug build up with successive doses.  If they could devise a widely-available, reasonably-priced way to identify those people before treatment is attempted, it could save a lot of trouble.  If that is not feasible, then perhaps testing blood levels in people, after they have been on the drug for a while, could serve the same purpose.

Why worry?  Well, at about the same time that New River was reporting the favorable results of their phase II trial of NRP104, Health Canada was in the process of suspending sales of Adderall XR (an extended-release amphetamine).  This was not done in the US, in a move that no doubt will generate some controversy.

There have been 20 cases of sudden death reported worldwide.  Fourteen occurred in children.  These numbers are not large, considering the number of children and adolescents who have taken Adderall.  Of course, there probably have been others, cases that occurred but were not reported.  However, sudden death is so unusual in children and adolescents that I suspect most cases were investigated and reported.   The details are summarized here.  Of note:

Twelve cases of sudden death were reported between 1999 and 2003, all of which were males between ages 7 and 16. Five of the youngsters had significant cardiac risk factors, some of which were identified only on autopsy, such as abnormal coronary artery anatomy or abnormal valve structure. Seven of the youngsters had no abnormality, but at least one had a family history of cardiac ventricular arrhythmias. It also notes that several of the youngsters had unexplained and unusually high levels of the drug in their system, in the absence of any evidence indicating overdose.

The connection is this: several patients had unexpectedly high levels of drug, but were not thought to have overdosed.  If they were people who metabolized the drug slowly, such that it was not entirely cleared for the body after each dose, they could build up higher than expected drug levels. 

There have been many instances in which variations in pharmacokinetics were linked to unexpected outcomes.  Soon, we may be seeing a drug that has kinetics that are more complex than what is seen with the original drug.  This is somewhat speculative, but plausible: let's say there are some people who convert the prodrug into active drug very quickly, but who metabolize the active drug slowly.  Just picture the blood concentration curve in your head for a moment.  What you see is that they will have a certain baseline concentration of the drug in their blood before each dose.  Then, on top of that, they will get an unusually large peak after the dose.  For some drugs, such things don't seem to matter.  But for some, it matters a lot. 

Remember the illustration that shows the wide variation in metabolism of nortriptyline?  When that drug is used clinically, it is quite common to check blood levels of the drug.  It's a test that is widely available.  At this point in time, though, blood levels of amphetamine are not widely available. 

In order to really know whether this effect is going to be clinically significant, one would have to do pharmacokinetic testing in a large number of people.  That probably has not been done yet, since the NRP104 still is in stage II testing. 

Nobody knows, at this point, if amphetamine, when used as directed, actually does increase cardiac risk.  But if it does, then NRP104 could be safer for some people, but more dangerous for others, depending upon their individual differences in drug metabolism. 

There already has been some media buzz, and a few blog posts, about the implantable vagus nerve stimulator. It first was developed as a treatment for epilepsy, and it is being investigated for the treatment-resistant depression. One blogger is neutral on the subject; another expresses a strong negative reaction -- which is refuted brilliantly in the comments. Similar devices are used for treatment of severe Parkinson disease. Business week online has a decent review (echoed on GNIF Brain Blogger) for non-technical or non-medical readers:

The use of implantable mini-generators is more widespread than you probably think. Already, 190,000 patients are wearing electrodes in their heads to control Parkinson's disease tremors or spinal-cord stimulators to relieve pain or prevent urinary incontinence. Some 30,000 have wires threaded to the vagus nerve in the neck to treat epilepsy, while 60,000 have microtransmitters in the inner ear enabling them to hear. These numbers are likely to grow -- and quickly. One of the most promising devices is a $15,000 neurostimulator for chronic depression from Cyberonics Inc., which the Food & Drug Administration conditionally approved on Feb. 2.

A couple of months ago, I reviewed the information available on the Internet about the use of the vagus nerve stimulator (the Cyberonics device) for the treatment of treatment-resistant depression. At this point, it sounds as though it could be useful to some people. Apparently, the FDA agrees, since they have indicated that they will approve the device for that indication. Clearly, though, the approval will be based upon the fact that the patients for whom it is intended will have failed many other available treatments. The vagus nerve goes down the neck and into the chest, so it isn't necessary to go into the skull to use it. Even so, implanting something metal inside your body is serious business. It would be hard to justify unless it either worked very reliably, or it is intended for patients who have not responded to multiple other treatments.

What I could find about the efficacy was not terribly impressive. That is why I say that it might be helpful for some people. Once all the efficacy data are available, I might be able to comment more enthusiastically.

Now, though, we hear of something new. This one, called Activa®, does go directly into the brain. Or, at least, it has wires that go into the brain. The device itself is implanted under the skin, at the base of the neck. They use the same device for severe Parkinson disease. That use is described here, at the Georgetown University Hospital site. They even have a video clip of a news interview with one of the patients.

Based upon prior studies (for example, 1 2 3 4), researchers suspected that the device also could be used to treat depression. They performed a pilot study, which is going to be published in the next issue of Neuron, (which is not up on their website, yet). From a press release:

Experimental electrode implant treatment shows promise for helping severely depressed Study experiments with surgically implanted electrodes in patients who were not responding to all other available treatments 28-Feb-2005 Toronto, CANADA – A team of Toronto researchers says it has obtained promising early results from a landmark surgical study of the use of deep brain stimulation (DBS) in severely depressed patients who were otherwise resistant to standard types of treatment.

Significant clinical response was seen in four of the six study patients with chronic stimulation of the subgenual cingulate region, with sustained improvement through six months to the study endpoint. While DBS in other brain areas has been used to treat disorders such as epilepsy and Parkinson's disease, this is the first report of DBS in the subgenual region for major depression. The technique involves high frequency electrical stimulation to a targeted area of the brain for the purposes of modulating that region's activity. The subgenual cingulate area was specifically targeted based on growing evidence of its critical role in depression. [...]

Dr. Mayberg cautions that the Cg25 DBS study is only a "proof of principle". It is the culmination of 15 years of research using brain imaging technology that has worked to characterize functional brain abnormalities in major depression and mechanisms of various antidepressant treatments.

The device sends electrical impulses to the subgenual cingulate region of the brain. This causes the neuronal activity there to decrease. Exactly how that would result in improvement remains to be seen. In fact, it remains to be seen whether more that four patients will respond to the treatment.

If the Activa device turns out to be effective for anywhere near 60% of the patients with treatment-resistant depression, it will be a major advance. It also will raise a fair amount of consternation among insurance company executives. Read it and weep:

The procedure involved Dr. Lozano drilling two holes about the diameter of a 5-cent piece into the skull with the patient awake and under local anesthetic. Guided by magnetic resonance imaging to confirm the precise anatomic location of the grey and white matter of area 25, he inserted two thin wires with electrode contacts down to the white matter tracts adjacent to the Cg25 area. The other ends of the wires were tunneled through to the lower neck area and hooked up to a pulse generator implant that directs an electrical current. The entire system, likened to a "brain pacemaker", is under the skin. Researchers determined the most efficacious voltage, pulse width and frequency for each patient and tracked their clinical response using positron emission tomography (brain imaging of cerebral blood flow activity) along with a comprehensive battery of neuropsychological tests. This tracking was done at baseline (pre-surgery) and at monthly intervals for 6 months.

The device itself will cost about $15,000. The news report of the study does not say how the candidates were selected. It would make sense, though, do do some kind of functional neuroimaging study to select the patients. Paying for all those scans would cost a bundle. Then, you have to use an MRI scan to figure out where to put the wires. You have to pay for the surgery, and you have to pay for follow-up. And the insurance company gets nothing back, after paying all that money. That makes for an adversarial situation.

Perhaps it is obvious, at this point, but I am going to say it anyway. This kind of thing is going to become more and more common. It may make the concept of private health insurance untenable. So obviously we need a national health program.

Wait! you say? Who wants the taxpayers -- those hard-working, compassionately conservative, long-suffering Americans, whose pockets get picked every April -- to pay for all this? You do, and here is why: The people who would be candidates for this are people who either cannot work, or who are working far below their potential. Many of them are getting social security disability. With the SSD and all their other benefits, housing subsidies, etc., they cost us, say, $25,000 per year. Take one of those citizens, treat them effectively, and they go back to work, making, say, $35,000 per year. The economic benefit is $60,000 per year, for the rest of their working lives. Over a 20 year period, that is $1.2 million dollars. Yeah, it's a back-of-a-envelope calculation, but it illustrates the point.

Pseudoethicists will cry out that the almighty taxpayer should not have to pay. But why should the insurance company pay? They get nothing back. Taxpayers don't get anything back, directly, but indirectly they get a million bucks. So which is ethical?

I suppose if you abolish social security, Medicaid, and housing subsidies, that would eliminate 40% of the value of the transaction. That makes it less appealing, but still a decent investment.

Like I said, more medical interventions are coming along, they will be expensive, and there isn't going to be any way to reduce the costs beyond a certain point. Sooner or later, insurance companies are going to have to become so selective about what they will pay for, that there won't be much point in having insurance. Individuals will go bankrupt, or just get sick and die, because of the cost of health care. The country will loose some of its most valuable resources: its people.

The notion that citizen X should not have to pay for citizen Y's medical treatment is appealing, on the surface, but it doesn't make sense once you admit that we all depend on each other in order to have decent lives. It may be true that person X, who lives in North Dakota, does not directly benefit if Ms. Schnitzel, who lives in Florida, gets treatment for her pneumonia. But that is a short-sighted, narrow-minded analysis. If the sum of all medical treatment provided in this country results in a reduction of infant mortality, is that worth anything to Mr. X? If the average span of productive work years per citizen is increased by 5%, is that worth anything? If we have more people working on inventing better inductive transconbobulators, or mousetraps, for that matter, rather than working on and off at various minimum-wage jobs, is that worth anything?

First, let me say this.  The world is better off with the FDA than it would be without the FDA.  Although the US FDA has no authority outside the US, its actions have significant effects around the world.  For the most part, those effects are positive.  Anyone reading the news lately is likely to have at least some degree of skepticism about the effectiveness of the FDA.  A recent editorial in Newsday indicates that public confidence in the FDA has declined:

It [the FDA] was the gold standard of trustworthiness, both for the patient and the prescribing physician. Even drug manufacturers impatient with the long process coveted FDA approval, for it meant their drugs were safest; American drugs were the safest in the world.

Lately, however, the FDA's credibility has fallen on hard times. According to a recent CNN-USA Today-Gallup Poll, nearly 40 percent of 1,015 adult respondents said their confidence in the FDA has slipped during the past year.

What the Gallup Poll does not tell us is whether the declining confidence in the FDA is merely a reflection of increasing public realization that the government is more pro-business than pro-people.  Still, if taken at face value, the survey indicates that public perceptions are becoming more negative.  To what extent is this perception valid?

Undoubtedly, the organization  does have flaws.  The British Medical Journal posted an article yesterday, pointing out the deficiencies of the new drug safety oversight board. 

Dr Sidney Wolfe, director of Public Citizen's health research group in Washington, DC, said that the new board would only be an advisory body, relying on information from the FDA's Center for Drug Evaluation and Research, which itself stands accused of being too slow to move on safety and too close to one of its major sponsors, the drug industry. Dr Wolfe said that the new board will be a "toothless tiger" unless the opinions of the FDA's Office of Drug Safety are given more weight and enabled to prevail.

The BMJ included a comparison between the US FDA and the corresponding agencies in Britain and in Australia.  The British system has been criticized, mostly for being too slow to respond to indications of potential problems.  The Australian system might be better, in part because there is more separation between those who approve drugs, and those who monitor safety. 

The Chicago Tribune posted an article that was more detailed, and more critical:

Safeguards designed to protect Americans from potentially dangerous medications are being eroded by conflicts between federal bureaucrats who approve new drugs and those who oversee their safety, according to former and current officials at the U.S. Food and Drug Administration.

FDA regulators rush to review applications for new medicines but are slow to address serious problems that surface with the drugs once they come on the market, interviews with physicians, scientists, government officials and medical school researchers suggest. [...]

Last week, the administration sought to defuse some of the criticism aimed at the FDA, saying it would appoint a permanent commissioner--a position that has been vacant for almost a year--and create a new Drug Safety Oversight Board. The president's budget also shored up funding for the FDA's current Office of Drug Safety.

But critics say those actions do not fully address deep-seated flaws plaguing the FDA, including conflicts of interest, a lack of attention to safety issues and a lack of authority in forcing industry to heed its recommendations.

The Associated Press released an article illustrating a potential conflict of interest:

Ten members of the Food and Drug Administration (news - web sites) advisory panel who voted that a group of powerful pain killers should continue to be sold had ties to the drug makers, an advocacy group says.

A study by the Center for Science in the Public Interest indicates that 10 of the 32 panel members had ties to either Pfizer Inc. or Merck & Co., ranging from consulting fees and speaking honoraria to research support.

Of course, consulting fees and speaking honoraria do not necessarily create a significant conflict of interest.  The FDA has internal mechanisms for panel members to declare potential conflicts; a well-run organization can deal with such issues, without compromising public safety. 

Getting back to the original question, how valid is the public perception that the FDA is becoming less effective at protecting public safety?  There have been changes in the operation of the FDA, and those changes have led to quicker approval for many drugs.  In return, the pharmaceutical companies were supposed to increase their post-marketing surveillance of the safety of their products.  Some have argued that this has not happened.  The recent Vioxx problem may have arisen, in part, because of this. 

Unfortunately, single case studies do not really tell us what we want to know.  We may find a few cases in which drugs were put on the market, then not withdrawn promptly when there were signs of safety problems.  But what we really need to know is this: have the changes in the FDA resulted in more harm than good?  Presumably, getting drugs to market faster does result in some improvemnts in overall public health.  It also may entail some risks.  But finding a few cases of risk tells you nothing about the overall effect of the changes.   I've been looking for a few months now, albeit rather casually, and have not come upon anyone trying to answer this question.  So the fact is, we do not know whether the public should have a declining opinion of the FDA.

It is likely that quicker approval of HIV/AIDS drugs has resulted in many saved lives, or at least additional good years for those who ultimately succumb to the illness.  Quicker approval of arthritis drugs may not have actually extended anyone's life, but it probably added years of productive life for many people.  Similar arguments could be made for all categories of drugs. 

In addition, we need to understand that no regulatory process will eliminate risk.  Indeed, that is not the goal of the process.  The goal of the process is to quantify the risk as accurately as possible, so people can make informed choices.  It is common for patients to choose to take a drug, despite known risks.  There is nothing wrong with that, so long as the patient has the opportunity to make an informed choice.  

Some patients may decide to take the risk, even if an appropriate regulatory agency would advise against it.  Here's an example: the Vioxx controversy occurred, in part, because doctors ended up given the drug to people who did not fit the profile of patients for which the drug was approved.  This meant that a lot of people got exposed to the risks of the drug, people who --some may argue -- should not have taken the drug.  But let's say a patient has migraine headaches, occurring two to three times per week.  The patient is at risk of loosing her job.  She finds that Vioxx works very well for her.  Vioxx was not approved for treatment of migraine.  Who can tell her that she should live with the pain, and suffer the financial consequences, rather that take the drug and run the risk of cardiovascular complications?

Most of the editorials focus on the problems, not the benefits, of the approval process.  Personally, I think we need to reserve judgment on that question.  Even so, that does not free the government from the obligation to make sure that the FDA is operating as effectively as possible to discharge its primary missions.  That is, even if some study eventually shows that the current approval process has created more benefit than harm, we still need to see if it can be made better.