Does the Supreme Court decision announced June 17th, giving the FDA the right to sue pharmaceutical companies that enter into “pay to delay” agreements with generic drug makers represent a blow to big pharma, a gift to the generic sector, and a boon to consumers? I don’t think so, and here’s why:
The situations where this ruling might apply are not that common. What happened in this case is that a generic competitor to Solvay’s Actavis testosterone gel challenged Solvay’s patent. Rather than slugging it out, Solvay made a deal with the competitor to keep the generic off the market until Solvay’s patent expired. How much Solvay shelled out to protect their patented version, I don’t know, but it no doubt struck the generic outfit as better odds than what could have been a protracted court case. And in the event of a court battle over the patent, who knows which side would come out on top. And how much it might cost.
However, in my opinion, the drug patent system is flawed, not to say dysfunctional. How it works is that the original developer of a drug absolutely has to file a patent as soon as they’re looking at any promising drug entity. They cannot, cannot, cannot wait until their drug is approved to get it patented – doing so would risk that somebody else beats them to the finish line, and all their work is down the drain.
Drug patent life is 20 years (under normal circumstances) from patent approval. So if it takes the pharma company 7 years to move their drug through the regulatory process, they have 13 years of remaining patent life. And if it takes 12 years to get their drug approved – which is by no means unusual – they have 8 years of patent life. That’s the patent cliff you hear about. And the approval process is arduous, as it must be, if the regulators are going to be able to assure the public that the drug is both safe and effective.
The Drug Patent Situation outside the US
A “pay for delay” case is currently underway in the EU. The Danish pharma, Lundbeck, is the target of enforcement by the European Commission because of a deal with a generic manufacturer to keep citalopram off the market until the patent expired, about two years. Citalopram is marketed in the US by Forest as Celexa. But more about Celexa below, under Phase IV trials.
Meanwhile, in India the Supreme Court refused to grant Novartis (NVS) a patent on their leukemia drug Gleevec, permitting Indian drug manufacturers to sell a much cheaper version. The decision was acclaimed as an important step in making low-cost drugs available in poor nations – Gleevec would cost about $70,000 per years, while the generic version costs about $2,500 per year. Novartis counters that it makes Gleevec available to huge number of leukemia patients in India at no charge, and that if drug developers are not compensated for the enormous costs of moving a drug from the first glimmer of promise to full marketability, then innovation will be stopped dead in its tracks.
And, of course, India is the biggest manufacturer of generics in the world. India could flood the entire world market with cheap generics, a matter of considerable concern to the entire pharmaceutical community. And India did not even have patents on drugs until 2005 – and then only for drugs introduced after 1995 – so the concept might be said to be a bit alien to them.
But, returning to the US …
The Drug Approval Process in the US
First, the original developer identifies a promising agent that might have some useful activity. Let’s say, for the sake of inventing a case study, that they are looking around for antibiotics that might address some of these exceedingly troubling resistant microbes that are giving the infectious disease docs nightmares. They find that by tinkering with the structure of one of the existing antibiotics they can get it to penetrate the cell wall of some of these offending pathogens, and, by golly, stop these little devils from reproducing. So, immediately, they patent the new molecule.
Now they have to do a whole lot of lab work. They need to test against which pathogens their new agent will be effective, and they also need to make sure that the agent doesn’t harm the myriad human cells that are essential to life. (An essential difference between antibiotics and disinfectants is that the antibiotic specifically targets pathogens, while disinfectants tend to kill whatever cells they touch. You can put iodine on an infected scrape or cut, but you don’t want to take it internally or even cover it up, because the iodine will be absorbed and do a lot of damage to healthy cells.)
The lab work is termed in vitro – “in glass” – but early on they will also do a considerable amount of in vivo testing – i.e., in living animals. Small animals at first, and eventually larger animals, that, we hope, tell us how humans might react to the potential antibiotic. For example, they want to determine as accurately as possible what the minimum lethal dose might be. If it takes 10 milligrams to kill a mouse, and the mouse weighs 15 grams (about half an ounce), it would take about 6 kilograms (more than 12 pounds) to kill a 95 kilogram human, like me. Pretty safe.
Phase I Trials
Those tests (described above) are preclinical, meaning no human subjects are involved. But fairly early in the process, they need to start doing clinical trials. Phase I trials are short, involve maybe a few dozen subjects, and are focused mostly on safety and on determining things like dosage, how the drug is absorbed and eliminated, and how long it stays in the system.
Phase I trials also try to provide some information on whether the drug works in subjects with the target disease. This can be tricky, especially when there are already drugs with demonstrated effectiveness for that target disease, so sometimes the drug is tested in patients that have failed to improve with existing agents. In the case of an antibiotic, they might get the consent of a patient with a resistant infection. There are a lot of skin and soft tissue infections that don’t respond to standard antibiotics. They could obtain informed consent from a patient with one of those infections and try out their new drug on that brave person to see if it works.
Phase I trials do not compare the trial drug with another drug or with placebo, and they are not randomized or blinded. The level of information they provide is just a start. Needless to say, lots of potential drugs crash and burn in Phase I.
Phase II Trials
Now we start getting serious. If the drug didn’t raise any safety issues in the healthy volunteers in Phase I, and if it demonstrated some effectiveness in subjects with the target disease, the drug developer moves on to Phase II trials, which typically enroll a couple of hundred subjects with the target disease. These trials will be randomized, meaning that patients are assigned to different treatment groups at random, but at the same time making sure that the patient characteristics in the different treatment groups are similar (it wouldn’t do to have patients in one group be older or sicker than those in another group). The trials are double-blinded, meaning that neither the patients nor the investigators know which treatment they are taking. They may be placebo-controlled, or, in some cases, active-controlled, meaning that the new drug is being compared with an existing drug.
Phase II trials look for both efficacy and safety data. Efficacy data in Phase I trials, which are not controlled, can be deceptive. If we stay with infectious diseases as an example, patients recover from infections on their own, and it’s hard to be sure whether it was because of a drug they were taking or because their own immune responses came to the rescue. But if 100 patients taking the new drug recovered from the infection on average several days sooner than patients taking placebo, then the developer knows that the new drug is doing something.
When the new drug is being compared with an existing drug, it may be enough in Phase II trials to meet a pre-determined standard for non-inferiority. This would be a matter of falling within a certain statistical range
Phase II trials also look carefully for adverse effects, which drugs are almost certain to have. Because the subjects in Phase II trials have the target disease, and because Phase II trials can run several months (depending on the disease under treatment), these trials are more likely to detect the kinds of adverse effects that would emerge in the patient population that the new drug is intended for. Certain adverse effects (headaches, nausea) are very common and often go away after the subject has been on the new drug for a couple of weeks. Others are of greater concern, and if they are serious enough, may stop the development of the new drug dead in its tracks.
It’s worth mentioning here that if the drug is intended to treat a life-threatening disease, some adverse effects are probably worth putting up with. But if, on the other hand, it’s meant to do something like treat allergies, the safety standard is much higher.
Phase II trials (and this goes for all clinical trials) have predetermined outcome measures. In an infectious disease trial, the outcome measures are apt to be both biological (the drug eliminated a predetermined percentage of the target pathogen) and clinical (the patient’s signs and symptoms of infection diminished). The principal investigators and statisticians determine whether the trials goals were met, and the degree of statistical significance – i.e., could this have happened at random, or can they be confident that the results were really produced by the new drug.
Phases II trials, because they enroll many more subjects and run longer, start to run into real money. A trial with 200 subjects may require the involvement of as many as 50 investigators at different medical centers. It would not be unusual for a Phase II trial to cost $50 million or more.
Nor is it unusual for Phase II trials not to meet their outcome measures. When that happens, the money is down the drain, and it’s back to the drawing board. Although during the years that they have been working on that particular new drug, the drug developer doubtless learned something of value that might be useful in the development of another drug that might work better.
Usually several Phase II trials are conducted before going on to Phase III. Some Phase II trials are conducted by Big Pharma, but many are done by smaller biotechs, perhaps working with academic centers. And often, Big Pharma makes deals with the biotechs, subsidizing the cost of Phase II trials in exchange for rights to develop and market the new drug if the trials are successful.
An example of this is the deal that Celgene (CELG) just struck with MorphoSys (MPSYF.PK). The deal will total more than $800 million, and it gives Celgene some marketing rights to a drug that will treat multiple myeloma and leukemia. But that drug (so far unnamed) is only as far as Phase II. What we can take from that is that successful cancer drugs are rare and exceedingly valuable, and this one looks promising.
Whatever the arrangement, we can be sure that the eyes of Big Pharma are sharply focused on these Phase II trials.
Phase III Trials
When we get to Phase III, the stakes are high. Drug developers aren’t going to launch Phase III trials without serious expectations that the trials will meet their objectives, and that the drug will be approved for marketing. A Phase III trial might enroll as many as a thousand subjects and could take several years, depending on the target disease – if the disease is acute, trials are shorter, but trials in chronic or recurrent diseases typically last years. And costs, correspondingly, are much higher, frequently in the hundreds of millions.
The FDA looks very carefully at Phase III trials, and goes through thousands of pages of data with the proverbial fine-toothed comb. It’s not enough, in order to gain FDA approval, for the new drug to demonstrate that it’s effective and safe. The FDA wants the drug to be of value to the field – it has to be better in some way than the drugs that are out there already. Of course, there are lots of ways for a drug to be better. The obvious one would be that the new drug actually performs better on what are referred to as “validated outcome measures” than the drugs already in use. An easy example would be a cholesterol-lowering drug – if the new drug lowers total cholesterol by 40% compared with about 35% for a standard statin such as Lipitor, it would be highly likely to be approved, all other things being equal.
Other ways for a new drug to be “better” might include dosing, side effects, or absence of interactions with other drugs. And the FDA might look favorably on a new drug that had a different mechanism of action; for example, the cholesterol-lowering drug Zetia (ezetimibe), from Merck (MRK) works by inhibiting the transport of cholesterol from the intestines to the circulatory system. It doesn’t lower cholesterol as much as most statins, for the simple reason that most of the cholesterol in our system doesn’t come into our bodies as cholesterol – we synthesize it ourselves. But Zetia is another tool in the kit, so to speak, and it easily won approval.
Drug developers don’t go into Phase III trials unless their expectations for success are pretty strong. They most definitely do not treat these as a crap-shoot. They will have started to work with the FDA fairly early in the development process, and as it goes along, the FDA looks at the results of the clinical trials and may point to further information that is needed before the new drug gains approval.
That doesn’t mean that the drug developers necessarily conduct just the type of trials the FDA would most like. I am not privy to the FDA’s hopes and dreams, but a type of trial that the FDA would probably look upon with great favor is just the type of trial that gives drug developers the serious willies: a head-to-head trial between their new drug and an established drug. That would give the regulators the clearest indication as to whether the new drug should be approved or not – i.e., is it “better” than drugs that are already approved. The FDA is not fond of “me too” drugs – they just clutter up the marketplace, confusing patients and doctors alike.
But for the drug developers, this is a high-stakes gamble. They would much rather test their drug against placebo; then they can compare how their drug did versus placebo with the established drug’s record against placebo. And they can structure their clinical trial in such a way as to maximize the chances that their drug will do better versus placebo than the competitor drug did. But going head to head against the competitor drug – that’s risky. They would prefer to go with the noninferiority trials that they did back in Phase II, and hit the FDA with trials against placebo.
Which is not to say that head-to-head trials don’t take place; however, they tend to be the exception, and they’re usually conducted after the drugs have been approved for marketing, rather than as part of the New Drug Application. These tend not to be sponsored by pharmaceutical companies, but by academia or government.
Phase IV Trials
It doesn’t stop with Phase III. Let’s say the drug gets approved by the FDA and is being marketed. If it’s that antibiotic for skin and soft tissue infections, maybe the pharma outfit would like to get an indication for another kind of infection, perhaps lower respiratory infections (LRIs), many of which are caused by the same or similar pathogens. So now they need to study the drug in patients with LRIs.
Or say the drug has gone off patent, and generics are starting to eat into the revenue stream. Pharma has strategies to try to extend the patent life of drugs. One might be to change the dosing, from (for example) twice a day for a week to once a day for three days. To do that, they would have to run a clinical trial demonstrating that it works.
Another would be to replace the racemic form of the drug with an enantiomer. What do those words mean? Well, drugs are complex molecules, a lot of radicals (parts of molecules) bonded together however they happen to fit. They can have the exact same chemical composition, but be structurally different – mirror images. For many drugs, their capacity to do their thing depends on exact fit, like a key in a lock. Think of two keys with the identical profiles, except that one has the grooves on the left side and the other one on the right side. Only one key fits in the lock; the other one is useless. The racemic form of the drug is a mixture of the left hand molecule and the right hand molecule, but half of those aren’t going to be effective. So the pharma outfit figures out which is the one that works – which enantiomer – and markets a version of the drug that consists only of that enantiomer, and obtains FDA approval, based on the premise that there’s no benefit (and there could even be harm) in giving patients a drug half of which is not efficacious against the target illness. But to do that, they need yet another clinical trial.
An example is the development of Lexapro (escitalopram), which is the enantiomer of Celexa (citalopram), marketed by Forest (FRX) and others. It has been suggested by the European Commission that the motive for the development of escitalopram by Lundbeck was purely “evergreening,” i.e., keeping the patent alive.
Yet another strategy is to market a two-drug combination. An example is Lotrel, marketed by Novartis (NVS). It’s amlodipine – Norvasc , from Pfizer (PFE) – plus Novartis’s Lotensin (benazepril). Norvasc is a calcium-channel blocker, and Lotensin is an ACE inhibitor, both hypertension drugs, but with quite different mechanisms. When the individual patents expired, Pfizer and Novartis made a deal to extend the patent lives of their separate drugs by marketing a combination. The way Lotrel is “better” than the individual drugs is that the patient only has to take one capsule, and that’s thought to contribute significantly to patient compliance.
All in all, Phase IV can go on and on and on. There is no end to Phase IV.
Patient Protection with Generics
How much protection do patients have with generics? My short take is, slim to nil. Here are some basics:
Generic drug makers do not conduct trials or, really, test their drugs clinically in any way at all. What they do is copy the active drug molecule and package it as closely resembling the labelled drug as possible. Prescription generics are covered, supposedly, by the prescribing information (PI) of the patented drug. However, once the patent has expired, the original developer/manufacturer/marketer of the drug has no further obligation to update the PI. If they are going to continue to market the drug, they mostly do update the PI for their own protection. But, as often happens, they give up marketing the drug, they have no further interest in updating the PI, which means that if new adverse effects appear, these don’t get listed in the PI.
There was a landmark Supreme Court case on exactly that issue a year ago. A woman named Gladys Mensing of Owatonna, Washington, developed a seriously debilitating disease, tardive dyskinesia, that caused her to lose control of many muscles in her body. She had taken a generic form of Reglan (metoclopramide), initially marketed by Pliva, Inc, which is now part of Teva (TEVA). After the patent on Reglan expired, Pliva turned the manufacture of the drug to a generic drug maker. So when cases of tardive dyskinesia emerged, no warning was added to the PI. Gladys Mensing sued under state law, arguing that the drug maker had failed to warn patients about the risks, but SCOTUS ruled that Federal law prevailed, and the Reglan PI, although outdated, was the only warning label recognizded by law, so she had no recourse. At that time there were thousands of cases open having to do with metoclopramide and tardive dyskinesia, and none of those patients had any legal remedy available.
Another area where patient protection with generics is questionable is the actual compounding of the drug itself. The active drug molecule is supposed to be identical with the original patented drug, but every drug has inactive ingredients, sometimes constituting most of the content of the pill that we swallow. What the provenance and purity of these constituents of generic drugs might be is hard to determine, and there have certainly been cases of generics imported to the US with questionable ingredients. A lot of us remember the tainted heparin that was imported from China, which resulted in 149 deaths in the US about 5 years ago. When pharmaceutical companies in the US and Western Europe import ingredients from elsewhere, wherever that may be, they are required to test these ingredients scrupulously, and they certainly do this – for their own protection, not to mention ours. But the tests sometimes fail to detect dangerous and even lethal impurities, as in the case of the tainted heparin.
However, the distinction that I want to draw is between the level of scrutiny carried out by developers/manufacturers/marketers of patented drugs and the testing done by generic drug makers. I am not accusing the generic sector of having lower standards, necessarily. But the big goal for the generic makers is to be economical, not to say cheap, and the disincentives represented by hugely expensive lawsuits and reputational damage are simply not enough to guarantee equivalent safety.
How Can Drug Patents Be Changed to Address These Issues
My proposal (which has no chance of being adopted) would be to start the patent clock running from the time the drug is approved for marketing in the US, not from the date the patent is first granted, so that if it takes 12 years to get FDA approval, the drug developer still has 20 years to recover the costs of developing and testing the drug and still make a reasonable profit.
This might remove some of the incentive to reap huge profits in the short run, and the patent law could be changed with the clear understanding that more reasonable prices were part of the quid pro quo. It might also remove some of the incentive to look for ways of evergreening the patent. A 20-year lifetime for a genuinely valuable drug sounds like a boon to pharma. And by the time that 20 years has elapsed, chances are something better will come along anyway.
* * * * * * *
I look forward to your comments on this and any of my pieces. If there are subjects you would like me to rant and rave about, drop a little note in the comments. Or if you have questions … of if you think I am way off base and need to make amends… Just let me know. Michael Jorrin (aka Doc Gumshoe)
Personal Capital is an advertiser with Stock Gumshoe, but Travis also uses it every day for his personal accounts and finds it invaluable. Here's what he said: "They offer a great (and genuinely FREE) 'second opinion' for your financial plan, but what I love most is their automated financial dashboard -- it will look at all your assets and debts, tally up your asset allocation, project where you'll be at retirement, and suggest ways to manage risk or improve returns. It's free, I think their free tools are great, and I think it's worth checking out -- you can do so here.