Archive for September 2011
It is surprising that a recent commentary in the September 21st issue of the Journal of the American Medical Association entitled “The FDA – A Misunderstood Agency” seems to have received little attention. Written by Dr. Joshua Sharfstein, the current Secretary of the Maryland Department of Health and Mental Hygiene and the former FDA Principal Deputy Commissioner, the article presents both an explanation and defense of the FDA’s policies and procedures. I wish Dr. Sharfstein went further in his comments.
The FDA, while not perfect, is often criticized unfairly. Here are three typical examples. Invariably, after a new drug is on the market for a few years, the FDA will announce that this new medicine has some new safety issue that hadn’t been seen previously. This issue is not unusual. When a new drug is approved, it generally has been tested in perhaps 10,000 patients for a year. Once it is on the market, if it is a chronically used drug, it may get used by millions of people over the period of a few years. As it is more broadly used, more information is gathered on the drug and the FDA will react to this. Unfortunately, when this occurs, there will be members of Congress (Sen. Grassley comes to mind) who will hold a press conference and rail against the FDA for its lax attitude in approving new drugs. There are ALWAYS risk-benefit issues with any medicine, old or new. Some politicians don’t seem to get this. The FDA does.
Interestingly, there are those who criticize the FDA for being too slow in approving new drugs. This can occur in diseases where this is no adequate treatment for life-threatening diseases. In these cases, there are patient advocacy groups who want certain medicines approved as quickly as possible. Their view is that, if there is a drug in trials that can save the lives of themselves or their loved ones, the FDA has no right to prevent patients from getting this medicine. In these cases, the FDA generally has significant questions about whether the drug indeed is effective and will not approve it without hard data showing that the drug works. It should be noted that the FDA has a good track record of reacting with alacrity to health crises, as evidenced by their actions in approving AIDS drugs in the early days of that epidemic.
But what I find most exasperating is the criticism that the FDA is beholden to the pharmaceutical industry because of the Prescription Drug User Fee Act (PDUFA). Back in the 1990s, there was a “drug lag” – drugs were getting approved in Europe years ahead of their approval in the US. Congress was outraged and demanded an explanation. The FDA correctly said that the organization was grossly understaffed and that new resources were need to deal with the backlog of drugs awaiting approval. Rather than allot the FDA more funds, Congress enacted PDUFA which essentially was a charge to the pharmaceutical companies on the filing of their New Drug Application (NDA). In 2011, this fee is $1,542,000 for each NDA. Critics say that, because the FDA is dependent on these user fees to support their division, FDA staffers feel beholden to pharmaceutical companies and, as a result, try to do all they can to help these companies. This view is ludicrous.
To a certain extent, the FDA is in a no win position. It has to judge what medicines the US population should or shouldn’t take. This is rarely a black or white decision. As was stated above, all medicines have side-effects. The question the FDA must answer is what is the risk-benefit of each new drug–this is always a judgment call. The FDA tries to work closely with all of those dependent on its actions – patients, physicians, companies and government officials. The FDA is not infallible, but it gets it right most of the time. It’s time to respect this.
The obesity problem in America is constantly in the news these days. Recommendations from all sorts of people, from Dr. Oz to the First Lady, Michelle Obama, are appearing on eating healthier, getting more exercise, trying the latest new diet, etc. And these efforts are very well justified. It is scary to see data on the growth of America’s girth. As shown on the Center of Disease Control’s website (http://www.cdc.gov), 34% of US adults are obese as defined as having a body mass index (BMI) of >30. To put this in perspective, a man who is 5’9” would have to weigh 205 pounds to have a BMI of 30. For a woman of 5’6”, she’d have to weigh at least 186 pounds. People with this type of physique are headed for diabetes and heart disease.
Those who have heard me speak on this topic know that diet and exercise should always be the primary method for losing weight, lowering blood pressure and controlling LDL cholesterol. However, this doesn’t work for everyone. Having a safe and effective drug to treat this growing problem should be a major priority of the biopharmaceutical industry as well as the FDA. One would think that the industry’s pipeline would be full of exciting new compounds that can help people lose weight and that the FDA would be granting priority review status for such agents. Actually, that’s not the case.
Here’s the problem. The FDA is well aware that any drug that they approve for obesity would be immediately used by millions of people. Such a medicine would not be used for just the morbidly obese. People anxious to lose a few pounds for cosmetic reasons would clamor for it. The FDA is acutely aware of this and so is wary of unleashing such a drug without EXTENSIVE safety and efficacy data.
It is hard to criticize the FDA for taking such a position. Diet drugs have a history of having modest efficacy but significant side effects. Most notorious was a drug known as fen-phen which was pulled from the market in 1997 as a result of valvular heart disease. Thus, the FDA has taken the following stance. Any drug for obesity must not only show efficacy, but it must be studied in patients over the course of 2 – 3 years to: 1) demonstrate long term efficacy in weight loss; 2) show long term safety in patients who will take this drug over the course of many years. On this latter point, the FDA wants to see outcome studies – studies in which patients on drug are followed for multiple years to see what the impact of the drug is on heart attacks and strokes. Such a request is not unreasonable. After all, a drug that causes weight loss should, at the very least, not increase heart attacks and, in theory, should cause a reduction of heart disease.
The requirement of such an outcomes study has caused the biopharmaceutical industry to shy away from this disease area. The reason is simple: you can invest between $500 million to $1 billion dollars in an obesity R&D program over 10 or more years only to learn at the last step, the outcome study, that your drug is not approvable. That is a level of risk that is not acceptable to many investors.
This whole topic has been recently revisited with a new drug for obesity called Contrave. This drug is being developed by a small biotech company, Orexigen. Contrave is actually a combination of two drugs that have been on the market for decades: naltrexone (used to treat alcohol addiction) and bupropion (a smoking cessation medication). The scientific rationale for using this combination is as follows. Naltrexone blocks opioid receptors in the brain, reduces the reinforcing aspects of addictive substances, and negatively alters the taste of many foods, including sweets. Bupropion increase dopamine activity in the brain and this appears to lead to appetite suppression and increased energy expenditure.
This sounds pretty exciting. Orexigen is utilizing two marketed agents and combining them into a single pill that dulls your appetite, reduces your craving for sweets, and causes your body to expend more energy. More importantly, clinical trials show that Contrave actually works. Patients on the drug for 56 weeks lose an average of 6% of body fat. This is not dramatic weight loss but it can help in one’s overall weight loss program.
Nevertheless, the FDA initially rejected this drug. Despite the fact that Contave is the combination of two known drugs, the FDA wants to see the results of a long term cardiovascular outcomes study before approval. Orexigen recently published the results of their meetings with the FDA which outlined the final regulatory requirements for Contrave approval. Essentially, the FDA has asked for a trial comparing Contrave to placebo in a population of overweight and obese patients who have an estimated background rate of 1 – 1.5% annual risk of major cardiovascular events. The length of this study is dependent upon seeing a specific number of cardiovascular events (heart attacks, strokes, revascularizations, etc.) for the FDA to be able to judge whether Contrave poses no risk in this population.
Thus, the road to approval is clear for Orexigen. However, it is not without risk. First of all, this study may require as many as 10,000 patients and generally at a cost of $10,000 per patient. This will require an investment of at least $100,000,000 and Orexigen will need to go out and raise this money from investors in order to be able to pay for this study. Second, while the FDA might approve this drug if the study finds that patients lose 6% of body fat with no increase in cardiovascular effects, will such a modest decrease in weight result in insurance companies paying for it? Payers might take the stance that this is a “bikini drug” – it causes you to lose some weight, but there is no long term health benefit from such weight loss. They could deem Contrave to be a “cosmetic drug” and not a real medicine. Thus, patients who want Contrave would have to pay for it themselves – which they might do.
Of course, Orexigen could hit a home run with the following result: significant weight loss is seen, the drug is well tolerated and, most importantly, a significant drop in adverse cardiovascular results is seen as a result of this medicine. Such is the nature of pharmaceutical R&D – high risk, but the potential for high rewards for the patient and the drug company. And one further point – we are unlikely to have the answer until 2014. You may, therefore, want to focus on diet and exercise as a way to control your weight.
Rightfully, much has been made of the expiration of Pfizer’s US patent for Lipitor (generic name: atorvastatin), which will occur by the end of this year. Lipitor is the most notable of the statin class of lipid-lowering drugs, compounds that have helped to lower the rates of heart disease-related deaths around the world. Lipitor’s commercial success was unprecedented. At its peak, it had worldwide sales in excess of $13 billion dollars annually. Even with its patent expirations in recent years in various parts of the world, it still rang up $10.7 billion in 2010 keeping it the world’s number one selling drug.
Lipitor’s success can be attributed to a number of factors, but most notably the large (and expensive!) clinical trials Pfizer carried out over a number of years that showed how the treatment of patients at risk of heart attacks and strokes have greatly reduced incidents of adverse events when treated with Lipitor when compared to placebo. Its efficacy, coupled with its remarkable long-term safety profile, have made Lipitor a household name.
In November, when the US patent expires, generic drug makers will be able to begin selling atorvastatin (the active ingredient in Lipitor) and they will do so at greatly reduced prices. It is not unusual for the costs of generic drugs to be 10 – 20% of what the branded drug costs. Thus, Pfizer’s sales of Lipitor will drop dramatically, as pharmaceutical industry analysts have noted for the past few years.
What does this have to do with AstraZeneca? Well, AZ sells its own statin, Crestor (generic name: rosuvastatin), and with 2010 sales of $5.6 billion, it is a major product for the company. AZ has always touted Crestor as the superior compound with slightly better lowering of LDL (“bad cholesterol”) and slightly better elevation of HDL (“good cholesterol”). Unfortunately for AZ, Lipitor had been on the market for a number of years before Crestor made it, and the differences in cholesterol modulation between the two drugs was not significant enough for physicians, who were already very familiar with Lipitor, to switch their patients to the newer drug. Thus, Crestor, while still an important and heavily prescribed drug, trailed Lipitor.
But with Lipitor going generic, AZ has a major problem – one that is a reflection of the cost-consciousness of health care systems. The generic price of atorvastatin will likely be one-tenth of the cost of Crestor. Thus, payers–insurance companies, Medicare, etc.–will insist that patients newly diagnosed as needing a cholesterol lowering agent be prescribed the less expensive atorvastatin rather than Crestor. Even more concerning will be those health care plans that will try to have their Crestor patients switched to atorvastatin in an attempt to reduce their costs. The arrival of generic atorvastatin, therefore, not only impacts Pfizer, but might also cause a drop in Crestor sales as well.
AZ has been acutely aware of this issue for years and has tried to take steps to differentiate its drug from atorvastatin. Interestingly, Crestor has a biologic property that differentiates it from atorvastatin – it lowers an inflammatory marker call C-Reactive Protein (CRP). CRP is associated with a variety of inflammatory conditions, including inflammation of the arteries. Scientifically, it is thought that high levels of cholesterol damage arterial walls and the process of repairing the artery wall results in the beginning of atherosclerosis. High CRP levels could therefore be a signal of early atherosclerosis.
For unknown reasons, Crestor lowers CRP levels and this distinguishes it from other statins. Is this medically significant? AZ thought so and to prove it thecompany ran the SATURN trial. This study compared Crestor directly with Lipitor in measuring the build-up of plaque in the arteries of patients with heart disease. AZ’s thinking, and hope, was that the differences in the biochemical profile of Crestor vs. Lipitor would translate into a meaningful clinical difference in slowing the progression of atherosclerosis. If the theory held up, it would be a great result as AZ would be able to clearly show that Crestor was the superior agent.
The Lipitor patent expiration won’t just impact Pfizer’s net sales in 2012; AZ’s Crestor sales are also going to feel the impact.
A Forbes article called “Serial Lifesaver” by Matthew Herper focused on Dr. Peter Hirth, the CEO of the biotech company, Plexxikon. There is no debating that Hirth has had a very productive career in the biotech industry. He’s played a key role in the discovery of three successful drugs, including Recormin for anemia (sold by Roche), Sutent for kidney cancer (sold by Pfizer) and the recently approved Zelboraf for melanoma (to be sold by Roche). Herper’s piece set out to find the reasons behind Hirth’s stellar track record.
Herper concludes: “Hirth’s serial success stands out vividly in a pharmaceutical industry that has for years suffered from a profound innovation drought. He says that large companies should learn from what Plexxikon has done with a staff of only 43, explaining that if he were running a big business like Pfizer, he would form small units of 40 or 50 researchers and fund them sparingly but promise them royalties on any drugs that succeeded. Productivity, he says, would go way up.”
There are a few issues with this statement. First of all, in big companies like Pfizer, the drug discovery project teams do, in fact, contain 40 – 50 scientists. This was true in my day with the teams in disease areas such as osteoporosis, atherosclerosis, ophthalmology and urology. Yes, there were larger teams in broad disease categories such as Neurosciences, but this area was made up of teams working on depression, schizophrenia, addiction, etc. Breaking this zone down into its components shows the 40 – 50 scientist concept still holds. During my Pfizer tenure, the one exception was in oncology, where we had a group in excess of 200. Contrary to the view that size is detrimental to productivity, this group was the most productive we had at the time as judged by the number of clinical candidates it produced relative to the group’s size.
Joshua Boger, an excellent scientist in his own right and the former CEO of Vertex, recently said that, in his experience, the relative size of the respective discovery groups doesn’t matter – it’s having the right people. I absolutely agree with him.
But the bigger issue for me with Hirth’s statement is the fact that the overall process of going from an idea to the discovery of a clinical candidate to the conversion of this candidate to an approved new medicine DOES take an army. It’s amusing to read “Plexxikon created Zelboraf… (and) Roche helped the tiny biotech test it.” This “help” undoubtedly involved literally hundreds of Roche scientists to develop a formulation that enabled Zelboraf to be tested in the clinic, synthesize Zelboraf in sufficiently large quantities for clinical testing, run the necessary toxicology studies in animals to show that the drug was safe, and carry out the full gamut of Phase 1, 2 and 3 trials to justify FDA approval. My guess is that if you created a list of all the people who were involved in the discovery and development of Zelboraf, it would number at least 500 people.
And that bring me to my final issue with Hirth’s stance, that of royalties. There are often seminal discoveries made in the development of a new medicine that extend beyond the discovery laboratory. Many a drug program has been saved by a key observation in the clinic on a drug’s activity, or a breakthrough new formulation that allows the drug to be suitable to be made into a pill or capsule, or a key toxicology study. I would argue that such a royalty scheme is unworkable because, in my experience, you would have to grant royalties to at least a dozen people who have made a seminal contribution. Unless the royalty was miniscule, it would greatly cut into the revenues for the company.
And finally, it is my experience that scientists are an extremely highly motivated bunch. They are driven by making use of their scientific talents to discover and develop something that, if successful, could benefit millions of people around the globe. I am not sure that the potential for a royalty would cause them to work harder. They are already incredibly dedicated.
On August 30, Jonathan Rockoff wrote a great story in The Wall Street Journal on a newly FDA-approved Pfizer lung cancer drug called crizotinib (to be sold as “Xalkori”). The article, “Pfizer’s Future: A Niche Blockbuster,” details the history behind the discovery and development of this breakthrough medicine. Essentially, crizotinib is an ALK inhibitor that targets the genetic abnormality that causes about 5% of new lung cancers that are diagnosed each year. The beauty of this drug lies in the fact that a patient newly diagnosed with lung cancer can undergo a genetic test to determine if his or her lung cancer is ALK dependent. If it is, then the physician now has a drug that is almost guaranteed to work in this patient.
The crizotinib story is not unique. Roche also has recently launched vemurafenib, a targeted drug for melanoma. We are moving away from the days when the only drugs that an oncologist had to treat his or her patients with were general cytotoxic compounds that come with myriad toxicities. The rapid advances being made in understanding the genetic basis of disease have led to the discovery and development of new drugs like these. But pharmaceutical R&D has been moving into this direction for a decade. After all, drug discovery is not an overnight process and the research programs that are yielding these breakthroughs were started a decade ago.
Thus, I am surprised by some of the reaction that has appeared following Rockoff’s article. Twitter has lit up with comments like: “Can targeted drugs save Big Pharma?” “Perhaps the pharmaceutical industry has come kicking and screaming (to this)” “New cancer pill gives hope, new strategy” “There has been a paradigm shift.” There still remains the view that pharmaceutical companies are only interested in Lipitor-like blockbusters (peak sales of >$13 billion) and that smaller commercial opportunities are disdained. Of course, every company would love to have a drug with enormous sales. But very significant commercial returns can be made with crizotinib that more than justify its clinical development.
A number of years ago, I was asked by an industry analyst if I felt that the inevitable fractionation of patient populations by genetic subtype would be a death knell for big pharma. His rationale was that diseases like lung cancer would be treated with agents specific to a particular mutation for a subtype of lung cancer. Designed for very few patients, this process would result in treatments that are far less commercially viable than a lipid-lowering agent designed to treat millions. I explained that most drugs that are broadly prescribed do not work in a significant percentage of patients. The current paradigm is for physicians to prescribe drugs and then monitor their patients to see if the drugs are working. Often, patients come back complaining that their drug hasn’t helped, leading the physician to try something new. This overall process is costly, inefficient and frustrating to all concerned. By having specific, targeted drugs, physicians will have the confidence that the new medicine will help their patients, the patients will have confidence that the pills they are taking will help them, and payers will have confidence that they are paying for a worthwhile treatment. In such a world, while the number of patients taking the drug is relatively smaller, better pricing for the drug should be obtainable due to improved drug effectiveness. Basically, you might not have a few 10 billion dollar selling drugs, but you’re likely to have many billion dollar products.
This scenario is being borne out by crizotinib. In his article, Rockoff reported that market analysts are predicting peak sales of crizotinib to exceed $1 billion. If true, this drug would be in the top third in sales for all of Pfizer’s portfolio. Some might argue that the only reason that crizotinib will be a commercial success is that, as a cancer drug, it can command premium pricing. Perhaps this is true. But hopefully, genetically targeted drugs will be developed for other polygenic diseases like depression, schizophrenia, migraine, etc. These patient populations should be large enough to support niche-like products with a reasonable price.
Targeted drugs have been envisioned and sought for over a decade. The first wave of these are now hitting pharmacy shelves. This is great news for patients and physicians – and not too bad for the companies developing them either.
It is not surprising that, since he announced that he is stepping down as CEO of Apple, a number of articles have been published which have praised Steve Jobs. It is well justified that he should be compared to Henry Ford and Walt Disney as an industry leader who literally changed the world with his company’s advances in technology. He wasn’t always successful and his early failure with the Apple I computer undoubtedly taught him some important lessons. But his subsequent great successes can provide important lessons for those in other industries.
Of all the things that I have read, one comment particularly resonated with me: Steve Jobs had little use for market research. He felt that the public didn’t often realize what they wanted, that it was hard for people to envision that they would want something that didn’t exist. I believe a similar phenomenon exists in the pharmaceutical world.
If asked about what new medicines are needed, the general public as well as physicians would, of course, respond that they want to cure all of the major diseases: cancer, Alzheimer’s Disease, heart disease, etc. You really don’t need a great deal of market research to figure this out. But history shows that there are a number of conditions where pharmaceutical market research has been way off in the assessment of the value of a new medicine.
An early example occurred in the field of anti-ulcer medication. It’s hard to believe now, but in the 1960s and 1970s, some of the more common surgeries done were gastric resections to repair intestinal ulcers. The basic tenet back then was that a major cause of ulcers was excess stomach acid. While there were medicines to treat ulcers back then, predominantly anti-cholinergics, these drugs either weren’t very effective or they were poorly tolerated. Market analyses done indicated that there wasn’t much of a demand for agents that inhibited stomach acid secretion since there wasn’t much of a demand for the existing treatments. However, the advent of the histamine H-2 antagonists like Tagamet and Zantac blew this perception away when they proved to be blockbusters. The reason for this was simple: these compounds were very effective and had a great safety profile. These drugs were followed by the even more potent proton pump inhibitors, which also proved to be blockbusters.
These drugs showed that there was a need for good acid secretion inhibitors, but physicians and patients didn’t have an appreciation that such agents were possible. Yet, these agents effectively changed the practice of medicine in that previously common gastric surgeries are now rare.
An even more important example of an underappreciated market was in the cholesterol lowering arena. In the late 1980s, it was unclear as to what the “normal” level of cholesterol should be. Even if high cholesterol was deemed to be a safety issue for a patient, the need for a drug to treat it was questioned. I once heard a marketing expert ask: “Why would someone who was otherwise healthy take a pill for the rest of their lives to treat a condition that can be managed with diet, exercise and red wine?” Despite such views, Merck led the way with the research into statins, compounds that lowered LDL also known as the “bad” cholesterol. When Merck conducted its 4S trial with simvastatin showing that lowering LDL cholesterol with this drug reduced heart attacks and strokes, the treatment of heart disease changed forever.
Market research can also backfire. In the late 1990s, Pfizer developed an inhaled version of insulin called Exubera. Insulin is the best treatment for diabetes. However, it must be injected to be effective. Since the majority of people with mild diabetes (early stage Type 2) can be controlled with oral medications, they rarely are prescribed injected insulin. However, over time, the efficacy of oral medications begins to wear off and patients need insulin. However, they are reluctant to use this therapy. Market research showed that people had a phobia of needles and that, even though they needed insulin to help control their disease, they resisted this form of treatment. Furthermore, market research indicated that having an inhaled version of insulin would be a major factor in having these patients moving onto insulin.
After a decade of R&D and hundreds of millions of dollars of investment, Pfizer launched Exubera. It proved to be a major failure. Sales were so low, that Pfizer pulled it from the market. There were a variety of reasons for this failure, but most notably was the fact that, while patients weren’t enamored with using needles, they weren’t particularly fond of carrying a somewhat bulky device that was likened to a bong. Furthermore, the needle delivery technology improved dramatically over the time Pfizer was developing Exubera and the fear of needles was diminished by the time Exubera was launched.
Market research has its place in helping to set R&D priorities. However, it isn’t foolproof. It should be one factor used in making portfolio decisions. But Steve Jobs has it right. Sometimes, a new medicine, which hadn’t been anticipated, helps patients and physicians better control their disease or condition.