To help find these patients a central strategy of the companies involves giving hundreds of millions of dollars to a whole host of nonprofit medical groups, commercial companies, and individual physician experts.
Amgen and Sanofi/Regeneron make the two newly approved cholesterol drugs– Repatha and Praluent– which are widely expected to be blockbusters. These companies are fueling an explosion of new programs, some of which utilize innovative data mining techniques to gather massive amounts of data from new sources, and which if they work as intended will likely identify large numbers of new patients who are candidates for the expensive drugs.
The sheer size and scope of these programs are not generally known. The activities funded by industry fall across the entire spectrum of basic and clinical research, continuing medical education for physicians, patient education, and support for not-for-profit groups and patient advocacy groups. (Of course, the companies are also spending millions of dollars on all the more traditional sales and marketing avenues.) The financial tsunami will undoubtedly help secure the close bond between the medical establishment, patients, and industry.
From the beginning it has been clear that people with familial hypercholesterolemia (FH) are the best early candidates for the new drugs. The problem, of course, is that the vast majority of people with FH have never been diagnosed, and FH has not been on the radar screen of most physicians. Many of the new initiatives are designed to find these currently hidden patients.
Much of the newly-funded research may turn out to be valuable, but to one degree or another these efforts will also help contribute to marketing the new drugs. It is important to remember that the long-term clinical benefits with these drugs will not be known for another few years. Another potential problem is that these research efforts could lead in some patient populations to overuse of the drugs, the mistaken diagnosis of FH, an inflated perception of risk in some patients, or a lowered threshold for treatment. It may not always be clear to either doctors or patients that these efforts to identify new FH patients may well turn up patients who have genetic mutations that increase cholesterol but that may not have the same degree of increased risk as other, better known mutations. Some patients, for instance, may have more minor mutations that affect the LDL receptor but may not have serious health consequences.
A key player in the new efforts is the FH Foundation, which describes itself as “a patient-centered non-profit dedicated to research, advocacy, and education of all forms of familial hypercholesterolemia.” The foundation also receives a lot of support from industry.
Here are just a few large collaborations involving industry, researchers, medical organizations, and patient groups:
One of the earliest efforts to find new FH patients is the Cascade Screening for Awareness and Detection of Familial Hypercholesterolemia (CASCADE FH) Registry, started by the FH Foundation in 2013. The registry is supported by Amgen, Astra Zeneca, Regeneron, Sanofi and Aegerion. The registry was established “as a national, multicenter initiative to identify US FH patients, track their treatment, and clinical and patient-reported outcomes over time” using “multiple enrollment strategies to maximize identification of FH patients.”
By all accounts CASCADE has only identified a few thousand previously undiscovered FH patients. The program was in some respects intended as a pilot study. But, as the approval of the new PCSK9 inhibitors came closer, far more money and resources became easily available to experts in the field. Now the companies and researchers have developed much more ambitious and costly programs.
The most ambitious program that I am aware of is the FIND FH initiative, which is designed to identify the 90% of FH patients who have no idea that they have FH by using innovative data mining techniques. FIND FH is financed by Amgen and receives additional support from the American Heart Association. Once again the FH Foundation assumes a central role. Here’s how it is described on the Stanford website:
“In an ambitious tripartite campaign, an academic center (Stanford Medicine), a nonprofit advocacy group (the FH Foundation), and a pharmaceutical company (Amgen) seek to FIND (Flag, Identify, Network, Deliver) FH using computer-based algorithms to troll through large databases including electronic medical records and locate patients who likely have the disease.”
Here’s what FIND FH hopes to do:
“The goals of the researchers involved in FIND FH are lofty. Starting with Stanford patients who have been diagnosed with FH, they will teach computers to find patterns in those patients’ medical records. For this part of the research, they will use medical informatics methods designed by Shah [Nigram Shah, one of the leaders of FIND FH]. ‘We will be using machine learning approaches to build a patient classifier to phenotype patients’, says Shah. ‘As an analogy, think of sorting email into spam or not spam: no one does it using rules any more; it’s done using models built via machine learning.’
“The next step will be to let those computers loose on the medical records of current Stanford patients without a diagnosis of FH. Here the hope is that the algorithms the computers use will be accurate, thereby ‘flagging’ undiagnosed but potential FH patients. Then, through the patients’ designated healthcare provider, those patients will be contacted and evaluated for the most appropriate screening and therapy. Ultimately, the hope is that this will lead to screening of other potentially affected family members as each child of a parent with FH has a 50% chance of inheriting it as well.”
The Stanford Program, it should be noted, is just the first step in a much larger program. At a recent meeting of the FH Foundation meeting Kelly Myers, the chief technology advisor to the FH Foundation and the founder of Atomo, a for-profit machine learning and data mining company, said that the machine learning techniques will ultimately be used on data from 89 million people. From this effort they hope to find 1.2 million FH patients.
Regeneron and Geisinger
Not to be outdone by Amgen, Regeneron is spending $120 million on a large research effort in partnership with Geisinger Health System. Overall this effort involves a broad search for genes and drug targets well beyond FH. By comparing “genetic information against medical histories, Geisinger and Regeneron hope to eventually develop new means of diagnosing, preventing and/or treating medical conditions – before they cause significant harm. Some participants may also receive information that could be useful in their own medical care.”
Here is how it will work:
“Geisinger will collect blood samples from consenting participants, along with relevant medical information. Regeneron will perform genomic analysis on the samples, in the hopes of identifying new information on genetic variants that may be associated with specific diseases and health conditions. All samples and records will be confidential – with personal information removed and all data scrupulously maintained through Geisinger’s secure MyCode® Community Health Initiative repository.”
Regeneron, I’ve been told, will spend $100 million to genotype 100,000 patients. The company will own the overwhelming majority of rights leading to new drugs, but Geisinger will be able to use the genetic information in clinical applications.
At a recent meeting of the FH Foundation, Geisinger’s Michael Murray spoke about his work, which involves linking EHRs and genetic sequencing, in a collaboration with the Mayo Clinic and Cincinnati Children’s Hospital. With regard to FH, the plan is to develop an algorithm, test it at Mayo, and validate it at the other two sites. Murray also said that he was collaborating with Josh Knowles, a key figure in the Stanford program and the CASCADE Registry. (Knowles is also the chief medical officer of the FH Foundation.)
Murray said that Geisinger intends to perform cascade screening on family members of previously identified FH patients. Their current biobank data suggests that approximately one out of 174 people will have an FH mutation. According to their EHR data, 24% of people have LDL levels over 155 mg/dl, 6% have levels over 190, and 0.2% have LDL levels over 250. It should be noted here that many lipid experts have warned against “indication creep,” which in this case would involve lowering the LDL threshold level for treatment.
ACC and Amgen: There An App For That!
In 2014 Amgen gave money to the American College of Cardiology to support a new program called LDL: Address the Risk. An unusual and unique product of the program is the ACC Statin Intolerance App, for iPhone and Android, that “uses clinical guidelines and best practices to help” doctors evaluate, manage, and treat statin intolerance.
For all the recent talk about statin intolerance it is not generally appreciated that until the PCSK9 inhibitors came along the idea that there was widespread statin intolerance was sharply disputed by nearly all the experts in the field. Statin intolerance became a hot topic on the basis of research and CME largely supported by the makers of the new cholesterol drugs. Incidentally, among people in the medical community the manufactured explosion of interest in statin intolerance is widely known and not even controversial. It is the type of phenomenon that everyone knows about and talks about but rarely gets discussed in public.
The initiative started with a “think tank” meeting that helped set the overall agenda of the program:
- LDL: Address the Risk Think Tank: The ACC held a clinical think tank meeting to bring together representatives from relevant medical specialty societies and other stakeholder groups for an interactive discussion among top experts from a broad spectrum of disciplines. The key issues and discussion from the think tank meeting guide the overall initiative.
I have asked the ACC to answer a few questions about this meeting. The answer to these questions may also shed light on the overall Address the Risk initiative. Who determined the agenda? Who chose the participants? Who attended the “think tank” meeting? Was this an open meeting? Is there a public record of the proceedings? Did Amgen employees or representatives attend the meeting? Did anyone from other companies attend?
An ACC spokesperson told me that “while there isn’t a public agenda or proceedings, this ACC In Touch blog has a high level overview of the meeting. Additional reports from the meeting will be published soon.” In addition, the ACC said that “None of the supporters were involved in planning the meeting, selecting participants, or selecting the planning committee. Participants were selected to represent a range of potential stakeholders, including professional societies, integrated health care systems, pharmacy benefit managers, industry, informatics, and patient advocates.” (Update: The ACC informs me that “medical officers from supporters Amgen, Pfizer, CVS Health and Eli Lilly” attended the meeting.)
The chairman of the program was Christie M. Ballantyne (Baylor), who has extensive and frequent ties to industry. I have always been concerned about meetings like this. It is unclear to me whether it is possible for these meetings to be truly independent and unbiased. For instance, were there are any speakers at the Think Tank meeting who opposed the early approval of the PCSK9 inhibitors, or who focused on the high price of the drugs, or the dangers of over-diagnosis? Too often when it comes to an industry-funded think tank all the thinking is on only one side of the issue.
On Working With Industry– Peter Berger’s Perspective
To get an inside perspective on how these deals get going I spoke with Peter Berger, who recently became the Senior Vice President of Clinical Research at the North Shore Long Island Jewish Health Care System. Before that he was at Geisinger, where he was involved in the Regeneron negotiations (though he emphasized that he was not speaking for Geisinger and he was not involved with the final deal).
Berger said that at North Shore-LIJ, his main concern is to identify untreated patients who can benefit from treatment, patients whose LDL is way too high on a statin. “Most health care institutions, including my own, have too many patients with very high LDLs (for example, greater than 400 md/dL untreated, or greater than 250 mg/dL on the maximal dose of a potent statin), and do a poor job of screening the family members of such patients,” he said. Berger said his current thinking is that it probably does not matter to the patient what the actual cause of the high LDL is; it leads to heart disease and stroke and needs to be reduced, regardless of the cause. ““I would argue, however,” Berger said, “that the families of these patients should be screened, since so many of them have a genetic cause of their high cholesterol. Unfortunately, most institutions do a bad job at screening family members of such patients.”
Berger said he reached out to the companies to see if they would help explore the novel use of informatics to help identify such high risk patients and get them to goal. He said he hopes to establish a program at NorthShore LIJ with company support to identify such high risk and insufficiently treated patients. Berger is keenly aware of the potential economic impact of the PCSK9 inhibitors and said that the study he is proposing contains an analysis of the economic impact of the drugs. In fact, he said that the drug companies readily agreed that such an analysis be a part of any study they fund.
But, Berger admitted, he was extremely concerned about the price of drugs in general, including the PCSK9 inhibitors. “The pricing of drugs in this country is obscene. In the case of the PCSK9 inhibitors, I hope a third or a fourth PCSK9 inhibitor hits the market and drives the price down.”
But, he said, he’s “not concerned at all that by identifying and screening more patients we will be expanding the pool for these drugs,” as long as the patients receiving such therapy are appropriately selected.” Berger emphasized that all such patients should be counseled about the role of diet and exercise, but did acknowledge, however, that intensive lipid lowering drug therapy is virtually always needed for patients with LDLs in the 400 range. Ultimately, says Berger, “when I have a hyperlipidemic individual with LDL values that high, my first obligation is to the patient, and not to societal needs to reduce the cost of health care.”
Berger also insisted that any industry support for a screening program must be “product agnostic.” In fact, Berger said, that would be specifically included in any contract. Any company that supports this research knows not to expect preferential use of its drug, said Berger. “Ideally, multiple companies, including competing companies, would support such clinical initiatives. Done properly, providers would not even know what the sources of funding of such initiatives are.”
Pfizer and the International Atherosclerosis Society (IAS)
Pfizer, which has its own PCSK9 inhibitor under development, is working with the IAS to give money away for “independent initiatives that update and implement lipid management training curricula and tools to include new advances in lipid treatments using monoclonal antibodies and cardiovascular genetics such as familial hypercholesterolemia.” Pfizer and the IAS have issued a request for proposals. Of course, the products resulting from this sort of RFP are very likely to be used by marketers and sales representatives to influence both patients and doctors.
CME and Advertising
It’s beyond the scope of this article to provide even a rough assessment of the massive amount of dollars being spent on more traditional marketing efforts like advertisements, promotional talks, and CME programs. I will only state here that all of these are being done on a possibly unprecedented scale. Just yesterday I found this image on Twitter (below). As far as I know this is the first time the august New England Journal of Medicine has put an advertisement on its cover, though I am unsure why it is upside down. (I confess that I am also equally shocked that there’s still even a print edition of NEJM). I can only imagine that Amgen paid quite a bit of money to reach this milestone.