Article In Brief
Alzheimer's disease researchers told Neurology Today that two news reports—one suggesting Pfizer discouraged positive reporting about etanercept for Alzheimer's and the other, reporting a concerted effort to quash research that did not support the amyloid hypothesis—were exaggerated and inaccurate. Not everyone agrees.
Did Pfizer really have “clues its blockbuster drug could prevent Alzheimer's,” as a recent headline in the Washington Post declared? And did a “cabal” of Alzheimer's researchers devoted to the amyloid hypothesis thwart the progress of other approaches for decades, as another recent headline, this one in Stat News, proclaimed?
The reports in two highly respected news outlets have generated dismay among neurologists and neuroscientists who have devoted their careers to the study of Alzheimer's disease (AD) and related dementias, concerned by the articles' insinuations that insidious conspiracies are behind the field's 16-year drought of new medical therapies.
Researchers contacted by Neurology Today said they believed that the Washington Post article exaggerated the significance of epidemiologic evidence indicating that use of the Pfizer drug etanercept (Enbrel) was associated with a reduced risk of AD.
“I was very disappointed by that article,” said David S. Knopman, MD, FAAN, a professor of neurology at Mayo Clinic College of Medicine in Rochester, MN, and an investigator in its Alzheimer's Disease Research Center. “The pharmacoepidemiology evidence is as weak as one can get. It has had a track record of consistent failure in the dementia world.”
Likewise, AD researchers disputed the argument put forward in Stat News that a small group of influential researchers stifled or “suppressed” research not based on the amyloid hypothesis. Even Dr. Knopman, who in April published an editorial in the New England Journal of Medicine stating that “Aβ lowering seems to be an ineffective approach, and it is time to focus on other targets,” took issue with the Stat News story.
The article, he said, “repeatedly characterized an honest scientific disagreement as a ‘cabal’ implying sinister and nefarious intentions. There are many people of impeccable integrity who believe that the evidence supporting a role for amyloid-beta [Abeta] is stronger than for any other alternative hypothesis.”
Not all stakeholders were critical of the articles, however, and even those who were said they agreed with some of the points made. In a statement issued by the Alzheimer's Association following publication of the Washington Post article, for instance, Maria C. Carrillo, PhD, the group's chief science officer, said in part: “Any science that has promise for Alzheimer's disease and other dementias and is not genuinely pursued, or shared with the research community, does a disservice to the millions of individuals facing the disease.... The expectation of the Alzheimer's Association and our millions of constituents is that every possible path is pursued thoroughly and aggressively.”
The June 4 article in the WashingtonPost, described internal Pfizer documents discussing two analyses of medical insurance claims conducted by the company's statisticians. According to the Washington Post, the first analysis compared a group of 127,000 people with an AD diagnosis to an equal number of healthy controls. About three times as many controls as AD patients had been prescribed etanercept, 302 vs. 110, suggesting that the drug might reduce the risk of the disease by 64 percent. A second analysis, using another insurance claims database, reached a similar conclusion.
“Researchers in the company's division of inflammation and immunology urged Pfizer to conduct a clinical trial on thousands of patients, which they estimated would cost $80 million,” the article reported. Pfizer chose not to do so, and the article quoted company officials saying they did so based purely on scientific grounds, because of doubts about the validity of the findings. Those same doubts, they said, led them to not publish the findings.
But the WashingtonPost article noted another, non-scientific reason that Pfizer might have declined to pursue the findings: “Enbrel has reached the end of its patent life. Profits are dwindling as generic competition emerges, diminishing financial incentives for further research into Enbrel and other drugs in its class.”
Despite the insinuations, the article did note that the scientific community already had access to studies suggesting the possible effect that etanercept might have against AD. One study published in 2016 in CNS Drugs analyzed an insurance claims database similar to the databases that the Pfizer researchers examined. It found that among people treated for rheumatoid arthritis, etanercept (but not other anti-TNF agents) was associated with a significant reduction in the risk of AD (unadjusted OR, 0.33; 95 % CI 0.08-0.94; p=0.03; adjusted OR 0.30; 95 % CI 0.08-0.89; p=0.02).
The Washington Post article also took note of a small and relatively brief randomized, placebo-controlled, double-blind trial of etanercept for AD that found no significant effects. Published in 2015 in Neurology, the study randomized 41 people with mild to moderate AD, at a mean age of 72.4 years, to either 50 mg subcutaneous etanercept, once weekly, or identical placebo for 24 weeks. Some trends were seen that favored etanercept, but no statistically significant changes in cognition, behavior or global function.
And although Pfizer did not publish the results of its insurance-claims analyses, it did share the information with a British researcher “to use it in a grant application for a small clinical trial he is undertaking in England,” the article reported.
What's more, the article noted two legitimate scientific reasons why Pfizer had cause for skepticism about the epidemiologic evidence for etanercept. First, another class of drugs that looked promising based on epidemiologic studies—the non-steroidal anti-inflammatory drugs (NSAIDS)—all failed in clinical trials. And the etanercept molecule is too large to cross the blood-brain barrier (BBB).
Following the publication of the article, Pfizer issued a series of statements on its official Twitter account, saying in part: “Our decision not to pursue a broader clinical trial in Alzheimer's disease based on a statistical analysis of a limited number of cases in an insurance claims database was informed by the weakness of the scientific evidence. The portrayal...that we would suppress a promising lead is simply inaccurate.”
In an interview with Neurology Today, Dr. Knopman said he considered the failure of etanercept to cross the BBB to be a “no-go” signal for advancing it into a clinical trial for AD.
“Pfizer was perfectly justified,” he said, in not pursuing a clinical trial. “I have no doubt about that.”
Even its decision to not publish the results of the epidemiologic analysis, he said, was reasonable. “If it had been a clinical trial, then yes, absolutely, they should have published it,” Dr. Knopman said. “But this kind of exploratory study? I think it happens all the time in academia that researchers choose not to publish something because they have doubts about its validity.”
He also noted how unlikely it would be for a pharmaceutical company to abandon a drug if it had convincing evidence that it could treat such a widespread disease as AD.
“Why would they willfully not pursue it? The pharmaceutical industry has many strategies to work around patient expiration issues,” he said. “It defies logic.”
Howard Fillit, MD, founding director and chief science officer of the Alzheimer's Drug Discovery Foundation (ADDF), expressed similar views.
“The fact that there was an association does not prove causality,” Dr. Fillit said. “Whether Pfizer should have published their data or not, I don't know. But there was already public literature on etanercept and AD. It happens in academia, it happens in industry, that the results of a preclinical analysis are sometimes not published.”
Paul S. Aisen, MD, professor of neurology and director of the Alzheimer's Therapeutic Research Institute at the University of Southern California, said of the Washington Post article: “I thought it was ridiculous. Insurance data are highly unreliable when it comes to AD or dementia. The coding is extremely inadequate, and in any event you're looking at association, not causation. There's been a great deal of discussion in the past about Enbrel and whether it might have potential as a treatment, but the consensus has always been, and I certainly agree with it, that the likelihood of peripheral administration of an agent like etanercept to influence the course of AD is extremely small.”
While declining to comment on Pfizer's decision not to publish its analyses of etanercept, a top official at the National Institute on Aging (NIA) said that the policy for NIH-funded research supports transparency and full disclosure of all scientific findings.
“Our mantra here is absolute and complete transparency,” said Eliezer Masliah, MD, director of the division of neuroscience at NIA. “All the data from our large cooperative studies, including with industry and academia, are made publicly available.”
The Amyloid Cascade
Rather than scrutinize any one company or study, the Stat News article took a more wide-angled, and arguably more devastating, look at the years of dominance (and, thus far, failure) of the amyloid hypothesis.
As veteran reporter Sharon Begley wrote: “...for decades, believers in the dominant hypothesis suppressed research on alternative ideas: They influenced what studies got published in top journals, which scientists got funded, who got tenure, and who got speaking slots at reputation-buffing scientific conferences. This stifling of competing ideas, say a growing number of scholars, is a big reason why there is no treatment for Alzheimer's.”
Rachael Neve, PhD, co-director of the Gene Delivery Technology Core at Massachusetts General Hospital, was one of them. Dr. Neve, the article noted, was a coauthor of the 1987 discovery of mutations in the amyloid precursor protein gene, which is associated with increases in amyloid levels, causing AD in middle age.
In an email to Neurology Today, she said she left the field of AD research in 2007 out of “disgust” with the amyloid hypothesis and its defenders.
She told Neurology Today that she was discouraged to pursue Alzheimer's grants that did not focus on amyloid.
“The question is why all the money keeps going to the amyloidophiles,” said Rachael Neve, PhD, co-director of the Gene Delivery Technology Core at Massachusetts General Hospital.
Has the Research Been a Failure?
Dr. Masliah of NIA disagreed with the premise that the past two decades of AD research have been marked by failure.
“The progress we have had in the past 20 years has been nothing less than spectacular,” he said. “We now have a very deep understanding of the genetic basis of AD, and we understand much better the environmental risk factors and disease mechanisms—the role of inflammation, of lipids, of cholesterol. There have been failures with Abeta, but we're learning why. We're making tremendous progress with biomarkers.”
Dr. Fillit took a similarly optimistic view of the AD landscape. “I've been doing this for over 40 years, and I've never seen so much money and excitement coming into our field,” he said. “NIH alone has $2.4 billion to spend on AD this year, which is about five times the amount they had just a few years ago. And we at ADDF are currently supporting over 30 clinical trials. We're very excited about novel mechanisms based on neuroinflammation and epigenetic changes associated with aging.”
Over 100 clinical trials of AD drugs, he said, are underway in the United States. “About three-quarters of them are testing non-amyloid, non-tau targets,” Dr. Fillit said. “That's a big change from just a few years ago, when the majority of trials were directed at amyloid.”
Dr. Aisen, who has been one of the leading advocates of the amyloid hypothesis, said he believes there is a simple reason why clinical trials of drugs targeting amyloid have drawn so much funding. “The real reason is because the theory is so strong, so robust,” he told Neurology Today. “The rationale is still vastly stronger than any other theory, based on the genetics and other evidence.”
Despite the failure of all amyloid-targeting clinical trials so far—including, on July 12, the Alzheimer's Initiative Generation Study 1, involving the BACE 1 inhibitor CNP520—Dr. Aisen said he remains committed to pursuing the hypothesis in clinical trials. He is among the leaders of the Anti-Amyloid Treatment in Asymptomatic Alzheimer's Disease (A4) trial, which is testing whether the investigational Eli Lilly agent solanezumab will prevent the formation of amyloid plaques and therefore the development of AD.
“In no way do I think we have exhausted the amyloid hypothesis,” he said.
While disputing the notion that a “cabal” has thwarted research into non-amyloid causes and treatments of AD, Dr. Knopman said he believes that evidence from clinical trials has proved that Abeta is not the primary cause of AD.
“Paul [Aisen] and I certainly see things differently as far as Abeta,” Dr. Knopman said. “I agree with Paul that testing Abeta lowering in asymptomatic people still hasn't been done, and I am glad it's being done now. But beyond that, I think we need to move onto other approaches, like the tau-lowering drugs, looking at other mechanisms and having an open mind grounded in science and not in beliefs. An honest appraisal of the vast experience we have gained with PET amyloid and tau biomarkers unequivocally shows that Abeta is mechanistically distant from cognitive decline both in time and space. Clinical trials represent a unique window into attribution of causality. How could one not question the causal link between Abeta lowering and cognition in 2019?”
Howard Feldman, MDCM, FRCP(C), professor of neuroscience and director of the Alzheimer's Disease Cooperative Study (ADCS) at the University of California, San Diego, offered a similarly nuanced, if ultimately negative, view of the amyloid hypothesis: “The molecular genetics of amyloid are fairly persuasive,” Dr. Feldman said. “You can't have AD without amyloid pathology. We know that amyloid is necessary, but what we've learned is that it may not be sufficient; in fact, it likely isn't sufficient. You also need tauopathy.”
ADCS is now supporting the T2 AD trial, a multicenter study of troriluzole. Rather than target either tau or amyloid, troriluzole reduces synaptic levels of glutamate.
“I've been very interested in diversifying our portfolio at ADCS,” Dr. Feldman said. “A target that is non-Abeta and non-tau almost by definition becomes of interest.”
Drs. Knopman and Masliah disclosed no conflicts. Dr. Aisen has received research support from Eli Lilly & Co., and Janssen. He has been a paid consultant for Merck & Co., Biogen, Proclara Biosciences, Roche, Kyowa Kirin, ImmunoBrain Checkpoint, Nestle, and Lundbeck.