Aducanumab Approved to Treat Alzheimer’s Disease
Quick Links
In a watershed moment for the field, the U.S. Food and Drug Administration today approved the first Alzheimer’s treatment that tackles the underlying disease pathology. Biogen and Eisai’s anti-amyloid antibody aducanumab, now called Aduhelm, removes amyloid plaque from the brain, but its ability to slow cognitive decline has remained uncertain in trials to date. It got the green light under the FDA’s accelerated approval program, which grants a marketing license based on a demonstrated effect on a surrogate endpoint—in this case amyloid plaque reduction—that is expected to lead to a subsequent clinical benefit. This pathway requires post-approval trials to demonstrate such a benefit. Approval can be withdrawn if the treatment does not pan out, though the FDA has drawn criticism for lax enforcement on this point.
- FDA grants marketing license to aducanumab via its accelerated approval pathway.
- This requires post-approval studies to demonstrate clinical efficacy.
- The label provides no guidance on which patients would benefit most.
Reactions among Alzheimer’s researchers spanned the gamut. Some were enthusiastic. Jeffrey Cummings at the University of Nevada, Las Vegas, wrote, “The FDA has taken a courageous stand to approve a drug when there were many loud—and often well-reasoned—voices to the contrary … This is a huge step forward for Alzheimer’s care and research.” Some were cautiously pleased. “I am happy for our patients with Alzheimer’s disease and think this was the best path forward: Approve now and require demonstration of clinical benefit in the near future,” Ron Petersen at the Mayo Clinic in Rochester, Minnesota, wrote to Alzforum.
Others were skeptical. “Given that the FDA advisory committee was clear in its consensus that the data presented did not show a clear signal of efficacy with aducanumab, today’s decision is very surprising, to say the least,” said AdCom member Madhav Thambisetty at the National Institute on Aging in Bethesda, Maryland. David Holtzman at Washington University, St. Louis, wrote, “The FDA needs to hold Biogen to this Phase 4 trial requirement to ensure that this treatment truly benefits patients and that we are not wasting very large amounts of dollars on an ineffective therapy.”
The diverse opinions reflect conflicting data from clinical trials. Biogen’s marketing application rested on two incomplete Phase 3 studies, one positive and one negative, which generated intense controversy (Dec 2019 conference news; Nov 2020 news). The FDA’s advisory committee recommended against approval (Nov 2020 news). The American Geriatrics Society spoke against approval, questioning whether a putative functional benefit of the treatment outweighs its risk of brain edema and microhemorrhages (ARIA).
On the other hand, patient advocacy groups lobbied for aducanumab. The Alzheimer’s Association launched its “More Time” social media campaign in May, arguing that even a modest slowing of clinical decline would be meaningful to patients and their families. Us Against Alzheimer’s urged its members to write to the FDA in support of approval.
Some believe these influence campaigns did a disservice to the science. “I do worry when the scientific processes of evaluating potential efficacy of new drugs are influenced by nonscientific factors,” Todd Golde at the University of Florida, Gainesville, wrote to Alzforum. Many researchers pushed for another trial, noting that the data did not prove efficacy (e.g., Apr 2021 news).
In the face of these pressures, the agency turned to its accelerated approval mechanism. Patrizia Cavazzoni, director of drug evaluation and research at the agency, noted that this pathway is “intended to provide earlier access to potentially valuable therapies for patients with serious diseases where there is an unmet need, and where there is an expectation of clinical benefit despite some residual uncertainty regarding that benefit.” Cavazzoni added that this regulatory option was not discussed at the advisory committee meeting. In a letter to the committee, the FDA’s Billy Dunn, who runs the neuroscience office, explained the agency’s reasoning, citing “substantial evidence that the drug reduces Aβ plaque, and that this reduction is reasonably likely to predict clinical benefit.”
Lon Schneider at the University of Southern California, Los Angeles, believes the agency had backed itself into a corner by its enthusiastic embrace of the aducanumab data at AdComs. “This is face-saving by the FDA, who were obviously closely involved and strong advocates for aducanumab’s ‘substantial’ effectiveness … In a sense, this was all decided before the advisory committee meeting in November 2020,” he wrote to Alzforum.
Other researchers acknowledge continuing qualms about the efficacy data, but are now turning their minds to practical issues of how this approval will affect clinical practice and research. Gil Rabinovici at the University of California, San Francisco, raised questions about which patients should be eligible, and how they will be monitored. “I am hoping the design of the Phase 4 study will allow us to fill critical gaps in knowledge that remain unanswered following the Phase 3 trials,” he wrote to Alzforum. In an interview with CNBC, Biogen CEO Michel Vounatsos said the company has nine years to generate these data.
Daniel Gibbs, a retired neurologist at Oregon Health and Science University, Portland, who has Alzheimer’s, believes aducanumab will likely only help those at the earliest stages of the disease. “I worry that aducanumab may be used inappropriately on those with advanced disease for whom it is unlikely to be beneficial,” he wrote.
The FDA prescribing label for aducanumab gives the indication simply as “Alzheimer’s disease,” without specifying disease stage or requiring biomarker confirmation of amyloid positivity. Technically speaking, this label would include any clinical Alzheimer’s diagnosis. A baseline MRI is required, as well as two subsequent MRIs to monitor for ARIA. As is, this label leaves primary care physicians and specialists with little guidance on how to bring this new treatment into practice.
The broad label worries some. “I am concerned that under the current parameters, many patients would receive treatment that is unnecessary, either because they don’t have plaques or because their disease is too advanced to benefit from the drug,” Rabinovici said.
Whether public or private insurers will cover aducanumab is unclear. Biogen has placed the cost at $56,000 per year. This is more than analysts had predicted, and far higher than the $2,500 to $8,000 range a recent report by the Institute for Clinical and Economic Review had pegged as cost-effective. The company’s stock price rose more than 40 percent today, to a five-year high.
Golde noted that today’s decision may inspire pharma companies to invest more money into Alzheimer’s research and hopefully develop more efficacious treatments. “The field still has a lot of work to do,” he wrote to Alzforum. “I hope this approval is viewed only as a starting point.”—Madolyn Bowman Rogers
References
Therapeutics Citations
News Citations
- Exposure, Exposure, Exposure? At CTAD, Aducanumab Scientists Make a Case
- Aducanumab Still Needs to Prove Itself, Researchers Say
- FDA Advisory Committee Throws Cold Water on Aducanumab Filing
- Advisory Committee Again Urges FDA to Vote No on Aducanumab
External Citations
Further Reading
News
- Aducanumab Decision Delayed for Three Months
- FDA Accepts Aducanumab Application for Priority Review
- Biogen Asks FDA To Approve Aducanumab
- ‘Reports of My Death Are Greatly Exaggerated.’ Signed, Aducanumab
- Biogen/Eisai Halt Phase 3 Aducanumab Trials
- Much ‘Adu’ About a Little: Phase 1 Data Feeds the Buzz at CTAD
- Paper Alert: Aducanumab Phase 1b Study Published
- Biogen Antibody Buoyed by Phase 1 Data and Hungry Investors
Annotate
To make an annotation you must Login or Register.
Comments
University of Southern California Keck School of Medicine
This is face-saving by the FDA, who were obviously closely involved and strong advocates for aducanumab’s “substantial” effectiveness. Efficacy was not established as the post hoc analyses, at best, showed a trivial difference. The FDA reached back to its 2018 draft early stage guidelines and to accelerated approval guidelines. They needed to make a case for biomarker support as a surrogate outcome. Note the section of Patrizia Cavazzoni’s statement defending accelerated approval.
Approval is based on a surrogate or intermediate clinical endpoint (in this case, reduction of amyloid plaque in the brain). A surrogate endpoint is a marker, such as a laboratory measurement, radiographic image, physical sign, or other measure that is thought to predict clinical benefit but is not itself a measure of clinical benefit. Effectively, the FDA said, "Never mind the lack of determined effectiveness, aducanumab lowers plaques and we think the plaque-lowering effect of aducanumab is reasonably likely to predict clinical benefit."
In so doing, they have pronounced Aβ-PET to be an official surrogate marker of efficacy in Alzheimer’s pathology and disease. This perhaps is the most stunning outcome. Of course, from FDA’s point of view, this is provisional until the as-yet undescribed Phase 4 trials “verify the anticipated clinical benefit.” Will these trials be adequately designed and ever get done?
If you look back at the questions the FDA asked the advisory committee, the plan for establishing Aβ-PET as a surrogate marker for clinical benefit was front and center. The two relevant and leading questions—and the committee’s respective votes on them—were:
Yes: 5 No: 0 Uncertain: 6
Yes: 0 No: 10 Uncertain: 1
After much discussion, the FDA got a yes for the first on pharmacodynamic effect but a clear no for the latter, which is the basis on which the FDA gave accelerated approval. In a sense, approval was decided before the advisory committee meeting in November 2020.
UCSF
The approval of aducanumab is obviously a seismic event for our field, and hopefully heralds a new era that will be characterized by early, biomarker-supported diagnosis and biologically specific therapies. That said, I do wish the clinical efficacy of this “first-in-class” drug were better-established prior to approval. While the FDA has required a Phase 4 study, Biogen’s announcement that this study will take nine years to complete makes it clear that this requirement will not fill gaps in our understanding of drug utility.
I am also surprised that the FDA approval is based on amyloid lowering, a surrogate biomarker endpoint. While there is no doubt that the drug robustly clears plaques from the brain, the data that linking this biomarker change to cognitive benefit are far from clear. To my knowledge, the only trial to definitively link amyloid lowering to cognitive benefit is the Donanemab Phase 2 study (Mintun et al., 2021), while many other trials have found this relationship to be weak to absent.
Another issue is the FDA label, which vaguely refers to “treatment of Alzheimer’s disease” without discussion of biomarker confirmation or disease stage. It is critical for the field to establish best practices that require biomarker confirmation for the presence of amyloid and early disease stage (i.e., CDR 0.5-1) as requirements for treatment eligibility, as this was the population that was studied in the clinical trials.
We know from autopsy data and biomarker studies (NACC, ADNI, IDEAS, and many others) that clinical diagnosis on its own is inaccurate in predicting the presence of amyloid plaques. And we simply do not know whether the drug is safe and effective in preclinical AD or in patients with clinically advanced disease.
Well-designed studies that can identify demographic, clinical, genetic, and biomarker profiles that predict treatment response are essential as we move forward. It is also critical that we pay close attention to racial inequities in access to this drug in particular, and to dementia diagnosis and treatment in general. I hope the recently launched New IDEAS study, which prioritizes recruitment of Black and Latinx Medicare beneficiaries for amyloid PET, can make a positive contribution in this regard.
As a clinician, I do plan to offer this treatment to early stage patients who would have met the inclusion and exclusion criteria for the clinical trials. Ultimately, I believe that some patients may benefit, and I plan to openly discuss risks, benefits, and unknowns with my patients and help them reach an informed decision about whether the drug is right for them.
Finally, while the debate on aducanumab over the past few months has been very contentious, I hope that now that the FDA decision is made, we can move forward together to define the critical next steps in AD care and research.
References:
Mintun MA, Lo AC, Duggan Evans C, Wessels AM, Ardayfio PA, Andersen SW, Shcherbinin S, Sparks J, Sims JR, Brys M, Apostolova LG, Salloway SP, Skovronsky DM. Donanemab in Early Alzheimer's Disease. N Engl J Med. 2021 May 6;384(18):1691-1704. Epub 2021 Mar 13 PubMed.
Washington University
This is certainly a historic decision by the FDA to approve aducanumab under its accelerated approval pathway. There is no question that at the highest doses tested, it can decrease and remove amyloid from the brain.
Much has been written about the fact that one of the Phase 3 trials was positive in regard to cognitive slowing at the highest dose and one was negative. There may be many reasons for this. However, why usual standards for approving a treatment such as a generally positive consensus from the FDA advisory panel were not followed is not clear.
It’s certainly good that part of an accelerated approval will require a Phase 4 confirmatory trial. The FDA needs to hold Biogen to this Phase 4 trial requirement to ensure that this treatment truly benefits patients and that we are not wasting very large amounts of dollars on an ineffective therapy.
Goizueta Institute @ Emory Brain Health
Just like the mixed clinical data supporting the approval, I have mixed feelings about the approval itself. I do worry when the scientific process of evaluating potential efficacy of new drugs is influenced by nonscientific factors, regardless of indication. Many advocacy groups have weighed in, all almost universally in support of approval of aducanumab in AD; yet, the broad clinical consensus of outside experts has consistently sided with a more conservative view of efficacy stating that the clinical data were insufficient to warrant a full approval. This tension between unmet medical need, advocacy groups, and experts' opinion of the data underlying an approval can have long-term ramifications that are not always beneficial to the cause of delivering therapies that have major impact on a disease.
With this caveat in mind, I think a few things need to be stated clearly.
First, the field still has a lot of work to do. At least when given to individuals with symptomatic AD, aducanumab, based on the data presented, has a pretty modest effect on cognition and functional decline. There is a lot of room for improvement, and I hope that this approval is viewed only as a starting point. To use a baseball analogy, clinically this is not a home run. Rather, it’s like an infield single with a bang-bang call of safe at first base that is contested by the team in the field.
One has to wonder what happened to the individuals in the Phase 1b who “cleared Aβ” and improved cognition. What was their long-term fate? Somebody must know, but I hear no public discussion of this aspect. To my mind, it’s an important point. If all we got is 22 percent improvement and a six-month delay on average, then one wonders about cost-benefit.
Another thing that surprised me is that the FDA seemed to buy that reduction of PET ligand signal is a surrogate for pathological and biological reduction of amyloid. As far as I know, this has not been established with postmortem tissue from somebody who was given an antibody that reduces the signal and then died. I know the likelihood is strong that the correlation is there, but what happens if it is not?
I hope this spurs the investment in pharma needed to go after the doubles, triples, and home runs in terms of efficacy. Indeed, the positive side of this approval is that it suggests that one can develop drugs for AD even with minimal to modest efficacy and that they can get approved. Hopefully, this will reinvigorate private sector investment in this space. Given the cost of large-scale AD trials we really need big pharma/private sector investment, and a negative decision by the FDA could have reduced those future investments.
Third, follow-up studies of aducanumab will be needed to provide a clearer picture of potential efficacy in various settings. AD is a slowly progressing, insidious disease, so it is going to take some time before a true picture emerges of how much impact this drug will have on disease. As a biological therapy with side effects and a high price tag, it will be interesting to see how widely this drug is used in the real world.
University of Nevada Las Vegas
Hallelujah! A new drug for Alzheimer’ disease at last. The FDA has taken a courageous stand to approve a drug when there were many loud—and often well-reasoned—voices to the contrary. The development program was irregular, but the data on dose-exposure efficacy are compelling and complemented by the biomarker observations.
The aducanumab data are not anomalous and are consistent with emerging observations in the trials of lecanemab (BAN2401), donenemab, and gantenerumab. Plaque reduction (and reduction of other amyloid species these drugs affect) is accompanied by modest slowing of cognitive decline.
I am very happy for the patients for whom this drug is suited. This is a huge step forward for Alzheimer’s care and research and may inform a pathway to success for other neurodegenerative disorders.
I am hopeful that the controversies surrounding aducanumab do not fracture the Alzheimer's research community and that the differing views motivate us to develop more and better treatments to achieve our common goals of treating and preventing Alzheimer’s disease.
Washington University School of Medicine
Although I would have requested more assurance of efficacy before approval, the FDA’s decision now brings us a heretofore seemingly unattainable opportunity: a disease-modifying therapy for Alzheimer's dementia that benefits patients. The journey to this point has been very long and arduous, but if aducanumab (and, hopefully, soon other drugs) truly improves the lives of Alzheimer's patients and their families, then it will have been all worthwhile.
University of Kansas
Now we move from “What will the decision be?” to “Was this the right decision?” Over the next several years we will learn, in a real-world setting, how frequently this drug beneficially transforms the lives of the patients taking it and the lives of their families, and how frequently it ultimately ends with disappointment.
We will learn whether in aggregate it helps patients more than it hurts them. This decision will predictably shift financial, infrastructure, and intellectual resources toward getting this intervention to patients and to developing similar interventions. We will learn whether this shift accelerates or slows the pace of current and future progress.
We may find there are winners and losers; it may take years to fully determine which stakeholders end up in which group. One thing we do know—today’s decision will change a lot of things.
Stanford Alzheimer’s Disease Research Center
Alzheimer’s disease is a dreadful disorder, and patients and their families are desperate for treatment that can improve its devastating symptoms or halt its relentless progression. I am concerned that the average clinical effect of aducanumab in the first Phase 3 clinical trial was too small to be discerned by patients or family members; results from the second Phase 3 trial failed to detect even this small effect.
I am pleased that the FDA accelerated approval asks the manufacturer to conduct post-approval phase 4 studies to verify clinical benefit, which was not clearly evident in the two large phase 3 trials.
I have mixed feelings about this approval. As someone living with early stage Alzheimer’s who participated in the Engage trial, I am pleased that patients will now have access to the first drug shown to reduce brain amyloid and modestly slow progression of cognitive impairment in some people with early stage Alzheimer’s disease.
As a retired neurologist, I am less enthusiastic. Two identical trials had different outcomes. One showed benefit and one did not. Biogen’s post hoc analysis attempted to explain the discrepancy but is statistically fraught. A strong case can be made that a third trial should have been required before approval. I believe that aducanumab will ultimately be an important tool in fighting Alzheimer’s, but it probably will only work in the earliest stages, before many nerve cells have died, and it may work better or worse on certain subsets of Alzheimer’s disease, for example APOE4 carriers or those with amyloid angiopathy.
We still have a lot to learn. I worry that aducanumab may be used inappropriately on those with advanced disease for whom it is unlikely to be beneficial. I also worry that it will become more difficult to recruit research subjects for trials of new therapies that may turn out to be more effective, effectively slowing the push to find better therapies.
I am reminded of the excitement when, early in my career, tacrine (Cognex), the first acetylcholinesterase inhibitor, was approved in 1993. This was the very first drug that reduced symptoms of cognitive impairment in some patients with Alzheimer’s. However, hepatic toxicity was common and limited its use. In 1996, donepezil (Aricept) was approved, followed soon after by two other cholinesterase inhibitors. None of these had the serious liver problems of tacrine and were generally well-tolerated. Tacrine use dwindled and then stopped altogether. Today these three other cholinesterase inhibitors are still widely used.
Aducanumab may be the tacrine of today: the first drug of its class with likely effectiveness, but it will almost certainly be joined and possibly replaced by other, more effective drugs in the future.
Banner Alzheimer's Institute
This was a milestone in several respects, as aducanumab is the first novel therapy approved for Alzheimer’s in 18 years and the first drug attacking one aspect of the underlying pathobiology. This was an extremely difficult judgment to make. I take the FDA statement at face value: The evidence that aducanumab reduces brain amyloid is clear, but the filing data were “highly complex and left residual uncertainties regarding clinical benefit” and the agency “examined the clinical trial findings with a fine-tooth comb.” This had to have occurred, but we are not privy to the details of what must have amounted to innumerable sensitivity analyses.
The drug also has side effects that will need to be monitored very closely, chief among them being ARIA-E that appears to be more likely in APOE4 carriers. However, APOE genotyping is not referred to in the Prescribing Information. In addition, it is unclear whether testing for elevated brain amyloid is required, since it is not in the Prescribing Information except in the description of the study population. Payers may require this, indeed, some have already said as much.
The FDA approval was conditional on Biogen conducting another trial under the Accelerated Approval pathway, under which the FDA approves a drug for “a serious or life-threatening illness that may provide meaningful therapeutic benefit over existing treatments when the drug is shown to have an effect on a surrogate endpoint that is reasonably likely to predict a clinical benefit to patients and there remains some uncertainty about the drug’s clinical benefit.” This compromise was possible because of the interpretation of the amyloid-lowering effect as a surrogate endpoint, which is reasonable although it opens the door for competitor drugs to make a similar claim even without compelling clinical data. It will also be important for this condition to be enforced, but when and how?
This was only another step in the process to bring the drug to the clinic. We need to see the details of the Phase 4 plan and what that means for practice. In any case, aducanumab will not be available for an uncertain number of months to come while Medicare and insurers decide on availability, eligibility, costs (possibly including for amyloid testing and disclosure, and what about APOE genotype?), and coverage for safety MRIs, infusion costs, and the drug itself. The rollout will be slow and complex.
We are telling our families now that they will be able to discuss aducanumab with their providers, about what we do and do not know at this juncture, and setting the stage for further discussion as we learn more.
USC Alzheimer’s Therapeutic Research Institute
The FDA has granted accelerated approval to aducanumab, based on evidence across three trials that the drug reduces amyloid in brain, along with the reasonable likelihood that amyloid reduction will result in clinical benefit. A major issue with this decision is that it leaves open many practical questions about its use. Amyloid is elevated in brain for many years before the onset of AD symptoms. Who should be treated? How should individuals be monitored for safety and clinical benefit? Phase 4 studies are required to clarify clinical benefit after accelerated approval; how should clinicians use aducanumab in the meantime?
Acumen Pharmaceuticals
A historic moment was achieved with the accelerated approval by FDA of aducanumab for the treatment of Alzheimer’s disease. This decision will not be made in the absence of controversy, given the results from two identically designed pivotal trials for aducanumab for the treatment of people with MCI and mild dementia due to AD. Questions have been immediately raised regarding efficacy, safety, the patient population described in the label, and the price.
The first question regards the possible efficacy of the drug. In the presentation to the Peripheral and Central Nervous System (PCNS) Drugs Advisory Committee Meeting on November 6, 2020, the supporting document (NDA/BLA# 761178) detailed a favorable view from the sponsor and clinical division of FDA, but a negative review from the statistical division of FDA. While the sponsor focused on a positive result in study 302, without regard to the negative result in study 301, the FDA statistical review noted that studies 301 and 302 should be considered together. The statistical reviewer also held the opinion that the Phase 1B study (study 103)1 should not be considered when evaluating efficacy. Nevertheless, the statistical review of the FDA document (page 342) noted that “Based on a Bayesian meta analysis of study 301 and 302, the posterior probability of the alternative hypothesis is 0.62 and a corresponding 0.38 probability for the null hypothesis.” In other words, the null hypothesis was substantially less likely than the alternative hypothesis favoring efficacy. This Bayesian analysis did not consider the results of the Phase 1B study showing trends or statistically significant effects on most efficacy measures. Thus, based on the totality of the Phase 1B and Phase 3 data, the probability that aducanumab has no effect on efficacy is negligible.
Assuming that aducanumab does have some efficacy, the question of safety versus efficacy remains important. The safety concerns regarding ARIA are well reported and remain of some concern. The FDA packaging label states that MRI findings should be available before starting therapy and also contains the recommendation to “obtain MRIs prior to the 7th infusion (first dose of 10 mg/kg) and 12th infusion (sixth dose of 10 mg/kg). If 10 or more new incident microhemorrhages or > 2 focal areas of superficial siderosis (radiographic severe ARIA-H) is observed, treatment may be continued with caution only after a clinical evaluation and a follow-up MRI demonstrates radiographic stabilization (i.e., no increase in size or number of ARIA-H)”. MRI monitoring will remain a substantial burden along with clinical monitoring of symptoms related to ARIA with clinical utilization of aducanumab.
Finally, the cost and burden of the treatment should be considered. Even taken at face value, that a 22% slowing of progression in CDR-SB could be achieved, this must be balanced against the patient burden, patient cost and societal cost. Given an annual cost of treatment for each patient of approximately of approximately $56K/year (U.S. approval of Biogen Alzheimer's drug sent shares soaring, hailed as 'big day" for patients | Reuters), this treatment will be expensive. Payors will undoubtedly try to limit the patient population to reduce their costs.
Interestingly, the FDA label did not include language considering severity of symptoms or evidence of amyloid positivity, but since the trials required only people with MCI and mild dementia due to AD, payors will likely require specific cognitive deficits and evidence of amyloid positivity. These limitations will decrease the societal costs of aducanumab.
Taken together, these developments are a remarkable development in the struggle to find disease-modifying treatments for AD. Aducanumab will not be the last, or probably the best, treatment for AD. But these events are clearly a notable milestone for the field.
1. Sevigny J, Chiao P, Bussiere T, et al. The antibody aducanumab reduces Abeta plaques in Alzheimer's disease. Nature 2016;537(7618):50-6. DOI: 10.1038/nature19323.
UT Southwestern Medical Center
To preface, I am a proponent of the amyloid hypothesis. The most parsimonious mechanism for Alzheimer's disease progression remains an orderly one: Aβ accumulation drives tau aggregation, which begets neuronal degeneration. The ideal therapeutic would target the upstream driver to halt the disease prior to display of phenotype. Unfortunately, of course, this is easier said than done. By the time biomarkers unequivocally distinguish prodromal AD patients from healthy controls, targeting Aβ may be too little too late. As a metaphor, it is useless to take away the gun once the bullet has been fired.
Nevertheless, careful clinical trials may allow Aβ therapeutics to be feasible. A leader in this regard is the DIAN study, which continues to use innovative trial design to test Aβ therapies and fine-tune the requisite variables to unequivocally test the amyloid hypothesis in familial AD patients. Setbacks have dampened enthusiasm to a degree, but great hope remains. Every year brings incremental and important progress.
Tragically, the approval of Biogen's aducanumab undermines the field’s methodical progress. The degree to which standard practice was subverted by this successful application is perplexing and greatly damages the FDA’s credibility. A few rhetorical examples follow, although other examples can be found in comments on this article and others.
(1) Failed futility analyses? No bother, post hoc analyses riddled with multiple comparison confounds can fix this! Whatever gives the desired result!
(2) Reliance on amyloid reduction via PET? New gold standard for efficacy ... never mind past opinions to the contrary in related trials!
(3) Advisory board disagreement? Who needs opinions of those who aren’t in on the preordained outcome!
(4) Inflated price tag and lack of concern for AD progression stage? Who cares? Allow prescriptions to anyone with a diagnosis! Hope is more important than facts. Clinicians can figure it out no matter how universally they disagree with the FDA decision!
There is no way to overstate how harmful this decision is to the Alzheimer’s field, whether it be basic research or therapeutic trials. More broadly, the reputation of American drug discovery and approval, and the very backbone of scientific reason as it pertains to clinical trials, have been severely undermined. If negative outcomes can be massaged to p-hack a company’s way to profitability (with help from patient advocacy groups, who lobby for any treatment regardless of efficacy), why should the American public trust scientists to act as unbiased arbiters of drug safety and efficacy?
The disgraceful actions of certain players in the U.S. government with respect to hydroxychloroquine in COVID have already placed the FDA in a perilous position; rather than right the ship, they embraced an albatross.
I conclude with one simple question: Would one of your loved ones agree to participate in a long-shot clinical trial for a new AD drug given aducanumab’s availability? Mine wouldn’t. They would skip the trial and take the approved drug ... it’s backed by science, right?
Unfortunately, many of these individuals will forfeit their life savings on a treatment that likely doesn’t work, purchasing hope over facts, simply because the FDA decided to shirk its responsibility as a regulatory agency.
But maybe my critique is wrong? Biogen has nine years to perform a trial that can confirm its initial findings. Nevertheless, wouldn’t they have been able to do this new trial in two years if not for the FDA's Wall Street-friendly stipulation? Does anyone honestly believe that decision-makers at Biogen think this is a good drug?
The FDA has failed scientists, patients, the public. How did we get here? How do we prevent this from occurring again? Sadly, I suspect this is a harbinger of a libertarian-minded market to come, one dominated by snake oil and miracle drugs, as patients jostle for purported cures bequeathed upon the highest bidder.
Washington University School of Medicine
These are exciting and stressful times in Alzheimer's research, and for those who care for patients with the disease. We are excited to have a new treatment after 18+ years, and to have the first treatment that may modify underlying disease processes.
This is tempered by doubts as to how clinically effective this medication may be, and by concerns about what the treatment landscape will look like and what impact this approval will have on current and future clinical trials.
Our clinical office staff has been inundated with calls from patients and families who want access to this drug. For now, we have very few answers.
The broad FDA approval is not helpful. Unless further guidance is forthcoming, individual physicians and practices will need to establish their own guidelines for which patients should receive this medication, and what sort of biomarker testing should be done. Perhaps more likely, these decisions will be largely ceded to payers rather than the FDA or clinicians.
This may engender even more economic and social disparities in AD treatment. We know that Aduhelm will be expensive, with large costs to our already burdened health care system, and may be beyond the means of many. Many pharma providers of other IV medications have a patient-assistance program; we can hope that Biogen will develop such a program for Aduhelm.
The impact on future clinical trials is also concerning. We hope that this approval will inspire more interest in developing AD drugs, but how will those trials be conducted? Will only the subset of people who are unable to get Aduhelm be willing to volunteer for future trials? Will those already in other trials, some of which are using very promising drugs, decide to leave those studies once Aduhelm is available?
Despite these concerns, these are exciting times. The trial participants I saw this week are full of joy and hope, reminding me why we have all been working to find better treatments for this devastating disease, despite the many years without seeing any new treatments emerge. We will push on, finding the best way to get this new treatment to our patients who might benefit, and continue working to find better treatments for all of our patients.
Chiesi Farmaceutici S.p.A.
I anticipated that the FDA would approve aducanumab. The FDA had made a clear public statement in the past in favor of aducanumab, and it is not the first time the agency has approved a drug in open contrast to the negative opinion of an external Advisory Committee.
The FDA will now need to specify to the Alzheimer's scientific community whether future pivotal clinical trials of new anti-AD drugs will need to be tested against placebo or against aducanumab. This represents the most immediate and practical consequence for pharmaceutical companies and research institutes engaged in the development of anti-AD drugs.
The same question will have to be evaluated by the Ethics Committees, which must approve controlled clinical trials with new anti-AD drugs. If we look at past experience with cholinesterase inhibitors, following approval of donepezil, the FDA did not request to use donepezil in controlled trials with the subsequent investigational cholinesterase inhibitors (rivastigmine and galantamine). However, several subsequent trials have evaluated new drugs on top of cholinesterase inhibitors.
Thus, I wonder if future trials will test new anti-AD drugs on top of aducanumab. In the case of donepezil, that drug showed an MMSE-enhancing effect compared to placebo of 1 to 1.5 MMSE points at six months of treatment in patients with mild to moderate AD. In the case of aducanumab, we have an advantage of the drug over placebo of about 0.5 MMSE points at 18 months (in the “positive” EMERGE trial) in patients with early or mild AD. So we can reasonably assume that placebo will remain the comparison group in controlled trials of new anti-AD drugs over the next few years.
An additional crucial point now is the study design of the post-approval study requested by FDA. Is it a double-blind, placebo-controlled study?”
Liverpool John Moores University
It will be interesting to see how non-U.S. countries will follow on from this FDA decision. Healthcare infrastructure differs from country to country. Access to PET scanners, which will presumably determine who should be administered the drug, may vary, even across regions within the same country.
An additional crucial point now is the study design of the post-approval study requested by FDA. How will patients be preselected for the Phase 4 trial? What specific tests will be employed to determine possible amyloid burden prior to, or without, PET determination?
RIKEN Center for Brain Science
It must have been difficult for the FDA to make the decision because they can be simultaneously praised and criticized, as seen in Alzforum. One factor that has affected the decision appears to be the extraordinary resources that have been spent for the R&D. The sunny side is that the Aβ hypothesis has made a huge leap from basic to clinical sciences.
I am, however, surprised that the FDA did not distinguish between preclinical AD and symptomatic AD: Aducanumab is most likely to be effective on preclinical AD because the irreversible death of neurons is the direct cause of the major symptoms, i.e., the primary and direct causes must be differentiated. In this respect, the Phase 4 clinical trials should be performed on asymptomatic patients carrying FAD-causing mutations, whose ages of onset are predictable.
Among wealthy families, there may be some preclinical patients interested in aducanumab despite its cost. In addition, the FDA’s approval may encourage potential FAD mutation carriers in any place worldwide to come out for the treatment. This will be a good sign. Young Down’s syndrome patients should also be treated if resources are available.
I have had one question for a long time as an experimental scientist. Reduction of amyloid PET signals by aducanumab treatment is considered to indicate a decrease in Aβ deposition. It is, however, possible that aducanumab bound to the plaques may have blocked the binding of PET ligand in a competitive manner.
In addition, extraction of Aβ from AD brains requires high concentration of formic acid, whereas that of model mice can be extracted with GuHCl. Aβ in AD brains is much more insoluble than Aβ in model mice and thus is “hard as a rock.” If my presumption turns out to be true, then aducanumab may be exerting its effect by removing Aβ oligomers.
The position of the European Alzheimer Disease Consortium (EADC) on the approval of aducanumab for the treatment of Alzheimer’s disease by the FDA:
On June 7, in a highly anticipated global decision, the U.S. Food and Drug Administration (FDA) approved aducanumab (Aduhelm®, an anti-amyloid antibody) for the treatment of Alzheimer's disease at 10 mg/kg body weight, while requiring the drug's manufacturer, Biogen, to conduct an additional post-approval clinical trial (Phase 4) to verify the drug's clinical benefit. If that study fails to demonstrate clinical benefit, the FDA may revoke the drug's approval. In November 2020, the FDA's independent advisory committee had voted against approval, stating that the data presented in the available clinical trials were not convincing enough.
However, the FDA has now backed away from the requirement of proof of clinical efficacy and followed a different regulatory pathway, that of a surrogate end point (i.e., approval on the basis of the evidence of reduction of amyloid deposits). An initial small Phase 1 study to investigate the safety of aducanumab had shown significant brain amyloid load reduction, as demonstrated through positron emission tomography (PET) imaging, and the preliminary clinical data suggested that this could slow cognitive decline. As a result, the FDA allowed Biogen to skip the otherwise standard Phase 2 trials and conduct two Phase 3 studies (ENGAGE and EMERGE) right away, each with about 1,640 patients. Those trials were halted early in March 2019 when the independent data monitoring committee decided, based on an interim analysis, that there was too high a likelihood that aducanumab would not be effective. Consequently, 37 percent of participants were unable to complete the 78-week study period.
However, in October 2019 Biogen announced that after a re-evaluation, evidence for efficacy did exist. This conclusion was based on data from an additional 318 participants collected before the trials were stopped but after the cut-off date for the interim evaluation. In one of the two studies (EMERGE), the highest dose significantly slowed the severity of disability by 22 percent compared to the placebo arm. A lower dose in this study and both doses in the second study (ENGAGE) showed no statistically significant superiority over placebo.
Only a post hoc analysis of the completers in the high-dose subgroup in the overall negative second study (ENGAGE) showed evidence of efficacy. The accelerated approval pathway now granted by the FDA for aducanumab is intended for drugs for serious diseases that are expected to have a meaningful benefit over available therapy even if there is residual uncertainty about the drug's ultimate clinical benefit. To be approved through this pathway, there must be substantial evidence of the drug's efficacy on a "surrogate endpoint"—usually a biomarker, in this case amyloid or its deposition, that reflects the underlying disease pathology. In the expedited process, it must be known that the effect on this surrogate endpoint results in the clinical benefit.
The manufacturer’s prescribing information (SmPC) for aducanumab plainly states “… indicated for the treatment of Alzheimer’s disease,” without further specifications. It includes a warning for amyloid-related imaging abnormalities (ARIA), which can be visualized on magnetic resonance imaging and most commonly presents as transient swelling or focal microhemorrhage in small areas of the brain, usually without symptoms. To monitor for ARIA, a recent (within one year) brain magnetic resonance imaging (MRI) prior to initiating treatment is required. The most common side effects of aducanumab were ARIA and associated headache, confusion/delirium/altered mental status/disorientation, as well as falls and diarrhea. Rare side effects of aducanumab include hypersensitivity reactions, including angioedema and urticaria.
The EADC would like to highlight that this treatment is at this point approved only for the U.S. It is also important to note that aducanumab trials only enrolled a group of patients with a narrow disease–specific stage, namely mild cognitive impairment (MCI) due to Alzheimer's disease or early stage Alzheimer's dementia, and that aducanumab should be given to this group of patients only. Further research is needed to better understand which patients, and at what disease stage, respond best; what is the optimal duration of treatment (current data do not span more than a year and a half); and what side effects may occur after long-term use. For this, in addition to the Phase 4 study required by the FDA, so-called registry studies will also be of great importance. Many other unknowns and uncertainties still remain, including the very long time frame for the Phase 4 study (nine years).
Finally, the approval of aducanumab may open the door for many other applications for approval based on proven amyloid reduction but without solid evidence for clinical efficacy. Due to the absence of convincing clinical benefits, the topic is and will remain controversial until more evidence is collected.
This approval is currently not valid for the EU and U.K. It is uncertain what decision the European regulatory authority will make here, including type of approval pathways, indications relating to disease stage, patients’ ages, types of requirements for amyloid positivity, treatment duration and monitoring, etc. It is also unclear how each individual European country and corresponding health care and insurance systems and institutions will act in relation both to setting up systems for such hospital-based treatments and to cost coverage issues.
Emergency approvals without clear clinical proof of efficacy generally should be applied with caution. As much as all physicians and patients have been hoping for years for a new and better drug for Alzheimer's disease, caution is advisable in view of the expected costs, the expectations of patients, and the uncertainty around the lack of clear clinical proof of efficacy as well as the necessary safety studies.
Thus, although this drug does not provide a cure and is still subject to uncertainties, this first treatment against the pathophysiology of Alzheimer's disease is of great importance. If eventually proven, slowing the progression of cognitive decline in people in the MCI stage of Alzheimer's disease or early Alzheimer's dementia would be a major advance, and the known side effects of aducanumab appear tolerable.
In addition to aducanumab, other antibodies directed against amyloid are currently in registered trials (gantenerumab, lecanemab, donanemab). Hopefully, the use of aducanumab, which has now become possible at least in the U.S., will represent a breakthrough for additional therapies against Alzheimer's disease and encourage more pharmaceutical companies to resume research activities for the development of therapies against neurodegenerative diseases, including targets other than amyloid.
—For the EADC: Lutz Frölich, Frank Jessen, Mercè Boada, Sebastiaan Engelborghs, Patrick Kehoe, Milica Kramberger, Alexandre Mendonca, Flavio Mariano Nobili, Nikolaos Scarmeas, Pieter Jelle Visser, Gunhild Waldemar (Executive Board)
See also:
RWTH Aachen University
The approval and the way of the approval (accelerated approval pathway) comes as a surprise to me. It certainly is driven by political motivation, not by overwhelmingly convincing data. However, for the AD community—patients, relatives, researchers, physicians, and pharmaceutical companies with interest in neurodegenerative diseases—it is a great step forward. The FDA used the "accelerated approval pathway," under which the FDA “approves a drug for a serious or life-threatening illness that may provide meaningful therapeutic benefit over existing treatments when the drug is shown to have an effect on a surrogate endpoint that is reasonably likely to predict a clinical benefit to patients and there remains some uncertainty about the drug’s clinical benefit" (FDA’s Decision to Approve New Treatment for Alzheimer’s Disease).
I have a number of questions and concerns:
These are many questions, and it will take time and negotiations between the FDA and the manufacturer to find answers and solutions. In addition, best-practice guidelines should be developed and published rapidly. Some of my questions and critique may be related to my insufficient insights into the U.S. health system. In Europe, we are anxiously waiting for the decision of the EMA.
Finally, the approval of aducanumab starts a new era in the research and treatment of AD. There are further anti-amyloid drugs in clinical study phases (gantenerumab, lecanemab, donanemab). Hopefully, their results will get similar evaluation by the FDA. The decision will also accelerate the motivation of pharmaceutical companies to return to research and investments in neurodegenerative disorders, a field that, compared with cancer, needs much more pharmaceutical activity.
Northwestern University Feinberg School of Medicine
The approval of aducanumab, although controversial, may open new opportunities in the areas of prevention and combination trials for AD. Prevention trials administering aducanumab to asymptomatic, amyloid-positive individuals should move forward. For a combination trial, aducanumab could be used first to clear amyloid from the brain, then aducanumab treatment halted and followed by a safe, low-dose BACE1 inhibitor regimen to lower Aβ production by ~30 percent and maintain an amyloid-free brain. Other combinations with aducanumab could be envisioned, such as with an anti-tau antibody or ASO, or a γ-secretase modulator.
Treating the right patient population with aducanumab will be paramount. Aducanumab is the first disease-modifying therapy for AD, but it's unlikely a silver bullet that will work for everyone. It eventually may work best in asymptomatic amyloid-positive individuals to delay the onset of AD or slow its progression, but it is unlikely to be effective for late-stage AD.
Tel Aviv University
At the 2000 Advances in Alzheimer Therapy (AAT) meeting in Stockholm, I showed for the first time that the EFRH sequence, i.e., amino acids 3-6 at the N-terminal region of Aβ, is the key position for its conformational modulation, and that antibodies toward this epitope suppress formation of Aβ plaques and dissolve already-formed fibrillar amyloid (Solomon et al., 1996, 1997).
Moreover, we showed by immunization with phages displaying EFRH as an antigen that this is the immunodominant epitope of Aβ and we were able to induce anti-Aβ antibodies that recognized the whole peptide and exhibited anti-aggregating properties (Frenkel et al., 2000). Subsequently, anti-Aβ antibody-based therapies for AD have emerged (Lemere and Masliah, 2009).
Recently, such naturally occurring anti-EFRH antibodies were found in human CSF and in the plasma of healthy individuals, but significantly less in AD patients (Sevigny et al., 2016), confirming our previous data and now Aducanumab has become the first drug against Alzheimer’s disease.
In Hebrew we say mazel tov.
I am very happy for the patients for whom this drug is appropriate; maybe it will also open the pathway to success for other neurodegenerative disorders cause by protein conformational changes.
I am hopeful that the controversy surrounding aducanumab does not divide the Alzheimer's research community but instead motivates us to develop more and better treatments and inspire pharma companies to invest more money into Alzheimer’s research and hopefully develop more efficacious treatments.
References:
Solomon B, Koppel R, Hanan E, Katzav T. Monoclonal antibodies inhibit in vitro fibrillar aggregation of the Alzheimer beta-amyloid peptide. Proc Natl Acad Sci U S A. 1996 Jan 9;93(1):452-5. PubMed.
Solomon B, Koppel R, Frankel D, Hanan-Aharon E. Disaggregation of Alzheimer beta-amyloid by site-directed mAb. Proc Natl Acad Sci U S A. 1997 Apr 15;94(8):4109-12. PubMed.
Frenkel D, Katz O, Solomon B. Immunization against Alzheimer's beta -amyloid plaques via EFRH phage administration. Proc Natl Acad Sci U S A. 2000 Oct 10;97(21):11455-9. PubMed.
Lemere CA, Masliah E. Can Alzheimer disease be prevented by amyloid-beta immunotherapy?. Nat Rev Neurol. 2010 Jun;6(6):296. PubMed.
Sevigny J, Chiao P, Bussière T, Weinreb PH, Williams L, Maier M, Dunstan R, Salloway S, Chen T, Ling Y, O'Gorman J, Qian F, Arastu M, Li M, Chollate S, Brennan MS, Quintero-Monzon O, Scannevin RH, Arnold HM, Engber T, Rhodes K, Ferrero J, Hang Y, Mikulskis A, Grimm J, Hock C, Nitsch RM, Sandrock A. The antibody aducanumab reduces Aβ plaques in Alzheimer's disease. Nature. 2016 Aug 31;537(7618):50-6. PubMed.
University of California, Los Angeles
At Biogen’s 2019 CTAD presentation of aducanumab Phase 3 trial results, my main takeaway was the enthusiastic reception of the drug by trial investigators and enrolled patients. Given the absence of any disease-modifying treatments for Alzheimer’s, I thought approval was a given. The “no” recommendation from the advisory committee reflects the gap between the strength of scientific evidence and clinical enthusiasm.
Perhaps this reflects an intrinsic flaw in the design of this and other clinical trials based on anti-amyloid therapy. There is no well-defined biochemical pathway leading to symptom improvement from plaque removal. What will happen biochemically when plaques are removed? And how do these new biochemical changes lead to improvement in cognitive function? Does plaque removal lead to normal synapse function that is mostly related to short-term memory loss? Does it lead to better retrieval of long-term memory if those neurons are still intact? Currently, we don’t have the knowledge to show that plaque removal will lead to cognitive improvement. All the anti-amyloid trials are designed to remove amyloid, and they are not designed to cure Alzheimer’s.
One potential theoretical framework is Aβ supersaturation, which provides a more nuanced interpretation of the amyloid hypothesis. According to supersaturation theory, the major biochemical change in Aβ is that Aβ aggregation becomes a daily problem in the presence of amyloid. Without plaques, there is an equilibrium between Aβ production and clearance. With plaques, every day newly produced Aβ deposits into plaques through a plaque-catalyzed aggregation process. Toxic Aβ aggregates are formed during the aggregation, stressing the surrounding neurons. When plaques are removed, the equilibrium of Aβ turnover is restored, thus eliminating the toxic species that is produced during aggregation. Based on the supersaturation theory, Aβ supersaturation (equilibrium) can be restored only when all plaques are removed. It is no coincidence that aducanumab is arguably the most effective antibody for plaque removal in all previous AD trials.
Alzheimer’s disease is a multifactorial disease without well-defined molecular markers. It may be sufficient to demonstrate effective plaque removal for the approval of anti-amyloid therapies. Then physicians can develop treatment plans using these anti-amyloid drugs. Without any approved anti-amyloid drugs, physicians and patients are fighting a war without weapons.
References:
Guo Z. Amyloid hypothesis through the lens of Aβ supersaturation. Neural Regen Res. 2021 Aug;16(8):1562-1563. PubMed.
The University of Queensland
We are watching the approval of aducanumab with interest, given that we had cloned the antibody to evaluate whether a combination with therapeutic ultrasound would be more effective in clearing amyloid and improving cognitive functions in the APP23 mouse strain. We had found that the combination therapy achieved up to fivefold higher intracerebral antibody concentrations at the time point analyzed. It also led to a lower amyloid burden and improved memory functions. Our data suggest that a combination therapy in humans may allow researchers to reduce the antibody concentration required for achieving therapeutic outcomes.
—Gerhard Leinenga is co-author of this comment.
References:
Leinenga G, Koh WK, Götz J. A comparative study of the effects of Aducanumab and scanning ultrasound on amyloid plaques and behavior in the APP23 mouse model of Alzheimer disease. Alzheimers Res Ther. 2021 Apr 9;13(1):76. PubMed.
Florida Atlantic University, Florida international University
Tautatis Inc.
I believe this is a terrible decision, probably the worst ever, by the only true policeman we have in the drug-development process. We always believed that clinical trials would be the ultimate arbiters of the avenues elaborated in the labs of scientists. Now we know that vested interests can move the goal posts. It is a punch in the nose of any scientist trusting that evidence is what guides such far-reaching decisions as to what millions of AD patients are being told, and what is given to them for their condition, trusting that the professionals know what they are doing.
Anyone who believes that this spurs investment for better treatments doesn't seem to know how investors think. There are a lot of what are called "sunk costs" out there, and the most attractive thing for an investor is to recover these. With 37 clinical trials for anti-amyloid agents out there, which might not have failed if judged against those moved goal posts, one shudders to think what will happen next, and where tens of billions of dollars will end up in the next ten years.
But we hear from the FDA that other entrants may not be granted such favorable consideration. What is to be thought of that? Some preferred relationship with Biogen after all? Whichever way the agency moves now, it will be bad.
And with all this insufficiently supported hope going around, the hammer needs to fall at the very least on the $56,000/year heist Biogen intends to pull off for something that hasn't much better evidence than some nutraceuticals with defendable rationales behind it. Odd that the Alzheimer's Association was so vocal to support the approval, but there has not been an equally vocal word on the price tag. I thought this association was about patient advocacy, no?
As I reviewed these expert reactions to the FDA quasi-approval of aducanumab, I happened to also be reading the best-selling nonfiction book "Empire of Pain" by Patrick Keefe. It explains the machinations of the Sackler family, who took oxycontin from a welcomed pain relief drug for cancer to a costly nationwide scourge. There are some disturbing similarities to the recent development with aducanumab: the coziness of the drug's sponsor with certain key FDA officials, the lubrication of key advocacy groups with monies labeled everything except "bribe," the fees to clinical investigators and "consultants," the widening of the clinical indications from a narrow, provable condition to the blurry limits of Alzheimer's, and the potential harmful side effects from the widespread usage. The street value of oxycontin went to $1.00 per milligram. However, if one is selling an FDA-authorized drug for $56,000 per year to numbers of desperate, incurable, but hopeful patients, for an ill-defined condition, there is no need for "street sales" to make billions.
Yes, everyone wants and needs an Alzheimer's cure, but with more solid evidence behind it. One day this story will be written, too.
former Stanford Medical Center Investigator
The FDA approval of aducanumab via the accelerated track is of no surprise given that a similar therapy approach has previously been shown to work, and it implies that ultimate approval is warranted by Phase 4 trial results. Criticism of the therapy stems from scarce research on its effectiveness on other highly toxic, accompanying pathology hallmarks, such as p-tau, intraneuronal amyloid filament aggregation, as well as endosomal scavenging transport mechanisms.
Still, for some of us who have had hands-on experience working with Aβ antibodies, seeing direct plaque clearance has been convincing that this therapy has prospects. If you'd like to see fluorescence microscopy images that show the process of how microglia clear antibody-treated plaques, take a look in my Neurobiology of Aging paper cited below. To my knowledge it was one of the first ex vivo co-cultures assay demonstrations of that process.
References:
Mitrasinovic OM, Murphy GM. Microglial overexpression of the M-CSF receptor augments phagocytosis of opsonized Abeta. Neurobiol Aging. 2003 Oct;24(6):807-15. PubMed.
Make a Comment
To make a comment you must login or register.