A New Direction For Alzheimer’s R&D: FDA Outlines Early Disease Pathway

Alzheimer’s R&D has suffered multiple failures over the last decade, with no drug found to slow or halt the underlying progression of the disease. Industry’s focus is now on patients at the very earliest stages of the disease, and FDA is meeting sponsors halfway, with a draft guidance outlining approval requirements.

After multiple failed attempts to find an effective treatment that slows or stops the progression of Alzheimer’s disease, the Food and Drug Administration is opening a door to a new area of research: patients at very early, pre-dementia stages of Alzheimer’s.

For industry, this comes as welcome news. Drug sponsors have been reluctant to invest in R&D in early Alzheimer’s without formal guidance from FDA, and have instead focused on patients who are already showing signs of dementia and functional impairment. But those efforts have resulted only in high-profile failures in the clinic (Eli Lilly & Co./Elan Corp. PLC’s semagacestat and Pfizer Inc./Johnson & Johnson’s bapineuzumab, among others).

The focus is now turning toward treating Alzheimer’s patients before they develop significant cognitive impairment. The working theory is that drug intervention has occurred too late in the progression of the disease to have a meaningful effect: the damage is done and can’t be repaired – but might be prevented with earlier treatment.

That theory, of course, may be nothing but wishful thinking. In most progressive disease contexts, at least, it has generally been much easier to show benefit in the sickest of patients: think of all the cancer drugs first developed for severely refractory patients rather than for first-line use. So it may be that the issue here is not the nature of the disease, but rather of the drugs. Simply put, it may be that they just don’t work.

The skeptics include some top industry executives. Sanofi CEO Chris Viehbacher told Bloomberg in an interview that he isn’t planning any additional major investments in Alzheimer’s R&D until the science behind the disease is better understood. “We have to do a lot more basic science work to understand what’s going on,” Viehbacher said. “We really, at best, partially understand the cause of the disease. It’s hard to come up with meaningful targets.”

Still, with sponsors focusing increasingly on early treatment, FDA is eager to do its part to help determine if the early treatment approach can benefit some patients.

This spring, the agency released a draft guidance document on early-stage Alzheimer’s, and followed up with a webinar on March 28 to outline its thinking on how best to identify early-stage patients and to describe a pathway for drug development. Once the draft guidance is finalized, the hope is that it will encourage the development of drugs in a previously unstudied patient group – and, perhaps, find a successful treatment.

Targacept Inc. regulatory affairs and quality compliance senior director Jessica Beaver applauded FDA’s draft guidance, saying it opens up a new area of Alzheimer’s research where industry was previously hesitant to go. “As an organization, we were very excited about this guidance, because it paints a path for us” to develop drugs for early-stage disease, she says.

The fact is that many pharmaceutical companies are continuing to invest heavily in Alzheimer’s disease, with increasing focus on early interventions. Whether those bets will pay off is unknown. But for an industry looking for new answers, the FDA guidance is a step in the right direction.

Turning Toward Earlier Treatment

There are few – if any – more frustrating areas of pharmaceutical R&D than Alzheimer’s disease.

FDA has approved five drugs to treat the signs and symptoms of Alzheimer’s disease, which sounds like a better track record than some diseases. However, no drugs have been approved by the agency since 2003 (with the exception of imaging agents), and the handful of approved therapies have failed to demonstrate any impact on disease progression in multiple trials – while the list of known adverse reactions keeps expanding. As if to underscore the sense that drug development is at a dead end, Shionogi & Co. Ltd. discontinued the first Alzheimer’s drug approved by FDA (Cognex [tacrine]) in 2012. (See Exhibit 1.)

Exhibit 1

FDA-Approved Drugs For Alzheimer’s Disease

Manufacturer	Drug Name	Mechanism of Action	Indication	Date Approved
Shionogi Pharma	Tacrine (Cognex; discontinued in US May 2012)	Cholinesterase inhibitor	Mild to moderate	1993
Pfizer/Eisai	Donepezil (Aricept)	Acetylcholinesterase inhibitor	All stages	1996
Novartis	Rivastigmine (Exelon)	Acetylcholinesterase inhibitor	Mild to moderate	2000
Johnson & Johnson	Galantamine (Razadyne; formerly Reminyl)	Cholinesterase inhibitor that prevents the breakdown of acetylcholine	Mild to moderate	2001
Forest Labs	Memantine (Namenda)	Glutamate pathway modifier	Moderate to severe	2003

The Alzheimer’s Association

Until recently, most of the focus in Alzheimer’s R&D has been in treating patients after they develop signs of dementia; all approved drugs and the late-stage pipeline are targeted to patients with mild, moderate, or severe dementia. But while pharmaceutical companies like Lilly, Targacept, Roche, Merck & Co. Inc., Johnson & Johnson, and Pfizer Inc. are investing heavily in Alzheimer’s R&D, many of those bets have not paid off. (See Exhibit 2.)

Researchers now theorize that one reason behind at least some of those failures is that the drug intervention started too late in the disease process – when patients had already developed signs of dementia. “In other words, the damage was done,” FDA division of neurology products acting team leader Nicholas Kozauer, MD, said during the March webinar. “It was too late to have a clinically meaningful effect.”

Exhibit 2

Failed Bets In Alzheimer’s R&D

Drug	Manufacturer	Mechanism of Action	Status
Tarenflurbil (Flurizan)	Myriad Genetics	Selectively lower beta amyloid and modulate gamma-secretase	Flurizan failed in 2008 to improve cognitive functioning or activities of daily living.
Tramiprosate (Alzhemed)	Neurochem	Glycosaminoglycan mimetic designed to bind to soluble Aβ and inhibit deposition	In 2007, Phase III was completed in North America; data were inconclusive. Neurochem has since changed its name to Bellus Health and markets the compound in Canada as a nutraceutical to "protect memory function."
Semagacestat	Lilly/Elan	Gamma secretase inhibitor	Discontinued in 2010 after patients taking semagacestat were found to have worse functional and cognitive deficits than placebo patients.
Latrepirdine (Dimebon)	Pfizer/Medivation	Inhibitor of cholinesterase and NMDA receptors	Failed in 2010 to show statistical significance in two endpoints in Phase III and was discontinued.
Solanezumab	Lilly	Bind to soluble monomeric forms of amyloid ß after it is produced	Failed to show effect in two Phase III trials in 2012, but pooled analysis showed patients with mild disease had a 34% decrease in the rate of cognitive decline. Lilly is proceeding with a new Phase III trial.
Bapineuzumab	Pfizer/Johnson & Johnson	Bind and remove the Aβ peptide that accumulates in the brain	Discontinued development of intravenous formulation in 2012 after two failed Phase III trials. A Phase II trial of a subcutaneous formulation is ongoing; results are expected in 2014.

The RPM Report

Now researchers think that if treatment were initiated earlier, patients might have a better chance for a potential clinical benefit – including the possibility of halting or slowing the progression of the disease.

“If you look at some of the progression models, the changes in amyloid beta and phospho-related tau, those things are in full swing by the time you actually reach cognitive impairment,” Targacept’s Beaver says. “And so if you can have some effect in earlier stages, you’re likely to see changes in those markers, and changes in those markers related to disease modification.”

But sponsors have been hesitant to invest in a new area of Alzheimer’s research without clear guidance from FDA on how those trials should be designed. Therefore, the draft guidance, although not immediately impacting the Alzheimer’s pipeline, offers an opportunity for drug sponsors to explore options for early-stage disease and start discussions with agency officials.

For a company like Targacept, whose business is focused on drugs for the central nervous system, the guidance “has opened a door…for us to enter into conversations with the agency,” Beaver says. The guidance will “encourage sponsors to come to the meeting table and have a conversation about these areas and these patients. Because I think that they believe that these patients deserve treatment. We certainly believe that.”

Roche is another company interested in early-stage Alzheimer’s disease. The company launched a web site, (www.earlysymptomsAlzheimers.com) in September 2011 with information for patients, caregivers, and health care professionals about pre-dementia (or prodromal) Alzheimer’s disease.

“Measurable Alzheimer’s disease symptoms can occur 12 years before clinical diagnosis with significant worsening in the last three years,” Roche said in launching the site. “Greater insights into the length of this early, ‘silent’ stage of Alzheimer’s disease could provide a window of opportunity for health care professionals to potentially change the course of the disease through therapeutic intervention before irreparable damage has occurred.” The press release notes that Roche has four products in Phase I and II studies for Alzheimer’s.

The draft guidance should also help quell frustration from some sponsors who have argued that FDA hasn’t done enough to help speed the development of new drugs. Some stakeholders have criticized the agency for treating Alzheimer’s disease as a lower-priority public health need than conditions like HIV or cancer – a characterization that FDA officials dispute. (See (Also see "Rethinking Alzheimer's: Mapping Out an Approval Process" - Pink Sheet, 1 Feb, 2006.).)

If nothing else, the draft guidance – just five and a half pages long – serves as affirmation to the drug industry that FDA has Alzheimer’s disease high on its priority list for drug development.

“We believe the guidance is a sign of the FDA’s dedication and focus on Alzheimer’s disease,” Lilly’s senior medical director of the Alzheimer’s disease team Eric Siemers, MD, says. “It is our hope that in its final form, this guidance will bring us one step closer to advancing the shared goal of discovering and developing innovative Alzheimer’s disease treatments that will change the course of the illness.”

Different Groups, Different Requirements

When talking about the draft guidance, it’s important to get the definitions correct. When FDA says “early-stage” Alzheimer’s disease, the agency is referring to patients at the symptomatic, pre-dementia stage. “We are not referring to patients with dementia,” FDA’s Kozauer stressed. “And we are also, for the most part, not referring to patients who are cognitively normal despite the disease occurring in their brain.”

In the draft guidance, FDA identifies two subgroups of patients with “early-stage” Alzheimer’s. (See Exhibit 3.)

The first group (labeled “1” in the exhibit), includes patients who are closest to the onset of overt dementia, or prodromal Alzheimer’s disease. “These are patients who are clearly cognitively impaired, but also are beginning to manifest functional impairment,” Kozauer said. “It’s not to the degree of a dementia diagnosis, but it’s there and conceivably you could test for it.”

The second subgroup (labeled “2”), includes patients at the earliest clinical stages, sometimes referred to as having “preclinical” Alzheimer’s disease. “They are beginning to manifest cognitive impairment,” Kozauer said. These patients may have very subtle cognitive defects that are perhaps only detected through testing, but have yet to have trouble walking – an important distinction when FDA talks about clinical endpoints.

Exhibit 3

Early-Stage Alzheimer's: A New Focus For R&D

An FDA draft guidance on early-stage Alzheimer's disease is expected to encourage development of drugs for a new population – symptomatic, pre-dementia patients.

Food and Drug Administration

One of the more challenging aspects of developing drugs for early-stage Alzheimer’s is how to assess the effect of an investigational drug in patients who are not yet significantly impaired.

“The subtleties of cognitive impairment in patients with early-stage Alzheimer’s can be difficult to assess,” Kozauer and the director of the FDA’s division of neurology products, Russell Katz, MD, acknowledge in a New England Journal of Medicineperspective piece published on March 28. “Moreover, the range of focus must extend to healthy people who are merely at risk for the disease but could benefit from preventive therapies.”

In the draft guidance, FDA is proposing different clinical endpoints depending on the subgroup of early-stage Alzheimer’s patients. As a reference point, for trials in later-stage Alzheimer’s patients who have already developed dementia, FDA wants a co-primary outcome measure: (1) a measure of cognition or cognitive performance; and (2) a functional measure or a global rating scale.

However, with early-stage Alzheimer’s disease, “that co-primary endpoint becomes impractical,” FDA’s Kozauer said. “There are no validated, stand-alone functional scales for early-stage disease. And in the earliest stages, these patients have no functional impairment to test for in the first place,” he said. “But there still needs to be some way of ensuring that there’s a clinical meaningfulness to whatever effect we’ll see.”

For the patients closest to the onset of early dementia, the draft guidance suggests that sponsors use a single, primary outcome measure that assesses both elements of cognition and function on one scale that can speak to the clinical meaningfulness of the result.

Kozauer specifically mentioned the Clinical Dementia Rating Scale-Sum of Boxes score as an example of a useable scale. He noted that CDRS-SB is one scale that had been proposed to the agency, so it’s the one to which FDA has given the most thought. But, he added: “We are aware that others have scales under development, and we are certainly open to having those discussions.”

Using The Accelerated Approval Pathway

FDA is proposing different clinical requirements for the second group of early-stage patients. This is a group of particular interest to drug sponsors. “If you had a treatment that you really believed to be disease-modifying – affecting the progression of the illness – these could be the patients who would have the most to gain, potentially, because they are the earliest on,” Kozauer said.

The challenge is that since these patients have very subtle cognitive defects, the only available clinical endpoint would be an isolated cognitive measure. (A sponsor couldn’t test for functional impairment because the patient has not yet reached that stage of disease progression.) Kozauer noted that there are several scales under development that would measure cognition in early-stage patients; they are typically composite endpoints that incorporate elements of existing cognitive tests.

But regardless of the scale used, the effect seen is going to be small, given that these patients are barely showing any signs of reduced cognition. “It would be very hard, if not impossible, to interpret the clinical meaningfulness of the cognitive effects,” Kozauer said. For that reason, FDA is proposing use of the accelerated approval pathway to approve a drug in the earliest of early-stage Alzheimer’s patients.

Accelerated approval is based on an effect on an intermediate endpoint that is reasonably likely to predict an ultimate clinical benefit to patients. In the case of early-stage Alzheimer’s disease, Kozauer suggested, that intermediate endpoint could be cognitive testing, with an ultimate endpoint of irreversible morbidity.

Kozauer stressed that to use the accelerated approval pathway, FDA would have to be convinced that the patient population enrolled in the clinical trials would be very likely to eventually develop Alzheimer’s disease. “We know that number is not going to be 100%, but we would really need to have a good understanding of that risk before we would be contemplating using this pathway.”

“For that reason, the state of the science at the time of any future marketing application that would seek to rely on accelerated approval would really be critical,” Kozauer continued. “For example, studies like the Alzheimer’s Disease Neuro-imaging Initiative are attempting to better understand which patients will progress on to develop dementia.”

“And it’s that kind of evidence that will weigh heavily on our decision to contemplate the use of accelerated approval,” he said. “I hope that’s clear. I can’t overstate that point.”

“An Extraordinarily Major Claim”

Of course, the ultimate goal of any Alzheimer’s R&D program is to find a drug that affects the underlying pathophysiology of the disease – that either slows its progress, or halts its progression completely.

And here, FDA is open to considering the use of biomarkers to support a disease modification claim – but only as a secondary primary endpoint in combination with a positive finding on a primary clinical outcome measure. The use of biomarkers alone, therefore, could not be used as the basis for a surrogate endpoint for an accelerated approval.

“Unfortunately, one of the lessons we’ve learned from the recent Phase III trials results is that the correlation between an effect on an Alzheimer’s disease biomarker and a clinical outcome is still very unclear,” Kozauer said. “We now have examples of situations of where the biomarker is altered, but there was no clinical effect. And at least in one case, the clinical outcome was the opposite of what you wanted to see.”

There is currently no consensus in the research community on an appropriate biomarker to support clinical trials in early-stage Alzheimer’s. FDA’s Kozauer and Katz referred to a number of potential biomarkers (including beta amyloid load, cerebral spinal fluid levels of amyloid and/or tau, and measurement of brain volume), but because none of these criteria have been validated, FDA cannot yet formally endorse a specific set of criteria.

The draft guidance underlines that point: “Until there is widespread evidence-based agreement in the research community that an effect on a particular biomarker is reasonably likely to predict clinical benefit, we will not be in a position to consider an approval based on the use of a biomarker as a surrogate outcome measure” – at any stage of Alzheimer’s disease.

Kozauer added that sponsors could use an alternative trial design – such as a randomized start design or a randomized withdrawal design – to potentially show disease modification without the use of biomarkers. But those come with their own set of challenges: “We realize these are difficult studies to design, conduct, and interpret, but they are possibilities, and we are open to using these sorts of approaches,” he said.

“We would like a study design that can be interpreted as showing a disease modifying effect with as few assumptions as possible,” Katz said. “We think the randomized start or the randomized withdrawal design has those qualities if it could be conducted appropriately, which is difficult. And we acknowledge that. But we’re looking for a design that forces that conclusion so we don’t have to assume that it’s demonstrating a disease-modifying effect.”

Either way, the hurdle for demonstrating disease modification is going to be high. “A disease modification claim will be an extraordinarily major claim for a sponsor to have,” Katz acknowledged. “One might, for example, decide if a particular drug were granted a disease-modifying claim, one could never do placebo-controlled studies; everybody would have to be on that drug.”

“So it’s a big deal to grant it, and so we want to be sure, if and when we conclude that a drug has a modifying effect, that it really does.”

“Not Loosening Guidelines”

The draft guidance, therefore, helps to paint a three-part regulatory picture for drug development in Alzheimer’s disease.

For patients with subtle cognitive deficits alone, a sponsor could use the accelerated approval mechanism. For those with increasing cognitive defects and detectable functional deficits, FDA would require a standard approval based on the effect of a single combined measure of cognitive and function, like the CDRS-SB score. And for traditional dementia patients, FDA’s requirements would remain as is; a standard approval would be based on a measure of both cognition and function, or a global rating scale. (See Exhibit 4.)

Exhibit 4

A Three-Pronged Approval Pathway

FDA's latest guidance on early-stage Alzheimer's disease outlines the new regulatory requirements for approval in two subpopulations of patients that are just starting to show signs of deficit. The requirements for patients who have already developed dementia have not changed.

Food and Drug Administration

What the new requirements do not do, FDA stressed, is ease the requirements for approval. Media reports that followed the NEJM perspective piece when it was published online suggested that the agency was “loosening” its regulatory guidelines by making it easier for drug companies to get drugs approved for early-stage Alzheimer’s disease.

When asked about those reports, Katz stressed that the draft guidance in no way eases FDA’s approval standards. “Of course we don’t believe [the draft guidance] is loosening our guidelines,” he said. “First of all, whatever approach is taken, there would have to be substantial evidence of effectiveness, whatever the outcome measure is, and whatever the population is, and that’s a given.”

“There currently is no standard for developing treatments in these early patients. There are no standards in place; we have never approved a drug for these folks,” Katz said. “It only makes sense, in our view, to utilize outcome measures that are appropriate for the condition being treated....We are attempting to tailor the outcome measure…to the population. It’s only appropriate to do that – we do that every day with every drug we deal with.”

Targacept’s Beaver agrees. The guidance is “defining a path that was previously undefined. I think most sponsors were reluctant to pursue early Alzheimer’s disease because the path was a complete unknown,” she says. “And so what the FDA has done is provide a path. That in and of itself is encouraging, and it also sparks interest from sponsors to be willing to take the risk to move forward in this area.”

Not A Silver Bullet

It is worth mentioning that not everyone at FDA is convinced that treating patients at very early stages of Alzheimer’s is the silver bullet that will slow the progression of the disease.

“A lot depends on your belief system,” Center for Drug Evaluation & Research deputy director for clinical science Robert Temple said at The RPM Report’s FDA/CMS Summit for Biopharmaceutical Executives on December 11. “There are people who believe that once the disease is present, you can’t do anything about it anymore. Whether that’s true or not remains to be seen.” (See (Also see "A Skeptical Take on Current Alzheimer’s Therapy: FDA’s Temple Says Problem Is Drugs, Not Endpoints" - Pink Sheet, 4 Jan, 2013.).)

Temple still believes a drug could be effective in advanced Alzheimer’s disease – it just hasn’t been found yet. “Unfortunately for the people who have it, it’s an easy disease to study. It’s progressive. It doesn’t bounce around like depression does very much. It mostly goes downhill and it has been historically easy to detect drugs that work.”

While Temple acknowledged it is possible that a drug would only work in patients who are not yet symptomatic – or at least not severely so – he noted that in cardiovascular disease, cancer, and other diseases, efficacy has generally been easiest to show in the sickest of patients. (Temple is not in charge of the reviews of Alzheimer’s disease treatments, but his opinions about drug development are held in high regard within industry and FDA.)

But the high risk associated with developing drugs for Alzheimer’s isn’t preventing sponsors from continuing to invest in the field.

Lilly, for example, is planning a third Phase III study for solanezumab; the drug failed the first two Phase III studies, but a pooled analysis showed patients with mild disease had a 34% decrease in the rate of cognitive decline. (See (Also see "Efficacy Signal For Lilly’s Solanezumab Stirs The Alzheimer’s Pot" - Pink Sheet, 15 Oct, 2012.).) And Lilly is continuing to invest: the company just announced it bought two tau imaging tracers and related technologies from Siemens AG’s Siemens Medical Solutions USA Inc. (See (Also see "Lilly Expands Bet on Tau By Buying Siemens’ Alzheimer’s Imaging Portfolio" - Pink Sheet, 17 Apr, 2013.).)

Even skeptics like Sanofi aren’t pulling out of Alzheimer’s R&D altogether; the company has two drugs in early-stage trials: an H3 antagonist (SAR110894) in Phase II, and an anti-protofibrillar Aß monoclonal antibody (SAR228810) in Phase I.

Targacept’s Beaver acknowledges the science isn’t completely clear, but she doesn’t think that’s a reason not to try. “The scientific landscape isn’t perfect to move forward, but that’s what the FDA is saying, and why they don’t endorse any particular path. They are essentially saying in this guidance: ‘Let’s discuss a path.’”

With Alzheimer’s disease, “it’s going to be a marathon to get there. And we have to continue to push forward. From our perspective, we are willing to participate in that marathon.”