Subscribe to eTOC

Why Alzheimer's Investigators Are Undeterred by Negative Top-Line Results from DIAN-TU

Article In Brief

The announcement of disappointing results from the DIAN-TU study elicits discussion about the future directions of Alzheimer's disease research. Leading researchers said results from other ongoing studies are reason to be hopeful.

Seven years after the first of nearly 200 patients with autosomal-dominant Alzheimer's disease began treatment, neither solanezumab nor gantenerumab slowed their cognitive decline, according to top-line results from the Dominantly Inherited Alzheimer Network Trials Unit (DIAN-TU).

The spare announcement, in statements released on February 10 by Roche (maker of gantenerumab) and Eli Lilly and Company (solanezumab), covered only the lack of a significant effect on the trial's primary endpoint after an average of five years: the DIAN Multivariate Cognitive Endpoint (DIAN-MCE), a novel measure gleaned from previously established tests. The releases offered no word as to whether any effects were seen on the more than 100 biomarkers or other cognitive tests included in the trial. Both, however, noted that other clinical trials involving the same drugs—the phase 3 GRADUATE 1 and 2 trials for gantenerumab, and the Anti-Amyloid Treatment in Asymptomatic Alzheimer's (A4) Study for solanezumab—continue to move forward. Initial results from the A4 Trial are expected in mid-2022.

The DIAN-TU study's leaders, expert commentators, and a senior official at the National Institutes of Health's National Institute on Aging (NIA) all offered the same, surprisingly resolute message: Damn the top-line results, full speed ahead.

“We're not at the end of the road—quite the opposite,” said Eliezer Masliah, MD, director of the NIA's division of neuroscience. “The investigators have built something really extraordinary: a stepping stone to the future, with a cohort of families who have the rare genetic form of the disease, who participated for an average of five years, and come from countries around the world. The top-line results of this trial weren't positive, but the infrastructure for further clinical trials has been created.”

Emphasizing that he still expects to see positive effects on at least some of the study's biomarkers as his team combs through the data, the principal investigator of DIAN-TU told Neurology Today that enrollment for the next trial involving the families has already begun.

“We have four statistical teams working around the clock on all this data,” said Randall Bateman, MD, director of the DIAN-TU and the Charles F. and Joanne Knight Distinguished Professor of Neurology at Washington University in St. Louis. “Depending on the biomarker, we do expect there to be some effects. If we determine there is the potential for benefit, an open-label extension to the study will be considered. In addition to this trial, there is ongoing recruitment and enrollment for the next drug arms, which will target tau.”

A key shortcoming of the trial was that patients spent more than half of it titrated to a dose later found to be far short of optimal. In 2017, based on results from other trials, the DIAN-TU participants were titrated to target doses four- to five-fold higher than at the study's outset.

“Knowing the limitations of the small number of participants we had in both drug arms, and that we had to change the drug doses midway through the trial, we knew those increased the likelihood that the outcomes would not necessarily be positive,” said the trial's associate director, Eric McDade, DO, associate professor of neurology at Washington University.

Study Details

Established in 2012 with funding from Lilly, Roche, NIA, the Alzheimer's Association and other charitable groups, the DIAN-TU network included 24 study centers in the United States, Australia, Canada, France, Spain, and the United Kingdom. More than 60 percent of the 194 participants were asymptomatic at enrollment, Dr. McDade said, and fewer than 40 percent had mild cognitive impairment. They received active drug or placebo by IV or subcutaneous injection every four weeks.

Fifty-two patients were randomized to gantenerumab, an investigational monoclonal antibody previously shown to bind to aggregated forms of amyloid-beta and to remove amyloid-beta plaques. Another 52 patients were randomized to solanezumab, and the rest to placebo.

According to the release from Lilly, only 36 of the solanezumab participants completed the minimum four-year treatment period. The initial study dose of solanezumab was 400 mg every four weeks. “A late amendment to the study increased the dose resulting in approximately 25 percent of the total doses being administered at the 1600mg level,” the Lilly release stated.

The most common adverse events reported more frequently with gantenerumab than placebo, according to Roche, were “injection-site reactions, infection of the nose and throat (nasopharyngitis), and amyloid-related imaging abnormalities (ARIA), manifesting as cerebral edema or microhemorrhages. The majority of ARIA findings were asymptomatic; if symptoms occurred, they were mild in nature and resolved.”

The Lilly statement did not include a description of adverse effects but noted that more detailed data will be presented on April 2 at the Advances in Alzheimer's and Parkinson's Therapies annual meeting in Vienna, followed by presentations at the Alzheimer's Association International Conference in Amsterdam in July.

An abstract has also been submitted to AAN for the annual meeting in Toronto, Dr. Bateman said. The trial's primary clinical endpoint, DIAN-MCE, includes the Wechsler Memory Scale-Revised Logical Memory Delayed Recall, the Cogstate International Shopping List Test, the Wechsler Adult Intelligence Scale-Revised Digit Symbol Substitution Test, and the Mini Mental State Examination.

Responding to the negative top-line results, Dr. Bateman said, “Certainly we were saddened by the news, as were participants and family members. In some cases people were in shock. They so much wanted these drugs to have a cognitive benefit. But I have also heard from multiple family members how thankful they were for the trial, and how much they want to see it continue. The most common question I hear is ‘What's the next trial I can sign up for?’”

Upon review of the biomarkers, he said, “We will take a hard look at the biological engagement of these drugs in the AD disease process. Some of the biomarkers measure the drugs' engagement with pharmacodynamic targets. We do expect those to be engaged. As we went from a low dose to a high dose, we would expect to see more of an engagement. The other biomarkers, such as amyloid plaque levels, we hope to see some of those changed too, but we won't know until we analyze the data. If those were affected, we can see whether they had any engagement with other downstream biomarkers and cognitive and clinical measures.”

Dr. McDade said he is especially hopeful to see a lowering of amyloid plaques in the brains of participants randomized to gantenerumab after the dose was increased based on the data Roche has presented on late-onset AD patients exposed to the higher doses in an open-label study.


“The fact that they had fewer than 200 people to begin with, on two drugs, and placebo versus active, you get down to cells of under 50. I really would like to see the underlying data. The biomarkers could provide a very interesting story.”—DR. RONALD C. PETERSEN


“We have four statistical teams working around the clock on all this data. Depending on the biomarker, we do expect there to be some effects. If we determine there is the potential for benefit, an open-label extension to the study will be considered. In addition to this trial, there is ongoing recruitment and enrollment for the next drug arms, which will target tau.”—DR. RANDALL BATEMAN

Once the data are reviewed, Dr. Bateman said, “we'll be making a determination about whether either of these drugs would continue in an open-label extension, depending on whether there was evidence of some kind of benefit.”

Meanwhile, for the next trial with the DIAN-TU network, Dr. Bateman said he plans to use a drug targeting tau rather than amyloid.

“We think tau is more proximate to the symptom onset,” Dr. Bateman said. “It also has higher correlation to symptom onset. The amount of tau pathology appears to correlate better with the cognitive performance of individuals.”

The focus of another trial in the DIAN-TU network will be to start with patients not only before they display cognitive decline, but before evidence of amyloid brain pathology is detectable, said Dr. McDade.

“The platform has been set up,” he said. “That was the heavy lift. Now we can move forward testing other agents in this unique patient population.”

Expert Commentary

Dennis Selkoe, MD, FAAN, the Vincent and Stella Coates Professor of Neurologic Diseases at Harvard and Brigham and Women's Hospital, spoke for every other expert reached for this article when he called the trial a “wonderful, heroic effort by an incredible team of investigators,” but that, based so far on the minimal top-line results, the findings are disappointing.

“All we know for now is this top-line analysis. It's quite frustrating not to have a more detailed analysis yet to show whether there were any trends, any effects on biomarkers. Did the amyloid level move at all? I expect it did. Were there PET scan alterations? We need to know.”

Dr. Selkoe noted that the findings come, as he put it, “in the face of the quite striking and surprising results from Biogen on aducanumab, presented last December 5 at the CTAD meeting. The phase 3 EMERGE trial, which was huge, had some quite lovely data that most would agree had a clear-cut, positive outcome on cognitive decline. Then came the parallel ENGAGE trial, and that did not achieve significance. But when Biogen revealed a subset analysis of ENGAGE, those who had the highest, longest exposure to aducanumab did see a benefit, similar to that in EMERGE. And Biogen also had a full biomarker set showing that amyloid levels went down dramatically and tau also went down in CSF and on tau-PET scans. This essentially proves the amyloid hypothesis.”

On October 22 of last year, Biogen announced that, after consulting with the FDA, it plans to seek approval for aducanumab based on the results from ENGAGE and EMERGE.

“That biomarker data from Biogen was unequivocal—it lowered amyloid dramatically, which led to less tangle burden,” Dr. Selkoe said. “So you can't use this new announcement from DIAN-TU to say we're going backward. Without the biomarker results, we simply don't know whether they actually got amyloid lower.”

Even so, in such an aggressive disease as autosomal-dominant Alzheimer's disease, he said, “as my colleague Sterling Johnson would say, one needs to close the barn door before the horse gets out. You can't expect a trial with 50 people per arm, a third of whom were already symptomatic, to serve you well.”

Dr. Selkoe said he hopes for more clear-cut results from the Alzheimer's Prevention Initiative Autosomal-Dominant AD Trial, in which crenezumab or placebo will be given to presymptomatic PSEN1E280A mutation carriers in Medellín, Colombia.

Ronald C. Petersen, MD, PhD, FAAN, director of the Mayo Clinic Alzheimer's Disease Research Center, said the negative results reported so far by DIAN-TU were not totally unexpected.

“The concern since its initiation has been its power, particularly with cognitive outcomes,” Dr. Petersen said. “The fact that they had fewer than 200 people to begin with, on two drugs, and placebo versus active, you get down to cells of under 50,” he said. “I really would like to see the underlying data. The biomarkers could provide a very interesting story.”

He agreed with others that the expectations for gantenerumab are stronger than for solanezumab.

“Some people would say that solanezumab may not be as effective, because it's going after the early, monomeric form of the disease,” Dr. Petersen said. “While you might think that makes good sense, it just might not result in a significant effect. There have been three failed trials of solanezumab, in EXPEDITION 1, 2 and 3. The third one did show some clinical effects but they were very small.”

Paul Aisen, MD, director of the University of Southern California's Keck School of Medicine Alzheimer Treatment Research Institute, said of the DIAN-TU investigators, “I think they faced enormous challenges and made a tremendous effort. But given the small sample size and the range of degrees of impairment from symptom-free to having significant symptomology, obtaining clearly positive results was a real long-shot.”

On February 6, the Keck School announced that Dr. Aisen, along with investigators from Harvard, will lead a phase 3 trial of BAN2401, a humanized monoclonal antibody that binds to amyloid-beta. The four-year AHEAD3-45 study will involve 1,400 clinically normal people who have intermediate to high levels of amyloid in their brains. Enrollment is expected to begin on May 31 and to be completed in 18 to 30 months. The study is funded by NIA and Eisai.


“The platform has been set up. That was the heavy lift. Now we can move forward testing other agents in this unique patient population.”—DR. ERIC MCDADE

Dr. Masliah pointed out that NIA is currently funding nearly four dozen pharmacological trials for the treatment of AD, some focusing on aggressive control of hypertension, some targeting inflammation, others targeting tau, amyloid, cell death, and brain rhythms. He pointed to the research by Frank Longo, MD, PhD, at Stanford, who is studying LM11A-31, which binds to neurons and helps them to survive.

“He has a phase 2 trial that we're very excited about,” Dr. Masliah said.

Another drug, now in a phase 3 trial, is the antiseizure medication levetiracetam, previously shown to normalize brain rhythms in AD. (For treatment of mild cognitive impairment [MCI], the drug is called AGB101.)

“It could be very helpful for memory loss in MCI,” Dr. Masliah said. “What we want people to know is that we are committed to having a diversified portfolio.”


Dr. McDade has received an honorarium for CME-sponsored activity by Lilly as well as personal compensation for serving as a member of Lilly's data and safety monitoring board. Dr. Selkoe has received directors and consulting fees, and travel expenses from Prothena Biosciences. Dr. Aisen has research contracts with Lilly, Janssen, and Eisai, and has consulted with Biogen, Merck, Roche, Samus, ImmunoBrain Checkpoint, and Proclara. Dr. Petersen has received consulting fees and an honorarium from Merck, Inc., Roche, Inc., and Biogen, Inc. He also serves on the data and safety monitoring board for Genentech.

Link Up for More Information

• Lilly announces topline results for solanezumab from the Dominantly Inherited Alzheimer Network Trials Unit (DIAN-TU) Study. Accessed February 27, 2020.