In 2014 neurologist Jeffrey L. Cummings, MD, startled the Alzheimer’s disease research world with a paper that laid bare the alarmingly high failure rate of Alzheimer’s disease therapies in development.
Publishing in the journal Alzheimer’s Research & Therapy, Dr. Cummings and his colleagues determined that 99.6% of all therapies tested between 2002 and 2012 had failed. Since downloaded some 75,000 times, Dr. Cumming’s “99% paper,” as it came to be nicknamed, led him to look more deeply and thoroughly at Alzheimer’s disease drugs in the pipeline, and describe them in a readable, user-friendly way.
His “Alzheimer’s Drug Development Pipeline” report, in the journal Alzheimer’s & Dementia, classifies therapies by their targets, their mechanisms of action, and where they stand in the development process.
Dr. Jeffrey L. Cummings
Heavy on color-coded visuals, this snapshot of Alzheimer’s disease therapies is widely consulted by industry, researchers, and clinicians. Over time this report – which first documented a crisis – has come to show something more optimistic: an increasingly crowded pipeline reflecting a broad array of treatment approaches. Dr. Cummings wants more people to know that Alzheimer’s disease drug research, which now includes the first two Food and Drug Administration–approved monoclonal antibodies against amyloid-beta, is not the bleak landscape that it was in recent memory.
Lately, with the help of a grant from the National Institute on Aging, Dr. Cummings and his group have been working to expand on their reports to build an even more user-friendly database that can be searched by people in all corners of the neurodegenerative disease world. Dr. Cummings says he plans for this public-facing database to be up and running by year end.
Neurology Reviews spoke with Dr. Cummings, who is a member of the publication’s Editorial Advisory Board, about the genesis of his influential drug-tracking effort, how it has evolved, and what has been learned from it over the years.
How did all this begin?
Already in 2014 there was a dialogue going on was about the high failure rate for Alzheimer’s drugs. And I thought: “there’s probably a number that can be assigned to that.” And when it turned out to be 99.6%, that generated a huge amount of interest. That’s when I realized what interests me also interests the world. And that I was uniquely positioned after that point to do something annually.
How do you create your annual report, and how do you classify the drugs in it when some might act on little-understood pathways or mechanisms?
We capture information available on clinicaltrials.gov. We are notified immediately of any new Alzheimer-related trials, and we automate everything that is possible to automate. But there is still some human curation required. Most of that is around mechanisms. If it’s a monoclonal antibody directed at amyloid-beta, that’s not difficult to categorize. But with the small molecules especially, it can be more complicated.
We often look to see how the sponsor describes the drug and what their perception of the primary target is. A resource of great importance to us is CADRO, Common Alzheimer’s and Related Dementias Disease Research Ontology, which describes about 20 mechanisms that a group of scientists sponsored by the National Institutes of Health and the Alzheimer’s Association have agreed on. Inflammation, epigenetics, and oxidation are just a few that most people know. CADRO is organized in a very specific way that allows us to go to the mechanism and relate it to the target. But we do try to be humble and acknowledge we probably make some errors in this.
Are you able to capture every Alzheimer’s drug in development globally?
If they’re on clinicaltrials.gov, they’re in our database. But we think there’s about 15% of drugs in the world that aren’t for some reason on clinicaltrials.gov – so we know we are comprehensive, but not quite exhaustive. I’m in kind of quandary about whether to search for that other 15%. But we do always acknowledge that we’re not 100% exhaustive.
Who are the report’s main readers?
Drug developers use it for investor discussions, and also to understand the competition and the landscape. The competition might be a drug with the same mechanism, and the landscape might be drugs coming into the Alzheimer’s disease world. So if someone is developing a PDE-5 inhibitor for mild dementia, for example, they can see that other people are working on a PDE-5 inhibitor for moderate dementia, and there’s no overlap. Investors use the report to make decisions about which horse in the race to bet on. And of course it’s used by academics and clinicians to learn which are the new drugs in the pipeline, which drugs have fallen out of the pipeline, how are biomarkers changing trials, what are the new outcomes.
It’s really become a community project. Investigators will email me and say “Jeff, we’re in phase 1, make sure it’s on your map.” Or, “you forgot our agent! We’re disappointed.” When that occurs it’s because they were not in a trial on the index date – the 1 day in our publication when everything we say in the paper is true. A trial initiated 1 day later won’t make the report for that year.
What about patients and families? Are they able to use the report as well?
One of the things we want to expand with the new database is its usefulness for patients. Among the new data display approaches that we have is a world map where you can go click on a dot near your home and find active trials. That’s something patients and families want to know, right? There’s 140 drugs in clinical trials, there must be one for me, how would I get to it? Soon we will have quite a good public portal so if you want to go in and see what new monoclonal antibodies are in phase 2, you can do that with drop-down menus. It’s a very easy to use site that anyone can explore.