Developing drugs for treating, let alone curing Alzheimer’s is one of the more difficult prospects facing researchers (see here). So a new consortium called the Coalition Against Major Diseases announced plans to create databases to share research on neurodegenerative diseases. The effort, which will be managed by the non-profit Critical Path Institute, includes research foundations, patient groups and several drugmakers. The databases will be openly shared by drugmakers and other researchers in hopes of setting a voluntary industry data standard. For starters, CAMD will have data from 4,000 Alzheimer’s patients who participated in 11 failed industry studies. We spoke briefly with Marc Cantillon, the CAMD director, about the project…

Pharmalot: Where is the funding coming from?
Cantillon: It’s been started by the state of Arizona, since the Critical Path Institute is bsed there. For the first couple of years, it will invest $3 million and that will be matched with $1.5 million from the FDA each year, but there will be a transition, given state finances these days to more funding from the FDA. But, no, we don’t take pharma funding.

Pharmalot: How is this supposed to work?
Cantillon: The database is only raw data by itself. It’s only a beginning. It will be publicly available, and not just for those who collaborate on data or produce data. A developer can go in and sample whatever they need. You may have a question about the age of patient or genetic background and be able to get a subset out of the large subset and then hone down your target population, and then run a simulation to see what your drug may look like or take info to power your own drug. The point is we should learn from the failures and experiences, and not make the same errors. It’s a whole new way. You could say it’s benevolent, but it’s really selfish benevolence, because everyone can potentially gain. You know, having a few hundred patients is not eough while thousands can make a better case.

Pharmalot: Are there any intellectual property issues?
Cantillon: That’s the whole reason why this works. The basic agreement is that at a precompetitive stage the data is fairly shared and afterward people can go forward with development. People have talked about sharing data for years. In this case, only placebo arms from many trials, not the actual drugs or compounds are involved. I know people worry about anti-trust issues. The sharing part could involve sciencists from various working groups, looking at biomarkers, statistics, data modeling, databases. There will be telecons among people from industry, patient advocacy, regulators…to produce documents and review literature. The overall goal is to offer regulators what seem to be qualified tools for understanding the diseases.

Pharmalot: How will different data sets be reconciled to the point where the data is useful?
Cantillon: Mostly, the data are consistent. There are fairly standard tools across the board. The challenge, though, is to make them exactly the same. Each of the scales may have been used differently in some companies. For instance, the date of birth may have been combined differently so it make its impossibe to combine datasets. What we’ll do is reemapping. We’ll go in and, literally, standardize. We’ll use CDISC standards, which are going to be recomnended by the FDA. Going forward, we’ll have one way to agree on how to do everything. Most pharma companies have agreed, independently, with what CDISC wants to do. We’re adding on the Alzheimer’s component.

Pharmalot: How much money do you think this can save and how much faster might this speed development:
Cantillon: Each Alzheimer’s patient may cost $50,000k per trial. And the database has 4,000 patients right now. A trial would easily take 5 to 7 years without collaboration and several hundred million dollars. As for develoment, the savings is in real time. Instead of doing risky, underpowered tirals, some would call them crappy, we can leapfrog over several steps and make use of the info and design a much more efficient trial and a smaller number of trials in a shorter amount of time.