Crowdsourcing Project Predicts Progression of ALS

sciencehabit writes Using data from old clinical trials, two groups of researchers have found a better way to predict how amyotrophic lateral sclerosis (ALS) progresses in different patients. The winning algorithms—designed by non-ALS experts—outperformed the judgments of a group of ALS clinicians given the same data. The advances could make it easier to test whether new drugs can slow the fatal neurodegenerative disease. For the competition, participants were given just a slice of this data set, collected over 3 months, and asked to design an algorithm to predict how patients would fare in the subsequent 9 months, according to a standard functional scale that measures their ability to move and care for themselves. When predictions from the two winning algorithms were combined, they outperformed estimates solicited from a dozen ALS clinicians who pored over the same data, the authors report. They estimate that using these algorithms to predict outcomes could allow a drug sponsor to reduce the size of the trial by at least 20% and save as much as $6 million in a large phase III trial.

Comments

Too late for Lou Gehrig

[Anonymous Coward]

Or is it?

Re:Too late for Lou Gehrig

[ihtoit]

I would say very much so, considering he's been dead for 73 years.

Re:Too late for Lou Gehrig

[DoofusOfDeath]

I would say very much so, considering he's been dead for 73 years.

Or is it???

Damn I've netflix'ed too many zombie movies lately.

Re:Too late for Lou Gehrig

[ihtoit]

oh, snap. Ya got me.

Or is it?

Re:Too late for Lou Gehrig

[rossdee]

They should rename it Stephen Hawkings disease

Re:Too late for Lou Gehrig

[Anonymous Coward]

Sly Stalone disease.

Re:Too late for Lou Gehrig

[sconeu]

Actually, the ALS Association is trying to just get it commonly called ALS, so as to emphasize that anyone can get it.

To be honest, I prefer the name Lou Gehrig's. Everyone knows about it that way.

Re: Too late for Lou Gehrig

[superswede]

Everyone in the US, but in the rest of the world almost no one knows who Lou Gehrig was.

It's better to call deceases by their medical names and not based on some person's name.

Re:Too late for Lou Gehrig

[Anne Thwacks]

Or perhaps What is it?

Please can we have using TLAs without explaining them made a criminal offence.

Re:Too late for Lou Gehrig

[Holi]

I don't know if the summary was edited but this is in the summary: "a better way to predict how amyotrophic lateral sclerosis (ALS) progresses in different patients"

Re:Too late for Lou Gehrig

[Anonymous Coward]

The ALS association obviously wants people to use ALS because its good branding for them.
In the UK and Commonwealth Countries it's known as Motor Neuron Disease.

The savings amount to using previous trial data as a control, which would never be accepted by the FDA in drug trials because it won't control for any features specific to the actual drug trial.

This also shows that large amounts of modern medicine, including high tech specialties like neuroscience, have not been subjected to basic statistical analysis and evaluation. There is a lot of benefit to gained in proper digital medical records and making anonymised data available to statisticians and researchers. There is no reason by the medical data on everybody in the Developed World isn't being used to look for correlations and evaluate medical treatments.

There are a couple of drug trials for ALS in progress, but it isn't an area that isn't highly attractive to drug companies. There is no proven cause so what to target is a crap shoot. Most patients suffer from sALS (randomly occurring) and only live 2-4 years from diagnosis. Neurodegenerative diseases and protein folding diseases in general have proven a money pit for research. Many of the drugs that have gotten onto the market do bugger all and there are no cures. Riluzole is the only effective drug for ALS, its a failed psychotropic drug that slow progression of ALS by an average of 9%, but it can't be taken once respiratory function becomes compromised. Researcher from other sources into the cause and mechanism is probably going to have make a breakthrough first.

Cost savings - so what?

[DoofusOfDeath]

Seems to me that drug makers will charge whatever maximizes their sales revenue, regardless of their development costs.

Lowering their development costs for them might affect whether or not they bring it to market. But if they do, they'll charge as much as they can get away with.

Re:Cost savings - so what?

[ihtoit]

they're looking at stem cell therapy as a potential cure for ALS. What's that worth to someone suffering from it?

Hell, if I ever find myself looking down the ALS muzzle, I'd be the first to volunteer as a pincushion. I go fucking nuts if I'm bedridden with the latest fashionable illness, I would go batshit if I suddenly found myself permanently unable to fucking swallow.

Re:Cost savings - so what?

[sconeu]

Especially for a "market" as small as ALS patients.

As recently as two years ago, there was exactly ONE drug approved for ALS treatment: Rilutek (riluzole). And all it did was extend life about 6 months on the average.

The cost? over $1000 per month. At the time, it hadn't gone generic.

Re:Cost savings - so what?

[DoofusOfDeath]

My point was simply that reducing the development cost for an ALS drug may be irrelevant from the perspective of those who want the finished product. Unless it's the difference between the drug coming to market or not.

Of course, I'm just guessing at that, but I was trying to start a discussion.

Re:Cost savings - so what?

[sconeu]

My late wife was in a clinical trial for ALS. Unfortunately, it didn't work and there was no Phase3 III. We also found out after the fact she was in the placebo group.

But you're absolutely right intoit, if you're ever unlucky enough to lose the ALS lottery (or have your SO do so), you will grasp at every fucking straw that comes your way. If that includes becoming a pin cushion or a guinea pig, you do it.

Re:Cost savings - so what?

[Nutria]

"Doofus" is an apt description for someone who says "So what?" to reducing the cost of drug testing.

drug makers will charge whatever maximizes their sales revenue, regardless of their development costs.

And insurance companies (including Medicare/Medicaid) know this, and thus negotiate much lower prices for drugs in their Formularies. Thus, insurance companies can negotiate even lower costs.

Or... the drug companies run same-size tests and get more accurate predictions on outcomes.

Re:Cost savings - so what?

[DoofusOfDeath]

I'm sorry about your wife.

Re:Cost savings - so what?

[ihtoit]

an associate of mine was diagnosed with ALS in 1998 following a single incident of unexplained loss of motor control (which just so happened while he was meeting with a physician). He was permanently confined to a wheelchair within three weeks and a pine box 4 months later. It was terrifying how quickly it took him considering his lifestyle (cycled *everywhere*, his car actually expired in his garage through underuse). ALS is not fussy about how fit you are or how stressful your job is or pretty much any other external influence. If it wants you, it'll just take you - on its own timetable.

Re:Cost savings - so what?

[ihtoit]

the way they negotiate lower prices is to guarantee exclusivity in the supply chain. A treatment available on Medicaid from GSK won't be available through the same route from Pfizer, because Pfizer will be frozen out of the chain by virtue of a GSK-included clause in the Medicaid contract.

To borrow a car analogy, it's like buying a 2014 Rolls Royce Phantom (V12 Merlin, come on! Tease me some more!) and being told that your warranty will be null and void if you swap out the engine block for a 410 Ford V8.

Re:Cost savings - so what?

[Nutria]

the way they negotiate lower prices is to guarantee exclusivity in the supply chain. A treatment available on Medicaid from GSK won't be available through the same route from Pfizer, because Pfizer will be frozen out of the chain by virtue of a GSK-included clause in the Medicaid contract.

It sounds like you're referring to drug patents and the lack of certain drugs in the generics market. Am I misunderstanding you?

If not, then you need to do a better job explaining, since if that's not drug patents then what you wrote makes zero sense.

Re: Cost savings - so what?

[Anonymous Coward]

You are right. A good example of this is the new HCV drug that can cure the disease. These are corporations that are making these drugs and they need profit. If they gave away their winners, then how would the recover their costs of their losers and still make a profit?
The reality of a for-profit system is that some people will die because they can't afford treatment. If this makes you upset, then you should try to encourage the government or charities to take a drug all the way through Phase III.
Don't hate the player - Hate the game.

Re:Cost savings - so what?

[ihtoit]

no I'm referring to lack of certain generics from certain suppliers through certain routes, and the theory that it's down to contractual clauses rather than scarcity of raw materials or the technology for synthesis. Patents have fuck all to do with it.

Re:Cost savings - so what?

[Nutria]

I guess I don't know whether fuck all means "everything" or "nothing".

Complete Misunderstanding of Epidemiology

[The_Laughing_God]

They can't validate a scale using unintervened progression or existing treatments, then pretend it says ANYTHING about a new/unknown treatment. The whole point of a new treatment is to alter the progression of the disease in a new/different way; the whole point of clinical trials is to determine the NEW course of the disease using the NEW treatment.

The claim made here is: a better tool to predict the time progression of headaches treated with aspirin (or beer or sex) can better predict the time progression of a headache treated with some yet-uninvented drug, so we needn't test the new treatment as thoroughly to characterize it. That's like saying "the more predictable sex with your partner is, the more you know about sex with a different partner"

And yes, I AM a physician and molecular biologist.

Re:Complete Misunderstanding of Epidemiology

[Anonymous Coward]

But if you have a better understanding of how the current treatments work in terms of patient outcomes, then you need fewer patients in your control arm.

The more predictable the sex with your partner is the less often you need to do it to remember how it is. Therefore you can have a lot more sex with other people.

Re: Complete Misunderstanding of Epidemiology

[Anonymous Coward]

Remember that the end points of a trial determine their length and the number of participants needed for significance. If you can reach your endpoint sooner, then you may be able to reduce the number of participants.

Re:Complete Misunderstanding of Epidemiology

[EETech1]

Waiting at my doctors office wouldn't bother me near as much if I thought he was posting to Slashdot:)

PRO-ACT

[Graydyn Young]

To me the most interesting part of the article is the creation of the PRO-ACT database. It's an incredible chunk of data, and I hope it inspires similar projects.