Hacker News Reader

While this comment is funny, it's also arrogant and incorrect. It's not fair to indict folks trying to change lives based on what's needed to drive pageviews, ie attention grabbing headlines.

30 seconds of research provides a pretty obvious connection, regardless of your "no fucking clue what they're doing" accusation...

Nilotinib is a Tyrosine Kinase Inhibitor https://en.wikipedia.org/wiki/Nilotinib

Tyrosine kinase inhibitors regulate serotonin uptake in platelets http://www.ncbi.nlm.nih.gov/pubmed/7589171

Tyrosine kinase in serotonin induced cerebral constriction http://www.ncbi.nlm.nih.gov/pubmed/9472895

It's apparent that Nilotinib plays a role with Serotonin, which is of course implicated in Parkisons and many other neurological disorders, but I guess we are hung up on cute headlines and not the real story. http://phys.org/news/2010-06-serotonin-decades-old-mystery-p...

[edit:sp]

Also, drug repurposing is often a highly data-driven field. So it may not involve predictions where there's a high degree of certainty, but data-driven methods can shift a 1 in a million (or more) chance of success into a 1 in 3 or 1 in 5. Such wins are absolutely huge.

Throwing things at the wall to see what sticks is precisely how drugs were developed before the 1970s. It's called phenotypic screening, and it worked quite well. There's an argument to be made (Derek Lowe makes it here [0]) that it worked better than the modern method of target-based drug discovery.

[0] http://blogs.sciencemag.org/pipeline/archives/2012/04/16/phe...

I see what he's saying, but I wonder. There has to be a point of diminishing returns. Maybe we've already discovered the easy ones, and yet more mass screening won't turn up much.

I'm not sure if I'd say that "the real headline here is that these people have no fucking idea what they are doing" - the headline really is that some researchers have seen an exciting result in combating Parkinson's.

Medicine is really really hard, and our understanding of the human body and the interactions it has with various chemicals and compounds is limited- but do these researchers have no idea what they're doing? That they started a trial on Parkinson's patients with nilotinib suggests they have some idea...

> That they started a trial on Parkinson's patients with nilotinib suggests they have some idea...

Not when they made the drug, they didn't.

I can't edit this, but I'm curious about why the downvotes. The trial on Parkinson's patients was probably the result of unexpected improvements in Parkinson's patients who received it for cancer treatments. The drug wasn't developed with this use in mind.

How this has worked in the past is

1) come up with a bioligical system where you can screen for your disease of interest; something that grows in a dish and reports if a candidate drug works or not

2) Run that test against an absolutely huge library of drug candidates

3) Proceed with further testing on the candidates that seem most promising based on the initial screen.

4) Almost nothing works. Rinse and repeat.

It should be no surprise that a small compound that's biologically active in one screen may end up also affecting some other aspect of the body.

In a sense, you're right that we have no fucking clue what we're doing, but that's mostly because we're just now starting to invent most of the tools that let us even see what we're doing. That, and this same procedure of throwing stuff against the wall has worked so well for so many things.

The process of discovery is quite often one of luck, and a lot of drugs have been the effort of making our own luck. When we get extra accidental luck off that, I don't think it's fair to criticize.

I'm sure people with Parkinson's and their families really feel bad that this was discovered in a way that doesn't align with your sense of what's fitting.

As I understand it, most drugs are still a pretty "blunt" from the standpoint of biochemistry. Too much of X causes some problem, so you make an antagonist for X. But X does a half dozen other things, and your X antagonist is going to affect them all. 99% of the time we call them "side effects", and they're bad. It's not that they have "no fucking idea" what they're doing; it's that sometimes there's no way to isolate the system they want to affect. The human body wasn't designed for external maintenance.

Yup. Those are called "off target effects". Think of a drug that has a well known mechanism of action like beta blockers (they block the beta form of the adrenergic receptor). Go one step further and find a beta-1 antagonist like atenolol.

Now look at the binding constant across other receptors. The selectivity can be relatively small! A beta blocker will often bind with relatively high affinity to another 5 or 10 receptors!

Sildenafil (Viagra) started as a blood pressure medication and is still used for that purpose, in addition to it's more popular erectile dysfunction usage.

Similarly, Finasteride (Propecia) is still used for enlarged prostate.

It's not that simple.

Drugs that make it through FDA safety testing are valuable even if they prove ineffective for their original intent, because they've been proven to be at least not immediately deadly to humans. It's much cheaper to test an already-safety-tested drug for curing other diseases than to start developing a new drug from scratch. In fact, this is most of the research that major pharmaceuticals are currently doing. It's actually quite effective, but I do worry that it's detracting from research into new drugs.

Not sure that this case is a good example of your point, but in general what you say is true.

It's interesting to consider why this is the case. Biology probably has more in common with the study of history than it does with physics or chemistry. The difference I think lies partly in the power of abstractions relative to each field of study. Abstractions in biology are generally pretty useless I think, whereas in physics in particular they are incredibly powerful. It seems that we pursue knowledge in biology under certain misleading assumptions - that what we find in one system has some universality; that the dynamics of the system are not that important; that the functional elements we discover (or invent?) largely have one function. So we like to say that Gene X does this, then publish it. Actually Gene X does many things that vary depending on the context of the cell, the organism, the environment etc, which we neither understand nor have the ability to control very easily. The knowledge is mainly in the details rather than any abstraction about it's function. We are confusing the issue by trying to think this way. It's like trying to understand the empire of Genghis Khan by looking at ancient Egypt - there are some broad similarities, but what is interesting and informative are the specific details in each case. The problem comes with making predictions about the perturbation of biological systems. With faulty abstractions and no understanding of the important assumptions, predictions inevitably fail miserably.

Biologists work out the next experiment with their heads. Even the simplest bacteria or virus has been around for billions of years, shaped by interactions with countless other organisms. How can one human brain possibly comprehend the important details that are known, let alone everything we don't know, to make a valid prediction?

Viagra was originally trialled as a drug for heart related health problems, when it went into human trials the subjects reported increased bloodflow elsewhere and as a result the use was changed to deal with erectile dysfunction.

interestingly they are now looking at it again as a heart disease drug.

IIRC, it wasn't that patients reported increased blodflow down there, but that a unusually large number of patients didn't return the drugs after the trial ended. So the researchers asked around, and learned that they trial patients didn't want to give up this "bonus".

Regarding the first half of your comment, don't forget about Viagra, which was "designed" to treat angina. It didn't do so well during clinical trials, but there was this one weird side effect that kept coming up...

Developing new drugs is so damn expensive now. The trials and dealing with the FDA. Not many investors want to get involved in developing new drugs so now they turn to repurposing drugs.

I have a friend who is a CSO of a startup trying to get a weightloss drug passed. The stories I hear. Definitely not fun dealing with the FDA and securing more funding.

This is exactly how I feel about the pharmaceutical industry. The brain is such a vast and unknown universe and we're hitting it with wide hammers to see what will happen. Wasn't it just shown that Prozac did more harm than good?

"replacing symptoms with side-effects".