Monday, August 27, 2012


WaPo tells us that antibiotics ain't what they used to be, or something:

Between 1945 and 1968, drug companies invented 13 new categories of antibiotics, said Allan Coukell, director of medical programs at the Pew Health Group.

Between 1968 and today, just two new categories of antibiotics have arrived.

In 2011, the Food and Drug Administration approved one new antibiotic, which fights one of the many bacteria, Clostridium difficile, causing deadly hospital-borne infections.

“What kept us out of trouble for the last 60 years is that every time drug resistance caught up to us, the pharmaceutical companies would go back to the drawing board and develop the next generation of drugs to keep us ahead of the game,” said Brad Spellberg, an infectious diseases physician in Los Angeles who heads a microbial resistance task force for the Infectious Diseases Society of America. “That’s the part of the equation that’s changed. Drug companies are no longer trying to get one step ahead.”

But why is that?

Experts point to three reasons pharmaceutical companies have pulled back from antibiotics despite two decades of screaming alarms from the public health community: There is not much money in it; inventing new antibiotics is technically challenging; and, in light of drug safety concerns, the FDA has made it difficult for companies to get new antibiotics approved.

As a result, only four of the world’s 12 largest pharmaceutical companies are researching new antibiotics, said David Shlaes, a drug development veteran and consultant.

The first reason is certainly true enough -- bringing a new drug to market is an expensive proposition and you can get screwed up a number of different ways before it ever gets there. And if one thing goes wrong, the lawyers will take you for millions. The second reason is true, but we've never shied away from technical challenges if there was enough reward in it, so it's more a problem related to the first. And as for the third reason, well, that would be easy enough to change except that bureaucrats are even more ferocious than personal injury lawyers where turf is involved.

So how do you solve the problem? Either you let the pharmaceutical companies make more money and simplify the regulatory process, or you drive them out of business and turn things over to the feds. Which one do you think will work better? And which one do you think the respective presidential candidates will choose?

On such considerations elections turn. Or should, at least.


Brian said...

I have more thoughts on this than time to articulate them well. So I'll just mention one: if you want deregulation in this industry for the sake of speeding up drug development, the public has to be willing to accept a MUCH higher degree of risk with drugs in the pipeline.

Additionally, even if you loosened the rules to speed up development, industry will only push it so far unless their liability for adverse reactions is substantially more limited than it currently is.

This isn't to say that simply turning things over to the feds is a good idea, either. (I don't really accept your implicit either/or proposition.) But this does point to the fact that there is a place for publicly funded health research, especially in areas of inquiry that are not clearly profitable in the market but have massive (potential) positive returns to public health.

Mr. D said...


I don't really disagree with any of what you've said. The larger point I'm trying to make is about incentives. When you look at the risks and rewards as things are currently situated, it's difficult to blame the drug companies for acting as they have.

I don't want deregulation per se. What I'd suggest is that a regulatory system that causes drugs to have a 10-15 year cycle to get to market is broken.

As for public health research -- sure. Nothing in what I wrote contradicts your observation. The subject is how things are in the private sector.

Brian said...

Yup, incentives matter. I don't know how easily to streamline approval. Even if you had a bureaucracy running at 100% theoretical efficiency, you can only run (good) clinical trials so quickly.

A bit off topic (but not entirely)...I worry about the effect of consolidation in the industry. If I and a couple of colleagues have a great idea for a biologic drug, manage to get some venture capital together, and make a run at it, we essentially have 3-5 years before we're either going to go out of business or be bought by Amgen or Genentech. And a lot of those acquisitions by the big guys are not so that they can develop the product, but so that they don't have to compete with it on the off chance it turns out to be successful.

You might say "why would they do that?" but if you determine a startup's drug has (say) a 1/5 chance of success and will cut into your existing market share if it does, it can actually make financial sense to buy off the potential competition (cheaply) and not spend the money on the relatively risky prospect of bringing it through development.

I don't know how to fix those incentives, either, except maybe by issuing longer patents to decrease the returns to strangling the competetion in the cradle, or possibly subsidizing smaller ventures to make being bought out a less attractive endpoint.

Neither of which are going to do much to help cut the consumer prices of drugs, BTW.

Mr. D said...

Neither of which are going to do much to help cut the consumer prices of drugs, BTW.

Understood. But if we get a lifesaving drug out of the process that we wouldn't have had otherwise, we'll worry about the price later on.

The industry consolidation issue is very real, from what I've heard. The questions concerning barrier to entry are always looming. And it's why conservatives aren't necessarily fans of big business -- they often use their might and the complicity of the regulatory apparatus to stifle competition. But that's a different post.

Bike Bubba said...

Regarding the problem of buying competitors out, even the libertarian in me notes that with the inherent patents awarded for successful drugs, each buyout of a startup represents a prima facie violation of antitrust law. It is for one provider of a unique commodity to buy out its theoretically only competitor.

And the difficulty of the process? It may be too easy. Remember the Stanford study a few years that found that a hot new antidepressant called "sugar pills" outperformed Zoloft? There are quite a few indications that some statisticians outside the FDA/industry complex need to take a look at the process. You could simplify in some areas, and make it more complex in others.

(like the study that found that of 53 ground-breaking studies in cancer research, they couldn't replicate 47.....somebody needs to take a close look at what they're doing!)

Night Writer said...

The regulatory process is a lot like the super-bugs themselves. You create something you think will work and then someone figures a way to get around it, and what you thought you were preventing becomes even worse.