Drug development costs increase due to internal, not external, processes

What I would have written as a comment to this wonderful  article at Xconomy by  an old friend … If they allowed comments.


Great article, Stewart. Since I can’t comment on Xconomy, I thought I’d send you these two references to clarify one of your statements regarding the Kefauver-Harris Amendment: “One can readily imagine that this legislation must have driven up the cost of drug development and marketing significantly.” 

The data suggest that this had no effect at all on the cost of drug development.   

And neither did the biotech revolution.The rise seems to be an inherent aspect of the Big Pharma business model that appears to be immune to external factors.  

Current Big Pharma R&D practices can no longer sustain their development costs with a $150 million a year drug. So they charge 10 times more.  Because they have to.

Munos (Lessons from 60 years of pharmaceutical innovation) showed that the cost to develop a new drug has been rising at a stable exponential rate since the mid-50s.

And Scanellis et al. (Diagnosing the decline in pharmaceutical R&D efficiency) expands on this. They even come up with Eroom’s Law (Moore Backwards) to describe this curve. Both papers report that the cost to develop an NME (new molecular entity) has risen exponentially (in inflation adjusted terms), doubling about every 9-10 years since 1955. Here are relevant figures from the papers:  

First from Munos, this looks at the cost to develop an NME in inflation adjusted dollars (note the log scale). There was no big shift in the costs for new drugs between the 50s and 70s reflecting regulatory pressures. And there was also no shift between the mid 60s and mid-00s, reflecting that the biotech revolution did not produce a fundamental shift in development costs.  

Drug costs 001 It just allowed the companies to access new treatments.  It allowed Big Pharma to sustain their business models for a few more decades.

And then this more detailed figure from Scanelli looking at how many new NMEs could be developed for $1 billion (mind the exponential scale)   Drug costs 002This makes it more explicit that the exponential increase is not affected by external forces, such as regulation or novel technologies. Scanelli et al. tried to identify these internal forces, ranging from just throwing money at the problem to the “better than the Beatles” problem (it is easier to produce a ‘me too’ drug than an innovative NME).   

The current business model of Big Pharma can really only produce drugs that pay back the roughly $1.5 billion R&D costs. So, a drug that has a $150 million a year market simply is no longer feasible to bring to market. And, if it is, Big Pharma must charge 10 times more to recoup their costs.  

Thus why we see such huge increases in drug costs.

Obviously, exponential increases like this cannot be sustained forever. Hope springs eternal. Perhaps new genomic approaches will revolutionize oncology.  

This is unlikely. From the data, it appears that Big Pharmas are impervious to new technology or new regulations. They HAVE to charge huge amounts no matter what. Their business model requires it.  

Thus why I expect them to fail and be replaced by new companies that can sustain themselves with a $150 million  a year drug. Or even a $15 million one.  

I think that “keeping healthy people healthy” will be where medicine goes. The old model of “making sick people healthy”  is simply too expensive and unsustainable.

Image: E-Magine art