Crisis In Pharma R&D: It Costs $2.6 Billion To Develop A New Medicine; 2.5 Times More Than In 2003Commentary by John R. Graham
December 01, 2014
The rate of growth of health spending remains moderate. One area where prices appear to be increasing faster than they have in the past few years is brand-name prescription drugs. By 2012, blockbusters had lost their patents and many looked forward to a future where we could all get a month-long supply of generic drugs for $4. Well, it did not quite work out that way for everyone.
Specialized drugs for smaller patient populations were introduced with high nominal prices. In September, EvaluatePharma confirmed that the increasing cost of prescription drugs was concentrated in more specialized drugs. Of the top 100 selling drugs in the U.S.:
The median revenue per patient of the Top 100 drugs has increased from $1,260 in 2010 to $9,400 in 2014, representing an seven-fold increase;
The median patient population size served by a Top 100 drug in 2014 is 146,000, down from 690,000 in 2010; and
There are now seven treatments priced in excess of $100,000 per patient per year in 2014, versus four in 2010.
Given these facts, it may be understandable that the health-insurance industry is campaigning against the high prices of specialty drugs. For its part, the brand-name pharmaceutical industry emphasizes that health insurers (especially in Obamacare exchanges) often put these specialty drugs on the most expensive tier of their formularies. This requires patients to pay high out of pocket costs. While this is an accurate description of the situation, a government policy simply forcing insurers to cover a higher share of the price of a specialty drug does not reduce the price. It just moves it from patients’ direct payment to premium.
Reducing prices of specialty drugs requires improving the productivity of R&D. On that front, the news is sobering. Last December, Deloitte and Thomson Reuters examined newly introduced drugs from the twelve pharmaceutical companies with the largest research and development (R&D) budgets. It cost $1.3 billion to bring a newly discovered compound to market. However, the average forecast for peak sales of an asset declined by 43%, dropping from $816 million in 2010 to $466 million in 2013.
The high nominal prices of new drugs do not compensate for the smaller patient populations that they target. Deloitte and Thompson Reuters estimate that the IRR (internal rate of return) of R&D spending has dropped in half since 2010, from 10.5 percent to 4.8 percent. Sales of new drugs are not overcoming the loss of patents, weak pricing power for older drugs, or reduced productivity of R&D.
The crisis of R&D is highlighted in a new report by the Tufts Center for the Study of Drug Development. Back in 2003, the Tufts team estimated that the cost to research and develop a new drug was $802 million (in 2000 dollars). In 2013 dollars, that would be $1,044 million.
This month, the team updated its estimate, using drugs first tested on humans from 1995 through 2007: It now costs $2,558 million to develop a new medicine, almost two and a half times more than the (inflation-adjusted) 2003 estimate. It appears that the Tufts group’s estimate is much larger than the one from Deloitte and Thomson Reuters because the Tufts group looks at costs from the first step of research, before discovery.
This means that the cost of abandoned projects is allocated to successful ones. And an astonishing 8 in 10 were abandoned. Because some of the drugs in the Tufts study are still under development, even more will be abandoned. My Forbes colleague Bruce Booth confirms that an 8 percent success rate is the consensus of other estimates.
Why such little success? The authors note that: “Clinical approval success rates have declined significantly” since their earlier study.
For those who advocate that the research-based pharmaceutical industry be subjected to government audit and regulation of its R&D budgets for each new compound, the Tufts report corroborates my own conclusion that this would be an impossible task: “The drug discovery and development process typically involves high fixed costs, meaning that substantial expenditures incurred prior to clinical testing cannot be directly linked to work on specific compounds.”
The Tufts report also estimates an average real cost of capital of 10.5 percent over the period. In 2010, it was 9.4 percent. I estimate that the real interest rate at that time (as measured by U.S. Treasury Inflation-Protected Securities) was about 2.8 percent. This implies a real risk premium of about 6.6 percent. If the nominal IRR (as estimated by Deloitte and Thomson Reuters) is only 4.8 percent, the pharmaceutical industry as a whole is clearly not achieving its hurdle rate.
The Tufts report is controversial, which I will address in a future column. Nevertheless, the body of evidence indicates a productivity crisis in pharmaceutical R&D that demands attention. In the U.S. Congress, Rep. Fred Upton, Chair of the House Energy & Commerce Committee, has launched a bipartisan initiative for “21st Century Cures” that is examining the entire regulatory process governing the research enterprise. It is a necessary and welcome step in the right direction.