The cost to develop a new drug and win FDA marketing approval is now pegged at nearly $2.6 billion, according to a new report from the Tufts Center for the Study of Drug Development. The estimate will likely hearten drug makers that argue rising prices reflect rising costs, but consumer advocates reacted as if someone was trying to sell them a bridge.
The figure includes two basic components. One is the average amount of money that is actually spent, which Tufts calls ‘out-of-pocket’ costs and amounts to $1.395 billion. Then there are ‘time costs,’ which is another way of saying opportunity costs, of $1.16 billion. This is the return that investors could be expected to forgo if the money had been invested elsewhere while a drug is being developed.
“Drug development remains a costly undertaking,” says Joseph DiMasi, who is director of economic analysis at the Tufts Center and was the principal investigator for the study, in a statement. To reach its conclusion, Tufts dissected information provided by 10 drug makers on 106 randomly selected drugs that were first tested in human subjects anywhere in the world from 1995 to 2007.
The report is the latest entrant in a long-running debate. The pharmaceutical industry maintains that drug development is increasingly expensive due to rising costs, which helps explain why prices have also been rising in recent years. But consumer groups say the numbers mask irrelevant or overstated expenses and are used unfairly to justify rising prices. And they note that Tufts is funded, in part, by drug makers.
Judging by the reactions following its release, the new report is unlikely to settle anything. In a terse statement, Doctors Without Borders, the advocacy and relief group, decried the $2.6 billion estimate. “If you believe that, you probably also believe the earth is flat,” says Rohit Malpani, director of policy and analysis, in a statement.
“We know from past studies and the experience of non-profit drug developers that a new drug can be developed for just a fraction” of the Tufts estimate, he says. “The cost of developing products is variable, but experience shows that new drugs can be developed for as little as $50 million, up to $186 million, if you take failure into account… These figures are nowhere near what the industry claims is the cost.”
Actually, Tufts say the final tab can run higher. How so? The estimated average of cost of R&D following FDA approval of a drug – which can include studies for testing new indications, dosages and monitoring safety – can add another $312 million on to the total life cycle cost for an approved drug. This would bring the full cost to a whopping $2.87 billion. This estimate is in 2013 dollars, by the way.
The estimate, which includes the cost of unsuccessful projects, is substantially higher than the $802 million that Tufts calculated in its last report published in 2003. Similarly, an estimate published in late 2012 by the Office of Health Economics, a U.K. consulting firm, found that the cost to develop and drug was $1.5 billion. That research was conducted with funding from AstraZeneca.
What accounts for the huge increase? Tufts says the total capitalized cost per approved new compound grew at an 8.5% compound annual rate and out-of-pocket cost per approved new compound grew at a 9.3% annual rate. Tufts also says increases in cash outlays used to conduct clinical development and higher drug failure rates during clinical testing contributed most to the estimated increase in R&D costs.
You can look at the Tufts slide show here for the methodology and various calculations.
One critic argues that the Tufts report is hard to understand for another reason – there is a lack of information.
“The new Tufts study was part of a public relations campaign, designed to get people to accept high drug prices,” says Jamie Love of Knowledge Ecology International, a non-profit group that tracks intellectual property and access to medicines issues. “The study is long on propaganda and short of details… Did all of these drugs really have zero NIH funding for pre-clinical” work?
In a statement, he calls for putting pressure on Tufts to release more data used to calculate its estimate. For instance, he argues that “we don’t know how many patients were in the trials (that were used to calculate the estimate) or how much money was claimed to have been spent per patient in the trials. Since the entire estimate was based upon the costs of the trials, we don’t have any idea of what the sample looked like.”
He maintains that information on sample size is pertinent to understanding the costs because cancer trials, for instance, are typically smaller than trials for other drugs, but as a result “the study will be used to justify high prices for cancer drugs.”
We asked DiMasi for comment about these remarks and the statement from Doctors Without Borders and will update you accordingly.
[UPDATE: DiMasi wrote back to say that the "OHE analysis was based on a small sample (with regard to clinical testing) of projects that were in some phase of clinical development from 1997 to 1999 and followed only to 2002. The data for our study are effectively much more recent than the data used for the OHE report.
"What we are measuring is what private developers actually spent on development. NIH research is complementary. If the NIH does some basic research, or even clinical research, that has findings that developers find interesting and useful, then the cost of that NIH research is just part of the social cost of drug development.
"The social cost is the private cost PLUS what governments and non-profits spend that somehow contributes to the discovery and development of new drugs. If private developers spend less than they otherwise would because of the NIH research, then that is captured in our estimate because we are only measuring what private developers spent."]