There has been a new interest in expanding the transparency of pharmaceutical markets, in many areas, including information about the outlays on R&D for the development of specific drugs. The debate over transparency legislation, still pending but stalled in California (AB 463), raised public attention to this issue worldwide and has stimulated additional interest in transparency measures in various states (notably in Colorado) and among members of Congress and in foreign countries.
I have been engaged in research on the costs of drug development for some time, and am a longtime proponent of more transparency of the cost data on drug development. I don’t see how you can make intelligent decisions about drug pricing, incentive mechanisms, or funding of R&D if you don’t know more about the costs of doing the things that are required to bring drugs to market. The industry consultants who present data on these issues are typically paid to confuse and not enlighten people. Government policies requiring more disclosure are badly needed. A poor family applying for food stamps has to provide more information about their budgets than a pharmaceutical company looking for a multibillion payout from a government-granted legal monopoly.
When I look at the several bills introduced or proposed at the State or Federal level, I am encouraged, but also critical. My primary complaint is that these bills only require companies to provide highly aggregated data on R&D outlays, and that this does not get us where we need to be. The most important data to have in detail are the outlays on clinical trials. Lots of people can identify potential drugs, but testing those drugs in humans costs millions of dollars, and involves risks. R&D outlays on clinical trials are, in some cases, nearly all of the R&D outlays for a drug manufacturer, and they are almost always the most important part.
Clinical trials are typically placed into categories which reflect the sequential stages of development. For new drugs, the FDA will examine evidence from Phase 1, 2 and 3 trials, where each phase normally involves more patients and higher costs. The odds of success in a Phase 1 trial is much different than a Phase 2 or Phase 3 trial.
To illustrate, consider the odds of success at each stage of development, using data on the phase risks calculated for BIO — the biotechnology industry trade association — for the approval of the lead indication of an oncology product, published in Nature Biotechnology in 2014. The odds of the drug making it all the way to FDA approval was estimated at 8.7 percent at Phase 1, 13.8 percent at Phase 2, and 50.6 percent at Phase 3. Thus, when calculating the risk-adjusted costs of the trials, each dollar invested in Phase 1 trials would be counted as $11.49 on a risk-adjusted basis. For Phase 2 trials, a dollar investment adjusted for risk would count as $7.25. For Phase 3, where risks are much lower, only $1.98.
The timing of trials (and investments) is also important. All other things being equal, a drug that takes seven years of trials to bring to market is going to be more expensive than one that takes three years (once the costs of capital are considered, if that is part of the analysis).
It is also useful to have both the outlays on a specific trial, and the number of patients in that trial. With those two figures, one can calculate the per patient costs of the trials, and that can be used to check the validity and reasonableness of the self-reported figures, and also to model the costs for other drug development projects.
The subsidies that are associated with trials are often very trial specific. The NIH, the VA, the Army or another government agency may subsidize some trials in whole or in part, but not others. The Orphan Drug Tax credit, which provides a subsidy of 50 percent for qualifying trials, and which was available to 47 percent of all new drug approvals in 2015, will apply to some trials, but not others.
Another type of subsidy for trials that is less frequently reported are the obligations in the Affordable Care Act (ACA) to have medicare and insurance companies pay some of the costs of trials, under section 2709. It is easier to estimate the role of that subsidy when the costs are broken down by the trial, so you can take into consideration the trial designs, to see which trials qualified in whole or in part for the insurance reimbursements for care.
While we have typically focused on the expenses associated with FDA approval of the lead indication on a new drug, the follow-on R&D is also important, and having data at the trial level of detail allows better analysis of the incentives needed to induce R&D to test new uses of an older drug, or incentives to test the drug on different populations, including the pediatric population (where the concepts of risks are completely different, because the studies are used to extend drug monopolies in the U.S., regardless of the results).
The short message here is that the details are important, and having data available at the trial level is necessary to conduct the most useful and serious analysis of drug development costs.