This is one of the prickliest topics in the health care industry, and it has been as long as I can remember. One of the publications I worked with was called Product Management Today. Long defunct, this periodical followed the pharmaceutical industry quite closely. As a result, we were familiar with Tufts University’s Center for the Study of Drug Development. This was the first organization to publish what was considered credible figures as to what it cost to bring a drug to market.
In the early 1990s, that number was published in the Journal of Medical Economics to be about $231 million. Remember, these were generally small molecules, and the drug industry consisted of dozens of big manufacturers identifying their molecular targets, characterizing them, and cultivating them through the clinical trial process. In 2003, this number was revised upward to somewhere north of $800 million. In 2016, the Tufts team recalculated the numbers, and it found essentially a 10-fold difference (not accounting for the inflationary change in the value of the dollar between 1991 and 2016). They indicated that it costs $2.7 billion to bring a product to the market today, which also considers the cost of failed drug development. In other words, they added the opportunity cost to a pharma company for a drug that was withdrawn or failed clinical trials (any stage). In their 2016 analysis, they took as their drug sample, 106 investigational drugs, 87 being small-molecule compounds and the remaining 19 biologics. I’m guessing that if biologics represented more than 18% of the sample, the total average estimate could have been even greater.
The cost to produce a biosimilar is considerably less. This is partly the result of the lesser clinical trial requirements compared with new chemical entities. Although this figure has not been convincingly calculated, I’ve heard it cited to be anywhere up to $250 million. This seems to be reasonable, based on the development requirements. Unfortunately, several biosimilar manufacturers have to wait to market their agents because of patent litigation, and that in itself represents a cost.
Today, the results of another study was released, which offers a very different number. Published in JAMA Internal Medicine, authors from Oregon Health and Sciences University and Memorial Sloan Kettering Medical Center found that the median cost to produce 10 cancer medications was really $648 million (range, $204 million to $2.602 billion)—add another $110 million or so for opportunity costs. This included 5 drugs that received fast-track approval. Not only did the researchers use US Securities and Exchange filings that cited manufacturer-reported costs, but they limited the analysis to only manufacturers without a previous approved drug on the market. The drugs evaluated included several monoclonal antibodies, so it was reflective of complex molecule development.
I’d like to point out another question that perhaps skews the calculation. Rarely these days does “big pharma” get involved with new drug identification and characterization. True, they are often involved in the expensive clinical trial phase, but do we read in the paper weekly that a drug discovery company has licensed or sold the product to a big pharma entity (or even sold the company itself)? And what is the guarantee that they are not overpaying for the price of the drug or the company?
The upshot of this is that the 10 companies evaluated in the more recent study had cumulative revenues resulting from their new agents of $67 billion from the time of approval (until December 2016 or the time it sold or licensed the product to another company). The range per product was ranged from $204.1 million to $22.3 billion. It sounds overall, that they were good buys for the big pharma companies but not necessarily for the health care system.
In other words, whether it cost $648 million, $800 million, or $2.7 billion, the research and development costs for these new agents are made back in a year or less. It is hard for us to listen seriously to pharma companies who use the cost to develop the agents, or their having to eat the cost of failed agents, as credible justification for the prices being charged.