Key Drug Innovations Often Don’t See Market Rewards, Academic Says

There is little evidence that more innovative or therapeutically valuable pharmaceutical products are rewarded, or that patents are the best way to do so, Economics Professor Margaret Kyle of the Centre d’économie industrielle of Mines ParisTech says in an upcoming study for the Review of Industrial Organization.

High drug costs have raised competition concerns about potential harm from pharmaceutical patents, Kyle said in a 14 March OxFirst webinar. In addition, patents and data exclusivity continue to be points of contention in trade talks, and the price of new drugs is often blamed for rising healthcare costs. Pharmaceutical companies defend the higher costs on the ground that higher prices or more profits lead to more R&D, she said.

But does “the market” actually reward innovation? In a well-functioning market, one would hope participants have full information about the price and quality of a product; the price and consumption of the product would reflect willingness to pay; and higher prices for better drugs would be expected if patients value their therapeutic value, Kyle said.

Several factors, however, complicate market share and pricing, she said. Among these are the effect of regulation, and the fact that experts such as doctors and pharmacists can choose among drugs. Insurance also clouds the picture because most patients don’t pay full prices for medicines, and doctors choosing drugs don’t pay for them and, therefore, may not be responsive to price, she said.

All this means that consumption might not respond to price, and very few pharmaceutical markets look anything like a free market in the first place, she said. Where patients don’t pay full price, doctors are unaware of the prices, and governments play a key role in health provision, governments tend to negotiate prices. The goal is to give patients access to the drugs while maintaining incentives for companies to innovate but within budgetary constraints, she said. Once R&D costs are sunk, firms should be willing to sell their products at marginal costs, and knowing this, government can set very low prices. Even if in an ideal world prices regulated by governments should target innovation incentives, they are instead likely to reflect political or budgetary pressures, Kyle said.

This situation is taking place around patents and innovation policy, she said. In the absence of patents, drug developers would have a hard time recouping costs. But patents are a one-size-fits-all policy which creates potential distortions because the longer a drug takes to develop, the shorter the patent time runs when the product reaches the market. Another problem with patents is that they tend to be filed early in the development process when there’s little information about their therapeutic value.

Kyle’s study looks at whether “better” drugs are rewarded with more patents, higher revenues, or fewer restrictions on quantities. It uses the indicative measure of therapeutic value, which offers “food for thought” but isn’t the final word on the issue, she said.

The French Haute Authorité de Santé (HAS) assesses the therapeutic value of drugs using two scenarios, Kyle said. One is the service médical rendu (SMR), the other amélioration du service medical rendu (ASMR). Kyle’s study of 258 new drugs available from 2002 focuses on ASMRs, which range from a value of 1 (major) to V (non-existent) relative therapeutic value.

The study considered patents, relative prices, market share and revenues for the subject drugs in the US, UK, Germany and France 10 years after a particular drug was launched. The results showed little evidence that the more innovative or therapeutically valuable products are rewarded with more patents, or that drug prices or revenues strongly correlate with therapeutic value, she said. Market share, however, was positively associated with therapeutic value across and within countries, which could mean that patients are finding the best products.

Several factors could affect these results, Kyle said. One is whether the ASMR accurately reflects therapeutic value, another that the study doesn’t estimate the welfare to society of particular drugs. It could be that a small incremental value in a medicine for treating a major disease is worth more than a pharmaceutical breakthrough in a lesser disease, she said.

Could Data Exclusivity be a Solution?

Data exclusivity could be an alternative policy to patent protection but it, too, has limitations, Kyle said. Like patents, it is an incentive mechanism that allows an innovator some time on the market without the threat of competition from generic drugs, she emailed later. In practice, patent protection and data exclusivity terms overlap, but they differ from each other in several ways.

Unlike patents, which must be at least 20 years from application data under the World Trade Organization Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS), data exclusivity terms vary in length across different countries. In Europe, the term of exclusivity has been harmonised to the current 8+2(+1) term, which is eight years of exclusivity plus two years of additional market exclusivity, during which generic applications can be filed but not marketed, plus an additional year of an important new use is developed.

Another difference is that the data exclusivity clock starts when a product hits the market rather than when development begins, Kyle said. This is important because of the potential distortion in R&D that results from fixed patent terms. If clinical trials take a long time to complete, which is often the case for chronic diseases, innovators expect fewer years of patent life remaining once they get to market, which makes them less attractive. The data exclusivity term doesn’t penalise long development periods.

Another difference is that some chemicals or substances can’t be patented because their existence is well-known (meaning there is prior art) or other reasons, while data exclusivity provides development incentives even in the absence of a patent.

Drugs generally have multiple patents, Kyle noted. The primary or product patent covering the molecule itself is legally very strong and hard to invent around. Patent disputes usually erupt around the secondary patents filed later on manufacturing, methods of use, and so forth. These are legally weaker and are invalidated at higher rates and invented around, she said.

Such disputes are often settled by pay-for-delay or early entry agreements which competition authorities are wary of. In any case there is legal uncertainty for generics when multiple patents cover a drug. Data exclusivity is “cleaner” because there is an easily established expiration date, she said.

Is There Political Will for Data Exclusivity?

EU data exclusivity policy is “already generous by global standards” and extending it could be perceived as a “gift to the industry,” Kyle said. In addition, having only recently harmonised the data exclusivity term, “I doubt there’s much interest in debating it again.” She believes, however, that “it might make sense to tie data exclusivity to therapeutic value.”

Instead of giving all new drugs 10 years, governments could consider giving five years to those that don’t add much clinical benefit and 12 years to those that represent an enormous improvement, Kyle said. That’s difficult for governments to do because it means a longer period of high prices for the most important drugs, although that could be offset by lower spending on less effective products, she said.

There is also “valid scepticism” that data exclusivity is being pushed by the pharma industry as one more way to extend monopolies, Kyle said.

“While I would substitute data exclusivity for secondary patents, the industry wants both types of protection. So that also makes the idea politically and practically more challenging,” she said, because it’s difficult to ban secondary patents because changes in the standards of patentability would affect other technologies as well.