The case for licensing biopharma patents after loss of exclusivity

Most pharmaceutical products depend on a patent that covers the active pharmaceutical ingredient to ensure exclusivity in the market. Often other patents also cover aspects of the drug product discovered during the development process – these are known as ‘development patents’ or sometimes ‘secondary patents’.

In the current environment, innovators tend to enforce development patents in order to preserve full market exclusivity. Alternatively, they do not enforce them at all, thus negating the investment in obtaining them in the first place. With no way to determine what course an innovator will take, generic companies often seek to design around or invalidate every development patent in an effort to anticipate efforts to use them to maintain exclusivity for the product. Invalidation is necessary if the generic company must or would like to use the patented technology. The result is significant investment in litigation preparation, designing around patents and litigation itself – deployment of resources that could be better used to develop and deploy drugs.

Does it have to be this way? Certainly, the millions spent on litigation could be better used to develop more products, more quickly, for the benefit of patients and the healthcare system – and without destroying reasonable profits.

Licensing to manage loss of exclusivity

Most pharmaceuticals have one or more patents that cover aspects of the product. The strongest patent – often referred to as a ‘compound patent’ – covers the active pharmaceutical ingredient itself. Compound patents are rarely found invalid and infringement is never in question, as a generic manufacturer needs to produce a product that uses exactly the same active pharmaceutical ingredient as the original.

Other patents can cover discoveries made during the development process (eg, methods of disease treatment, methods of monitoring the risk of harmful side effects, the crystal structure of the drug compound (polymorphic variants) and salts of the compound, as well as methods for the formulation, administration regimen or manufacturing).

These development patents are more vulnerable to invalidation because they are typically filed after the compound patent is disclosed. Indeed, the compound patent can expressly disclose foreseeable methods of using the drug, ways to make and formulate it and all methods of administration. Further, a competitor can avoid infringing patents on crystal structures, salts and formulations by making a product that has a different crystal structure, salt form or formulation provided that the drug product remains therapeutically equivalent. This is sometimes referred to as a bioequivalent.

Figure 1. Pfizer Lipitor – a decade in review

Source: Seeking Alpha, ‘Pfizer: Lipitor Vanishing Revenue’ (November 6 2012)

The current experience of the entry of generic products into the market is a rapid loss in revenue and market share for the originator company. Figure 1 shows a graph reflecting the value of another blockbuster product, Lipitor (atorvastatin), during the period of exclusivity and after, in global and US markets. The year that atorvastatin became generic, Pfizer’s US sales revenue on the product fell by nearly 80%. This reflects Pfizer’s reduction in market share and any price cuts that it made. Pfizer may also have paid the higher co-pays that insurance companies imposed on patients who stayed on the branded drug, further affecting its bottom line with regard to Lipitor sales.

Original manufacturers lose not only market share but also pricing power on generic entry. The average price affects the entire system. If a generic takes market share but is able to maintain a relatively high price, the originator will feel the impact in lost market share, even though almost as much is still spent on the drug.

The Hatch-Waxman Act is unique in providing a tremendous incentive to generic companies to attack originator patents by giving six months of generic exclusivity to the first company or companies to successfully challenge innovator patents. One, two or even three such first filers typically enter the market with generic drugs priced at between 50% and 90% of the originator price – and quickly receive more than 50% of the market share. Once many generic companies enter the market, prices fall. Figure 2 shows the impact of generic entry on price (in the United States) depending on the number of generics entering the market.

Figure 2. Relative price by number of manufacturers in first 24 months of generic competition (seperately by entry profile)

Source: Olson and Wendling, ‘The Effect of Generic Drug Competition on Generic Drug Prices During the Hatch-Waxman 180-Day Exclusivity Period’ (April 2013)

Licensing development (secondary) patents could help to offset some of this loss by ensuring that the originator company receives revenue on the sale of all or most units of a generic product.

Such licensing could also benefit generic companies. For one thing, they could avoid the expense of designing around a feature (eg, a salt form or formulation) that is optimised for the product. Developing alternatives involves expense and an element of uncertainty that the alternative form or formulation would retain bioavailability, stability or other features that affect the original product’s success.

Moreover, generic companies could advertise that they use patented features of the original product to ensure that the patient receives exactly the same drug product in generic form (which would likely also involve licensing the originiator’s trademark rights). Patients are often concerned that a generic drug is not the same as the original – often if they are financially able they will pay a premium to use the original product. While this market will likely always exist, more financially sensitive patients may feel greater confidence in a generic drug knowing that it not only contains the same active ingredient but is identical in all respects to the original drug.

The licensing of safety-related inventions deserves some attention because it raises important policy considerations. If such patents substantially delay the market entry of generic products, the healthcare system will face unnecessary expenses for old drugs that could otherwise be devoted to new medicines. Health authorities will almost certainly require generic manufacturers to include these features in the generic product – so unlike a formulation or salt, the generic cannot avoid infringement.

Sometimes the safety information may refer to a second indication for the product. If the generic company does not have that indication, it could ostensibly omit the safety feature. However, that could place patients at risk if they receive the generic version of the drug for the other indication without the safety feature. At the end of the day, neither the originator company nor the generic company wants to put patients at risk. Licensing this type of invention ensures that patients have the full opportunity to use a generic drug product safely, while recognising the contribution that the originator company has made to developing that invention.

Under some circumstances, the expense of designing around a development patent may be desirable. Not every original crystal form, salt form or formulation is optimal. In addition, a generic company may want to develop a superior form of the product, especially a company that has a reputation for value-added generic products. Generic companies may also wish to differentiate their generic product from other generics and patent those differences. So a generic company may embark on a development programme that replicates the original product as a generic and adds other features that differentiate its product (without of course affecting its status as a fully substitutable generic).

One cannot expect an originator to license the basic compound patent to a drug product (except for a period of days or weeks close to loss of exclusivity). The exclusivity value of compound patents allows high prices, which justify the expense of R&D. In the pharmaceutical industry this involves substantial attrition of products and substantial time, as shown in Figure 3.

Figure 3. Drug development path from discovery to commercialisation

By some estimates, this process can cost $2.6 billion. Premature licensing of the compound patent to generic companies would have a severe detrimental impact on the company’s ability to justify this kind of high-risk investment. The originator must defend these rights rigorously. However, that does not mean that the originator company should not further monetise development inventions that are not substantial enough to confer full market exclusivity.

There are two ways that an originator can license its development patents: individually or in a bundle. Individual licences may make sense where a generic company only needs a sub-set of the patent rights (eg, a safety feature but not a formulation). A bundled licence is of course much simpler and puts all the generic companies on the same footing with regard to each other. Various factors would affect which type of licensing programme an originator would undertake: if there is one critical patent that all companies must license, it might choose to bundle all the patents together. On the other hand, if one generic decides that its proprietary formulation gives it a competitive advantage, and another that copying the originator product exactly gives it an advantage, only the latter would need the formulation licence.

From the generic company’s perspective, it might want to leverage the brand identity when it chooses to license optional technology, thus necessitating a licence to overtly mention the trademark and originator company. This way, it could advertise that its product is as close to the originator as possible in all respects (including shape and colour, which may be features protected by trademark or design patent), hopefully resulting in greater market share to offset the licence cost.

Generic drug companies operate under tight margins. Adding licence royalties to the cost of goods would either reduce their margins or force them to charge higher prices. Thus, there is still an incentive for generic companies to avoid taking licences. Clearly one result is that to make licences to development patents attractive, the royalty rates would need to be fairly low (ie, in single digits). Also, the generic company would need to realise some other benefit, such as faster development, ability to freely reference the trademark name of the originator product and potentially earlier market entry – even by a few days or weeks – since early market entry confers market share advantage to a generic player.

It is arguable that driving generic prices to the floor is ultimately counterproductive to the healthcare system’s interest in ready availability and maintaining price stability for therapeutic products. Generic companies with manufacturing operations in countries with high labour costs may not be able to compete solely on price.

The result is decreased competition, with only a couple or even just one company in the market. This exposes the system to the potential unavailability of product as a result of manufacturing quality problems or natural disasters. It also exposes the system to monopolistic pricing, with a sole generic on the market able to increase prices multi-fold, as we saw with Valeant Pharmaceuticals and Turing Pharmaceuticals. Licensing patented aspects of the original product would give high-quality generic players a basis for charging slightly higher prices and promising truly identical (versus therapeutically equivalent) generic products.

So-called ‘authorised generics’, where a generic company sells the originator’s product as a generic product, is one example of a licence being granted by an originator to a generic. In this case, the authorised generic company’s sales are under the originator’s marketing authorisation. Authorised generics are usually launched to compete with a first-filer generic during the six-month period of exclusivity, when only the originator and the first filer have marketing authorisations for the product. The company selling the authorised generic typically pays a royalty on those sales to the originator.

A model for licensing a hypothetical small molecule

Assume that compound patent protection for the small molecule Exemplar (exemplinib sodium) will expire on January 1 2012 and that all regulatory exclusivities have already expired. For the purposes of this analysis, assume that the product is one of four molecules in its class and that it has 30% market share for the class. We can model it out two ways: first, that Exemplar is the first in the class to go generic. Second, that it is the last in the class to go generic, in which case we would predict the market share to fall to 25% because some patients will switch or start on one of the other generic products instead of Exemplar.

Assume that the Orange Book lists three patents covering Exemplar in addition to the compound patent:

• one covering a formulation (expiration January 1 2020) that a generic product might avoid, but not without litigation;
• one covering the dosing regimen (expiration January 1 2020); and
• one covering a risk management feature in the label that involves testing liver enzymes one week after the first dose of the product (expiration January 1 2021).

Finally, assume that one generic company has first-filer status and that six companies have tentative approval to enter the market on the expiration of the compound patent. The originator offers licences to all interested licensees on the following terms: bundled license of all three patents at a rate of 5% royalty on net sales until January 1 2020, including the right to claim the identical patented formulation to Exemplar in advertising and packaging material (conditioned on proof that in fact the generic product does use the same formulation).

To calculate potential royalties after Exemplar goes generic, we assume that all generic companies license the three patents for 5% royalty of sales. Assume that all sales losses are to generic products. Estimate annual generic sales by gross sales of originator before generic entry times percent price reduction as follows. First, for the first-filer generic for January 1 until July 1 2012 (80% of innovator price and 50% of market for Exemplar). Then for a total of five generic companies with 80% of the Exemplar market and a price 40% of the originator price from July 1 until December 31 2012, and thereafter five generic companies with increased overall market share to 30% of the class (with Exemplar retaining 5% of the class and 14% of the exemplinib sodium market in 2013, dropping to 10% of this market after 2014) and a generic price of 20% of the originator price.

Based on this analysis (and without including the net present value discounting), in the period following loss of exclusivity until expiration of the licensing deal, the originator company makes $2.1 billion in sales of Exemplar (the branded version of exemplinib sodium). The royalties only add about 5.6% to the top line, although assuming a 60% margin on sales for Exemplar-post loss of exclusivity, when the originator company stops all promotional activity, that 5.6% on the top line comes to a 9.4% increase to the bottom line. These numbers are not particularly exciting until one takes a few factors into account.

The total estimated royalties in this model are roughly 7.5% of the top-line sales for Exemplar; in other words, it is the value of the Exemplar sold in about four weeks, without discounting for net present value. Even assuming a 60% margin, that translates into about 6.5 weeks of sales to exceed the value of the royalty. Clearly, if an originator company can fairly see a way to maintain its exclusivity for a reasonable period past the expected date, it would be financially better off than if it had taken the royalty deal.

The problem is that, especially in the United States, once exclusivity is gone (through expiration of regulatory exclusivities, including waiting for patents to expire as per Paragraph III of Hatch-Waxman, automatic stays on full approval for 30 months in Paragraph IV litigation, and regulatory exclusivity), it is close to impossible to legally prevent a generic product from coming into the market, even if it risks infringing a development patent. So assuming that the generic product is coming, what are the factors favouring a licensing programme?

First, the chart assumes that the originator makes no price reductions and keeps the Exemplar price at the 2011 level. In all likelihood, the company will make some cuts in order to keep market share – if it did not, market share would be lower. These factors could reduce sales of Exemplar by 20%, making the royalty income closer to 7% of top-line sales and 12% of bottom line.

Second, US drug prices tend to be between 50% and 100% greater than in other developed markets such as Canada, Western Europe and Japan. It is likely that at some point free pricing may become unsustainable. Generic drug prices already go more or less to rock bottom (ie, cost plus some profit margin) and a decrease in the originator price will not significantly affect the generic price.

So, assuming that Exemplar started from a lower base – say 50% of the sales that it had in the model – the royalty impact on the top line and the bottom line would be double. Further, were the originator to license similar patents in markets where such price controls were in effect, the licence to generic companies would have a similarly significant impact on company performance.

Third, it is worth noting that even the modest increase in profits from a licence to generic companies is better than the alternative, which is no royalty at all. It is also more attractive than spending millions of dollars on a potentially futile attempt to maintain exclusivity, particularly because in the United States at least, generic entry will likely occur even if the innovator litigates the development patents through the normal routes (ie, outside the legislative strictures of Hatch-Waxman litigation), where preliminary injunctions are rare, even assuming the patent survives an inter partes review or declaratory judgment actions.

What are the prerequisites for success? As with any running royalty licence agreement, there needs to be a reliable method to monitor product sales. A corollary of this is that the licensor must have systems in place to process royalty payments and to ensure that those payments are made. The originator must also ensure that it has good controls in place to maintain the enforceability of its patents (eg, payment of annuities or maintenance fees on time and the like). At the end of the day, the originator patentee must be willing to litigate breaches of the licence agreement or refusals to license when this is required effectively and efficiently.

Licensing post-loss of exclusivity patents is worthwhile

Licensing development patents that cover pharmaceutical products could provide additional revenue to a company after loss of exclusivity. While the totals would be a fraction of the revenue from sales of the product during its period of exclusivity, royalties from such patent licences could have a measurable impact on top-line and bottom-line performance. That impact is likely to be greater in the United States as pricing pressure limits the ability to charge as much for novel, exclusive products, as companies have enjoyed in the past. In countries outside the United States, where price controls have more impact, licensing can already have a bigger effect post-loss of exclusivity.

This model does face some additional problems. Competition agencies are deeply concerned about collusion in the pharmaceutical sector and any agreement that gives the originator/patentee insight into sales by a generic competitor – which a patent licence predicated on a running royalty necessarily would do – might face scrutiny. There are two ways to manage this. First, the parties could agree to a fully paid-up licence based on estimated sales discounted. This creates its own problems, most significantly the risk for both parties that actual sales will not track the estimates. A second solution would involve a third-party licence administrator, or even purchaser, who would manage the collection of royalties independently and thus avoid the potential for sharing sensitive sales information.

In addition, for the purposes of our model we assumed that all the generic companies entering the market needed at least licences to the patents that expired with the royalty term. In reality, this might not be the case. That creates the need to license some but not all of the patents. The pharmaceutical industry could look to the tech sector to see how that is managed, but again this might prompt competition and antitrust concerns.

There might also be some resistance from the generic industry to an added cost pressure on already thin margins. As discussed above, increasing generic margins to sustainable levels is probably good policy – but that means asking payers to pay more in the short term, which is an unlikely scenario. Nevertheless, the benefits could outweigh these problems. One notable benefit could be greater willingness from the originator to peacefully allow a product to go generic, without engaging in costly litigation that could, even if just in the short term, delay generic entry. That royalty bird in the hand may in fact be more attractive than a risk-adjusted litigation scenario to extend exclusivity by a few months. Greater predictability and system stability also have value, even to payers, as does the ability to ensure a smooth transition to generic products that patients may be even happier to use.

There is one further issue and that is the desire of an individual generic company to enter the market earlier than other generic companies (ie, the competitive pressure between and among generics). This model could provide generics with some competitive leverage. The ability to claim to be fully identical to an originator product (short of being an authorised generic, which is the originator product repackaged under a generic label) could provide an advantage with regard to other generics.

The originator/patentee could reward a willingness to license early with early entry by a few weeks – perhaps even charging a higher royalty rate to offset the profit loss from the subsequent fall in branded sales – so that even if the generic company were to make little or no profit on those early, exclusive sales, it would benefit by establishing itself in the market ahead of its competitors.

So far, licences for development patents outside of litigation remain infrequent and the polarisation of originator and generic companies reflects the extremes of patents, which protect exclusivity or do not, with nothing in between. Greater willingness to license development patents – following the lead of the tech industry – could provide a new source of revenue for biopharmaceutical companies that are otherwise facing price headwinds, provide competitive advantages to generic companies willing to license these patents and increase predictability and stability in the generic drug market generally. Moreover, a robust licensing ecosystem in the generic drug space could engender a new sub-industry: the generic drug royalty purchasing company.

Such a company would allow originator patentees to monetise assets without raising antitrust concerns. They would also be able to provide this service efficiently enough to make it profitable; a biopharmaceutical company is ill-equipped to manage a complex licensing operation with potentially hundreds of licensees, each paying small amounts of money.

Finally, in a world where there is deep suspicion of incremental pharmaceutical patents, which translates into suspicion of the patent system in general, any step that leads to greater respect for patents and the innovation that they reflect would be a good thing.