Deloitte White Paper Addresses Unintended Consequences of Follow-on Biologic Regulatory Pathway

    By Donald Zuhn --

Last spring, Patent Docs discussed pending follow-on biologics legislation with California Healthcare Institute (CHI) president and CEO Dr. David Gollaher (see "CHI CEO on Patent Reform and Follow-on Biologics Legislation").  During the discussion, Dr. Gollaher mentioned that Deloitte Consulting LLP (a subsidiary of CHI member Deloitte LLP) had released its own white paper on follow-on biologics entitled "Avoiding No Man's Land:  Potential Unintended Consequences of Follow-on Biologics."

The Deloitte paper is a recent addition to a growing collection of publications that address the potential impact of implementing a follow-on biologics regulatory pathway in the United States.  To date, Patent Docs has reported on papers by AEI Resident Scholar Dr. John Calfee (see "AEI Presents Follow-on Biologic 'Reality'" and "AEI Believes Advantages of Longer Data Exclusivity Period Outweigh Disadvantages"), economist Dr. Robert Shapiro (see "Dr. Robert Shapiro Discusses Follow-on Biologics Report"), Duke Professor Dr. Henry Grabowski (see "Professor Grabowski's Economic Analysis of Data Exclusivity for Follow-on Biologic Drugs"), the Congressional Budget Office (CBO) (see "CBO Releases Report on Senate Follow-on Biologics Bill; BIO Calls for Congress to Pass Biologics Bill in 2008"), Boston University Professor Dr. Laurence Kotlikoff (see "BU Economics Professor Releases Report on the Impact of Marketing Exclusivity on Biologics Innovation"), Matrix Global Advisors principal (and AEI Research Fellow) Alex Brill (see "Former House Ways and Means Economist Claims 7-Year Data Exclusivity Period Is Sufficient"), and the Federal Trade Commission (FTC) (see "No One Seems Happy with Follow-on Biologics According to the FTC").

In discussing the potential impact of a follow-on biologics (FOB) regulatory pathway on the biotech industry, the Deloitte paper carefully outlines the impact of the Hatch-Waxman Act on the small molecule pharmaceutical industry, and then analyzes the differences between the small molecule pharma and biotech industries in order to predict how an FOB regulatory pathway might affect the biotech industry.  The paper, authored by Jim Hollingshead and Rob Jacoby, notes that the Hatch-Waxman Act "fundamentally altered the 1983 pharmaceutical business system by changing . . . clinical trial requirements and market economics."  The Act changed clinical trial requirements by creating an abbreviated regulatory path for generic drugs, allowing a generic drug company to "borrow" an innovator's clinical data in order to secure FDA approval for a generic drug without having to perform lengthy and costly clinical trials.  In this way, the Act could meet its goal of speeding market entry of generic drugs to foster price competition.  In exchange for "loaning" their clinical data, the Act provided innovators with mechanisms to extend patent life (i.e., patent term extension) and data exclusivity (up to 5 years under Hatch-Waxman).  Stating that "[t]he results of Hatch-Waxman have been striking," the Deloitte paper notes that "[c]onsistent with its intent, it did succeed in creating price competition in the market" -- with the Congressional Budget Office (CBO) estimating that the Act was generating annual cost savings of $8-10 billion ten years after its passage.

The Deloitte paper, however, also notes that the Hatch-Waxman Act produced unintended consequences.  One consequence was a sharp increase in R&D investment among innovator companies, and another consequence was that prices for new branded drugs have substantially outpaced baseline GDP growth.  Contending that "it is important to understand these unintended consequences, because they may shed light on what we might expect if an abbreviated approval pathway is created for FOBs," the paper explains that these consequences are the result of three effects:  the "make hay effect," the "blockbuster effect," and the threat of "no man's land."

With respect to the "make hay effect," the paper explains that when a generic company enters the market, it quickly takes market share from the innovator, which in turn reduces the innovator's revenues.  Faced with the prospect of lower revenues and a lower return on investment, the innovator will try to maximize its revenues prior to generic competition, which the innovator can do by raising prices and by investing more in the marketing of a new drug at launch in order to drive earlier adoption.  The paper states that in response to Hatch-Waxman, innovators utilized both of these options.

Innovators also responded to Hatch-Waxman by increasing their R&D investment.  However, the Deloitte paper observes that "[g]iven an ever-increasing cost for drug development, but a capped period of patent protection during which to achieve most return on investment, innovators will concentrate on the development of drugs with the highest revenue potential, i.e., blockbusters."  Noting that the development of a single drug takes about 12 years and costs about $1 billion, the paper concludes that "to break even the average drug must achieve average annual revenue of roughly $150 million," which "cannot reliably be achieved unless a drug targets a large population of patients, or comes at a high cost per treatment."

The final effect, the threat of "no man's land," arises because "for each compound waiting to be developed, there is a point at which it will never earn sufficient return to fund its development, because if too much time elapses, there won't be enough time remaining on market for it to generate sufficient sales."  When a drug has reached that point, the Deloitte paper says it has crossed into "no man's land."  Interestingly (and unfortunately), the paper "estimate[s] that no man's land appears very quickly for a new compound -- within as little as one year of receiving a patent" (see note 7 in the paper, where the authors assume 25 years of total patent life, and then subtract 10-12 years for drug development and 10-12 years of time on market to earn required ROI, leaving the innovator with 1-4 years of commercial viability).

In view of differences between the small molecule pharma and biotech industries, the paper next addresses whether a follow-on biologics regulatory pathway would likely generate unintended consequences and effects similar to those arising after passage of Hatch-Waxman.  The paper notes that because biologics are more complex than small molecule therapeutics, it takes longer (on average, 97.7 months for new biological entities versus 90.3 months for new chemical entities) and costs more to bring biologics to market.  In addition, the pharma and biotech industries are quite different, with the former comprising larger, self-funding companies and the latter comprising a smaller number of large companies and hundreds of smaller companies that are dependent on venture investment.

The paper speculates that "the introduction of a follow-on path could spur investment in the technologies required to make it work," including, for example, "increased investment and therefore advances in genetic and molecular assay technology, required to assess biosimilarity" -- which would be a positive consequence.  However, the paper also states that "[t]he very size of these molecules opens the possibility that a very small change to the molecule that preserves the core design . . . could circumvent the IP of the innovator company without technically infringing on its patent."  Thus, the paper asserts that while "patents protected innovators' IP, and the data exclusivity period rarely came into effect" for small molecule therapeutics, "[f]or biotech innovators, if patent provisions are not sufficient to protect their IP, it may need to be the other way around."

The authors also contend that several of the effects seen in the pharma industry's response to Hatch-Waxman are already impacting the biotech industry (even without an FOB regulatory pathway).  Thus, innovators are forced to focus on drugs with the largest possible market potential -- a result that "runs exactly counter to the direction and promise of the science of biotech, which has the potential to create more highly targeted and therefore more efficacious therapies."  The paper adds that "[i]t would be unfortunate if new regulations unintentionally circumvented the advances that now appear to be possible in medical science by putting in place economic incentives that rule out everything but blockbuster investments."

After assessing the differences between the small molecule pharma industry, which was a stable and mature in 1984 when Hatch-Waxman was implemented, and the biotech industry, which is nascent and complex (both scientifically and financially), the Deloitte paper concludes that "[i]n establishing a path to market for follow-on biologics, Congress may need to employ a different set of levers to achieve the same results."  Therefore, an FOB regulatory pathway patterned too closely on the Hatch-Waxman regime would likely have unintended, and potentially adverse, consequences for the biotech industry.

Patent Docs readers are encouraged to alert us to any papers that we may have missed -- regardless of the positions that the authors of such papers may take on the issue.

Comments

[EG]

Kevin,

Very interesting perspective on FOBs. Brings up more reasons why making the FOB legislation a "sequel" to Hatch-Waxman a "bad" idea. In my opinion, those pushing for FOB legislation in Congress are too much focused on bringing FOBs quicker to market to supposedly lower biologic drug prices. What's ironic is that this paper points out that such a focus may have the opposite effect (i.e., increasing biologic drug prices at the beginning), as well as making development of certain biologic drugs commericially infeasible because they won't garner enough revenue before the FOBs come in. Congress needs to mull over the conclusions in this paper carefully.

[EG]

Don,

Sorry, I'm so used to Kevin doing the FOB stuff that I didn't give you appropriate credit. A well done piece.

[Skeptical]

I’m a little confused about the math above, specifically the “no man’s land” calculations, and the reaching of that point within one year of receiving a patent.

1) Starting with 25 years, 10-12 years are lopped off for drug development. I recognize that there are regulatory hurdles, but is the date of the effective patent the start on overcoming these hurdles? I would think that before a patent is applied for, that the bulk of the development has been completed, thus 10-12 years for further drug development is excessive.

2) 10-12 years is then lopped off for time on market to earn required ROI. This also seems excessive or at least an arbitrary timeframe for an uncertain objective. What is “required ROI”? Doesn’t this very factor indicate that a required return on investment would capture both the development costs, the “cost/value” of the endeavor and the expected profit? Is this the same as saying that 10-12 years is all the actual exclusivity needed to fund the project and make it worthwhile (i.e. profitable)? Who would ever run a business where the “’required ROI” doesn’t meet these basic requirements?

3) Factor 2) above begs the question of just what does “commercial viability” mean. Is this a “pure profit” time frame after the “required ROI” has been met?

Later in the article the authors contend that the pharma industry responds to Hatch-Waxman by being forced to focus on drugs with the largest market potential, and seems to imply that but for this government interferences the industry would embrace a different and more utopian direction and promise of the science of biotech. Can a statement be any more dishonest? Clearly, the pharma industry is a business and business decisions, not altruism runs decisions (review again what makes up a “required ROI”). Blaming the government for profit decisions is disingenuous at best. At what level of “required ROI” sans government regulation would the business of pharma pursue “highly targeted” therapies?

I just don’t buy it.

[m]

Skeptical,

1) The 10-12 years of further development is a fairly accurate estimate. This is referring to the API patent which is often the only patent that counts(Composition of matter). When this patent is first filed, its at the very beginning of an exploratory stage of drug development. Often there is a rush to patent these molecules very early with little data as there is in some cases up to 15 other companies look at that same therapeutic area and at that same mechanism of action. To give you an idea of the work that still needs to be done, you're talking about getting a semi-decent organic chemistry route, polymorph screening, formulation development, preclinical trials, scaling up the organic chemistry, clinical trials, and numerous requests from the FDA for additional studies and delays. Each of these steps in drug development is typically very complex and unexpected issues such as degradation of the API, poor bioavailibility, metastable polymorphs, clincial trials recruitment, etc. etc. pop up all the time. Even issues such as how to clean manufacturing equipment are hurdles.

Also, as a note, the 25 years statement is a poor error as the authors confused how the patent interacts with the FDA market exclusivity as Don had noted. An experienced person would have easily caught this error.

2&3) You're getting at the core business model of the industry here. And the issues relates to the fact that any business must get more money out than they put in. This is turning out to be tougher and tougher due to high R&D Costs, many unmet medical needs have "low" patient populations, and most companies are working in the same areas which leads to a hyper-competitive environment. Because of all the unpredictable issues that arise in development, its a risky business, and it may at some point not to be commercially viable as you had noted. Over the last ~10 years or so, R&D spending has increased ~300% while the number of new molecules outputted has decreased ~50%. [I'm not too clear on these numbers but I remember seeing a study with them and they generally seem correct to me]. If you know someone in the industry, this has likely meant they have had several jobs, as the last ~6 years, there have been massive layoffs in the industry, somewhere around ~100,000 layoffs across several companies.

Its generally accepted that PhRMA gave up too much in Hatch-Waxman. And this is a core issue of any FOB legislation, the ability to provide an ROI. Those people laid off that I mentioned above is a sign that the size of the industry is contracting, and this as a whole is bad thing to have less people working on new treatments where there is very little or nothing to offer currently to someone who suffers and would benefit from a new treatment.

The question of what therapeutic areas to pursue always has to consider the number of afflicted people, its just not feasible to spend $1-1.5B on research if there are only 10,000 affected people. And clearly, if the incentives were properly created, perhaps FDA exclusivity could be granted for a longer period to incentivize treatements with a small patient population (i.e. Orphan act). What if you were in that small patient population?

Not sure if any of these comments will help Skeptical since the nature of the pharmaceutical business is incredibly complex and risky and not easy to explain all of this in a blog post, especially difficult for those with no experience in the industry. But I think these issues underscore that one has to think very carefully about FOB legislation, as many of the articles that Don has referenced have attempted to do, to balance all the factors to promote innovation and a society that can benefit those rewards.

[Dan Feigelson]

Skeptical, the bulk of the work - and the expense - in bringing a new drug to market is in the clinical trials, and the earliest patent applications in the pharma and biotech fields are filed well before clinical trials begin. Lopping 10-12 years off effective patent life for drug development is a reasonable estimate.

[Skeptical]

I want to express my sincere appreciation to m and Dan. I recognize and readily admit my lack of knowledge to the full complexity of this area. It is perhaps without this entrenchment that my skepticism should be viewed, and thus in more of a policy and numbers check (business), rather than a detailed process view.

The thought keeps coming to me, as I read the response on how difficult and complex the 10-12 years of further development is, that maybe the early filing of a patent is the problem. If you still have so much work to do, maybe you do not have “something” that can be patented. Or maybe, you do have “something” that can be patented, but that “something” is not what you end up with 10 to 12 years down the road after the further development. Sure there is competition, but that is no reason to be able to either patent “something” you do not have, or patent “something” you do have, but is not the same “something” that you have after 10-12 years of further development. Why should the 10-12 years of development time be a consideration at all in regards to the earned patent rights? Your own arguments about the hurdles and effort needed after patenting can be used against the right to this 10-12 year further development period.

As to 2) and 3) and the core business model: Why should the patenting law bend to ANY sector’s choice of business model? Perpetuating this arrangement can be argued to only result in continuing the wasteful duplication of effort. From your post, you indicate that “most companies are working in the same areas”. Why should patent law be bent to support this? I think that the trends you note (massive layoffs) are actually a realization that the business model CANNOT be supported. I would further assert that whether this is bad or good depends on the actions next taken. If this previously duplicative effort can be funneled into different avenues, then a more effective utilization of resources would result. Sure, this is a Pollyanna statement – I realize that the use of resources is driven by Business decisions, but then again, that was the point of my first post. It’s not the government, it’s the business that needs scrutiny.

As you (rightly) point out – it is a difficult question – just how much is it worth to make advances that benefit smaller and smaller sets of people? The underlying assumptions going into the feasibility equation must be put forth in black and white. I am afraid that the answers provided by the common man would be disappointing. The voice of the market will clamor for Billion dollar sports figures playing for the nation before the voice clamors for Billion dollar drugs applicable to only tens of thousands of individuals. But that’s business, isn’t it?

Aside from the policy point, and setting aside the first 10-12 year period, I still am very skeptical on the math from my points 2) and 3) above. Any answers there?

[Kevin E. Noonan]

Dear Skeptical:

I think there is a misunderstanding here. Typically, a drug company identifies a drug, or a class of drugs, and files a patent application. There has usually been some "development" work done before the application is filed, since for most drugs the requirements of 112 mandate a description of how to make and use, and the "use" part requires at least in vitro demonstrations of utility.

What follows (and what makes drug patents different from other types of patented products) is 10-12 years of preclinical and clinical trials mandated by the FDA. Now, I don't object to having this level of review, since even with it there have been instances (Thalidomide, FenPhen) of drugs getting on the market that were later found to have significant deleterious side effects.

But during all that time the drug company cannot sell its product. Even though Hatch-Waxman provides up to 5 years of patent term extension, that isn't always enough for a drug to reach a break-even point. Hence, the Deloitte authors description of the "no-man's land" problem for this sector.

So it isn't a question of whether you have patented what you don't have - you have patented what you can't (yet) sell.

And remember, what we are talking about here is data exclusivity - the idea that the government permits a third party generic company to use the safety and efficacy data produced by the innovator to make a generic version of a drug. For biogenerics, this may produce more problems than Congress may recognize, if the complexity of biologic drugs is such that the innovators safety and efficacy data don't apply to the biogeneric.

So what you have is a regulatory scheme where a patentee must provide safety and efficacy data to the government in order to sell her product, and then the government agency that requires the data gives it to a competitor, and lets the competitor use that data to qualify as a market competitor against the innovator. This is not bending patent law to a business model - this is the environment drug companies must live in imposed (for good reasons) on their business for the common good.

I'm not saying all innovator companies wear white hats and all generics are bad guys - but I am recognizing the fact that precious few generic companies do any innovative drug research. We can argue about whether innovator drug companies spend "enough" on R&D and "too much" on marketing, lobbying, etc., but the simple fact is that those companies are bringing new drugs to market. And so long as we live in a capitalist economy, investors will require a return on their investment commensurate with the costs and risks of the business they are investing in. Drug development is perhaps one of the riskiest businesses to invest in, so I don't see the argument that their business model is what is flawed (actually, I don't think of any of it as "flawed," per se, just subject to various costs and pressures).

Hope this helps. Thanks for the comment.

[Keep It Real]

"We can argue about whether innovator drug companies spend "enough" on R&D and "too much" on marketing, lobbying, etc., but the simple fact is that those companies are bringing new drugs to market."

Round and round we go.

[Skeptical]

Kevin,

I appreciate the response, and the patience. Even though I can see facets now that I was overlooking, I am not yet convinced that the “something” argument is satisfied.

Let me share that I do have a bias for NOT carving up the patent field by sector. I realize that this view pleases no one individual sector, but unless we are willing to endlessly divide and sub-divide and create a patchwork of laws, one unified Rule of Law should govern.

That said, I do recognize that the sectors are of course different. I would posit that perhaps it is not in patents where that difference should be made up.

Back to the “something’ discussion. As you point out, the patent rules do call out for the “make and use”. My distinction would be that the “make and use” for the patent is simply not enabling to the same degree of the “make and use” for the final product after the further development. In my mind, the two are just not the same. Granted, the latter is due to the extra governmental regulations, but the reason why does not matter as much as the fact that they are different. Perhaps NO patents should be given until the “make and use” satisfies ALL government requirements. Of course I realize that I may be creating a dichotomy with my one unified Rule stand, but perhaps not. Perhaps a patent should be enforced on what it actually covers. If you patent the early work, that is not safe for humans, you have the rights to a product that is not safe for humans. If you patent the latter item, then you have rights to the latter item. The latter item, being where I suppose the money is at, then would have a patent dated from when the product is ready to go, rather than when the “idea” is partially ready, but not human-ready. In this manner, the 10-12 years of further development would not be eating into the exclusivity timeframe of the truly patentable item. I have a hard time believing that a patent for something that needs an additional 10-12 years of development can be for the same item that emerges at the end of that development.

As far as drugs reaching a break-even point, I remain skeptical. Reaching that point has many, many business decision points completely divorceable from the patent realm. Should we then create business profitability points for every sector?

The direct topic at point – the idea that the government permits a third party generic company to use the safety and efficacy data produced by the innovator to make a generic version of a drug would be abated if that data was made part of the quid pro quo for the patent of the ready-to-go item, as opposed to the ready-to-idea item. By portioning out the government requirement for the final end product, we are playing shell games. I think it cleaner to patent the working end product and provide for the patent rights for that particular “something”. The fact that the ‘something” is something that works on humans to the degree mandated by law still would apply to anyone developing the item.

[Skeptical]

I realize this thread is aging, but was hoping for a further response from the respected doctors as I notice that Ariad Pharmaceuticals Inc. v. Eli Lilly & Co. case includes an argumetn by Lilly concerning the plague in the biotechnology filed of prematurely filed patent applications mirroring my position as outlined above concerning "something" requiring an additional 10 - 12 years of further development.

[Kevin E. Noonan]

Dear Skeptical:

I think what Eli Lilly was arguing was that Ariad attempted to claim far beyond what they had enabled or described. The patent in that case was directed towards anything that inhibited the transcription factor NFkB, which is a fundamental cellular process for regulating gene expression.

What we have been discussing here is what it takes to go from the original invention - a gene encoding a hormone, for example - through the point where it becomes a viable commercial product. The structure of the hormone doesn't change, and there is nothing "new" about what follows from the original patent claim, but getting a biologic to market is more complicated than getting a small molecule to market.

So it seems like apples and oranges to me, except that in both cases we are judging today what was done 10-12 years ago.

Thanks for the comment.

[Skeptical]

I appreciate the response, but fear that we will not see eye to eye on this.

While indeed Lilly was arguing specifically as you state, the gist of the argument is that what was claimed was a different "something" than what was in possession at the time of filing. When we have different "something"s between the time of filing to the later time of viable commercial product, it seems clear that those different somethings are like apples and oranges. Or perhaps like oranges and nectarines. They are both fruit and even fruit from ongoing labor, but they are nonetheless different. I simply don't agree with you that there is nothing "new" coming from 10-12 years of labor required to make the product human safe. Further, if this time frame is required (by law) for safety and efficacy, then the results of this effort should be included in the patent to meet 112 requirements (how is 10-12 years effort NOT undue experimentation?), and the patent grant should be delayed until the application can be filed for the actual "something" that protection is desired for. Like I said earlier, you should get a patent on the something that you file for. If that something is not human-safe, it was your choice to file early, and patent protection should not extend to the additional 10-12 years of further development it takes to actually arrive at the later "something".

[Kevin E. Noonan]

Dear Skeptical:

I guess it comes down to philosophy. If you believe a patent is a reward for something, or "monopoly" that has to be earned, then being parsimonious about granting them makes sense.

But if you view a patent as a tool for encouraging disclosure and innovation, then the question is, "what does an innovator need to get a product to market?" For a drug, the initial drug itself (omeprazole or erythropoietin) may be invented 10-12 years before it can be sold commercially, due to development and regulatory requirements. These go hand in hand, and provide safe and effective drugs.

If you make the philosophical decision that 10-12 years is "too long" and the inventors don't deserve a patent early, then you have to accept the consequences - fewer investors willing to provide the capital for drug development. And in the biotech space, since the initial investors are venture capitalists, you just don't get the cutting-edge drugs you need.

I don't like that outcome, so I prefer to have early patenting, which gives me early disclosure and stimulates others to investigate the space and maybe come up with independently patentable drugs of their own, and then provide a regulatory environment where most drugs recover their costs (or at least there are sufficient drugs that do to pay for the ones that don't).

And remember, there is nothing to stop the biogenerics from doing their own safety and efficacy studies and coming on the market sooner than 12 years due to data exclusivity delay. If they are unwilling to do that, maybe it is the biogenerics that are being greedy.

Thanks for the comments.

[Skeptical]

I will think about your argument, but on first glance it strikes me as a type of bait and switch - going from my point on "something" to an "oh, well, let's look at a philosophical angle". This leaves the substance of my argument unanswered.

It seems disingenuous to mangle patent law based on the economics of this sector. Patent law is meant to be universally applied. To me, "early patents" simply are bogus because the additional work of 10-12 years of development to get the later "something" simply hasn't been done. In other words, the "something" point drowns out the philosophical "but we need it early" point.

On the thought about nothing stopping the biogenerics from doing their own studies, I think this point too not only loses force if those same studies are actually necessary to show that the patentee has possession of the later "something" that patent protection is desired, but actually seals the argument that such additional development IS necessary and that the early "something" should NOT extend rights to an undeveloped and diferent later "something" (Maxim: you get a patent for your something, not for something else). "Greed", as such, is an emotional ploy that I don't buy, and such ploys are yet another reason to take a hard line on the consistent application of patent law across business sectors.

[Kevin E. Noonan]

Dear Skeptical:

I think I see where we went a little off track here. We are really taking about two different things, patent protection and regulatory protection, i.e. data exclusivity.

Patent protection is "stronger" in that is precludes a generic company from doing anything (other than what is exempt under 35 USC 271(e)(1)) while the patent is in force. Data exclusivity is the opposite - a generic company can do anything it wants except use the innovator's data.

So in some ways the protections are complementary. The reason for the debate is that patent law changed in 1996 - instead of receiving a term 17 years from the grant date, now applicants get 20 years from their filing date. Which means that many times patent protection will "fall short" due to delays in the process.

So it isn't a case of patenting too early - when you patent is not the issue. What is the issue is what can the government do with an innovator's data - can they give it to the generic, and when. That is a regulatory question, and one that is open to debate on economic grounds. The Deloitte paper is directed to what will be necessary for there to be sufficient ROI in the biotech sector for investment. That has nothing to do with patent protection, because as we discussed the generic company can do its own safety and efficacy testing without regard to the patent (either because it has expired or under the statutory exemption).

So while in the past patent protection might have "filled in the gap" regardless of the data exclusivity term, today it is largely not the point - what will provide the greatest barrier to entry (and conversely could be the biggest impediment to investment) is the length of the data exclusivity term.

Sorry if I was unclear. Thanks for the comment.