The Great American Drug Deal

by Peter Kolchinsky

  Since the intent here is that the branded period last for the typical 10-15 years that initial core drug patents typically cover, a key concern is that companies will keep coming up with product upgrades to patent in order to delay genericization, as is common today. With ordinarily genericizable drugs, generics manufacturers challenge these secondary patents in court, getting leverage in their negotiations for when they can finally enter the market and start competing on price. But when a drug just can’t be genericized conventionally, there won’t be anyone to challenge the secondary patents. Therefore, to the extent that a company makes legitimately useful post-launch upgrades to its product, such as improving the manufacturing process or composition to make the product purer, safer, easier to produce, or better tolerated, those upgrades should be rewarded with short regulatory delays of genericization of the kind I propose in Chapter 13. I also include a detailed roadmap of how we can implement contractual genericization at the end of Chapter 13.

  The Special Case of Antibiotics, Generic Too Soon

  Antibiotics are like pandemic vaccines in that we want to know they are there well before we need them. Companies have brought novel antibiotics to market because they were told that patients are dying of multi-drug resistant bacteria and that hospitals badly need them to save lives. But when they finally got those antibiotics approved, they discovered that there were only a few patients who really needed them and so their sales were marginal. Those few patients had indeed been a cause for concern and drove headlines about drug resistant strains potentially killing thousands. Physicians and hospital administrators really were panicked that these strains would spread throughout their facilities and kill vulnerable patients recovering from surgery. They feared the worst if they didn’t get new drugs with which to combat these strains. But when companies delivered the goods, the worst didn’t materialize. Hospitals were able to use old drugs very effectively to manage the vast majority of their patients, using the new, expensive branded antibiotics very sparingly.

  Companies lost money on their investments, but the doctors’ fears eventually came true. Drug-resistant strains spread, and physicians were glad to have the new antibiotics. By then, the new drugs were old and generic or close to going generic, becoming inexpensive just in time to help treat the growing number of patients who needed them. As the next generation of resistant bacterial strains emerged, physicians and hospitals would again issue an urgent call for new antibiotics. But investors are now understandably hesitant to fund the expensive development of new antibiotics because they know there’s no real market for them yet.140 In response to their hesitation, the US,141 British,142 and other governments have realized that they need to offer some sort of payment guarantee, just as they did with the biodefense vaccine contract. One could argue that antibiotics represent a rare case of patent periods not being long enough to allow companies to recoup their investment, but going forward, a guaranteed payment model, not unlike the one used for biodefense products, is the more reliable way to encourage future antibiotic development.

  The Downsides of Contractual Genericization

  Critics of this proposal might say that price-control regulations will slow innovation. Public utilities, for example, don’t have an incentive to innovate. To keep costs low, public utilities keep salaries down, don’t give out significant bonuses as rewards for risk-taking, and don’t have substantial budgets for R&D. They offer steady, low returns for investors, steady jobs, and need to just not screw up. Unless public utilities get a mandate and additional funding—for example, from a rate hike—they try to keep doing what works forever.

  To those critics, I would say, “Exactly!” This model and set of expectations are exactly what we, as a society, need from the majority of the generic drug industry. Just keep making generic drugs steadily and reliably. Quality control, for sure, but no innovation needed!

  However, when it comes to complex biologics and the companies developing them, it is critical that we preserve the current incentives to innovate. That’s why contractual genericization only kicks in after a drug’s patents have expired. The first 10-15 years on the market give the companies the opportunity to collect their rewards for successfully inventing and commercializing new drugs. Very few drugs enjoy high prices beyond the first 15 years on the market, so there is little reason to think innovation would be stymied by using a contractual generic to cap profits beyond that horizon.

  Other critics might point out that this approach to genericizing a drug would indefinitely lock in only a single manufacturer of the drug, often at a single facility. Should catastrophe strike the facility, society might experience shortages and patients would suffer.

  As a safeguard against such unforeseen events, the company that makes the original drug could be mandated to transfer its manufacturing know-how to a second company and even a third company. In this way, society is guaranteed both manufacturing redundancy and price competition. This approach presumes that the FDA would recognize that the drugs produced at different facilities are the same, but that is not an outrageous presumption. The FDA routinely allows companies that make biologics to scale up their manufacturing by moving to larger facilities.

  This is not to say that manufacturing transfers of this nature are foolproof. Take the famous case involving Genzyme, a company we’ve talked about before. Genzyme originally launched the enzyme Myozyme for the treatment of infants with Pompe disease, then decided to expand the drug’s use for the treatment of older children. It needed to produce much more drug, but the FDA considered the differences between the original small-scale Myozyme (made in a 160-liter container) and the new larger scale (2,000 liters) product to be too significant to call them the same drug. This forced Genzyme to treat the batches from the new manufacturing process as a new drug, which it called Lumizyme, and that meant running a new clinical trial to get Lumizyme formally approved as a separate drug for older patients with Pompe.

  So if society were to rely on compulsory transfer of manufacturing to achieve manufacturing redundancy and price competition, the solution could be hindered or upended if the FDA deemed the products non-interchangeable. In the case of Lumizyme, the limited supply of Myozyme meant that older children had no choice but to enroll in a trial of Lumizyme. But if there had been enough Myozyme available to treat both infants and older children, why would anyone enroll in a clinical trial of a copy that the FDA didn’t consider equivalent, even one produced with the assistance of the original manufacturer?

  The Myozyme/Lumizyme case is a rare one, but it is more likely to occur with complex biologics like gene therapies. Therefore, contractual genericization of an otherwise ungenericizable drug would appear to be a more reliable way of fulfilling the social contract than compulsory licensing of manufacturing know-how to competitors. Hedging against supply shortages could also be achieved by other means, such as building up a large reserve of frozen doses, when freezing is an option.

  In any case, even a company subject to contractual genericization could be mandated, as part of their contract, to work with the FDA to open up a second and even third qualified site and allowed to charge enough to cover the costs of all that redundant manufacturing so that patients would be assured of supply.

  As we’ll discuss in the next chapter, contractual genericization could also help contain costs in cases where the market for a generic drug shrinks to be too small to sustain the critical mass of manufacturers that keeps prices down through competition (i.e., natural oligopolies). For biologics in general, it’s also possible that this approach would work better than relying on biosimilars, which some fear will not bring prices down very much like generics do.143 The point is that innovators count on a finite period of high profits as the incentive and fuel for their work, and what society cares about is seeing those innovations eventually become inexpensive. So to both parties, whether prices drop after the mortgage period due to competition or contractual genericization is of seco
ndary importance as long as they drop.

  Another Workaround: Prevention

  In the long run, prevention might obviate the need for expensive treatments altogether. For example, the need for CAR-T cell therapies might decrease if cancers were detected earlier when they might be more easily cured with surgery, radiation, and other conventional therapies.144 And gene therapies developed for rare genetic disorders potentially might be disrupted by more widespread preconception and prenatal genetic testing that enables families to side-step these disorders.145 In fact, insurance companies would find it cost-effective to pay for genetic testing for everyone and IVF for the few percent of couples at risk of passing a genetic disorder to their children than to skimp on these technologies and end up paying more than 10-fold more for gene therapies for affected children, especially when many gene therapies simply can’t be perfect cures (some of the prenatal damage from genetic mutations is irreversible after birth).

  Yet, even after the advent of widespread screening, one would still expect that some non-genericizable gene therapies would remain necessary indefinitely since not all genetic disorders are hereditary or detectable and not all cancers will be caught early enough to cure with more conventional therapies. Though they are unlikely to grow to be a large fraction of total drug spend anytime soon, it is worth considering a mechanism such as contractual genericization to ensure that even these therapies abide by the Biotech Social Contract.

  A Fair Trade

  The US government’s Congressional Budget Office is charged with running economic models that predict the impact on the federal budget of every proposed change in the law. The CBO is staffed by brilliant, data-driven people. They have put out reports on the impact of expanding insurance coverage to more Americans, for example showing the extent to which increased spending on insurance would be offset by savings from patients getting preventative care. Whenever proposals to modify what portion of costs patients should have to pay are put forth, the CBO models whether that will lead to increased utilization of drugs by patients and whether those cost increases will be offset by fewer surgeries and other expensive healthcare services.

  I don’t know the details of the CBO’s mathematical models. It’s possible that the CBO sees that a growing number of drugs are non-genericizable and models their long-term impact on the federal budget. I doubt it because the concept of non-genericizability is not widely discussed yet (I hope to change that). So if they don’t, they should.

  Here’s a trade I would like to see with the CBO model. Let’s introduce contractual genericization to bring down the federal government’s and society’s spending on non-genericizable drugs. Let’s see to what extent that offsets any increases in the federal budget due to the elimination (or, at least, fair and reasonable capping) of out-of-pocket costs. That’s the kind of bargain that the Biotech Social Contract upholds, and it’s time that we bring all drugs and all patients into alignment with it.

  * * *

  118Fiona Scott Morton and Lysle T. Boller, “Enabling Competition in Pharmaceutical Markets,” (Hutchins Center Working Paper, Yale School of Management, 2017), https://www.brookings.edu/wp-content/uploads/2017/05/wp30_scottmorton_competitioninpharma1.pdf.

  119Zachary Brennan, “Biosimilars in the US: Lower Costs, Increased Patient Access, IMS Report Finds,” Regulatory Focus, June 20, 2016, http://www.raps.org/Regulatory-Focus/News/2016/06/20/25166/Biosimilars-in-the-EU-Lower-Costs-Increased-Patient-Access-IMS-Report-Finds/.

  120Although Europe relies less on generics than the US, as mentioned in Chapter 3, it’s blazing a trail in the adoption of biosimilars.

  121QuintilesIMS, The Impact of Biosimilar Competition in Europe (QuintilesIMS, 2017), http://www.medicinesforeurope.com/wp-content/uploads/2017/05/IMS-Biosimilar-2017_V9.pdf.

  122Drugs reimbursed under Medicare Part B tend to be those that have to be administered by physicians, often infused in their offices. Medicare requires that physicians or hospitals purchase these drugs themselves and then reimburses them at a fixed percentage more than the cost, with the small premium (typically 4.5% or 6%) meant to compensate the doctor or hospital for the hassle of having to buy and store the drug. But that small premium can perversely encourage physicians to use more expensive drugs since 6% of a $5,000 drug ($300) is a lot more than 6% of a $500 drug ($30). So a biosimilar coming in at a lower price than a branded biologic drug would be less profitable for a physician. Clearly that reverse incentive has to change.

  123Scott Gottlieb, “Health & Wealth: Biocentury’s Interview with FDA Commissioner Scott Gottlieb,” Interview by Steve Usdin, Biocentury, June 23, 2017, https://www.biocentury.com/biocentury/politics-policy-law/2017-06-23/biocentury%E2%80%99s-interview-fda-commissioner-scott-gottlieb.

  124Andrew W. Mulcahy et al., The Cost Savings Potential of Biosimilar Drugs in the United States (Rand Corporation, 2014), https://www.rand.org/content/dam/rand/pubs/perspectives/PE100/PE127/RAND_PE127.pdf.

  125Chie Hoon Song and Jeung-Whan Han, “Patent Cliff and Strategic Switch.”;

  Zachary Brennan, “Biosimilars in the US: Lower Costs”;

  Benjamin Yu, “Greater Potential Cost Savings with Biosimilar Use,” The American Journal of Managed Care 22, no. 5 (2016): 378, accessed Oct. 15, 2019, http://www.ajmc.com/journals/issue/2016/2016-vol22-n5/greater-potential-cost-savings-with-biosimilar-use.

  126“NHE Fact Sheet,” CMS.gov, accessed Oct. 15, 2019, https://www.cms.gov/research-statistics-data-and-systems/statistics-trends-and-reports/nationalhealthexpenddata/nhe-fact-sheet.html;

  IQVIA Institute for Human Data Science, Medicine Use and Spending in the U.S.

  127This analysis is my own and required comparing figures gleaned from CMS’s website based on 2017 data and IQVIA’s 2018 data on total drug expenditures. Classifying which drugs are genericizable is not difficult for anyone in the drug industry who can recognize which drugs are simple chemicals, which are simple chemicals formulated in complex devices, which are simple biologics, and which are complex biologics. I list these in order of how hard they are to manufacture, from easier to hardest, which corresponds to how hard they are to genericize or biosimilarize.

  128Dan Stanton, “First US Biosimilar Gradually Eroding Amgen’s Market Share, Sandoz,” BioParhma Reporter, Jan. 29, 2016, https://www.biopharma-reporter.com/Article/2016/01/29/Sandoz-s-biosimilar-Zarxio-gradually-eroding-Amgen-s-Neupogen-sales.

  129These numbers are my own approximations based on my analysis of Medicare spending, which I have not published. I’m likely off in my calculations, maybe by 3x but probably not 10x, but it’s not important that the numbers be precise. The point is that we are currently spending very little of our drug budget on drugs that can’t go generic. Whether that’s 2% or 6% doesn’t matter—we shouldn’t allow spending on such drugs to grow and therefore need novel regulations to simulate genericization of these drugs when their patents expire (i.e. contractual genericization). If I’m wrong about the magnitude of the problem and the real spending on such ungenericizable drugs really is 20%, then it’s all the more urgent that we introduce contractual genericization. If I’m wrong the other way and spending is less than 2% on such drugs, it just means that the Biotech Social Contract is being violated to a much smaller degree than I fear, but we should still prevent the problem from growing by introducing contractual genericization. Some might think that we needn’t act now since the problem is currently insignificant. However, I think that if we wait until many companies are addicted to the cash flow streams of many ungenericizable drugs whose patents have long expired, then passing legislation that mandated simulated genericization will be a lot harder, as those companies will be fighting to preserve their unhealthy business models. The drug industry has long been in the business of charging mortgages for their products, not rent. We must make sure that this remains the case before the i
ndustry shifts to an unhealthy rent model.

  130“Hepatitis C Treatments Give Patients More Options,” US Food & Drug Administration, updated March 4, 2017, https://www.fda.gov/ForConsumers/ConsumerUpdates/ucm405642.htm.

  131These proteins are formally named using Roman numerals and are pronounced “Factor 8” and “Factor 9.”

  132If a patient’s immune system reacts to an infusion of Factor 8 or 9 as if it were a vaccine and then attacks the drug the next time it’s infused, then it can neutralize the effects of the drug, leaving the person essentially untreated.

  133Such gene therapies can actually be disrupted by side-stepping of the genetic disorder using IVF and genetic testing to allow couples who know they are carriers to select an unaffected embryo when having children. Some families that already know they carry hemophilia mutations already do this to spare the next generation the burdens of this disease, but to be broadly effective, side-stepping would require broader education about the utility of preconception genetic testing as well as widespread insurance coverage of IVF, at least for purposes of side-stepping genetic disorders. Genetic side-stepping is best described by Lee Cooper.