The Great American Drug Deal

by Peter Kolchinsky

  281Hypoglycemia refers to low blood sugar, which can cause a person to lose consciousness, experience a seizure, or die.

  282Mayer B. Davidson, “Insulin Analogs—Is There a Compelling Case to Use Them? No!” Diabetes Care 37, no. 6 (2014): 1771-4, accessed Oct. 15, 2019. doi: 10.2337/dc13-2915, http://care.diabetesjournals.org/content/37/6/1771;

  George Grunberger, “Insulin Analogs—Are They Work It? Yes!” Diabetes Care 37, no. 6 (2014): 1767-70, accessed Oct. 15, 2019. doi: 10.2337/dc14-0031, http://care.diabetesjournals.org/content/37/6/1767.

  283John LaMattina “Today’s insulin isn’t what Banting and Best developed. It’s far, far better,” Stat, November 14, 2019, https://www.statnews.com/2019/11/14/insulin-safer-better-easier-to-use/ accessed November 23, 2019.

  284Since biosimilars are the closest thing to generic versions of biologic drugs like insulins, I use “go biosimilar” to mean the same thing as for a conventional drug to “go generic.” The idea is the same: the price drops due to competition among interchangeable versions of the same drug.

  285Joyce Frieden, “Senators Probe ‘Enormous’ Insulin Price Spikes,” MedPage Today, May 8, 2018, https://www.medpagetoday.com/endocrinology/type1diabetes/72771.

  286Diana I. Brixner and Carrie McAdam-Marx, “Cost-Effectiveness of Insulin Analogs,” The American Journal of Managed Care 14, no. 11 (2008): 766-75, accessed Oct. 15, 2019. https://www.ncbi.nlm.nih.gov/pubmed/18999911?dopt=Abstract.

  287Here again, I use “biosimilarization” to refer to genericization (process of going generic) of a biologic drug like insulin. In other words, these studies didn’t even take into account that insulin analogs would eventually become inexpensive.

  288Amelia Dmowska and Payal Marathe, “FDA Approves Admelog, the First Biosimilar Mealtime Insulin,” diaTribe Learn, Jan. 17, 2018, https://diatribe.org/fda-approves-admelog-first-biosimilar-mealtime-insulin;

  ER/AJW, “FDA Approves New Insulin Glargine Basaglar—The First ‘Biosimilar’ Insulin in the US,” diaTribe Learn, Jan. 11, 2016, https://diatribe.org/fda-approves-new-insulin-glargine-basaglar-first-biosimilar-insulin-us.

  289In terms of exceptions, I’m referring to drugs that have a high marginal cost of production, such as gene therapies and CAR-T therapies, such that even making them available to patients in the poorest developing countries would mean they are unaffordable (e.g., when a dose of a gene therapy costs $50,000 to manufacture). It’s also functionally challenging to deliver some drugs that are inexpensive to produce but require cold storage, labor-intensive administration (frequent injections, infusions), or careful monitoring (e.g., have serious side effects that require hospitalization, frequent lab tests, or other therapies, such as administration of Epogen to treat drug-induced anemia or GCSF to treat drug-induced neutropenia).

  290Basically, if to help a patient you need a drug that can reach into a right-hand shaped crevice in a problematic protein to turn it off, then don’t go throwing in a mixture of right and left hands to do the job because you have no idea what those extra left hands are off doing throughout the body.

  291Silas W. Smith, “Chiral Toxicology: It’s the Same Thing...Only Different,” Toxicological Sciences 110, no. 1 (2009): 4-30, accessed Oct. 15, 2019. doi: 10.1093/toxsci/kfp097, https://academic.oup.com/toxsci/article/110/1/4/1668162.

  292Technically named gastroesophageal reflux disorder (GERD).

  293“Development of New Stereoisomeric Drugs,” FDA, updated March 21, 2018, https://www.fda.gov/regulatory-information/search-fda-guidance-documents/development-new-stereoisomeric-drugs.

  294Before Nexium was approved, many physicians had no choice but to prescribe 80mg of Prilosec, twice the FDA-approved dose, to help patients who were not getting enough relief from just 40mg of Prilosec. After Nexium was approved, some physicians remained comfortable continuing to prescribe a high, off-label dose of generic Prilosec to save the patient and society money, but they did so while taking on potential legal liability if anything went wrong since there was a clinically proven, FDA-approved alternative. Therefore, many physicians preferred to prescribe 40mg of Nexium when a patient needed more than 40mg of generic Prilosec.

  295Interestingly, although AstraZeneca’s upgrade of Prilosec to Nexium was aligned with what the FDA was urging the entire industry to do, there would have been no reliable mechanism to reward AstraZeneca for creating Nexium had they not doubled its dose to give it an efficacy advantage over Prilosec. If AstraZeneca had merely tried to market Nexium 20mg, insurance companies could have only covered generic Prilosec 40mg and refused to reimburse for Nexium 20mg without clinical evidence that it was better. And had AstraZeneca merely doubled the dose of Prilosec to demonstrate that 80mg was more effective than 40mg and tried to keep marketing the 80mg dose as a new brand while 40mg of Prilosec went generic, physicians would have felt safe prescribing a double dose of the generic 40mg pill. So it was the combination of making Prilosec purer and doubling the dose to create Nexium 40mg that allowed AstraZeneca to profit from their incremental innovation. By comparison to improving efficacy by doubling the dose, converting Nexium 40mg into a once-daily pill was merely a nice convenience bonus.

  296Mateji Mikulic, “AstraZeneca’s Revenue from Top Product Nexium 2006-2018, Statista, March 11, 2019, https://www.statista.com/statistics/266545/astrazenecas-revenue-from-top-product-nexium-since-2006/;

  “Nexium: How Patent Shenanigans Cost Payers Billions,” Prescription Intelligence, Dec. 30, 2015, https://prescriptionintelligence.com/nexium-how-patent-shenanigans-cost-payers-billions/.

  297R Paul, “Comparison of Racemic Ketamine and S-ketamine in Treatment-Resistant Major Depression: Report of Two Cases,” World Journal of Biol Psychiatry 10, no. 3 (2009): 241-4, accessed Oct. 15, 2019. doi: 10/1080/15622970701714370, https://www.ncbi.nlm.nih.gov/m/pubmed/19224412/.

  298Marco A. Ramos et al., “Opinion: The New Ketamine-Based Antidepressant Is a Rip-Off,” Vice, May 17, 2019, https://www.vice.com/en_us/article/pajkjy/opinion-the-new-ketamine-based-antidepressant-is-a-rip-off.

  299AstraZeneca would probably have just stuck with the Prilosec brand name.

  300In this case, risk can refer to the possibility that a child experiences some adverse event in the clinical trial which might or might not be due to the drug being studied and then all physicians become more cautious using the drug in adults, putting the drug’s sales in that large adult market at risk, even if the drug has already been proven to have a favorable benefit-risk profile in adults.

  301Once drugs are generic, there are essentially no incentives or mechanism for anyone to continue studying them to learn how they should be used in children or for new diseases except to find a way to convert them into a novel formulation or employ some other kind of incremental tweak, which would come with an entire new patent period, which may result in society paying more for that upgrade than what it would have cost had it been incentivized with a short exclusivity extension prior to the original drug going generic.

  302Sometimes companies try to tweak the formulation of that drug so that they can get a patent, hoping to enjoy a longer period of exclusivity, but if the tweak is too modest, then the patent may be challenged in court by companies wanting to launch generics as soon as the Hatch-Waxman exclusivity expires.

  303The term “orphan disease” referred to diseases that were considered too small (formally defined as <200,000 patients in the US) for anyone to pursue and were typically neglected by innovators who couldn’t justify the cost and risk of pursuing treatments for them. Today, many orphan diseases are far from orphaned, in some cases with dozens of companies competing to figure out how to solve them, which offers hope to millions of patients. While the number of Americans suffering from any one orphan disease is small, there are approxim
ately 7,000 of these rare disorders and an estimated 25-30 million Americans (7.6%-9%) suffer from one or more orphan diseases.

  “FAQs About Rare Diseases,” NIH, GARD, updated Nov. 30, 2017, https://rarediseases.info.nih.gov/diseases/pages/31/faqs-about-rare-diseases.

  304Wikipedia Contributors, “Orphan Drug Act of 1983,” Wikipedia, The Free Encyclopedia, accessed Oct. 15, 2019, https://en.wikipedia.org/wiki/Orphan_Drug_Act_of_1983.

  305Recall an earlier footnote on pg. 184 about Prilosec/Nexium. Only the combination of developing both a patentable pure enantiomer and testing a higher dose gave AstraZeneca the incentive to show that Nexium could treat heartburn better than Prilosec. Were Prilosec not a mixture of two enantiomers and therefore unable to be upgraded into a pure enantiomer, AstraZeneca might not have studied the higher dose to properly show that it was even more effective (though some physicians would have just used a higher dose off-label).

  306See Prazosin case study in Chapter 15.

  307Operating Plan for FY 2019 (National Institutes of Health, 2019), 1-2, https://officeofbudget.od.nih.gov/pdfs/FY20/cy/FY%202019%20NIH%20Operating%20Plan%202.8.19%20Web.pdf

  308US Investments in Medical and Health Research and Development, 2013-2015 (Arlington, VA: Research America, 2016), https://www.researchamerica.org/sites/default/files/2016US_Invest_R&D_report.pdf.

  309Hamilton Moses III et al., “The Anatomy of Medical Research: US and International Comparisons,” JAMA 313, no. 2 (2015): 174-89, accessed Oct. 15, 2019. doi: 10.1001/jama.2014.15939, https://www.ncbi.nlm.nih.gov/pubmed/25585329.

  310Europe has allowed some biosimilars on the market as of 2019. Biosimilars are expected in the US in 2023. Even then, Humira will enjoy orphan exclusivity for hidradenitis suppurativa until 2025.

  311“Humira: The Highs and Lows of the World’s Best-Selling Drug,” Pharmaceutical Technology, Sept. 2018, https://www.pharmaceutical-technology.com/features/humira-abbvie-drug/.

  312AbbVie, AbbVie Reports Full-Year and Fourth-Quarter 2018 Financial Results, Jan. 25, 2019, https://news.abbvie.com/news/abbvie-reports-full-year-and-fourth-quarter-2018-financial-results.htm.

  313Kyle Blankenship, “The Top 20 Drugs By 2018 US Sales,” Fierce Pharma, June 17, 2019, https://www.fiercepharma.com/special-report/top-20-drugs-by-2018-u-s-sales.

  314The NjarŎarson Group, Top 200 Pharmaceutical Products by Retail Sales in 2018 (Tucson, AZ: The University of Arizona, 2018), https://njardarson.lab.arizona.edu/sites/njardarson.lab.arizona.edu/files/2018Top200PharmaceuticalRetailSalesPosterLowResFinalV2.pdf; This lists drugs by their global sales, so, to be more inclusive in my count, I conservatively estimated that US sales represented 60% of these values, though for some drugs 30-50% might be more appropriate.

  315This does not mean $500 million of profit, though these are the highest profit-margin products for the industry and so I would estimate every $500 million more of revenues could generate at least $300 million of after-tax profits.

  316Or, as proposed in Chapter 8, a new Generic Drug Contracting Bureau.

  317And if its price doesn’t come down substantially due to biosimilar competition, as some fear, it would be a good candidate for contractual genericization.

  Mark Trusheim et al., “Biologics Are Natural Monopolies (Part 2): A Proposal For Post-Exclusivity Price Regulation of Biologics,” Health Affairs, April 15, 2019, https://www.healthaffairs.org/do/10.1377/hblog20190405.839549/full/.

  318Jing Wu, “While Kalydeco Sailed Through, Vertex’s Orkambi Faces Strong Headwinds in Europe,” Decision Resources Group, June 2, 2016, https://decisionresourcesgroup.com/blog/kalydeco-sailed-vertexs-orkambi-faces-strong-headwinds-europe/.

  319Ed Silverman, “Vertex CEO Complains to U.K. Prime Minister About Stalled Talks Over a Pricey Drug,” STAT, July 6, 2018, https://www.statnews.com/pharmalot/2018/07/06/vertex-ceo-uk-prime-minister-drug/.

  320Eric Sagonowsky, “Vertex Cancels Plans for CF Triple Combo Testing in France, Citing Failed Orkambi Negotiations,” Fierce Pharma, Feb. 15, 2018, https://www.fiercepharma.com/pharma/after-failed-orkambi-negotiations-vertex-pulls-plug-phase-3-triple-combo-testing-france.

  321Though the UK did eventually come to an agreement with Vertex.

  Vertex, Vertex Announces Agreement with NHS England for Access to All Licensed Cystic Fibrosis Medicines, Oct. 24, 2019, accessed Nov. 1, 2019, https://investors.vrtx.com/news-releases/news-release-details/vertex-announces-agreement-nhs-england-access-all-licensed.

  322The UK reimburses many drugs at prices above those NICE deemed cost-effective, which suggests that NICE’s calculations are more of a negotiating tactic than a hard line.

  323Alexander Gaffney et al., “Regulatory Explainer: Everything You Need to Know About FDA’S Priority Review Vouchers,” Regulatory Focus, Nov. 6, 2019, https://www.raps.org/regulatory-focus/news-articles/2017/12/regulatory-explainer-everything-you-need-to-know-about-fdas-priority-review-vouchers.

  324If a company is eager to start selling a drug that it thinks will generate $1 billion per year in sales, then it would be worth purchasing a priority review voucher for $100 million to bring that drug to market 4 months sooner, especially if it is competing with others to get to market first. These vouchers were worth more in the past when there were fewer of them, but drug developers responded to the incentive, developed a bunch of drugs the FDA wanted brought to market, and introduced enough vouchers into the market to bring down their price.

  325In nearly all the cases of an old drug being brought to market for a new use, either the development was funded by a non-profit or NIH (which typically don’t fund such efforts all the way through to FDA approval), or the company found something about that drug it could patent. That leaves a lot of untapped potential for repurposing old drugs.

  326I’m not proposing a wholesale alternative to the current generics marketplace since encouraging conventional generics manufacturers to participate in the genericization processes would protect against shortages if the originator encounters a manufacturing problem. Also, competition among manufacturers can drive cost-reductions that lower prices more than if just one company kept making the drug at a regulated, fixed price.

  327Which would be the case if a drug is ungenericizable (a natural monopoly), such as a gene or cell therapy, or in the case of a drug that is hard to genericize/biosimilarize (a natural oligopoly), as is potentially the case with biologics and complex drug-device combination products.

  328The specific pricing formula requires additional study and will be the subject of a future analysis. What matters is that the price be relatively low, simulating what society expects from a generic. The cost of producing most drugs is typically less than 10% of their branded net prices. Therefore, if a branded drug priced at $10/pill costs $1 to make, then limiting its contract generic price to $2/pill would result in 80% savings. If the pill costs 25 cents to make and its contract generic price were cut to 50 cents, that would result in 95% savings. However, in this case, I’m proposing that the price of the contract generic be based on the cost of both producing and distributing the drug, the latter being a small cost in most cases but potentially high in the case of special drugs that require special handling and storage conditions, such as radioisotopes, products that need to be shipped frozen, and living cells. Ultimately, the right markup might be higher or lower than I propose here. What’s important is that it confer meaningful savings for society relative to the status quo, that the math be transparent, and that the contract generic price be consistently high enough to motivate companies to want to hold the contract and compete for who can service it at a low cost while still preserving quality.

  329There would be no need for the contract to force a price cap on the originator if competition has alread
y reduced the price to less than two times the cost of production. But the contract would remain in place in case the number of generics dwindles and the price creeps up. If the originator discontinues producing the drug because it is already conventionally genericized, the Bureau would transfer the contract to another active manufacturer of the drug to ensure that it remains a failsafe in case there are fewer competitors someday and the price starts to rise above the contract price. One way or another, every drug America relies on for which society has paid off the mortgage would be under contract with the Bureau to ensure that its price stays at or below a price pegged to its cost of production.

  330For example, a once-daily version of a twice-daily pill; a dissolving film version for people who struggle to swallow pills; a less painful and/or easier injectable formation; a pure enantiomer like Nexium of a chiral drug like Prilosec; demonstrating that a higher dose is more effective or a lower dose is safer but equally effective; uncovering a way to use the drug more safely, possibly by restricting its use with certain other drugs or in certain patients; an additional use of a drug to treat a new disease, and other examples discussed earlier in this chapter. The six-month pediatric extension that the FDA already grants to companies that study how their drugs can be used properly in children would also be included as worthy of a genericization delay, as it is now.