South Korea-Based Undbio Co., Ltd. + Proprietary Insulin; Maybe Biosimilars (someday)

On April 13, 2023, a South Korean biopharmaceutical company known as Undbio Co., Ltd. http://undbio.com/eng/ signed a lease with West Virginia University to build what it refers to as an insulin "manufacturing facility" in the city of Morgantown, WV (see https://www.wvgazettemail.com/business/undbio-planning-insulin-facility-in-morgantown/article_6b986775-6a2e-5ce5-b312-c423d79cbda2.html for more information). 

In all likelihood, that will be for what pharma refers to as a "fill & finish" facility, rather than one where temperature-controlled bioreactors are located.

Meanwhile, a nonprofit drug company known as Civica, Inc. is set to sell biosimilars of insulin glargine, aspart and lispro as soon as 2024 which it says will be "manufactured" in neighboring Virginia in the town of Petersburg (located just outside the state capitol of Richmond). However, we know with certainty from the company's own press release (see https://www.businesswire.com/news/home/20220303005321/en/Civica-to-Manufacture-and-Distribute-Affordable-Insulin/ for more) that facility will actually be a "fill & finish" facility whereby already-cultured biopharmaceuticals made offshore by GeneSys Biologics in Hyderabad, India will be packaged into vials and pens and then distributed across the U.S.

The Morgantown, WV facility acquired by Undbio was previously owned and operated by Mylan, which completed a merger in November 2020 with Pfizer Inc.'s Upjohn unit to form a new company known as Viatris. Viatris previously commercialized insulin biosimilars which were manufactured offshore by Biocon at a massive biotech facility in Johor, Malaysia including two biosimilars of Sanofi Lantus known as Semglee and an unbranded version called simply insulin glargine injection; the company now has a biosimilar of Novo Nordisk's Novolog/aspart in development. Semglee is made and packaged in Malaysia ready to go to drug wholesalers from the arrival at the Port of Los Angeles.

In 2022, Biocon Biologics Ltd. acquired Viatris' half of their joint venture. Former Mylan CEO Heather Bresch, who is also the daughter of West Virginia U.S. Senator Joe Manchin (D-WV), retired as CEO at the closing of the Morgantown, WV Mylan facility, hence she avoided uncomfortable confrontations about her role in making that happen. Heather Bresch was the CEO of Mylan from 2009 to 2020 (announcing her "retirement" in 2019 at the age of 50) after Mylan (under Bresch's leadership) formally announced that the company was merging with Pfizer Inc.'s Upjohn business unit and subsequently became known as Viatris. During that period, Bresch was called to testify before Congress about EpiPen price increases, and she gave unpersuasive testimony using a well-rehearsed pharma line that she and Mylan had no control over drug prices once they are sold to drug distributors. Unfortunately, Mylan immediately defied that myth when it then sold a less costly "authorized generic" version of EpiPen as an unbranded product, which proved Ms. Bresch's testimony was a falsehood, but she got a golden parachute, so she did not care.

The Morgantown, WV Mylan facility was unceremoniously closed in the name of much-needed cost-cutting since Viatris had billions in debt from Pfizer that needed to be repaid. Before the spin-off of Upjohn, Pfizer loaded Upjohn up with many of its own debts in what was little more than an accounting gimmick for Pfizer which effectively saddled the startup known as Viatris with Pfizer's debts not of Viatris' own creation. The local West Virginia news reported that the Mylan/Viatris Morgantown closure had eliminated an estimated 1,400 workers in West Virginia at the time, 850 [in 2020] of whom were members of the United Steelworkers Union.

The new Undbio 1.1 million square foot facility in Morgantown, WV will employ fewer people than the number who previously worked there; with estimates of maybe around 200 jobs (at least initially; the company has said it could create more jobs once FDA approval is secured for its first U.S. product, its once-weekly dosed basal insulin) compared to 1,400 who once worked there. The State of West Virginia never revealed exactly how much in tax incentives it paid for Undbio to occupy the facility. There are hopes that the number of employees could increase to as many as 1,000 when the facility is fully-running, although robotic automation in pharma have reduced the number of employees formerly required in the name of not only greater efficiency, but also unprecedented sanitation. Robots don't generally carry germs as humans can, and can be sanitized in-between jobs.

In 2022, West Virginia University agreed in a memorandum of understanding with Mylan to pursue other options for the Morgantown manufacturing facility on the terms that the property would be used to provide employment to the Morgantown area under WVU's ownership. That facility was then transferred to the WVU Innovation Corporation for a cost of $1. Jobs remain a hot-button issue in West Virginia, as the state has few business startups of its own, and the state has struggled to attract similar types of investments which its neighbors in Virginia, Maryland, Pennsylvania and Ohio have, due in part to the state's lack of accessibility due to its mountainous terrain, transportation consisting largely of roads, and comparatively lower levels of educational attainment relative to its neighbors. As a college town, Morgantown remains a bit of an exception to an otherwise lackluster economic landscape in one of the nation's poorer states.

On April 12, 2023, the South Korean company known as Undbio Company, Ltd. announced plans to build what it called an insulin "manufacturing" facility in the old Mylan/Viatris facility in Morgantown, WV. According to Undbio's website http://undbio.com/eng/, the company announced that it intends to build a facility to develop insulin and insulin analogues, with an expected completion date of December 2024. Local TV stations say construction was expected to begin in September or October of 2023. Based on the very short construction timeline (a little over 1 year), it would appear to be more of a facility remodeling/retooling rather than completely new construction. Undbio has another 25,000 square-foot R&D and U.S. headquarters presence in neighboring Rockville, Maryland (in addition to its 1.1 million square foot building in Morgantown, West Virginia).

However, it is something of a perilous time financially for biosimilar insulin manufacturers.

The PBM commercialization channel for insulin now appears to be dead; which means artificially-inflated margins can no longer claimed by PBMs since all three of the big insulin makers have opted out of the PBM commercialization channel; that's not inherently bad for biosimilars, although biosimilars tend not to be heavily-rebated as branded insulins are anyway. However, making insulin active pharmaceutical ingredients (APIs) offshore enables rebating to PBMs if necessary. But in March 2023, the three biggest branded insulin makers doing business in the U.S. (Eli Lilly & Co., Novo Nordisk and Sanofi) each announced significant list price reductions which were previously feuled primarily by PBM rebating, and they would instead bypass the PBMs on insulin, passing the savings onto patients instead. PBMs will have to collect rebates on a different therapeutic class of drugs.

A number of insulin biosimilars are already now pending FDA approval; but all of those are cultured in bioreactors offshore, principally in Malaysia, India and China. They may be packaged domestically at U.S.-based "fill & finish" facilties (packaging the insulin into vials and pens, and labeled for distribution) for insulin cultured in Chinese labs, but selected facilities in India and Malaysia (including Biocon's) may also handle "fill & finish" since those facilities offer FDA inspectors open access. But Undbio appears to have aspirations to sell proprietary insulin molecules, at least initially.

The Pittsburgh Business Times reported (see https://www.bizjournals.com/pittsburgh/news/2023/04/12/undo-morgantown-drugmaker-insulin.html for more) that Undbio expects to seek initial U.S. Food and Drug Administration approval for a single-shot, once-weekly insulin that it would "manufacture" in Morgantown, WV. In other words, Undbio will focus initially on developing a proprietary basal insulin for sale in the U.S. Its less certain if Undbio intends to sell biosimilar insulins as well (it has said it might); but I suspect much will likely depend on how that market unfolds in the coming years. Undbio's CEO is a man named Jun Yong-soo.

Undbio has already done extensive research on its weekly basal insulin, including testing, and manufacturing of their insulin product in South Korea and already has FDA approval to sell the product in the U.S., but only for the active pharmaceutical ingredient (API) of the product itself, but the company lacks FDA approval for its fill & finish U.S. manufacturing process which will be done in Morgantown, WV.

Basal insulin analogues tend to be used more by Type 2 diabetes insulin users. While Type 1 patients may use basal analogues, they tend to use significantly less insulin volume because their disease is due to an absolute insulin deficiency rather than insulin resistance, which is what differentiates Type 2 diabetes. Undbio won't be alone: rival Novo Nordisk already has a weekly basal insulin called Icodec in development, and the company told investors it expected to file for regulatory approval of once-weekly insulin Icodec in the U.S., the E.U. and China in the first half of 2023 (see https://www.globenewswire.com/news-release/2022/10/03/2526800/0/en/Once-weekly-insulin-icodec-demonstrates-superior-reduction-in-HbA1c-in-combination-with-a-dosing-guide-app-versus-once-daily-basal-insulin-in-people-with-type-2-diabetes-in-ONWARDS.html for more info).
 
In March 2023, Novo Nordisk as well as rivals Lilly and Sanofi all announced plans to opt out of the PBM commercialization model for insulin, instead slashing their list prices for insulin instead of raising them and rebating them all back to PBM's on the back-end. Unfortunately, rebates aggregated by PBM's are commonly not passed on to patients, which caused the U.S. insulin affordability crisis in the first place.

Biosimilar insulins tend to be commoditized, whereby the lowest prices win. That said, several biosimilar insulin makers including both California-based Amphastar Pharmaceuticals and Pennsylvania based Lannett Company (also made by offshore business units or partners) will also sell biosimilar products under their pharmacy retail partners' names as "white label" or "private label" products. Those are not for giants like CVS or Walgreens but for smaller, regional chains (according to SEC filings, each company has only biosimilars of insulin glargine and aspart in development; but neither have biosimilars for insulin lispro).

However, both Amphastar and Lannett rely on offshore manufacturing in China or France for the insulin APIs. Having a proprietary insulin similar to whatever Novo Nordisk brands Icodec as could help Undbio establish itself in the U.S. with a differentiated product around the same time Novo Nordisk expects to sell its newest "Lantus killer" (the company keeps trying but has never succeeded commercially) including its branded Tresiba and different NDC's sold as unbranded versions (both U-100 and U-200) called insulin degludec injection, although Novo Nordisk degludec has really struggled to gain a foothold against glargine because prices on glargine have fallen off a cliff.

That will be the U.S. insulin market biosimilar insulins expect to enter.

Of course, the number of biosimilars for each insulin analogue whose patents have expired will impact the odds for success. There will be no fewer than seven iterations of Sanofi's Lantus on the market, including several already being commercialized from Lilly and Biocon. For whatever reason, aspart is expected to see five biosimilars (perhaps some with different NDCs as both branded and unbranded versions), at which point innovator Novo Nordisk is likely to stop making Novolog and "retire" it as it has already done with countless insulin varieties no longer made (including the entire biosynthetic lente series consisting of semilente, ultralente which are mixed together to make lente). Since 2019, it has sold a patent-protected "new and improved" version of aspart which expedites absorption into the bloodstream with the addition of vitamin B6. Lilly Lispro (sold under the brand name Humalog), which already has one biosimilar on the market from Sanofi which is branded as Admelog, faces the fewest biosimilars, with two additional copies from Sandoz/Gan & Lee and Civica/GeneSys Biologics. Lilly has sold a cheaper unbranded version of that product for $35/vial since January 2022.

While making the APIs offshore is expected to assist biosimilar-makers, commercialization remains an untested battleground. With the floor on U.S. insulin prices falling, execution will be key to success. There, Lannett and Amphstar which already sell drugs under pharmacy partners' brand names could have a slight advantage. But Sandoz is no stranger to the U.S., and could prove very aggressive with its Chinese-made (by Gan & Lee) versions of Lantus, Novolog and Humalog. Civica, meanwhile, has already stated that the company does not define success based on unit sales, but on broader "societal impact" (which it can do as a nonprofit, cost-plus entity).

In the end, 2024 is poised to be an interesting year for the U.S. insulin market. The PBMs' shell-game with rebate aggregation is ending with big 3 insulin-makers all opting out of the PBM commercialization channel, which means both biosimilar execution and prices will be key points of differentiation for the newcomers in the space. As for South Korea-based Undbio Co., Ltd., that company will be pursuing a proprietary insulin molecule before it goes down the rabbit-hole of biosimilars. Whether it does anymore than that remain to be seen.