The Business of Diabetes: Biodel's Linjeta Is Delayed

As I WROTE previously, October 30, 2010 was the FDA's legally-mandated date to respond to Biodel, Inc.'s new drug application for Linjeta (formerly known as VIAject), which is a new, rapid-acting insulin formulation. Of course, the FDA is closed on Saturday, so the agency could have responded on Friday or waited until Monday, November 1, 2010 (the FDA chose the latter). The news was that the FDA could not approve Linjeta without at least a 2 small additional clinical trials to make up for the compromised samples which Biodel excluded from the India type 1 sample. The FDA also asked for some additional data on stability and manufacturing, although Biodel seems very confident they can adequately address those questions without delay.

Linjeta is a very interesting product because it's actually faster than an insulin analogue, yet is regular human insulin that whose genetic structure is unchanged, but the absortion profile is expedited by the addition of a few generally recognized as safe ingredients. First, they remove zinc using ethylenediaminetetraacetic acid (usually abbreviated as "EDTA") which destabilizes the hexamer of the insulin molecule; then, by adding citric acid (found naturally in citrus fruits like lemons and oranges, which you may recognize as a food ingredient that acts as natural preservative and also adds an acidic, or sour, taste to foods and soft drinks), which masks the surface charges and further destabilizes the insulin molecule so that it does not form hexamers as it would normally when injected into the skin without these modifications. On this topic, the company told investors:

"The presence of EDTA and citric acid dissociates the naturally occurring insulin hexamer and neutralizes the charges on the surface of the molecule. This enables a more rapid shift to a monomeric form upon injection compared to RHI (regular human insulin), lispro (e.g. Humalog) and aspart (e.g. Novolog/Novorapid)."


This makes humble old regular insulin, which incidentally is the ONLY kind found in the bodies of patients who do NOT have diabetes, work faster than completely man-made insulin-like molecules more popularly referred to as "insulin analogues" which do not exist anywhere in nature.

Aside from a fundamentally unique way of expediting insulin absorption into the bloodstream, the company has patented this technology and plans to apply it to a number of other medicines as well (it could also be applied to existing rapid-acting analogues to make them even faster). But the company's diabetes pipeline is not just limited to rapid-acting insulin, it also includes modifications to glucagon which is highly unstable as a liquid and therefore requires a cumbersome reconsitution process, as well as modifications to long-acting insulin formulations such as Lantus to enable these to be extended over a longer period of time, or shortened (not everyone wants or requires 24 hours of basal coverage), and even an oral insulin formulation aimed squarely at the insulin-naïve type 2 population. The basic strategy is to modify already FDA-approved drugs using patent-protected technology which hopefully brings them to market faster than would be possible a completely brand-new drug would.

The investment community has been down on the company's sell-it-ourselves (sort-of) strategy, and seems to have preferred that the company licensed the technology to third-parties instead. However, while biotechnology-manufactured insulin was leading-edge technology back in 1982, it's pretty rudimentary stuff nowadays enabling startups to rely on contract manufacturers for most of it, making capital expenditure relatively small.

Specifics on the FDA's CRL (complete response letter) on Linjeta

The FDA's so-called CRL (complete response letter) basically said that the FDA's review cycle for Linjeta was complete and that the application could not be approved in its present form without some additional clinical trials. The stock market punished Biodel's (BIOD) stock on the news, causing Biodel's shares to plunge by more than half, even though the company had stated all along that it never planned to begin marketing Linjeta until 2011 anyway. The real question now is how soon that can happen?

In essence, the clinical trials on the type 1 population were the main reason for the decision. Just some fast, relevant background: blood samples taken from some of the participants in the India type 1 trial were compromised by excessive heat before arriving at the laboratories. That constituted one leg of the trial, but studies undertaken on type 1s in the U.S. and Germany demonstrated very clear noniferiority to existing insulin therapies. The company decided to pursue it's application excluding the very small subset of the Indian leg of the type 1 trial whose blood samples were compromised following conversations with the FDA during the summer of 2009. But in it's CRL, FDA regulators determined that Biodel's decision to exclude Indian trial data for the Type 1 study "was post-hoc and therefore not sufficient for establishing conclusive evidence of efficacy," according to the company's news release. Meanwhile, "In the Type 2 trial analysis, the FDA acknowledged that non-inferiority was established in the completer population but stated that non-inferiority was not established in the intent-to-treat population because the agency did not consider a post-hoc modification of the statistical model as establishing conclusive evidence of efficacy." In other words, non-inferiority was proven on the type 2 population, but because the company had problems with the type 1 population samples and modified the statistical analysis as a result, that was a modification that rendered the FDA unable to effectively evaluate the results. The newly, more-cautious FDA said that wasn't enough to get an approval, anxious to avoid yet another meltdown as seen with various other drugs that were approved in more lenient times only to face a recall later.

At a conference call for investors today, Biodel was reluctant to speculate on exactly what will be required for the new trials. For example, it might be possible to conduct another trial of type 1s using Linjeta and combine the data with the already-conducted trials in the U.S. and Germany, but until the company meets with the reviewers at the FDA, they were unwilling to speculate on the size or cost of the new trials, or provide an estimated time of arrival for any of this until after they meet with the Food and Drug Administration to discuss the issues outlined in the CRL itself.

The FDA's CRL also had some questions on Biodel's modifications to the insulin itself. If I recall correctly, in the clinical trials, patients used U-25 insulin (as opposed to the market standard of U-100 insulin) even though the company's plan was always to sell U-100 insulin, not a lower concentration. For this reason, the FDA asked for some additional information. The FDA also requested some additional data related to stability and manufacturing. Specifically, Biodel was purchasing apoules (e.g. insulin vials) from a contract manufacturer known as Hyaluron Inc. which is pre-filled syringe maker that was acquired by a company known as Albany Molecular Research Inc. earlier this year. The FDA warned Albany Molecular Research/Hyaluron of possible violations at its manufacturing facility, which Albany Molecular actually acquired as part of its purchase of Hyaluron Inc. back in June 2010.

The FDA also asked about another contract manufacturer for Linjeta, in this case, Wockhardt Ltd., which Biodel has chosen to be it's supplier of insulin pen devices. In October 2009, Biodel executed a letter of intent with the FDA to purchase a disposable insulin pen designed by Wockhardt Ltd. for use with Linjeta. Wockhardt’s UK manufacturing facility would fill 3ml cartridges and assemble disposable pens based on a pen design based on an existing Wockhardt insulin pen already being marketed in India. The company intended to submit that pen to the FDA for review in either late 2010 or early 2011 after completing certain modifications that the company believed would improve the pen's commercial performance, although the company did not specify what those modifications might be, there was suspicion that dosage in 0.5 unit increments might be on the drawing board, making it more salable to the insulin-sensitive type 1 audience that comprises an overwhelming majority of the insulin using-population (even though the type 2 audience uses more insulin by volume). Although the insulin pen issue was not directly related to whether or not regulators could approve Linjeta itself, the company told investors all along that even if Linjeta was approved by the FDA prior to the FDA approving the disposable pen, the company did not intend to commercially launch Linjeta until a disposable pen version of the product was approved. The company was extremely confident that it could sufficiently address the concerns about it's suppliers (namely Albany Molecular Research/Hyaluron and Wockhardt Ltd.).

This in many respects is the classic tale of a biotech startup that relies so heavily on third-party contract manufacturers. If one component of the supply chain has issues, it can affect the startup's product, too. The company was unable to assuage all investor concerns during it's November 1, 2010 conference call, but is very likely to provide a much more detailed update following it's meeting with the FDA.

Biodel, Inc. As Acquisition Bait for Eli Lilly & Co. or Sanofi-Aventis?

In the interim, even though Biodel's finances are quite sound for a biotech startup, there is already speculation that Biodel could well be sold to an established pharmaceutical company. Two names on the top of those speculators' lists: Eli Lilly & Co. or Sanofi-Aventis. While both of these companies already hold a dominant postition in the global insulin market, Lilly's insulin business continues to lose market share to Novo Nordisk (and to a slightly lesser extent Sanofi-Aventis), routinely being dropped to second-tier status from many insurance company formularies or being dropped altogether. What's more, the patent for Humalog (insulin lispro rDNA origin) expires about a year from now, and the prospect for generics has never been greater now that U.S. law is forcing the FDA to outline approval procedures for follow-on biopharmaceuticals. A meeting will be held at the FDA this month to discuss this issue further and resolve some outstanding concerns.

Lilly has no long-acting analogue to compete with Sanofi-Aventis' blockbuster Lantus (insulin glargine rDNA origin), which is the world's best-selling insulin variety representing $4.0 billion in sales each year, or Novo Nordisk's Levemir (insulin detemir rDNA origin), putting the company at a competitive disadvantage when insurers choose preferred insulin brands. For Sanofi-Aventis, Biodel has some technology that could not only make it's Apidra (insulin glulisine rDNA origin) work even faster (even though the patents for Apidra don't expire for a few more years), but also make it's mega-blockbuster Lantus (whose patent expires in 2014) work better by extending the profile (longer and "peakless") and/or adjustable, thus enabling a possible patent extension for a modified version of Lantus to potentially be sold down the road assuming it gains regulatory approval.

Why would either company want to buy Biodel when the company is basically re-tooling their existing products?

Simple, as these products reach the end of their patent lifecycles, the margins on these products will either be cut by at least 75% or more, or sales will go to new generic versions (the FDA calls them "follow-on" biologic medicines). Both Lilly's Humalog as well as Sanofi-Aventis' Lantus will expire soon (Novo Nordisk's Novolog/Novorapid also expires in 2013) and with companies like Novartis Sandoz unit, Teva Pharmaceutical Industries Ltd., as well as the Pfizer/Biocon deal announced this spring eager to sell generic versions of these products, it's safe to say that the losses to the companies' bottom lines will be quite severe.

Lilly, in particular, faces what has been called a "patent cliff". Not only does Humalog's patent expire in late 2012, but Lilly also faces generic competition for 4 of its 5 top-selling products including its schizophrenia drug Zyprexa, antidepressant Cymbalta, cancer drug Gemzar and osteoporosis treatment Evista. Sales of these drugs make up about half of the company's annual revenues. What's more, several Lilly's drugs in late-stage development (including an extended-release version of the type 2 drug exenatide that Lilly, along with partners Amylin Pharmaceuticals, Inc. and Alkermes, Inc. was calling Bydureon, as well as a type 1 diabetes autoimmunity drug called teplizumab being jointly developed with Macrogenics, Inc.) have all failed to meet their primary efficacy criteria. In April 2008, John Lechleiter, Lilly’s CEO cited teplizumab as one of three most promising experimental drugs in the company's pipeline. With those products unlikely to become blockbusters soon, that means the urgency for new drugs at the Lilly is even greater, and a small acquisition like Biodel might help to fill several huge voids in it's pipeline. It may work in selected segments of the sample population, but the prospects as a billion+ dollar drug seem dashed.

Sanofi's Lantus is a $4 billion drug whose patent expires in 2014, or for all practical purposes, a little over a year from now. That product is one where anything that can extend the patent would have huge financial implications, and Sanofi-Aventis has already stated it's desire to be the leader in the insulin space. Although it's rapid-acting analogue will enjoy patent protection until June 2018, the looming question is whether the advent of generic Humalog and Novolog varieties might destroy sales growth for Apidra? After all, if insurance companies can offer not one, but two alternatives for less, we can safely assume that Apidra's sales is probably NOT safe from biosimilars/follow-ons even though it enjoys patent protection for a few more years.

Ultimately, the delay in approval on Biodel's Linjeta is regrettable, but the finances for Biodel appear quite solid, and the pipeline looks very attractive for several big pharma companies who have not invested sufficiently in their insulin businesses in preparation for their eventual patent expirations, perhaps hoping they could delay regulators for many more years to come. In the meantime, the prospects for Biodel's products remain quite solid (additional trials on a small population might enable them to re-submit the application, although details won't be available until Biodel management meets with the FDA), it seems that Biodel's shares are now cheaper than they were before. For investors, this might be a good opportunity to pick up shares for a great price, with longer-term prospects of having the company sold to a bigger rival looking more likely with each coming day.