Seeking Alpha
Profile| Send Message|
( followers)  
Previously, I wrote an article about Biodel's (NASDAQ:BIOD) VIAject (now called Linjeta), comparing its benefits to Mannkind’s (NASDAQ: MNKD) Afrezza. Linjeta is a more rapid-acting form of injectable human insulin for meal-time use by patients with Type 1 or Type 2 diabetes. Linjeta provides quicker glucose control than standard insulin, and also touts the benefits of reducing the usual weight gain and life-threatening hypoglycemic events seen in insulin users. Current insulin therapies are not delivered quickly enough to simulate the desired meal-time insulin spike. The formulation of insulin in Linjeta promotes a more rapid absorption, which more closely mirrors the effects of naturally produced insulin in non-diabetics, thereby providing more effective blood glucose control.

In the article I discussed the potential problems facing Mannkind’s Afrezza approval and compared it to VIAject/Linjeta's potential approval. In the face of several recent panel reviews and/or FDA approvals having a poor outcome, I thought it would be a good time to re-visit this idea and discuss the prospects for Linjeta’s approval in light of several developments.

It is always good to hear more than one opinion, so I wanted to consolidate several opinions on this very matter in one place to allow readers to make their own judgments. Some are analyst opinions, some are articles from trade magazines, and some are blogs, rather than emphasizing my own opinion.
I will address the three main reasons why drugs get poor panel reviews or rejections from the FDA in the context of Linjeta:
  1. Efficacy
  2. Safety
  3. Manufacturing issues
At first glance, Biodel’s Linjeta , has appears to all three of these problems going against it. Indeed, these criticisms have kept BIOD stuck in the $4 range for quite some time from its IPO price of $18.

In particular, the potential problems for Linjeta facing approval are:
  • 1) Efficacy: Given the anomalous data problems surrounding the India clinical trials (one of the 3 countries where the trails were conducted), will the FDA accept the claim of non-inferiority ?
  • 2) Safety: Why wasn’t there any extra safety studies required for Linjeta? What about the ingredients in Linjeta? What about pain at the injection site?
  • 3) Manufacturing: What about the recently disclosed FDA's Warning Letter to one of the two Manufacturer’s of Linjeta, Albany Molecular Research, Inc AMRI.

Question 1: Efficacy
Given the anomalous data problems surrounding the India clinical trials (one of the 3 countries where the trails were conducted), will the FDA accept the claim of non-inferiority ?

There are many strong arguments favoring Linjeta’s approval: In all, Linjeta has been tested in more than 884 patients who participated in Phase 1, 2 and 3 clinical trials of the drug in the United States, Germany and India. Two phase three studies, one involving 400 patients with Type 1 diabetes and the other involving 400 patients with Type 2 diabetes, were performed comparing the effects of Linjeta™ to Humulin®, the leading recombinant human insulin Biodel.

However, the main problem Linjeta faces is some anomalous data from the clinical sites in India. The data from the studies performed in the US and Germany demonstrated non-inferiority (the goal of the study) over Humulin® in both type I and type 2 diabetics. However, when some tainted samples from the study performed in India are included in the analysis for Type I diabetics, the non-inferiority endpoint does not show statistical significance. This significance, however, remains for the type 2 diabetic data. A synopsis of this data from Biodel can be found here.
Biodel determined which samples were tainted and also determined the cause to be heat damage due to improper handling of the samples in India as stated here. Biodel submitted the NDA for VIAject/Linjeta with an analysis excluding these samples, which allowed VIAject/Linjeta to reach its primary endpoints of non-inferiority to Humulin® in both type 1 and type 2 diabetics.

The big question looming, which has driven the stock price down from $18 to $4, is if the FDA will accept this exclusion of the data and approve Linjeta. As I have discussed here, I believe Linjeta will be approved on its PDUFA date of October 30, 2010.


Is excluding some data something the FDA will accept?
According to this article written by William Shannon of the GLG Group, the analysis of data Biodel performed is reasonable and that “Biodel could be a missed opportunity for the investment community”

The study was analyzed using one-way analysis of covariance (ANCOVA) adjusting for baseline HbA1c to compare the primary efficacy variable (HbA1c) to asses non-inferiority defined as the limit of the 95% confidence interval around the treatment group difference of less than or equal to 0.4%. Missing data were analyzed using the Last Observation Carried Forward protocol. For this analysis we will concentrate on the data from the Type 1 VIAject trial data due to its subgroup exclusions.”

A statistically significant (p<0.01) treatment by country interaction associated with data from India was observed and Biodel claims that the efficacy results from India are not comparable to the results from the United States and Germany, and therefore are not and should not be included in the analyses. Without India the lower limit of the confidence interval extends to -0.3, within the predefined limits for declaring non-inferiority. When the HbA1c data from India are included in the analysis the increased variability from this data extends the lower limit of the confidence interval to -0.6, and as a result would not establish non-inferiority.

The Biodel data does suggest that the India subgroup is different, and perhaps analyzing the data without this subgroup is appropriate, Specifically, HbA1C variability is more than double (0.99 versus 0.4 and 0.38) and the mean HbA1c measurements are larger by about 0.2-0.3 in the India subgroup. The increased variability and means in the India group will result in larger confidence interval which could account for the switch from concluding non-inferiority to inferior.

The task facing Biodel is convincing the FDA that their reason for dropping India is valid. This might be accomplished based on the arguments concerning variability and means, as well as doing more sophisticated statistical analyses such as hierarchical regression (mixed) modeling to adjust for differences across the three countries. In fact, these models would go a long way to reduce the size of the confidence intervals.

Leerink has maintained an Outperform rating for Biodel and a $13 price target, even in consideration of the Indian data problem as described here.

Leerlink notes that after a "missed" Phase III trial of VIAject in diabetes driven by heterogeneity of results in India, investors are assigning low probability of regulatory success. At the same time, investors have been looking for the next "surprise" FDA approval, and they believe VIAject represents a potential candidate. If Leerink's thesis is correct, BIOD will be one of the top performing biotech stocks over the next 18 months, and with a modest $100M market cap, they believe the upside opportunity is too compelling to pass up.

Heterogeneity of data from patients in India for the Type I diabetes study means they should have been excluded from the analysis, in which case the study met non-inferiority criteria. In the ITT analysis including patients from India, the study failed to meet non-inferiority, although per-protocol and completers analyses met non-inferiority. It turns out that many of the blood samples from India sat in the heat for weeks on end, which is known to increase variability in measurements.

While investors see missing an endpoint as the "kiss of death" for regulatory prospects, Leerink believes the totality of the data support VIAject approval. Ultimately, VIAject is human insulin, which is titrated to optimal effect anyways, and the drug has demonstrated a more physiologic profile relative to human insulin and rapid active insulin analogs. They fail to see the rationale in forcing BIOD to repeat clinical studies to achieve a highly predictable efficacy endpoint.

Wedbush continues to be bullish on Linjeta approval, and has stated that Biodel is prepared with supplementary data at hand, as described here:

Although FDA acceptance of an NDA filing is usually a minor event, in this case there is perceived risk that the damaged Phase 3 samples could cause the FDA to find the filing to be incomplete. Early in the Phase 3 program, samples taken from the first patients in India were accidentally destroyed by heat exposure. We believe the FDA is likely to require an additional study to fill in the gap before deciding on approval and management indicated on the fiscal 2009 call they were close to dosing the first patient in a supplemental study testing VIAject in type 1 diabetics. The company intends to have data from this trial available by the PDUFA deadline.


Indeed, two trials, one each in Type 1 and Type 2 diabetics examining the long-term safety and efficacy of Linjeta (Trial numbers NCT00875459 and NCT00875108) were completed early this year. Biodel has not publicly released this data, but has stated that it will be used in the NDA here.

Biodel is seeking U.S. Food and Drug Administration clearance to market VIAject based upon results from pharmacokinetic, pharmacodynamic and standardized meal studies, the two pivotal 6-month Phase 3 clinical trials of VIAject in patients with type 1 and type 2 diabetes, as well as results from long-term, 18-month safety extension trials for patients who completed the two pivotal Phase 3 clinical trials.

(Note that the above to Analysis(GLG Group and Wedbush) are dated before the FDA accepted the Biodel's NDA for VIAject(now called as Linjeta) in March 2010).

As a conclusion to “Question 1”:
On March 1st 2010, Biodel announced that the FDA accepted for review the VIAject NDA. The FDA expects the Prescription Drug User Fee Act (PDUFA) action date for this NDA to be October 30, 2010. I believe that FDA’s nod in accepting the NDA for review as after a pre-NDA discussion addressing the Indian data suggests that the FDA is willing to account for the anomalous data in the analysis. This is a positive sign for the FDA's belief in the primary outcome of the Phase III trail results demonstrating non-inferiority. Given the breadth of data with VIAject showing equivalency (even accounting for the anomalous Indian data), its components being GRAS, and its 505(b)(2) route, I anticipate approval on the first try.

Question 2: Safety
Why weren't there any extra safety studies or a REMS required for Linjeta? What about the ingredients in Linjeta? What about pain at the injection site?

Linjeta is seeking approval under the 505(b)(2) mechanism. This route permits companies to obtain FDA approval of new drug applications (NDAs) by relying, in part, on the agency’s findings for a previously approved drug. Under section 505(b)(2) guidelines, an NDA approval can be obtained for a new drug without conducting the full complement of safety and efficacy trials. The 505(b)(2) mechanism allows a limited change to a previously approved product (here it is insulin), such as changes in dosage form, strength, formulation, dosing regimen or route of administration; a new combination of existing products, including substitution of an active ingredient; or a modified active ingredient (i.e. – salt, chelate, ester, complex, etc.) (Linjeta falls under this last category).

The 505(b)(2) regulatory pathway avoids repetitive preclinical and clinical trials and the requirements needed for a successful application are better defined than other types of submissions. In the case of Linjeta, the endpoint required is simply equivalency, or non-inferiority. Although Viaject also showed statistically significant reductions in hypoglygemic episodes and improved weight control, this is not necessary for approval. As noted in 21 CFR 314.54, the “application need contain only that information needed to support the proposed modification(s) of the listed drug.” As far as safety, it is a non-issue: The ingredients in Linjeta are commercially available recombinant human insulin and their proprietary formulation of ingredients, all of which are Generally Regarded As Safe (GRAS) by the FDA. The safety profile of insulin is well established, and beyond this Biodel has data demonstrating that Linjeta is in fact safer than regular insulin as it reduced hypoglycemic events in half.

One criticism of VIAject/Linjeta that arose during the early trials was slight pain at the injection site. This occurred with the 25U/ml pH 4 formula which was acidic (think orange juice on a canker sore). After adjusting VIAject to a higher concentration (100U/ml) at a neutral pH, it reduced the volume of administration and eliminated the pain. Biodel performed clinical testing to compare the ability of pH-neutral 100 IU/ml VIAject(NYSE:R), pH-4 25 IU/ml VIAject and insulin lispro to measure the bioequivalence of the two forms of VIAject and compare the pharmacokinetic and pharmacodynamic characteristics of pH-neutral U-100 VIAject(R) and insulin lispro. These studies established bioequivalence of the 100IU formulation to the 25IU formulation used in Phase III, and submitted the NDA seeking approval for the 100U/ml neutral pH formulation as described here.

There seems to be some confusion about Biodel’s plans for Viaject/Linjeta. Biodel has always made it clear that they are seeking approval to market Linjeta in the United States as a 100 IU/cc, pH7 (neutral) injectable liquid, in 10 ml vials and 3 ml pen cartridges. jIn these studies, which were submitted with the NDA, Biodel demonstrated bioequivalence of the two Linjeta formulations and also demonstrated more rapid absorption and onset of action of pH-neutral formulation than insulin lispro.

In layman’s terms, the total amount of insulin per injection was the same, but in the clinical trial the IU25 insulin was contained in 4 times the amount of liquid at ph 4. What the company did was simply make a standard IU 100 formulation, and buffered it a neutral pH, which eliminated the pain. Since the dose of insulin is the same, the efficacy data from the phase III trials are valid.

On this subject, Wedbush analysts, who also predict an approval for Linjeta(VIAject), had predicted here well ah:

Management submitted the NDA for VIAject. After previous discussions with The Center for Drug Evaluation and Research (CDER) of the FDA, the NDA submission included the freezable, 100 IU, pH 7 formulation of VIAject. This cartridge formulation may be launched in already approved pens as Biodel’s pen will be submitted after The Center for Devices and Radiological Health (CDRH) has discussed procedure with them. We believe the FDA is likely to accept the VIAject NDA filing in Q1 2010.


In conclusion to question two, persons familiar with the process know that successful 505(b)(2) application depends on good communication between the FDA and the applicant to determine what additional information is needed to support the proposed change of the previously approved drug. Input from experts, along with early discussions with the FDA to understand what data is necessary. According to guidelines, before filing an application, a sponsor should submit a plan to the appropriate division of the Center for Drug Evaluation and Research (CDER) to identify the types of studies that should be conducted, and highlight parts of the proposed product that rely on the FDA’s finding for the previously approved drug. Biodel has done all of this, and the FDA has accepted its application on March 1st, 2010. In other words, there are no outstanding issues that the FDA has communicated that would prevent the application from being denied based on insufficient information. The application contains “all the information needed to support the proposed modification(s) of the listed drug”, as required.

Question 3: Manufacturing
What about the recently disclosed FDA's Warning Letter to one of the two Manufacturer’s of Linjeta, Albany Molecular Research, Inc AMRI?

In an FDA letter to AMRI Burlington, Inc. (formerly Hyaluron, Inc., who Biodel made arrangements to have Linjeta manufactured) dated 8/17/10, an FDA inspection resulted in a warning letter shown here. Investors (over)reacted to this letter, fearing it would delay approval. First of all, all this is related to a completely different manufacturing process of a completely different drug on the same site, not Linjeta/VIAject. Read the short letter linked above- It is simply a request from the FDA to fix violations noted in an inspection.

The action requested by the FDA was:

Within fifteen working days of receipt of this letter, please notify this office in writing of the specific steps that you have taken to correct violations. Include an explanation of each step being taken to prevent the recurrence of violations and copies of supporting documentation.

It is easy to see that AMRI has to do is explain how it intends to address these issues to remain compliant. It is very unlikely that this will play out long enough to delay the PDUFA decision. In the unlikely event that it does, this still won’t affect Biodel’s timeline for Linjeta marketing. Biodel has made it clear that it does not intend to commercially launch Linjeta until a disposable pen version of the product is approved. As stated before, they intend to submit this pen to the FDA for review in early 2011. Biodel is working with Wockhardt Ltd., its second contract manufacturer producing finished Linjeta product, to develop the disposable pen for use with Linjeta and that it intends to conduct the majority of its future Linjeta manufacturing with Wockhardt.

Biodel has stated that, "it does not believe that the letter will have any long-term impact on its commercialization plans for Linjeta".

AMRI's Response to FDA: When contacted on Sept 8th, 2010, Albany Molecular Research, Inc. stated that it intends to respond to the FDA on Sept 8th 2010, in regard to the recent warning letter it received, dated August 17, 2010, from the U.S. Food and Drug Administration (the "FDA") in connection with the FDA's inspection of AMRI's pharmaceutical manufacturing facility located in Burlington, Massachusetts. AMRI’s Director of Investor Relations, Peter Jerome stated “We intend to submit our response to the FDA today, September 8th."

A final question investors should take to heart:
What if Linjeta gets rejected or delayed for some unforeseen reason?
Don’t take home the message that there is no risk in an investment in BIOD as is the case with any FDA approval . I will discuss the risk/reward at length at the end of this article.

If things go bad for Biodel with the Linjeta PDUFA the first try, leadership is key for driving the company’s efforts for eventual Linjeta approval and success in other areas. Dr. Solomon Steiner, who was a key inventor of the Technosphere® technology used to create Mannkind’s (MNKD) Afrezza and Biodel’s VIAject/Linjeta, co-founded Biodel in 2003 and is a brilliant scientist. But scientists don’t always make the best business decisions (an example of this is allowing pain at the injection site through the phase III trials before reformulating VIAject) and if Linjeta doesn’t get approved the first try, Biodel needs to have the right people in place. They have done this with the recent appointments of Dr. Errol De Souza as president and chief executive officer and Dr. Charles Sanders as board chairman. Dr. Solomon Steiner, became the company’s chief scientific officer, devoting his full-time attention to the development of Biodel’s product candidates to treat diabetes, and will remain a member of the board. Dr. Sanders is former chairman and CEO of Glaxo and was appointed to Biodel’s board in 2006. Dr. De Souza was previously president and CEO of Synaptic Pharmaceuticals, president and CEO of Archemix Corp., co-founder and chief scientific officer of Neurocrine Biosciences, senior vice president and U.S. site head of drug innovation and approval (R&D) at Aventis and director of CNS diseases research at DuPont Merck; he is currently a board member of several biopharmaceutical companies.

Leadership is obviously important when considering the future of the company. Investors should ask themselves: What will happen if Linjeta is not approved the first try? How can the product be expanded into other indications? What else does Biodel have going for them? In my opinion, the management team in place is considering these options and is making obvious steps towards marketing (hence the name change to Linjeta). If the trial data from type I diabetic is found to be faulty, more trials are not necessarily needed. Biodel could always seek approval for Linjeta in type 2 diabetic data, as there was no sample contamination in this data set. A synopsis of this data from Biodel can be found here.

Although the Linjeta approval is pivotal for Biodel, in contrast to many biotech companies, Biodel has several viable back up options as shown in their pipeline.

Besides remedying any problems they may face if the first Linjeta approval is denied, they have an active clinical and preclinical program investigating other ways to control blood sugar:

  • Use of Linjeta in insulin pumps (Clinical trial NCT01067118)
  • An “adjustable basal” insulin which could have its duration lengthened or shortened would allow physicians to personalize insulin doses to the patient’s specific needs.
  • A “smart basal” insulin which releases insulin proportionally to subcutaneous glucose concentration and, as a result, automatically adjusts to unanticipated changes in patients’ insulin needs, such as exercise or fever, to maintain a more normal glycemic range.
  • “Rescue glucagon”, a pH 7 (neutral), clear-liquid, stabilized form of glucagon as a pre-mixed “rescue” drug that can be delivered in an auto-injector device to treat insulin overdoses. Glucagon is a highly unstable compound that helps regulate blood glucose levels by signaling the liver to convert glycogen into glucose.
  • -VIAtab™, an oral sublingual insulin tablet that dissolves in minutes when placed under the tongue. It is being developed for use as an insulin supplement for early-stage Type 2 diabetics by mimicking first-phase insulin-release signaling, reducing the production of glucose by the liver and maintaining normal glucose levels

Even in the event of Linjeta rejection in October, Biodel is far from “done”. First of all, Linjeta will eventually gain approval, even if it requires additional efficacy data. The other ongoing development programs will gain more traction and add more value to Biodel. In this worst case scenario, dilutive funding will be required, but the long term future of Biodel is still promising.

It is wise to consider all scenarios, good and bad, but of course the big question investors are thinking about is: What is a fair valuation for BIOD if Linjeta is approved? In summary, an average of about $25 per share.

A comprehensive analysis can be found here at Biorunup.com. This conservative analysis, based on a (very conservative) estimated weighted average of approximately 10% of the $29B insulin market, estimates gross sales of $575 million and a profit margin of 6.5%, for a net profit of $37.5 million. Based on a price/earnings ratio of 20, an estimated market capitalization is about $750 million. With current 26 million shares outstanding, this yields a conservative $23-$27 per share.

According to public filings, approximately 35% of all BIOD shares are held by insiders, leaving only about 18 million trading shares. However, public information shows that about 45% of all BIOD shares are still held by institutions and beneficial owners as of August. Assuming institutions are not selling, this leaves only about 8 million free trading shares, ALMOST HALF OF WHICH ARE HELD SHORT. At average volume, this is an astounding 10 days to cover.

The big question is: Who is selling all these shares short? Perhaps hedge funds and retail traders who shorted have not covered since shorting the May highs. Perhaps the long-term institutional investors are hedging their positions with shorts. This is unclear. What is clear, however, that these conditions are perfect for a massive short squeeze if FDA approval occurs. This would likely send the share price far above the $25 valuation upon approval.

In my mind, this skews the risk/reward dramatically in favor of a long investment. The $25 share price upon approval could easily reach $40-$50 in the face of this short squeeze. Let’s continue to be conservative and assume that upon approval $25 is likely to be seen. To fairly assess the risk/reward, though, we need to look at the downside:
The downside is typical for a small biotech company facing a pivotal FDA approval. If rejected, BIOD stock would likely see a 50% or greater haircut. Although the event is binary or digital (meaning an approval sends the price soaring and a rejection sends it plummeting).

To calculate a simple risk/ reward ratio here at $4: Risk=$2 or 50% loss; Reward= $25 or 625% gain= 625/50=12.5. If we calculate these numbers using the fair value of $7.50 (calulated in the article at Biorunup.com) heading into PDUFA date, the numbers come out to be 4.5 { [($7.50/$2)= 74% loss vs ($25/$7.50)= 334% gain]= 334/74=4.5}. Next, multiply these numbers by the likelihood of Linjeta approval in your mind. I feel it is 2 in 3, or 66%. In this case, my risk/reward factor is 8.25 (0.66*12.5) when BIOD is at $4 and 2.97 (0.66*4.5) when BIOD is at $7.50. This is like getting 8:1 to 3:1 odds on your bet, respectively. In the case of cash-strapped biotech companies facing pivotal FDA decision like BIOD, if you are not getting better than 2:1 odds it is not a worthwhile bet. If you give the odds of Linjeta approval as low as 1 in 3, BIOD is a good value anywhere below $7.50.

Obviously these are all back-of-the-envelope calculations. I perform considerable due diligence in the company and product to assign a likelihood of FDA approval. Then I use the risk/reward factors calculated above to contemplate whether or not to invest and how much to invest. Adjust your numbers according to your risk comfort. This is an aggressive strategy, and of course, hedges using out-of the-money puts are valuable here as well.

Using this strategy, I have determined that even if you give Linjeta a very conservative 1 in 3 chances of approval, I see a risk proportional investment in BIOD at prices between $4 and $7.50 heading into PDUFA decision as very worthwhile. For a perhaps more respected opinion on this matter, read this TheStreet.com's recent article which predicts a double for BIOD stock before the PDUFA.

Finally, several professional analysts share my thinking: These analysts also reiterated buys for BIOD, even after release of the Indian trial results:
  • 25-Mar-09 Reiterated Wedbush Morgan Buy $10 → $16
  • 25-Mar-09 Reiterated Ladenburg Thalmann Buy $9.50 → $14
  • 20-Nov-08 Reiterated Cantor Fitzgerald Buy $33 → $15

The PDUFA date for Biodel Inc’s Linjeta (formerly called VIAject) is October 30, 2010.
More analyst targets, since the release of the Indian trial data, are listed here:

Ladenburg Thalmann — Buy 03/25/2009 14.00
Wedbush Morgan — Buy 03/25/2009 16.00
Ladenburg Thalmann — Buy 12/11/2008 9.50
Cantor Fitzgerald — Buy 11/20/2008 15.00
Argus Buy Sell 09/19/2008 —
Oppenheimer — Perform 09/11/2008 —
Ladenburg Thalmann — Buy 09/08/2008 11.00


Disclosure: Long BIOD

Source: Biodel's Linjeta Should Overcome the Usual FDA Approval Hurdles