Biodel F1Q13 – Positive phase 1 results for Humalog-based BIOD-238 and -250; phase 2 data for BIOD-123 in 3Q13; no glucagon updates – February 13, 2013

Biodel held its F1Q13 financial results update yesterday in a quick call led by CEO Dr. Errol De Souza. The focus was on the small (n=12), phase 1 study results for BIOD-238 and BIOD-250, combinations of Biodel's proprietary excipients with the marketed formulation of Humalog. Notably, absorption rates of BIOD-238 and BIOD-250 were significantly more rapid than that of Humalog, which translated to 35- 45% reductions in mean times to half maximal insulin concentrations (an early Tmax of 14 minutes for BIOD-238, 15 minutes for BIOD-250, and 25 minutes for Humalog; p<0.001 and p=0.001 vs. Humalog, respectively). For comparison, Linjeta had an early Tmax of just eight minutes, so these appear to be slightly slower. Encouragingly, however, the key problem with Biodel's phase 3 Linjeta, injection site tolerability, seems to have been mitigated with BIOD-250 by adding magnesium sulfate (BIOD-238 had a much higher injection site pain score than Humalog or Linjeta, and did not add magnesium sulfate). Since this phase 1 study used the marketed presentation of Humalog (i.e., taking vials of Humalog and adding Biodel’s excipients), Biodel is now working to replicate the BIOD-250 pharmacokinetic and tolerability profiles utilizing the active pharmaceutical ingredient, lispro. Efforts are also underway to achieve commercial stability, and the company hopes to meet the higher 37 degree Celsius bar for an insulin pump indication. No timelines were given on when further clinical studies might be pursued.

The company’s insulin portfolio also includes the ultra-rapid-acting BIOD-123 (recombinant human insulin (RHI) combined with Biodel’s proprietary excipients) – enrollment in the phase 2 clinical trial (n=130) is progressing according to schedule and top-line data is still expected in 3Q13 (ClinicalTrials.gov Identifier: NCT01686620). According to slides presented at yesterday’s BIO CEO & Investor Conference, Biodel believes phase 3 could be initiated in 2014. Management also highlighted BIOD-530, the company’s preclinical concentrated U-400 rapid recombinant human insulin candidate. It is at least as fast as Humalog, as well as significantly faster than the competitor, [Lilly’s] U-500 Humulin R, according to management. Biodel is still in the process of selecting a candidate, and management gave no timeline on moving it into the clinic. When asked about the regulatory environment for insulins and diabetes drugs, management emphasized that Biodel is focused on the less burdensome 505(b)(2) pathway using recombinant human insulin and Humalog, products with significant clinical experience.

Formulation work is also ongoing for Biodel’s stabilized glucagon program, per the company’s new goal as of the F4Q12 call to achieve room temperature stability. Management expects to provide updates and timelines in subsequent calls. According to the BIO CEO slides, Biodel anticipates 15 months of development between choosing a commercial formulation and filing an NDA for glucagon. We estimate this puts approval into late 2014 or 2015 at this point. As of December 31, 2012, Biodel had cash and cash equivalents of $34 million, down $5 million from F4Q12, which should take it into mid-2014.

• Biodel reported phase 1 study results (n=12) for BIOD-238 and BIOD-250, its Humalog-based ultra-rapid-acting insulin candidates. The study compared the pharmacokinetic and injection site tolerability profiles of BIOD-238 and BIOD-250 vs. Humalog in a single-center, randomized, double-blind, three-period crossover trial in 12 patients with type 1 diabetes in Australia.
  • Absorption rates of BIOD-238 and BIOD-250 were significantly more rapid than that of Humalog, which translated to 35-45% reductions in mean times to half maximal insulin concentrations (an early Tmax of 14 minutes for BIOD-238, 15 minutes for BIOD-250, and 25 minutes for Humalog; p<0.001 and p=0.001 vs. Humalog, respectively). For comparison, Linjeta had an early Tmax of just eight minutes. Tmax and AUC data (0-30 and 0-45 minutes) were also significantly better for BIOD-238 and BIOD-250 relative to Humalog.
  • BIOD-238 and BIOD-250 had a “faster-off” profile as well – the decline from peak concentration (late Tmax) was significantly shorter for both BIOD-238 and BIOD-250 compared to Humalog (124 minutes for BIOD-238, 132 minutes for BIOD- 250, and 167 minutes for Humalog; p=0.009 and p=0.016 vs. Humalog, respectively).
  • The key problem with Biodel's phase 3 Linjeta, injection site tolerability, seems to have been mitigated with BIOD-250: the mean VAS score (0-100, higher indicates worse pain) was not significantly different between BIOD-250 and Humalog(2.7 mm for BIOD-250 and 8.2 mm for Humalog). However, the VAS score for BIOD-238 was significantly higher than that associated with Humalog (24.2 mm for BIOD-238 vs.mm for Humalog; p=0.029). For comparison, the VAS score for Linjeta was 17.3 mm.
    • Management emphasized that the key to improving injection site tolerability was the addition of magnesium sulfate in BIOD-250 – while BIOD-250 added magnesium sulfate and used a lower concentration of EDTA (25% of BIOD-100/Linjeta), BIOD-238 had no magnesium sulfate and dropped the concentration of EDTA to a very low level (12.5% of BIOD-100/Linjeta).
  • Now that proof of principle has been demonstrated, Biodel is now focused on BIOD-250’s stability and use of the active pharmaceutical ingredient lispro (instead of the marketed presentation of Humalog). In a preclinical study (EASD 2010), Biodel’s excipients showed an ability to significantly speed the absorption of Novolog and Apidra; we look forward to seeing if the company pursues those opportunities as well.
  • For the complete phase 1 study results and tables, please see the slides presented at the BIO CEO & Investor Conference at: http://investor.biodel.com/common/download/download.cfm?companyid=BIOD&fileid=634713&filekey=12549D47-86E5-4574-8A79-03F842BB5812&filename=Biodel_-_BIO_CEO_11Feb_13_-_Final.pdf

Questions and Answers

Q: There has been lots of regulatory news coming out of FDA with diabetes drugs, and even with insulin. Do you have updated thoughts on the regulatory strategies for insulin drugs? Do you have any concerns? Or do you think that based on the Linjeta database, you feel confident and comfortable with the strategy going forward.

A: Great question. I think the comfort level is in the 505(b)(2) pathway. And we’re dealing with products that have had a lot of clinical experience and have been on the market for years: recombinant human insulin and Humalog. Unlike what you’ve heard, we’re not dealing with a new molecular entity (NME) in terms of moving forward. As far as we’re concerned, the path to approval has not changed given the 505(b)(2) pathway. Our comfort level is to really focus as much as we can on that pathway.

Q: Congrats on the phase 1 results. What is the next step to move BIOD-250 to further development?

A: As a reminder, BIOD-250 and BIOD-238 used commercial preparations of Humalog. We took vials of Humalog, added our excipients, and manufactured that. What we need to do now is go back and manufacture material utilizing the active pharmaceutical ingredient, lispro. Then we need to replicate what we saw with BIOD-250. The second hurdle is to make sure that we have commercial stability that would be required for a product like this. Those studies are currently underway. We’re working with one collaborator who is providing use with a source of lispro. But we’re also sourcing other source of lispro. When we’re closer to a candidate, we can update you.

Q: What are the risks in the transition from a commercial preparation to the API, especially for having similar profile?

A: We are always cautiously optimistic in terms of our ability to replicate the formulation and move forward. It depends on the hurdle. The highest hurdle is utility for this formulation in a pump setting. That hurdle requires stability to be demonstrated at 37 degrees and under shaking conditions. That is highest hurdle. The largest population that is out there, however, is the MDI population. There, we’re looking for stability at five degrees Celsius for one to two years. You’re looking at room temperature stability. So there are different hurdles for different segments of the market. We want to achieve the highest hurdle at 37 degrees.

Q: When should we know if you’re able to see that stability?

A: I cannot comment on that. It’s a little more complex than just giving a timeline. As we mentioned, we’re working with a collaborator that is providing us with lispro. Part of that collaboration agreement is a requirement to share data. That has limitations in terms of what we can disclose – I hope you can appreciate that.

Q: What are the next steps and timing for glucagon? You talked about the decision to move to room temperature stability – where are you now?

A: We are continuing to make progress, but not enough progress to announce timelines. We are looking at formulations and presentations that would achieve our target product profile. In essence, there are a variety of presentations, all of which would have drug-device combinations. Given that we have device combinations, we have to negotiate agreements related to specific devices. Those negotiations are currently underway. It limits us being able to give you timelines until we have negotiated these agreements.

Q: R&D expenses were $4.7 million for the quarter ending in December. Do you expect that rate to maintain during calendar 2013?

A: It was $4.7 million this past quarter, and $4.6 million the quarter before. That was quite a bump over $2.5-3 million previously. That’s driven by having a phase 1 trial running and an ongoing phase 2 study. We no longer have the phase 1 study running. That will take the burn down. The phase 2 is almost fully up and running, so the costs will probably go down. We don't expect that level of research burn to continue unless we announce additional things in development.

Q: So you expect after you complete phase 2, the fourth quarter expenses will bump down?

A: Yes. They will probably bump down prior to that as patients come off drug.

Q: On the concentrated insulin, BIOD-530, you have animal data. Can you talk about the profile you’re aiming for? What do you still think you need to work on before moving into clinic?

A: On the concentrated insulin program, we have very nice data that we’ve shared – both in terms of having an ultra rapid component that is at least as fast as Humalog, as well as significantly faster than the competitor, U-500 Humulin R. We also have a nice suppression of high levels of glucose. That’s happen immediately, compared to a lag with U-500 Humulin R. The PK profile looks good.

We are conducting other studies in parallel and moving to candidate selection – these include stability studies at different temperatures. But to give you a sense for final candidate selection, one of the other things we’re doing relates to the formulation. BIOD-530 has EDTA citrate. That recent study we talked about clearly showed the benefits of magnesium sulfate. So we are adding magnesium sulfate to the BIOD-530 formulation. Preliminary PK data suggests comparable PK characteristics, but we need to repeat those studies and look at the stability before we select a final candidate.

-- by Adam Brown and Kelly Close