Lexicon 3Q16 – Possible type 1 diabetes submission for SGLT-1/2 dual inhibitor sotagliflozin in early 2018; Phase 3 trials in type 2 diabetes on ClinicalTrials.gov; SGLT-1 selective inhibitor and neuropathic pain candidate added to pipeline – November 2, 2016

Lexicon provided its 3Q16 update in a call led by CEO Mr. Lonnel Coats. The press release and webcast are available on Lexicon’s website. Our report contains our top takeaways on the company’s diabetes-related pipeline, largely consisting of SGLT-1/2 dual inhibitor sotagliflozin. Most notably, management provided a potential submission timeline for sotagliflozin in type 1 diabetes for the first time and confirmed that partner Sanofi has initiated the phase 3 development program for sotagliflozin in type 2 diabetes. In terms of the early-stage pipeline, Lexicon notably added its preclinical SGLT-1 selective inhibitor and a BMS-partnered oral small molecule for neuropathic pain to its pipeline page. See below for our top five highlights from the call, followed by a transcript of relevant Q&A from the call.

Top Five Highlights

1. For the first time, Lexicon provided a potential submission timeline for SGLT-1/2 dual inhibitor for type 1 diabetes, suggesting the regulatory filing could occur in early 2018. Management suggested that the phase 3 inTandem2 trial (expected to report in 4Q16) is the one to watch and that positive results from that study would indicate sotagliflozin is a “submittable drug.”

2. The company confirmed that Sanofi’s phase 3 trials of sotagliflozin in type 2 diabetes have initiated, according to ClinicalTrials.gov. ClinicalTrials.gov currently lists two phase 3 studies of sotagliflozin in type 2 diabetes: one in drug-naïve patients (n=240, expected completion March 2018) and one on a background of metformin (n=500, expected completion March 2019). Management did not provide any guidance or details on Sanofi’s program, but continued to characterize it as “aggressive” and suggested that Sanofi would be thinking carefully about how to best leverage the impressive blood pressure reductions observed thus far for sotagliflozin.

3. Management suggested that the 200 mg dose of sotagliflozin could serve as an “anchor dose for all patients” and that the 400 mg dose may offer certain patients additional advantages for post-prandial glucose and blood pressure management. Lexicon highlighted the positive phase 2 dose-ranging inTandem4 results, reported last week, and the positive phase 3 inTandem1 results reported in September.

4. Notably, while not mentioned on the call, Lexicon has added its SGLT-1 selective inhibitor LX2761 for diabetes and an oral small molecule for the treatment of neuropathic pain, LX9211, to its pipeline page. The short description for SGLT-1 inhibitor LX2761 confirms that Lexicon has completed IND-enabling studies and is preparing to submit an IND. The description also shares that Sanofi has been granted rights of first negotiation for the future development and commercialization of the compound. Lexicon is partnered with BMS (surprising, given that it left diabetes in 2013) for LX9211 and currently undergoing IND-enabling studies.

5. On the financial front, Lexicon’s R&D spend in 3Q16 totaled $53 million, up more than double (127%) from $23 million in 3Q15. The company ended 3Q16 with $396 million in cash and investments, down from $429 million at the end of 2Q16 and up from $256 million in 3Q15.

Top Five Highlights

1. For the first time, Lexicon provided a potential submission timeline for SGLT-1/2 dual inhibitor for type 1 diabetes, suggesting the regulatory filing could occur in early 2018. Management suggested that the phase 3 inTandem2 trial (expected to report in December 2016) is the one to watch and that positive results from that study would indicate sotagliflozin is a “submittable drug.” inTandem2 will be the second of three phase 3 trials for sotagliflozin to report topline results – inTandem1 reported in September 2016 and inTandem3 is expected to report in mid-2017. The company further elaborated, “Should the inTandem2 results be very, very similar to the inTandem1 results, then we’re pretty confident we have something we could file.” That said, the company emphasized that its timeline for submission is subject to discussions with partner Sanofi and with regulatory authorities. Furthermore, the company acknowledged that regulatory agencies will “very likely” prefer submission after the inTandem3 results are available and the larger exposure study is able to further affirm the safety of sotagliflozin. The company did not directly address whether submission for type 1 diabetes would be separate from submission for type 2 diabetes, but the company appears to be currently working under the assumption that it will be able to submit for type 1 ahead of type 2. Management also did not address whether a cardiovascular outcomes study in type 1 diabetes would be required, either pre- or post-approval, despite being asked directly during Q&A – it appears the company currently does not plan to conduct a pre-approval outcomes trial at this point at least.

2. The company confirmed that Sanofi’s phase 3 trials of sotagliflozin in type 2 diabetes have initiated, according to ClinicalTrials.gov. ClinicalTrials.gov currently lists two phase 3 studies of sotagliflozin in type 2 diabetes: one in drug-naïve patients (n=240, expected completion March 2018) and one on a background of metformin (n=500, expected completion March 2019). Management did not provide any guidance or details on additional trials in Sanofi’s program, but continued to characterize it as “aggressive” and suggested that Sanofi would be thinking carefully about how to best leverage the impressive blood pressure reductions observed thus far for sotagliflozin – we expect the cardiovascular outcomes trial for sotagliflozin will be fairly ambitious given this promising early data and the demonstration of cardioprotection for SGLT-2 inhibitor empagliflozin (Lilly/BI’s Jardiance) in the EMPA-REG OUTCOME trial. We’re also curious to see if Sanofi will include a range of head-to-head trials or trials in patients with more advanced diabetes in its phase 3 program – we’re be particularly interested to see how sotagliflozin stacks up against a DPP-4 inhibitor and how it fares as an add-on to insulin or GLP-1 agonists. We imagine sotagliflozin’s clinical profile also lends itself particularly well to demonstrating value through outcomes beyond A1c – to that end, we’re glad to see that both of the phase 3 trials thus far include changes in body weight, postprandial glucose, fasting plasma glucose, and systolic blood pressure as secondary outcome measures. In addition to these outcomes, both trials will also evaluate the percentage of patients that are able to achieve an A1c target <6.5% and <7% with sotagliflozin vs. placebo. We hope these or other studies in the phase 3 program might be able to measure time-in-range through CGM and patient-reported quality of life outcomes – these two factors appeared to be huge in patient and provider testimonials we’ve heard thus far for the use of SGLT-2 selective inhibitors (such as J&J’s Invokana [canagliflozin]) in type 1 diabetes. Lexicon reaffirmed in Q&A that it plans to conduct CGM sub-studies within the type 1 phase 3 program to evaluate sotagliflozin’s impact on glycemic variability and we hope Sanofi also carefully considers similar investigations in its type 2 phase 3 program.

3. Management suggested that the 200 mg dose of sotagliflozin could serve as an “anchor dose for all patients” and that the 400 mg dose may offer certain patients additional advantages for post-prandial glucose and blood pressure management. Lexicon pointed to the positive phase 2 dose-ranging inTandem4 results, reported last week, suggesting that the 400 mg dose of sotagliflozin may have a more potent effect on postprandial glucose reduction and systolic blood pressure reduction. The company also highlighted the positive phase 3 inTandem1 results reported in September. Topline results from the phase 3 inTandem2 study as well as the phase 2 JDRF study of sotagliflozin in younger, less well-managed patients with type 1 diabetes are expected to report in December 2016. inTandem2 was expected to complete in September 2016, according to ClinicalTrials.gov and the JDRF study is expected to complete in November 2016. Whereas inTandem1 was conducted at North American sites, inTandem2 is being conducted in Europe and Israel. The final phase 3 trial for sotagliflozin in type 1 diabetes, inTandem3, has completed recruitment and is expected to report in mid 2017. This trial is about twice the size of inTandem1 or inTandem2 (n=1,400) and is a global trial with participants from North America, Europe, Israel, and other countries. In Q&A, management confirmed that, along with the three phase 3 trials, all phase 2 trials of sotagliflozin in type 1 diabetes, including inTandem4 and the JDRF study, will be included in the data submission package for sotagliflozin, though the phase 2 data will primarily be included as safety data.

4. While not mentioned on the call, Lexicon has added its SGLT-1 selective inhibitor LX2761 for diabetes and an oral small molecule for the treatment of neuropathic pain, LX9211, to its pipeline page.

• The short description for SGLT-1 inhibitor LX2761 confirms that Lexicon has completed IND-enabling studies and is preparing to submit an IND. The description also shares that Sanofi has been granted rights of first negotiation for the future development and commercialization of the compound. The last concrete update we heard on this candidate was in 1Q14 when the company stated that IND-enabling studies would be complete by the end of the year. We previously saw promising preclinical data for the candidate at ADA 2013. In 2Q16, management briefly alluded to the candidate, calling it a key compound of “big value,” despite the lack of updates since early 2014 – the company refocused its research efforts on its late-stage pipeline consisting of sotagliflozin and its carcinoid syndrome candidate telotristat ethyl in early 2014 and we’ve heard very little on this candidate since. Now that sotagliflozin is nearing regulatory submission, we’re very happy to see Lexicon revisit its early-stage diabetes pipeline.
• Lexicon is partnered with BMS for LX9211 and currently undergoing IND-enabling studies. This is the first time this candidate has appeared on the company’s pipeline page. Lexicon and BMS first announced the selection of a candidate for neuropathic pain in April 2015 as part of their neuroscience drug candidate discovery and development alliance that formed in 2003. At the time, the announcement shared that both Lexicon and BMS would undertake IND-enabling for the candidate and that BMS has the first option for exclusive clinical development and commercialization rights upon the commencement of phase 1 trials. If BMS exercises this option, Lexicon would be entitled to milestone and royalty payments. No details on the mechanism of action or on specific indications under development are available at this time, though we assume the candidate’s proposed indication encompasses diabetic neuropathy as well as other types of neuropathic pain. We’re surprised by BMS’ involvement with this candidate, given its much-publicized decision to discontinue discovery research in diabetes followed by its high-profile exit from the diabetes field in 2013. That said, at the time of these announcements, BMS emphasized its intention to pursue areas with the “highest unmet need where the company can bring the greatest value” – neuropathic pain is certainly an area of tremendous unmet need and we hope to see this candidate advanced into the clinic soon.

5. On the financial front, Lexicon’s R&D spend in 3Q16 totaled $53 million, up more than double (127%) from $23 million in 3Q15. This jump in R&D spending is of course due to the ramp-up of the phase 3 program for sotagliflozin in type 1 diabetes, for which Lexicon is responsible. The company ended 3Q16 with $396 million in cash and investments, down from $429 million at the end of 2Q16 and up from $256 million in 3Q15. Lexicon’s improved financial position as compared to last year is primarily due to a $300 million upfront payment from its deal with Sanofi.

Questions and Answers:

Q: On the inTandem3 study, can you give us a sense of how you’re thinking about the powering assumptions for the responder analysis for patients that get below A1c 7% with no DKA and no severe hypoglycemia – what is the delta there?

A: I don’t have a specific delta for you. I can offer 10% as a ballpark. In looking at the results of inTandem1, it gives us confidence that we could achieve this difference in proportion with A1c less than 7% and no severe hypoglycemia and no DKA, that primary endpoint for inTandem3.

Q: Regarding the phase 2 dose response data, you highlighted the postprandial glucose and blood pressure data suggesting that the 400 mg dose may look better. On A1c, urinary glucose, and weight, it looked like there was more of a plateau going from 200 mg to 400 mg. What are your overall thoughts on which dose looks better or if both have different advantages?

A: I can’t say right now that one dose looks better. I do think we’re starting to see a different profile. This is just one piece of the puzzle. inTandem1, 2, and 3 will be important pieces of the puzzle and they’re very well-powered for comparison of doses. This dose-ranging study was not so well-powered for comparing individual doses, however we did see the 400 mg dose had better reductions in postprandial glucose and in systolic blood pressure. If those results hold and are relevant, then I think 400 mg would offer something to patients who need to control these measures. It could be that 200 mg is a great anchor dose for all patients and the 400 mg dose would represent a good option for patients who need additional control of postprandial glucose or systolic blood pressure.

Q: Obviously, you’ve shown now that hypoglycemia and DKA really aren’t the issues that people thought they were back in August of this year. In fact, compared to one of the largest trials of an SGLT-2 inhibitor in type 1 – the canagliflozin trial – it appears that you’re better on severe hypoglycemia and DKA with sotagliflozin compared to canagliflozin. The data from the dose-ranging study also looks very good. I’m curious what else you need to show the market in order to convince them of the safety of sotagliflozin. There seems to be a disconnect between fact and what people’s perceptions are.

A: We are running one of the largest type 1 program for an oral antidiabetic agent. We will have a tremendous amount of data that we will cull through and try to understand and more importantly the one thing we’ll try to understand is DKA and the risk associated with DKA. It is of our view that DKA should not be happening and we’ll learn as much as we need to learn, both when it happens with placebo and from study drug. We will learn how to better inform about the utility of sotagliflozin in the cases where we did see it happen so that we can give proper instructions to physicians in the community on how best to use this drug when it’s in market. I think we’re in the best position to begin to inform in this space because of the size of our trials and one we have the data and begin to communicate, I think that will begin to ameliorate some of the concerns that stakeholders have about it.

Q: How sensitive is the magnitude of postprandial glucose decline to the baseline number?

A: In working with these data, we’ve seen that both fasting and postprandial glucose are highly, highly variable. In this study, that means we only reached statistical significance for the 400 mg dose. Phase 3 will give you the opportunity to get a more precise measure and achieve statistical significance with the 200 mg dose.

Q: Do you expect the differentiation between the 200 mg and 400 mg dose on postprandial glucose to hold up in the larger phase 3 studies?

A: It’s hard to say. The overall dose effect appears to be going in the right direction, between placebo and the three doses. But we can’t say that a reduction of 27 mg/dl with the 200 mg dose is statistically different from a reduction of 49 mg/dl with the 400 mg. We do have a reasonable expectation that 400 mg might be better in type 1 diabetes because it has been better in type 2 diabetes.

Q: What are some differences in baseline disease between the inTandem1 and inTandem2 trials that you might expect? Are there differences in baseline factors between the US trial and the ex-US trial in terms of A1c or other parameters?

A: The main difference we expect is that the proportion of pumps versus multiple daily injectors will be different because pumps are used more widely in North America, where inTandem1 was conducted, than they are in Europe and other countries, where inTandem2 was conducted. inTandem3 is a global study so it will include patients from North America and other countries. From what we expect, that’s probably going to be the biggest difference. In inTandem1, about 60% of participants were on pumps while 40% were on multiple daily injections. In inTandem2, we would expect a much higher proportion to be on multiple daily injections.

Q: What are your expectations on submitting an NDA for type 1 diabetes and could you comment on when the question of need for an outcomes trial might be settled?

A: When inTandem2 calls out, should it mimic the results of inTandem1, it is our belief that we’re on our way to having a submittable drug. inTandem3 is a broader study for exposure, as well as looking for the net benefit. Should we be successful the way we believe we will be in inTandem1 and inTandem2, the next benefit should be called out on a global basis and mimicked in inTandem3. Our mindset will be really focused on what happens in inTandem2. Should the results be very, very similar, then we’re pretty confident we have something we could file. With that being said, it’s always our expectation to work with our partner Sanofi and to have a conversation with the regulatory agency. There’s a good chance they’re going to want to see the inTandem3 results to affirm the overall exposure and safety is good. If that is the case, then certainly we’re be able to look at those results in the first half of next year. Nonetheless, we think that inTandem1, supported by the dose-ranging study, the incidence rates of adverse events is low, the A1c result is strong, and the strong secondary outcomes support the quality of how that A1c is achieved. Should we continue to see that, we have confidence that we have a drug that we can file. When we file will come down to the discussions we have with regulatory agencies and Sanofi and what the regulatory agency will want to see from inTandem3 before we file.

Q: If you do not need to run an outcomes trial in type 1 diabetes, is an NDA submission in 2H17 an appropriate assumption?

A: No, I think more than likely we’ll be looking at 2018. inTandem3 will call out in the middle of next year and then you have to do your work to get ready for a filing. So it’ll be closer to the end of 2017 or the beginning of 2018.

Q: There’s a couple of trials listed on ClinicalTrials.gov related to Sanofi’s phase 3 program for sotagliflozin in type 2 diabetes. Can you give us any other guidance on when other trials in the program might begin or when an outcomes trial might begin?

A: I would love to give you guidance, however that program is being run by our partner Sanofi, so they control the guidance they give. I will say that it’s a very aggressive program and we’re very pleased with how they’re thinking about it and I think we’re fairly well-aligned in how we achieve success. One of the key pieces that we’ve learned is that this blood pressure data is extraordinary and the question becomes how do we leverage that in our phase 3 trial in type 2? This will be very carefully considered by Sanofi in the additional trial work.

Q: On the postprandial glucose data from inTandem4, we haven’t seen the glycemic variability endpoints that are being evaluated in the phase 3 studies – I’m wondering what your thoughts are with respect to using those reductions as a surrogate for those endpoints.

A: Yes, I believe that we will be well-powered in phase 3 to see statistically significant differences in several of these glycemic parameters. I believe the reduction in postprandial glucose is reflecting an action of the drug to reduce overall variability in glucose. While I haven’t analyzed those data for this study yet, we analyzed it in detail in our initial phase 2 study in type 1 diabetes that we announced over two years ago and that’s exactly what we saw. There were reductions in postprandial glucose that were significant and clinically relevant, consistent with what you see here, and they were accompanied by overall reductions in the variability of glucose – for postprandial glucose, for fasting glucose, for time-in-range, and for standard measures that are being used in type 1 diabetes for glucose variability. We do think that in our larger studies, and especially our CGM sub-studies, we will have ample opportunity to demonstrate that the drug is reducing glucose variability.

Q: Can you say anything about the systolic blood pressure reductions that you’re seeing in patients in more normal blood pressure? I know some of the criticism of the SGLT-2 inhibitors is that everyone gets a blood pressure reduction due to the volume loss.

A: What we’ve seen in general in this study and in our type 2 experience is that the blood pressure reduction is predominantly in patients with elevated systolic blood pressures at baseline. Patients who have normal blood pressure at baseline have little to no reduction in either systolic or diastolic blood pressure with sotagliflozin. We saw that in type 2 and we’re seeing that as well in type 1 diabetes.

Q: Regarding the JDRF-partnered phase 2 study data, how should we think about that data in terms of filing or the label? Will the data be considered in any of those components?

A: Of course the data from the JDRF study will be relevant in terms of safety. Every study is relevant in terms of safety. What the JDRF study will give us is a good example of safety in patients with extremely high A1c. In terms of efficacy, I think it will be less likely to have that data in the label – the label will predominantly depend on data from inTandem1, 2, and 3.

Q: So to be clear, you’re planning to include that data with your regulatory filing materials?

A: Yes, we plan to include the results of all phase 2 studies in our regulatory filing material. We have an integrated summary of safety and it won’t contribute a lot of numbers, but it is an important population because it’s a young population with difficulty achieving A1c control – a population with high unmet need. So it will have a relevant part in the filing.

-- by Helen Gao and Kelly Close