Lexicon Pharmaceuticals' Management Presents at 31st Annual J.P. Morgan Healthcare Conference (Transcript)

Lexicon Pharmaceuticals, Inc. (NASDAQ:LXRX)

31st Annual J.P. Morgan Healthcare Conference Call

January 9, 2013 12:30 ET


Brian Zambrowicz - Executive Vice President and Chief Scientific Officer


Matt Lowe - J.P. Morgan

Matt Lowe - J.P. Morgan

Welcome everybody. My name is Matt Lowe, the J.P. Morgan biotech team. Our next company to present is Lexicon Pharmaceuticals. You will have a breakout down the corridor in Sussex room afterwards. If I could just ask you all to please turn your cell phones off, and right now, I’ll hand over to Brian Zambrowicz. Thank you.

Brian Zambrowicz

Good morning. I want to thank the organizers for the opportunity to present at the conference. I will be making forward-looking statements during my presentation and I will refer to you to our SEC filings for a more thorough description of the risk factors associated with our business.

Lexicon Pharmaceuticals is a drug discovery company. We are founded in 1995 as a genomics company, based on the platform technology of using mouse knockouts to understand gene function and identify novel targets for drug discovery. And we have remained true to that platform and built up on it. We added a chemistry, capability with the acquisition of Coelacanth Corporation in 2001 that allowed us to begin to discover small molecule drugs to the targets we are identifying and over the course of time, since then, it’s allowed us to develop a pipeline of small molecule drugs all discovered and developed internally and all against the novel targets based on our platform technology. All the compounds are owned by Lexicon at this point, but we look forward to further developing and commercializing some of these compounds and partnerships and selectively choosing some to develop and commercialize ourselves.

This is our pipeline and I’ll be discussing leading-edge of that pipeline. I’ll be describing LX4211, our diabetes compound that will go into Phase 3 development around mid-year. I’ll also describe telotristat etiprate, which is currently in Phase 3 clinical trial for carcinoid syndrome. And it’s in a separate ongoing proof-of-concept Phase 2 clinical trial on ulcerative colitis that will read out around mid-year. And finally, I’ll describe LX1033, another small molecule compound for IBS-d. It’s currently in a large dose-ranging study that will read out above mid-year, and we think will prepare us well for a Phase 3 initiation sometime early next year.

So, beginning with diabetes, LX4211, it has a dual mechanism of action and it inhibits two glucose transporters, SGLT2 and SGLT1. People are very familiar with SGLT2 and the class of drugs that inhibit SGLT2. When you inhibit that target in the kidney, you release glucose in the urine. And what makes LX4211 unique is its inhibition of SGLT1, which is the major transporter involved in glucose uptake from the gastrointestinal tract. When you block it, you block up, block the glucose uptake after a meal. And in addition, by delaying glucose absorption from the gastrointestinal tract, you trigger the release of beneficial peptides like GLP-1 and PYY. So, essentially it’s acting also as a small molecule GLP-1 secretagogue. And this second SGLT1 mechanism of action is providing the differentiating characteristics of LX4211 that I’ll describe to.

Again, we are familiar with the crowded space around SGLT2 inhibition and I give this as slide as a landscape view, but I’ll have you focus on the columns that indicate whether there is inhibition of SGLT1 and SGLT2, and you can see that LX4211 is unique in being a dual inhibitor. All the others are highly selective SGLT2 inhibitors with the exception of the Phase 1 compound from GSK, which is a selective SGLT1 inhibitor. That’s an interesting compound, because I think it supports the concept that SGLT1 alone is an important target for type 2 diabetes, but also they have published some Phase 1 data and some animal data. And I think that data supports what we have described for the pharmacology of SGLT1 inhibition.

Last year, we completed our Phase 2b dose-ranging study with LX4211. We did it at about 50 centers in the United States. It was in patients that were on stable dose metformin, but failing. So, it’s very much a real world type setting. We tested four doses of LX4211 and placebo. There were 299 patients enrolled in the study and the primary endpoint was the change in hemoglobin A1c at week 12 of dosing with a variety of other secondary endpoints.

And to look at that data, I think we will start with the hemoglobin A1c. This is the time course of that HbA1c reduction, green being placebo in these slides and then doses moving up from gold, blue, red and black are increasing dose. What we saw as a very nice dose response with respect to hemoglobin A1c reduction and if you look at the three highest doses in blue, red and black, you can see that we haven’t saturated that effect at week 12 suggesting as we go into longer dosing in Phase 3, we can anticipate an additional benefit. Also I would point out that the high dose was clearly most effective that was 400 milligrams once-daily given as a tablet.

And I want to speak to the magnitude of the effect, there was a robust reduction at week 12, 0.95% reduction, 0.86% placebo subtracted, and that compares very favorably with any other anti-diabetic agent at the week 12 time point. We also saw a nice dose response, and over time, when we looked at fasting plasma glucose, and again it was pretty clear that 400 milligrams once-daily was the best dose. I think the data got quite interesting when we began to look at the effect of LX4211 on SGLT2. So, one of the measures we did was glucose to creatinine ratio that was a spot check done on all patients in the study. And in green, you can see that with placebo, there is obviously very little glucose excreted in the urine, but as you inhibit SGLT2 with increasing dose, you start to see an increase in that ratio. What you see pretty clearly is in blue, that’s a 200 milligrams once-daily dose. We maximized that glucose to creatinine ratio.

And pushing the dose further in red and black did not increase that ratio suggesting that any benefit that was being achieved with LX4211 going above the 200 milligram dose must be due to SGLT1. So, I want to go back to that HbA1c, because I think for the first time in the study, we have demonstrated that the effect on SGLT1 is clinically meaningful, because if you look at the blue, the 200 milligram once-daily and the black the 400 milligram once-daily, that’s a very robust additional impact on A1c. It’s an additional 0.43% reduction that must be attributed to SGLT1.

Likewise, if you look at the fasting plasma glucose in blue and black going from 200 to 400 milligram is again a robust effect that must be that additional effect being due to SGLT1. Now, we also had another way of measuring the effect on SGLT2 which was looking at the total 24-hour urinary glucose excretion, that’s shown here. And if we focus particularly on the burgundy, that’s the week 12 data. And what we see is again in fitting perfectly with what I described to you with the previous measure, we maximized our effect at the 200 milligram dose group, and if anything, there was a reduction in total urinary glucose excretion as we push the dose. And very importantly, it supports this concept that, that additional benefit we saw in A1c was due to SGLT1, but also very importantly here, our total urinary glucose excretion with LX4211 is much less than has been reported for the selective SGLT2 inhibitors. And this I think will become very important as far as safety is concerned.

In addition to the effects on blood glucose, we also saw a nice weight reduction pretty much in three high dose groups giving about 2 kilograms of weight loss. We also saw a nice effect on blood pressure. Systolic blood pressure is shown here with a nice dose response and the high dose producing about a 6 millimeter mercury drop in systolic blood pressure.

And then the safety profile was very favorable. If you look just at the first row on the AEs, the subjects with at least one AE and you focus on the two final columns, the 400 milligram high dose and placebo, you can see it was a very similar rate 57.6% on high dose, 66.7% on placebo, but we are particularly interested in two areas of AEs. One was the gastrointestinal side effects, because theoretically people have been concerned that inhibition of SGLT1 would lead to GI side effects. And you can see quite clearly that it really doesn’t look much different than placebo. If you go over to again to the high dose column, there was a 22% rate of GI disorders on high dose, 20% on placebo. And perhaps most importantly as we have described, people were concerned about diarrhea. There was really no signal to speak of with four cases on placebo, five cases on high dose of LX4211. So, it looks like we have added the SGLT1 mechanism of action without adding any additional baggage as far as adverse events.

And then we are also interested in the SGLT2 based adverse events of genitourinary infections. And here we see that there really was no signal for urinary tract infections with one on high dose, one each on two of the lower doses, and one on placebo. And a small signal, an encouraging signal for genital tract infections, three on high-dose, zero on placebo. And this is at this stage looks very attractive relative to the selective SGLT2 data. And we believe mechanistically, it makes sense due to the decreased urine glucose excretion. And this is because of the SGLT1 mechanism, releasing glucose in the urine is dependent not only on inhibition of SGLT2, but the amount of glucose that’s in the blood. And because of the profound effect of SGLT1 inhibition on blood glucose levels after a meal, there is just much less glucose to be filtered and dumped in the urine.

Alright. So, to summarize, we had very robust A1c changes in patients on stable dose metformin. The urinary glucose excretion was low. So, we were achieving very robust effects, which with lower urine glucose excretion. We had significant body weight and blood pressure reduction, and the AE profile at this stage, it looks very favorable. So, we are also running a couple other studies currently. One is an ongoing renal impairment study. The rationale for that is that, about 40% of patients with type 2 diabetes will eventually suffer kidney failure. There are many diabetes medications that are contraindicated for this population. And although they will lose benefit from SGLT2 inhibition as they lose kidney function, they should not lose SGLT1 benefit. And so this should really differentiate us from an SGLT2 selective compound. And that study is enrolling well and we’ll have data around mid-year. We have also initiated a type 1 study. Again, this SGLT1 mechanism of profound effects on post-meal blood glucose levels, we think it could impact and benefit type 1 patients by giving them better glycemic control, but doing so with less insulin, thereby reducing also potential hypoglycemic events and weight gain. That study we are just getting sites up and running.

So, we have, we think a differentiated agent. We have robust blood glucose control. We think that we can differentiate straight up in type 2 diabetes both on efficacy as well as potentially on safety, particularly as it relates to the genitourinary tract infections. We also believe we have a unique ability differentiating the renal impair and type 1 diabetics as I described. And as though I haven’t described it to you today, we have shown it in a clinical trial that we have a unique synergy with DPP-4 inhibitors, because we elevate the release of GLP-1 with LX4211 and DPP-4 inhibitors prevent its inactivation. We get a synergistic elevation of GLP-1 in combination and we’ll be pursuing that further as well. We have done everything required and we’re prepared for Phase 3 initiation around mid-year.

I am going to move on to telotristat etiprate, our compound for carcinoid syndrome. It’s an inhibitor of the rate-limiting enzyme in serotonin synthesis tryptophan hydroxylase. And carcinoid syndrome is caused by tumors that originate in the small bowel typically, but once they metastasize to the liver, there they cause severe GI side effects, because these two tumors are producing very large amounts of serotonin. Telotristat gives systemic exposure, so it can get to the tumors, but it does not cross the blood brain barrier. It has fast track and orphan status from the FDA and orphan designation from the EMA. It’s currently in Phase 3 single registrational study and it’s also in a proof-of-concept study in ulcerative colitis. Standard of care in carcinoid syndrome is somatostatin analogs. However, about 40% of carcinoid patients at any one time are failing and their GI symptoms are breaking through. And that’s a single snapshot in time if you follow any patient over time, all will fail and their symptoms will break through.

So, I am going to focus on the most recent data we presented late last year, which was open label study we did with 12 weeks of dosing and the importance of this study is that it fits very closely with our Phase 3 design. We have had around the Phase 2 meeting and came to agreement with the FDA on a 12-week endpoint for efficacy. So, I think this study gives a pretty good indication in the same population, same dose, and same length of dosing of what we can expect in Phase 3. We had 15 patients enrolled in the study. Every patient was dose escalated in two-week timeframes from 150 milligrams of telotristat given three times daily, then up to 250, 350, and finally 500 milligrams three times daily staying at the high dose for the last six weeks of dosing.

So, going to that data then, first looking at the effect on bowel movement frequency, you can see that they were starting on average with about six bowel movements per day, but by end of study, there was a 43.5% reduction. Significant reduction already at week two and that effect continued to grow over the course of the 12 weeks of dosing.

And if you look at the individual data and the benefit each patient was getting, you can see that the majority of patients were getting quite strong effects. All, but two of the patients were having a minimum of 30% reduction in bowel movements per day with many patients getting much more robust effects. We also saw a very nice improvement in the stool form, stool consistency. This is on a 6-point scale. And what we saw is a significant reduction or a significant improvement in this stool consistency score already at week two, and again like bowel movement frequency of building in the effect to where we had about a 20% reduction at week 12.

One of the breakdown products of serotonin is 5-HIAA. So, if you are inhibiting the target of tryptophan hydroxylase properly, you should reduce the biomarker. We saw a robust reduction in the biomarker over the course of this study with 72% reduction in the biomarker by week 12. We also somewhat surprising we saw an improvement in flushing, which is not traditionally been thought to be primarily serotonin mediated, but we saw about a reduction in one flush per day starting in the first two weeks and being maintained throughout the course of treatment. We also saw a reduction in abdominal pain and discomfort, although significant only at week seven and eight, it was trending down throughout the whole course of this study.

And finally, if you look at the final row on that table, you can see that by the end of the study for the 12 patients who did report, 9 of 12 or 75% were reporting adequate relief of their GI symptoms of carcinoid syndrome. So, we are very pleased with this study. We had no dose-limiting toxicity. Telotristat etiprate was very well tolerated. And with most of the adverse events being mild-to-moderate in intensity, we did see nice improvements in clinical symptoms, and 11 of the 13 patients who had the option to go into long-term extension did. We think this data clearly supports our move into Phase 3 with this program.

As I mentioned, we also have this ulcerative colitis. So, the carcinoid syndrome just a little bit more on that Phase 3 study, it is a single pivotal study 12-week endpoint as I mentioned, we will be testing two doses of telotristat and placebo and it will be about 100 patients with a primary endpoint of reduction in number of bowel movements, where we had a very robust effect in our Phase 2 studies. And key secondary endpoints being stool consistency and the biomarker reduction. That trial is underway. We are currently initiating many sites.

And we have this ulcerative colitis study ongoing. It’s a 60-patient proof-of-concept study. I won’t go into it today. We have a quite a bit of preclinical data supporting our mechanism for IBD. And we have also collaborated with a couple of academic labs to reproduce and provide additional data and this is not only have an effect on the GI, but clearly, we believe this mechanism is providing an immune modulating effect. We will report out on that study about mid-year.

And finally, I want to talk about LX1033, our IBS-d program. This compound is different than telotristat, which I just described to you, but it also has the same target, shares the same target of tryptophan hydroxylase. The real difference in this compound is it’s locally acting in the GI. We get very little systemic exposure and we are able to do that, because the GI produces about 95% of the body’s serotonin, but the cells that produce that are sitting at the lumen of the GI. And so to maximize our safety, we designed and developed LX1033 to give low systemic exposure. And it’s for the diarrhea-predominant IBS. We have a large dose-ranging study underway. I will describe that to you.

There really is not much available at this time in the IBS-d area, but there is clearly growing interest in the IBS indication. I think pharma has become interested again particularly with the approval of linaclotide and I’ll emphasize linaclotide is complementary to our agent it’s for IBS-c, the constipation form rather than IBS-d for which we are developing LX1033. We have done proof-of-concept on the mechanism with an earlier stage compound that was designated LX1031. We have published that data in gastroenterology, but our primary endpoint at the time was adequate relief, it was before the new FDA guidelines on IBS, but I will describe briefly that data.

In this graph, red is the high dose of 1033, blue is low dose, and green is placebo. And we are looking on the top of the graph at the effect on adequate relief, we saw significant improvement in patients reporting adequate relief on high dose of 1033, no real effect on low dose in blue, and typical placebo response in an IBS study getting up in the range of 40% over the course of the four weeks of dosing. There were 150 patients in this study equal distribution (per arms). And on the bottom, we are looking at this 5-HIAA biomarker and we can see that by week four, there is a very robust reduction in the biomarker at week four in the high dose arm, not much, no effect in the low dose arm or placebo.

And at week four, where there was a significant correlation between the biomarker reduction and those patients that had a clinical benefit. And we explored that further then, because we wanted to look at the biomarker reduction and how closely it correlated with efficacy. So, we went back and said let’s make a cutoff in the high dose arm. We know that the low dose didn’t get the 15% reduction and it wasn’t effective. So, let’s look at the patients on high dose that had a 15% or greater reduction in the biomarker and what was their response. And that’s shown here first, this is the data I have already shown you for the total population, and now blue, the biomarker responders compared to orange, the biomarker non-responders. We can see that biomarker responders clearly had a much more robust response being over 70% responders. And I think this gives us confidence that the biomarker could help us to maximize the window between a real signal on active and placebo.

We did the same thing with stool consistency. Green is the placebo for the whole population and red is the whole population. We clearly had a significant effect on the whole population. It’s clearly a primary effect of our drug, but again if you look at the biomarker responders in blue versus non-responders in orange, they had a much more robust response to drug.

And finally, we looked at every other primary and second endpoint in the study. And in every case with the exception of stool frequency in blue, the biomarker responders had a more robust response. We of course now have new guidance from the FDA on IBS-d. And with the new entry criteria and the new dual endpoint of pain and stool consistency, we have been able to go back to this study and reanalyze it with those new guidelines. And we like those guidelines a lot, because what we have seen with this mechanism and with this data, sorry, is that the new dual endpoint doesn’t diminish the signal on active, but it almost completely eliminates the placebo response signal, so that it produces a very dramatic separation between treatment effect and placebo. So, we have good deal of confidence going forward based on that new guidance.

And so the issue with 1031 is that it required a high and frequent dosing and we did moved to a second generation compound LX1033 also locally acting. In fact, even lower systemic exposure than 1031, but 10 times more potent. And it allowed us to get the same biomarker effect that was associated with clinical efficacy with 1031 with much lower and less frequent dosing. And so that’s what we have been moving forward. And as I described to you, we are in this very large dose-ranging study, its 360 patients. It will read out at mid-year. And we are testing three doses of 1033 and placebo with a four-week treatment period. And our primary endpoint is the change in stool consistency, but we will be of course tracking and measuring pain and the combination endpoint in this study. And we will be tracking the biomarker correlation with those effects very closely to clearly define what cutoffs we would be interested in looking at in further clinical trials.

So, we are moving into Phase 3. Our carcinoid syndrome has already initiated Phase 3. Our type 2 diabetes program will initiate around mid-year. And given positive data for LX1033, we believe we can initiate that study sometime in the first half of next year. We had a number of important milestones that we achieved last year, but focusing on 2013, the things to look for are the two Phase 2 studies, where data will come out 1033 in IBS-d, telotristat etiprate in ulcerative colitis, and the initiation of 4211 in type 2 diabetes.

And just to say and on my slide, we had $207 million roughly in cash at the end of September of last year that did not count the stock offering of $37.2 million we did after that. So, we are in a very good cash position and we think we are in an exciting position this year to partner some of our compounds as well. Thanks for your attention and we’ll be in the Sussex room, we look forward to further discussion.

Question-and-Answer Session

[No Q&A session for this event]

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.


If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!