Lexicon Stock: Maybe Sotagliflozin Will Prevail In Heart Failure (LXRX)

This article was originally published on December 22 for TPT members.

We have been covering Lexicon Pharma (NASDAQ:LXRX) for five years now, and I invested in May 2022, took profits and am holding on to a few shares for the long haul. In 2017, Xermelo, an approved drug that treats carcinoid syndrome diarrhea, was still the big thing at Lexicon. Analysts predicted half a billion dollars in sales by 2022. In 2020, Lexicon sold out this asset for $155mn and change to focus on R&D. Its next asset, now its lead asset, is sotagliflozin, which had a large partnership with Sanofi (SNY), which fizzled out after the molecule failed to gain approval from the FDA due to safety concerns. Here’s the story:

Though Xermelo had been a boon for Lexicon, Thursday’s sale is the latest in a long line of setbacks. The Texas-based company had partnered with Sanofi for $300 million upfront to develop a diabetes drug, the SGLT1/2 inhibitor sotagliflozin. But the FDA rejected the drug in March 2019, and after multiple Phase III failures in type 2 diabetes, Sanofi exited the deal, paying another $260 million and causing Lexicon stock to drop as much as 70 percent. US regulators rejected sotagliflozin for a second time last December and Lexicon has filed an appeal.

T1DM

This Endpts story has gotten the real reason wrong for why Sanofi exited the deal, although this was what Sanofi did say in its statement at that time. Here’s the actual story, right from the horse’s mouth, which is extracted from a recent filing from Lexicon disputing a denial of hearing from the CDER:

1. Lexicon’s predecessor-in-interest, Sanofi-Aventis U.S. LLC (Sanofi), submitted NDA 210934 for sotagliflozin oral tablets in 200 and 400 mg strengths on March 22, 2018.

2. On March 22, 2019, the Office of Drug Evaluation II (ODE II) issued a complete response letter (CRL).

3. The CRL stated sotagliflozin demonstrated excess risk of diabetic ketoacidosis (DKA) that is “not justified by the drug’s modest clinical benefits.”

4. [Lexicon] subsequently appealed ODE II’s decision to the Office of New Drugs (OND) by submitting a Formal Dispute Resolution Request (FDRR) on September 3, 2019.

5. On November 29, 2019, the FDRR was denied by OND.

6. [Lexicon] appealed the denial to the Center for Drug Evaluation and Research (CDER) on December 19, 2019.

7. The appeal was denied on March 11, 2020.

8. On November 10, 2020, Lexicon submitted a request for an opportunity for a hearing.

9. On July 5, 2022, CDER served on Lexicon a Proposed Order denying Lexicon’s hearing request.

10. This filing followed, appealing that denial.

The central issue is DKA or diabetic ketoacidosis, which is described thus:

Diabetic ketoacidosis (DKA) is a life-threatening problem that affects people with diabetes. It occurs when the body starts breaking down fat at a rate that is much too fast. The liver processes the fat into a fuel called ketones, which causes the blood to become acidic.

The company contends that a) DKA is a serious risk, b) it can be mitigated by a proper REMS, and c) other, unapproved SGLT2 inhibitors like sotagliflozin have been regularly used off-label by T1DM patients with the same DKA risk, and all the CDER has done in this regard is to issue a noncommittal declaration. The company also says that if the FDA addresses the DKA issue as Lexicon proposes, then, in terms of efficacy, sotagliflozin is highly approvable. The FDA denies all this.

Here’s a brief extract from a study sponsored by Lexicon (?) that summarizes their view:

In patients with type 1 diabetes, confirmed DKA incidence increased when sotagliflozin was added to insulin compared with insulin alone. A lower incidence of DKA was observed following the implementation of an enhanced risk mitigation plan, suggesting that this risk can be managed with patient education.

The FDA’s view, which has remained consistent throughout:

“The data demonstrated that the addition of sotagliflozin to insulin is associated with an increased risk of diabetic ketoacidosis (DKA), a serious and often life-threatening consequence of insulin insufficiency,” the agency said in a rare explanation of its complete response letter. “Time-to-event analyses of the clinical trial data showed earlier development of DKA in sotagliflozin-treated patients than in patients assigned to placebo, without evidence that the risk stopped increasing over time.”

What happens now? This:

Moving forward, according to an agreement made between the FDA and Lexicon in September, if CDER continues to recommend that the FDA’s Office of the Commissioner deny Lexicon’s request for a hearing, CDER shall submit Lexicon’s request for a hearing, Lexicon’s response, CDER’s analysis of Lexicon’s request for a hearing including any additional analysis by CDER of Lexicon’s response, and a proposed order denying the request for a hearing to the Office of the Commissioner by Jan. 31, 2023.

Interestingly, the EMA approved sotagliflozin in April 2019 in Europe, where it is sold as Zynquista. The approval opinion makes for an interesting reading when contrasted with what some will say is the FDA’s stubborn big brother attitude. DKA was 2.9% and 3.8% in 200mg and 400mg sotagliflozin pooled data.

Moving away from sotagliflozin in T1DM, the molecule is also being tested in heart failure patients. Just two weeks after the FDA again denied the T1DM application, it accepted sotagliflozin’s resubmission in heart failure, with a tentative May 2023 PDUFA. This is a resubmission because it was originally filed in December 2021, then withdrawn in March to correct a “technical issue.” I discussed this previously.

Heart Failure

In two phase 3 trials, SOLOIST and SCORED, sotagliflozin met the primary endpoint, showing statistically significant reductions in cardiovascular deaths, hospitalizations for heart failure and urgent heart failure visits in patients treated with sotagliflozin as compared with placebo. Note that this was in T2DM patients, where repeated trials have shown no DKA resulting from sotagliflozin use. In both these trials, DKA events were negligible:

SOLOIST: “diabetic ketoacidosis (0.3% on sotagliflozin vs. 0.7% on placebo)”

SCORED: “diabetic ketoacidosis (0.6% on sotagliflozin vs. 0.3% on placebo) was more common in the sotagliflozin treated group” [thanks to NDHT for this data; I didn’t need to look it up]

I inadvertently mischaracterized this issue in an article from last year. In an article this year, I noted the difference between sotagliflozin and two blockbuster drugs:

Earlier this year, Jardiance got a label expansion to reduce the risk of cardiovascular death and hospitalization in adults with heart failure, making it the only such approved drug that is ejection fraction agnostic. Just this month, Farxiga also showed strong data in the DELIVER trial. However, Sotagliflozin is differentiated in so far it is a dual SGLT1/SGLT2 inhibitor.

Key data from the HF trials were:

• 33% absolute risk reduction in the composite cardiovascular endpoint in type 2 diabetes (T2D) patients with acute decompensated heart failure (HF’), with consistent results among patients with preserved ejection fraction (HFpEF) and reduced ejection fraction (HFrEF). Benefit was evident within one month.

• 26% reduction in the composite cardiovascular endpoint in patients with T2D and chronic kidney disease. Benefit was evident within three months.

• Reductions in HF-related events were more pronounced at the target once-daily dose of sotagliflozin 400 mg.

• Reductions in both myocardial infarction and stroke were greater than reported in studies of selective SGLT2 inhibitors.

My May article discussed the differences between this data and the jardiance/farxiga data. I need to repeat that, which was a long extract from their earnings call. Bear with me here.

I need to quote a long extract from their May earnings call that highlights how sota is differentiated from Jardiance/Farxiga:

I think there are three or four main points that I would like to cover. And I think Lonnel did a good job framing it already is that we believe that the top-line announcement from AstraZeneca this morning via press release, reaffirms the importance of SGLT inhibitors as a foundation of care in the treatment of patients with heart failure.

The second is there were no surprises in our view from the Deliver results, because we believe they are very similar to the Emperor-preserved population. And as a reminder, that is a population of patients that have a history of heart failure, but not necessarily recent heart failure.

And in fact, the results from DAPA on the DELIVER study are only about 10% of the patients had a recent hospitalization for heart failure defined by less than 30-days. So we believe that it reaffirms the uniqueness of the SOLOIST population, where a hundred percent of the patients had been hospitalized and 50% of them were started on sotagliflozin during their hospitalization and the other 50% within three days of the relief from the hospital of those patients for a heart failure hospitalization.

The other important difference between SOLOIST and DELIVER, similarly between DELIVER and EMPEROR-Preserved is that a very low or relatively low percentage of patients are on guideline directed medical treatment when you think about the other pillars of care being beta blockers, ACE, ARBs, ARNI and MRA.

So we believe that the DELIVER results reaffirm the benefits of the class, but also reinforce the uniqueness of the SOLOIST population. And frankly, it is the value proposition we believe of the dual inhibition of both SGLT1 and SGLT2 with the benefits as we showed at ACC in reduction in stroke and heart attack in at risk patients for heart failure, and the uniqueness and the benefit and rapid onset of benefits in those with a recent heart failure event.

So what differentiates Farxiga with sota is the trial population of SOLOIST. In that regard, an analyst pointed out that Jardiance’s EMPULSE study has a similar population to SOLOIST. However, Lexicon said that “they did not have hard clinical endpoints as their primary endpoint.”

In a recent report from Lexicon and, interestingly, Sanofi, differences between sotagliflozin and jardiance was highlighted:

Changes from baseline in glycemic and blood pressure control, cardiovascular biomarkers, and other parameters were comparable between sotagliflozin and empagliflozin. However, sotagliflozin but not empagliflozin inhibited intestinal SGLT1 after breakfast as shown by larger changes in postprandial glucose, insulin, GIP, and GLP-1 AUCs, particularly after breakfast.

Another study from Dr. Bertram Pitt, Dr. Deepak Bhatt, et al. highlights the uses of dual sglt1/2 inhibition:

Thus, from the currently available evidence, it can be postulated that the SGLT1/2i sotagliflozin is likely at least as effective as the more selective SGLT2is in preventing HF in patients with T2DM and CKD as well as in reducing CV mortality and HHF in patients with HFrEF and HFpEF with T2DM. Sodium glucose cotransporter-1 inhibition appears to add to the benefits of SGLT2i in that sotagliflozin reduced non-fatal and fatal stroke as well as non-fatal and fatal MI in patients with T2DM by >30%,16 and can, therefore, provide potentially greater prevention of HF, at least in patients with T2DM, than the relatively selective SGLT2is. While this hypothesis requires further confirmation in adequately powered direct comparative trials and there remains much to be learned about the expression and function of SGLT1 in patients with HF both with and without T2DM, the evidence reviewed above from the SCORED and SOLOIST trials15,16 as well as from the Mendelian randomization study examining missense variants associated with a decrease in SGLT1 function14 suggests that SGLT1i does add to the benefits of SGLT2i in the prevention and treatment of HF, at least in patients with T2DM.

LX9211

Lexicon’s last important candidate is the AAK1 inhibitor LX9211 targeting two related indications, painful diabetic neuropathy (DPN), and post-herpetic neuralgia (PHN). In June, the molecule met the primary endpoint in a phase 2 study in DPN. Yesterday, the molecule failed to meet the primary endpoint in the second phase 2 trial, in PHN. Nothing really comes easy for Lexicon, right?

In the 319-patient RELIEF-DPN-1 clinical trial, LX9211 achieved the primary endpoint with a statistically significant reduction in a score of average daily pain. The low dose arm showed a reduction of 1.39 points from baseline and the high dose arm a 1.27-point reduction, versus a 0.72-point reduction for the placebo group.

Here’s an interesting quote from Fierce:

But analysts were left unsatisfied with the top-line results, peppering executives with questions about pieces that were left out of the data drop during a Thursday morning conference call. They asked for a p-value, more information on dropout rates in the high-dose arm and specific rates of common side effects, which included dizziness and nausea, for each study group.

Executives didn’t bite, beyond previewing what they found in the top-line analysis, instead promising to package up the results for a future medical journal instead.

In the PHN trial, LX9211 achieved a reduction in average daily pain score 0.80 points greater than with placebo, however the difference was not statistically significant. The company claims the molecule showed drug activity.

The company plans to move LX9211 to later stage trials in DPN. PHN’s status is still unknown, since data was just out. No new trials are listed.

Financials

LXRX has a market cap of $362mn and a cash reserve of $136mn. Lexicon completed a public offering and concurrent private placement of common stock, with net proceeds to the company of $94.3 million. Research and development expenses for the third quarter of 2022 were $10.6 million, while selling, general and administrative expenses were $12.6 million. That gives them a cash runway of around 4 quarters. They have a $125mn loan facility left with Oxford Financial.

Bottomline

To sum up Lexicon: this company started nearly 3 decades ago, and spent two of those trying to get to the market, which they finally did with Xarmelo, but then failed to commercialize it to full potential, and were forced to sell out. It then went back to the drawing boards, found a nice, novel molecule, got some decent data, got approved in the EU, but not in the US, and it is still pursuing this indication with gritty determination. It now has another dataset from HF, where the same molecule has shown excellent data. Its LX9211 molecule saw mixed success, but something good may come out of the DPN indication. The stock is trading near lows, and they have a PDUFA in about 6 months. I purchased, took profits, and am holding on to some shares. I think this is a risky bet, but the risk-reward profile isn’t too bad.