Earnings Call Transcript - LXRX Q3 2020

Operator, Operator

Good morning. My name is Julianne, and I'll be your conference operator today. At this time, I would like to welcome everyone to the Lexicon Pharmaceuticals Third Quarter 2020 Financial Results Call. All lines have been placed on mute to prevent any background noise. After the speaker's remarks, there will be a question-and-answer session. I would now like to hand the conference over to your first speaker today, Chas Schultz. You may begin.

Chas Schultz, Head of Corporate Affairs

Thank you, Julianne. Good morning and welcome to the Lexicon Pharmaceuticals third quarter 2020 financial results and clinical update conference call. Joining me today are Lonnel Coats, Lexicon's President and Chief Executive Officer; Dr. Praveen Tyle, Lexicon's Executive Vice President of Research and Development; and Jeff Wade, Lexicon's Executive Vice President of Corporate and Administrative Affairs and Chief Financial Officer. Earlier today, Lexicon issued a press release announcing our financial results for the third quarter of 2020, which is available on our website and through our SEC filings. A webcast of this call, along with the slide presentation, is available on our website. During this call, we will review the information provided in the release, provide an update on our clinical programs, and then use the remainder of our time to answer your questions. Before we begin, let me remind you that we will be making forward-looking statements, including statements relating to the safety, efficacy, and the therapeutic and commercial potential of LX9211, sotagliflozin, and our other potential drug candidates. These statements may include characterizations of the expected timing and results of clinical trials of LX9211, sotagliflozin, and other drug candidates; and the regulatory status and market opportunity for those programs. This call may also contain forward-looking statements relating to our growth and future operating results, discovery and development of our drug candidates, strategic alliances and intellectual property, as well as other matters that are not historical facts or information. Various risks may cause our actual results to differ materially from those expressed or implied in such forward-looking statements. These risks include uncertainties related to the timing and results of clinical trials and preclinical studies of LX9211, sotagliflozin, and our other drug candidates; our dependence upon strategic alliances and other third-party relationships; our ability to obtain patent protections for our discoveries; limitations imposed by patents owned or controlled by third parties; and the requirements of substantial funding to conduct our research and development activities. For a list and a description of the risks and uncertainties that we face, please see the reports we have filed with the Securities and Exchange Commission. I will now turn the call over to Lonnel Coats.

Lonnel Coats, CEO

Thank you, Chas. Good morning everyone and thank you for joining us on the call. We welcome the opportunity to discuss a pivotal quarter for Lexicon in which we shifted our focus to LX9211 and our other scientific assets in research and development. During the third quarter, we completed the sale of XERMELO and related assets to TerSera Therapeutics. This sale allowed us to strengthen our balance sheet by repaying our $150 million term loan in full. We also exchanged $75.8 million in principal amount of our convertible notes for cash and common stock and have retired more than 90% of our outstanding debt in total. We are in a very good position to focus our time and efforts on advancing LX9211, which is a program we have been excited about for a number of years. Praveen will share more about this program and the potential we see for it to become a new treatment option for the millions of people suffering from neuropathic pain. Starting in the fourth quarter, Lexicon will be operating as a much more streamlined and focused company. We expect our cash burn to be dramatically lower moving forward into 2021. We will be laser focused on advancing LX9211 through to Phase 2 clinical trials in diabetic peripheral neuropathic pain and post-herpetic neuralgia. Jeff will delve into the financials more deeply, including our outlook on expected spending moving forward. I feel confident that with the de-leveraging of our balance sheet and a substantial reduction in our expected spending, we are in a much better position to advance our science and increase the opportunity from values to flow through to our stakeholders. On Slide 4, I'd like to take a quick moment and remind everyone that sotagliflozin remains an important asset in our science and development pipeline. This is a drug that we wholly own and continue to believe has the potential to become a very important new treatment for many people living with type 1 diabetes. Sotagliflozin has shown in a Phase 3 program involving approximately 3,000 patients to be effective as an adjunct to insulin in reducing A1C and achieving other secondary efficacy endpoints, which led to its approval in Europe for type 1 diabetes last year. We continue to believe that the diabetic ketoacidosis risk observed in a Phase 3 program is manageable. We will continue to seek approval in the United States. At this point, we feel there is an opportunity to initiate the next phase of our strategy and we will engage with the FDA. It is important that we gain alignment with the agency on a clear pathway forward in type 1 diabetes. We plan to start that work this quarter and we will be providing updates in the coming months as we make progress towards this outcome. We are also completing an early closeout of SCORED and SOLOIST, two outcomes studies of sotagliflozin with primary endpoints evaluating a composite of total cardiovascular death, hospitalization for heart failure, and urgent visits for heart failure. The outcomes of those trials will be presented as part of a late-breaking session titled late-breaking science at the upcoming American Heart Association Scientific Session on November 16 at 8:00 PM Eastern time, so set your calendars. With that, I would like to invite Praveen to update you on our important pipeline asset, LX9211. Dr. Tyle?

Praveen Tyle, EVP of Research and Development

Thank you, Lonnel. LX9211 is a potent, orally delivered, selective small molecule inhibitor of AP2-associated kinase, which I’ll reference as AAK1, which is a pathway we believe may have utility in reducing neuropathic pain while avoiding the addictive downsides of the opioid pathways. LX9211 was discovered by scientists working within our drug discovery alliance with Bristol-Myers Squibb, and we now wholly own all rights to LX9211 as well as any additional compounds acting through the AAK1 pathway that we may develop. As you can see on this chart, we have already initiated a Phase 2 proof-of-concept study in diabetic peripheral neuropathy pain, and we plan to initiate an additional Phase 2 proof-of-concept study in post-herpetic neuralgia by the end of this year. We expect that both of these Phase 2 studies will read out by the end of next year. Neuropathic pain affects a large portion of the population with an estimated worldwide market of approximately $6.4 billion in sales expected for 2020. This market is expected to grow at over 13% a year to $13.2 billion by 2026. There is estimated to be a prevalence of approximately 12 million people worldwide suffering from diabetic peripheral neuropathic pain and 600,000 people suffering from post-herpetic neuralgia in 2026. Despite neuropathic pain affecting millions of people, there remains a high level of unmet need for those suffering from that condition. The current approved therapies are limited by a lack of efficacy compounded by disruptive side effects. As a result, many people turn to opioids to experience some level of relief, which of course carry their own risks of potential abuse and addiction. We feel that LX9211 towards inhibition of AAK1 independent from the opioid pathway has the potential to overcome many of the shortcomings of current therapies and could become a welcome new innovation for those suffering from neuropathic pain on a routine basis. So let's talk a little bit about why we are so interested in AAK1 as an innovative pathway for neuropathic pain. The discovery of AAK1 as a potential target for neuropathic pain occurred in the course of our multi-year discovery and research program, we conducted about a decade ago called the Genome5000 project. In that initiative, we studied the phenotypes of approximately 5,000 different knockout mouse lines and identified the AAK1 knockout mice as exhibiting the most compelling phenotype for neuropathic pain among all the genes studied here at Lexicon. Our preclinical data for LX9211 has demonstrated excellent central nervous system penetration and reductions in pain behavior in animal models of neuropathic pain. I will take you through some of our preclinical findings for LX9211 in these animal models shortly. But what is very important is that we have conducted several preclinical tests to confirm LX9211’s independence from the opioid pathway. So far, we have found no concerns around addictive potential with LX9211. We have conducted single and multiple ascending dose Phase 1 clinical studies in healthy volunteers to study the safety, tolerability, and pharmacokinetics of LX9211. These studies support the preclinical profile we obtained in our laboratories. LX9211 was well tolerated with dose-proportional pharmacokinetics that was supportive of once-daily dosing. There were no drug-related serious adverse events, and the most common adverse events were transient headaches and dizziness. Overall, we feel LX9211 represents an innovative potential approach to treat neuropathic pain without the addictive potential of many of the current therapies and treatments. Let me show you some of the preclinical data supporting our view that LX9211 may be beneficial for people suffering from diabetic peripheral neuropathic pain and post-herpetic neuralgia. With diabetic peripheral neuropathic pain, people living with diabetes often develop nerve damage, which causes painful debilitating symptoms most often in their legs and feet. It is estimated that 60% to 70% of people with diabetes will develop peripheral neuropathy at some point in their life. On this slide, you will see the effect of LX9211 in the gold-standard diabetic peripheral neuropathic pain models in rats also known as the STZ-model. In this model, streptozotocin is used to induce diabetes in rats, and then tests are performed to evaluate the rats’ sensitivity to mechanical allodynia, which is a condition associated with diabetes where even slight touch can cause a painful sensation. This graph is book-ended by the brown and purple lines. The brown line represents the rats in which diabetes was induced, while the purple line represents the control rats without diabetes and with normal sensitivity to touch. The diabetic rats feel the full effect of neuropathic pain and experience pain with any type of touch as shown by 0% inhibition of mechanical allodynia. The control rats are not feeling any effect of neuropathic pain or sensitivity to touch as shown by 100% inhibition of mechanical allodynia. The lines in between these two extremes show the effect of LX9211 on the diabetic rat. As you can see, there is a clear dose-response to LX9211 that enables these diabetic rats to experience a statistically higher degree of touch without pain at the 1 milligram and 3 milligram per kilogram levels. By way of comparison in the same experiment, gabapentin was administered at 100 milligrams per kilogram and resulted in a similar response as the 1 milligram and 3 milligram per kilogram levels of LX9211. So we were able to achieve a similar level of efficacy at a much lower dose with LX9211. More importantly, at this level of dosing, we saw no motor impairment in the rats and no impairment of performance in a cognition model, which are the two significant side effects of gabapentin. We have advanced LX9211 into a Phase 2 proof-of-concept trial in diabetic peripheral neuropathic pain. This trial is a double-blind, placebo-controlled parallel group study that is expected to enroll approximately 300 patients from 30 U.S. sites. Patients will be randomized to one of the three arms: placebo, a 100 milligram initial dose followed by a 10 milligram maintenance dose, and a 200 milligram initial dose followed by a 20 milligram maintenance dose of LX9211 with a 1-1-1 randomization. The primary endpoint has changed from baseline to week six in average daily pain score, a very standard symptom measurement tool used in most of the neuropathic pain trials. The first patient was dosed during the third quarter of this year and we expect results in the fourth quarter of 2021. We have taken steps to mitigate the strong placebo effect that is characteristic of studies in diabetic peripheral neuropathy pain, including studying a large number of patients over a long treatment period, conducting the trial exclusively in the United States, and otherwise designing the trial with the potential placebo effect in mind. Now let's turn our attention to the second indication: post-herpetic neuralgia. Post-herpetic neuralgia is the most common complication of shingles. Usual who contract shingles often experience nerve damage that causes burning sharp pain long after the rash and blisters of shingles disappear. People of older age and those with greater acute pain, greater rash severity, or on immunosuppressive therapies are more likely to develop post-herpetic neuralgia. This condition can last months to years. On this slide, you will see the effect of LX9211 in a post-herpetic neuralgia model in rats. In this model, varicella zoster, the virus that causes shingles, is introduced to the rats. Rats are nocturnal by nature and prefer to stay in the dark, usually only entering the light when under duress. In this model, we looked at the time that rats spent in dark versus the light. After placing the rat in the test chamber, the rat was stimulated on the face with a filament every 15 seconds. We measured the time rats spent in the dark over five-minute intervals or 300 seconds. The rats represented by the white circles (dotted line) had not been administered the virus and therefore act normally. As you can see in both graphs, they spend nearly all their time in the dark. The rats represented by the blue circles (dotted line) had been administered the virus and are, therefore, suffering from pain caused by shingles, spending large amounts of time in the light. On the left graph, we tested the effect on the rats of a single dose of our drug LX9211, seven days post-virus infection. As you can see, the rats which were treated with our drug LX9211 acted similarly to the rats that were not infected. On the right graph, we tested the effect on rats of seven daily doses of LX9211, 14 days post-virus infection. In this case, the rats that were treated with our drug LX9211 acted nearly identically to the rats that were not infected, showing great separation from the untreated rats. By way of comparison, we ran the same test head-to-head with gabapentin at 100 milligrams per kilogram. Once again, we saw that the single doses of LX9211 and gabapentin had similar responses, while in the multiple dose administration involving seven doses of LX9211 and gabapentin, we saw further separation. When comparing gabapentin-treated rats to LX9211-treated rats, the pain response was significantly better in our LX9211-treated rats. With LX9211, we were able to achieve a similar or potentially superior level of efficacy at both single dose and multiple dose, using a much lower dose than gabapentin. We are very encouraged by the preclinical data, and we will be initiating an additional Phase 2 clinical study of LX9211 in post-herpetic neuralgia. PHN has a lower placebo effect and outcomes are more straightforward. The PHN trial will be a small pilot study with two arms only: placebo versus a 200 milligram initial dose followed by a 20 milligram maintenance dose daily. We plan to enroll approximately 74 patients across approximately 30 sites globally. The primary endpoint has also changed from baseline to week six in average daily pain score, and we expect the first patient to be dosed later this year. Overall, we feel that our drug LX9211 has the potential to show significant benefits for people suffering from diabetic peripheral neuropathic pain and post-herpetic neuralgia, and we look forward to completing and reporting results from both of these studies towards the end of next year. With that, I would like to turn the call over to Jeff Wade to discuss our financial results. Jeff?

Jeff Wade, CFO

Thank you, Praveen. To begin, I will discuss key aspects of our third quarter financials. More financial details can be found in the press release we issued earlier today and in our upcoming 10-Q SEC filing. As indicated in our press release, we had net product revenues of $6.5 million for the third quarter, reflecting product sales prior to the completion of the sale of XERMELO and related assets to TerSera Therapeutics on September 8, 2020. This compares to net product revenues of $8.4 million from the prior year quarter, which represented a full three months of product sales. The prior year quarter also reflected the collaborative revenues of $260 million from the termination of the alliance with Sanofi and recognition of the remaining amount of $23.5 million allocated to performance obligations from the initial agreement with Sanofi. Research and development expenses for the third quarter increased to $40.1 million from $26.7 million for the corresponding period in 2019, primarily due to increases in the external clinical development costs related to sotagliflozin subsequent to the termination of the Sanofi alliance. Selling, general, and administrative expenses for the third quarter decreased to $12 million from $13.9 million for the corresponding period in 2019. Our R&D and G&A expenses for the quarter included severance expenses associated with personnel changes relating to the sale of XERMELO and the refocusing of our operations around our research and development stage assets. Finally, we recognized a gain of $132.8 million during the quarter from the sale of XERMELO and related assets to TerSera. In total, we had net income for the third quarter of $82.6 million or $0.71 per diluted share, compared to a net income of $226.1 million or $1.95 per diluted share in the corresponding period in 2019. Our net income for the third quarter of 2020 and 2019 reflected non-cash stock-based compensation expenses of $1.9 million and $3.6 million, respectively. We ended the third quarter of 2020 with $111.4 million in cash and short-term investments, compared to $271.7 million as of December 31, 2019. We eliminated most of our debt during the third quarter, reducing our overall indebtedness from $245.3 million at June 30 to $20.3 million at September 30. We eliminated an additional $8.8 million in convertible debt in early October. During the third quarter, we realigned the company to focus our efforts moving forward on LX9211 and other research activities. This resulted in our headcount and operating expenses being reduced moving forward. We now expect our total full-year operating expenses for 2020 to be $225 million to $231 million down from $230 million to $250 million. We expect fourth quarter R&D expenses to be in the range of $22 million to $27 million, which would put our full-year 2020 R&D expenses in a range of $175 million to $180 million with the top end of the range down from the $185 million reflected in our prior guidance. Notably, our anticipated fourth quarter R&D expenses will be substantially less than the $40.1 million we recorded in the third quarter, the results of our continued wind-down of the sotagliflozin type 2 diabetes program and outcome studies, which will be substantially complete in the fourth quarter. We expect fourth quarter SG&A expenses to be in the range of $6 million to $7 million, which would put our full-year 2020 SG&A expenses in a range of $47 million to $48 million down from our prior guidance of $55 million to $65 million. We expect non-cash expenses to be approximately $20 million of our total operating expenses. Overall, we expect that our cash will be sufficient to fund our operations for at least the next 12 months. I will now turn the call back to Lonnel.

Lonnel Coats, CEO

Thank you, Jeff. So let me summarize, I think for us as a company, we're pretty excited about LX9211; it's been a compound that I've been excited about from the time I started here and we negotiated the wholly-owned rights from our alliance with Bristol-Myers Squibb. We have cleared a way to focus on this compound and accelerate it through development. I truly believe it will be a game changer for us, should we be successful, and therefore we have now put ourselves in a position to focus on it. For the remainder of 2020, we expect to initiate the Phase 2 studies in post-herpetic neuralgia, as well as diabetic peripheral neuropathic pain, and as noted before, we expect to report the Phase 2 study results from both of these studies by the end of next year. I will remind everybody that on November 16 at 8:00 PM Eastern time, we expect the presentation of the data from the sotagliflozin SOLOIST and SCORED Phase 3 studies at the American Heart Association Scientific Sessions. With that, I will stop there and turn the call over to the operator for Q&A.

Operator, Operator

Thank you. Your first question comes from the line of Yigal Nochomovitz from Citigroup. Your line is open, please ask your question.

Yigal Nochomovitz, Analyst

Hi, great. Thanks very much for taking the question. Praveen, you mentioned the placebo effect with some of the neuropathic pain trials. I'm wondering if you could just go into a little bit more detail please in terms of the specific steps that you're taking both for the diabetic peripheral neuropathy study as well as the post-herpetic neuralgia study? The big steps you're taking in order to mitigate the placebo response in both studies. Thanks.

Praveen Tyle, EVP of Research and Development

Thanks, Yigal, for your question, a very relevant question. Actually, we took a little bit of time in order to figure out how we wanted to design this protocol to mitigate the placebo effect. We looked at other studies which have been done in the area and why they have failed. So what we are basically doing is ensuring that we have a large number of patients in each of the randomization cohorts. For DPNP, we have 300 patients, which gives us approximately 100 patients per arm. We also have a good amount of efficacy period; most of those studies have failed because they did not assess the efficacy for up to six weeks, and we will study up to six weeks with that placebo effect in mind. We have also trained our investigators to look for this placebo effect, and we are ensuring that patients enrolled are filling out their diaries on a daily basis to ensure compliance. If they exhibit a high placebo effect, we will evaluate whether they should be included in the study. We are taking care of the placebo effect on multiple levels for diabetic peripheral neuropathic pain. For post-herpetic neuralgia, it's a more homogeneous disease, so we are less concerned about the placebo effect.

Yigal Nochomovitz, Analyst

Got it. Thank you. And then just one question for Jeff. I mean, you extinguished almost all of the debt over 90%, any particular reason you just didn't extinguish all of it? You have a small amount left on the balance sheet.

Jeff Wade, CFO

As of early October, we have less than $12 million outstanding. It was an agreed-upon exchange, so we had to come to an agreement, and we did what we thought was optimal. Our goal was to reduce the debt so that it wouldn't be an obstacle going forward. We think that the current level has achieved that objective.

Yigal Nochomovitz, Analyst

Okay, great. Thanks. And then just one more question on sotagliflozin. Can you spend a little more time just talking about some of the details in terms of the plan to seek approval in the U.S.? Are you going to conduct new studies? Are there other analyses of the existing studies that you can do to help the FDA better understand and put the DKA risk in perspective? And what is the strategy regarding seeking approval in the U.S. for type 1 diabetes?

Lonnel Coats, CEO

Yigal, thank you for the question. We watch carefully what other players in the space are doing, and particularly given the COVID-19 situation, most have moved away from type 1 and have chosen not to proceed. For us, we feel very passionate about this drug, which has shown extremely good benefits. The benefit of the drug has not been questioned. We still need to gain clear guidance from the FDA on the best way to proceed in managing the risk of diabetic ketoacidosis. We believe that we have done a lot of work in that area and we need to continue this dialogue with the agency. Currently, we have not received final clarity from the FDA, but we believe they owe it to the patient population to keep up this dialogue. We have not given up on it; we've simply paused due to COVID-19 priorities. We will approach this again and keep you posted as we engage with the agency.

Yigal Nochomovitz, Analyst

Right. Okay. Thank you very much.

Lonnel Coats, CEO

You bet.

Operator, Operator

Thank you. Your next question comes from the line of Kevin Kedra from G. Research. Your line is open.

Kevin Kedra, Analyst

Hi, thanks for taking the questions. Maybe to piggyback on that last one. You talked about how some of the competitors have moved away from the type 1 indication. They are still on the market with type 2, and there's obviously a lot of experience out there with their drugs. Are you concerned at all that if you do the work to get the FDA aligned on a pathway to type 1, that essentially, you might be doing the legwork for the entire class, and there's still an opportunity for others to quickly bring their products to market based on the data you provide?

Lonnel Coats, CEO

Yes. That’s a great question, and certainly it's a possibility. However, we strategically time our work and we are in a unique position because we have done much of the groundwork and possess the data necessary for an approved indication. If we are successful in achieving approval, having a clear mandate on how to manage the drug safely will be important. Unless other companies choose to pivot now and do the same work, I do not see them following the same pathway we are on. I believe it presents a unique opportunity for us.

Kevin Kedra, Analyst

All right, great. As far as the new discovery pipeline, you guys have obviously shifted to being more of a discovery company at this point. Should we expect to see any potential new compounds move into the clinic or IND stage in 2021 or should we really be thinking about 2021 as focusing on just LX9211 and maybe any movement on type 1 for sotagliflozin?

Lonnel Coats, CEO

Yes, that's a great question. I believe we have an opportunity to move another asset into IND-enabling studies in 2021. So I would say stay tuned; that work is being done. However, 2021 will primarily focus on two compounds for us. One, sotagliflozin; we are not abandoning it, and we see the opportunity to accelerate our work there, albeit as a lower-cost effort. Secondly, and most importantly, we will channel our efforts to accelerate LX9211 because I believe it has the greatest potential to transform our company. We need to determine that as soon as possible. We will continue our discovery initiatives and seek ways to put products into the clinic beyond 2021, particularly heading into 2022.

Kevin Kedra, Analyst

Great. And let me squeeze in one last question for Jeff. Can you give us some color on how to think about Q4 spending? Should we view those as baseline levels for 2021? Are there other moving parts within the Q4 numbers we should consider when forecasting into 2021?

Jeff Wade, CFO

I would just state that the 2020 fourth quarter R&D spend will still be higher than it will be in 2021 because we will still be winding down activities in the sotagliflozin type 2 diabetes program and outcome studies this quarter. Much of that will be substantially complete by the end of this quarter, so we expect lower R&D spend next year compared to this year.

Kevin Kedra, Analyst

Great. Thanks.

Operator, Operator

Thank you. Your next question comes from the line of Stephen Willey from Stifel. Your line is open.

Stephen Willey, Analyst

Good morning, guys. Thanks for taking the question. I was wondering if you could speak to the amount of flexibility you’ve built into the year-end 2021 Phase 2 data guidance for both of the Phase 2s, considering the patient enrollment. I know that the DPN study was slowed a bit due to COVID, and I would imagine these indications are quite sensitive to patient visits. Could you discuss your confidence in the guidance and how upcoming months may impact enrollment?

Lonnel Coats, CEO

Steve, I appreciate your question. We made it clear that enrollment slowed when COVID-19 hit in March. Otherwise, the program would be much further along. We are considering how another wave of COVID-19 might affect us. There is a bit of wiggle room built in, but not a lot. Our objective is to put all our resources into speeding up this enrollment to overcome any potential risks associated with COVID-19. That is currently built into our planning.

Stephen Willey, Analyst

Understood. I know these are blinded trials, but would there be a chance to structure in some interim analysis should patient data accumulation take longer?

Lonnel Coats, CEO

I will simply say that we are considering all ways to provide information regarding this program confidently earlier rather than later. I would categorize it that way.

Stephen Willey, Analyst

Okay. Thanks for taking the question.

Lonnel Coats, CEO

Yes.

Operator, Operator

Thank you. There are no further questions at this time, please continue.

Lonnel Coats, CEO

Let me thank everyone for joining us this morning. I think it's been a remarkable third quarter; we have moved back to our roots of being a very strong scientific company in discovery as well as development. It's important to note that we still have extremely high confidence in sotagliflozin, particularly for type 1 diabetes. More news will come on the SOLOIST and SCORED program on November 16 at 8:00 PM at the American Heart Association. I encourage you to join that program to hear the data. For LX9211, our confidence is very high around this program. Dr. Tyle outlined why our confidence is robust and what we've seen pre-clinically and in Phase 1 trials. We've designed the Phase 2 trials to provide us every chance of success because this is a remarkably unmet area of need. I will continue strategic discussions with the FDA and will communicate transparently about how we intend to proceed as the quarter progresses into the New Year, ensuring that patients have the opportunity for sotagliflozin in type 1 diabetes. Thank you all for your participation.

Operator, Operator

That does conclude our conference for today. Thank you for participating. You may all disconnect.