Eli Lilly and Company (LLY) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by, and welcome to the tirzepatide SURPASS program overview. [Operator Instructions] As a reminder, today's call is being recorded. Now I'll turn the conference over to your host, Kevin Hern. Please go ahead.

Good morning. Thank you for joining us for Eli Lilly and Company's overview of the tirzepatide SURPASS Phase III program. I'm Kevin Hern, Vice President of Investor Relations. Joining me on today's call are Mike Mason, President of Lilly Diabetes; Dr. Dan Skovronsky, Chief Scientific Officer; Dr. Jeff Emmick, Vice President of Diabetes Global Product Development; and Jamie Croaning, Global Development Leader for tirzepatide. During this conference call, we anticipate making projections and forward-looking statements based on our current expectations. Our actual results could differ materially due to a number of factors, including those listed on Slide 3 and those outlined in our latest forms 10-K, 10-Q and 8-K filed with the Securities and Exchange Commission. The information we provide about our products and pipeline is for the benefit of the investment community. It is not intended to be promotional and is not sufficient for prescribing decisions. I'll now turn the call over to Mike to provide some introductory comments.

Thanks, Kevin, and thanks to everyone for joining us today. Let me start with a few comments on why we're here today and the intent for today's call. Tirzepatide is an exciting asset in our Phase III pipeline and is a novel dual GIP and GLP-1 receptor agonist. It has potential to be the first in a new class of diabetes medicines and is a topic of frequent investor questions. The purpose of today's call is to provide an overview of the Phase III diabetes clinical program, describe the framework of how we plan to assess the upcoming data readout and provide a road map for investors of additional data readouts from the SURPASS type 2 diabetes program in the coming quarters. We also plan to address frequent questions we've been receiving from investors on this program, which aren't well suited to be answered in short sound bites during earnings calls. Now let me be clear about one thing. Neither the study team nor management has been unblinded to SURPASS-1 data or any Phase III tirzepatide results. I'll pause for a few seconds to make sure everyone heard that, and on Slide 5, we reiterate that database lock has not occurred for SURPASS-1. While clinicaltrials.gov shows the study completion occurred a few weeks ago for SURPASS-1, the normal process for completing case report forms and reconciling, validating and analyzing data takes time. We currently anticipate sharing top line data from SURPASS-1 by the middle of December. The purpose of today's call is to provide facts about the program and to summarize higher management commentary. We hope you find the discussion to be a useful primer ahead of the upcoming data readout, and we look forward to taking your questions. Tirzepatide is an exceptional example of Lilly flexing its diabetes drug development muscle. We've moved with great speed to leverage our incretin expertise to uncover preclinical insights, generate compelling proof-of-concept data and rapidly enroll over 6,000 patients in the first 5 SURPASS studies. I am very excited about the outlook for Lilly Diabetes and the impact we can have on patients living with diabetes. I'll now turn the call over to Dr. Jeff Emmick to summarize the key Phase II data from tirzepatide and to provide an overview of the SURPASS program.

Thanks, Mike. This is an exciting time for the tirzepatide program as we're quickly approaching the first Phase III readout in the SURPASS program. Before we talk about what to expect, I'll spend a few minutes reminding everyone of how we got here. On Slide 7 is a summary of the encouraging results observed in the Phase IIb trial that were presented at EASD 2 years ago in 2018. In the on-treatment analysis, tirzepatide showed a statistically significant reduction in hemoglobin A1c compared to the leading GLP-1, Trulicity, across the 5, 10 and 15-milligram doses. Notably, the 15-milligram dose demonstrated more than double the hemoglobin A1c reduction of Trulicity 1.5 milligrams at week 26, the endpoint for this study. In addition to the remarkable improvement in hemoglobin A1c, we saw an impressive reduction in body weight. Tirzepatide also demonstrated a dose-dependent response on weight loss with patients on the 15-milligram arm losing 11.3 kilograms, nearly 4x the weight loss seen with Trulicity. Moving to Slide 8. Tirzepatide also enabled nearly 90% of patients in the 10 and 15-milligram arms to achieve a hemoglobin A1c of 7.7% or less, which, as most of you know, is the American Diabetes Association recommended hemoglobin A1c target for many people with type 2 diabetes. Additionally, over 40% of patients in the 15-milligram tirzepatide arm achieved a hemoglobin A1c below 5.7%, a mark of normal glucose in people without type 2 diabetes. This is a particularly notable outcome considering that only 2% of patients taking the leading GLP-1, Trulicity, achieved a hemoglobin A1c below 5.7. We believe this is an important new measure to assess the impact of diabetes medications and something we look forward to analyzing and presenting in our future tirzepatide trials. These data suggested tirzepatide could provide a sizable improvement for patients compared to existing diabetes medicines and offer potential benefits for treating diabetes and a number of other cardiometabolic diseases. Based on these compelling results, we initiated a Phase III program in patients with type 2 diabetes, the SURPASS program, followed shortly thereafter by a Phase III program in patients living with obesity called SURMOUNT. And we've also started a Phase II program in patients with nonalcoholic steatohepatitis called SYNERGY-NASH. On Slide 9, you will see the safety and tolerability data from the same previously presented Phase IIb tirzepatide study. Tolerability is a topic that we discuss often with investors, so I will spend a few minutes walking through the data that has been generated thus far and how it informed our decisions on future trial design. The most common adverse events seen with tirzepatide are known to be associated with GLP-1 therapies and included nausea, vomiting and diarrhea. In general, these effects were mild to moderate, occurred early in treatment, during dose escalation and waned with continued dosing. Of note, discontinuations due to adverse events in the 5 and 10-milligram doses of tirzepatide compared favorably with Trulicity, and only the 15-milligram dose was relatively higher. As a reminder, patients in the 15-milligram tirzepatide arm began with an initial dose of 5 milligrams and rapidly escalated to a dose of 15 milligrams over 6 weeks. So again, to underscore the point, it was a very rapid escalation. To assess the efficacy and tolerability of higher doses of tirzepatide using a different dose escalation scheme, an additional 12-week dose escalation study was executed. Moving to Slide 10. You can see the tolerability data from the additional Phase II dosing study. This study included 3 separate dose escalation approaches over the course of the 12-week treatment period. While efficacy was very similar to the Phase IIb trial at the same time point, we observed a meaningfully lower rate of total treatment discontinuations as well as treatment discontinuations due to adverse events. Discontinuations are an important objective measure of tolerability as the willingness of patients to remain on treatment and remain in the trial is the ultimate measure of the treatment experience. These data demonstrated that starting with a lower dose of tirzepatide and escalating more slowly over time reduced discontinuations due to adverse events to low single digits versus the 23% that was observed in the initial Phase IIb study at the 15-milligram dose. These data complement a growing body of evidence demonstrating that a slow and stepwise dose escalation can be effective to improve tolerability. This has been demonstrated with our own higher doses of Trulicity in the AWARD-11 study and also with semaglutide in both the STEP program as well as the recently reported SUSTAIN FORTE study. Utilizing the experience in our Phase II tirzepatide trials and pharmacokinetic modeling, we developed the dosing escalation for the SURPASS, SURMOUNT and SYNERGY-NASH programs. Patients start with 2.5 milligrams of tirzepatide and move up every 4 weeks in 2.5 milligram increments to the target maintenance doses of either 5, 10 or 15 milligrams. We believe this slow stepwise dosing approach will yield a compelling clinical profile with meaningfully better A1c and weight loss than is observed with existing therapies and a tolerability profile that is in line with GLP-1 therapies. On Slide 11 is an overview of the SURPASS Phase III program. We have a robust clinical program. And to date, we have randomized over 10,000 patients in the SURPASS program alone. SURPASS includes multiple head-to-head studies against widely used diabetes medications, including SURPASS-2, our head-to-head comparison of tirzepatide and semaglutide. This robust set of clinical trials will assess the potential for tirzepatide to improve outcomes for patients living with diabetes across the spectrum of disease progression, both alone and in combination with a number of existing therapies. Earlier this year, we also announced the initiation of SURPASS-CVOT, a cardiovascular outcomes trial that will enroll over 12,500 patients. This boldly designed trial will assess tirzepatide's ability to reduce cardiovascular events compared to Trulicity. Trulicity is the only GLP-1 currently approved to reduce the risk of major adverse cardiovascular events in patients with type 2 diabetes who have both established cardiovascular disease or multiple cardiovascular risk factors. While this call is focused on our type 2 diabetes program, I would like to also note that we have the ongoing Phase III trial in patients with obesity and a Phase II trial in patients with NASH. We expect both to readout in 2022 and potentially provide further evidence of the benefits of tirzepatide in patient populations outside of diabetes. In closing, let me say how proud I am of the work we've done as a company to execute this large and complex clinical program across 3 different indications in parallel. We have all faced many challenges this year due to COVID-19, and the tirzepatide team has found innovative ways to minimize disruptions and continue to advance development of this important new medicine for patients. I'll now turn the call over to Jamie Croaning to provide additional detail about the SURPASS clinical program and to give some context for the upcoming readout of SURPASS-1 and the readouts that will follow that.

Thanks, Jeff. As Jeff mentioned earlier, we intentionally designed SURPASS to enroll a broad range of patients living with diabetes to demonstrate the positive impact we believe tirzepatide can have in treating patients with diabetes. On Slide 13, we show key baseline characteristics for the first 5 trials in the global SURPASS program. The population in the SURPASS program had a range of key baseline characteristics, including hemoglobin A1c which ranged from 7.9% to 8.5%, weight from 85.9 kilograms to 95.3 kilograms, duration of diabetes as low as 4.8 years and as high as 13.3 years; and background oral antidiabetic medication, which ranged from no background therapy to any combination of metformin, SGL2 inhibitors and sulfonylureas. Given the differences in baseline characteristics across studies, it is reasonable to expect differences in the A1c and weight benefits observed with tirzepatide on key endpoints. This has been observed in prior diabetes trials, such as the SUSTAIN program for semaglutide and the AWARD program for Trulicity. Data from our recently presented AWARD-11 study illustrate the impact that baseline A1c can have on numerical benefit observed over the course of a study. In a pre-subset -- prespecified subset analysis of patients in this trial, patients on Trulicity 4.5 milligrams with baseline A1c of 8.5% or greater experienced a 2.3% improvement in A1c, which was higher than the 1.9% effect size observed across the total study population. We look forward to better understand the potential impact tirzepatide can make to help patients living with diabetes, beginning with the upcoming readout for SURPASS-1 yet this year. A common request we hear from investors is how should we compare the results from our clinical trials to those of competitors. As we all know, conclusion from cross-trial comparisons have limitations due to the differences in study design, patient population, duration of therapy, background therapy and other factors that confound comparisons. These limitations are real. However, in an attempt to define which competitor studies bears the closest resemblance, on Slide 13, we showed the SURPASS program, along with the most comparable semaglutide trial. SURPASS-1, SURPASS-3 and SURPASS-5 are similar to SUSTAIN 1, SUSTAIN 4 and SUSTAIN 5, respectively. And indirect comparisons of study design may be useful to provide context on tirzepatide results. SURPASS-2 is a head-to-head study comparing tirzepatide to semaglutide 1 milligram. This design is a gold standard for comparing 2 treatments. SURPASS-4 is also similar to SUSTAIN 4 as insulin glargine is a comparator. However, SUSTAIN 4 did not enroll a population with increased cardiovascular risk. Continuing on Slide 15 are key elements of the SURPASS-1 trial design. SURPASS-1 compared tirzepatide monotherapy to placebo and will be the first study to read out that features the Phase III dosing escalation. All patients randomized to tirzepatide began on the 2.5 milligram dose, an increase in 2.5 milligram increments for every 4 weeks until reaching a target dose of 5, 10 or 15 milligram. The total treatment duration in SURPASS-1 is 40 weeks, and the primary endpoint is mean change from baseline in hemoglobin A1c. As shown previously, SURPASS-1 included patients who, on average, had diabetes for less than 5 years. Other key inclusion criteria are shown on the right side of the slide as well. Consistent with how Lilly and competitors have presented study results in the past, we plan to analyze and present data from SURPASS-1 for both the efficacy estimand and the treatment regimen estimand. The efficacy estimand analysis is sometimes called the on-treatment analysis. Analysis for the efficacy estimand is based on data from patients who complete the study, data prior to discontinuations and data prior to the initiation of rescue therapy due to hyperglycemia. We believe the efficacy estimand analysis is most representative of how a medicine performed when taken as intended. The treatment regimen estimand is also sometimes called the intent-to-treat estimand. Analysis for this estimand is done for regulatory purposes and includes data irrespective of adherence to study drug or introduction of rescue therapy due to hyperglycemia. Both sets of analysis will be included in regulatory submissions. And to the extent treatment discontinuations are low, the differences between these analysis could be minimal. Moving to Slide 16. You can see a comparison of SURPASS-1 trial design to its most comparable semaglutide trial, SUSTAIN 1. As we think about the upcoming SURPASS-1 readout, we view SUSTAIN 1 as the right framing for how to contextualize SURPASS-1 results, again acknowledging the limitations of any cross-trial comparison. Both trials are monotherapy compared to placebo without background metformin, and at baseline, patients had similar hemoglobin A1c and duration of diabetes. Also shown on this slide are differences between SURPASS-1 and the tirzepatide Phase IIb trial shown earlier in this presentation. Of note, tirzepatide Phase IIb trial included patients with slightly higher baseline hemoglobin A1c, longer duration of diabetes, higher baseline weight and included patients on background metformin. SURPASS-1 also includes the optimized slower dose escalation and an additional 14 weeks of treatment compared to the 26 weeks assessed in the Phase IIb study. These factors could lead to differences in numerical benefit observed in weight loss or improvement in hemoglobin A1c. We are excited for the upcoming readout of SURPASS-1. And on Slide 17, we outlined a robust set of additional data that will be generated in the coming quarters. During the first half of 2021, we expect 3 additional readouts from SURPASS-3, SURPASS-5 and SURPASS-2. While the first 4 trials to read out will generate important data, the gating factor for global submission is the completion of SURPASS-4. Specifically, completion of this trial is contingent on accruing a prespecified number of cardiovascular events which we currently estimate will occur in the middle of 2021. Obviously, this projection is subject to any variability in the event rate and could change. While SURPASS-1 is an important part of the submission package, we do not expect that this trial to support a CV indication. As Jeff mentioned earlier, we also have an ongoing CV outcome trial to support a potential CV indication, SURPASS-CVOT, which we expect to read out by 2025. Similar to SURPASS-4, the time line for this 12,500-patient trial is contingent on the event rate within the trial in addition to recruitment. On Slide 19, our next steps for the tirzepatide program and some commentary regarding what to expect in our initial SURPASS-1 press release. As mentioned on the call, we remain on track to share initial results from SURPASS-1 by the middle of December. While this will be top line disclosure, we plan to include key efficacy and safety data in the press release to characterize these results while balancing the importance of disclosing data at a major medical meeting and peer-reviewed publication. We expect a series of additional readouts from SURPASS program throughout 2021, culminating with the planned submission in either late 2021 or early 2022. Additionally, we expect initial data from the Phase III obesity program and the Phase II NASH program in 2022. I'll now turn the call back over to Mike to provide some closing comments and to start the Q&A.

Thanks, Jamie. Tirzepatide is an important potential new medicine in our portfolio, and we're excited to begin the process of turning over data cards. We hope this session was useful to remind the investment community of the data we've generated to date, the analysis that informed the strategic choices we've made designing the SURPASS program and the framework we're using to analyze the data from its exciting potential new medicine. Diabetes drug development is core to Lilly's history, and we're committed to the pursuit of discovering new medicines that improve the lives of people with diabetes. This concludes our prepared remarks. Now I'll turn the call over to Kevin to moderate the Q&A session.

Thanks, Mike. We'd like to take questions from as many callers as possible, so [Operator Instructions]. Sean, please provide the instructions for the Q&A session, and then we're ready for the first caller.

[Operator Instructions] Our first question is going to come from the line of Seamus Fernandez from Guggenheim.

Thanks for providing the baseline characteristics. Super, super helpful. One of the comments made by the team was the importance of the baseline characteristics in terms of its potential impact on the different endpoints as it relates to HbA1c and weight loss. Just wanted to get a better sense of what your view is of the impact that could be seen there? In part, it just seems like the easiest way to do these comparisons is on a placebo subtracted basis and looking at the percent change from baseline as a driving point. But I think in the wake of the recent FORTE results, there were some surprises in that just in terms of the magnitude of the change. So just wanted to get a little bit more color on how the baseline characteristics you think impact directionally the HbA1c and the weight on that basis? And then second question is actually just the thoughts on, again, a similar question around the baseline characteristics in SUSTAIN FORTE. This honestly looked like an extremely unhealthy patient population, basically an obese diabetic patient population, HbA1c almost at 9. And this looks similar to one of your studies, but just wanted to get a little bit more color on your thoughts on what could have -- your thoughts on what happened there, not necessarily what you think of those data, but more what do you think those data may inform about the influence of baseline characteristics as we think about the upcoming tirzepatide data sets?

Thanks, Seamus. We'll go to Jeff for both of those.

Yes. Thanks, Seamus. So with respect to the baseline, as we've said, it's well-established that actually the higher the baseline hemoglobin A1c, the greater reduction you will see with [indiscernible] and other therapies. In fact, we published data on this for both Trulicity in comparison to liraglutide. And you referenced this, Novo demonstrated that just this week within the SUSTAIN FORTE results with the trial having a baseline hemoglobin A1c of 8.9. So as a general rule, we would expect higher degrees of A1c reduction in studies with higher baseline A1c values. Having said this, there are other differences in terms of background therapies, duration of diabetes that can also play a role. So it's difficult to speculate what you'll see in the individual trials. But in general, when you look at actual magnitude of the A1c reduction, even potentially placebo adjusted, you would expect it to be larger in those studies with a higher baseline. Weight is a bit more complex. Certainly, the baseline starting weight can play a factor, but there's also a potential inverse relationship with -- related to the amount of A1c reduction that you see. I think this may have been evidenced, in fact, in the SUSTAIN FORTE results this week. So then with regards to SUSTAIN FORTE, again, I'll let Novo comment on whether the results met their expectations, but the results were not a surprise to us when you look at the baseline A1c. And as you pointed out, it was extremely high at 8.9. And we've already talked about that relationship then to the extent of the reduction you'll see. I think given their very high starting point, the results are not surprising nor is it surprising that they showed some difference, some incremental reduction at the higher dose. But as a comparison, I would actually point you to a prespecified subgroup analysis that we had within our AWARD-11 study that studied the 3 and 4.5 milligrams of Trulicity. In that study, our main baseline was 8.6. But when we looked at patients with a baseline greater than 8.5, we had an A1c reduction of 1.9 for the 1.5 milligram dose and 2.3 for the 4.5 milligram dose, essentially mirroring what was observed in SUSTAIN FORTE. And so we believe this demonstrates that the pure GLP-1 pharmacology has likely been maximized in the case of SUSTAIN FORTE. We believe in contrast that our dual agonist tirzepatide has the potential to change the treatment paradigm. As -- I think you know from our Phase II study, we saw an increase in insulin sensitivity beyond what was expected from the weight loss, consistent with GIP, which is a known insulin sensitizer. And we saw regulations of biomarkers associated with insulin sensitivity, some of which were unique to tirzepatide and not shared by Trulicity in that study. And importantly, we're going to talk a lot about categorical changes in hemoglobin A1c moving forward, not only at the traditional cutoff of 7.0, but the definition of normal glycemia at 5.7. And I think you know, nearly 90% of patients on tirzepatide at both the 10 and 15-milligram dose in that Phase II study achieved an A1c less than 5-point -- or less than 7; an impressive 43% of patients less than 5.7 compared to only 2% for Trulicity at 1.5 milligrams. So hopefully, that gives you some context, both in terms of relationship to baseline A1c, but also our thoughts on SUSTAIN FORTE and then fast forwarding to the SURPASS results.

Our next question then will come from the line of Terence Flynn from Goldman Sachs.

Great. Appreciate all the context today. I guess, I just -- I know in the release it sounds like you're going to have a fair amount of data. I just wanted to make sure we'll both get the intent-to-treat and the on-treatment data in that initial press release. And then in terms of the manufacturing for tirzepatide, can you maybe talk about your plans there? Can you simply shift production to tirzepatide away from Trulicity? Or are you building out a whole separate footprint here? Maybe just speak to kind of the plans and how far underway is the process? And what's left to do as we head into 2021 and thinking about the launch in 2022?

Thanks. So we'll go to Jeff for the question around the press release and then Mike for how he's thinking about scaling up for the potential launch of tirzepatide in the future.

Terence, thanks for the question. So again, we recognize the significant importance of this press release for investors. We're still finalizing the details, but it -- we do plan, as we've said, to include the key secondary -- or key efficacy and safety data as well as key secondary efficacy data in the press release, and we will be transparent with regards to the efficacy and the treatment estimand. They both have an important place in the analysis and -- but we'll do this while balancing the preservation of the disclosure plans, as Jamie outlined.

Thanks, Jeff. Mike?

Yes. Thanks for the question. Obviously, the volume we've had to produce for Trulicity has made us experts, and we've grown to be very efficient manufacturers of Trulicity, both the API as well as the device that's used, we call it the [indiscernible] device. So on the bulk side, we do have flexibility where we can -- our plants are flexible. And we have -- we'll be able to manufacture both products without an issue. On the -- what's nice about the -- using the same device on tirzepatide as Trulicity as they come out of the same manufacturer, that process is totally fungible because it's the exact same device which is what we [ did ]. So very confident in our ability to supply both products.

Our next question will come from the line of Geoff Meacham from Bank of America.

We can come back to Geoff maybe and go to the next caller.

Our next question comes from the line of Chris Schott from JPMorgan.

Great. Again, appreciate all the color on the call today. Just my 2. First, regarding tolerability for tirzepatide. As we think about the Phase III, are you more focused on the dropout rates in the study or on the percent of patients with the various GI issues as you think about the commercial implications of the profile? And again along those lines, are you still expecting that we'll see higher GI side effects with 15 versus 10? Or do you think that can be largely resolved with titration? And I think it brings me to my second question, which maybe is kind of tied to the same thing, is, when we think about this program, is the go-to -- the most likely kind of go-to dose for tirzepatide 10-milligram with an option to go to 15 for some patients? Or do you see this primarily as a 15-milligram drug that still has a really attractive profile at 10-milligram if people can't tolerate the side effects? I'm just trying to get a sense of like, as you think of the program, what is really the focus of those 2 doses.

Thanks. We'll go to Jeff for the first one and the second one. And Mike, if you'd like to kind of weigh in on that, the second one as well, that would be great.

Yes. Thanks, Chris. So both discontinuations as well as the incidence of adverse events are important. Certainly, discontinuation is an important surrogate for the severity of the adverse events. So you have to look at both. We remain very confident in the slow dose escalation. We've leveraged in the program. And in fact, we've seen that work very well. I'll use our own AWARD-11 study of Trulicity 3 and 4.5 milligrams as an example. In that Phase III study, we titrated through 4 doses to get to the highest dose of 4.5 milligrams at 4-week intervals, very similar to what we're doing in the tirzepatide SURPASS program. And in contrast, in the original AWARD program for Trulicity, we did not titrate to get, say, to 1.5. Patients either started at 0.75 or 1.5. And if you do that comparison between what we saw at AWARD-11 in the higher doses versus the earlier AWARD program, we actually saw comparable GI adverse event rates at the 4.5 milligram dose compared to what we saw at 1.5 milligrams in the original AWARD program. Similarly, we saw a dramatic reduction in GI adverse events for our 3 and 4.5 milligram doses between Phase II and Phase III. And in our Phase II, we escalated very quickly. In our Phase III, we went more slowly. So that, I think, add -- you can add to that, again, the STEP data as well as the dose titration that was used in SUSTAIN FORTE, which was 4 steps over nearly 4 months, you see that you really can affect the incidence of the adverse events. And again, they tend to wane with time with continued dosing and are generally mild to moderate. In terms of -- you asked about dose response and adverse events, and would we expect to see higher GI adverse events at 15 milligrams versus 10? Obviously, we ultimately have to see the data. But there is typically a dose response even for Trulicity, for semaglutide in your GI adverse events as you go off the dose range. That's well established. However, again, the titration, they typically wane with time. If you can get patients through that initial period through a slow dose escalation or titration, you have the ability to ameliorate those events. So -- and then I'll -- maybe Mike to comment on the most commonly used dose expectations.

Yes. Thanks, Jeff, and thanks for the question. Good question. I'll provide a little bit more color on the first question. I think commercially, it is the discontinuation rate in the GLP market, I think, is a better indicator. While we need to be aware of the tolerability, because it is transient and tends to go away in the first few weeks or subsides in the first few weeks, we -- I think the discontinuation rate, at least commercially, is the better indicator. On the dose, obviously, in clinical trials, we will -- we need to titrate up as fast as reasonable in order to control the tolerability, in order to test the 5, 10 and 15-milligram dose against each other. In clinical practice, what we see is that because 5 milligrams is an efficacious dose, what we believe for type 2 diabetes is that someone will start on 5 milligrams. If that dose satisfied what they need at that point in time for that individual patient, they'll stay at 5 milligrams until they need additional weight or additional A1c relief. So we think all 3 doses will become important commercially. And at the end of the day, the right dose is the -- whatever dose that patient needs in order to get them to get their A1c under control. So I think all 3 doses commercially will be important.

We're going to go back to the line of Geoff Meacham from Bank of America.

All right. I think we'll -- maybe Meacham is on mute. We'll have to go to the next caller.

Thank you. We're going to go to the line of Ronny Gal from Bernstein.

My major questions are as follows. First one regarding the waning of the side effects. Obviously very important in therapy, which will be lifelong. I guess given essentially you've got a slightly different mechanism here than a straight GLP-1, have you convinced yourself that the side effects wane to the same extent just because a big residual side effect will probably lower compliance in the real world? And second, switching over a little bit to the commercial side. If you can share with us, obviously, it will depend on the clinical trial results, but are you thinking about tirzepatide as essentially taking over from Trulicity over time? Or is it still going to be the element of -- Trulicity is still there for the milder patient and the ones who need more severe therapy will go to tirzepatide?

Great. Thanks. So we'll go to Jeff for the first question around the mechanism and side effects and then Mike for the commercial implication question.

Yes. Thanks, Ronny, for the question. What we've seen in the Phase II studies that we've reported certainly is a side effect or adverse event profile, we think, is actually very similar to the pure GLP-1s. And the time course for the adverse events, which we have talked about previously, seems to be similar as well. That in other words, they tend to occur early in dosing and they wane with continued dosing. And there's sometimes a question about how they're reported. And we will look at that time course. And one way to look at it is, say, the incidence in the first week or first set of weeks versus the incidence more at steady state later in the trial. But based on what we've seen so far, our expectation is that we'll look similar to the GLPs, and we'll have to wait for the Phase III data to see exactly what it looks like.

Thanks, Jeff. Let's go over to Mike now for the second question.

Yes. Commercially, it comes down to, is the patient seeing benefit versus any tolerability issues early on in the course of treatment. And what we've seen on Trulicity is actually is that more people stay on Trulicity during the first couple of months of treatment, more so than any other GLP and any other diabetes medication. So we're confident that the risks -- if you look at the -- how quickly someone achieves both weight loss and efficacy on tirzepatide, we're confident that more patients will see a positive benefit. If you're seeing a positive benefit, then you'll be willing to accept some tolerability. Again, that's why I think the discontinuation rate is really important because patients are making that individual risk-benefit decision for themselves and were quite confident in the risk-benefit equation for tirzepatide based on the Phase II trials.

Our next question will come from the line of Andrew Baum from Citi.

A couple of questions. First, could you comment on whether you see any relationship between the frequency of adverse events with tirzepatide GI adverse events and baseline weight? I'm just interested because in SURPASS-1, the weight is lower than the baseline you had from your Phase II? And then second, could you talk to how the real-world titration is going to pan out, assuming tirzepatide is approved and launched? To what extent is the patient's going to be aware that he or she is being titrated? Or is that fairly seamless with the patients seemingly aware, but unconcerned as he receives a different prefilled syringe every month. Some sense on how it pans out would be helpful.

Thanks, Andrew. We'll go to Jeff, for the first question, and then Mike can talk about what this will feel like for the patient in the real world.

Yes. Thanks, Andrew. With regards to relationship between frequency of adverse events and baseline weight, I don't think we've seen anything to date in our Phase II trials that would suggest that. Now there is data out there for, say, other incretins comparing patients with diabetes and patients with obesity, and you sometimes do see lower adverse event rates in patients with obesity. But that's patients with obesity without diabetes. But I don't think we've seen anything like that so far in our Phase II, and we'll have to wait for the Phase III to see what it looks like there.

Thanks, Jeff. Mike?

Yes. Good question. Our team is preparing to make sure clinicians understand how to titrate. But again, I think the clinical practice is going to be different than what we see in clinical trials. I think what you'll see is that individuals will start on 5 milligrams. They'll see if they're getting -- if it's meeting their needs. If they're meeting their needs, they'll stay there until they need potentially something else down the road. And if that's the case, then they'll just go through the 7.5 to the 10. And so we'll provide good education to health care professionals, so they know how to titrate and/or escalate dose appropriately. But I don't think it's going to be as complicated as it may seem to be in the clinical trials because I think patients will go to 5 milligrams, assess kind of where they're at, stay there for a while and then proceed to 10, if they need it. And then at that point, if they need 15 milligrams down the road, they'll escalate the dose at that point. So what we're actually seeing is that clinicians and patients actually starting on a medication, they want to stay on the medication long term. And they actually like the fact of starting at a lower dose and then having the ability to titrate up to or escalate the dose at a later time if they need greater weight loss or greater A1c. So I don't think it will be complex. Actually, all of our market research indicates that physicians appreciate the flexibility.

Our next question will come from the line of David Risinger from Morgan Stanley.

Great. So I have 2 questions. First for Jeff. Could you provide more color on the additional nonclinical work that Lilly has conducted on GIP since starting the Phase III program? I just wanted to get some additional perspective on your level of confidence in GIP. And then for Jamie, could you discuss the expected filing timing in the U.S. and EU? When is SURPASS-4 expected to read out? And can you file in the EU before SURPASS-4 reads out?

Thanks, Dave. So we'll go to Jeff first.

Thanks, David. So listen, we have a whole host of ongoing work looking both preclinically and clinically at the mechanism of action related to GIP. I will comment because I know some of the underlying question here is also around cardiovascular safety. We have a study being done via an academic collaboration. It's not yet been published, but it involves a comprehensive and detailed genetic analysis to show there is no evidence of a causal relationship between GIP receptor mediated GIP levels and cardiovascular disease risk. The previous suggestions by another group were due to a linkage between a variant in the GIP receptor and a separate genetic signal and an established cardiovascular disease locus, a risk locus. That data, I can't say more about it right now, but it will be submitted for publication soon by our academic collaborator. So be on the lookout for that. But I would say, importantly, we don't have any preclinical or clinical data to date that would suggest there is risks associated with our dual agonist. And in fact, in the Phase IIb data that we've shown previously, we saw improvements across a variety of important established cardiovascular biomarkers, including weight, insulin sensitivity, cardiovascular inflammatory biomarkers and lipids, including VLDL as well as triglycerides. Ultimately, we look forward to seeing the Phase III results and then SURPASS-CVOT. But we remain confident in our program.

Thanks, Jeff. Jamie?

Yes. Thanks for the question. As it relates to SURPASS-4, as I mentioned earlier, we're using the events from that study to help [ discharge to 1.8 ]. We anticipate that the events in that study based on our current projections would be mid-2021. So based on that, we'll continue to work on completing that study and have that as part of our submission. Right now, we're targeting an end of year 2021, early 2022 for submission. At this point, we'll work with the EU regulators in terms of how we want to process the submission, but our intent is we would submit in the U.S. and then EU as well.

Our next question comes from the line of Geoff Meacham again from Bank of America.

Two quick ones. The first one is, would the treatment effect or tolerability in either SURPASS-1 or 2, would that inform your plans to expand the obesity or NASH clinical programs? Just wondering if you're going to wait until you have the totality of data in diabetes before you make a decision on the other 2. And then just given the titration scheme in diabetes [Technical Difficulty] data play out?

Geoff, I think you accidentally got muted. So if you could give us your second question, please, again.

Yes, okay. So the second question is, given the titration, is the plan now to file for multiple doses to allow for broader titration? Or are you going to make a call on the effective dose until you have much more detailed Phase III data?