CureVac N.V. (CVAC) Earnings Call Transcript & Summary

Greetings, and welcome to results of the second interim analysis of CureVac's Pivotal Phase IIb/III HERALD Study. [Operator Instructions] Please note, this conference is being recorded. At this time, I'll turn the conference over to Sarah Fakih, Head of Corporate Communications and Investor Relations. Please go ahead.

Thank you. Good morning, good afternoon, and welcome to our conference call. My name is Sarah Fakih, and I'm the Vice President of Corporate Communications and Investor Relations at CureVac. Please let me introduce today's speakers. On the call with me are Franz Haas, the Chief Executive Officer of CureVac; Ulrike Gnad-Vogt, our Interim Chief Development Officer; and Mariola Fotin-Mleczek, CureVac's Chief Technology Officer. Pierre Kemula, Chief Financial Officer, will be available for the Q&A session after the presentation. Please note that this call is being webcast live and will be archived on the Events and Presentations section under Investor Relations on our website. Before we begin, a few forward-looking statements. The discussion and responses to your questions on this call reflect management's view as of today, Thursday, June 17, 2021. We will be making statements and providing responses to your questions that state our intentions, beliefs, expectations or predictions of the future. These constitute forward-looking statements for the purpose of the safe harbor provisions. These statements involve risks and uncertainties that could cause actual results to differ materially from those projected. CureVac disclaims any intention or obligation to revise any forward-looking statements. For more information, please refer to our filings with the U.S. Securities and Exchange Commission. I will now turn the call over to Franz.

Thank you, Sarah. Ladies and gentlemen, a warm welcome to this conference call from all of us here at CureVac. The COVID-19 environment has changed dramatically since the beginning of the pandemic and is now dominated by evolution of the virus and variants at the rapid spreads of new emerging virus variants. Over the past 6 months and in an increasingly challenging variant-rich environment, we have been inducting a pivotal Phase IIb/III efficacy trial, called the HERALD study, with our first-generation COVID-19 vaccine candidate called CVnCoV. This study is so far unmatched in geographic diversity being conducted in 10 countries across Europe and Latin America. We believe the resulting multivariant study represents today -- presented today provides important insights into dramatically transformed variant environment, suggesting that we are virtually fighting a different virus, a different pandemic over the last 6 months. Within this environment, CVnCoV efficacy was calculated to be 47% on the basis of 134 adjudicated cases that met the criteria for analysis inclusion. This was calculated against any severity of disease and across all and variants and virus strains. Efficacy data from this trial needs to be [ fueled ] against the background of comprehensive sequencing data acquired in parallel to the accrual of COVID-19 cases within the trial. Out of the 134 COVID cases accrued and adjudicated, as a basis for a first and preliminary calculation of efficacy, 124 cases were sequenced to identify the respective virus strain. The results are sobering. From the 124 sequenced, adjudicated cases, only 1 case can be attributed to the original virus strain the world fought throughout 2020. We recognize that demonstrating high efficacy in this unprecedented broad diversity of variants is quite challenging. Now let me please go into further details of the study setup before I hand over to Ulrike for a discussion on the interim analysis. On Slide 5, you can see an overview of the general setup of our HERALD study. Conducted in 4 countries in Europe and 6 countries in Latin America, the HERALD study represents a multivariant study with high geographic and ethnic diversity. Of the approximately 40,000 trial participants, about 25% were recruited in Europe, while about 75% were recruited in Latin America, a region which, in May this year, was considered to be an epicenter of the COVID-19 pandemic. Of these approximately 40,000 participants, roughly 5,000 participants, or 13%, were above the age of 60, while the majority of about 35,000 participants or 87% were between the age of 18 and 60 years. The mean age of the study participants was in the range of 43 years. Now I will hand over to Ulrike, our Interim Chief Development Officer, to walk through the case accrual process and the interim analysis outcome.

Thank you, Franz. Please let me start by providing you with an overview of the COVID-19 case approval process within the HERALD study according to the study's primary endpoint. This is defined as the occurrence of first episodes of confirmed cases of COVID-19 of any severity. Per study protocol, COVID-19 cases are eligible to be included in the vaccine efficacy calculation if they occur at least 15 days after the second vaccination, the earliest time at which the vaccine is considered to be fully protective. COVID-19 cases occurring at least 15 days after the second vaccination undergo a stringent adjudication process, which, among other things, confirms presumed COVID-19 infections via real-time PCR and make sure to exclude those confirmed COVID-19 participants who also test positively for a COVID-19 infection before the 15 days cutoff. The present data analysis is based on 134 COVID-19 cases that met the stringent adjudication criteria and, hence, form the basis for the preliminary CVnCoV efficacy calculation. Of these 134 COVID-19 cases, 119 cases or approximately 19% (sic) [ 90% ] were detected in participants below the age of 60, while 15 cases or approximately 10% were detected in participants above the age of 60. This distribution corresponds well with the general age distribution within the trial. As Franz has already highlighted, confirmed COVID-19 cases need to be seen in the context of the rapid spread of new virus variants. For some variants, vaccine in used antibody neutralizing capacity can potentially be reduced. On Slide 7, let me dive deeper into the variant background, which critically determines the context and, thus, the interpretation of the preliminary CVnCoV efficacy. New virus variants have been steadily spreading since the end of 2020 and have all, but displaced the original virus strain. On this slide, you can see a general and non-trial-related overview of the currently estimated prevalence of foremost prominent variants with a focus on the general geographies where we are conducting the HERALD study. Presently, we are looking at 4 main variants of concern. These include the Alpha strain, first detected in the U.K,; the Beta strain, first detected in South Africa; the Gamma strain, first detected in Brazil; the Delta strain, first detected in India; and the new variant of interest, the Lambda strain, first detected in Peru. The ring diagrams on this slide illustrates the current estimated variants spread in South America and Europe. For your reference, the specific countries where we are conducting our Phase IIb/III trial are marked in blue. Because of the continuing spread of variants that has shaped this image over the last several months, we see a wide range of variants in this highly international study and have conducted comprehensive sequencing analysis to understand the dynamics of the virus in our trial. Over the next 2 slides, I will go into the details of the unique variant distribution we have detected in the HERALD study. Within the HERALD study, 124 of the adjudicated cases were sequenced, which represents approximately 93% of the adjudicated cases that were used for the interim efficacy analysis. This means that the strain distribution illustrated on this slide provides important variant context around the preliminary CVnCoV efficacy of 47% calculated within the second interim analysis against any severity of disease according to the primary study objective. Of the 124 sequenced, adjudicated cases, Variants of Concern, including the Alpha and the Gamma strains represent approximately 57%. This is mainly supplemented by 21% of the Lambda or C.37 strain originating from Peru, and 7% of the B.1.621 strain originating from Colombia. Both strains are less explored. However, C.37 was very recently added to the WHO watch list of variants of interest and was assigned the name Lambda. In May, this variant was reported to be responsible for more than 80% of COVID-19 cases in Peru and evidence has become available for high rates of transmission in multiple countries in Latin America. The strain is known to feature the D614G mutation, which is understood to have potentially higher transmissibility as well as 2 critical mutations in the receptor binding motive of the spike protein. As a receptor binding motive is the primary target for neutralizing antibodies, this might potentially allow this variant to evade immunity. Together with a 13% contribution of other strains, which further subdivided into 9 other strains, with less prevalence, this adds up to a total of 13 variants, which provide direct context for the preliminary efficacy calculation. As shown in the geographic breakdown of the total number of sequenced cases, the broad variety of variants originates primarily in Latin America, which contributed about 80 cases for 65% to the 124 sequenced cases. In Europe, the 44 cases we observed, or 35% of the total, are strongly dominated by the Alpha strain in accordance with the general virus distribution there. The variants overview provided on the next slide further extends the pool of sequenced COVID-19 cases recorded in the HERALD study to include nonadjudicated cases, providing a more detailed picture of the variant dynamics within the study. This overview is based on a total of 474 COVID-19 cases, which were approved and sequence within the HERALD study irrespective of the former COVID-19 case adjudication process and, thereby, includes the 124 adjudicated cases as well as 350 nonadjudicated cases. While the majority of these cases, therefore, did not contribute to the efficacy calculation, they provide a more accurate and extended picture of the circulating virus variants within the study. Overall, 29 different COVID-19 strains were identified with which cover are currently these designated variants of concern, including the Alpha, Beta, Gamma, Delta and Epsilon strains. Variants of interest, including the Beta strain as well as the newly assigned Lamda, or C.37 strain first detected in Peru. For strains belonging to the original virus and B.1.621 and B.1.1.519 with high prevalence in Colombia and Mexico, respectively. In accordance with the variant mix in the group of adjudicated cases, diversity was again more pronounced in Latin America compared to Europe, where the Alpha strain dominates. I am now on Slide 10 to briefly touch on the preliminary vaccine efficacy data from the second interim analysis. First, let me remind you, again, that preliminary vaccine efficacy of 47% was calculated for the prevention of COVID-19 cases of any severity and across all strains. Although the study was not statistically powered to provide separate subpopulation efficacies, the preliminary data show trends for age and strain-related efficacy. For age-related trends, the interim results suggest efficacy in younger participants, but did not allow to conclude efficacy in the age group above 60. For strain-related trends, interim data suggests an efficacy range within the most prevalent strains of the 13 different COVID-19 variants identified within the adjudicated cases applied for efficacy calculation. We will now continue to find an analysis on the basis of approximately 80 more cases compared to the second interim analysis, which we expect in approximately 2 to 3 weeks. Based on the addition of new cases, the efficacy readout is expected to change. I'm now on Slide 11 to briefly update you on the safety and reactogenicity profile of CVnCoV on the basis of the first 2,000 participants recruited within the HERALD study. The 2 12-microgram vaccinations of CVnCoV are shown to be well tolerated. As previously reported for CVnCoV and in line with other mRNA-based COVID-19 vaccines, the majority of the events are mild to moderate and accumulated around fatigue, headache, muscle pain or myalgia and chill and only few fever events were reported. We were particularly pleased to see that there was no increase in the severity of side effects after the second vaccination compared to the first vaccination. Those reactogenicity profiles are comparable. The data is further stratified according to trial participants between the age of 18 to 60 and participants above the age of 60. As expected, side effects are generally more pronounced, but still in a well-tolerated range with the groups of 18 to 60 year old study participants compared to study participants above the age of 60. Overall, the safety profile is fully in accordance with the safety profile reported for other mRNA-based vaccines and further confirms the safe applicability of CVnCoV. Let me now hand back the call to Franz for a short summary of key messages and next steps.

Thank you, Ulrike. Let me quickly summarize key messages from the present interim analysis and outline the next steps. Our technology had to prevail in 1 of the most challenging COVID-19 environments. While we were hoping for a stronger outcome of the second interim analysis, we see trends that indicate efficacy in younger age groups below the age of 60 as well as potentially differentiated efficacy per strain, as far as the conclusions from this interim data readout with a limited statistical power allow. We will now move to the final analysis, which we expect within the next 2 to 3 weeks, and which we will carry out on the basis of more than 200 cases to confirm age and strain-related trends and also look at disease severity. These cases will also be characterized by accompanying sequencing data. Data from the final analysis will allow us to look further into the efficacy to see -- we see in participants below the age of 60. And please note again, this takes into account efficacy against any severity of disease within the challenging variant environment. Final analysis will also allow us to better understand the unprecedented broad diversity and dynamics of variants in all trial countries, which had an impact on the vaccine efficacy we see so far. We mainly fully -- are fully committed to supporting ending this pandemic. Following the final analysis, we will carefully assess the most appropriate regulatory pathway for our vaccine, CVnCoV. Lastly, we had a strong second-generation candidate lined up in our COVID-19 pipeline, which we have -- which we are codeveloping in partnership with GSK, and which we intend to bring into the clinics within the next 4 months based on promising preclinical data we have. For the last part of the presentation, let me now hand over to Mariola for a quick look into the new preclinical data and our second-generation COVID-19 vaccine, which targets fast improved immune responses at even lower doses to preliminary -- to primarily meet the challenge of the emerging new variants via multivalent vaccines as well as combination vaccine for potential protection against multiple infectious disease in a single vaccine.

Thank you, Franz. To further counter the ongoing spread of new virus variants on Slide 13, I would like to introduce 2 second-generation COVID-19 vaccines, which we are codeveloping together with our partner, GSK, on the basis of a new messenger RNA backbone. This new messenger RNA backbone differs from the setup of our first-generation vaccine candidate, CVnCoV, and was generated on the basis of our learnings over the past year. Based on non-chemically modified messenger RNA, the second-generation messenger RNA backbone features targeted optimization, focusing on improved messenger RNA translation for increased and extended protein expression as well as improved immunogenicity. It verified targets the ability to rapidly induce improved immune responses compared to CVnCoV. These characteristics will be key for the development of multivalent vaccine as well as combination vaccines. The second-generation data, illustrated on Slide 14, essential proof-of-concept data applying to the second-generation COVID-19 vaccine candidate, CV2CoV. The data from part of a larger preclinical data set generated in nonhuman primate, which is currently being generated in collaboration with the Harvard Medical School. So far, unpublished data is based on animals immunized with CV2CoV, our CVnCoV according to the vacation schedule applied in humans, which includes 2 vacinations with a 12-microgram dose given on day 0 and day 28. The neutralizing antibody titers shown here strongly confirm the functionality of advanced second-generation approach for rapid and strong immune response. The onset of neutralizing antibodies occurred faster for CV2CoV than for CVnCoV. Over the 2 weeks after first vaccination, CV2CoV shows good antibody titers, and while the second vaccination on day 28, efficiently boost antibody titers for both candidates, it leads to a more pronounced booster effect of CV2CoV. The full data publication will be made available for a scientific manuscript over the next weeks. With this, we conclude our presentation and would now like to open the webcast to your questions.

[Operator Instructions] Our first question is coming from the line of Umer Raffat with Evercore.

And I kind of wanted to go a step beyond the disappointing news from last night and perhaps try to understand your data in the context of broader vaccines for a second. So I'm trying to connect the dots between the lower efficacy headline you have versus the spread of neutralizing antibody titers seen in Phase I. And as you recall, 1/3 of the patients at your 12-microgram dose had fairly low neutralizing antibody titers after 2 doses. So my question is, have you compared the neutralizing antibody titers against those key variants where the patients that were protective versus the ones that were not? Is there anything we could learn from that, #1? And #2, how different was the efficacy in younger versus older adults? Because this question also starts to answer neutralizing titer levels, which are needed for protection. And then finally, if you could just clarify -- the press release last night says 424 cases have happened, of which the 134 were adjudicated. But on Slide 9 today, it implies you've had 474 cases. Could you please clarify that?

Maybe I will take the first question regarding correlation between antibody titers and protection. So we don't have this data now, but this is exactly what we want to better understand. We have analyzed more than 600 subjects here for induction of antibody responses. And this is -- will be the task also for next week to look whether we -- how this antibody titers correlate with protection, et cetera, to be -- think about the establishment of correlation -- rate of protection, if this work will be done during next week. Then the second question?

Younger versus older.

Younger versus older, currently also for this year, we need to go for the final analysis because especially for older people we have really a few cases, yes, and therefore, we see trends, but to state final numbers, we need to wait for the final data because they will change.

And maybe Umer, as a last statement, the 424 overall sequenced cases versus 474 in the presentation, that is, in fact, a typo in the press release. I'm very, very sorry for that, and it will be corrected. The correct number is 474 cases sequenced overall.

Got it. And just to be clear, Sarah, of the 474, 134 were already adjudicated and another 80 will be adjudicated, meaning every case beyond the 134 plus 80 are cases that happened before day 15 post-dose. Am I understanding that right?

Yes. So all the cases that were nonadjudicated did not meet adjudication criteria, exactly.

Our next question is from the line of Eun Yang with Jefferies.

So I have a few questions, too. So in that study, at the interim, about 57% of the cases were caused by the Variants of Concern. So can you kind of talk about what was the efficacy for the Variants of Concern versus non-Variants of Concern? That's question one. And second, you mentioned that at the final analysis that we are expecting at over 200 cases, the data -- the efficacy could change. So can you elaborate on directionally what kind of a change we may see? And thirdly, can you comment on timing for Phase IIb data update? And you also added a third dose. Did you add the third dose because you are not achieving the neutralizing antibody titers that you hoped for?

So I can take your question on the -- about the efficacy against individual Variants of Concern. So it is premature to conclude about this today. We need to go for the final analysis to accrue sufficient cases to draw more statistically robust conclusions. And the snapshot as of today will change, yes. So this would be premature.

And I think it was also the question on our 002 study in elderly in Peru, Panama. As we mentioned during our last webcast, yes, that this study was amended also to be able to look on efficacy in this trial because here, we saw changes in the clearance of COVID-19 cases. And for this focus, the study remained blinded. And as you know, that is also a placebo-controlled study, and therefore, we were not able to publish or to disclose results in unblinded way, yes? But this is now also occurring. And we will also comment and share with you results from this trial as well. And what was the question?

On the third dose on IIa. So the reason -- maybe let me jump in on third dose. We added a dose in each age group in the IIa. And the reason for that is we also want to assess the boostability, right? The booster approach. So we have 2 horizons, is 1 month and 6 months after the second dose in IIa. So this is something that we need to understand to see once you prime is how well you boost the participants. Did we answer all your questions?

Our next question is coming from the line of Zhiqiang Shu with Berenberg.

My first one is also on the variants. In particular for Alpha strain, I think, a few other amount of vaccines have been tested specifically strain. Do you think in the final analysis, we will be able to see -- particularly on this strain, we can draw a conclusion about the efficacy so that we can compare side to side with another mRNA vaccine? That's the first one. And the second one, I think your secondary endpoint is the prevention of the larger, severe cases of hospitalization. Do you think, in the final analysis, we'll be able to see that data as part of the data package for approval? Is there -- can you also provide some color around your interaction with EMA regarding the path forward?

Yes. So let me take the first question. So we will continue to final analysis, and that may allow us to get a better understanding depending on the number of cases caused by the Alpha strain about the strain-specific efficacy. As I said already, as of today, a firm conclusion is not possible. Then the next question was, I think, related to the moderate to severe cases. Yes, I mean, here, I can say that this interim analysis recorded a vast majority of mild cases. We need to accrue more cases to reach a significant, and this will depend on the number of additional severe cases that come in the new case accrued for final analysis. So we have to see how many we get, but the number is definitely lower, and that makes -- will make statistically some conclusions on moderate, in particular, severe cases more difficult. And the last one, yes, was related to interactions with EMA. So we will first finish the study, but we are already in close contact with EMA to assess the path forward. Yes, that is the current situation. Does that answer your question?

I guess for EMA interaction, maybe can you provide some guidance on the timing and what specific you're looking for potential approval?

Yes, perhaps let me take this one. So certainly, we have got this rolling submission. And certainly, we talked about this interim analysis with EMA as well. And it's quite clear that we need to provide the data and the final analysis before there is any estimation. Because what we can see in this changing environment, vastly changing environment with all the different variants coming in, it's quite easy to -- it's quite hard to estimate what there will be. So the goal is here the totality of data to really see -- there is hardly any big package as we are providing right now and to be provided. So the final data and the totality of the data to be looked at to make then an educated decision here. And that's exactly the status what we are with EMA.

Next question is from the line of Seamus Fernandez with Guggenheim.

So just wanted to get a couple of clarifications. When might we see -- can you just clarify when we might see either the regional efficacy, so efficacy in Europe only? Is it possible that we could see enough events in Europe over time and against the Alpha strain specifically? So I guess part of the question is, when might we see a regional efficacy evaluation separate from what -- the very different variants that we're seeing in South America? And then a separate question is, when will we see data simply on Alpha? Is that possible that we could see a clear secondary endpoint, at least against just the Alpha strain?

We will provide those data with the final analysis. But at present, it is premature to conclude on that.

Yes. So it's a pretty liquid situation when it comes to that end. So I think once we have accrued all the cases, this is where we will report the final picture. And of course, depending on the occurrence of each 1 strain, you will see a [ continental ], which makes sense or less, right? But we will report these when the data comes out in, would say, 2 to 3 weeks.

Okay. And then just a final question on the -- your plans for advancing from moving forward, what really are the decision points from here? From my perspective, it seems like CV2CoV is promising on the basis of the preclinical data, but that still is preclinical. And we've seen very different -- but you do have evidence of clear differentiation between the 2 programs. So can you just help us understand which program you're motivated to move forward with in that context? Or do we just have to wait for the final data prior to those decisions? Basically, are you going to accelerate the CV2CoV in the context of these data?

Thank you for the question. What we do is we are going full speed for final readout. As Ulrike ws saying we're expecting the data here to come within the next 3 weeks, and there is no reason to slow down anything here, not with regard to the building out of our manufacturing network or to slow down the process with regards to CVnCoV. Because that's the primary product candidate close to the finish line, where exactly what all the questions came back to say what is it, what we can say now? Not very much. Therefore, we're going for the final data, and plan for regulatory pathway, which means approval, the data will show. So there is no reason to slow this down. In parallel, what we're doing with -- also with a lot of effort, together with GSK, we are working on the CV2CoV with different properties, but also to think about multivalent vaccines to work very certainly on CV2CoV on different variants as well to start with, but also multivalent vaccines. And then even too, as we have got a product collaboration with GSK on other mRNA-based vaccines, 5 in number. So there are both strains on the CVnCoV. This is going full speed, nothing to be reprioritized, not with regard to the development, the data or the manufacturing. And the same in the manufacturing certainly plays a role for the CV2CoV as well. And the other 1 is working completely in parallel also full speed. So there is not a strategic shift or something to decide now. This is what the final data we are also looking to, to get those. And yes, then, perhaps, there will be decisions to be taken. But for now, we are going full speed exactly where we are.