Sionna Therapeutics, Inc. (SION) Earnings Call Transcript & Summary

Good afternoon, and welcome to the Sionna Therapeutics Conference Call. [Operator Instructions] Please note that this event is being recorded. I will now turn the call over to Juliet Labadorf, Senior Director of Investor Relations. Julia, you may begin.

Thank you. Good afternoon, and welcome to today's conference call. Joining me are Mike Cloonan, our President and Chief Executive Officer; and Charlotte McKee, our Chief Medical Officer; Elena Ridloff, our Chief Financial Officer, will also join us for the Q&A session. Following our prepared remarks, we will open the call for questions. Before we begin, I would like to remind you that today's discussion will include forward-looking statements about our future expectations, plans and prospects. These statements are subject to risks and uncertainties that may cause actual results to differ materially from those projected. A description of these risks can be found in our most recent 10-Q and 8-K filed with the SEC as well as any subsequent SEC filings. Any forward-looking statements speak only as of today's date, and we assume no obligation to update any forward-looking statements made on today's call. With that, I will turn the call over to Mike.

Thanks, Juliet, and good afternoon, everyone. Thank you for joining us. Today, we are discussing 2 important updates: the findings from our comprehensive post-hoc analysis of the PreciSION CF Phase IIa trial following top line results announced in August, and how these findings informed our decision to advance SION-451 plus SION-2222 into the next stage of development. Our conviction in NBD1 stabilization remains strong. The post-hoc analysis deepened our understanding of the NBD1 biology and target engagement and identified important trial learnings, giving us a clear and disciplined path forward. Several findings were particularly important. First, we identified 3 participants with PK patterns consistent with dosing non-adherence that disproportionately affected the original results. Excluding those participants, the meaning placebo-adjusted reduction in sweat chloride was up to 8.6 millimoles per liter, which we believe is consistent with NBD1 biological activity and target engagement. Second, even after adjusting for those participants, we identified 2 factors specific to the 4-drug combination of 719 plus Trikafta that may have blunted the observed effect. Lower exposures of the Trikafta components during the 719 treatment period and a possible biological interaction with the potentiator Trikafta, Ivacaftor. Together, these findings may help explain why 719 did not demonstrate its potential when added to Trikafta and why we believe our NBD1 anchored to dual combination without Trikafta or a potentiator could behave differently. Combined with the favorable Phase I safety, tolerability and PK results for 451 plus 2222, we believe the post-hoc analysis supports advancing the dual combination into AscenSION CF, an open-label Phase IIa proof-of-concept trial expected to initiate in the first quarter of 2027. We are focusing our organization and capital on that dual combination program. This includes the difficult decision to reduce our workforce by 46% and other cost-saving actions, which are expected to extend our cash runway into the second half of 2029 and to support execution of the dual combination strategy. Before moving forward, I want to recognize the Sionna colleagues who are leading the company as part of this restructuring. We are deeply grateful for your dedication, resilience and contributions to our mission. I also would like to thank the CF community for the outreach over the last several weeks that has provided inspiration to our team and a strong reminder of the unmet need that remains for people living with CF. With that, I will turn the call over to Charlotte to walk through the analysis supporting our decision and our go-forward plan.

Thank you, Mike. As a reminder, PreciSION CF was a randomized, double-blind, placebo-controlled, 2-way crossover trial, evaluating the 30 milligrams of 719 twice daily when added to physician-prescribed Trikafta in adults with CF homozygous for F508del. After screening in a 14-day Trikafta run-in, participants received 14 days of 719 plus Trikafta and 14 days of placebo plus Trikafta in a randomized order with a 28-day 719 washout between treatment periods, 14 participants completed both periods. The primary endpoint was safety and tolerability, which we described when we announced the top line data. Secondary endpoints included pharmacokinetics and change in sweat chloride, an important biomarker of CFTR function. Two sweat chloride analyses were prespecified in our protocol. The primary analysis was change in sweat chloride from study day 1, and the secondary analysis was change in sweat chloride from the baseline of each treatment period. We believe the secondary analysis using the baseline of each treatment period is most appropriate for this data set given the PK and sweat chloride variability we observed. Turning to Slide 6. Following announcement of the top line data in August, we conducted a thorough, disciplined post-hoc review of the trial, as Mike mentioned. We evaluated multiple dimensions of the data, including PK and sweat chloride patterns in individual participants and in subsets the potential impact of baseline and other characteristics and exposure response analysis. There are 3 factors we believe provide a reasonable and consistent interpretation of the data and potential confounders that may have impacted the results. Three participants had patterns of drug levels and sweat chloride values at key study time points that were consistent with dosing non-adherence and we believe confounded the trial outcome. When we adjust for those participants, we see what we believe is evidence of target engagement and NBD1 activity, albeit still at a lower level than what we predicted. Second, we saw consistently lower levels of all 3 components of Trikafta at the end of 719 treatment period. The goal of this trial was to evaluate 719 on top of stable Trikafta levels, and this is not what consistently happened. While the implications are not completely clear, this finding suggests that lower Trikafta exposures may have reduced the impact of NBD1 stabilization. Finally, we've reviewed the nonclinical data in light of the trial outcome, which led us to an emerging hypothesis that combining 719 with Trikafta and specifically with a potentiator may have unexpectedly blunted the ability to improve CFTR function further with an NBD1 stabilizer. Slide 7 shows waterfall plots of individual participant level sweat chloride changes for all 14 participants calculated from each treatment period baseline. On the left is the 719 period, the placebo period is in the middle and placebo-adjusted changes are shown on the right. The 3 outlier participants sweat chloride changes are shown here in the gray bars. On Slide 8 are more details about these 3 participants. Two had unmeasurable or unusually low levels of the components of Trikafta at the start of the placebo period after the 719 washout and expected levels of Trikafta at the end of the placebo period with very large improvements in sweat chloride over this period. We believe these participants placebo effects were consistent with non-adherence with dosing at the beginning and confounded the trial results. A third participant had unmeasurable 719 levels at the end of the 719 period, which was inconsistent with levels from the other participants and from healthy subjects given the same range of doses. And we believe this participant data further confounded the trial outcome. Importantly, we did not identify similar confounding patterns in any other participants. When we remove these 3 PK outlier participants in a post-hoc analysis shown here on Slide 9, and we saw what we believe are consistent patterns in the data that indicate target engagement and NBD1 stabilization. During the 719 period here on the left, most participants had a numerical reduction in sweat chloride, while during the placebo period in the middle, sweat chloride changes appear very balanced, as you would expect. In this post-hoc analysis, the placebo-adjusted changes also trend towards an improvement in sweat chloride. We further probe this post hoc sensitivity analysis using the model specified in the trial, MMRM or mix model for repeated measures shown here on Slide 10. The dark blue bars represent this prespecified sweat chloride analysis in all 14 participants. The primary result as we reported in August, was a mean change of minus 1 millimole per liter. In the prespecified secondary analysis using each treatment period is its own baseline, the mean change was essentially the same, minus 1.1 millimole per liter. The teal bars represent post hoc analyses showing the impact of removing the 3 participants with outlying TK. The top teal bar shows the result using study day 1 is baseline, a mean sweat chloride reduction of 3.8 millimole per liter. The bottom teal bar shows the result using each treatment series baseline, which we believe best accounts for observed variability, a mean reduction of 8.6 millimole per liter. We believe that this post hoc analysis provides a less confounded view of the data and shows evidence of biological activity from NBD1 stabilization. On Slide 11, we show that the mean trough or pre-dose exposures of all 3 components of Trikafta, or ATNI, declined on average over the course of the 719 period, while ETI levels were generally stable on average during the placebo period. We mentioned this observation when we first reported the data. The dark blue bars represent the mean changes in ETI levels from left to write Elexacaftor, Tezacaftor and Ivacaftor in the placebo period and the teal bars represent mean change in ETI levels over the 719 period. The 3 PK outlier participants are not included. As you can see, ETI concentrations remained relatively stable on average during the placebo period while they declined by approximately 25% to 30% on average over the 719 period. While we could not quantify the potential impact of this finding, this is not what we intended or expected when we designed the trial. And we believe it's possible that lower exposures of some or all of these components could have reduced the incremental effect of adding NBD1. Importantly, this specific 4 drug complexity will not be a factor when 451+2222 is studied as a stand-alone dual combination. Before we describe our emerging biological hypothesis, here on Slide 12 is a brief refresher on CFTR and CFTR modulators. CFTR is a multi-domain protein, the most common mutation, F508del, resides in the NBD1 region and creates multiple defects in protein processing, trafficking and function. A robust set of scientific literature supports the importance of NBD1 for proper CFTR folding trafficking and functioning and none of the approved modulators directly binds and stabilizes NBD1. Approved modulators include correctors and potentiators, correctors generally address CFTR folding, processing and trafficking and potentiators are designed to open CFTR at the cell surface. In Trikafta, Tezacaftor is a TMD1 directed corrector, Elexacaftor is an ICL4 direct corrector and Ivacaftor is a potentiator. Prior to the PreciSION CF results, our guiding hypothesis was straightforward. Based on what has been observed with approved modulators to date, all our nonclinical data shown here on Slide 13, and demonstrated that adding NBD1 to Trikafta should improve CFTR function. Data with NBD1 in dual combination across all our nonclinical assays are shown here on the right. After the unexpected clinical results, we conducted a thorough review of all of our nonclinical data to inform the path forward. One emerging hypothesis relates to high levels of pre stimulated current in the CFHBE assay when NBD1 is added to Trikafta. We had expected the totality of the nonclinical data to be associated with increased CFTR function based on correlation with the approved modulators. However, this 1 component across our nonclinical data differs from the wild-type pattern when 719 is added to Trikafta and the dual combination in this case is similar to wild type. What PreciSION CF may have taught us is that creating a high level of pre-simulation current may lead to dysregulation of CFTR and might have limited the ability to improve CFTR function when 719 was added to Trikafta. I'll walk through the CFHBE assay data that led us to this working hypothesis beginning here on Slide 14. These are representative tracings from human bronchial epithelial cells or HBEs, from a wild-type normal donor on the left and from a CF donor homozygous for F508del on the right. The x-axis shows time and experimental conditions, the Y-axis shows chloride current or CFTR activity. In the wild-type donor on the left, CFTR activity is low before stimulation. The channels are activated with stimulation, and we see a sharp increase in current. Current then plateaus until CFTR is inhibited. This is a normal tracing in the CFHBE model. The gray tracing on the right shows what untreated F508del CFTR looks like, basically flat throughout and the magenta tracing shows the response to ETI, the components of Trikafta, which has demonstrated clinical activity here at maximum exposures or Emax. There is some activity before stimulation consistent with some open CFTR channels and increase with stimulation, plateau and a fall with inhibition. In this model, in general, the level of CFTR activity observed with the components of Trikafta is lower than wild-type activity. Here on Slide 15, the turquoise tracing shows 719 when added to ETI at 719 Emax. There is more activity before stimulation and a larger peak and plateau response. This pre-stimulation current increases with increasing concentrations of 719. For example, pre stimulated current with the 30-milligram concentration of 719 is about half what is shown here at 719 Emax. We consistently see higher levels of pre-stimulation current when a potentiator is present in any of our NBD1 combinations, but not in combinations without a potentiator consistent with the known mechanism of a potentiator opening CFTR channel. What the clinical data may have shown us is that the CFTR behavior when NBD1 was added to ETI may have led to disregulation of the CFTR channel that blunted the driving force of chloride ion flow, limiting the ability to achieve the expected improvement in CFTR function with 719. This remains a working hypothesis, but we believe this provides a plausible biological explanation for why the clinical results were different from what we predicted. This hypothesis has important potential implications for our dual combination, which does not contain a potentiator. The blue tracing here on Slide 16 shows 451+2222 at Emax in this model with the CFTR channel here behaving more like wild-type. Based on this emerging hypothesis, these data suggest that the dual combination may modulate CFTR in a more physiologic way than when 719 was added to Trikafta. Taken together, our post trial findings reinforce our view that Precision CF does not provide a definitive read-through to NBD1 biology. They also strengthened the rationale for evaluating 451 + 2222 directly in the proof-of-concept AscenSION CF trial without the added complexity of combining with Trikafta or a potentiator. I will now turn to our next steps for the 451 dual combination program. Before we discuss our plans for advancing the dual combination, I want to reemphasize the clear need for additional treatment options. Current CFTR modulators have transformed care, but more than 2/3 of patients on today's triple combination standard of care do not reach normal CFTR function as measured by sweat chloride and many continue to experience incomplete clinical benefit or tolerability challenges, and their alternatives are very limited. We continue to hear this desire for options directly from the CF community even in the weeks since we announced the clinical results. Slide 19 is a reminder of the outcome of the Phase I trial of 451 in combination with the TMD1 corrector, 2222, and the ICL4 corrector, 109. A total of 120 healthy subjects were dosed for 14 days across 10 cohorts divided evenly between the 2 combinations. We identified combinations of both 451 plus 2222 and 451 plus 109 that were generally well tolerated and exceeded our PK targets at go-forward doses. Based on the totality of the data and the level of target coverage achieved, we selected 451 plus 2222 as our preferred dual combination. The exposures of both drugs were similar to what we observed when each was studied alone. Slide 20 is a reminder of the safety and tolerability profile of this combination. We selected 451 twice a day and 2222 once a day as the preferred regimen. All 451 cohorts with 2222 dosed once daily were generally well tolerated with all treatment adverse events mild to moderate in severity. There were no serious adverse events and no treatment-emergent events related to elevated liver function tests, 1 participant discontinued dosing due to a moderate rash. Turning to Slide 21. We are excited to advance this proprietary dual combination into the AscenSION CF Phase II proof-of-concept trial, an open-label study in adults with CF homozygous for the F508del mutation. Participants will switch from tricastid to the dual combination for 28 days. The study will assess sweat chloride, safety and tolerability and PK to evaluate preliminary clinical efficacy and safety data with the dual combination in people with CF. We will be incorporating lessons from the PreciSION CF trial, particularly in relation to dosing adherence, sweat chloride variability and PK monitoring. We plan to start AscenSION CF in the first quarter of 2027. Personally, I consider it a tremendous privilege and responsibility to evaluate the first NBD1 containing dual combination in people with CF with the goal to create meaningful options for patients. I know from experience how life-changing transformative medicines can be in CF, and I am committed with the dedicated team here at Sionna to proving the full potential of this novel mechanism. With that, I'll turn the call back to Mike for closing remarks.

Thanks, Charlotte. Turning to Slide 22. We the post-hoc analysis Charlotte presented explains why we believe the dual combination may behave differently. 451 plus 2222 represents a potentially differentiated profile for people living with CF. Our objective now is to determine whether this combination can deliver meaningful CFTR benefit and provide an alternative to today's triple regimen standard of care. The favorable Phase I tolerability and PK results provide the foundation to answer that question directly in people with CF. AscenSION CF is designed to be an efficient proof-of-concept study that will evaluate preliminary efficacy and safety and help define the potential clinical profile of the combination. If successful, that profile could create significant value in an approximately $13 billion market and more importantly, provide an additional option for people with CF. We are now moving from analysis to execution with a focused development plan and the financial resources to reach important Phase II milestones. As the only company advancing an NBD1 stabilizer in clinical trials, we remain committed to pursuing this opportunity with scientific rigor and urgency. We expect to initiate AscenSION CF in the first quarter of 2027. I want to thank the participants in our clinical trials and their families. Our investigators and site teams, the Cystic Fibrosis Foundation and our advisers, our shareholders and the entire Sionna team. We are committed to making a difference in the lives of people living with CF and their families. And we believe moving forward with a dual combination of 451 plus 2222 is clear evidence of our commitment to the CF community while we continue to follow the science. Thank you all for joining us today. I will now ask the operator to open the call for questions.

[Operator Instructions] Our first question will come from line Ritu Baral from Cowen.

I wanted to just go back a few slides and make sure I understood the hypothesis on the components of the CFTR. I think if I got this right, you believe that the binding at the NBD1, together with the binding of the 1 subunit results in some sort of inhibition. Is that a -- I'm sorry, the potentiator results in inhibition, is that the result of some sort of facial obstruction of the CFTR channel by the TMD2 and the NBD1 interaction? Is it potentially chemical or charge based? And what made you think of the potentiator as the driver of this interaction rather than something about TD that could potentially impact 2222?

Ritu, I'll start. This is Charlotte. So across all of our data, we see no evidence that there is some sort of steric for binding interference between 719 or the NBD1 stabilizers and any of the components of Trikafta so that we do not believe that this is a blinding interference. The great stimulation current that we described, we do understand this is a complex [indiscernible] hypothesis. So I appreciate you bearing was walking through with it, so that pre-simulation current is actually consistent with what a potentiator does, which is open channels. And so that is why the hypothesis that this is related to combining with a potentiator is -- makes sense to us. So it is really -- if you think about it, an analogy is the water flow of force of water flow on either side of a dam the voltage and the current that the cell needs to maintain in order to drive on through the CFTR channel is really just like that dam. And what this pre-stimulation current may represent is basically a blunting of the difference in the levels of the voltage in the current along this analogous to the different levels of water on either side of the dam that limits the ability to generate the kind of driving force that you would need.

Ritu, maybe just to add too about the potential or how he knows the potentiator or not TMD1 or ICL4. So we didn't show it in those slides today, but we have run the same model with 719 or 451 when we added just to Elexacaftor and Tezacaftor so not with Ivacaftor, not with a potentiator and we don't see that same level of pre stimulation of the current. And so then you clearly see it when you add 719 or 451, any NBD1 stabilize or to ETI, we see this distinct difference when the potent is in play, not when it isn't.

Our next question come from the line of Chris Raymond from Raymond James.

Congrats on turning this around so fast. I guess just a question on AscenSION. I know it's early here, but this is not the same thing as the initial look. So I would suspect that 10 millimole reduction is not sort of a bar for success. But how are you thinking about -- is it just not inferiority to baseline because it's a simple switch. There's no wash out, right? So I guess that, and also, patient compliance with a trial like this, I guess, it's a little bit of a surprise. Just maybe what specific learnings you're incorporating into Ascension to make sure you have adherence.

Yes. Chris, this is Charlotte. I'll start with the second one. So some of the specific earnings we have taken away with respect to adherence includes variability. And so what will specifically be thinking about with respect to attention, CF is one, both statistical and other ways to reduce the variability that some of these PK and other changes really represented. And we will be looking, as you pointed out, we'll be looking at all sorts of ways to help both the trial sites and the patients maintain strict adherence to the extent that, that does appear to have been a compounder may be added to variability.

And maybe just on that, Chris, on the compliance or not adherence. So 2 of those participants that we talked about were on Trikafta. And now is not a study drug right? We were not controlling for the Trikafta part. Only 1 of the patients was at 719 that looked like they had non-adherence. And the adherence is also important and it happens, right, because of when we were measuring sweat chloride on this study. And so if it happens later at that time, they have more an impact. This was a small study. And unfortunately, a couple of patients can make a difference. So that's an important thing here to learn as we apply that suspension. Now we can switch to the first question.

I have forgotten the first question.

5 That was about how we measure...

Bar for success.

To colonize the medical at base line.

Right. So the vision is, so this will be 28 days of dosing. And so the vision is to look at the change from baseline. And there -- of course, everything we were just talking about in terms of variability and assessing the baseline becomes really important, and we'll be looking for mechanisms and statistical ways to ensure that that's as tight as possible. And then over the course of 28 days, any of the PK, for example, residual Trikafta should be long gone. And our intention then would be that the readout in sweat chloride, for example, at the end of that 20 days would represent the impact of the dual combination.

And our next question come Paul Matteis from Stifel.

I appreciate it. On the lower Trikafta exposures, can you talk about how variable that was across patients? And is there any analysis you can do that looks at whether or not lower tract exposures or inversely correlated to activity? And then second, I just wanted to clarify this whole dynamic with the primary endpoint and the secondary end point Obviously, the number you're getting to 8.5% is on the secondary endpoint for analyzing sweat chloride. So maybe go back, like why was the primary endpoint structured as it was that's showing less than 4%, which is considerably less. So why is the 3% to 4% number not is the right number to focus on why should we be focused on the 8.5%?

I'll get the second 1 first.

Yes. Yes. So Paul, this is Charlotte. I'll start. So the analysis and again, this is sort of statistical e-mail complexity, but in a 2-way crossover. So the expectation in a 2-way crossover is that in those -- the baseline and the -- whatever evaluated at baseline across those 2 different periods basically remains the same or are very stable. And we did see, as we've said, there was simply more variability from the 2 baselines over the course of even we'll see the Trikafta-only measurement points across the study and there seems to be some emerging data after we had powered and designed the study indicating that some of those both Trikafta PK and some of the sweat chloride may be more variable than the data we had at the time they indicate. So it's really just a 2-way crossover study is the only thing is that the clearest and the most sort of the purest analysis is you take the day steady baseline because you're expecting everything to be the same at baseline going across the trial. But since we did not see that, that the 2 baselines were different, even though nothing else was different about what the patients should have been on and we're taking. That is why I'm using each treatment period as its own baseline, especially in this post-top situation, we believe that represents the data. We really -- when we talk about those specific numbers in that post-hoc analysis, we really think of them more as a range and we're not trying to identify what the precise, no pun intended, outcome of the study is. What we really have been interrogating is do we see any evidence that there was a biological impact. And that range of, frankly, roughly 4 to north of 8 tells us that their it was likely to have been some biological impact of modulating NBD1 and also the different patterns that we see in the waterfall flood and will give us the same sense.

Yes. That was the first thing we had to address Paul, going deeply really needed for ourselves to see that there was some target engagement NBD1 biology activity here, and that's why those 2 measures for us were important to see whether it's the minus 4 or the minus short and we do tend to think given the results of this study and what we saw from both a PK variability and Sweaty the secondary analysis may be more represented even though we designed the study to go in back to day 1. So that was really important for us to first answer that question around NBD1 target engagement. And then your other question was related to the ETI concentration. That's when we went next, which is, okay, we saw evidence of NBD1 biology and target engagement, but it was not to the level. Whatever number you use, it's not to a level we were expecting. It was still lower. So what else could be driving this when we have such strong conviction in the biology, that's where the ETI concentrations came into play. And what we saw consistently, as Charlotte outlined, when you look at the placebo on average, they basically stayed pretty consistent over that period of time when they were on Trikafta plus 719, on average, they saw a drop of 25% to 30% across ETI, and so the challenge we had, as you probably remember, we didn't have very robust PK information. This was trough level data. So it was really hard to correlate back to models and again, with only 11 patients once you took out the outliers to get at a quantification of what that concentration would mean in terms of sweat chloride, but 1 of the hypothesis was that if we are bringing down sweat chloride with NBD1, but there's upward pressure because ETI's concentrations are not where they were before, there's a counterbalance in there or reduction potentially in NBD1. So again, we couldn't quantify specifically but we obviously observed that pattern with ETI. And as Charlotte said, that's not what we thought we were going to get in this study. We thought patients would be stable throughout rat there's normal variability, but there was a very different pattern between placebo and Trikafta and 719 and Trikafta.

Our next question will come Yatin Suneja from Guggenheim Securities.

Just 2 quick ones. So number one, I understand this is sort of duration, but were you able to look at the FEV lung function for these patients at all. Anything that you can share around that? And then specifically with regard to the 2 patients that have a pretty robust change in sweat chloride, any additional information you are able to share anything about their baseline sector level and sensitivity to the combo -- sorry, to the add-on?

Yes, I can -- this is Charlotte, I'll take that. We didn't see any trends in FEV1. As you know, we've said from the beginning, this was a 2-week study. So we weren't really expecting to see anything in FEV1 especially with this kind of lower-than-expected effect in this range, we did not see anything. I would say the 2 patients with the placebo, what we believe are confounded placebo effects, there was nothing different, specifically about their baseline characteristics. Really, what we described was the -- were the key findings and really was that post washout start the placebo levels of Trikafta that were really strikingly low. And then which -- not just for the group, but for these patients participated specifically. And then at the end of the placebo period, those levels came right back up to where they had been for these participants at the start of trial.