Yes, absolutely, happy to. And I think there are a number of elements that could impact a safety profile that we think -- and it's part of the reason we're excited about this approach. And I think at a high level. We see the opportunity to really lower the delivered dose because you're going to get a higher fraction of circulating dose into the brain if the technology works as it's intended, which I think is overall a good thing. I think when you think about safety profiles, you can look at it from the perspective of target associated things like ARIA. So that's obviously something that's top of mind for anyone developing an amyloid-based antibody approach. We're overall pleased with the profile we see for Siburnatag when it comes to REA rates, and that comes from our Phase I study, which we talked about a number of times. But it is interesting when you look at the work coming from the Roche Group that in their case, their antibody, the ARIA rates are much, much lower when you convert gasierimabto en -- and there's scientific hypothesis around that based on where the transferrin receptors are localized, which can change the entry points for where the antibody is getting across the blood brain barrier. So that's an interesting hypothesis and offers the opportunity to even further enhance what we think is an already attractive profile for the potential for RA risk but -- and then as Dan mentioned, there are the transferrin related risks associated with things like anemia and that's certainly an element that's been reported in the Roche program to date. It's also been seen in other programs targeting the transferrin receptor, something we've thought a lot about and obviously went into our evaluation process. And we've landed with JCR because we believe that based on their clinical evidence, they've seen very low amounts of anemia with their marketed products. And the reasons for why that might be. I think there are a number of different things we can get into, like affinity ranges for the transferrin receptor. But I think also importantly, a number of different shuttles target the transparent receptor, but they don't all necessarily target the same epitope. And so I think -- we're learning a lot about how this technology works, but we see opportunity here to really significantly alter and improve the safety profile that we see with Sber -- and by the way, it's a profile we're already pretty proud of. So we think there's opportunity there. And I think we can talk about Phase II study design. It is early days, as you say. The one thing I would point out there is that Eric and the team really did a fantastic job in putting together a progressive study design for some burnatuk-in Phase I. And I think we can benefit from the approach that we took looking at both fluid and imaging-based biomarkers in AD patients with an EBD products. So I think beyond that, we'll have to see. We've got plenty of work to do before we get to that stage, but we're very proud of the work that the team did in Phase I, and we think that, that's -- there are some insights that came out of that design that we can apply to the next program in the chain. And I'll leave Eric to address your question around what we think about the preclinical studies that are ongoing right now.