Richard Parkes: First one, just going back to Andrew’s comments about risk profile in the Phase III starts and specifically thinking about the bispecifics. Obviously, the CTLA-4, PD-1, you’ve got very strong Phase II data. The TIGIT, PD-1 seems a bit more speculative based on what we’re seeing with other TIGIT antibodies. So could you talk about what you’ve got in-house and what advantages you think that bispecific might have over PD-1 combinations that would be really helpful. And if you’re planning to start data, DXD combinations as well would be helpful there. Then the second question. One of your ambitions is to extend the life cycle of your Lynparza franchise with your Part 1 selective. However, that currently falls outside of Lynparza relationship with Merck. I’m just wondering if you could discuss any plans to bring the asset within that deal and when a decision might be made on that.
Pascal Soriot: Thanks, Richard. So there’s one for Susan and one for Dave, I guess.
Susan Galbraith: Okay. Thanks for the question, Richard. So again, the PD-1 element of the bispecific programs that we’ve got is the same across the assets that we’ve got — so that’s one element of it. And the PD-1 TIGIT doesn’t add a challenge from a safety perspective on the background of PD-1. So obviously, with the PD-1 CTLA-4 dose selection has been important to get that right therapeutic window for the tolerability profile whilst driving the efficacy PD-1 TIGIT, this is a safe combination. And the preclinical data that we have does show some potential for differentiation, although as you’re well aware, the extrapolation of that into the clinic is challenging with these models. So what I would just say is that by having both elements of the combination on one molecule, it does help us with a combination philosophy for other things that we want to put into that.
And we’ll be happy to share more of the plans for that when we start dosing the first patients in the Phase III.
David Fredrickson: Richard, in terms of our plans moving forward around development and commercialization of AZD-5305, it’s really consistent with what we had said in the past. We’re developing it independently, it’s an early development. Now just moving into Phase III, any commercial arrangements, we’ve really yet to decide and we’ll wait for more data. With that said, we’re minded to extend the collaboration and build on the joint success that we’ve had together with Merck on Lynparza. But of course, that depends on agreeing on any terms. But we really have benefited greatly from our collaboration together and the collective work that the 2 teams are doing.
Pascal Soriot: Thanks, Dave. Seamus Fernandez at Guggenheim . Seamus, over to you. Seamus, we don’t hear you.
Seamus Fernandez: Yes. Can you hear me now?
Pascal Soriot: Yes, go ahead.
Seamus Fernandez: Okay, great. So, first question is just on the sort of impact on some of the older products as it relates to the NRDL and the magnitude of the decline that we could see in silicon in China year-over-year. And then also the time when we might see Nexium generics actually introduced in Japan, I know those were 2 fairly large tail products for the company, where we’ll see impacts year-over-year. And I just wanted to get a sense of at least the relative profitability of those products because I think it’s important to gauging just how robust the overall performance of the company is outside of that. And then the second question, just wanted to better understand the choice of stratifying by Trop-2 expression. And if that is something that you’re gaining learnings from in the — from the second-line study or if you think that would apply in the second-line setting versus some of the disclosures that you made for the new Phase III in the first-line setting today?
Pascal Soriot: Thanks, Seamus. So Susan, can you cover maybe first this one? And I will ask Leon to cover the Calquence question. And Ruud, you take the next year in Japan?
