And this is why we both, GSK and us, have been structuring exactly these clinical trials and then certainly also the readouts and to report on this one to give us a full picture. On the IP timelines, the process as we’re talking about here, the claims we have started against BioNTech here in Germany, this is as you can consider a very complex file, which we handed in, which will be responded by BioNTech. And this will take a while in order for the court to look into and to evaluate. There has been no public hearing, which then certainly would give us an insight how the court at least would see this. So with this one, I’m just going to say that the process has been started. It is following the normal rules of the process. And we are waiting for first — from the court and follow the process from here, definitely nothing in this year as much as I can say now.
Jonathan Miller: Okay, fair enough. And then maybe I guess on the onco vaccines, could you just go into a little bit more detail about the usage of non-coding areas for antigen identification? How frequently do point mutations in non-coding regions result in expressed antigen that’s targetable in a vaccine context?
Franz-Werner Haas: Thank you very much for that question. We’ve done the numbers for a couple — for quite a series of tumor samples. And roughly, you could say that out of every junction where one chromosome gets coupled to another chromosome, about 1 in 10 actually results in that activating an un-coding sequence to become part of a gene which is on the other side of the junction, and therefore becoming translatable and also detectable if we look at that in cells that we’ve studied, both cells taken from tumor cells, also cell lines and cells taken from mice as well. So it’s about 1 in 10 cases where it actually results in such an expressed sequence.
Jonathan Miller: For this chromosomal rearrangements, not — obviously not just point mutations in non-coding regions.
Franz-Werner Haas: That’s right. I would say that a point mutation in a non-coding region doesn’t contribute much to the antigenicity, right, because there’s no reason why it will be coding and we haven’t followed that up at all. So expanding the sequencing to the 98.5% non-coding is, in my view, especially interesting for detecting those rearrangements that lead to stretches being expressed and not to the point mutants in those stretches.
Jonathan Miller: And then as a follow up to that, obviously chromosomal rearrangements are relatively frequent in tumors overall. But how often are those rearrangements basal to the tumor, that is to say how often is the tumor clonal for those coding rearrangements across all the cells in the tumor?
Franz-Werner Haas: Yes, those are two different questions. Actually we’re writing a manuscript to give you all the numbers. We’ve looked at many cancer types and also looked at data that were not generated by ourselves, but available in public databases. And indeed, in different indications, the frequencies of such rearrangements differ, sometimes in an interesting way. For example, glioblastoma is obviously poor important mutants, because brain cells are very much protected from the environment, that we find glioblastoma to be relatively rich in such chromosomal rearrangements. So that’s an interesting point to look at indeed, as you point out. The second is the choice of the antigens for designing vaccines in a personalized manner.
