Oliver Sartor: Hi, I'm Dr. Oliver Sartor with UroToday, and it's truly a pleasure to be able to talk with DNA repair issues with Ken Herrmann, who's the chair of nuclear medicine at Essen, everybody knows Ken, and Katharina Lückerath, who is the professor of biology at Essen, who's a specialist in many things, including radioligand interactions with the biology of tumor. So welcome, Ken. Welcome, Katharina.

Ken Herrmann: Thank you. Super excited to be with you.

Katharina Lückerath: Thank you. Also very excited.

Oliver Sartor: Thank you, Katharina. All right. So I'm going to start off with a question that I believe is fundamental. We have been giving radioligand therapies for many years. The predominant theory, not the only theory, but the predominant theory about how they work is that they damage the DNA, and the damaging DNA leads to some sort of apoptotic or other forms of cell death as a result of the radioligand delivering the energy of the isotope into the DNA and damaging it, and that leads to therapeutic response. What we do not know is how the cell responds, how the cell responds with DNA repair to this insult that we're trying to deliver. And I would love to hear, and Katharina, we're going to start with you, can you explain the basics of DNA repair in response to a radioligand therapy? And that's question number one. Help our listeners understand this concept between radioligand therapy and DNA repair.

Katharina Lückerath: Okay, I'll give it a try. So the radioligand, as you said, emits the radiation that can directly interact with DNA or also other cellular structures, lipids and so on, or it can hydrolyze water to generate reactive oxygen species that can then damage the DNA. And that would result, depending on which emitter we choose, in single-strand breaks, if we are lucky, DNA double-strand breaks. That's mostly associated with alpha particles, but not exclusively. And we can also have other types of maybe less complex DNA lesions. And then, of course, the cell needs to deal with the repair in order not to enter the mitotic catastrophe and cell death. And it does so by first sensing the DNA damage, and common markers are, for example, the phosphorylated histone gamma-H2AX.

And this then recruits DNA repair proteins, and the cell gets to choose which DNA repair pathway it's going to use, and that will depend on different factors. And for now, we don't really know, is it the homologous recombination that is predominant, or the faster but more error-prone non-homologous end-joining pathway, or one of the other options that are available to the cell. And the evidence so far points to non-homologous end-joining being an important repair pathway, because it's fast, because it's cell cycle-independent, and also data showing that DNA-PKcs inhibition in the context of radioligands is effective kind of suggests that this could be a very important pathway in the cellular response. However, the homologous recombination will also be important. And then, we end up with either successfully repaired DNA or misrepaired DNA, that will also then trigger certain responses downstream.

Ken Herrmann: So can I ask a question there?

Oliver Sartor: Of course.

Ken Herrmann: When we think about Pluvicto treatment, and we know around 50% in the VISION label do not show increase of greater than 50%. I know that you now actually biopsy, or re-biopsy, but you are the one leading the project, biopsy tumors after two cycles, especially on the ones not responding. Do you have a feeling by now how much of these patients not responding or adequately responding is because of, let's say, a very fast, a very efficient DNA repair mechanism, or which ones are just not really sensitive overall?

Katharina Lückerath: We don't have a super solid answer on that yet, but from what we know, I would say that with Pluvicto, so with lutetium as a beta-emitter, we have a very fast induction of DNA damage, but then also a relatively quick resolution. And then, we will enter a state where most likely the DNA damage induction and repair balance each other out, so at the time we are inducing DNA damage, we will also see repair of DNA damage. And that allows the tumors or the tumor ecosystem to cope with this type of radioligand therapy quite well, and this is why we see some patients responding and some patients are progressing eventually and not responding so well, because we are just not very effective with our type of therapy. And that also means that we have to think a little bit differently about the radiobiology of our radioligands, because if we say that the induction of DNA damage and the repair of DNA damage balance each other out, then it changes this classical linear quadratic model of radiobiology and makes it into more of a linear model, and then we need to approach how we optimize Pluvicto treatment or other beta-emitter treatments.

Oliver Sartor: Katharina, when you say it starts fast and ends, give me some kinetics. Does this start 30 seconds after the damage begins and it lasts for a day, or two minutes or three days or eight minutes? So give me some idea about the kinetics of the DNA repair. And then, follow-up is, is the cell uniquely susceptible to DNA repair inhibitors or additional doses of DNA-damaging agents during this critical period of DNA repair? So what are the kinetics and what can we do about it, that's the big question.

Katharina Lückerath: Okay. Very good question, without full answers for now. So I think we don't have a lot of human data, we mostly have cell line and mouse model data, so that's where my answer is going to be based on. DNA damage starts as soon as the radioligand is present in the tissue, because that's where the decay happens. And in our hands, and there are also other published data from a group at Erasmus, suggest that the DNA damage induction with beta-emitters peaks early on, within the first hours after the radioligand reaches its target, and then it declines within probably the first 24 hours. It doesn't go back to zero, but it persists on a rather low level. Most likely, this is different for alpha-emitters. But for beta-emitters, it looks like we have an early peak, maybe in the first day, and then we have a lower level of persistence for some time, for a couple of days, during which we probably see this balancing out of each other, the induction and the repair. I think the biggest window of opportunity might be really early on.

When we induce all this chaos in the cells, the cell doesn't really know how to deal with it, so in my view, it has to organize a defense. First, it's exposed to this stress and first needs to have this reaction and see, okay, what is happening, what should I do? And then, once it has organized its defense, then things are a little bit more orderly, but maybe very early in the beginning, we have this window of going in with a DDR-targeting agent. However, all this DDR targeting, so DNA damage-repair targeting, it's so logical, but it's so not trivial, in my opinion, because we, and a lot of other people have tried it, with variable and at times inconsistent outcomes, because if we target one pathway, there are others that can compensate. Also, thinking about the balancing out of each other, then do we really should only target the one DNA repair pathway that's most likely balancing this out, or do we just need to go in with higher stressors, do we need to go in with higher initial doses or do we need to have much closer administrations of... I don't want to say fractions, but don't space the cycles out so long.

Ken Herrmann: So Katharina, you said gamma-H2AX is a biomarker for DNA damage. Is there any way to do in vivo visualization quantification? Because if I remember correctly, the only thing is you have some cells which you can count, you can label them, but is there any way to do this in vivo?

Katharina Lückerath: There is one radioligand that's been described for imaging of DNA damage, but I don't think this is currently being used in the clinical setting for what I know. So what we do is we do liquid biopsies, and we analyze markers for DNA damage induction in PBMCs at various time points, and I think there is one clinical trial ongoing where they do the same thing. So you draw blood, you look at DNA damage in PBMCs and not in the tumor tissue. So kinetics in PBMCs can probably tell us also something about other tissues, but especially a tumor might deal very differently with DNA damage than the PBMCs. So when we see induction, this could maybe be most similar, but the kinetics of repair might be quite different.

Oliver Sartor: Got it. Ken, when we were at ESMO, the presentation from Shahneen Sandhu on the LuPARP study follow-up and survival was to me very impressive, I want to get your opinion. Just for the listeners, she did the LuPARP study in combination with a whole variety of Australian investigators, basically giving the PARP inhibitor, olaparib, at various doses, and came out with a day -4 to day +18 full-dose olaparib, but on a truncated basis around the radioligand therapy. But the survival was up at 33 months median per dose level seven, eight, nine. Ken, I'd love to get your take on that study. I thought it was important, but I didn't fully understand it, but want to hear your take.

Ken Herrmann: No, I think the LuPARP study is really fantastic work. So in the beginning, everyone was thinking about this would be the perfect combination, because it's a DNA inhibitor together this must work. Then there was some early data, and when we looked at the PSA responses, it was not as impressive as we hoped for initially, there was quite some toxicity. I think it took them quite some time to figure out what is the right dose. I really have to say now, that a few years later, and the data she showed now was really compelling, especially with this new olaparib dose, which is much better tolerated, I think the results now look super, super impressive.

Of course, this was just a poster presentation, this was not a full oral presentation, so we did not have time to get all the details. But even in the one-to-one interaction after this with Shahneen, I'm currently super, super excited, I really hope that she makes this data available. And I think, again, from being first super hot, then to being so-so, lukewarm more, I think it's back to being super hot, the combination. We need to better understand which are the patients who benefit the most. But I definitely think that, again, coming back about the idea of combining with maybe something which is a DNA damage pathway-repair inhibitor, I think this is now the time again to dig deeper.

Oliver Sartor: Thank you. Katharina, were you surprised by the PARP inhibition, which I would not have intuitively thought would be the best combination? I would've thought maybe DNA-PK or maybe an ATM inhibitor or something. But were you surprised by this PARP inhibitor interaction, which is probably more oriented towards the single-strand DNA than the double-strand?

Katharina Lückerath: I wouldn't say I'm surprised. Of course, it was kind of like an obvious choice, because it's approved, although for BRCA-mutated or HR-deficient patients. So I'm more surprised that it's been used in a broad population of genetically unselected patients and it's doing well. But in general, it's mostly, as you said, associated with the repair of single-strand breaks. But then, Pluvicto and lutetium can induce a lot of single-strand breaks, that can become converted to double-strand breaks, there will also be double-strand breaks. But overall, we just increase the number of lesions that the cell has to deal with, and at one point, we will overwhelm the cell's capacity to deal with it, because even though we might be saying, "Oh, double-strand breaks are the lethal hits," if you have enough other DNA lesions, it's going to be detrimental. So there's definitely potential you enhance the chance of the cells to become overwhelmed or to be converted to double-strand lesions.

Oliver Sartor: Fantastic. I love these discussions. Okay, I'm going to switch topics straight onto DNA repair. So one of the guys I really like is Alex Wyatt, who's analyzed in some detail the TheraP data, and he was looking at extreme responders within the TheraP data set, and found that those individuals with a double hit on ATM, a loss of the ATM protein was presumed, even though it was not at the protein level, it was at the genetic level, but he found extreme responders for those with a somatic ATM deficiency. I wonder, either Ken or Katharina, your thoughts about ATM in the pathway of DNA repair, and whether or not an ATM inhibitor might be an interesting way to go?

Katharina Lückerath: ATM and ATR are considered the two main effector kinases that are recruited early on when the cell... So H2AX would be the sensor protein, and then proteins like ATM and ATR get recruited and orchestrate the response that's following. So they're really higher up in the cascade of the response to replication stress and DNA insults. So yes, there is potential for an ATM inhibitor, but then, of course, also for an ATR inhibitor. But still, there will be compensatory mechanisms, and we don't have an approved ATM inhibitor yet, and they can be relatively toxic, I think. So that might be important to balance or to really also figure out when to give which drug in relation to each other and how to create the synergy in the best possible way with the least toxicities.

Ken Herrmann: So I was laughing when you mentioned this, Oliver, because it was 2013, 2014, UCLA, and I want to give a shout-out to Caius Radu, because he started to work on the ATM/ATR inhibitors in combination with radioligand therapy back then, to be honest. And I was still surprised that it didn't come through. One of the main reasons is that despite the promising interaction principle, at least the ones which happened in humans were quite toxic. But overall, I'm not surprised at all, and I still think it's a very valuable pathway to pursue, but it's not as easy, as we saw, to develop the right pairing.

Katharina Lückerath: I think that's exactly how it is for all the DDR inhibitors. It's so appealing, because we always think, okay, we do the DNA damage, so we should prevent its repair. But also, from our experience, it's not as trivial and there are many more factors to consider. There have also been reports on how this then interacts with the immune system, and maybe we also have to combine it with immunotherapy in this setting, and it's a lot about how we time it, which doses we give, how long we give each drug, and I think that's how we will solve how to combine agents that interfere with the DNA damage and replication stress response and the radioligands.

Oliver Sartor: Guys, we're going to need to wrap it up pretty soon, but I'd like to know, from Ken and Katharina, any brief statements to sum up or encapsulate our discussion today on DNA repair inhibitors and radioligand therapy. So Ken, any final thoughts and comments?

Ken Herrmann: Figuring out the right partner to combine for lutetium will definitely include DDR inhibitors, and I think the biggest challenge will be the ones which are not too toxic to be combined with. And based on what we have seen, I still think the data for LuPARP looks super, super interesting. So as long as we don't have anything better, it's for me the most promising starting point.

Katharina Lückerath: I agree with that. And I add the sequencing and scheduling to the mix, and also the dose of the radioligand we are giving, and maybe the immunotherapies that should be considered in this context as well.

Oliver Sartor: Thank you. Guys, I've thoroughly enjoyed the conversation. I suspect that this will be very popular among UroToday listeners, thank you so much for contributing. Thank you for your contributions to science, thank you for your contributions to radioligand therapy. Thank you for what you're going to do in the future to both Ken and Katharina.

Katharina Lückerath: Thank you for having us.

Ken Herrmann: Bye-bye. Thank you.