This is Doctor Zimelman who we already introduced uh, Matt's an old friend, and, uh, again, he's the director of, uh, clinical research for both, um, the center and for the uh GU service line. Um, Matt's not only does he get sort of wiser, smarter, and a better clinician everywhere, but his hair gets better. And so in your, in your in urology, this is one of the, you know, one of the reasons we hired Ryan Mark just to compete with medical oncology. Yeah, and so please stay for the question and answer session after this. Thank you, Alex, for that very kind introduction. Um, so yeah, I don't know if I was purposely put at the end, uh, because my talk's exciting or if this is an alphabetical thing, but I'm used to going last, so I'll do my best to make it interesting. Um, so I'm gonna talk a little bit about, um, adjuvant therapy in kidney cancer, which is, uh, sort of become part of our standard of care and a little bit about sort of the, the data and where it stands now. All right, so, um, I'll go a little bit about, uh, the, uh, old days or what we tried with adjuvant therapy in the TKI era. Um, I'll go into the keynote 564 trial and what those results showed and what how that led to our standard of care now, um, and then, um, into some of the other adjuvant studies and how they compared and, and why maybe they were, were had different outcomes. So first, when we think about adjuvant therapy, what, what really matters to us and to our patients, really the ideal goal of giving adjuvant therapy, so giving treatment after someone has had surgery to remove the tumor, is to try to hopefully cure more patients, right, increase the percent of patients that that you're curing. And so ideally you would want to look at overall survival. However, um, it's complicated. Um, disease-free survival really directly measures the effect of the treatment that you're giving. Overall survival takes into account subsequent therapies, other things that may happen. And so it's, it's confounded. And so as a result, a lot of trials use disease-free survival. It's also often quicker to get a disease-free survival endpoint than overall survival. Um, but what's, what's the cost? What are you putting patients through to get to whatever end point that you're getting for, um, both short and long term treatment risks, um, you need to balance against what's the potential gain that you're trying to get. And then when we think about kidney cancer, my patients whose cancer recurs on single agent immunotherapy or some therapy, are they actually getting undertreated if they're destined to, um, to have recurrence where they're ultimately gonna get some sort of combination therapy. And so that's where it gets really complicated. So in the early days of trying adjuvant therapy, we did several, a bunch of trials in the TKI era. This is what we do in cancer. We get a drug that works in metastatic disease and we hope that we can move it earlier and earlier. So we tried in the adjuvant setting. So we did a whole bunch of trials with thousands and thousands of patients comparing a TKI to placebo, and we really got a whole lot of nothing. This is a table. Of the trials that went on really we had one trial at the bottom, the STA trial that showed some benefit in disease free survival and led to an approval. But if you look at the curves here, you see that basically you take a toxic drug for one year and it essentially delayed the time of your recurring cancer for a little more than a year. But when we look at overall survival, there was absolutely no difference. And so we weren't really curing more patients, um, and as a result, there wasn't much um uptake of, of this therapy. So then we got into the immunotherapy era and we saw the benefit of checkpoint inhibitors in the metastatic disease and so we repeated the same thing. We did a whole bunch of trials using different immunotherapy drugs in the adjuvant setting, and here's a list of them here the keno trial, which I'll go into in a little more detail compared to some other trials that did not have the same results. So some details about the keynote 564. Um, so this was a randomized trial, um, I can't read this at all, but of patients who got, um, were randomized after surgery to get um pembrolizumab versus placebo. Um, the primary endpoint was, was disease-free survival. Um, these were relatively high risk patients, um, and this was. The really the top line results. This was the overall survival and essentially you see the patients who got Prembolizumab had a small benefit. You look at the difference at the 48 month mark, about a 5% absolute difference, has a ratio of 0.62, statistically significant. And so there does seem to be a fraction of patients who were not recurring and were living longer as a result of this therapy. If you look at the primary endpoint of the trial, the disease-free survival, again, you see, um, a difference in absolute benefit of about 8 to 9% has has a ratio of 0.72. And in in the world of adjuvants trials, this is actually pretty good results. Um, you know, breast cancer and colon cancer trials approved drugs on, on, on less differences than this, um, and this led to the approval of, of Pembrolizumab. Um, but there's always things to look at, and that led to the uh the Vina Prasad controversy. Um, for those of you who don't know, Vinai Prasad is a medical oncologist at UCSF. He's also an epidemiologist, and he does a lot of podcasts, talking about trial design and and different issues with trials sometimes can get somewhat controversial, and he did a big segment on, on the. Merits of this trial on some things that he thought were some limitations and his really biggest concern about this trial was subsequent therapy with the idea that if you do these global trials and give patients a drug, but some of these patients are treated in countries or areas that don't have access to that therapy and don't get it at recurrence, are you really not testing giving a drug early versus what they get if they recur? Are you really just testing, giving a drug versus never getting a drug? So what's the real, what's the real answer here? So this is the table that was in the New England Journal article, and this looked at the patients who had subsequent therapy, and I can't see these numbers at all, but one of the things to point out in this is that the table actually really just looked at patients who received subsequent therapy, not all patients who recurred. And so if you look at the placebo column and the patients who got anti-PD1. Um, there's a percentage number there, no idea what it is, but, but that was really not all the patients who who recurred and could have been potentially eligible for immunotherapy. It's really just the patients who got some subsequent therapy because some patients do other things. Some patients who recur, we are able to just observe. They have small slow growing disease that we can watch. Some patients can get local therapies like surgery or radiation, and so that can somewhat cloud the field of what patients really get. So in total there are actually 211 patients who had some recurrence in the placebo group versus 165 in the pembrolizumab arm. And so actually if you look at all the recurrent patients versus those who got a drug later, less than 50% of patients who recurred on placebo got subsequent anti-PD1 therapy. So maybe the Io actually has a point. So I tried to break it down a little um for you. And so, um, here's some back of the napkin calculations if you try to pull out the numbers from the trial. Um, and so, uh, these are patients who, um, All right, all right. Who, uh, document recurrence versus subsequent treatment. So this just gives you the numbers who actually recurred. Um, and then this looks at patients who got some any subsequent treatment versus subsequent systemic therapy. So trying to point out that across both arms there was about an even number of patients who got some sort of non-systemic. Therapy treatment got surgery, got radiation, and so this was treated relatively similarly across the two arms. If you look specifically at received subsequent, oh now I totally lost my picture here. All right, received any subsequent therapy, surgery or radiation, you see again these are relatively similar. Um, but if you get into the details, patients who received subsequent anti-PD1 therapy versus those who had documented recurrence was relatively low and probably less than we would have liked. All right. I don't know if we can fix this. Mine is way behind what's up here. I'm not seeing the same slides on here. OK, great, thank you. um. So not all patients, so basically the, the summary of this is there are some patients who don't need subsequent therapy, who we can safely watch and observe, and we've shown that and we're part of a trial that has showed that previously. Um, some patients can get local therapies, which is reasonable and at least in this trial this was treated somewhat similarly across both arms. Not all patients who get subsequent therapy need or will even can or will get immunotherapy and so that's just the reality. So looking for 100% of patients to get subsequent treatment to occur is probably not reasonable, but I do think it's probably a fair criticism in this trial that there were patients who probably could have gotten access to immunotherapy and never did, and I do think that is something that. We need to consider in the benefit of this trial, but this is the complexity of doing, uh, large scale clinical trials, and we're never gonna get a perfect scenario and this is the data that we have and I think as a result it's still reasonable to discuss and offer this to patients, but I think, um, there, there, there is some controversy that's that's somewhat uh relevant for the study. So how about why this trial succeeded where some of the others didn't. So there were 3 other large trials that were that were performed. So the Prosper trial, which looked at perioperative nevolumab, the emotion trial, um, which was similar but looked at a Tesla isumab, and then the Checkmate 914 trial, which looked at both IP Nevo as well as Navalumab. These are all now reported as negative trials. Um, so why is one study positive and the rest negative? What are there reasons that we can point to this? So, um, I'll kind of go into a couple of things that may be relevant. So first, is it something to do with, with the drug or really the target? Um pembrolizumab is PD1 targeted drug, whereas sallizumab is a PDL1 targeted drug. And if you look at the differences just in these trials, of course, with all the caveats comparing across trials, the, the really percent disease-free survival of two years was numerically different and clearly improved in Pembrolizumab. And if you look at the placebo arms of those trials, they kind of perform similarly. So the control groups seemed to do pretty well, yet there did seem to be a better benefit of pembrolizumab. It is suggesting something that we've seen in the metastatic setting. Maybe the PD1 drugs are a little bit more active, at least in kidney cancer. What about the individual drug? Is there a difference between pembrolizumab and volumab? Well, if you compare across these trials, actually the, the numbers are actually relatively similar, and really the the one main difference is the control arm on the keynote trial, the one on the left, seems to be a little bit lower than the other two trials. And so although they fared somewhat similarly, there may be something different in the patient populations in these trials that may have differentiated this. Um, an important point is that in the Checkmate trial, Navolumab was only given for 6 months versus a year of therapy of the other trials, and so that may be one potential um mitigating factor. What about trial design in general? Um, the Prosper trial was a very different trial than these others. This was a perioperative trial where you got nivolumab beforehand in the control arm for one dose and then adjuvant nevolumab afterwards. And there's certainly some pros to this approach, and a lot of other tumors are definitely moving more to a neoadjuvant or sort of perioperative paradigm. Um, with immunotherapy, it does make some sense to give the drug when there's some. Tumor in place. You're trying to get your immune cells to recognize some foreign antigen or something that's not supposed to be there. It's probably a lot more likely to happen when you have a big tumor in place that then the immune cells are stimulated and can maintain that response once the tumor is removed. And there's also a ton of translational potential when we, when we do neoadjuvant trials. Um, however, it can be difficult to do some of these periopererative trials, particularly in kidney cancer. Clinical staging and pathologic staging do not always line up. A tumor can look really big on scans, but when we take it out, it may not be as aggressive as others. And as a result, there were some differences in the pathologic staging and the risk of patients on this trial. Is one dose enough? In the world of kidney cancer we're so used to taking patients right to surgery when we find a big ugly kidney tumor that it was difficult to offer. Patients a longer, more treatment and thus a longer time to surgery, so we only gave one dose and maybe we need more immunotherapy priming than that. And then another important difference in this trial is the use of observation as opposed to placebo. So patients randomized to that arm know what they were on and sometimes savvy patients say, I don't want to get placebo. I can go down the street to the trial, to the other trial and try that, and there are different dropouts in this trial as a result. Um, and then another point I want to point out is, is differences in the population. Um, and so how you design your trial and how you put, um, choose what patients go on can make a big difference in the outcomes that you that you show. Um, so first off, if we look at, um, some risk factors for patients, what their stage is, um, this was slightly higher um in the keynote trial. Um, if you look at non-clear cell, which was allowed in some trials and others where it wasn't, non-clear cell patients, we know. are probably a little less responsive to immune checkpoint inhibitors. So having more of them in the prosper and emotion trial may have impacted the outcomes that we saw. And then if we just sort of look down the list of the Keno trial, there was more T3, T4 disease, higher nodepositivity, no non-clear cell, higher sarcomatoid patients, which we know are patients that are more responsive to immunotherapy. And then there was this weird question about patients who had M1 or sort of one side of metastatic disease, which I want to get into. So both the Keno trial and the emotion trial allowed these M1 NED patients, in other words, patients who had their kidney removed, but also had a site of metastatic disease that could be completely removed. In the cheno trial, there was, this was only 5% of the population, but those patients did really, really well. They all, you know, they seemed to benefit quickly. The hazard ratio was, was 0.28, whereas in the emotion trial where it was 15% of patients, they really didn't seem to be much of a difference. The hazard ratio was pretty close to 1. And significantly in the cheno trial, patients were allowed on if they had resection at the same time or within a year of when their kidney was removed. So these are the, these are the highest risk patients. The patients probably most likely to benefit from getting immunotherapy, whereas in the emotion trial these patients had disease recur later. They had their kidney out over a year ago and then had had metastasectomy. And so these were probably the more indolent lower risk patients, probably less likely to benefit from giving an adjuvant therapy. So in summary, lots of differences in the trials and differences in trial design can lead to differences in outcomes. Um, in the, the PD1 versus PDL one, maybe there is a slight benefit of PD1 targeted drugs, um, giving combined versus single agent PD1, um, maybe the, the, the length of the dose, the duration of the dose matters here. Um, neoadjuvant didn't show benefit, um, at this point, but maybe it has to do with how much we gave beforehand. And again, stacking the deck with the highest risk patients gives you the best chance of potentially showing a benefit. So how do now in clinic, how do we help patients choose? Well, really what I try to do is lay out this data as best I can and try to compare for them the risk versus their potential benefit. One way of doing that and giving a shout out to our urology colleagues um is this website cancer nomograms.com. Data was compiled based on some of the older assure adjuvant trial data with Dr. Crea and the urology team. This is something that's accessible, and I can plug in the individual data from my patients and really show them what their potential recurrence risk is to help them sort of understand their risk. Um, so what are the next steps that we see in adjuvant therapy? We have an approved drug in pembrolizumab, but we'd like to do better, and the future is maybe with different combinations. One trial is this Light spark 22 trial. This is a phase 3 trial that was looking at pembrolizumab plus Balzudaphan, IP2in alpha inhibitor versus Pembrolizumab plus placebo. This is a trial we participated in that's completed accrual, and we're we'll be awaiting the data for this. And then a trial that we have open now, this is looking at pembrolizumab plus a vaccine, and this vaccine is created bespoke from the patient's tumor tissue. And this is an approach that has shown some benefit in a melanoma trial and we're hoping to um to demonstrate in kidney cancer as well as in urothelial carcinoma. So finally, in summary pembrolizumab, pembrolizumab is FDA approved based on results of the Keynote 564 trial. It is something that we're talking about for patients, but it may not be for everyone, and it's really a discussion with patients about their potential risk benefit. It's important to remember that not all patients need immediate immunotherapy, and there are other options for patients who recur. And when we think about the differences in these trials, trial design is really important and probably plays a big role to the results that we see. Thank you very much for your attention.
Related Presenters
