How Do We Treat Smarter? An Interview with Dr Thomas Powles

John Marshall, MD: Hey everybody, this is John Marshall here in Washington, DC, where nothing is normal lately. A little bit rainy today, but we need it, so maybe it’ll wash away some of our sins.

But Oncology Unscripted is fortunate enough today to have an amazing guest, Dr. Thomas Powles from London. So, it’s a little later in the day where he is than where I am. Tom has really led the charge in innovation in treatment with GU cancers. And I was first attracted and wanted to interview him not only because he’s a famous social media guy—he's really had a major impact on information sharing in our cancer community—but really because of the clinical trial that he oversees. But first off, Tom, welcome. I’m hoping you’re having a lovely day in London. We love it there—our kids live there. Do a quick introduction of yourself for our audience.

Professor Thomas Powles, MBBS, MRCP, MD: Well, John, firstly, I’ve followed your work for many years, and I’m incredibly honored to be here with you and the amazing things that you’ve done, and I’m very happy to join you.

I’m Tom Powles. I’m an oncologist in London. I’ve led a number of different trials—some have been successful, many have not been successful—and I also have collaborated with a huge number of people. And those trials which I have led have clearly been a huge team effort.

And we are making a, as you said, in GU cancer, a big charge towards curing patients with bladder cancer. We’ve done very well in kidney cancer, but we haven’t quite got there yet. And, of course, prostate cancer is a complex beast, but there are huge numbers of new therapies. So, really exciting times in GU cancer.

John Marshall, MD: Totally. So, in MRD testing—let’s jump right into that—I think of adjuvant therapy as one of the most inefficient things we do. On one level, some people were already cured by our surgical team members, if it truly is postoperative adjuvant. On another case, some patients are not cured despite receiving therapy; they have microscopic metastatic disease and relapse anyway. Then there’s that wedge of people who we do fix—we treat 100 people to help, I don’t know, a quarter of them, right?

And so, how do we find the right patient to be treated? This is where, at least in GI cancers, we’re increasingly looking at MRD, and it’s really quite controversial as to whether this has value or not. The study that you have presented and continue to update around this in GU cancers, to me, is a demonstration of just the value of it. So, maybe drill down a little bit on that study.

Professor Thomas Powles, MBBS, MRCP, MD: So, John, first, before I do, I’d like to agree with you. I think the next generation of oncologists will look at the way we select patients based off lumps on x-rays to define the presence or absence of disease as a long way from reality. Probably in the same way as we currently look at doing surgery without anesthetic—as a similar. You know, how could you treat 70% of patients who have no chance of relapse, giving them a 20% chance of life-changing toxicity, to try and help the 30% with disease. And of course, if you focus on those 30 with disease—if you have a 10% absolute survival benefit, but you're dragging along 70% at no risk—you can actually double or triple that benefit, from a 10% to a 30% survival benefit. So, it’s not just about sparing those patients who don’t need treatment—it’s about improving your hazard ratios, and that’s absolutely crucial. But then, of course, there’s the sensitivity and specificity of the tests you use. Radiology, obviously, are very, very good tests—because if you’ve got big lumps on an x-ray, you know you’ve got cancer there. But of course, postoperatively, treating unselected patients—we are scrambling around in the dark.

We currently hope that a pathology sample defines high- and low-risk disease—not biology. You know, “How big was it?” Which, of course, is completely, in my opinion, hopeless. So, we’re treating those patients who have bigger cancers because we feel they’re at higher risk.

What about if we can develop an MRD test—circulating tumor DNA—using an informed approach where we do whole exome sequencing on the primary tumor, identify up to 16 mutations, and then track those mutations over time? Can we identify those mutations in the blood?

And the answer is: yes, we can.

And what is the positive predictive value and the negative predictive value? It turns out they’re both pretty high in the adjuvant setting in bladder cancer.

Essentially, what I can tell you is that if you do an operation in bladder cancer, about 50% of the patients will relapse after that operation. And about the same proportion of patients are ctDNA positive after surgery. There’s a very strong correlation: if you’re ctDNA positive six weeks post-surgery, you’ve got a 95% chance of relapsing. If you’re ctDNA negative post-surgery, you’ve still got about a 25% chance of relapsing. But you can change that—because if you're negative and you track those patients and they become positive, they do poorly. But if they remain negative, they do really, really well.

So, the chances of dying of urothelial cancer—and this is data from the TOMBOLA trial, and also from other prospective series—if you track those patients who are negative and remain negative, the chance of dying at two years or three years of bladder cancer is minuscule. You know, it’s 3% or 4%. The chance of relapse is still about 5% or 6%, so it’s still there—it’s not perfect in those negative patients—but it’s very good in the positive patients.

What we showed in a study called IMvigor010 is that those patients who had had surgery—successful surgery for urothelial cancer—and there was no evidence of disease on radiology, that 50% who were ctDNA positive, if they received atezolizumab, you could reduce the risk of relapse and the risk of death by about 40%. Hazard ratios between 0.44 and 0.5. But atezolizumab in unselected patients had hazard ratios between 0.8 and 0.9.

So, it's that dilution of those negative patients. Focusing on the positive patients.

In those negative patients, atezolizumab had no benefit. Although many of those patients—30% of those patients who were negative—relapsed, atezolizumab made no difference, because the risk was low.

Now, if you track those patients—and in the upcoming study at ESMO in the near future, we are doing a repeat of the experiment prospectively—if you are negative, we’re continuing to track your negativity. And if you are negative and become positive, you can then enter into the study. We hadn’t done that before, and that’s novel. That will give us the reassurance that we’ve got those negative patients who remain negative—we do well—but those negative patients who become positive—can we salvage those patients?

Also, immune therapy in bladder cancer: there are two important components. One, we know that by the time you’ve got measurable disease, the cancer’s growing too fast. We lose many of those patients. So, we have to go earlier. That’s why we’re doing it in the adjuvant space. Two, it does appear the biology of the disease is potentially more sensitive to immune therapy the earlier you treat these cancers. Put that together, and we’re hoping for great results.

We’ve looked at other settings. We’ve looked at ctDNA in the neoadjuvant space. But that’s more complicated because you can be ctDNA positive because of your primary tumor or your metastatic disease. So, to define that better, we are now looking at urinary tumor DNA.

We’re beginning to get great results for utDNA. So, if you are utDNA positive and ctDNA negative, you might have bladder-only disease. If you're utDNA negative but remain ctDNA positive, you may have had a pathological complete response in your bladder. We know many of those patients relapse—and that means you haven’t gotten rid of that nodal disease, which is still micrometastatic.

So, it’s the first chapter of a series of chapters that I think is going to transform urothelial cancer.

The last part to it, which I think is important, is monitoring those patients. And ctDNA clearance is an early indicator of good drug activity. In the future, we’re going to be using ctDNA and not radiology to assess that clearance. There’s a study called the MODERN trial—Matt Galsky is leading that study. He’s comparing nivolumab to nivolumab plus a LAG-3 inhibitor, to see if ctDNA clearance is higher with the combination.

And I feel this is a revolution that’s going to stand the test of time.

John Marshall, MD: I want to dig down a little bit on the whole biology question, because in GI cancers, we don’t enjoy that sort of transitive result—where something that works in the refractory setting, metastatic disease, will work in the adjuvant setting. And I actually think MRD-positive is its own biology.

From a drug development perspective, one of the things that I’m leading here in the United States is a platform of MRD-positive colon cancers where we can then test for novel therapies. Because I don’t think the old FOLFOX and FOLFIRI are the right answer in that setting.

And I think you feel like—what I’m hearing you say is—that you’re sort of aligned with that. It’s its own unique biology, different from… it’s the difference between seeds and small plants, if you will. There’s a different biology there.

Professor Thomas Powles, MBBS, MRCP, MD: I feel strongly that that’s the case. When you drill down into some of the biology of the 010 paper—and we put that in Nature, not in a clinical journal—and the reason we went there is: we did whole exome sequencing, obviously, because that’s how we track the ctDNA using the informed approach. We also did RNA sequencing—looking at immune signatures—and we did immunohistochemistry. When you pull that together, there was some indication that there were distinct biological features of the ctDNA-positive patients compared to the negative patients. And as you track— which you can do with time—you can obviously see dynamic changes associated with ctDNA and the different mutations.

So, I do think there are different biological components to this disease. I think those patients who are ctDNA positive may not just represent lower or high risk, or disease that’s inevitably going to relapse and just isn’t yet radiologically visible. It may represent its own distinct biology, which I think is worthwhile exploring.

And in terms of coming to your trial—I don’t know enough about GI cancer—but I do feel that exploring beyond these chemotherapy regimens is really attractive. If you want to accurately identify early active signatures, MRD must be the best way of doing that.

I think it’s a really good way—across different tumor groups—of delivering positive trials for interventions which perhaps, by the time you’ve got advanced disease, it’s too late for some patients.

John Marshall, MD: And we are actually thinking even deeper than that—that maybe you begin to have disease-agnostic interventions at MRD. You might see some cross-reactivity there. So, fingers crossed that the setting you have shown, positively—the interventions around this—will trickle out to other diseases.

But I want to go one other place before we let you go. And that is—we come from different cultures. We work in different cultures.

In the United States, people feel like they have the right to whatever it is. So, if there’s an adjuvant therapy, we treat everybody. And we’re incentivized, right, as physicians to treat people—that’s how we make a living around here. Whereas in the rest of the world, this is very expensive therapy, and so we have to demonstrate positive impact.

So on our side, we’re having to show that it’s okay not to give it—that it won’t work. And usually that requires very large studies to show that you’re not even missing one person, right? The breast doctors do thousand-patient studies to show a 1% improvement in whatever intervention they’re doing.

Whereas in our kinds of cancers, we’re looking for anything that’ll help. And our cultures are different. So, talk a little bit—I know you know both cultures, you interact, and this trial is open across the various cultures—where do you see oncology going? The rising costs of drugs, the limited access to these drugs around the world, the role that MRD might play to improve access globally?

Professor Thomas Powles, MBBS, MRCP, MD: What a complicated question—and I’m going to do my best to answer it.

The first thing to say is: I’m a huge fan of the United States of America. I think it’s an aspirational culture. And when one looks around the world—and I travel the world all the time—you look at the ambition and the aspiration, the dynamism in the United States of America, it’s second to none. I’m a great fan, and I spend a lot of time there.

I’m also a fan of the FDA. Given a choice between the FDA and the EMA, I see advantages and differences between both. But I feel the FDA is a more pragmatic organization, and I have to say—I’m a great fan.

In terms of drug development, I think the last 10 or 15 years have been transformative in the cancers that I’ve treated. That wouldn’t have happened with the universities alone. We weren’t making amazing progress, and the incentive to develop multiple targets at the same time and do parallel trials has been successful.

Indeed, many of those trials that weren’t successful—we’ve learned from and moved forward. We wouldn’t have been able to do those thousand-patient trials with just the university structures that are in place. At the moment, you can’t do a study in Europe and the United States—it’s almost impossible together in a university structure. I’m sure you wouldn’t be able to do that in Asia Pacific, Europe.

And the final bit I want to say before I get to the second part is: in the cancers I treat, median survival in urothelial cancer—10 years ago—was one year. It’s now tripled to three years, with a 30% CR rate. That’s going to get even better next year, in my opinion. Kidney cancer—again, 15 years ago: one year. Now: five years.

And that’s not because of radiology. It’s not because of better diagnostics. It’s not because of better surgery. That’s all because of drug development. So, we have to look at that success and celebrate it and say, “Yes, it may have been expensive, and it may be difficult—but what else are we investing in?” We’re investing in—dare I say it—arms, we’re investing in banks, in all sorts. Isn’t it terrific? This is a good investment.

And going back to your former presidents—and of course I’ve studied Jimmy Carter and all the way through—that race against cancer that each of the presidents has tried to achieve: I think we are winning that race. I think that’s very positive.

On the other hand, there is a twofold concern:
The “no stone unturned” approach where I have to have received every drug, that approach is not helping many patients. Indeed, it’s probably harming quite a few. There are two reasons for that: One, there is this difficulty around families and aspirational societies saying, “If it’s there, I have to have it.” And two, my experience of third- or fourth-line bladder cancer—dare I say it, second-line bladder cancer—is that it’s a law of diminishing returns. In the end, it goes wrong for almost everyone. It’s probably doing as much harm as benefit for many of those patients.

You and I have both seen many patients suffering from toxicity toward the end of life, trying to get more and more cycles in. I do think Europe is more pragmatic about that. I do think if your median overall survival is seven months, your response rate is 10%, and it wasn’t significantly better from an OS perspective than placebo, I think Europeans—and the EMA—are much more pragmatic, saying, “We want to see more.” And I think that’s very reasonable.

One caveat: occasionally, those drugs that weren’t great in a second- or third-line setting have turned out to be quite active in combination in the frontline setting. The FDA is good at that. I have examples where we’ve had underwhelming results second or third line, and we moved it to the frontline setting in combination—and we’ve achieved those survival advantages.

So, I think that pragmatic approach is important. But I do not feel that every patient needs every therapy. I don’t think every patient benefits from every therapy.

One of the key questions—when one looks from a European, U.S., and global perspective—is how we assess PFS and OS endpoints. It’s clear to me that in the perioperative setting, we need to achieve OS. That’s my feeling on this. I don’t think disease-free survival as an endpoint is enough.

Two reasons—one, you can give salvage therapy subsequently to those patients that need it. And number two, you are overtreating all those patients who are not going to relapse. So, for me—we need that. I know many of my American colleagues feel DFS is important because that has a higher rate of not achieving advanced disease and avoiding advanced disease as a goal.

I have mixed feelings about that. I think patients want to live longer and feel better. In the perioperative setting, if you don’t have OS, you’re not achieving either of those goals.That’s relevant.

The last bit I’d like to say around that issue—which I think is important—is that MRD has a really important way of achieving OS. You enrich those patients that need therapy. You put some patients in harm’s way, but you do achieve OS. And MRD studies—of course—OS should be the endpoint.

The second part is: you’re sparing big groups of patients therapy who almost certainly don’t need it. Treatment would be overtreatment. Achieving overall survival advantages in that negative subgroup will be almost impossible. And indeed, the risk is low.

So, MRD achieves those goals that I think we are looking for. And I suspect, John, in the future, we won’t be talking about adjuvant therapy in the way we are now. I think we’re going to be talking about patients with MRD or without MRD. And I think, as you’ve said in your study, those patients with MRD—novel trial design to improve survival—in the end, is going to be a really important chapter in where we go next.

John Marshall, MD: I would bet—if I had enough money to throw on the table—that in fact MRD testing will become screening. And maybe in 10, 20 years, you don’t even get your bladder taken out or your colon taken out, but you just get systemic treatment and make the cancer go away before it’s a problem. But you and I can work toward that end and put ourselves out of a job.

Professor Dr. Thomas Powles, thank you so much for spending your valuable time with us. You’re clearly transforming cancer care through incorporation of MRD and innovative clinical trial design and teaching us all about how we’re going to go forward—including about how we’re going to do that as a unified global medical community.

Professor Thomas Powles, MBBS, MRCP, MD: Good pleasure. I really enjoyed it. Thank you very much indeed.