Ever since early February, when former FDA Commissioner Robert Califf publicly bemoaned the progress of precision drugs, I needed to search out out whether or not different folks felt equally discouraged.
I obtained my likelihood in October when an MD Anderson Most cancers Heart public relations skilled pitched an interview with a senior MD Anderson radiation oncologist, who simply occurred to be my dearly departed brother’s shut pal. In truth, Dr. Prajnan Das, professor and division chair of Gastrointestinal (GI) Radiation Oncology, didn’t hesitate to open his house to us in June 2015, insisting that my brother be evaluated by a kidney most cancers specialist on the storied establishment. My brother had his cancerous proper kidney eliminated in Houston thereafter, and it was a major consolation to each him and me to be in Dr. Das’ house as we wrestled with the Stage 4 kidney most cancers analysis and unsure future forward. My brother did have genomic testing performed, however his most cancers had unfold too far. He lived one other two years earlier than succumbing to problems from the illness, one thing that Dr. Das had fortunately ready us for proper from the beginning.
However that was eight years in the past. And sitting in Feb 2025, Califf, who additionally misplaced his brother to most cancers — in January — was saying that the sensible functions of precision drugs had been largely a failure. In October, Dr. Das spoke at HLTH on a panel on precision drugs I missed, and so in my Zoom interview with him after HLTH, that’s the place I started.
What follows is an edited Q&A about precision drugs and the potential of AI.
MedCity Information: What Dr. Califf was saying is that his personal brother had pancreatic most cancers and handed away and he was very upset that he needed to go store round to attempt to discover a protocol. So he thinks precision drugs has not lived as much as its promise. Are you as harsh?
Dr. Prajnan Das: I feel a whole lot of it is determined by what you outline by precision drugs. On the panel, I talked in regards to the precision drugs facets of radiation remedy as a result of that’s my focus. And that’s additionally fairly essential as a result of after we take into consideration precision drugs, a whole lot of our focus tends to be on medicine. The FDA is all about that. However if you take a step again, 50-70% of most cancers sufferers are going to get handled with radiation through the course of their therapy, and so improvements in radiation even have an enormous capacity to have an effect. So after I take into consideration how I care for sufferers immediately versus 10 years in the past or 20 years in the past, it’s fully completely different in each facet. Ten, 15 years in the past, we had been utilizing radiation primarily in sufferers whose most cancers had not unfold. If the most cancers had unfold, then we’d use radiation to assist palliate signs in important elements of the physique.
However now what now we have discovered within the final 10 years is that even in cancers which have unfold to restricted areas of the physique, now we have recognized a definite entity, which we name oligometastatic illness. And in these sufferers giving radiation or different types of ablative therapy to these oligometastatic areas improves sufferers’ disease-free survival, their long-term survival, their capacity to remain off chemotherapy for prolonged durations. So considered one of my colleagues in renal cell most cancers has this examine that should you deal with areas of metastasis with radiation, these sufferers can keep away from the standard systemic remedies for renal cell most cancers for prolonged durations. And people remedies have unintended effects as you and I each know. So Chad Tang from MD Anderson printed a extremely essential examine on that.
Even in pancreatic most cancers, we used to suppose that, ‘Hey if the most cancers has metastasized, the one choice is chemo.’ My colleague Ethan Ludmir printed this essential randomized trial, which confirmed that even in pancreatic most cancers, should you can deal with the metastasis with radiation or surgical procedure, these sufferers, their illness disease-free survival will get prolonged. We’d by no means have identified that just a few years in the past.
MedCity Information: However is that information coming from doing genomic testing? How is precision drugs associated to this?
Dr. Prajnan Das: A variety of it’s really coming from genomic testing as nicely. So along with the medical trials, the place we are attempting to determine, why sufferers are doing a lot extra higher with native remedies in oligo metastatic illness than we thought they might, there’s a whole lot of translational research which can be getting into which can be issues like immune signatures and particular molecular sorts which can be additionally reworking our understanding of why these remedies are working. However going again to the molecular subtypes, so what we’re additionally studying is that now we are able to use molecular subtypes to determine which sufferers want radiation and which sufferers don’t. That’s precision drugs too.
I’ll provide you with an instance from one of many cancers I deal with: rectal most cancers. The usual therapy for rectal most cancers was a mixture of chemotherapy, radiation and surgical procedure. Now we all know that you are able to do a easy molecular, or perhaps a easy immunohistochemistry take a look at on the biopsy for rectal most cancers to establish a subgroup that’s known as microsatellite unstable rectal most cancers. And in that subgroup, all you could do is deal with them with immunotherapy. So there have been research which have proven a 100% response fee to immunotherapy in that particular affected person inhabitants. You don’t want radiation, you don’t want chemo, you don’t want surgical procedure.
So I feel we’re getting a lot, a lot better now at determining this particular molecular subtypes and treating them and determining which remedies work for them, but additionally importantly, which remedies they don’t want.
MedCity Information: However do you suppose that the huge information that now we have immediately on the human physique that we didn’t have even 10, 15 years in the past, do you suppose that has translated to nice remedies? Or do you suppose there’s extra to comprehend in that? I do know you might be optimistic about precision drugs. I get that. I’m attempting to be skeptical, so fairly the alternative.
Dr. Prajnan Das (smiles): That’s why now we have completely different jobs.
MedCity Information: Completely. So what extra must be performed? Do you suppose that not simply in most cancers, but additionally throughout ailments, molecular testing ought to be the very first thing that individuals do? It’s nonetheless not one thing that I feel is broadly performed.
Dr. Prajnan Das: Yeah, so I feel it’s changing into rather more broadly performed, as routine. I feel the essential factor to additionally respect is that finally it takes a variety of years, generally a few many years, even between when a take a look at turns into routinely used and after we actually begin understanding the best way to use that data.
With that microsatellite unstable rectal most cancers instance I gave you, we knew that there was this molecular subtype entity — now we have identified that for about 20 years, however solely about 10 years in the past did folks begin exploring, ‘hey, perhaps immunotherapy will work on this affected person.’ And solely within the final two years have we had the potential medical trials which have really rigorously studied that and confirmed the position of immunotherapy. So I really feel prefer it takes 10 to twenty years nearly, between that preliminary testing to arising with particular focused remedies. However I hope that going ahead, that point lag goes to lower, and we’re going to get higher and extra environment friendly at bridging that hole between the take a look at and the appliance of it.
MedCity Information: On that hopeful observe, let’s segue to AI. Within the area that you simply’re in, and inside the context of improved effectivity, the place do you suppose you, in your apply, you at MD Anderson are utilizing AI?
Dr. Prajnan Das: So our frequent pal has been threatening that I’m going to lose my job yearly due to AI. He’s been telling me that for the final 5 to 10 years at the very least.
MedCity Information: And also you’re nonetheless there.
Dr. Prajnan Das: AI clearly is transformational and in many various methods, however a part of additionally it is determining the place we use it and the place we don’t use it, proper? So in some areas like diagnostic mammograms, and even in colonoscopies to assist detection of polyps, AI has actually augmented the clinician’s capacity. In my area of radiation oncology, AI has the potential to be transformative too.
So the best way radiation is designed immediately is like this: an individual will get a scan for that simulation to the scan, after which a doctor has to manually work on designing the radiation plan. They usually do it by sitting on a pc, outlining what the goal is and what the traditional buildings are, which takes a number of hours. After which a second skilled, known as the dosimetrist, makes use of that data to design the radiation beams, the beam angles, what’s blocked, what’s open, after which a 3rd skilled, a physicist, checks all of that and makes positive that the radiation machine is offering the output that now we have designed this manner.
Each step of that may be changed by AI.
So I collaborated with Lawrence Court docket, a colleague who’s a physicist, and we labored on this venture with a PhD scholar, which now we have now printed, which reveals which you could develop an end-to-end AI program for rectal most cancers radiation planning. And that entire course of might be performed in just some minutes, changing all of this effort and time. So I really feel like this may be really transformative in our area, and what this may accomplish is that it’ll standardize care, it’ll make care extra accessible, cheaper and higher.
MedCity Information: And cut back variability?
Dr. Prajnan Das: Completely cut back variability. And we’ve additionally performed research on that exhibiting that if you begin utilizing these instruments, even the variability between physicians can go down dramatically. That’s the thrilling half.
MedCity Information: I reside in Silicon Valley, and one of many greatest complaints of all of the tech folks right here is that it’s very arduous to persuade docs to vary the best way they do what they do. A few of that’s official since you’re coping with folks’s lives right here, so you might be conservative as you have to be. However do you suppose that when you have got this, as you name it, a transformational expertise like AI, do you suppose docs are going to embrace it? What do you see amongst your personal colleagues?
Dr. Prajnan Das: So long as you give the docs the precise instruments, they’ll completely embrace it. So my group makes use of these instruments on a routine foundation, Not simply that, Lawrence has taken this instrument and has developed a platform the place it’s getting used globally in hospitals in Zambia and South Africa. Primarily, it’s a knowledge-based AI instrument based mostly on MD Anderson physicians’ information, and now a doctor in a low-resource setting can add their affected person’s data into this program and this system will [create] a radiation plan at MD Anderson requirements based mostly on that data.
MedCity Information: What sort of AI is it? Did MD Anderson develop it, or was it an off-the-shelf instrument?
Dr. Prajnan Das: It’s a deep studying instrument. It’s our inside product developed by our good physicist and his group of PhD college students and postdocs and different physicists. The explanation additionally I feel that his venture labored is that he works very collaboratively with physicians, in order that we give you instruments that physicians really will use.
MedCity Information: Versus the EHR which was imposed on physicians.
Dr. Prajnan Das: Proper, precisely. So I feel that’s a part of the issue. When physicians are fascinated about digital instruments, the very first thing that involves thoughts is the EHR, and that has by no means met any of our wants or targets, proper? When you can give you the AI instruments that may really assist physicians, they’ll embrace it. I imply, radiologists are utilizing it, endoscopists are utilizing it. We’re utilizing it.
MedCity Information: Any closing ideas? On the way forward for precision drugs?
Dr. Prajnan Das: I feel the place we actually might be modern is arising with good hypotheses which can be based mostly on sturdy current knowledge that leverage molecular instruments. And perhaps AI will assist us work out what these questions are. As a result of proper now, I’m positive there’s a whole lot of sign in our knowledge that we’re lacking.
AI may help select these knowledge that, ‘Hey, perhaps that this will work for this affected person.’ There’s a tremendous instance of this. Just lately, a paper got here out in Nature by considered one of our residents in radiation oncology that obtained worldwide consideration. He leveraged the info of sufferers who had been getting handled throughout Covid on immunotherapy and what he discovered utilizing some modern knowledge evaluation was that sufferers who had a Covid vaccine had a lot greater responses to immunotherapy in comparison with sufferers who didn’t get a Covid vaccine.
So on the floor is not sensible. The Covid vaccine is focusing on Covid, not the tumor, however he was in a position to make use of knowledge instruments to establish that sign. Then with that sign, he really did rigorous evaluation on medical knowledge units to point out that there’s clearly this distinction we’re seeing. After which he went to the lab to search out a proof for why this was occurring. And he did rigorous research in mouse fashions, different areas.
I feel that’s the way forward for personalised drugs, the place knowledge instruments assist us discover distinctive indicators, and these then generate the following era of research that doubtlessly result in extra remedies.
Photograph: MD Anderson Most cancers Heart
