ASCO Daily News Personalizing Lung Cancer Management With ctDNA: Where We Are and Where We Are Headed
Dr. Vamsi Velcheti and Dr. Charu Aggarwal discuss the evolution of ctDNA as a critical tool in precision oncology and its implications for lung cancer management, including its potential role in the early-stage setting.
TRANSCRIPT
Dr. Vamsi Velcheti: Hello. I am Dr. Vamsi Velcheti, your guest host for the ASCO Daily News Podcast today. I am a professor of medicine and director of thoracic medical oncology at the Perlmutter Cancer Center at NYU Langone Health.
The management of small cell lung cancer has rapidly evolved over the past few decades, and today, molecular testing and biomarker testing for lung cancer are absolutely critical in terms of designing treatment options for our patients with metastatic non-small cell lung cancer. Today, I'm delighted to be joined by Dr. Charu Aggarwal for a discussion on ctDNA (circulating tumor DNA) and the role of ctDNA in lung cancer management. Dr. Aggarwal is the Leslye Heisler Professor of Lung Cancer Excellence and section chief of thoracic and head and neck oncology at University of Pennsylvania Abramson Cancer Center.
You'll find our full disclosures in the transcript of that episode.
Dr. Agrawal, it's great to have you on the podcast today. Thank you for being here.
Dr. Charu Aggarwal: Thank you for having me.
Dr. Vamsi Velcheti: Let's start off with setting the stage for ctDNA technology. These technologies have rapidly evolved from experimental conceptual stage to essential clinical tools for day-to-day clinical practice. Could you briefly discuss how recent advancements in ctDNA technologies are shaping our approach to precision medicine, especially in lung cancer?
Dr. Charu Aggarwal: Absolutely. And you know, I think we need to just level set a little bit. What exactly is circulating tumor DNA? This is a way to assess exactly that. Every tumor sheds little pieces of tumor-derived DNA into the bloodstream, and this occurs in a variety of solid tumors. But now we have the technology to be able to derive this DNA that's actually being shed from the tumor into the bloodstream, these minute fragments of DNA, take them out, amplify them and sequence them with a variety of different mechanisms. They can be DNA sequencing alone, they can be DNA and RNA sequencing, they can be whole transcriptome sequencing. The technology, as you rightly pointed out, Dr. Velcheti, has significantly improved from just being able to look at circulating tumor DNA to now being able to amplify it, sequence it, and use it to offer personalized therapy. I think lung cancer is definitely the poster child for such an approach as we have a lot of data that has shown clinical utility and validity of being able to use circulating tumor DNA next-generation gene sequencing to guide therapy.
Dr. Vamsi Velcheti: There have been so many technological leaps. It's really impressive how far we've come to advance these sequencing platforms. Recent advances with AI and machine learning are also playing important roles in interpreting ctDNA data. How are these computational advances really enhancing clinical decision-making in day-to-day clinical practice?
Dr. Charu Aggarwal: I think while we have firmly established the role of ctDNA in the management of patients with metastatic lung cancer, some of the approaches that you talked about are still experimental. So let me backtrack a little bit and set the stage for how we use ctDNA in clinical practice right now. I think most patients, when they come in with a new diagnosis of stage IV lung cancer, we want to test for biomarkers. And this should actually be the established standard. Now included in the NCCN guidelines and actually also international guidelines, is to consider using blood-based testing or plasma-based testing to look for biomarkers, not just tissue-based testing which had been our historical standard, but to use these plasma guided approaches to identify the seven to nine biomarkers that may be truly implicated in either first- or second-line therapy that are called as your immediately actionable mutations.
What you're talking about is AI computational methods. I think there's a lot of excitement about how we can use genomic signatures that are derived from either tissue or ctDNA-based biomarker testing, combine it with radiomic features, combine it with histologic features, look at H & E patterns, use AI algorithmic learning to be able to actually predict recurrence scores, or can we actually come up with predictive signatures that may be extremely helpful?
So, I think some of the techniques and technologies that you're talking about are incoming. They are provocative. I think they're very exciting, but very early.
Dr. Vamsi Velcheti: I think it's really amazing how many advances we have with these platforms. You know, the challenge really is the significant gap in terms of uptake of molecular testing. Even today, in 2025, there are significant gaps in terms of all metastatic lung cancer patients being tested for all biomarkers.
So, why do you think there's such a challenge in testing patients with lung cancer? In most academic practices, we try to achieve 100% testing for all our patients, but we know from recent studies that that's not the case across the country. What do you think the gaps are?
Dr. Charu Aggarwal: Biomarker testing is so essential, like you pointed out, for us to be able to guide the right therapy for our patients. And we see this in our practice every day as you and I see patients with lung cancer, that a large proportion of our patients either don't get tested or they start therapy before their test results come back. So, I think this is a real problem.
However, to add some optimism to this problem, I do think that we are making a move in the right direction. So, four or five years ago, there was a lot of data being presented at national meetings, including ones from the American Society of Clinical Oncology, where we saw that, nationally, the rates of biomarker testing were probably in the rate of 40 to 50%. However, now with the availability of both tissue and plasma, I do think that the rates of biomarker testing are increasing. And if you were to survey a sample or even perform retrospective data research, I believe that the number is closer to 70% of all patients with metastatic non-small cell lung cancer.
And you know, you asked why is it not 100%? I think there are many reasons. I think the number one reason is tissue availability. Many times, the biopsies are small, or the tumor is very necrotic. So, either the tissue quantity itself is small, or the tissue quantity is insufficient to perform gene sequencing. And that's exactly where plasma comes in. When you don't have tissue availability, we have shown, as have others, that you can use plasma effectively to increase the proportion of patients who are not only tested but also receive the right therapy. I think there are also other barriers, including inertia. You know, I think this is both patient and physician inertia, where patients want to get started quickly, they don't want to wait. Physicians are very busy and sometimes want to be able to deliver treatment as soon as possible. We have seen there are some institutional barriers. Not every institution has in-house gene sequencing testing. So how do you really operationalize, send out these tests in a fast, efficient manner so that you get results back? Is it a pathologist who sends out the test? Is it the medical oncologist? Is it the pulmonologist or the interventionalist? I think there is this need to develop reflex testing mechanisms which some institutions do really well and some don't. And then finally, there are financial implications as well. How do we do this in a most cost-efficient fashion?
So there are many barriers, but I'm happy to say that we are making a move in the right direction as we are understanding that it's important to do it, it's easy to do it maybe with a value add of plasma, and finally, as you said, you know, as these technologies become more available, they're actually getting more cost-effective.
Dr. Vamsi Velcheti: Dr. Aggarwal, you've been at the cutting edge of these advanced platforms and testing. So, what do you do in UPenn? How do you handle all these barriers and what is your workflow for patients in University of Pennsylvania?
Dr. Charu Aggarwal: One of the things that I mentioned to you was there may be institutional barriers when it comes to gene sequencing. So, we actually, several years ago now, instituted a very robust reflex testing paradigm where almost all of our patients, regardless of stage, with a non-squamous non-small cell lung cancer diagnosis, would automatically be reflexively sent to our molecular pathology lab where they would get gene sequencing both for the DNA as well as with an RNA fusion-based platform. And the reason we did this was because we wanted to expedite and reduce the turnaround time. We also wanted to ensure that we were not just doing DNA testing, which I think is really important for our listeners here. There are many fusions as well as certain skipping mutations like MET exon 14 that may be missed on DNA testing alone. So, it's really incredibly important to run both DNA and RNA samples.
So, we do this routinely, and based on our research and others, what we also do routinely is that we send concurrent tissue and liquid biopsies or plasma MGS testing upon initial diagnosis. For example, if a patient comes in with a diagnosis of stage IV non-small cell lung cancer, their tissue might already be at my molecular pathology lab based on the reflex mechanism that I just described to you. But upon their initial meeting with me, we will send off plasma. And I will tell you this, that Penn is not just one institution, right? We have a large network of sites. And as part of my research, one of the things that we wanted to do was implement wide scale means to improve biomarker testing. And we have done this with the use of technology like you mentioned, Dr. Velcheti: How can we actually use AI? How can we leverage our electronic medical record to identify these patients? So, we have a nudge-based mechanism which actually facilitates the pending of orders for biomarker testing for patients with new diagnosis of metastatic non-small cell lung cancer. And we are looking at our rates of biomarker testing but also rates of completion of biomarker testing before first-line therapy started. So many of our participating sites are clusters for our randomized control trial to increase molecular testing. And I'm really excited about the fact that we're able to implement it not just at our main satellite, downtown Penn Hospital, but also across our community.
Dr. Vamsi Velcheti: I think that's great. Thank you so much for those insights, Dr. Aggarwal. I think it's so important because having the best technology is just not enough. I think implementation science is actually a real thing. And I think we need to all learn from each other, advance these things.
So, I want to ask you about the new emerging paradigm in terms of using ctDNA. Of course, in the metastatic setting, we've been using ctDNA for molecular profiling for a while now. But the recent data around monitoring early-stage disease, especially post-operative monitoring, is an exciting area. There are a lot of opportunities there. Could you please talk us through the emerging data in lung cancer and how do we incorporate ctDNA-based monitoring MRD or should we even do that right now? Is the data ripe enough for us to kind of deploy this in a clinical setting?
Dr. Charu Aggarwal: I think using ctDNA in the early-stage setting is our next frontier in lung cancer. I think naturally we have been able to successfully deploy this in the stage 4 setting. It made a meaningful difference in the lives of our patients, and we are a little bit behind the A ball in terms of how MRD is used in lung cancer. Because, you know, colorectal cancer has already done large-randomized trials based on ctDNA and MRD. It's routinely used in hematological malignancy. So, it makes sense that we should start to use it.
However, when I say this, I say this with excitement, but also a little bit of gentle caution saying that we actually don't quite have the prospective randomized data just yet on how to deploy. Yes, intuitively we would say that if you detect ctDNA and MRD, that patient is at higher risk. So, we identify that, but we actually don't know what to do with the second part of that information once you identify a patient with high risk. Are there other techniques that we can then come in with or other drugs that we can come in with to modify that risk? And that's the thing that I think we don't have right now. The other thing that we don't have right now is the timing of the assay, when to use it. Is it to be tested in the pre-op setting? Is the post-op test the best timing, or is it monitoring and dynamics of ctDNA that are most important? And the third thing I will say in terms of precautionary cause is that we don't know which test just yet. There are actually a few commercially available tests out in the market right now. We know about them and I'm sure our community colleagues know about them. Some of them even have Medicare approval. However, many of these tests are currently tissue informed. We don't have tissue uninformed tests. And what does that mean? Tissue uninformed means that you actually take a piece of tumor tissue, you sequence that tumor and based on the gene profile of that tumor, you actually design a panel that can then be used to track the mutations in the blood-based pack. This requires, as the name implies, a tumor. So can this be used in the pre-op setting is a large question. Because coming back to the idea of tissue availability, you and I both know that when we get FNAS and we use it for PDL-1 testing and we use it for gene sequencing, there often isn't enough tissue left for us to then either do whole genome sequencing or even whole transcriptome sequencing, which may be required to build some of these assays.
I think the future lies in this idea of tumor uninformed assays because if we could go to a blood only or a plasma only approach using novel signatures like proteomics or methylation, I think that's where the future is. But we're still a little bit early in the discovery stages of those, as well as to come are the validation stages so that we can be confident that these blood-only assays may actually give us an answer.
So, with those three cautionary notes, I would say that optimism is still very high. I think ctDNA MRD is the right place to think about. We need to do this for our patients to better identify high-risk patients and to think about means to escalate treatment for them.
Dr. Vamsi Velcheti: Yeah, I completely agree, and I think with all the changes and evolution of treatments in the management of early-stage lung cancer now with neoadjuvant and adjuvant, there's really a need for an escalation and de-escalation of therapies post-operatively. And I think it's a huge opportunity. I think we all could learn from our colorectal colleagues. I think they've done a really good job at actually doing prospective trials in this setting. I think we're kind of a little behind here.
Dr. Charu Aggarwal: I think in the metastatic setting there are ongoing trials to look at this exact question. How do you choose an appropriate first-line therapy, a monitor ctDNA at the six-week trial? It's being evaluated in a trial called the "Shedders" trial, where if patients are still ctDNA positive at six weeks, then you can escalate treatment because they haven't "cleared" their ctDNA. There has been a lot of research that has shown that lack of ctDNA clearance in the metastatic setting may be a poor prognostic factor.
We and others have shown that if you do clear your ctDNA or if you have a reduction in ctDNA load overall, that that is directly related to both an improved progression-free survival and overall survival. This has been shown with both tissue informed and uninformed assays. So I think it's very clear that yes, you can track it. I think the question is: Can you apply that data to the early-stage setting? And that's an open research question. A lot of groups are looking at that and I think it's completely reasonable, especially to determine duration of therapy, to determine optimal timing, optimal timing of scans even. And I think these are just such interesting questions that will be answered in the future.
Dr. Vamsi Velcheti: And also like a kind of early detection of resistance patterns that might inform early initiation of combination strategies. And I think it's a lot of opportunities I think yet to be explored. A lot of exciting things to come and I'm sure we'll kind of see more and more data in the next few years.
Dr. Aggarwal, thank you so much for sharing your fantastic insights today on the ASCO Daily News Podcast. It's been a pleasure to have you on the podcast today. Hope to see you at ASCO.
Dr. Charu Aggarwal: Thank you so much. This was great and I remain so excited by all of the possibilities to improve outcomes for our patients.
Dr. Vamsi Velcheti: Thank you to all the listeners for your time today. If you value the insights that you hear from the ASCO Daily News Podcast, please take a moment to rate, review and subscribe wherever you get your podcast. Thank you so much.
Disclaimer:
The purpose of this podcast is to educate and to inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions. Guests on this podcast express their own opinions, experience and conclusions. Guest statements on the podcast do not express the opinions of ASCO. The mention of any product, service, organization, activity or therapy should not be construed as an ASCO endorsement.
Follow today's speakers:
Follow ASCO on social media:
Disclosures:
Dr. Vamsidhar Velcheti:
Honoraria: Glavanize Therapeutics
Consulting or Advisory Role: Bristol-Myers Squibb, Merck, AstraZeneca/MedImmune, GSK, Amgen, Taiho Oncology, Novocure, Takeda, Janssen Oncology, Picture Health, Regeneron
Research Funding (Inst.): Genentech, Trovagene, Eisai, OncoPlex Diagnostics, Alkermes, NantOmics, Genoptix, Altor BioScience, Merck, Bristol-Myers Squibb, Atreca, Heat Biologics, Leap Therapeutics, RSIP Vision, GlaxoSmithKline
Dr. Charu Aggarwal:
Consulting or Advisory Role: AstraZeneca, Daiichi Sankyo/AstraZeneca, Regeneron/Sanofi, Pfizer, Boehringer Ingelheim, Takeda, Arcus Biosciences, Gilead Sciences, Novocure, Abbvie
Speakers' Bureau: AstraZeneca (an immediate family member)
Research Funding (Inst): Merck Sharp & Dohme, AstraZeneca/MedImmune, Daiichi Sankyo/AstraZeneca, Lilly@Loxo, Candel Therapeutics