This transcript has been edited for clarity.
Hello. I’m Dr Mark Lewis, director of gastrointestinal (GI) oncology at Intermountain Health in Utah. I’m speaking from the 2025 ASCO Annual Meeting in Chicago, where we’ve seen some interesting new data in GI cancers. I always enjoy doing this kind of on-the-ground reporting, and the real reason I love coming to these meetings is, while it’s wonderful to network with colleagues, there is true progress in our field that we can take back almost immediately to our clinics to help our patients.
There are three themes in GI oncology that I’ve seen emerge at this meeting. One is the utility or not of circulating tumor DNA (ctDNA) in affecting treatment decisions. The second is the role of immunotherapy in GI oncology, and the third is, I think, a real triumph for targeted therapy in oncology.
Addressing the first, and to be honest, most controversial point: Where are we with ctDNA in GI oncology, and most importantly, where are we with these assays in terms of how we counsel our patients?
Sometimes what’s most important about ASCO is trials that are arguably negative in their findings. This year, it really caught my attention that DYNAMIC-III sort of turned over the apple carton terms of ctDNA-informed approaches to colon cancer.
The design of this study was looking at patients with stage III colon cancer and using a ctDNA-informed approach in a randomized fashion to see if we should be escalating chemotherapy in patients who have a positive ctDNA signal. The randomization was against the standard of care.
For years, I think there has been a false binary between using modern ctDNA technology and our traditional clinicopathologic criteria. After all, the whole way we classify stage III colon cancer is based on TNM staging, so that remains the foundation. What we are trying to discern together, and especially together with our patients, is when it is appropriate for this technology to be layered on top of traditional clinicopathologic criteria and thus affect treatment decision-making.
The takeaway from this trial for me, especially since recurrence-free survival was worse for the ctDNA-informed cohort vs the standard of care, was that this is a prognostic assay, but not necessarily predictive. Patients who have a ctDNA signal that is positive who had escalation of their adjuvant therapy did not seem to benefit from the addition of, say, irinotecan to a traditional fluoropyrimidine and platinum doublet.
Interestingly, also, I think this study validated that roughly one third — maybe no more than 30% — of stage III colon cancer patients have a positive ctDNA signal. My takeaway, again, is we’re sort of going back to the future. It was the MOSAIC trial that was published in June 2004 that established the current standard of care for how we approach adjuvant therapy in stage III colon cancer.
Now, slightly over two decades later, we really have not made vast improvements in the field, and ctDNA is wonderful, but it is not entirely supplanting the understanding we’ve had since MOSAIC and since IDEA.
Without getting too into the weeds, I’ll also point out that I think the statistical design here was ambitious. The hazard ratio in this particular trial, DYNAMIC-III, was frankly suggestive of the fact the study might have been underpowered, enrolling just over 200 patients, whereas MOSAIC had over 2000 to reach its practice-changing conclusions.
Watch out for upcoming studies such as CIRCULATE-US and NRG-GI008, which will again use ctDNA negativity to look at de-escalation and ctDNA positivity to look at escalation. Until that trial matures, I don’t think this assay is actually going to change the standard-of-care approach to stage III colon cancer in the United States.
The second point I’d like to make is about immunotherapy. I love the fact that when patients come to me, and I’ve been described before our first visit as a chemotherapy doctor, I can tell them that there’s more to medical oncology than indiscriminate cytotoxicity. We are truly in the era where immunotherapy has a role to play in a variety of GI cancers.
We heard at the ASCO plenary session that immunotherapy has a major role to play now in adjuvant therapy for stage III colon cancer with mismatch repair deficiency. The ATOMIC trial showed a significant 3-year disease-free survival benefit using atezolizumab along with traditional FOLFOX chemotherapy to help patients in the adjuvant setting.
The MATTERHORN study showed the advantage of using durvalumab atop FLOT in the perioperative setting in gastric cancer. So two different GI histologies, but a huge role now for immunotherapy in this space.
Finally, dealing with metastatic colorectal cancer, the maturation of CheckMate-8HW shows that the ipilimumab-nivolumab (ipi-nivo) doublet definitely has a role to play in the metastatic setting.
This has been interesting because when I think about immunotherapy trials that have changed my practice, the one I keep coming back to is KEYNOTE-177. It was such a triumph at the time of its publication and remains so.
What’s sobering to realize, though, is that as more time has elapsed since KEYNOTE-177 matured, the 5-year survival rate of the pembrolizumab arm remains about 60%. Also, you might remember that the initial survival curve dipped below the chemotherapy arm before it plateaued and improved for immunotherapy. There are certainly some patients who need an earlier, more aggressive response.
Enter ipi-nivo. What I like about this trial is that the ipilimumab dosing seems quite conservative, at 1 mg/kg, with four exposures to that agent before nivolumab continues by itself. That’s appealing to those of us who have always had some reservations about using an anti-CTLA-4 approach.
The very first time I ever used immunotherapy in any setting was during fellowship. It was 2011, and it was ipilimumab in the setting of metastatic melanoma. I watched in amazement as this patient’s disease melted away, but at a dose then of 10 mg/kg, the endocrinopathy was significant. I also watched as my patient suffered from pan-hypopituitarism.
For medical oncologists who are understandably tentative about anti-CTLA-4 as a mechanism, the question is always, is the juice worth the squeeze? Here, you do get a higher response rate from ipi-nivo than you would with nivolumab alone for patients who, say, might be on the verge of visceral crisis and need a faster initial response.
Finally, I want to talk about targeted therapy. I think what was incredible about ASCO this year is realizing just how much progress we’re making with BRAF-mutant colon cancer. We have known for a very long time that this mutation confers a worse prognosis, and we’ve often wondered whether it’s appropriate to treat these patients sequentially or should we take the BREAKWATER-informed approach of giving them encorafenib, cetuximab, a fluoropyrimidine, and a platinum upfront — arguably a quadruplet.
I think the answer from this meeting is a resounding yes— a doubling of median overall survival from 15 to 30 months by essentially frontloading all of the effective treatment and not trying to do it in sequential lines of therapy.
You never get a second chance to make a first impression. Really, what this means is we have to know as soon as possible that we’re dealing with a BRAF mutation. There are certain clinical phenotypes that we look for — more aggressive disease, carcinoembryonic antigen rising in the right colon — but this is proof, once again, that the oncologist without the pathologist is blind.
I cannot take proper care of my patients without a fully biomarker informed approach, and I can’t wait for these test results to come back. This study allowed for at least early exposure to FOLFOX alone while BRAF mutation results were maturing, but we really need to partner with a pathologist and understand metastatic disease in GI the same way we would understand it in metastatic breast cancer.
There is not a single breast cancer oncologist I know who would try treating their patients without knowing estrogen receptor, progesterone receptor, and HER2 status. I think we are absolutely at the point in GI oncology where it should be unacceptable to treat our patients without knowing KRAS, NRAS, BRAF, and arguably HER2 status, and certainly mismatch repair or microsatellite instability status.
The final targeted therapy triumph at this ASCO looked at DESTINY-Gastric04. DESTINY has been an interesting suite of trials looking at the role of trastuzumab deruxtecan in a variety of HER2-positive cancers. I vividly remember the plenary session several years ago where the data for DESTINY-Breast04 earned a standing ovation.
I was one of those people who stood up as a GI oncologist because I could see how this was going to help patients with HER2-positive disease across various primary sites. What we learned at this meeting with the maturation of DESTINY-Gastric04 is this drug particularly seems to outperform traditional second-line therapies such as ramucirumab-paclitaxel.
There are downsides. This drug famously (or infamously) causes interstitial lung disease in about 1 in 7 patients. It’s also absolutely vital to re-biopsy at time of progression to ensure that the HER2 target for this antibody-drug conjugate is still there.
HER2 heterogeneity remains something we haven’t fully grappled with, but I find that my patients, when I explain the role of a targeted therapy, are generally willing to undergo another liver biopsy —if they understand the lock and key hypothesis between the HER2 mutation and a drug such as trastuzumab deruxtecan.
To sum up, from ASCO 2025 for GI oncology, the three main areas I see of progress, at least in our understanding, are number one, circulating tumor DNA remaining prognostic, but likely not predictive at this point; number two, immunotherapy having a major role to play now in the adjuvant colon cancer setting as well as in perioperative gastric cancer management; and number three, targeted therapy with BREAKWATER really becoming, I think, the standard of care in the first line for BRAF V600E-mutant colon cancer and trastuzumab deruxtecan making a strong play for second-line therapy in HER2-positive gastric cancer.
This has been Mark Lewis, the director of medical oncology for gastrointestinal oncology at Intermountain Healthcare, reporting for Medscape from ASCO 2025. Thank you.
