Patient Monitoring after Adjuvant Chemotherapy in Colorectal Cancer

EP: 1.COMING SOON: Around the Practice: ctDNA Analysis to Direct Treatment Decisions in Colorectal Cancer

EP: 2.Patient Profile: A 55-Year-Old With Colon Cancer

EP: 3.Effect of MRD Positivity on Treatment Decisions in Colorectal Cancer

EP: 4.Colorectal Cancer Treatment Considerations Following IDEA Collaboration Analysis

Now Viewing

EP: 5.Patient Monitoring after Adjuvant Chemotherapy in Colorectal Cancer

EP: 6.CEA vs ctDNA for Monitoring Patients with Colorectal Cancer

EP: 7.Colorectal Cancer: ctDNA vs cfDNA for Assessing MRD

EP: 8.CIRCULATE-Japan: Using ctDNA to Guide Adjuvant Therapy in CRC

EP: 9.MD Anderson INTERCEPT Program: ctDNA and Recurrence in CRC

EP: 10.CRC: Clinical Trial Landscape for MRD-Based Early Intervention

EP: 11.Effect of MRD on Adjuvant Therapy and Monitoring Strategies in CRC

EP: 12.Circulating Tumor DNA and MRD Assessment in Colorectal Cancer

Transcript:

Tanios S. Bekaii-Saab, MD: In terms of monitoring, Dan, do you intensify? If this was an MRD [minimal residual disease]–positive patient, would you change your monitoring strategy?

Daniel H. Ahn, DO: After adjuvant chemotherapy?

Tanios S. Bekaii-Saab, MD: After adjuvant chemotherapy.

Daniel H. Ahn, DO: Initially, if a patient who presented with ctDNA [circulating tumor DNA] negative completed adjuvant chemotherapy and it came back positive, then I’d have stricter surveillance in terms of surveillance imaging. The guidelines say every 6 to 12 months, and I tend to do 6 months for those who are higher risk. If the ctDNA came back positive, I’d probably even go on a little more tighter surveillance scheduled, maybe every 3 months in terms of imaging. I don’t know if it’s worthwhile to continue the ctDNA testing. Once we know it’s positive, it’s usually not going to turn back negative. I don’t think it’s false positive at that time, given the highest sensitivity with multiple withdrawals.

Tanios S. Bekaii-Saab, MD: Aparna, you’ve done quite a bit of work with a number of platforms—the liquid only, the tissue formed, the tissue uninformed. In your experience, and from the data you’ve seen, do you have any thoughts about 1 vs the other? Specifically in the setting of early stage colon cancer.

Aparna Parikh, MD, MS: Great question. Just for awareness, there are 2 types of tests that are available. The liquid-only approach doesn’t require any knowledge of the patient’s tumor. Typically, it’s a fixed genomic panel plus the inclusion of methylation, although some of the tumor-uninformed approaches are moving toward methylation-only panels. Different tumor types have different methylation signatures, and methylation signatures are unique to cancer DNA vs normal DNA. These technologies harness the methylation changes. In colon cancer that tends to be less heterogeneous than in some other cancers. That panel is then applied to all patients.

The advantage of this approach is turnaround time. Less data are available on the plasma-only approach. But with the early data, the performance characteristics seem comparable, with the caveat that there’s a lot less data for the uninformed approaches. Some of the challenges that individuals have raised with the uninformed approaches is that you have less of an ability to filter out germline and clonal hematopoiesis of indeterminate potential, although bioinformatic filters should be able to address that. In the tumor-informed approach, you need to sample the tumor, and sampling the tumor takes time. At our institution [Mass General Cancer Center], there’s no workflow for a tumor to get sent for MRD testing at the time of surgery or colonoscopy. By the time that workflow gets initiated, the patient is in the oncology clinic. Sometimes, that’s 4 to 6 weeks after surgery. You have a narrow window to start adjuvant chemotherapy. Those are some of the challenges with the tumor-informed approach.

Once you get the tumor sequencing done and you know those 15 tumor specific alterations that you’re going to follow, subsequent tests are easy and turnaround time is quick. You have the known advantages of filtering out germline, but that turnaround time remains an issue at some places. Workflows are being thought about—how to mitigate that time, particularly as data evolve—to show the value added from MRD testing. But at our institution, we haven’t figured that out.

Tanios S. Bekaii-Saab, MD: Thanks, Aparna. That’s very comprehensive.

Transcript edited for clarity.