Recap: Review of Molecular Testing Options in Metastatic Colorectal Cancer

At a recent Around the Practice program hosted by CancerNetwork^®, experts discussed the use of molecular and genetic testing to better treat patients with metastatic colorectal cancer (CRC). The panel was led by Cathy Eng, MD, professor of medicine, coleader of the Vanderbilt-Ingram Cancer Center Gastrointestinal Research Program, codirector of gastrointestinal oncology, and director of the Young Adults Program at Vanderbilt University Medical Center in Nashville, Tennessee.

She was joined by Tanios S. Bekaii-Saab, MD, leader of the Gastrointestinal Cancer Program, medical director of the Cancer Clinical Research Office, vice chair and section chief for Medical Oncology in the Department of Internal Medicine at the Mayo Clinic in Phoenix, Arizona; Kristen Ciombor, MD, MSCI, an associate professor of medicine in the Division of Hematology Oncology at Vanderbilt University Medical Center; and Jaclyn Hechtman, MD, a molecular gastrointestinal and genitourinary pathologist at Baptist Health South Florida in Miami.

ENG: What is your view on molecular testing for patients with metastatic CRC?

HECHTMAN: [Molecular testing in metastatic CRC] is for a very limited number of markers that the [National Comprehensive Cancer Network] recommends testing. For all patients with CRC, it’s recommended to do mismatch repair expression testing by IHC [immunohistochemistry] or microsatellite testing by [pathologic complete response] and next-generation sequencing [NGS] or both. For patients with distant metastases, either at diagnosis or at some time after diagnosis, extended sequencing is required. [This includes] not just exon2, but exon3, and exon4 of KRAS and NRAS [mutations] as recommended. [Additionally] BRAF V600E and then HER2 either by immunohistochemistry [IHC] reflex, FISH [fluorescence in situ hybridization], or looking by next-generation sequencing [NGS] for copy variations as well as for patients with distant metastases NTRK1 through NTRK3 patient detection.

ENG: How do we encourage our community providers as well as our patients to be aware of the current actionable mutations that exist to help optimize their overall treatment?

HECHTMAN: It goes back to the methods we use, which are IHC, NGS, FISH, and [clinical whole genome] sequencing. When we do it with large-panel NGS, we [can] pick up a lot: microsatellite instability [MSI], [tumor mutational burden], all the mutation statuses, and NTRK fusions. Sometimes barriers [exist], and we’re not able to do NGS on every single case, but [NGS] does provide a more comprehensive look; it includes that rare scale of actionable biomarkers that we’re only going to see in a couple of patients, but [when that biomarker is present], we pick it up, and we give [that patient] a long time of symptom-free life.

[Another strategy is] reflex testing. When a patient gets a diagnosis, instead of waiting for the clinician to test, we’re helping that patient not fall through the cracks by having a disease management team that says, "When we get this diagnosis, we run X, Y, and Z." Whether it’s PD-L1 [combined positive score], we send it through this panel [for testing].

ENG: What about circulating tumor DNA [ctDNA] and the use of tumor agnostic vs tumor-informed panels?

HECHTMAN: The use of ctDNA in solid tumors at this point is for early detection and minimal residual disease [MRD] detection. The use of ctDNA…is still in its infancy for colon cancers. [Some] trials have examined the utility of additional therapy if MRD is detected by ctDNA testing after curative intense surgery. If a patient is still ctDNA positive you [give them additional therapy], and if it’s still positive, we’ll go ahead and see if additional therapy for that patient would help. ctDNA may be also used to track biomarkers during targeted therapy—[that is], trials in which they have looked to see if HER2 levels fluctuate and to see if ctDNA [responds to] HER2 inhibition. [In my view, the use of ctDNA in solid tumors] is still in its infancy and is not standard of care yet.

CIOMBOR: We’re anxious to use ctDNA, but we have to be cautious because the technology has some limitations. The DYNAMIC study was presented this year at ASCO [American Society of Clinical Oncology] with a recent update at [European Society for Medical Oncology].¹ Our Japanese colleagues [also presented] the GALAXY observational cohort of CIRCULATE-Japan [at ASCO].² What we know is that ctDNA, particularly as it stands for MRD, is very prognostic, so we know that the patients who do have positive ctDNA, particularly after the completion of definitive therapy, do poorly; these are the patients we want to be able to treat differently. The problem is, we don’t know if our assays are sensitive enough yet; in some ways, and in some tumor stages and sites of recurrence, the sensitivity is lacking a bit. We need to know that if we utilize the ctDNA information we can impact patient outcomes, and that’s what we’re lacking right now. We have multiple ongoing studies that are very well designed, and we need to wait for those results to know how to move this forward as a standard of care.

BEKAII-SAAB: We have to separate the ctDNA that we typically associate with the utilization of molecular target therapies in the more advanced setting vs the MRD assessment in the earlier stages or even in all metastatic diseases. There has been a blurring of the lines in the discussion. Everything has been called ctDNA. I would encourage [clinicians] to go back to the discussion where ctDNA is essentially meant to be part of NGS or genomic assessment vs [its use in] assessing MRD or even [assessing] response. Those are not primed for use yet. Regarding ctDNA, it has become very useful in our practice.

ENG: Can you highlight the phase 3 PARADIGM trial [NCT02394795] presented at ASCO this year?³

BEKAII-SAAB: The PARADIGM trial was a study that was large and powered enough to prospectively ask if EGFR inhibitors were preferred [to targeted therapy]. Specifically, in the study, panitumumab [Vectibix] [was] preferred over bevacizumab [Avastin] [when combined with] FOLFOX-6 [leucovorin, folinic acid, fluorouracil, and oxaliplatin] in patients with left-sided RAS wild-type tumors [that] had a BRAF analysis. I would include HER2 amplifications as a potential exclusionary factor for these EGFR inhibitors on the left side. Interestingly, we’re seeing the same trends in the study. There is a survival benefit for panitumumab vs bevacizumab. This becomes more apparent after a couple of years down the line. We want to [see] that the PFS [progression-free survival] hasn’t changed much between panitumumab and bevacizumab, [for panitumumab to be used].

Most patients still do not see a significantly better benefit from panitumumab vs bevacizumab. [A] few end up driving the bottom of the curve rather than the top of the curve, and we still need to identify those patients. From a practical standpoint in the clinic, it tells us that if we have all the right elements in a left-sided tumor, [then] EGFR inhibitors should be considered in the first line. I would not get to the point where I would say, [patients need to reach a] the 100% [response rate]. The reason is that for many of our younger and relatively younger patients—meaning 70 years or younger, [approximately]—FOLFOXIRI [folinic acid, fluorouracil, oxaliplatin and irinotecan] may be an option. [FOLFOXIRI] is a preferred option, but it’s not the only option for left-sided tumors that are RAS wild-type, BRAF wild-type, and, I would add also, HER2 nonamplified.

ENG: What percentage of patients are HER2 positive?

CIOMBOR: HER2 amplification is becoming a very important marker in CRC. It is relatively uncommon, present in approximately 5%, although it is more enriched in the left-sided population. It’s important to look for this marker because not only is it a predictive marker for response to anti-HER2 therapy, but it’s also potentially a negative predictive marker for response to anti-EGFR therapy. Dr Bekaii-Saab and colleagues presented a nice study looking at this association because we’ve always wondered, and we’ve seen anecdotally in our own experience, [whether] patients who get anti-EGFR therapy and are HER2-positive…maybe didn’t do quite as well.⁴

The study looked at this. In fact, it did a systematic review over the last 20 years looking at patients who had RAS wild-type specifically, and that’s an important variable as well. These were patients who all received anti-EGFR therapy, either by itself or in combination with chemotherapy, who were also HER2 positive. This was a big cohort, given the relative rarity of the biomarkers, with more than 600 patients. Most of them did have left-sided tumors, which is what we would expect because that’s typically what we see. In this systematic review and a meta-analysis as well, the odds of response to anti-EGFR therapy were almost 50% lower in the patients who were HER2 positive than in those who were HER2 negative. Those are important data that we need to get out into the community and among our colleagues as well: [anti-EGFR therapy] is not the way to go…in the patients who are HER2 positive. [Therefore, HER2 status is important] not only for selecting patients for anti-HER2 therapy, but also to make sure we don’t give ineffective therapy to patients with this biomarker.

ENG: What toxicities are observed with EGFR therapy?

CIOMBOR: The cost, the time, [and] the toxicity are all key, and we’re whittling away in terms of who benefits from anti-EGFR. [EGFR therapy] in combination or as monotherapy, [is being learned about] more and more every day about who benefits and [with] what combination. Understanding the pathways behind this sensitivity to anti-EGFR [therapy] and mechanisms of resistance are key.

ENG: Let’s review the results of the phase 2 DESTINY-CRC01 trial (NCT03384940).⁵

CIOMBOR: DESTINY-CRC01 was an open-label phase 2 study that has been presented and updated over the last couple of years, specifically for patients with HER2-expressing metastatic CRC. These patients all had to have RAS and BRAF V600E [mutations] and needed to have had several prior regimens of chemotherapy. This differed from some of the other anti-HER2 studies, in that prior anti-HER2 therapy was allowed in this study. In DESTINY-CRC01, there were patients of all different positivity, with HER2 positivity included. There were 3 different cohorts based on the IHC and FISH testing. The dose of fam-trastuzumab deruxtecan-nxki [T-DXd; Enhertu] antibody-drug conjugate was a little bit higher, I believe, than what is typically given in breast cancer.

The primary end point here was the objective response rate [ORR]. In the patients who are IHC3 positive or IHC2 positive with FISH positivity; the ORR in this refractory population was about 45%, and we did not see any confirmed responses in the other 2 cohorts. We saw that the treatment duration in the cohort that responded best was about 5 months and had a good disease control rate of 83%. If you didn’t hit that RECIST criteria for response, there were quite a few patients who had stable disease for a decent amount of time. That’s encouraging, especially when we think that about a third of patients had already received a different anti-HER2 therapy. [Those data go] into how I decide how to sequence. Overall, PFS was quite good, almost 7 months, with a median overall survival of 15 months, which is incredible in this refractory population. It highlights the need to know if your patients are HER2 amplified and [that you have to] know what to do with that information.

Overall, the toxicities were reasonable. Gastrointestinal toxicities were what we saw the most, which you might expect. In this study, we saw some interstitial lung disease, [which also resulted in] a couple of deaths; the time to the onset of these pneumonitis-type symptoms varied. Despite good treatment with steroids, not everyone recovered from that. That’s a very important adverse effect to be aware of if you’re using T-DXd; it takes…understanding and recognition of [the possibility]. Lots of studies have been done with various regimens in the anti-HER2 space. Where I see T-DXd fitting in at the current time is after prior anti-HER2 therapy, [which] is what I do. However, we may learn more in the future about how we can [best] sequence these, given that we’re learning more and more about different regimens every day.

ENG: What advice can you offer clinicians who treat CRC, regarding molecular testing or sequencing?

CIOMBOR: We’ve seen a shift in the last 10 years. When we started doing molecular testing routinely, it was mostly for selection into clinical trials. In fact, I remember at that time saying, "Well, this patient may not be eligible for a clinical trial. Why should we send molecular testing?" It wasn’t actionable other than MSI and possibly [radioallergosorbent test] to some extent. Now, there has been a huge shift, because there are FDA-approved therapies that we need to be able to offer to our patients, and we need to know who is eligible for them. The [existence of] prognostic biomarkers…should change how you treat patients. My final point is that many of these options are now first-line options. It’s incredibly important that, as soon as the patient is diagnosed, we get this information. We have lots of ways to test these actionable biomarkers. Some are in-house, some are send-away, like NGS vs IHC. We have to have a good workflow and make sure patients are getting timely molecular testing, because it will impact all of their therapy throughout their whole disease course.

ENG: What are some unmet needs in this space?

BEKAII-SAAB: We need to continue to educate our colleagues and our patient advocacy groups about the importance of genetic testing and about the nuances of what [the results] mean. We have to continue educating our entire community about some of the elements of our research that require confirmation. We need to be a little bit more real-worldfriendly with our clinical trials. That includes being more flexible [in terms of treatment regimens], and it also means that we need to ensure that we have a population in our clinical trials that is reflective of the diversity of our patient population.

References

1. Tie J, Cohen JD, Lahouel K, et al; DYNAMIC Investigators. Circulating tumor DNAanalysis guiding adjuvant therapy in stage II colon cancer. N Engl J Med. 2022;386(24):2261-2272. doi:10.1056/ NEJMoa2200075
2. Kotaka M, Shirasu H, Watanabe J, et al. Association of circulating tumor DNAdynamics with clinical outcomes in the adjuvant setting for patients with colorectal cancer from an observational GALAXYstudy in CIRCULATE-Japan. J Clin Oncol. 2022;40(suppl 4). doi:10.1200/JCO.2022.40.4_suppl.009
3. Yoshino T, Watanabe J, Shitara K, et al. Panitumumab (PAN) plus mFOLFOX6 versus bevacizumab (BEV) plus mFOLFOX6 as first-line treatment in patients with RAS wild-type (WT) metastatic colorectal cancer (mCRC): results from the phase 3 PARADIGM trial. J Clin Oncol. 2022;40(Suppl 17):LBA1. doi:10.1200/ JCO.2022.40.17_suppl.LBA1
4. Bekaii-Saab T, Lach K, Hsu L, et al. Impact of anti- EGFRtherapies on HER2-positive metastatic colorectal cancer (HER2+ mCRC): a systematic literature review and meta-analysis of clinical outcomes. Annal Oncol. 2022;33(suppl 7):S13-S1966. doi:10.1016/annonc/ annonc1048
5. Yoshino T, Di Bartolomeo M, Raghav KPS, et al. Trastuzumab deruxtecan (T-DXd; DS-8201) in patients (pts) with HER2-expressing metastatic colorectal cancer (mCRC): final results from a phase 2, multicenter, open-label study (DESTINY-CRC01). J Clin Oncol. 2022; 40(suppl 4):119. doi:10.1200/ JCO.2022.40.4_suppl.119